6430F SAM3U User Guide

User Manual: Pdf

Open the PDF directly: View PDF .
Page Count: 1187

Download
Open PDF In Browser	View PDF

Features
• Core

•

– ARM® Cortex®-M3 revision 2.0 running at up to 96 MHz
– Memory Protection Unit (MPU)
– Thumb®-2 instruction set
Memories
– From 64 to 256 Kbytes embedded Flash, 128-bit wide access, memory accelerator,
dual bank
– From 16 to 48 Kbytes embedded SRAM with dual banks
– 16 Kbytes ROM with embedded bootloader routines (UART, USB) and IAP routines
– Static Memory Controller (SMC): SRAM, NOR, NAND support. NAND Flash
controller with 4 Kbytes RAM buffer and ECC
System
– Embedded voltage regulator for single supply operation
– POR, BOD and Watchdog for safe reset
– Quartz or resonator oscillators: 3 to 20 MHz main and optional low power 32.768
kHz for RTC or device clock.
– High precision 8/12 MHz factory trimmed internal RC oscillator with 4 MHz Default
Frequency for fast device startup
– Slow Clock Internal RC oscillator as permanent clock for device clock in low power
mode
– One PLL for device clock and one dedicated PLL for USB 2.0 High Speed Device
– Up to 17 peripheral DMA (PDC) channels and 4-channel central DMA
Low Power Modes
– Sleep and Backup modes, down to 2.5 µA in Backup mode
– Backup domain: VDDBU pin, RTC, 32 backup registers
– Ultra low power RTC: 0.6 µA
Peripherals
– USB 2.0 Device: 480 Mbps, 4-kbyte FIFO, up to 7 bidirectional Endpoints,
dedicated DMA
– Up to 4 USARTs (ISO7816, IrDA®, Flow Control, SPI, Manchester support) and one
UART
– Up to 2 TWI (I2C compatible), 1 SPI, 1 SSC (I2S), 1 HSMCI (SDIO/SD/MMC)
– 3-Channel 16-bit Timer/Counter (TC) for capture, compare and PWM
– 4-channel 16-bit PWM (PWMC)
– 32-bit Real Time Timer (RTT) and RTC with calendar and alarm features
– 8-channel 12-bit 1MSPS ADC with differential input mode and programmable gain
stage, 8-channel 10-bit ADC
I/O
– Up to 96 I/O lines with external interrupt capability (edge or level sensitivity),
debouncing, glitch filtering and on-die Series Resistor Termination
– Three 32-bit Parallel Input/Outputs (PIO)
Packages
– 100-lead LQFP, 14 x 14 mm, pitch 0.5 mm
– 100-ball TFBGA, 9 x 9 mm, pitch 0.8 mm
– 144-lead LQFP, 20 x 20 mm, pitch 0.5 mm
– 144-ball LFBGA, 10 x 10 mm, pitch 0.8 mm

AT91SAM
ARM-based
Flash MCU
SAM3U Series

6430F–ATARM–21-Feb-12

1. ATSAM3U4/2/1 Description
Atmel's SAM3U series is a member of a family of Flash microcontrollers based on the high performance 32-bit ARM Cortex-M3 RISC processor. It operates at a maximum speed of 96 MHz
and features up to 256 Kbytes of Flash and up to 52 Kbytes of SRAM. The peripheral set
includes a High Speed USB Device port with embedded transceiver, a High Speed MCI for
SDIO/SD/MMC, an External Bus Interface with NAND Flash controller, up to 4xUSARTs
(SAM3U1C/2C/4C have 3), up to 2xTWIs (SAM3U1C/2C/4C have 1), up to 5xSPIs
SAM3U1C/2C/4C have 4), as well as 4xPWM timers, 3xgeneral purpose 16-bit timers, an RTC,
a 12-bit ADC and a 10-bit ADC.
The SAM3U architecture is specifically designed to sustain high speed data transfers. It includes
a multi-layer bus matrix as well as multiple SRAM banks, PDC and DMA channels that enable it
to run tasks in parallel and maximize data throughput.
It can operate from 1.62V to 3.6V and comes in 100-pin and 144-pin LQFP and BGA packages.
The SAM3U device is particularly well suited for USB applications: data loggers, PC peripherals
and any high speed bridge (USB to SDIO, USB to SPI, USB to External Bus Interface).

1.1

Configuration Summary
The ATSAM3U4/2/1 series differ in memory sizes, package and features list. Table 1-1 summarizes the configurations of the six devices.

Table 1-1.

Configuration Summary
Flash
Organization SRAM

Number
of PIOs

Number
of
USARTs

Number
of TWI

FWUP,
SHDN
pins

External Bus
Interface

HSMCI
data
size

Device

Flash

SAM3U4E

2x128
Kbytes

dual plane

52
Kbytes

Yes

8 or 16 bits,
4 chip selects,
24-bit address

8 bits

SAM3U2E

128
Kbytes

single plane

36
Kbytes

Yes

8 or 16 bits,
4 chip selects
24-bit address

8 bits

SAM3U1E

64
Kbytes

single plane

20
Kbytes

Yes

8 or 16 bits,
4 chip selects,
24-bit address

8 bits

SAM3U4C

2 x 128
Kbytes

dual plane

52
Kbytes

FWUP

8 bits,
2 chip selects,
8-bit address

4 bits

SAM3U2C

128
Kbytes

single plane

36
Kbytes

FWUP

8 bits,
2 chip selects, 8- 4 bits
bit address

SAM3U1C

64
Kbytes

single plane

20
Kbytes

FWUP

8 bits
2 chip selects,
8-bit address

Note:

4 bits

Package

ADC

LQFP144

2 (8+ 8
channels)

BGA144
LQFP144
BGA144
LQFP144
BGA144
LQFP100
BGA100
LQFP100
BGA100
LQFP100
BGA100

2 (8+ 8
channels)
2 (8+ 8
channels)
2 (4+ 4
channels)
2 (4+ 4
channels)
2 (4+ 4
channels)

1. The SRAM size takes into account the 4-Kbyte RAM buffer of the NAND Flash Controller (NFC) which can be used by the
core if not used by the NFC.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
2. ATSAM3U4/2/1 Block Diagram

TST
PCK0
-PCK2

SLAVE

PLLA

OSC
3-20 M

PMC

VDDUTMI

SM
BOD
RC 32K

FLASH
2x128 KBytes
1x128 KBytes
1x64 KBytes

8
GPBREG

SHDN
FWUP

SUPC

RTC

VDDBU
NRSTB
ERASE
NRST

POR

PDC
8-channel
12-bit ADC
10-bit ADC

RSTC

D
VB U T
D G MII
FS
D DP
F
D SD
H M
D SD
H P
SD
M

L
SE
S

Flash
Unique
Identifier

RTT

PIOA

DMA
I/D

OSC
32K

XIN32
XOUT32

NAND Flash
Controller
& ECC

MPU

WDT

EBI

Voltage
Regulator

USB
Device
HS

SysTick Counter N
V
Cortex-M3 Processor
I
Fmax 96 MHz
C

RC Osc.
12/8/4 M

VDDCORE

JT
AG

In-Circuit Emulator

UPLL
XIN
XOUT

HS UTMI
Transceiver

JTAG & Serial Wire

System Controller

MASTER

144-pin SAM3U4/2/1E Block Diagram
TD
TDI
TMO/T
R
TC S/S AC
K/ WD ES
SW IO W
O
CL
K

Figure 2-1.

NAND Flash
SRAM
(4KBytes)

5-layer AHB Bus Matrix

SRAM0
32 KBytes
16 KBytes
8 KBytes

PDC
TWI0
TWI1

SRAM1
16 KBytes
16 KBytes

ROM
16 KBytes

Peripheral
DMA
Controller

Peripheral
Bridge

NCS0
NCS1
NRD
NWR0/NWE
NWR1/NBS1

4-Channel
DMA
Static
Memory
Controller

APB

PDC

UART

USART0
USART1
USART2
USART3

PDC
PWM

TC0
TC1
TC2

NANDRDY
D0-D15
A0/NBS0
A1
A2-A20

NWAIT
A23
A21/
NANDALE
A22/
NANDCLE
NCS3

SPI

SSC

HSMCI

PIOB

NCS2
NANDOE,
NANDWE

AD
VR

-A

2B
V
VD RE
-A DA F
D1 NA
AD
2
12 A BT
B0 D0 RG
-A -A
D
T D1 7
TW WD 2B
C 0- 7
K0 TW
-T D
W 1
C
K
U 1
R
C U XD
TS T
X
R 0-C D
SCTSO TS
R K0 -RT 3
D - S
TXX0 SC 3
D -RDK3
0- X
TX 3
D D3
C
D
0
PW
D RI0
SR
M
PW H D 0
0
T
M -P R0
L0 W
TC -P MH
L WM 3
TI K0- L3
O T
TI A0 CL
O -T K2
N B0- IOA
PC T 2
S0 IOB
-N 2
PC
SP S3
C
M K
O
M SI
IS
O
TK
TF
TD
R
D
R
F
R
DA K
0DA
C 7
DA
C
K

PIOC

3
6430F–ATARM–21-Feb-12

TST
PCK0
-PCK2

In-Circuit Emulator

PLLA

OSC
3-20 M

SysTick Counter N
V
I
C

PMC

SM
BOD
RC 32K

FLASH
2x128 KBytes
1x128 KBytes
1x64 KBytes

8
GPBREG

SHDN
FWUP

SUPC

RTC

VDDBU
NRSTB
ERASE
NRST

POR

UT
DO

DI
VD

Flash
Unique
Identifier

RTT
PDC
4-channel
12-bit ADC
10-bit ADC

RSTC

NAND Flash
SRAM
(4KBytes)

5-layer AHB Bus Matrix

SRAM0
32 KBytes
16 KBytes
8 KBytes

SRAM1
16 KBytes
16 KBytes

ROM
16 KBytes

Peripheral
DMA
Controller

Peripheral
Bridge

NCS0
NCS1
NRD
NWE

4-Channel
DMA
Static
Memory
Controller

APB

PDC

TWI

UART

USART0
USART1
USART2

NANDRDY
D0-D7
A0
A1
A2-A7

NANDALE
NANDCLE

PDC
PWM

TC0
TC1
TC2

SPI

SSC

HSMCI
NANDOE,
NANDWE

2B
V
VD RE
-A DA F
D1 N
AD
2 A
12 A BT
B0 D0 RG
-A -A
D D3
12
B3
TW
TW D
C 0
K0
U
R
C U XD
TS T
X
R 0-C D
SCTSO TS
R K0 -RT 2
D - S
TXX0 SC 2
D -RDK2
0- X
TX 2
D D2
C
D
0
PW
D RI0
SR
M
PW H D 0
0
T
M -P R0
L0 W
TC -P MH
L WM 3
TI K0- L3
O T
TI A0 CL
O -T K2
N B0- IOA
PC T 2
S0 IOB
-N 2
PC
SP S3
C
M K
O
M SI
IS
O
TK
TF
TD
R
D
R
F
DA RK
0DA
3
C
DA
C
K

PIOB

AD
VR

-A

PIOA

DMA
I/D

OSC
32K

XIN32
XOUT32

NAND Flash
Controller
& ECC

MPU

WDT
VDDUTMI

EBI

Voltage
Regulator

USB
Device
HS

Cortex-M3 Processor
Fmax 96 MHz

RC Osc.
12/8/4 M

VDDCORE

HS UTMI
Transceiver

JTAG & Serial Wire

System Controller

UPLL
XIN
XOUT

D
VB U T
D G MII
F
D SD P
FS
D DM
H
D SD
H P
SD
M

SLAVE

MASTER

100-pin SAM3U4/2/1C Block Diagram
TD
I
TD
O
TM / T
S RA
TC / S C E
K/ WD SW
SW IO O
CL
K
JT
AG
SE
L

Figure 2-2.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
3. Signal Description
Table 3-1 gives details on the signal names classified by peripheral.
Table 3-1.

Signal Description List

Signal Name

Function

Type

Active
Level

Voltage
Reference Comments

Power Supplies
VDDIO

Peripherals I/O Lines Power Supply

Power

1.62V to 3.6V

VDDIN

Voltage Regulator Input

Power

1.8V to 3.6V

VDDOUT

Voltage Regulator Output

Power

1.8V

VDDUTMII

USB UTMI+ Interface Power Supply

Power

3.0V to 3.6V

GNDUTMII

USB UTMI+ Interface Ground

Ground

VDDBU

Backup I/O Lines Power Supply

Power

GNDBU

Backup Ground

Ground

VDDPLL

PLL A, UPLL and OSC 3-20 MHz Power Supply

Power

GNDPLL

PLL A, UPLL and OSC 3-20 MHz Ground

Ground

VDDANA

ADC Analog Power Supply

Power

GNDANA

ADC Analog Ground

Ground

VDDCORE

Core, Memories and Peripherals Chip Power
Supply

Power

GND

Ground

1.62V to 3.6V

1.62 V to 1.95V

2.0V to 3.6V

1.62V to 1.95V

Clocks, Oscillators and PLLs
XIN

Main Oscillator Input

Input

XOUT

Main Oscillator Output

XIN32

Slow Clock Oscillator Input

XOUT32

Slow Clock Oscillator Output

Output

VBG

Bias Voltage Reference

Analog

PCK0 - PCK2

Programmable Clock Output

Output

VDDPLL

Output
Input

VDDBU

VDDIO

Shutdown, Wakeup Logic

SHDN

Shut-Down Control

FWUP

Force Wake-Up Input

Output

Input

VDDBU

Low

push/pull
0: The device is in
backup mode
1: The device is running
(not in backup mode)
Needs external pull-up

Serial Wire/JTAG Debug Port (SWJ-DP)
TCK/SWCLK

Test Clock/Serial Wire Clock

Input

TDI

Test Data In

Input

No pull-up resistor
VDDIO

(4)

TDO/TRACESWO

Test Data Out/Trace Asynchronous Data Out

Output

TMS/SWDIO

Test Mode Select/Serial Wire Input/Output

Input

JTAGSEL

JTAG Selection

Input

No pull-up resistor

No pull-up resistor
High

VDDBU

Internal permanent
pull-down

5
6430F–ATARM–21-Feb-12

Table 3-1.
Signal Name

Signal Description List (Continued)
Function

Type

Active
Level

Voltage
Reference Comments

Flash Memory
ERASE

Flash and NVM Configuration Bits Erase
Command

Input

High

VDDBU

I/O

Low

VDDIO

Input

Low

Internal permanent 15K
pulldown

Reset/Test
NRST

Microcontroller Reset

NRSTB

Asynchronous Microcontroller Reset

VDDBU
TST

Test Select

Input

Internal permanent
pullup
Internal permanent
pullup
Internal permanent
pulldown

Universal Asynchronous Receiver Transceiver - UART
URXD

UART Receive Data

Input

UTXD

UART Transmit Data

Output
PIO Controller - PIOA - PIOB - PIOC

PA0 - PA31

PB0 - PB31

PC0 - PC31

Parallel IO Controller A

Parallel IO Controller B

Parallel IO Controller C

I/O

•Schmitt Trigger (1)
Reset State:
•PIO Input
•Internal pullup enabled

I/O

•Schmitt Trigger (2)
Reset State:
•PIO Input
•Internal pullup enabled

VDDIO

•Schmitt Trigger(3)
Reset State:
•PIO Input
•Internal pullup enabled

I/O

External Bus Interface
D0 - D15

Data Bus

A0 - A23

Address Bus

NWAIT

External Wait Signal

I/O
Output
Input

Low

Static Memory Controller - SMC
NCS0 - NCS3

Chip Select Lines

Output

Low

NWR0 - NWR1

Write Signal

Output

Low

NRD

Read Signal

Output

Low

NWE

Write Enable

Output

Low

NBS0 - NBS1

Byte Mask Signal

Output

Low

NAND Flash Controller - NFC
NANDOE

NAND Flash Output Enable

Output

Low

NANDWE

NAND Flash Write Enable

Output

Low

NANDRDY

NAND Ready

Input

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 3-1.

Signal Description List (Continued)

Signal Name

Function

Type

Active
Level

Voltage
Reference Comments

High Speed Multimedia Card Interface - HSMCI
CK

Multimedia Card Clock

I/O

CDA

Multimedia Card Slot A Command

I/O

DA0 - DA7

Multimedia Card Slot A Data

I/O

Universal Synchronous Asynchronous Receiver Transmitter - USARTx
SCKx

USARTx Serial Clock

I/O

TXDx

USARTx Transmit Data

I/O

RXDx

USARTx Receive Data

Input

RTSx

USARTx Request To Send

CTSx

USARTx Clear To Send

DTR0

USART0 Data Terminal Ready

DSR0

USART0 Data Set Ready

Input

DCD0

USART0 Data Carrier Detect

Input

RI0

USART0 Ring Indicator

Input

Output
Input
I/O

Synchronous Serial Controller - SSC
TD

SSC Transmit Data

Output

SSC Receive Data

Input

SSC Transmit Clock

I/O

SSC Receive Clock

I/O

SSC Transmit Frame Sync

I/O

SSC Receive Frame Sync

I/O
Timer/Counter - TC

TCLKx

TC Channel x External Clock Input

Input

TIOAx

TC Channel x I/O Line A

I/O

TIOBx

TC Channel x I/O Line B

I/O

Pulse Width Modulation Controller- PWMC
PWMHx

PWM Waveform Output High for channel x

Output
only output in
complementary mode
when dead time
insertion is enabled

PWM Waveform Output Low for channel x
PWMLx

PWMFI0-2

Output

PWM Fault Input

Input
Serial Peripheral Interface - SPI

MISO

Master In Slave Out

I/O

MOSI

Master Out Slave In

I/O

SPCK

SPI Serial Clock

I/O

NPCS0

SPI Peripheral Chip Select 0

I/O

Low

NPCS1 - NPCS3

SPI Peripheral Chip Select

Output

Low

7
6430F–ATARM–21-Feb-12

Table 3-1.

Signal Description List (Continued)

Signal Name

Function

Type

Active
Level

Voltage
Reference Comments

Two-Wire Interface - TWI
TWDx

TWIx Two-wire Serial Data

I/O

TWCKx

TWIx Two-wire Serial Clock

I/O

12-bit Analog-to-Digital Converter - ADC12B
AD12Bx

Analog Inputs

AD12BTRG

ADC Trigger

AD12BVREF

ADC Reference

Analog
Input
Analog
10-bit Analog-to-Digital Converter - ADC

ADx

Analog Inputs

ADTRG

ADC Trigger

ADVREF

ADC Reference

Analog
Input
Analog
Fast Flash Programming Interface - FFPI

PGMEN0-PGMEN2 Programming Enabling

Input

PGMM0-PGMM3

Programming Mode

Input

PGMD0-PGMD15

Programming Data

I/O

PGMRDY

Programming Ready

Output

High

PGMNVALID

Data Direction

Output

Low

PGMNOE

Programming Read

Input

Low

PGMCK

Programming Clock

Input

PGMNCMD

Programming Command

Input

VDDIO

Low

USB High Speed Device - UDPHS
DFSDM

USB Device Full Speed Data -

Analog

DFSDP

USB Device Full Speed Data +

Analog

DHSDM

USB Device High Speed Data -

Analog

DHSDP

USB Device High Speed Data +

Analog

Notes:

VDDUTMII

1. PIOA: Schmitt Trigger on all except PA14 on 100 and 144 packages.
2. PIOB: Schmitt Trigger on all except PB9 to PB16, PB25 to PB31 on 100 and 144 packages.
3. PIOC: Schmitt Trigger on all except PC20 to PC27 on 144 package.
4. TDO pin is set in input mode when the Cortex-M3 Core is not in debug mode. Thus an external pull-up (100 kΩ) must be
added to avoid current consumption due to floating input.

3.1

Design Considerations
In order to facilitate schematic capture when using a SAM3U design, Atmel provides a “Schematics Checklist” Application note.
Please visit http://www.atmel.com/products/AT91/ for additional documentation.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
4. Package and Pinout
The SAM3U4/2/1E is available in 144-lead LQFP and 144-ball LFBGA packages.
The SAM3U4/2/1C is available in 100-lead LQFP and 100-ball TFBGA packages.

4.1
4.1.1

SAM3U4/2/1E Package and Pinout
144-ball LFBGA Package Outline
The 144-Ball LFBGA package has a 0.8 mm ball pitch and respects Green Standards. Its dimensions are 10 x 10 x 1.4 mm.
Figure 4-1.

Orientation of the 144-ball LFBGA Package
TOP VIEW
12
11
10
9
8
7
6
5
4
3
2
1
A B C D E F G H J

K L M

BALL A1

4.1.2

144-lead LQFP Package Outline
Figure 4-2.

Orientation of the 144-lead LQFP Package

108

109

144

37
1

9
6430F–ATARM–21-Feb-12

4.1.3

144-lead LQFP Pinout

Table 4-1.

144-pin SAM3U4/2/1E Pinout

TDI

DHSDP

VDDANA

109

PA0/PGMNCMD

VDDOUT

DHSDM

ADVREF

110

PC0

VDDIN

VBG

GNDANA

111

PA1/PGMRDY

TDO/TRACESWO

VDDUTMI

AD12BVREF

112

PC1

PB31

DFSDM

PA22/PGMD14

113

PA2/PGMNOE

PB30

DFSDP

PA30

114

PC2

TMS/SWDIO

GNDUTMI

PB3

115

PA3/PGMNVALID

PB29

VDDCORE

PB4

116

PC3

TCK/SWCLK

PA28

PC15

117

PA4/PGMM0

PB28

PA29

PC16

118

PC4

NRST

PC22

PC17

119

PA5/PGMM1

PB27

PA31

PC18

120

PC5

PB26

PC23

VDDIO

121

PA6/PGMM2

PB25

VDDCORE

122

PC6

PB24

VDDIO

PA13/PGMD5

123

PA7/PGMM3

VDDCORE

GND

PA14/PGMD6

124

PC7

VDDIO

PB0

PC10

125

VDDCORE

GND

PC24

GND

126

GND

PB23

PB1

PA15/PGMD7

127

VDDIO

PB22

PC25

PC11

128

PA8/PGMD0

PB21

PB2

PA16/PGMD8

129

PC8

PC21

PC26

PC12

130

PA9/PGMD1

PB20

PB11

PA17/PGMD9

131

PC9

PB19

GND

PB16

132

PA10/PGMD2

PB18

PB12

PB15

133

PA11/PGMD3

PB17

PB13

PC13

134

PA12/PGMD4

VDDCORE

PC27

PA18/PGMD10

135

FWUP

PC14

PA27

100

PA19/PGMD11

136

SHDN

PB14

PB5

101

PA20/PGMD12

137

ERASE

PB10

PB6

102

PA21/PGMD13

138

TST

PB9

PB7

103

PA23/PGMD15

139

VDDBU

PC19

PB8

104

VDDIO

140

GNDBU

GNDPLL

PC28

105

PA24

141

NRSTB

VDDPLL

PC29

106

PA25

142

JTAGSEL

XOUT

PC30

107

PA26

143

XOUT32

XIN

PC31

108

PC20

144

XIN32

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
4.1.4

144-ball LFBGA Pinout

Table 4-2.

144-ball SAM3U4/2/1E Pinout

VBG

DFSDM

PB0

PB7

VDDUTMI

DHSDM

PC26

PC31

PB9

GNDPLL

PB2

PC29

PB10

PC14

PC25

PB3

PB19

PB21

PB1

PB4

PC21

PB23

GND

PA14/PGMD6

PB26

PB24

GND

PA16/PGMD8

TCK/SWCLK

PB28

VDDCORE

PA18/PGMD10

PB30

TDI

PC4

PC20

A10

TDO/TRACESWO

D10

VDDBU

G10

PA6/PGMM2

K10

PA1/PGMRDY

A11

XIN32

D11

PA10/PGMD2

G11

PA7/PGMM3

K11

PC1

A12

XOUT32

D12

PA11/PGMD3

G12

PC6

K12

PC2

VDDCORE

PC22

PC24

PC30

GNDUTMI

PA28

PC27

ADVREF

XOUT

PC19

PA27

AD12BVREF

PB14

VDDCORE

PB12

PA22/PGMD14

PB17

GND

PB11

PC17

PB22

VDDIO

GND

PC10

PB25

GNDBU

VDDCORE

PC12

PB29

NRST

PB16

PA19/PGMD11

VDDIN

PB31

PB15

PA23/PGMD15

B10

JTAGSEL

E10

PA12/PGMD4

H10

PC3

L10

PA0/PGMNCMD

B11

ERASE

E11

PA8/PGMD0

H11

PA5/PGMM1

L11

PA26

B12

SHDN

E12

PC8

H12

PC5

L12

PC0

DFSDP

PA31

PB5

VDDANA

DHSDP

PA29

PB6

GNDANA

XIN

PC23

PC28

PA30

VDDPLL

VDDCORE

PB8

PC15

PB18

VDDIO

PB13

PC16

PB20

GND

VDDIO

PC18

PB27

GND

PA13/PGMD5

PA15/PGMD7

TMS/SWDIO

VDDIO

PA17/PGMD9

PC11

VDDOUT

PC9

PC13

PA20/PGMD12

C10

NRSTB

F10

PA9/PGMD1

J10

PA2/PGMNOE

M10

PA21/PGMD13

C11

TST

F11

VDDCORE

J11

PA3/PGMNVALID

M11

PA24

C12

FWUP

F12

PC7

J12

PA4/PGMM0

M12

PA25

11
6430F–ATARM–21-Feb-12

4.2
4.2.1

SAM3U4/2/1C Package and Pinout
100-lead LQFP Package Outline
Figure 4-3.

Orientation of the 100-lead LQFP Package

75
76

100

26
1

4.2.2

100-ball TFBGA Package Outline
Figure 4-4.

Orientation of the 100-ball TFBGA Package
TOP VIEW
1 2 3 4 5 6 7 8 9 10

A
B
C
D
E
F
G
H
J
K

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
4.2.3

100-lead LQFP Pinout

Table 4-3.

100-pin SAM3U4/2/1C1 Pinout

VDDANA

PA0/PGMNCMD

TDI

DHSDP

ADVREF

PA1/PGMRDY

VDDOUT

DHSDM

GNDANA

PA2/PGMNOE

VDDIN

VBG

AD12BVREF

PA3/PGMNVALID

TDO/TRACESWO

VDDUTMI

PA22/PGMD14

PA4/PGMM0

TMS/SWDIO

DFSDM

PA30

PA5/PGMM1

TCK/SWCLK

DFSDP

PB3

PA6/PGMM2

NRST

GNDUTMI

PB4

PA7/PGMM3

PB24

VDDCORE

PA28

PA13/PGMD5

GND

VDDIO

PA29

PA14/PGMD6

VDDIO

GND

PA31

PA15/PGMD7

PA8/PGMD0

PB23

VDDCORE

PA16/PGMD8

PA9/PGMD1

PB22

VDDIO

PA17/PGMD9

PA10/PGMD2

PB21

GND

PB16

PA11/PGMD3

PB20

PB0

PB15

PA12/PGMD4

PB19

PB1

PA18/PGMD10

FWUP

PB18

PB2

PA19/PGMD11

ERASE

PB17

PB11

PA20/PGMD12

TST

PB14

PB12

PA21/PGMD13

VDDBU

PB10

PB13

PA23/PGMD15

GNDBU

PB9

PA27

VDDIO

NRSTB

GNDPLL

PB5

PA24

JTAGSEL

VDDPLL

PB6

PA25

XOUT32

XOUT

PB7

PA26

XIN32

XIN

100

PB8

13
6430F–ATARM–21-Feb-12

4.2.4

100-ball TFBGA Pinout

Table 4-4.

100-ball SAM3U4/2/1C Pinout

VBG

PB22

PB1

PA15/PGMD7

XIN

TMS/SWDIO

PB12

PA18/PGMD10

XOUT

NRSTB

VDDIO

PA24

PB17

JTAGSEL

PA31

PA1/PGMRDY

PB21

C10

VDDBU

VDDIO

H10

PA2/PGMNOE

PB23

DFSDM

GND

PB6

TCK/SWCLK

DHSDM

PB16

PB8

VDDIN

VDDPLL

PA6/PGMM2

ADVREF

VDDOUT

VDDCORE

PA30

A10

XIN32

PB20

F10

PA7/PGMM3

PB3

VDDCORE

ERASE

PB11

PA16/PGMD8

GNDUTMI

TST

PB2

PA19/PGMD11

VDDUTMI

FWUP

PB0

PA21/PGMD13

PB10

PA11/PGMD3

PB13

PA26

PB18

D10

PA12/PGMD4

VDDCORE

J10

PA0/PGMNCMD

PB24

PA29

GND

PB7

NRST

GND

PB15

VDDANA

TDO/TRACESWO

PA28

PA3/PGMNVALID

GNDANA

TDI

PB9

PA5/PGMM1

AD12BVREF

B10

XOUT32

GNDBU

G10

PA4/PGMM0

PB4

DFSDP

VDDIO

VDDCORE

PA14/PGMD6

DHSDP

VDDCORE

PB5

PA17/PGMD9

GNDPLL

PA10/PGMD2

PA27

PA20/PGMD12

PB14

PA9/PGMD1

PA22/PGMD14

PA23/PGMD15

PB19

E10

PA8/PGMD0

PA13/PGMD5

K10

PA25

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
5. Power Considerations
5.1

Power Supplies
The ATSAM3U4/2/1 product has several types of power supply pins:
• VDDCORE pins: Power the core, the embedded memories and the peripherals; voltage
ranges from 1.62V to 1.95V.
• VDDIO pins: Power the Peripherals I/O lines; voltage ranges from 1.62V to 3.6V.
• VDDIN pin: Powers the Voltage regulator
• VDDOUT pin: It is the output of the voltage regulator.
• VDDBU pin: Powers the Slow Clock oscillator and a part of the System Controller; voltage
ranges from 1.62V to 3.6V. VDDBU must be supplied before or at the same time than VDDIO
and VDDCORE.
• VDDPLL pin: Powers the PLL A, UPLL and 3-20 MHz Oscillator; voltage ranges from 1.62V
to 1.95V.
• VDDUTMI pin: Powers the UTMI+ interface; voltage ranges from 3.0V to 3.6V, 3.3V nominal.
• VDDANA pin: Powers the ADC cells; voltage ranges from 2.0V to 3.6V.
Ground pins GND are common to VDDCORE and VDDIO pins power supplies.
Separated ground pins are provided for VDDBU, VDDPLL, VDDUTMI and VDDANA. These
ground pins are respectively GNDBU, GNDPLL, GNDUTMI and GNDANA.

5.2

Voltage Regulator
The ATSAM3U4/2/1 embeds a voltage regulator that is managed by the Supply Controller.
This internal regulator is intended to supply the internal core of ATSAM3U4/2/1 but can be used
to supply other parts in the application. It features two different operating modes:
• In Normal mode, the voltage regulator consumes less than 700 μA static current and draws
150 mA of output current. Internal adaptive biasing adjusts the regulator quiescent current
depending on the required load current. In Wait Mode or when the output current is low,
quiescent current is only 7μA.
• In Shutdown mode, the voltage regulator consumes less than 1 μA while its output is driven
internally to GND. The default output voltage is 1.80V and the start-up time to reach Normal
mode is inferior to 400 μs.
For adequate input and output power supply decoupling/bypassing, refer to “Voltage Regulator”
in the “Electrical Characteristics” section of the product datasheet.

5.3

Typical Powering Schematics
The ATSAM3U4/2/1 supports a 1.8V-3.6V single supply mode. The internal regulator input connected to the source and its output feed VDDCORE. Figure 5-1, Figure 5-2, Figure 5-3 show the
power schematics.

15
6430F–ATARM–21-Feb-12

Figure 5-1.

Single Supply

VDDBU

VDDUTMI

VDDANA

VDDIO

Main Supply (1.8V-3.6V)

VDDIN
Voltage
Regulator
VDDOUT

VDDCORE

VDDPLL

Note:

Restrictions
With Main Supply < 2.0 V, USB and ADC are not usable.
With Main Supply ≥ 2.4V and < 3V, USB is not usable.
With Main Supply ≥ 3V, all peripherals are usable.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 5-2.

Core Externally Supplied

VDDBU

VDDUTMI

VDDANA

Main Supply (1.62V-3.6V)

VDDIO

VDDIN
Voltage
Regulator
VDDOUT

VDDCORE Supply (1.62V-1.95V) VDDCORE

VDDPLL

Note:

Restrictions
With Main Supply < 2.0 V, USB and ADC are not usable.
With Main Supply ≥ 2.4V and < 3V, USB is not usable.
With Main Supply ≥ 3V, all peripherals are usable.

17
6430F–ATARM–21-Feb-12

Figure 5-3.

Backup Batteries Used
FWUP

SHDN

Backup Batteries VDDBU

VDDUTMI

VDDANA

VDDIO

VDDIN
Main Supply (1.62V-3.6V)

Voltage
Regulator
VDDOUT

VDDCORE

VDDPLL

Note:

Restrictions
With Main Supply < 2.0 V, USB and ADC are not usable.
With Main Supply ≥ 2.4V and < 3V, USB is not usable.
With Main Supply ≥ 3V, all peripherals are usable.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
5.4

Active Mode
Active mode is the normal running mode with the core clock running from the fast RC oscillator,
the main crystal oscillator or the PLLA. The power management controller can be used to adapt
the frequency and to disable the peripheral clocks.

5.5

Low Power Modes
The various low power modes of the ATSAM3U4/2/1 are described below:

5.5.1

Backup Mode
The purpose of backup mode is to achieve the lowest power consumption possible in a system
which is performing periodic wake-ups to perform tasks but not requiring fast startup time
(<0.5ms).
The Supply Controller, zero-power power-on reset, RTT, RTC, Backup registers and 32 kHz
Oscillator (RC or crystal oscillator selected by software in the Supply Controller) are running.
The regulator and the core supply are off.
Backup Mode is based on the Cortex-M3 deep-sleep mode with the voltage regulator disabled.
The SAM3U Series can be awakened from this mode through the Force Wake-Up pin (FWUP),
and Wake-Up input pins WKUP0 to WKUP15, Supply Monitor, RTT or RTC wake-up event. Current Consumption is 2.5 μA typical on VDDBU.
Backup mode is entered by using WFE instructions with the SLEEPDEEP bit in the System Control Register of the Cortex-M3 set to 1. (See the “Power Management” description in The “ARM
Cortex M3 Processor” section of the product datasheet).
Exit from Backup mode happens if one of the following enable wake up events occurs:
• FWUP pin (low level, configurable debouncing)
• WKUPEN0-15 pins (level transition, configurable debouncing)
• SM alarm
• RTC alarm
• RTT alarm

5.5.2

Wait Mode
The purpose of the wait mode is to achieve very low power consumption while maintaining the
whole device in a powered state for a startup time of less than 10 μs.
In this mode, the clocks of the core, peripherals and memories are stopped. However, the core,
peripherals and memories power supplies are still powered. From this mode, a fast start up is
available.
This mode is entered via Wait for Event (WFE) instructions with LPM = 1 (Low Power Mode bit in
PMC_FSMR). The Cortex-M3 is able to handle external events or internal events in order to
wake-up the core (WFE). This is done by configuring the external lines WKUP0-15 as fast
startup wake-up pins (refer to Section 5.7 “Fast Start-Up”). RTC or RTT Alarm and USB wake-up
events can be used to wake up the CPU (exit from WFE).
Current Consumption in Wait mode is typically 15 μA on VDDIN if the internal voltage regulator
is used or 8 μA on VDDCORE if an external regulator is used.

19
6430F–ATARM–21-Feb-12

Entering Wait Mode:
• Select the 4/8/12 MHz Fast RC Oscillator as Main Clock
• Set the LPM bit in the PMC Fast Startup Mode Register (PMC_FSMR)
• Execute the Wait-For-Event (WFE) instruction of the processor

Note:

5.5.3

Internal Main clock resynchronization cycles are necessary between the writing of MOSCRCEN
bit and the effective entry in Wait mode. Depending on the user application, Waiting for
MOSCRCEN bit to be cleared is recommended to ensure that the core will not execute undesired
instructions.

Sleep Mode
The purpose of sleep mode is to optimize power consumption of the device versus response
time. In this mode, only the core clock is stopped. The peripheral clocks can be enabled. This
mode is entered via Wait for Interrupt (WFI) or Wait for Event (WFE) instructions with LPM = 0 in
PMC_FSMR.
The processor can be awakened from an interrupt if WFI instruction of the Cortex M3 is used, or
from an event if the WFE instruction is used to enter this mode.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
5.5.4

Low Power Mode Summary Table
The modes detailed above are the main low power modes. Each part can be set to on or off separately and wake up sources can be individually configured. Table 5-1 below shows a summary
of the configurations of the low power modes.

Table 5-1.

Mode

Backup
Mode

Low Power Mode Configuration Summary

SUPC,
32 kHz
Oscillator
RTC RTT
Backup
Registers,
Core
POR
Memory
(VDDBU
Region)
Regulator Peripherals

Mode Entry

PIO State
Potential Wake Up Core at while in Low PIO State Consumption Wake-up
(2) (3)
Sources
Wake Up Power Mode at Wake Up
Time(1)

Previous
state saved

PIOA &
PIOB &
PIOC
Inputs with
pull ups

Any Event from: Fast
startup through
ON
Powered
Clocked
+SLEEPDEEP WKUP0-15 pins
back
RTC alarm
bit = 0
SHDN =1 (Not clocked)
+LPM bit = 1 RTT alarm
USB wake-up

Previous
state saved

Unchanged 13 μA/20 μA (5) < 10 μs

Entry mode =WFI
Interrupt Only; Entry
mode =WFE Any
WFE or WFI Enabled Interrupt
ON
Powered(7) +SLEEPDEEP and/or Any Event
Clocked
from: Fast start-up
back
bit = 0
SHDN =1 (Not clocked)
through WKUP0-15
+LPM bit = 0 pins
RTC alarm
RTT alarm
USB wake-up

Previous
state saved

Unchanged

FWUP pin
WKUP0-15 pins
OFF
OFF
+SLEEPDEEP BOD alarm
SHDN =0 (Not powered)
RTC alarm
bit = 1
RTT alarm
WFE

Reset

2.5 μA typ(4)

< 0.5 ms

WFE

Wait
Mode

Sleep
Mode

Notes:

(6)

1. When considering wake-up time, the time required to start the PLL is not taken into account. Once started, the device works
with the 4/8/12 MHz Fast RC oscillator. The user has to add the PLL start-up time if it is needed in the system. The wake-up
time is defined as the time taken for wake up until the first instruction is fetched.
2. The external loads on PIOs are not taken into account in the calculation.
3. BOD current consumption is not included.
4. Current consumption on VDDBU.
5. 13 μA total current consumption - without using internal voltage regulator.
20 μA total current consumption - using internal voltage regulator.
6. Depends on MCK frequency.
7. In this mode the core is supplied and not clocked but some peripherals can be clocked.

21
6430F–ATARM–21-Feb-12

5.6

Wake-up Sources
The wake-up events allow the device to exit backup mode. When a wake-up event is detected,
the Supply Controller performs a sequence which automatically reenables the core power
supply.

Figure 5-4.

Wake-up Source
SMEN
sm_int
RTCEN
rtc_alarm
Core
Supply
Restart

RTTEN
rtt_alarm
FWUPDBC
SLCK
FWUPEN

FWUP

WKUPT0

WKUP0

WKUPIS0

WKUPDBC
WKUPEN1

WKUPIS1

WKUPS

SLCK
Debouncer

Falling/Rising
Edge
Detector
WKUPT15

WKUP15

WKUPEN0

Falling/Rising
Edge
Detector
WKUPT1

WKUP1

FWUP
Debouncer

Falling
Edge
Detector

WKUPEN15

WKUPIS15

Falling/Rising
Edge
Detector

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
5.7

Fast Start-Up
The ATSAM3U4/2/1 device allows the processor to restart in a few microseconds while the processor is in wait mode. A fast start up can occur upon detection of a low level on one of the 19
wake-up inputs.
The fast restart circuitry, as shown in Figure 5-5, is fully asynchronous and provides a fast startup signal to the Power Management Controller. As soon as the fast start-up signal is asserted,
the PMC automatically restarts the embedded 4/8/12 MHz fast RC oscillator, switches the master clock on this 4/8/12 MHz clock and reenables the processor clock.

Figure 5-5.

Fast Start-Up Sources
USBEN
usb_wakeup
RTCEN
rtc_alarm

RTTEN
rtt_alarm

FSTT0

WKUP0

High/Low
Level
Detector

fast_restart
FSTT1

WKUP1

High/Low
Level
Detector

FSTT15

WKUP15

High/Low
Level
Detector

23
6430F–ATARM–21-Feb-12

6. Input/Output Lines
The SAM3U has different kinds of input/output (I/O) lines, such as general purpose I/Os (GPIO)
and system I/Os. GPIOs can have alternate functions thanks to multiplexing capabilities of the
PIO controllers. The same GPIO line can be used whether it is in IO mode or used by the multiplexed peripheral. System I/Os are pins such as test pin, oscillators, erase pin, analog inputs or
debug pins.
With a few exceptions, the I/Os have input schmitt triggers. Refer to the footnotes associated
with “PIO Controller - PIOA - PIOB - PIOC” on page 6 within Table 3-1, “Signal Description List”.

6.1

General Purpose I/O Lines (GPIO)
GPIO Lines are managed by PIO Controllers. All I/Os have several input or output modes such
as, pull-up, input schmitt triggers, multi-drive (open-drain), glitch filters, debouncing or input
change interrupt. Programming of these modes is performed independently for each I/O line
through the PIO controller user interface. For more details, refer to the “PIO Controller” section
of the product datasheet.
The input output buffers of the PIO lines are supplied through VDDIO power supply rail.
The SAM3U embeds high speed pads able to handle up to 65 MHz for HSMCI and SPI clock
lines and 35 MHz on other lines. See “AC Characteristics” of the product datasheet for more
details. Typical pull-up value is 100 kΩ for all I/Os.
Each I/O line also embeds an ODT (On-Die Termination), (see Figure 6-1 below). ODT consists
of an internal series resistor termination scheme for impedance matching between the driver
output (SAM3) and the PCB track impedance preventing signal reflection. The series resistor
helps to reduce I/Os switching current (di/dt) thereby reducing in turn, EMI. It also decreases
overshoot and undershoot (ringing) due to inductance of interconnect between devices or
between boards. In conclusion, ODT helps reducing signal integrity issues.
Figure 6-1.

On-Die Termination schematic
Z0 ~ Zout + Rodt

ODT
36 Ohms Typ.

Rodt

Receiver
SAM3 Driver with
Zout ~ 10 Ohms

6.2

PCB Trace
Z0 ~ 50 Ohms

System I/O Lines
System I/O lines are pins used by oscillators, test mode, reset, flash erase and JTAG to name
but a few.

6.3

Serial Wire JTAG Debug Port (SWJ-DP)
The SWJ-DP pins are TCK/SWCLK, TMS/SWDIO, TDO/SWO, TDI and commonly provided on
a standard 20-pin JTAG connector defined by ARM. For more details about voltage reference
and reset state, refer to Table 3-1, “Signal Description List”

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The JTAGSEL pin is used to select the JTAG boundary scan when asserted at a high level. It
integrates a permanent pull-down resistor of about 15 kΩ to GNDBU, so that it can be left unconnected for normal operations.
By default, the JTAG Debug Port is active. If the debugger host wants to switch to the Serial
Wire Debug Port, it must provide a dedicated JTAG sequence on TMS/SWDIO and
TCK/SWCLK which disables the JTAG-DP and enables the SW-DP. When the Serial Wire
Debug Port is active, TDO/TRACESWO can be used for trace.
The asynchronous TRACE output (TRACESWO) is multiplexed with TDO. So the asynchronous
trace can only be used with SW-DP, not JTAG-DP.
All the JTAG signals are supplied with VDDIO except JTAGSEL, supplied by VDDBU.

6.4

Test Pin
The TST pin is used for JTAG Boundary Scan Manufacturing Test or fast flash programming
mode of the ATSAM3U4/2/1 series. The TST pin integrates a permanent pull-down resistor of
about 15 kΩ to GND, so that it can be left unconnected for normal operations. To enter fast programming mode, see the “Fast Flash Programming Interface” section of the product datasheet.
For more on the manufacturing and test mode, refer to the “Debug and Test” section of the product datasheet.

6.5

NRST Pin
The NRST pin is bidirectional. It is handled by the on-chip reset controller and can be driven low
to provide a reset signal to the external components or asserted low externally to reset the
microcontroller. It will reset the Core and the peripherals, except the Backup region (RTC, RTT
and Supply Controller). There is no constraint on the length of the reset pulse and the reset controller can guarantee a minimum pulse length.
The NRST pin integrates a permanent pull-up resistor to VDDIO of about 100 kΩ.

6.6

NRSTB Pin
The NRSTB pin is input only and enables asynchronous reset of the ATSAM3U4/2/1 when
asserted low. The NRSTB pin integrates a permanent pull-up resistor of about 15 kΩ. This allows
connection of a simple push button on the NRSTB pin as a system-user reset. In all modes, this
pin will reset the chip including the Backup region (RTC, RTT and Supply Controller). It reacts as
the Power-on reset. It can be used as an external system reset source. In harsh environments, it
is recommended to add an external capacitor (10 nF) between NRSTB and VDDBU. (For filtering values refer to “I/O Characteristics” in the “Electrical Characteristics” section of the product
datasheet.)
It embeds an anti-glitch filter.

6.7

ERASE Pin
The ERASE pin is used to reinitialize the Flash content and some of its NVM bits. It integrates a
permanent pull-down resistor of about 15 kΩ to GND, so that it can be left unconnected for normal operations.
This pin is debounced by SCLK to improve the glitch tolerance. When the ERASE pin is tied high
during less than 100 ms, it is not taken into account. The pin must be tied high during more than
220 ms to perform the reinitialization of the Flash.

25
6430F–ATARM–21-Feb-12

Even in all low power modes, asserting the pin will automatically start-up the chip and erase the
Flash.

7. Processor and Architecture
7.1

ARM Cortex-M3 Processor
• Version 2.0
• Thumb-2 (ISA) subset consisting of all base Thumb-2 instructions, 16-bit and 32-bit.
• Harvard processor architecture enabling simultaneous instruction fetch with data load/store.
• Three-stage pipeline.
• Single cycle 32-bit multiply.
• Hardware divide.
• Thumb and Debug states.
• Handler and Thread modes.
• Low latency ISR entry and exit.

7.2

APB/AHB Bridges
The ATSAM3U4/2/1 product embeds two separated APB/AHB bridges:
• low speed bridge
• high speed bridge
This architecture enables to make concurrent accesses on both bridges.
All the peripherals are on the low-speed bridge except SPI, SSC and HSMCI.
The UART, 10-bit ADC (ADC), 12-bit ADC (ADC12B), TWI0-1, USART0-3, PWM have dedicated
channels for the Peripheral DMA Channels (PDC). These peripherals can not use the DMA
Controller.
The high speed bridge regroups the SSC, SPI and HSMCI. These three peripherals do not have
PDC channels but can use the DMA with the internal FIFO for Channel buffering.
Note that the peripherals of the two bridges are clocked by the same source: MCK.

7.3

Matrix Masters
The Bus Matrix of the ATSAM3U4/2/1 device manages 5 masters, which means that each master can perform an access concurrently with others to an available slave.
Each master has its own decoder and specifically defined bus. In order to simplify the addressing, all the masters have the same decoding.
Table 7-1.

List of Bus Matrix Masters

Master 0

Cortex-M3 Instruction/Data

Master 1

Cortex-M3 System

Master 2

Peripheral DMA Controller (PDC)

Master 3

USB Device High Speed DMA

Master 4

DMA Controller

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
7.4

Matrix Slaves
The Bus Matrix of the ATSAM3U4/2/1 manages 10 slaves. Each slave has its own arbiter, allowing a different arbitration per slave.
Table 7-2.

7.5

List of Bus Matrix Slaves

Slave 0

Internal SRAM0

Slave 1

Internal SRAM1

Slave 2

Internal ROM

Slave 3

Internal Flash 0

Slave 4

Internal Flash 1

Slave 5

USB Device High Speed Dual Port RAM (DPR)

Slave 6

NAND Flash Controller RAM

Slave 7

External Bus Interface

Slave 8

Low Speed Peripheral Bridge

Slave 9

High Speed Peripheral Bridge

Master to Slave Access
All the Masters can normally access all the Slaves. However, some paths do not make sense,
for example allowing access from the USB Device High speed DMA to the Internal Peripherals.
Thus, these paths are forbidden or simply not wired, and shown as “–” in Table 7-3 below.

Table 7-3.

ATSAM3U4/2/1 Master to Slave Access

Slaves

Masters

Cortex-M3
I/D Bus

Cortex-M3 S
Bus

PDC

USB Device
High Speed
DMA

DMA
Controller

Internal SRAM0

–

Internal SRAM1

–

Internal ROM

–

Internal Flash 0

–

Internal Flash 1

–

USB Device High Speed Dual Port RAM (DPR)

–

NAND Flash Controller RAM

–

External Bus Interface

–

Low Speed Peripheral Bridge

–

High Speed Peripheral Bridge

–

27
6430F–ATARM–21-Feb-12

7.6

DMA Controller
• Acting as one Matrix Master
• Embeds 4 channels:
– 3 channels with 8 bytes/FIFO for Channel Buffering
– 1 channel with 32 bytes/FIFO for Channel Buffering
• Linked List support with Status Write Back operation at End of Transfer
• Word, HalfWord, Byte transfer support.
• Handles high speed transfer of SPI, SSC and HSMCI (peripheral to memory, memory to
peripheral)
• Memory to memory transfer
• Can be triggered by PWM and T/C which enables to generate waveforms though the
External Bus Interface
The DMA controller can handle the transfer between peripherals and memory and so receives
the triggers from the peripherals listed below. The hardware interface numbers are also given in
Table 7-4 below.
Table 7-4.

7.7

DMA Controller

Instance name

Channel T/R

DMA Channel HW interface
Number

HSMCI

Transmit/Receive

SPI

Transmit

SPI

Receive

SSC

Transmit

SSC

Receive

PWM Event Line 0

Trigger

PWM Event Line 1

Trigger

TIO Output of TImer
Counter Channel 0

Trigger

Peripheral DMA Controller
• Handles data transfer between peripherals and memories
• Nineteen channels
– Two for each USART
– Two for the UART
– Two for each Two Wire Interface
– One for the PWM
– One for each Analog-to-digital Converter
• Low bus arbitration overhead
– One Master Clock cycle needed for a transfer from memory to peripheral
– Two Master Clock cycles needed for a transfer from peripheral to memory
• Next Pointer management for reducing interrupt latency requirement

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The Peripheral DMA Controller handles transfer requests from the channel according to the following priorities (Low to High priorities):
Table 7-5.

7.8

Peripheral DMA Controller

Instance name

Channel T/R

TWI1

Transmit

TWI0

Transmit

PWM

Transmit

UART

Transmit

USART3

Transmit

USART2

Transmit

USART1

Transmit

USART0

Transmit

TWI0

Receive

TWI1

Receive

UART

Receive

USART3

Receive

USART2

Receive

USART1

Receive

USART0

Receive

ADC

Receive

ADC12B

Receive

Debug and Test Features
• Debug access to all memory and registers in the system, including Cortex-M3 register bank
when the core is running, halted, or held in reset.
• Serial Wire Debug Port (SW-DP) and Serial Wire JTAG Debug Port (SWJ-DP) debug access
• Flash Patch and Breakpoint (FPB) unit for implementing break points and code patches
• Data Watchpoint and Trace (DWT) unit for implementing watch points, data tracing, and
system profiling
• Instrumentation Trace Macrocell (ITM) for support of printf style debugging
• IEEE® 1149.1 JTAG Boundary-scan on all digital pins

29
6430F–ATARM–21-Feb-12

8. Product Mapping
Figure 8-1.

ATSAM3U4/2/1 Memory Mapping
0x00000000

Code

0x00000000

Address memory space

0x40000000

Code

0x40004000

Peripherals
MCI

Boot Memory
0x00080000

SSC

Internal Flash 0
0x00100000

0x40008000

0x20000000

SPI

Internal Flash 1
0x00180000

0x4000C000

Internal SRAM

0x40080000

0x40000000

0x00200000
Reserved
0x1FFFFFFF

1 MByte
bit band
region

+0x40

Peripherals
Internal SRAM

+0x80

0x60000000

SRAM0
0x20080000

TC1
TC2

TWI1

NFC (SRAM)
0x4008C000

Reserved

PWM

UDPHS (DMA)
0x20200000

0x40090000

0xE0000000

USART0

Undefined
0x22000000

0x40094000

System
32 MBytes
bit band alias

USART1
0x40098000

0xFFFFFFFF

USART2

Undefined
0x40000000

0x400E0000

System Controller

0x4009C000
USART3

SMC
0x400A0000

0x400E0200

External SRAM

PMC
0x400E0600

Chip Select 1

0x400E0740

0x61000000

UART

0x62000000

UDPHS
5

0x400A8000

0x400AC000

ADC12B
ADC

CHIPID
Chip Select 2

0x400B0000

0x400E0800

0x63000000

EFC0
Chip Select 3

0x400E0A00

0x64000000

EFC1
reserved

0x400E0C00

0x68000000

PIOA
NFC

0x400E0E00

reserved

0x400E1000

0x69000000

PIOB

0x9FFFFFFF

PIOC
0x400E1200
RSTC
ID

+0x10

DMAC
6

24
18
19
25
13
14

1 MByte
bit band
region

15
16

0x400E0000

0x400E2600

RTT
WDT
+0x60
RTC
SYSC

26
27
28

0x40100000

0x42000000

0x44000000

System Controller

Reserved
Reserved
32 MBytes
bit band alias
Reserved

SUPC

+0x50

0x400B3FFF

Reserved

0x60000000

+0x30

+0x90

0x400A4000

0x400E0400

Chip Select 0

Reserved

MATRIX

block
peripheral

TC0

0x40088000

0xA0000000

0x20180000

offset

TC0

TWI0

0x20100000

0x60000000

TC0

0x40084000

External SRAM
SRAM1

0x24000000

Reserved

Internal ROM

0x20000000

3
4
2

GPBR

0x400E1400
reserved
0x4007FFFF

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
9. Memories
The embedded and external memories are described below.

9.1
9.1.1

Embedded Memories
Internal SRAM
The SAM3U4 (256 KBytes internal Flash version) embeds a total of 48 Kbytes high-speed
SRAM (32 Kbytes SRAM0 and 16 Kbytes SRAM1).
The SAM3U2 (128 KBytes internal Flash version) embeds a total of 32 Kbytes high-speed
SRAM (16 Kbytes SRAM0 and 16 Kbytes SRAM1).
The SAM3U1 (64 KBytes internal Flash version) embeds a total of 16 Kbytes high-speed SRAM
(8 Kbytes SRAM0 and 8 Kbytes SRAM1).
The SRAM0 is accessible over System Cortex-M3 bus at address 0x2000 0000 and SRAM1 at
address 0x2008 0000. The user can see the SRAM as contiguous at 0x20078000-0x20083FFF
(SAM3U4), 0x2007C000-0x20083FFFF (SAM3U2) or 0x2007E000-0x20081FFFF (SAM3U1).
The SRAM0 and SRAM1 are in the bit band region. The bit band alias region is from 0x2200
0000 and 0x23FF FFFF.
The NAND Flash Controller embeds 4224 bytes of internal SRAM. If the NAND Flash controller
is not used, these 4224 bytes of SRAM can be used as general purpose. It can be seen at
address 0x2010 0000.

9.1.2

Internal ROM
The ATSAM3U4/2/1 product embeds an Internal ROM, which contains the SAM-BA Boot and
FFPI program.
At any time, the ROM is mapped at address 0x0018 0000.

9.1.3
9.1.3.1

Embedded Flash
Flash Overview
The Flash of the SAM3U4 (256 KBytes internal Flash version) is organized in two banks of 512
pages (dual plane) of 256 bytes.
The Flash of the SAM3U2 (128 KBytes internal Flash version) is organized in one bank of 512
pages (single plane) of 256 bytes.
The Flash of the SAM3U1 (64 KBytes internal Flash version) is organized in one bank of 256
pages (single plane) of 256 bytes.
The Flash contains a 128-byte write buffer, accessible through a 32-bit interface.

9.1.3.2

Flash Power Supply
The Flash is supplied by VDDCORE.

9.1.3.3

Enhanced Embedded Flash Controller
The Enhanced Embedded Flash Controller (EEFC) manages accesses performed by the masters of the system. It enables reading the Flash and writing the write buffer. It also contains a
User Interface, mapped within the Memory Controller on the APB.

31
6430F–ATARM–21-Feb-12

The Enhanced Embedded Flash Controller ensures the interface of the Flash block with the 32bit internal bus. Its 128-bit wide memory interface increases performance.
The user can choose between high performance or lower current consumption by selecting
either 128-bit or 64-bit access. It also manages the programming, erasing, locking and unlocking
sequences of the Flash using a full set of commands.
One of the commands returns the embedded Flash descriptor definition that informs the system
about the Flash organization, thus making the software generic.
The SAM3U4 (256 KBytes internal Flash version) embeds two EEFC (EEFC0 for Flash0 and
EEFC1 for Flash1) whereas the SAM3U2/1 embeds one EEFC.
9.1.3.4

Lock Regions
In the SAM3U4 (256 KBytes internal Flash version) two Enhanced Embedded Flash Controllers
each manage 16 lock bits to protect 32 regions of the flash against inadvertent flash erasing or
programming commands.
The SAM3U4 (256 KBytes internal Flash version) contains 32 lock regions and each lock region
contains 32 pages of 256 bytes. Each lock region has a size of 8 Kbytes.
The SAM3U2 (128 KBytes internal Flash version) Enhanced Embedded Flash Controller manages 16 lock bits to protect 32 regions of the flash against inadvertent flash erasing or
programming commands.
The SAM3U2 (128 KBytes internal Flash version) contains 16 lock regions and each lock region
contains 32 pages of 256 bytes. Each lock region has a size of 8 Kbytes.
The SAM3U1(64 KBytes internal Flash version) Embedded Flash Controller manages 8 lock bits
to protect 8 regions of the flash against inadvertent flash erasing or programming commands.
The SAM3U1(64 KBytes internal Flash version) contains 8 lock regions and each lock region
contains 32 pages of 256 bytes. Each lock region has a size of 8 Kbytes.
If a locked-region’s erase or program command occurs, the command is aborted and the EEFC
triggers an interrupt.
The lock bits are software programmable through the EEFC User Interface. The command “Set
Lock Bit” enables the protection. The command “Clear Lock Bit” unlocks the lock region.
Asserting the ERASE pin clears the lock bits, thus unlocking the entire Flash.

9.1.3.5

Security Bit Feature
The ATSAM3U4/2/1 features a security bit, based on a specific General Purpose NVM bit
(GPNVM bit 0). When the security is enabled, any access to the Flash, SRAM, Core Registers
and Internal Peripherals either through the ICE interface or through the Fast Flash Programming
Interface, is forbidden. This ensures the confidentiality of the code programmed in the Flash.
This security bit can only be enabled, through the command “Set General Purpose NVM Bit 0” of
the EEFC User Interface. Disabling the security bit can only be achieved by asserting the
ERASE pin at 1, and after a full Flash erase is performed. When the security bit is deactivated,
all accesses to the Flash, SRAM, Core Registers and Internal Peripherals either through the ICE
interface or through the Fast Flash Programming Interface are permitted.
It is important to note that the assertion of the ERASE pin should always be longer than 200 ms.
As the ERASE pin integrates a permanent pull-down, it can be left unconnected during normal

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
operation. However, it is safer to connect it directly to GND for the final application.
9.1.3.6

Calibration Bits
NVM bits are used to calibrate the brownout detector and the voltage regulator. These bits are
factory configured and cannot be changed by the user. The ERASE pin has no effect on the calibration bits.

9.1.3.7

Unique Identifier
Each device integrates its own 128-bit unique identifier. These bits are factory configured and
cannot be changed by the user. The ERASE pin has no effect on the unique identifier.

9.1.3.8

Fast Flash Programming Interface
The Fast Flash Programming Interface allows programming the device through either a serial
JTAG interface or through a multiplexed fully-handshaked parallel port. It allows gang programming with market-standard industrial programmers.
The FFPI supports read, page program, page erase, full erase, lock, unlock and protect
commands.
The Fast Flash Programming Interface is enabled and the Fast Programming Mode is entered
when TST, NRSTB and FWUP pins are tied high during power up sequence and if all supplies
are provided externally (do not use internal regulator for VDDCORE). Please note that since the
FFPI is a part of the SAM-BA Boot Application, the device must boot from the ROM.

9.1.3.9

SAM-BA® Boot
The SAM-BA Boot is a default Boot Program which provides an easy way to program in-situ the
on-chip Flash memory.
The SAM-BA Boot Assistant supports serial communication via the UART and USB.
The SAM-BA Boot provides an interface with SAM-BA Graphic User Interface (GUI).
The SAM-BA Boot is in ROM and is mapped in Flash at address 0x0 when GPNVM bit 1 is set
to 0.

9.1.3.10

GPNVM Bits
The ATSAM3U4/2/1 features three GPNVM bits that can be cleared or set respectively through
the commands “Clear GPNVM Bit” and “Set GPNVM Bit” of the EEFC User Interface.
The SAM3U4 is equipped with two EEFC, EEFC0 and EEFC1. EEFC1 does not feature the
GPNVM bits. The GPNVM embedded on EEFC0 applies to the two blocks in the SAM3U4.
Table 9-1.

General-purpose Non-volatile Memory Bits

GPNVMBit[#]

Function

Security bit

Boot mode selection

Flash selection (Flash 0 or Flash 1) Only on SAM3U4 (256 Kbytes internal
Flash version)

33
6430F–ATARM–21-Feb-12

9.1.4

Boot Strategies
The system always boots at address 0x0. To ensure a maximum boot possibilities the memory
layout can be changed via GPNVM.
A general purpose NVM (GPNVM1) bit is used to boot either on the ROM (default) or from the
Flash.
The GPNVM bit can be cleared or set respectively through the commands “Clear General-purpose NVM Bit” and “Set General-purpose NVM Bit” of the EEFC User Interface.
Setting the GPNVM Bit 1 selects the boot from the Flash, clearing it selects the boot from the
ROM. Asserting ERASE clears the GPNVM Bit 1 and thus selects the boot from the ROM by
default.
GPNVM2 enables to select if Flash 0 or Flash 1 is used for the boot. Setting the GPNVM2 bit
selects the boot from Flash 1, clearing it selects the boot from Flash 0.

9.2

External Memories
The ATSAM3U4/2/1 offers an interface to a wide range of external memories and to any parallel
peripheral.

9.2.1

Static Memory Controller
• 8- or 16- bit Data Bus
• Up to 24-bit Address Bus (up to 16 MBytes linear per chip select)
• Up to 4 chips selects, Configurable Assignment
• Multiple Access Modes supported
– Byte Write or Byte Select Lines
• Multiple device adaptability
– Control signals programmable setup, pulse and hold time for each Memory Bank
• Multiple Wait State Management
– Programmable Wait State Generation
– External Wait Request
– Programmable Data Float Time
• Slow Clock mode supported

9.2.2

NAND Flash Controller
• Handles automatic Read/Write transfer through 4224 bytes SRAM buffer
• DMA support
• Supports SLC NAND Flash technology
• Programmable timing on a per chip select basis
• Programmable Flash Data width 8-bit or 16-bit

9.2.3

NAND Flash Error Corrected Code Controller
• Integrated in the NAND Flash Controller
• Single bit error correction and 2-bit Random detection.
• Automatic Hamming Code Calculation while writing
– ECC value available in a register
• Automatic Hamming Code Calculation while reading

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
– Error Report, including error flag, correctable error flag and word address being
detected erroneous
– Supports 8- or 16-bit NAND Flash devices with 512-, 1024-, 2048- or 4096-byte
pages

35
6430F–ATARM–21-Feb-12

10. System Controller
The System Controller is a set of peripherals, which allow handling of key elements of the system, such as power, resets, clocks, time, interrupts, watchdog, etc...
The System Controller User Interface also embeds the registers used to configure the Matrix.
See the system controller block diagram in Figure 10-1 on page 37.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 10-1. System Controller Block Diagram
VDDBU

VDDIN

vr_standby
Software Controlled
Voltage Regulator

FWUP

VDDOUT

SHDN
WKUP0 - WKUP15
NRSTB

Supply
Controller

VDDIO
PIOA/B/C
Input / Output Buffers

Zero-Power
Power-on Reset

PIOx
VDDANA

General Purpose
Backup Registers

ADVREF

ADC (front-end)

ADx
SLCK

RTC

rtc_alarm
bodbup_in

SLCK

RTT

VDDUTMI

Supply
Monitor

bodbup_on
rtt_alarm

USBx

USB

osc32k_xtal_en
VDDCORE

vddcore_nreset
XIN32

XTALSEL

Xtal 32 kHz
Oscillator

XOUT32

Embedded
32 kHz RC
Oscillator

Slow Clock
SLCK

bodcore_on

Brownout
Detector

bodcore_in
supc_interrupt

osc32k_rc_en

SRAM

Backup Power Supply
Peripherals
vddcore_nreset

proc_nreset
periph_nreset
ice_nreset

Reset
Controller

NRST

Cortex-M3

Matrix
Peripheral
Bridge

FSTT0 - FSTT15(1)
Embedded
12 / 8 / 4 MHz
RC
Oscillator

XIN

Main Clock
MAINCK

3 - 20 MHz
XTAL Oscillator

XOUT

MAINCK

PLLACK
PLLA

MAINCK

Flash

SLCK

Power
Management
Controller

Master Clock
MCK

SLCK

Watchdog
Timer

UPLLCK
UPLL

Core Power Supply

FSTT0 - FSTT15 are possible Fast Startup Sources, generated by WKUP0-WKUP15 Pins,
but are not physical pins.

37
6430F–ATARM–21-Feb-12

10.1

System Controller and Peripheral Mapping
Please refer to Figure 8-1“ATSAM3U4/2/1 Memory Mapping” on page 30 .
All the peripherals are in the bit band region and are mapped in the bit band alias region.

10.2

Power-on-Reset, Brownout and Supply Monitor
The SAM3U embeds three features to monitor, warn and/or reset the chip:
• Power-on-Reset on VDDBU
• Brownout Detector on VDDCORE
• Supply Monitor on VDDUTMI

10.2.1

Power-on-Reset on VDDBU
The Power-on-Reset monitors VDDBU. It is always activated and monitors voltage at start up
but also during power down. If VDDBU goes below the threshold voltage, the entire chip is reset.
For more information, refer to the “Electrical Characteristics” section of the datasheet.

10.2.2

Brownout Detector on VDDCORE
The Brownout Detector monitors VDDCORE. It is active by default. It can be deactivated by software through the Supply Controller (SUPC_MR). It is especially recommended to disable it
during low-power modes such as wait or sleep modes.
If VDDCORE goes below the threshold voltage, the reset of the core is asserted. For more information, refer to the “Supply Controller” and “Electrical Characteristics” sections of the product
datasheet.

10.2.3

10.3

Supply Monitor on VDDUTMI
The Supply Monitor monitors VDDUTMI. It is not active by default. It can be activated by software and is fully programmable with 16 steps for the threshold (between 1.9V to 3.4V). It is
controlled by the Supply Controller. A sample mode is possible. It allows to divide the supply
monitor power consumption by a factor of up to 2048. For more information, refer to the “Supply
Controller” and “Electrical Characteristics” sections of the product datasheet.

Reset Controller
The Reset Controller is capable to return to the software the source of the last reset, either a
general reset, a wake-up reset, a software reset, a user reset or a watchdog reset.
The Reset Controller controls the internal resets of the system and the NRST pin output. It is
capable to shape a reset signal for the external devices, simplifying to a minimum connection of
a push-button on the NRST pin to implement a manual reset.

10.4

Supply Controller
The Supply Controller controls the power supplies of each section of the processor and the
peripherals (via Voltage regulator control).
The Supply Controller has its own reset circuitry and is clocked by the 32 kHz Slow clock
generator.
The reset circuitry is based on a zero-power power-on reset cell. The zero-power power-on reset
allows the Supply Controller to start properly.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The Slow Clock generator is based on a 32 kHz crystal oscillator and an embedded 32 kHz RC
oscillator. The Slow Clock defaults to the RC oscillator, but the software can enable the crystal
oscillator and select it as the Slow Clock source.
The Supply Controller starts up the device by enabling the Voltage Regulator, then it generates
the proper reset signals to the core power supply.
It also enables to set the system in different low power modes and to wake it up from a wide
range of events.

10.5

Clock Generator
The Clock Generator is made up of:
• One Low Power 32768 Hz Slow Clock Oscillator with bypass mode
• One Low Power RC Oscillator
• One 3 to 20 MHz Crystal Oscillator, which can be bypassed
• One Fast RC Oscillator factory programmed, 3 output frequencies can be selected: 4, 8 or 12
MHz. By default 4 MHz is selected. 8 MHz and 12 MHz output are factory calibrated.
• One 480 MHz UPLL providing a clock for the USB High Speed Device Controller. Input
frequency is 12 MHz (only).
• One 96 to 192 MHz programmable PLL (PLL A), capable to provide the clock MCK to the
processor and to the peripherals. The input frequency of the PLL A is between 8 and 16 MHz.
Figure 10-2. Clock Generator Block Diagram
Clock Generator
XTALSEL
On Chip
32k RC OSC

XIN32
XOUT32

XIN

Slow Clock
SLCK
Slow Clock
Oscillator

12M Main
Oscillator
Main Clock
MAINCK

XOUT
On Chip
12/8/4 MHz
RC OSC

MAINSEL

PLL B

HSCK

Divider
/6 /8

UPLL Clock
UPLLCK

PLL and
Divider A

PLLA Clock
PLLACK

Status

Control

Power
Management
Controller

39
6430F–ATARM–21-Feb-12

10.6

Power Management Controller
The Power Management Controller provides all the clock signals to the system. It provides:
• the Processor Clock HCLK
• the Free running processor clock FCLK
• the Cortex SysTick external clock
• the Master Clock MCK, in particular to the Matrix and the memory interfaces
• the USB Device HS Clock UDPCK
• independent peripheral clocks, typically at the frequency of MCK
• three programmable clock outputs: PCK0, PCK1 and PCK2
The Supply Controller selects between the 32 kHz RC oscillator or the crystal oscillator. The
unused oscillator is disabled automatically so that power consumption is optimized.
By default, at startup the chip runs out of the Master Clock using the Fast RC Oscillator running
at 4 MHz.

Figure 10-3.

Power Management Controller Block Diagram
Processor
Clock
Controller

HCK
int

Sleep Mode

Divider
/8

SystTick

FCLK
Master Clock Controller
SLCK
MAINCK
PLLACK
PLLBCK

Prescaler
/1,/2,/4,...,/64

MCK
Peripherals
Clock Controller

periph_clk[..]

ON/OFF

Programmable Clock Controller
SLCK
MAINCK
PLLACK
PLLBCK

ON/OFF
Prescaler
/1,/2,/4,...,/64

pck[..]

USB Clock Controller
ON/OFF
HSCK

UDPCK

The SysTick calibration value is fixed at 10500, which allows the generation of a time base of
1 ms with SystTick clock to 10.5 MHz (max HCLK/8).

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
10.7

Watchdog Timer
• 16-bit key-protected once-only Programmable Counter
• Windowed, prevents the processor from being in a dead-lock on the watchdog access

10.8

SysTick Timer
• 24-bit down counter
• Self-reload capability
• Flexible system timer

10.9

Real-time Timer
• Real-time Timer, allowing backup of time with different accuracies
– 32-bit Free-running back-up Counter
– Integrates a 16-bit programmable prescaler running on slow clock
– Alarm Register capable to generate a wake-up of the system

10.10 Real-time Clock
• Low power consumption
• Full asynchronous design
• Two hundred year calendar
• Programmable Periodic Interrupt
• Alarm and update parallel load
• Control of alarm and update Time/Calendar Data In

10.11 General-Purpose Back-up Registers
• Eight 32-bit general-purpose backup registers

10.12 Nested Vectored Interrupt Controller
• Thirty maskable interrupts
• Sixteen priority levels
• Dynamic reprioritization of interrupts
• Priority grouping
– selection of preempting interrupt levels and non preempting interrupt levels.
• Support for tail-chaining and late arrival of interrupts.
– back-to-back interrupt processing without the overhead of state saving and
restoration between interrupts.
• Processor state automatically saved on interrupt entry, and restored on
– interrupt exit, with no instruction overhead.

41
6430F–ATARM–21-Feb-12

10.13 Chip Identification
• Chip Identifier (CHIPID) registers permit recognition of the device and its revision.
Table 10-1.

ATSAM3U4/2/1 Chip IDs Register - Engineering Samples

Chip Name

Flash Size
KByte

Pin Count

CHIPID_CIDR

CHIPID_EXID

SAM3U4C

256

100

0x28000960

0x0

SAM3U2C

128

100

0x280A0760

0x0

SAM3U1C

100

0x28090560

0x0

SAM3U4E

256

144

0x28100960

0x0

SAM3U2E

128

144

0x281A0760

0x0

SAM3U1E

144

0x28190560

0x0

• JTAG ID: 0x0582A03F

Table 10-2.

ATSAM3U4/2/1 Chip IDs Register - Revision A Parts

Chip Name

Flash Size
KByte

Pin Count

CHIPID_CIDR

CHIPID_EXID

SAM3U4C (Rev A)

256

100

0x28000961

0x0

SAM3U2C (Rev A)

128

100

0x280A0761

0x0

SAM3U1C (Rev A)

100

0x28090561

0x0

SAM3U4E (Rev A)

256

144

0x28100961

0x0

SAM3U2E (Rev A)

128

144

0x281A0761

0x0

SAM3U1E (Rev A)

144

0x28190561

0x0

• JTAG ID: 0x0582A03F

10.14 PIO Controllers
• 3 PIO Controllers, PIOA, PIOB, and PIOC, controlling a maximum of 96 I/O Lines
• Each PIO Controller controls up to 32 programmable I/O Lines
– PIOA has 32 I/O Lines
– PIOB has 32 I/O Lines
– PIOC has 32 I/O Lines
• Fully programmable through Set/Clear Registers
• Multiplexing of two peripheral functions per I/O Line
• For each I/O Line (whether assigned to a peripheral or used as general purpose I/O)
– Input change, rising edge, falling edge, low level and level interrupt
– Debouncing and Glitch filter
– Multi-drive option enables driving in open drain
– Programmable pull up on each I/O line
– Pin data status register, supplies visibility of the level on the pin at any time
• Synchronous output, provides Set and Clear of several I/O lines in a single write

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
11. Peripherals
11.1

Peripheral Identifiers
Table 11-1 defines the Peripheral Identifiers of the ATSAM3U4/2/1. A peripheral identifier is
required for the control of the peripheral interrupt with the Nested Vectored Interrupt Controller
and for the control of the peripheral clock with the Power Management Controller.
Note that some Peripherals are always clocked. Please refer to the table below.

Table 11-1.

Peripheral Identifiers
PMC
Clock Control

Instance ID

Instance Name

NVIC
Interrupt

SUPC

Supply Controller

RSTC

Reset Controller

RTC

Real Time Clock

RTT

Real Time Timer

WDT

Watchdog Timer

PMC

Power Management Controller

EEFC0

Enhanced Embedded Flash Controller 0

EEFC1

Enhanced Embedded Flash Controller 1

UART

Universal Asynchronous Receiver Transmitter

SMC

Static Memory Controller

PIOA

Parallel I/O Controller A,

PIOB

Parallel I/O Controller B

PIOC

Parallel I/O Controller C

USART0

USART 0

USART1

USART 1

USART2

USART 2

USART3

USART 3

HSMCI

High Speed Multimedia Card Interface

TWI0

Two-Wire Interface 0

TWI1

Two-Wire Interface 1

SPI

Serial Peripheral Interface

SSC

Synchronous Serial Controller

TC0

Timer Counter 0

TC1

Timer Counter 1

TC2

Timer Counter 2

PWM

Pulse Width Modulation Controller

ADC12B

12-bit ADC Controller

ADC

10-bit ADC Controller

DMAC

DMA Controller

UDPHS

USB Device High Speed

Instance Description

43
6430F–ATARM–21-Feb-12

11.2

Peripheral Signal Multiplexing on I/O Lines
The ATSAM3U4/2/1 features 3 PIO controllers, PIOA, PIOB and PIOC that multiplex the I/O
lines of the peripheral set.
Each PIO Controller controls up to 32 lines. Each line can be assigned to one of two peripheral
functions, A or B. The multiplexing tables in the following pages define how the I/O lines of
peripherals A and B are multiplexed on the PIO Controllers. The two columns “Extra Function”
and “Comments” have been inserted in this table for the user’s own comments, they may be
used to track how pins are defined in an application.
Note that some peripheral functions which are output only, might be duplicated within the tables.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
11.2.1

PIO Controller A Multiplexing

Table 11-2.

Multiplexing on PIO Controller A (PIOA)

I/O Line

Peripheral A

Peripheral B

Extra Function

PA0

TIOB0

NPCS1

WKUP0

PA1

TIOA0

NPCS2

WKUP1(1)(2)

PA2

TCLK0

ADTRG

WKUP2(1)(2)

PA3

MCCK

PCK1

PA4

MCCDA

PWMH0

PA5

MCDA0

PWMH1

PA6

MCDA1

PWMH2

PA7

MCDA2

PWML0

PA8

MCDA3

PWML1

PA9

TWD0

PWML2

WKUP3(1)(2)

PA10

TWCK0

PWML3

WKUP4(1)(2)

PA11

URXD

PWMFI0

PA12

UTXD

PWMFI1

PA13

MISO

PA14

MOSI

PA15

SPCK

PWMH2

PA16

NPCS0

NCS1

WKUP5(1)(2)

PA17

SCK0

AD12BTRG

WKUP6(1)(2)

PA18

TXD0

PWMFI2

WKUP7(1)(2)

PA19

RXD0

NPCS3

WKUP8(1)(2)

PA20

TXD1

PWMH3

WKUP9(1)(2)

PA21

RXD1

PCK0

WKUP10(1)(2)

PA22

TXD2

RTS1

AD12B0

PA23

RXD2

CTS1

PA24

(3)

TWD1

SCK1

WKUP11(1)(2)
WKUP12(1)(2)

PA25

TWCK1(3)

SCK2

PA26

TCLK2

PA27

PCK0

PA28

PWMH0

PA29

PWMH1

PA30

TIOA2

PA31

TIOB2

Notes:

Comments

(1)(2)

AD12B1

1. Wake-Up source in Backup mode (managed by the SUPC).
2. Fast Start-Up source in Wait mode (managed by the PMC).
3. Only on 144-pin version

45
6430F–ATARM–21-Feb-12

11.2.2

PIO Controller B Multiplexing

Table 11-3.
I/O Line

Multiplexing on PIO Controller B (PIOB)
Peripheral A

Peripheral B

Extra Function

Comments

(1)(2)

PB0

PWMH0

WKUP13

PB1

PWMH1

WKUP14(1)(2)

PB2

PWMH2

WKUP15(1)(2)

PB3

PWMH3

AD12B2

PB4

TCLK1

AD12B3

PB5

TIOA1

AD0

PB6

TIOB1

D15

AD1

PB7

RTS0

A0/NBS0

AD2

PB8

CTS0

AD3

PB9

DTR0

PB10

DSR0

PB11

DCD0

PB12

RI0

PB13

PWMH0

PB14

PWMH1

PB15

PWMH2

PB16

PWMH3

PB17

NANDOE

PWML0

PB18

NANDWE

PWML1

PB19

NRD

PWML2

PB20

NCS0

PWML3

PB21

A21/NANDALE

RTS2

PB22

A22/NANDCLE

CTS2

PB23

NWR0/NWE

PCK2

PB24

NANDRDY

PCK1

PB25

PWML0

Only on 144-pin version

PB26

PWML1

Only on 144-pin version

PB27

D10

PWML2

Only on 144-pin version

PB28

D11

PWML3

Only on 144-pin version

PB29

D12

Only on 144-pin version

PB30

D13

Only on 144-pin version

PB31

D14

Only on 144-pin version

Notes:

1. Wake-Up source in Backup mode (managed by the SUPC).
2. Fast Start-Up source in Wait mode (managed by the PMC).

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
11.2.3

PIO Controller C Multiplexing

Table 11-4.

Multiplexing on PIO Controller C (PIOC)

I/O Line

Peripheral A

Peripheral B

Extra function

Comments

PC0

Only on 144-pin version

PC1

Only on 144-pin version

PC2

Only on 144-pin version

PC3

NPCS1

Only on 144-pin version

PC4

NPCS2

Only on 144-pin version

PC5

NPCS3

Only on 144-pin version

PC6

PWML0

Only on 144-pin version

PC7

PWML1

Only on 144-pin version

PC8

A10

PWML2

Only on 144-pin version

PC9

A11

PWML3

Only on 144-pin version

PC10

A12

CTS3

Only on 144-pin version

PC11

A13

RTS3

Only on 144-pin version

PC12

NCS1

TXD3

Only on 144-pin version

PC13

RXD3

Only on 144-pin version

PC14

NPCS2

Only on 144-pin version

PC15

NWR1/NBS1

PC16

NCS2

PC17

AD12B4

Only on 144-pin version

AD12B5

Only on 144-pin version

NCS3

AD12B6

Only on 144-pin version

PC18

NWAIT

AD12B7

Only on 144-pin version

PC19

SCK3

PC20

A14

Only on 144-pin version

PC21

A15

Only on 144-pin version

PC22

A16

Only on 144-pin version

PC23

A17

Only on 144-pin version

PC24

A18

PWMH0

Only on 144-pin version

PC25

A19

PWMH1

Only on 144-pin version

PC26

A20

PWMH2

Only on 144-pin version

PC27

A23

PWMH3

Only on 144-pin version

PC28

PWML3

NPCS1

Only on 144-pin version

MCDA4

AD4

Only on 144-pin version

PC29

PWML0

MCDA5

AD5

Only on 144-pin version

PC30

PWML1

MCDA6

AD6

Only on 144-pin version

PC31

PWML2

MCDA7

AD7

Only on 144-pin version

Notes:

1. Wake-Up source in Backup mode (managed by the SUPC).
2. Fast Start-Up source in Wait mode (managed by the PMC).

47
6430F–ATARM–21-Feb-12

12. Embedded Peripherals Overview
12.1

Serial Peripheral Interface (SPI)
• Supports communication with serial external devices
– Four chip selects with external decoder support allow communication with up to 15
peripherals
– Serial memories, such as DataFlash and 3-wire EEPROMs
– Serial peripherals, such as ADCs, DACs, LCD Controllers, CAN Controllers and
Sensors
– External co-processors
• Master or slave serial peripheral bus interface
– 8- to 16-bit programmable data length per chip select
– Programmable phase and polarity per chip select
– Programmable transfer delays between consecutive transfers and between clock
and data per chip select
– Programmable delay between consecutive transfers
– Selectable mode fault detection
• Very fast transfers supported
– Transfers with baud rates up to MCK
– The chip select line may be left active to speed up transfers on the same device

12.2

Two Wire Interface (TWI)
• Master, Multi-Master and Slave Mode Operation
• Compatibility with Atmel two-wire interface, serial memory and I2C compatible devices
• One, two or three bytes for slave address
• Sequential read/write operations
• Bit Rate: Up to 400 kbit/s
• General Call Supported in Slave Mode
• Connecting to PDC channel capabilities optimizes data transfers in Master Mode only
– One channel for the receiver, one channel for the transmitter
– Next buffer support

12.3

Universal Asynchronous Receiver Transceiver (UART)
• Two-pin UART
– Implemented features are 100% compatible with the standard Atmel USART
– Independent receiver and transmitter with a common programmable Baud Rate
Generator
– Even, Odd, Mark or Space Parity Generation
– Parity, Framing and Overrun Error Detection
– Automatic Echo, Local Loopback and Remote Loopback Channel Modes
– Support for two PDC channels with connection to receiver and transmitter

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
12.4

Universal Synchronous Asynchronous Receiver Transmitter (USART)
• Programmable Baud Rate Generator
• 5- to 9-bit full-duplex synchronous or asynchronous serial communications
– 1, 1.5 or 2 stop bits in Asynchronous Mode or 1 or 2 stop bits in Synchronous Mode
– Parity generation and error detection
– Framing error detection, overrun error detection
– MSB- or LSB-first
– Optional break generation and detection
– By 8 or by-16 over-sampling receiver frequency
– Hardware handshaking RTS-CTS
– Receiver time-out and transmitter timeguard
– Optional Multi-drop Mode with address generation and detection
– Optional Manchester Encoding
• RS485 with driver control signal
• ISO7816, T = 0 or T = 1 Protocols for interfacing with smart cards
– NACK handling, error counter with repetition and iteration limit
• SPI Mode
– Master or Slave
– Serial Clock programmable Phase and Polarity
– SPI Serial Clock (SCK) Frequency up to MCK/6
• IrDA modulation and demodulation
– Communication at up to 115.2 Kbps
• Test Modes
– Remote Loopback, Local Loopback, Automatic Echo

12.5

Serial Synchronous Controller (SSC)
• Provides serial synchronous communication links used in audio and telecom applications
(with CODECs in Master or Slave Modes, I2S, TDM Buses, Magnetic Card Reader, ...)
• Contains an independent receiver and transmitter and a common clock divider
• Offers a configurable frame sync and data length
• Receiver and transmitter can be programmed to start automatically or on detection of
different event on the frame sync signal
• Receiver and transmitter include a data signal, a clock signal and a frame synchronization
signal

12.6

Timer Counter (TC)
• Three 16-bit Timer Counter Channels
• Wide range of functions including:
– Frequency Measurement
– Event Counting
– Interval Measurement

49
6430F–ATARM–21-Feb-12

– Pulse Generation
– Delay Timing
– Pulse Width Modulation
– Up/Down Capabilities
– Quadrature Decoder Logic
• Each channel is user-configurable and contains:
– Three external clock inputs
– Five internal clock inputs
– Two multi-purpose input/output signals
• Two global registers that act on all three TC Channels

12.7

Pulse Width Modulation Controller (PWM)
• 4 channels, one 16-bit counter per channel
• Common clock generator, providing Thirteen Different Clocks
– A Modulo n counter providing eleven clocks
– Two independent Linear Dividers working on modulo n counter outputs
– High Frequency Asynchronous clocking mode
• Independent channel programming
– Independent Enable Disable Commands
– Independent Clock Selection
– Independent Period and Duty Cycle, with Double Buffering
– Programmable selection of the output waveform polarity
– Programmable center or left aligned output waveform
– Independent Output Override for each channel
– Independent complementary Outputs with 12-bit dead time generator for each
channel
– Independent Enable Disable Commands
– Independent Clock Selection
– Independent Period and Duty Cycle, with Double Buffering
• Synchronous Channel mode
– Synchronous Channels share the same counter
– Mode to update the synchronous channels registers after a programmable number
of periods
• Connection to one PDC channel
– Offers Buffer transfer without Processor Intervention, to update duty cycle of
synchronous channels
• Two independent event lines which can send up to 8 triggers on ADC within a period
• Four programmable Fault Inputs providing asynchronous protection of outputs

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
12.8

High Speed Multimedia Card Interface (HSMCI)
• Compatibility with MultiMedia Card Specification Version 4.3
• Compatibility with SD Memory Card Specification Version 2.0
• Compatibility with SDIO Specification Version V2.0.
• Compatibility with CE-ATA Specification 1.1
• Cards clock rate up to Master Clock divided by 2
• Boot Operation Mode support
• High Speed mode support
• Embedded power management to slow down clock rate when not used
• HSMCI has one slot supporting
– One MultiMediaCard bus (up to 30 cards) or
– One SD Memory Card
– One SDIO Card
• Support for stream, block and multi-block data read and write
• Supports Connection to DMA controller
– Minimizes Processor intervention for large buffer transfers
• Built in FIFO (32 bytes) with large Memory Aperture Supporting Incremental access
• Support for CE-ATA Completion Signal Disable Command

12.9

USB High Speed Device Port (UDPHS)
• USB V2.0 high-speed compliant, 480 MBits per second
• Embedded USB V2.0 UTMI+ high-speed transceiver
• Embedded 4-Kbyte dual-port RAM for endpoints
• Embedded 6 channels DMA controller
• Suspend/Resume logic
• Up to 2 or 3 banks for isochronous and bulk endpoints
• Seven endpoints, configurable by software
• Maximum configuration: seven endpoints:
– Endpoint 0: 64 bytes, 1 bank mode
– Endpoint 1 & 2: 512 bytes, 2 banks mode, HS isochronous capable
– Endpoint 3 & 4:64 bytes, 3 banks mode
– Endpoint 5 & 6: 1024 bytes, 3 banks mode, HS isochronous capable

12.10 Analog-to-Digital Converter (ADC)
Two ADCs are embedded in the product.
12.10.1

12-bit High Speed ADC
• 8-channel ADC
• 12-bit 1 Msamples/sec. Cyclic Pipeline ADC
• Integrated 8-to-1 multiplexer
• 12-bit resolution

51
6430F–ATARM–21-Feb-12

• Selectable single ended or differential input voltage
• Programmable gain for maximum full scale input range
• External voltage reference for better accuracy on low voltage inputs
• Individual enable and disable of each channel
• Multiple trigger sources
– Hardware or software trigger
– External trigger pin
– Timer Counter 0 to 2 outputs TIOA0 to TIOA2 trigger
– PWM trigger
• Sleep Mode and conversion sequencer
– Automatic wakeup on trigger and back to sleep mode after conversions of all
enabled channels
12.10.2

10-bit Low Power ADC
• 8-channel ADC
• 10-bit 384 Ksamples/sec. or 8-bit 533 Ksamples/sec. Successive Approximation Register
ADC
• -2/+2 LSB Integral Non Linearity, -1/+1 LSB Differential Non Linearity
• Integrated 8-to-1 multiplexer
• External voltage reference for better accuracy on low voltage inputs
• Individual enable and disable of each channel
• Multiple trigger sources
– Hardware or software trigger
– External trigger pin
– Timer Counter 0 to 2 outputs TIOA0 to TIOA2 trigger
– PWM trigger
• Sleep Mode and conversion sequencer
– Automatic wakeup on trigger and back to sleep mode after conversions of all
enabled channels

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13. ARM Cortex® M3 Processor
13.1

About this section
This section provides the information required for application and system-level software development. It does not provide information on debug components, features, or operation.
This material is for microcontroller software and hardware engineers, including those who have
no experience of ARM products.
Note: The information in this section is reproduced from source material provided to Atmel by
ARM Ltd. in terms of Atmel’s license for the ARM Cortex™-M3 processor core. This information
is copyright ARM Ltd., 2008 - 2009.

13.2

About the Cortex-M3 processor and core peripherals
• The Cortex-M3 processor is a high performance 32-bit processor designed for the
microcontroller market. It offers significant benefits to developers, including:
• outstanding processing performance combined with fast interrupt handling
• enhanced system debug with extensive breakpoint and trace capabilities
• efficient processor core, system and memories
• ultra-low power consumption with integrated sleep modes
• platform security, with integrated memory protection unit (MPU).
Figure 13-1. Typical Cortex-M3 implementation
Cortex-M3
Processor

NVIC

Debug
Access
Port

Processor
Core

Memory
Protection Unit

Flash
Patch

Serial
Wire
Viewer

Data
Watchpoints

Bus Matrix
Code
Interface

SRAM and
Peripheral Interface

The Cortex-M3 processor is built on a high-performance processor core, with a 3-stage pipeline
Harvard architecture, making it ideal for demanding embedded applications. The processor
delivers exceptional power efficiency through an efficient instruction set and extensively opti-

53
6430F–ATARM–21-Feb-12

mized design, providing high-end processing hardware including single-cycle 32x32
multiplication and dedicated hardware division.
To facilitate the design of cost-sensitive devices, the Cortex-M3 processor implements tightlycoupled system components that reduce processor area while significantly improving interrupt
handling and system debug capabilities. The Cortex-M3 processor implements a version of the
Thumb® instruction set, ensuring high code density and reduced program memory requirements.
The Cortex-M3 instruction set provides the exceptional performance expected of a modern 32bit architecture, with the high code density of 8-bit and 16-bit microcontrollers.
The Cortex-M3 processor closely integrates a configurable nested interrupt controller (NVIC), to
deliver industry-leading interrupt performance. The NVIC provides up to 16 interrupt priority levels. The tight integration of the processor core and NVIC provides fast execution of interrupt
service routines (ISRs), dramatically reducing the interrupt latency. This is achieved through the
hardware stacking of registers, and the ability to suspend load-multiple and store-multiple operations. Interrupt handlers do not require any assembler stubs, removing any code overhead from
the ISRs. Tail-chaining optimization also significantly reduces the overhead when switching from
one ISR to another.
To optimize low-power designs, the NVIC integrates with the sleep modes, that include a deep
sleep function that enables the entire device to be rapidly powered down.
13.2.1

System level interface
The Cortex-M3 processor provides multiple interfaces using AMBA® technology to provide high
speed, low latency memory accesses. It supports unaligned data accesses and implements
atomic bit manipulation that enables faster peripheral controls, system spinlocks and thread-safe
Boolean data handling.
The Cortex-M3 processor has a memory protection unit (MPU) that provides fine grain memory
control, enabling applications to implement security privilege levels, separating code, data and
stack on a task-by-task basis. Such requirements are becoming critical in many embedded
applications.

13.2.2

Integrated configurable debug
The Cortex-M3 processor implements a complete hardware debug solution. This provides high
system visibility of the processor and memory through either a traditional JTAG port or a 2-pin
Serial Wire Debug (SWD) port that is ideal for microcontrollers and other small package devices.
For system trace the processor integrates an Instrumentation Trace Macrocell (ITM) alongside
data watchpoints and a profiling unit. To enable simple and cost-effective profiling of the system
events these generate, a Serial Wire Viewer (SWV) can export a stream of software-generated
messages, data trace, and profiling information through a single pin.

13.2.3

Cortex-M3 processor features and benefits summary
• tight integration of system peripherals reduces area and development costs
• Thumb instruction set combines high code density with 32-bit performance
• code-patch ability for ROM system updates
• power control optimization of system components
• integrated sleep modes for low power consumption
• fast code execution permits slower processor clock or increases sleep mode time
• hardware division and fast multiplier

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• deterministic, high-performance interrupt handling for time-critical applications
• memory protection unit (MPU) for safety-critical applications
• extensive debug and trace capabilities:
– Serial Wire Debug and Serial Wire Trace reduce the number of pins required for
debugging and tracing.
13.2.4

Cortex-M3 core peripherals
These are:

13.2.4.1

Nested Vectored Interrupt Controller
The Nested Vectored Interrupt Controller (NVIC) is an embedded interrupt controller that supports low latency interrupt processing.

13.2.4.2

System control block
The System control block (SCB) is the programmers model interface to the processor. It provides system implementation information and system control, including configuration, control,
and reporting of system exceptions.

13.2.4.3

System timer
The system timer, SysTick, is a 24-bit count-down timer. Use this as a Real Time Operating System (RTOS) tick timer or as a simple counter.

13.2.4.4

Memory protection unit
The Memory protection unit (MPU) improves system reliability by defining the memory attributes
for different memory regions. It provides up to eight different regions, and an optional predefined
background region.

13.3

Programmers model
This section describes the Cortex-M3 programmers model. In addition to the individual core register descriptions, it contains information about the processor modes and privilege levels for
software execution and stacks.

13.3.1

Processor mode and privilege levels for software execution
The processor modes are:

13.3.1.1

Thread mode
Used to execute application software. The processor enters Thread mode when it comes out of
reset.

13.3.1.2

Handler mode
Used to handle exceptions. The processor returns to Thread mode when it has finished exception processing.
The privilege levels for software execution are:

13.3.1.3

Unprivileged
The software:
• has limited access to the MSR and MRS instructions, and cannot use the CPS instruction

55
6430F–ATARM–21-Feb-12

• cannot access the system timer, NVIC, or system control block
• might have restricted access to memory or peripherals.
Unprivileged software executes at the unprivileged level.
13.3.1.4

Privileged
The software can use all the instructions and has access to all resources.
Privileged software executes at the privileged level.
In Thread mode, the CONTROL register controls whether software execution is privileged or
unprivileged, see “CONTROL register” on page 65. In Handler mode, software execution is
always privileged.
Only privileged software can write to the CONTROL register to change the privilege level for
software execution in Thread mode. Unprivileged software can use the SVC instruction to make
a supervisor call to transfer control to privileged software.

13.3.2

Stacks
The processor uses a full descending stack. This means the stack pointer indicates the last
stacked item on the stack memory. When the processor pushes a new item onto the stack, it
decrements the stack pointer and then writes the item to the new memory location. The processor implements two stacks, the main stack and the process stack, with independent copies of
the stack pointer, see “Stack Pointer” on page 58.
In Thread mode, the CONTROL register controls whether the processor uses the main stack or
the process stack, see “CONTROL register” on page 65. In Handler mode, the processor always
uses the main stack. The options for processor operations are:
Table 13-1.
Processor
mode

Used to
execute

Privilege level for
software execution

Stack used

Thread

Applications

Privileged or
unprivileged (1)

Main stack or process
stack(1)

Handler

Exception
handlers

Always privileged

Main stack

Summary of processor mode, execution privilege level, and stack use options

See “CONTROL register” on page 65.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.3.3

Core registers
The processor core registers are:
R0
R1
R2
R3

Low registers

R4
R5
R6

General-purpose registers

R7
R8
R9
High registers

R10
R11
R12

Stack Pointer

SP (R13)

Link Register

LR (R14)

Program Counter

PC (R15)
PSR

PSP‡

MSP‡

‡

Banked version of SP

Program status register

PRIMASK
FAULTMASK

Exception mask registers

Special registers

BASEPRI
CONTROL

Table 13-2.

CONTROL register

Core register set summary
Type

Required
privilege

Name

(1)

(2)

Reset
value

Description

R0-R12

Either

Unknown

“General-purpose registers” on page 58

MSP

Privileged

See
description

“Stack Pointer” on page 58

PSP

Either

Unknown

“Stack Pointer” on page 58

Either

0xFFFFFFFF

“Link Register” on page 58

Either

See
description

“Program Counter” on page 58

PSR

Privileged

0x01000000

“Program Status Register” on page 59

ASPR

Either

0x00000000

“Application Program Status Register” on
page 60

IPSR

Privileged

0x00000000

“Interrupt Program Status Register” on page
61

EPSR

Privileged

0x01000000

“Execution Program Status Register” on page
62

PRIMASK

Privileged

0x00000000

“Priority Mask Register” on page 63

57
6430F–ATARM–21-Feb-12

Table 13-2.

Core register set summary (Continued)
Type

Required
privilege

Name

(1)

(2)

Reset
value

Description

FAULTMASK

Privileged

0x00000000

“Fault Mask Register” on page 63

BASEPRI

Privileged

0x00000000

“Base Priority Mask Register” on page 64

CONTROL

Privileged

0x00000000

“CONTROL register” on page 65

Describes access type during program execution in thread mode and Handler mode. Debug
access can differ.

An entry of Either means privileged and unprivileged software can access the register.

13.3.3.1

General-purpose registers
R0-R12 are 32-bit general-purpose registers for data operations.

13.3.3.2

Stack Pointer
The Stack Pointer (SP) is register R13. In Thread mode, bit[1] of the CONTROL register indicates the stack pointer to use:
• 0 = Main Stack Pointer (MSP). This is the reset value.
• 1 = Process Stack Pointer (PSP).
On reset, the processor loads the MSP with the value from address 0x00000000.

13.3.3.3

Link Register
The Link Register (LR) is register R14. It stores the return information for subroutines, function
calls, and exceptions. On reset, the processor loads the LR value 0xFFFFFFFF.

13.3.3.4

Program Counter
The Program Counter (PC) is register R15. It contains the current program address. Bit[0] is
always 0 because instruction fetches must be halfword aligned. On reset, the processor loads
the PC with the value of the reset vector, which is at address 0x00000004.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.3.3.5

Program Status Register
The Program Status Register (PSR) combines:
• Application Program Status Register (APSR)
• Interrupt Program Status Register (IPSR)
• Execution Program Status Register (EPSR).
These registers are mutually exclusive bitfields in the 32-bit PSR. The bit assignments are:

• APSR:
31

Reserved

Reserved
15

Reserved
7

Reserved

• IPSR:
31

Reserved
23

Reserved
15

Reserved
7

ISR_NUMBER

• EPSR:
31

Reserved
23

ICI/IT

Reserved
15

ICI/IT
7

Reserved
4

Reserved

59
6430F–ATARM–21-Feb-12

The PSR bit assignments are:
31

ICI/IT

T
17

Reserved
15

ICI/IT
7

Reserved

ISR_NUMBER

Access these registers individually or as a combination of any two or all three registers, using
the register name as an argument to the MSR or MRS instructions. For example:
• read all of the registers using PSR with the MRS instruction
• write to the APSR using APSR with the MSR instruction.
The PSR combinations and attributes are:
Table 13-3.

PSR register combinations

Type

PSR

RW (1),

IEPSR

IAPSR

Combination
(2)

EPSR and IPSR
(1)

APSR and IPSR

(2)

APSR and EPSR

EAPSR

APSR, EPSR, and IPSR

The processor ignores writes to the IPSR bits.

Reads of the EPSR bits return zero, and the processor ignores writes to the these bits.

See the instruction descriptions “MRS” on page 157 and “MSR” on page 158 for more information about how to access the program status registers.
13.3.3.6

Application Program Status Register
The APSR contains the current state of the condition flags from previous instruction executions.
See the register summary in Table 13-2 on page 57 for its attributes. The bit assignments are:

• N
Negative or less than flag:
0 = operation result was positive, zero, greater than, or equal
1 = operation result was negative or less than.
• Z
Zero flag:
0 = operation result was not zero
1 = operation result was zero.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• C
Carry or borrow flag:
0 = add operation did not result in a carry bit or subtract operation resulted in a borrow bit
1 = add operation resulted in a carry bit or subtract operation did not result in a borrow bit.
• V
Overflow flag:
0 = operation did not result in an overflow
1 = operation resulted in an overflow.
• Q
Sticky saturation flag:
0 = indicates that saturation has not occurred since reset or since the bit was last cleared to zero
1 = indicates when an SSAT or USAT instruction results in saturation.
This bit is cleared to zero by software using an MRS instruction.
13.3.3.7

Interrupt Program Status Register
The IPSR contains the exception type number of the current Interrupt Service Routine (ISR).
See the register summary in Table 13-2 on page 57 for its attributes. The bit assignments are:

• ISR_NUMBER
This is the number of the current exception:
0 = Thread mode
1 = Reserved
2 = NMI
3 = Hard fault
4 = Memory management fault
5 = Bus fault
6 = Usage fault
7-10 = Reserved
11 = SVCall
12 = Reserved for Debug
13 = Reserved
14 = PendSV
15 = SysTick
16 = IRQ0
45 = IRQ29
see “Exception types” on page 77 for more information.

61
6430F–ATARM–21-Feb-12

13.3.3.8

Execution Program Status Register
The EPSR contains the Thumb state bit, and the execution state bits for either the:
• If-Then (IT) instruction
• Interruptible-Continuable Instruction (ICI) field for an interrupted load multiple or store
multiple instruction.
See the register summary in Table 13-2 on page 57 for the EPSR attributes. The bit assignments are:

• ICI
Interruptible-continuable instruction bits, see “Interruptible-continuable instructions” on page 62.
• IT
Indicates the execution state bits of the IT instruction, see “IT” on page 147.
• T
Always set to 1.
Attempts to read the EPSR directly through application software using the MSR instruction
always return zero. Attempts to write the EPSR using the MSR instruction in application software
are ignored. Fault handlers can examine EPSR value in the stacked PSR to indicate the operation that is at fault. See “Exception entry and return” on page 82
13.3.3.9

Interruptible-continuable instructions
When an interrupt occurs during the execution of an LDM or STM instruction, the processor:
• stops the load multiple or store multiple instruction operation temporarily
• stores the next register operand in the multiple operation to EPSR bits[15:12].
After servicing the interrupt, the processor:
• returns to the register pointed to by bits[15:12]
• resumes execution of the multiple load or store instruction.
When the EPSR holds ICI execution state, bits[26:25,11:10] are zero.

13.3.3.10

If-Then block
The If-Then block contains up to four instructions following a 16-bit IT instruction. Each instruction in the block is conditional. The conditions for the instructions are either all the same, or
some can be the inverse of others. See “IT” on page 147 for more information.

13.3.3.11

Exception mask registers
The exception mask registers disable the handling of exceptions by the processor. Disable
exceptions where they might impact on timing critical tasks.
To access the exception mask registers use the MSR and MRS instructions, or the CPS instruction to change the value of PRIMASK or FAULTMASK. See “MRS” on page 157, “MSR” on page
158, and “CPS” on page 153 for more information.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.3.3.12

Priority Mask Register
The PRIMASK register prevents activation of all exceptions with configurable priority. See the
register summary in Table 13-2 on page 57 for its attributes. The bit assignments are:

Reserved
23

Reserved
15

Reserved
7

Reserved

PRIMASK

• PRIMASK
0 = no effect
1 = prevents the activation of all exceptions with configurable priority.
13.3.3.13

Fault Mask Register
The FAULTMASK register prevents activation of all exceptions. See the register summary in
Table 13-2 on page 57 for its attributes. The bit assignments are:

Reserved
23

Reserved
15

Reserved
7

Reserved

FAULTMASK

• FAULTMASK
0 = no effect
1 = prevents the activation of all exceptions.
The processor clears the FAULTMASK bit to 0 on exit from any exception handler except the NMI handler.

63
6430F–ATARM–21-Feb-12

13.3.3.14

Base Priority Mask Register
The BASEPRI register defines the minimum priority for exception processing. When BASEPRI is
set to a nonzero value, it prevents the activation of all exceptions with same or lower priority
level as the BASEPRI value. See the register summary in Table 13-2 on page 57 for its attributes. The bit assignments are:

Reserved
23

Reserved
15

Reserved
7

BASEPRI

• BASEPRI
Priority mask bits:
0x0000 = no effect
Nonzero = defines the base priority for exception processing.
The processor does not process any exception with a priority value greater than or equal to BASEPRI.
This field is similar to the priority fields in the interrupt priority registers. The processor implements only bits[7:4] of this field,
bits[3:0] read as zero and ignore writes. See “Interrupt Priority Registers” on page 172 for more information. Remember
that higher priority field values correspond to lower exception priorities.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.3.3.15

CONTROL register
The CONTROL register controls the stack used and the privilege level for software execution
when the processor is in Thread mode. See the register summary in Table 13-2 on page 57 for
its attributes. The bit assignments are:

Active Stack
Pointer

Thread Mode
Privilege
Level

Reserved
23

Reserved
15

Reserved
7

Reserved

• Active stack pointer
Defines the current stack:
0 = MSP is the current stack pointer
1 = PSP is the current stack pointer.
In Handler mode this bit reads as zero and ignores writes.
• Thread mode privilege level
Defines the Thread mode privilege level:
0 = privileged
1 = unprivileged.
Handler mode always uses the MSP, so the processor ignores explicit writes to the active stack pointer bit of the CONTROL register when in Handler mode. The exception entry and return mechanisms update the CONTROL register.
In an OS environment, ARM recommends that threads running in Thread mode use the process stack and the kernel and
exception handlers use the main stack.
By default, Thread mode uses the MSP. To switch the stack pointer used in Thread mode to the PSP, use the MSR instruction to set the Active stack pointer bit to 1, see “MSR” on page 158.
When changing the stack pointer, software must use an ISB instruction immediately after the MSR instruction. This
ensures that instructions after the ISB execute using the new stack pointer. See “ISB” on page 156

65
6430F–ATARM–21-Feb-12

13.3.4

Exceptions and interrupts
The Cortex-M3 processor supports interrupts and system exceptions. The processor and the
Nested Vectored Interrupt Controller (NVIC) prioritize and handle all exceptions. An exception
changes the normal flow of software control. The processor uses handler mode to handle all
exceptions except for reset. See “Exception entry” on page 83 and “Exception return” on page
84 for more information.
The NVIC registers control interrupt handling. See “Nested Vectored Interrupt Controller” on
page 165 for more information.

13.3.5

Data types
The processor:
• supports the following data types:
– 32-bit words
– 16-bit halfwords
– 8-bit bytes
• supports 64-bit data transfer instructions.
• manages all data memory accesses as little-endian. Instruction memory and Private
Peripheral Bus (PPB) accesses are always little-endian. See “Memory regions, types and
attributes” on page 68 for more information.

13.3.6

The Cortex Microcontroller Software Interface Standard
For a Cortex-M3 microcontroller system, the Cortex Microcontroller Software Interface Standard
(CMSIS) defines:
• a common way to:
– access peripheral registers
– define exception vectors
• the names of:
– the registers of the core peripherals
– the core exception vectors
• a device-independent interface for RTOS kernels, including a debug channel.
The CMSIS includes address definitions and data structures for the core peripherals in the Cortex-M3 processor. It also includes optional interfaces for middleware components comprising a
TCP/IP stack and a Flash file system.
CMSIS simplifies software development by enabling the reuse of template code and the combination of CMSIS-compliant software components from various middleware vendors. Software
vendors can expand the CMSIS to include their peripheral definitions and access functions for
those peripherals.
This document includes the register names defined by the CMSIS, and gives short descriptions
of the CMSIS functions that address the processor core and the core peripherals.
This document uses the register short names defined by the CMSIS. In a few cases these differ
from the architectural short names that might be used in other documents.
The following sections give more information about the CMSIS:
• “Power management programming hints” on page 88

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• “Intrinsic functions” on page 92
• “The CMSIS mapping of the Cortex-M3 NVIC registers” on page 165
• “NVIC programming hints” on page 177.

67
6430F–ATARM–21-Feb-12

13.4

Memory model
This section describes the processor memory map, the behavior of memory accesses, and the
bit-banding features. The processor has a fixed memory map that provides up to 4GB of
addressable memory. The memory map is:
0xFFFFFFFF
Vendor-specific
memory

511MB

0xE0100000
0xE00FFFFF
Private peripheral
1.0MB
bus
0xE0000000
0xDFFFFFFF

External device

1.0GB

0xA0000000
0x9FFFFFFF

External RAM

0x43FFFFFF

1.0GB

32MB Bit band alias
0x60000000
0x5FFFFFFF

0x42000000
0x400FFFFF
1MB Bit band region
0x40000000

Peripheral

0.5GB
0x40000000
0x3FFFFFFF

0x23FFFFFF
32MB Bit band alias

SRAM

0.5GB
0x20000000
0x1FFFFFFF

0x22000000
0x200FFFFF
1MB Bit band region
0x20000000

Code

0.5GB
0x00000000

The regions for SRAM and peripherals include bit-band regions. Bit-banding provides atomic
operations to bit data, see “Bit-banding” on page 72.
The processor reserves regions of the Private peripheral bus (PPB) address range for core
peripheral registers, see “About the Cortex-M3 peripherals” on page 164.
This memory mapping is generic to ARM Cortex-M3 products. To get the specific memory mapping of this product, refer to the Memories section of the datasheet.
13.4.1

Memory regions, types and attributes
The memory map and the programming of the MPU split the memory map into regions. Each
region has a defined memory type, and some regions have additional memory attributes. The
memory type and attributes determine the behavior of accesses to the region.
The memory types are:

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.4.1.1

Normal
The processor can re-order transactions for efficiency, or perform speculative reads.

13.4.1.2

Device
The processor preserves transaction order relative to other transactions to Device or Stronglyordered memory.

13.4.1.3

Strongly-ordered
The processor preserves transaction order relative to all other transactions.
The different ordering requirements for Device and Strongly-ordered memory mean that the
memory system can buffer a write to Device memory, but must not buffer a write to Stronglyordered memory.
The additional memory attributes include.

13.4.1.4

Shareable
For a shareable memory region, the memory system provides data synchronization between
bus masters in a system with multiple bus masters, for example, a processor with a DMA
controller.
Strongly-ordered memory is always shareable.
If multiple bus masters can access a non-shareable memory region, software must ensure data
coherency between the bus masters.

13.4.1.5

13.4.2

Execute Never (XN)
Means the processor prevents instruction accesses. Any attempt to fetch an instruction from an
XN region causes a memory management fault exception.
Memory system ordering of memory accesses
For most memory accesses caused by explicit memory access instructions, the memory system
does not guarantee that the order in which the accesses complete matches the program order of
the instructions, providing this does not affect the behavior of the instruction sequence. Normally, if correct program execution depends on two memory accesses completing in program
order, software must insert a memory barrier instruction between the memory access instructions, see “Software ordering of memory accesses” on page 71.
However, the memory system does guarantee some ordering of accesses to Device and
Strongly-ordered memory. For two memory access instructions A1 and A2, if A1 occurs before
A2 in program order, the ordering of the memory accesses caused by two instructions is:
Normal
access

Non-shareable

Shareable

Stronglyordered
access

Normal access

Device access, non-shareable

Device access, shareable

Strongly-ordered access

A2
A1

Device access

Where:
- Means that the memory system does not guarantee the ordering of the accesses.

69
6430F–ATARM–21-Feb-12

< Means that accesses are observed in program order, that is, A1 is always observed before A2.
13.4.3

Behavior of memory accesses
The behavior of accesses to each region in the memory map is:
Table 13-4.

Memory access behavior

Address
range

Memory
region

Memory
type

Description

0x000000000x1FFFFFFF

Code

Normal (1)

Executable region for program code. You can also put
data here.

Executable region for data. You can also put code
here.
This region includes bit band and bit band alias areas,
see Table 13-6 on page 72.

0x200000000x3FFFFFFF

SRAM

Normal

0x400000000x5FFFFFFF

Peripheral

Device (1)

This region includes bit band and bit band alias areas,
see Table 13-6 on page 72.

0x600000000x9FFFFFFF

External
RAM

Normal (1)

Executable region for data.

0xA00000000xDFFFFFFF

External
device

Device (1)

External Device memory

0xE00000000xE00FFFFF

Private
Peripheral
Bus

Stronglyordered (1)

This region includes the NVIC, System timer, and
system control block.

0xE01000000xFFFFFFFF

Reserved

Device (1)

Reserved

(1)

See “Memory regions, types and attributes” on page 68 for more information.

The Code, SRAM, and external RAM regions can hold programs. However, ARM recommends
that programs always use the Code region. This is because the processor has separate buses
that enable instruction fetches and data accesses to occur simultaneously.
The MPU can override the default memory access behavior described in this section. For more
information, see “Memory protection unit” on page 210.
13.4.3.1

Additional memory access constraints for shared memory
When a system includes shared memory, some memory regions have additional access constraints, and some regions are subdivided, as Table 13-5 shows:
Table 13-5.

Memory region share ability policies

Address range

Memory region

Memory type

Shareability

0x000000000x1FFFFFFF

Code

Normal (1)

0x200000000x3FFFFFFF

SRAM

Normal (1)

0x400000000x5FFFFFFF

Peripheral (2)

Device (1)

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 13-5.

Memory region share ability policies (Continued)

Address range

Memory region

0x600000000x7FFFFFFF

Memory type

Shareability
WBWA (2)

External RAM

Normal (1)

0x800000000x9FFFFFFF

WT (2)

0xA00000000xBFFFFFFF

Shareable (1)
External device

Device (1)

0xC00000000xDFFFFFFF

13.4.4

Nonshareable (1)

0xE00000000xE00FFFFF

Private Peripheral
Bus

Stronglyordered(1)

Shareable (1)

0xE01000000xFFFFFFFF

Vendor-specific
device(2)

Device (1)

See “Memory regions, types and attributes” on page 68 for more information.

The Peripheral and Vendor-specific device regions have no additional access constraints.

Software ordering of memory accesses
The order of instructions in the program flow does not always guarantee the order of the corresponding memory transactions. This is because:
• the processor can reorder some memory accesses to improve efficiency, providing this does
not affect the behavior of the instruction sequence.
• the processor has multiple bus interfaces
• memory or devices in the memory map have different wait states
• some memory accesses are buffered or speculative.
“Memory system ordering of memory accesses” on page 69 describes the cases where the
memory system guarantees the order of memory accesses. Otherwise, if the order of memory
accesses is critical, software must include memory barrier instructions to force that ordering. The
processor provides the following memory barrier instructions:

13.4.4.1

DMB
The Data Memory Barrier (DMB) instruction ensures that outstanding memory transactions complete before subsequent memory transactions. See “DMB” on page 154.

13.4.4.2

DSB
The Data Synchronization Barrier (DSB) instruction ensures that outstanding memory transactions complete before subsequent instructions execute. See “DSB” on page 155.

13.4.4.3

ISB
The Instruction Synchronization Barrier (ISB) ensures that the effect of all completed memory
transactions is recognizable by subsequent instructions. See “ISB” on page 156.
Use memory barrier instructions in, for example:
• MPU programming:
– Use a DSB instruction to ensure the effect of the MPU takes place immediately at
the end of context switching.

71
6430F–ATARM–21-Feb-12

– Use an ISB instruction to ensure the new MPU setting takes effect immediately after
programming the MPU region or regions, if the MPU configuration code was
accessed using a branch or call. If the MPU configuration code is entered using
exception mechanisms, then an ISB instruction is not required.
• Vector table. If the program changes an entry in the vector table, and then enables the
corresponding exception, use a DMB instruction between the operations. This ensures that if
the exception is taken immediately after being enabled the processor uses the new exception
vector.
• Self-modifying code. If a program contains self-modifying code, use an ISB instruction
immediately after the code modification in the program. This ensures subsequent instruction
execution uses the updated program.
• Memory map switching. If the system contains a memory map switching mechanism, use a
DSB instruction after switching the memory map in the program. This ensures subsequent
instruction execution uses the updated memory map.
• Dynamic exception priority change. When an exception priority has to change when the
exception is pending or active, use DSB instructions after the change. This ensures the
change takes effect on completion of the DSB instruction.
• Using a semaphore in multi-master system. If the system contains more than one bus
master, for example, if another processor is present in the system, each processor must use
a DMB instruction after any semaphore instructions, to ensure other bus masters see the
memory transactions in the order in which they were executed.
Memory accesses to Strongly-ordered memory, such as the system control block, do not require
the use of DMB instructions.
13.4.5

Bit-banding
A bit-band region maps each word in a bit-band alias region to a single bit in the bit-band region.
The bit-band regions occupy the lowest 1MB of the SRAM and peripheral memory regions.
The memory map has two 32MB alias regions that map to two 1MB bit-band regions:
• accesses to the 32MB SRAM alias region map to the 1MB SRAM bit-band region, as shown
in Table 13-6
• accesses to the 32MB peripheral alias region map to the 1MB peripheral bit-band region, as
shown in Table 13-7.
Table 13-6.

SRAM memory bit-banding regions

Address
range

Memory
region

0x200000000x200FFFFF

SRAM bit-band
region

Direct accesses to this memory range behave as SRAM
memory accesses, but this region is also bit addressable
through bit-band alias.

0x220000000x23FFFFFF

SRAM bit-band alias

Data accesses to this region are remapped to bit band
region. A write operation is performed as read-modify-write.
Instruction accesses are not remapped.

Instruction and data accesses

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 13-7.

Peripheral memory bit-banding regions

Address
range

Memory
region

0x400000000x400FFFFF

Peripheral bit-band
alias

Direct accesses to this memory range behave as peripheral
memory accesses, but this region is also bit addressable
through bit-band alias.

0x420000000x43FFFFFF

Peripheral bit-band
region

Data accesses to this region are remapped to bit band
region. A write operation is performed as read-modify-write.
Instruction accesses are not permitted.

Instruction and data accesses

A word access to the SRAM or peripheral bit-band alias regions map to a single bit in the SRAM
or peripheral bit-band region.
The following formula shows how the alias region maps onto the bit-band region:
bit_word_offset = (byte_offset x 32) + (bit_number x 4)
bit_word_addr = bit_band_base + bit_word_offset

where:
• Bit_word_offset is the position of the target bit in the bit-band memory region.
• Bit_word_addr is the address of the word in the alias memory region that maps to the
targeted bit.
• Bit_band_base is the starting address of the alias region.
• Byte_offset is the number of the byte in the bit-band region that contains the targeted bit.
• Bit_number is the bit position, 0-7, of the targeted bit.
Figure 13-2 shows examples of bit-band mapping between the SRAM bit-band alias region and
the SRAM bit-band region:
• The alias word at 0x23FFFFE0 maps to bit[0] of the bit-band byte at 0x200FFFFF: 0x23FFFFE0 =
0x22000000 + (0xFFFFF*32) + (0*4).
• The alias word at 0x23FFFFFC maps to bit[7] of the bit-band byte at 0x200FFFFF: 0x23FFFFFC =
0x22000000 + (0xFFFFF*32) + (7*4).
• The alias word at 0x22000000 maps to bit[0] of the bit-band byte at 0x20000000: 0x22000000 =
0x22000000 + (0*32) + (0 *4).
• The alias word at 0x2200001C maps to bit[7] of the bit-band byte at 0x20000000: 0x2200001C =
0x22000000+ (0*32) + (7*4).

73
6430F–ATARM–21-Feb-12

Figure 13-2. Bit-band mapping
32MB alias region
0x23FFFFFC

0x23FFFFF8

0x23FFFFF4

0x23FFFFF0

0x23FFFFEC

0x23FFFFE8

0x23FFFFE4

0x23FFFFE0

0x2200001C

0x22000018

0x22000014

0x22000010

0x2200000C

0x22000008

0x22000004

0x22000000

1MB SRAM bit-band region
7

0x200FFFFF

0x20000003

13.4.5.1

0x200FFFFE

0x20000002

0x200FFFFD

0x20000001

0x200FFFFC

0x20000000

Directly accessing an alias region
Writing to a word in the alias region updates a single bit in the bit-band region.
Bit[0] of the value written to a word in the alias region determines the value written to the targeted bit in the bit-band region. Writing a value with bit[0] set to 1 writes a 1 to the bit-band bit,
and writing a value with bit[0] set to 0 writes a 0 to the bit-band bit.
Bits[31:1] of the alias word have no effect on the bit-band bit. Writing 0x01 has the same effect as
writing 0xFF. Writing 0x00 has the same effect as writing 0x0E.
Reading a word in the alias region:
• 0x00000000 indicates that the targeted bit in the bit-band region is set to zero
• 0x00000001 indicates that the targeted bit in the bit-band region is set to 1

13.4.5.2

13.4.6

Directly accessing a bit-band region
“Behavior of memory accesses” on page 70 describes the behavior of direct byte, halfword, or
word accesses to the bit-band regions.
Memory endianness
The processor views memory as a linear collection of bytes numbered in ascending order from
zero. For example, bytes 0-3 hold the first stored word, and bytes 4-7 hold the second stored
word. or “Little-endian format” describes how words of data are stored in memory.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.4.6.1

Little-endian format
In little-endian format, the processor stores the least significant byte of a word at the lowestnumbered byte, and the most significant byte at the highest-numbered byte. For example:
Memory
7

0
31

13.4.7

Address A

A+1

A+2

A+3

lsbyte

24 23
B3

16 15
B2

8 7
B1

0
B0

msbyte

Synchronization primitives
The Cortex-M3 instruction set includes pairs of synchronization primitives. These provide a nonblocking mechanism that a thread or process can use to obtain exclusive access to a memory
location. Software can use them to perform a guaranteed read-modify-write memory update
sequence, or for a semaphore mechanism.
A pair of synchronization primitives comprises:

13.4.7.1

A Load-Exclusive instruction
Used to read the value of a memory location, requesting exclusive access to that location.

13.4.7.2

A Store-Exclusive instruction
Used to attempt to write to the same memory location, returning a status bit to a register. If this
bit is:
0: it indicates that the thread or process gained exclusive access to the memory, and the write
succeeds,
1: it indicates that the thread or process did not gain exclusive access to the memory, and no
write is performed,
The pairs of Load-Exclusive and Store-Exclusive instructions are:
• the word instructions LDREX and STREX
• the halfword instructions LDREXH and STREXH
• the byte instructions LDREXB and STREXB.
Software must use a Load-Exclusive instruction with the corresponding Store-Exclusive
instruction.
To perform a guaranteed read-modify-write of a memory location, software must:
• Use a Load-Exclusive instruction to read the value of the location.
• Update the value, as required.
• Use a Store-Exclusive instruction to attempt to write the new value back to the memory
location, and tests the returned status bit. If this bit is:
0: The read-modify-write completed successfully,

75
6430F–ATARM–21-Feb-12

1: No write was performed. This indicates that the value returned the first step might be out
of date. The software must retry the read-modify-write sequence,
Software can use the synchronization primitives to implement a semaphores as follows:
• Use a Load-Exclusive instruction to read from the semaphore address to check whether the
semaphore is free.
• If the semaphore is free, use a Store-Exclusive to write the claim value to the semaphore
address.
• If the returned status bit from the second step indicates that the Store-Exclusive succeeded
then the software has claimed the semaphore. However, if the Store-Exclusive failed, another
process might have claimed the semaphore after the software performed the first step.
The Cortex-M3 includes an exclusive access monitor, that tags the fact that the processor has
executed a Load-Exclusive instruction. If the processor is part of a multiprocessor system, the
system also globally tags the memory locations addressed by exclusive accesses by each
processor.
The processor removes its exclusive access tag if:
• It executes a CLREX instruction
• It executes a Store-Exclusive instruction, regardless of whether the write succeeds.
• An exception occurs. This means the processor can resolve semaphore conflicts between
different threads.
In a multiprocessor implementation:
• executing a CLREX instruction removes only the local exclusive access tag for the processor
• executing a Store-Exclusive instruction, or an exception. removes the local exclusive access
tags, and all global exclusive access tags for the processor.
For more information about the synchronization primitive instructions, see “LDREX and STREX”
on page 114 and “CLREX” on page 116.
13.4.8

Programming hints for the synchronization primitives
ANSI C cannot directly generate the exclusive access instructions. Some C compilers provide
intrinsic functions for generation of these instructions:
Table 13-8.

C compiler intrinsic functions for exclusive access instructions

Instruction

Intrinsic function

LDREX, LDREXH, or
LDREXB

unsigned int __ldrex(volatile void *ptr)

STREX, STREXH, or
STREXB

int __strex(unsigned int val, volatile void *ptr)

CLREX

void __clrex(void)

The actual exclusive access instruction generated depends on the data type of the pointer
passed to the intrinsic function. For example, the following C code generates the require
LDREXB operation:
__ldrex((volatile char *) 0xFF);

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.5

Exception model
This section describes the exception model.

13.5.1

13.5.1.1

Exception states
Each exception is in one of the following states:
Inactive
The exception is not active and not pending.

13.5.1.2

Pending
The exception is waiting to be serviced by the processor.
An interrupt request from a peripheral or from software can change the state of the corresponding interrupt to pending.

13.5.1.3

Active
An exception that is being serviced by the processor but has not completed.
An exception handler can interrupt the execution of another exception handler. In this case both
exceptions are in the active state.

13.5.1.4

13.5.2

13.5.2.1

Active and pending
The exception is being serviced by the processor and there is a pending exception from the
same source.
Exception types
The exception types are:
Reset
Reset is invoked on power up or a warm reset. The exception model treats reset as a special
form of exception. When reset is asserted, the operation of the processor stops, potentially at
any point in an instruction. When reset is deasserted, execution restarts from the address provided by the reset entry in the vector table. Execution restarts as privileged execution in Thread
mode.

13.5.2.2

Non Maskable Interrupt (NMI)
A non maskable interrupt (NMI) can be signalled by a peripheral or triggered by software. This is
the highest priority exception other than reset. It is permanently enabled and has a fixed priority
of -2.
NMIs cannot be:
• Masked or prevented from activation by any other exception.
• Preempted by any exception other than Reset.

13.5.2.3

Hard fault
A hard fault is an exception that occurs because of an error during exception processing, or
because an exception cannot be managed by any other exception mechanism. Hard faults have
a fixed priority of -1, meaning they have higher priority than any exception with configurable
priority.

77
6430F–ATARM–21-Feb-12

13.5.2.4

Memory management fault
A memory management fault is an exception that occurs because of a memory protection
related fault. The MPU or the fixed memory protection constraints determines this fault, for both
instruction and data memory transactions. This fault is used to abort instruction accesses to
Execute Never (XN) memory regions, even if the MPU is disabled.

13.5.2.5

Bus fault
A bus fault is an exception that occurs because of a memory related fault for an instruction or
data memory transaction. This might be from an error detected on a bus in the memory system.

13.5.2.6

Usage fault
A usage fault is an exception that occurs because of a fault related to instruction execution. This
includes:
• an undefined instruction
• an illegal unaligned access
• invalid state on instruction execution
• an error on exception return.
The following can cause a usage fault when the core is configured to report them:
• an unaligned address on word and halfword memory access
• division by zero.

13.5.2.7

SVCall
A supervisor call (SVC) is an exception that is triggered by the SVC instruction. In an OS environment, applications can use SVC instructions to access OS kernel functions and device
drivers.

13.5.2.8

PendSV
PendSV is an interrupt-driven request for system-level service. In an OS environment, use
PendSV for context switching when no other exception is active.

13.5.2.9

SysTick
A SysTick exception is an exception the system timer generates when it reaches zero. Software
can also generate a SysTick exception. In an OS environment, the processor can use this
exception as system tick.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.5.2.10

Interrupt (IRQ)
A interrupt, or IRQ, is an exception signalled by a peripheral, or generated by a software
request. All interrupts are asynchronous to instruction execution. In the system, peripherals use
interrupts to communicate with the processor.
Table 13-9.
Exception
number (1)

Properties of the different exception types
IRQ
number (
1)

Exception
type

Vector address
or offset (2)

Activation

Reset

-3, the
highest

0x00000004

Asynchronous

-14

NMI

-2

0x00000008

Asynchronous

-13

Hard fault

-1

0x0000000C

-12

Memory
management
fault

0x00000010

Synchronous

0x00000014

Synchronous when
precise,
asynchronous when
imprecise

Priority

Configurable
(3)

Configurable

-11

Bus fault

(3)

-10

Usage fault

(3)

0x00000018

Synchronous

7-10

Reserved

-5

SVCall

(3)

0x0000002C

Synchronous

12-13

Reserved

-2

PendSV

(3)

0x00000038

Asynchronous

-1

SysTick

(3)

0x0000003C

Asynchronous

16 and
above

0 and
above (4)

Interrupt (IRQ)

0x00000040 and
above (6)

Asynchronous

Configurable

Configurable
Configurable
Configurable
(5)

To simplify the software layer, the CMSIS only uses IRQ numbers and therefore uses negative
values for exceptions other than interrupts. The IPSR returns the Exception number, see
“Interrupt Program Status Register” on page 61.

See “Vector table” on page 81 for more information.

See “System Handler Priority Registers” on page 190.

See the “Peripheral Identifiers” section of the datasheet.

See “Interrupt Priority Registers” on page 172.

Increasing in steps of 4.

For an asynchronous exception, other than reset, the processor can execute another instruction
between when the exception is triggered and when the processor enters the exception handler.
Privileged software can disable the exceptions that Table 13-9 on page 79 shows as having configurable priority, see:
• “System Handler Control and State Register” on page 193

79
6430F–ATARM–21-Feb-12

• “Interrupt Clear-enable Registers” on page 168.
For more information about hard faults, memory management faults, bus faults, and usage
faults, see “Fault handling” on page 84.
13.5.3

Exception handlers
The processor handles exceptions using:

13.5.3.1

Interrupt Service Routines (ISRs)
Interrupts IRQ0 to IRQ29 are the exceptions handled by ISRs.

13.5.3.2

Fault handlers
Hard fault, memory management fault, usage fault, bus fault are fault exceptions handled by the
fault handlers.

13.5.3.3

System handlers
NMI, PendSV, SVCall SysTick, and the fault exceptions are all system exceptions that are handled by system handlers.

13.5.4

Vector table
The vector table contains the reset value of the stack pointer, and the start addresses, also
called exception vectors, for all exception handlers. Figure 13-3 on page 81 shows the order of
the exception vectors in the vector table. The least-significant bit of each vector must be 1, indicating that the exception handler is Thumb code.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 13-3. Vector table

Exception number IRQ number
45

.
.
.
18

-1

-2

Offset
0x00B4
.
.
.
0x004C
0x0048
0x0044
0x0040
0x003C
0x0038

IRQ29
.
.
.
IRQ2
IRQ1
IRQ0
Systick
PendSV
Reserved
Reserved for Debug

12
11

Vector

-5

0x002C

SVCall

Reserved

8
7
6

-10

-11

-12

-13

-14

0x0018
0x0014
0x0010
0x000C
0x0008
0x0004
0x0000

Usage fault
Bus fault
Memory management fault
Hard fault
Reserved
Reset
Initial SP value

On system reset, the vector table is fixed at address 0x00000000. Privileged software can write to
the VTOR to relocate the vector table start address to a different memory location, in the range
0x00000080 to 0x3FFFFF80, see “Vector Table Offset Register” on page 184.
13.5.5

Exception priorities
As Table 13-9 on page 79 shows, all exceptions have an associated priority, with:
• a lower priority value indicating a higher priority
• configurable priorities for all exceptions except Reset, Hard fault.
If software does not configure any priorities, then all exceptions with a configurable priority have
a priority of 0. For information about configuring exception priorities see
• “System Handler Priority Registers” on page 190
• “Interrupt Priority Registers” on page 172.

81
6430F–ATARM–21-Feb-12

Configurable priority values are in the range 0-15. This means that the Reset, Hard fault, and
NMI exceptions, with fixed negative priority values, always have higher priority than any other
exception.
For example, assigning a higher priority value to IRQ[0] and a lower priority value to IRQ[1]
means that IRQ[1] has higher priority than IRQ[0]. If both IRQ[1] and IRQ[0] are asserted, IRQ[1]
is processed before IRQ[0].
If multiple pending exceptions have the same priority, the pending exception with the lowest
exception number takes precedence. For example, if both IRQ[0] and IRQ[1] are pending and
have the same priority, then IRQ[0] is processed before IRQ[1].
When the processor is executing an exception handler, the exception handler is preempted if a
higher priority exception occurs. If an exception occurs with the same priority as the exception
being handled, the handler is not preempted, irrespective of the exception number. However,
the status of the new interrupt changes to pending.
13.5.6

Interrupt priority grouping
To increase priority control in systems with interrupts, the NVIC supports priority grouping. This
divides each interrupt priority register entry into two fields:
• an upper field that defines the group priority
• a lower field that defines a subpriority within the group.
Only the group priority determines preemption of interrupt exceptions. When the processor is
executing an interrupt exception handler, another interrupt with the same group priority as the
interrupt being handled does not preempt the handler,
If multiple pending interrupts have the same group priority, the subpriority field determines the
order in which they are processed. If multiple pending interrupts have the same group priority
and subpriority, the interrupt with the lowest IRQ number is processed first.
For information about splitting the interrupt priority fields into group priority and subpriority, see
“Application Interrupt and Reset Control Register” on page 185.

13.5.7

13.5.7.1

Exception entry and return
Descriptions of exception handling use the following terms:
Preemption
When the processor is executing an exception handler, an exception can preempt the exception
handler if its priority is higher than the priority of the exception being handled. See “Interrupt priority grouping” on page 82 for more information about preemption by an interrupt.
When one exception preempts another, the exceptions are called nested exceptions. See
“Exception entry” on page 83 more information.

13.5.7.2

Return
This occurs when the exception handler is completed, and:
• there is no pending exception with sufficient priority to be serviced
• the completed exception handler was not handling a late-arriving exception.
The processor pops the stack and restores the processor state to the state it had before the
interrupt occurred. See “Exception return” on page 84 for more information.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.5.7.3

Tail-chaining
This mechanism speeds up exception servicing. On completion of an exception handler, if there
is a pending exception that meets the requirements for exception entry, the stack pop is skipped
and control transfers to the new exception handler.

13.5.7.4

Late-arriving
This mechanism speeds up preemption. If a higher priority exception occurs during state saving
for a previous exception, the processor switches to handle the higher priority exception and initiates the vector fetch for that exception. State saving is not affected by late arrival because the
state saved is the same for both exceptions. Therefore the state saving continues uninterrupted.
The processor can accept a late arriving exception until the first instruction of the exception handler of the original exception enters the execute stage of the processor. On return from the
exception handler of the late-arriving exception, the normal tail-chaining rules apply.

13.5.7.5

Exception entry
Exception entry occurs when there is a pending exception with sufficient priority and either:
• the processor is in Thread mode
• the new exception is of higher priority than the exception being handled, in which case the
new exception preempts the original exception.
When one exception preempts another, the exceptions are nested.
Sufficient priority means the exception has more priority than any limits set by the mask registers, see “Exception mask registers” on page 62. An exception with less priority than this is
pending but is not handled by the processor.
When the processor takes an exception, unless the exception is a tail-chained or a late-arriving
exception, the processor pushes information onto the current stack. This operation is referred as
stacking and the structure of eight data words is referred as stack frame. The stack frame contains the following information:
• R0-R3, R12
• Return address
• PSR
• LR.
Immediately after stacking, the stack pointer indicates the lowest address in the stack frame.
Unless stack alignment is disabled, the stack frame is aligned to a double-word address. If the
STKALIGN bit of the Configuration Control Register (CCR) is set to 1, stack align adjustment is
performed during stacking.
The stack frame includes the return address. This is the address of the next instruction in the
interrupted program. This value is restored to the PC at exception return so that the interrupted
program resumes.
In parallel to the stacking operation, the processor performs a vector fetch that reads the exception handler start address from the vector table. When stacking is complete, the processor starts
executing the exception handler. At the same time, the processor writes an EXC_RETURN
value to the LR. This indicates which stack pointer corresponds to the stack frame and what
operation mode the was processor was in before the entry occurred.

83
6430F–ATARM–21-Feb-12

If no higher priority exception occurs during exception entry, the processor starts executing the
exception handler and automatically changes the status of the corresponding pending interrupt
to active.
If another higher priority exception occurs during exception entry, the processor starts executing
the exception handler for this exception and does not change the pending status of the earlier
exception. This is the late arrival case.
13.5.7.6

Exception return
Exception return occurs when the processor is in Handler mode and executes one of the following instructions to load the EXC_RETURN value into the PC:
• a POP instruction that includes the PC
• a BX instruction with any register.
• an LDR or LDM instruction with the PC as the destination.
EXC_RETURN is the value loaded into the LR on exception entry. The exception mechanism
relies on this value to detect when the processor has completed an exception handler. The lowest four bits of this value provide information on the return stack and processor mode. Table 1310 shows the EXC_RETURN[3:0] values with a description of the exception return behavior.
The processor sets EXC_RETURN bits[31:4] to 0xFFFFFFF. When this value is loaded into the PC
it indicates to the processor that the exception is complete, and the processor initiates the
exception return sequence.
Table 13-10. Exception return behavior

13.6

EXC_RETURN[3:0]

Description

bXXX0

Reserved.

b0001

Return to Handler mode.
Exception return gets state from MSP.
Execution uses MSP after return.

b0011

Reserved.

b01X1

Reserved.

b1001

Return to Thread mode.
Exception return gets state from MSP.
Execution uses MSP after return.

b1101

Return to Thread mode.
Exception return gets state from PSP.
Execution uses PSP after return.

b1X11

Reserved.

Fault handling
Faults are a subset of the exceptions, see “Exception model” on page 77. The following generate a fault:
– a bus error on:
– an instruction fetch or vector table load
– a data access

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• an internally-detected error such as an undefined instruction or an attempt to change state
with a BX instruction
• attempting to execute an instruction from a memory region marked as Non-Executable (XN).
• an MPU fault because of a privilege violation or an attempt to access an unmanaged region.
13.6.1

Fault types
Table 13-11 shows the types of fault, the handler used for the fault, the corresponding fault status register, and the register bit that indicates that the fault has occurred. See “Configurable
Fault Status Register” on page 195 for more information about the fault status registers.
Table 13-11. Faults
Fault

Handler

Bus error on a vector read

Bit name

Fault status register

VECTTBL

Fault escalated to a hard fault

FORCED

“Hard Fault Status
Register” on page 201

MPU mismatch:

Hard fault

on instruction access
on data access
during exception stacking

Memory
managem
ent fault

during exception unstacking

DACCVIOL
MSTKERR

during exception stacking

STKERR

during exception unstacking

UNSTKERR

during instruction prefetch

Bus fault

IBUSERR

Precise data bus error

PRECISERR

Imprecise data bus error

IMPRECISER
R

Attempt to access a coprocessor

NOCP

Undefined instruction

UNDEFINSTR

Attempt to enter an invalid instruction
set state (2)

“Memory Management
Fault Address Register” on
page 202

MUNSKERR

Bus error:

13.6.2

IACCVIOL (1)

Usage
fault

INVSTATE

Invalid EXC_RETURN value

INVPC

Illegal unaligned load or store

UNALIGNED

Divide By 0

DIVBYZERO

“Bus Fault Status Register”
on page 197

“Usage Fault Status
Register” on page 199

Occurs on an access to an XN region even if the MPU is disabled.

Attempting to use an instruction set other than the Thumb instruction set.

Fault escalation and hard faults
All faults exceptions except for hard fault have configurable exception priority, see “System Handler Priority Registers” on page 190. Software can disable execution of the handlers for these
faults, see “System Handler Control and State Register” on page 193.

85
6430F–ATARM–21-Feb-12

Usually, the exception priority, together with the values of the exception mask registers, determines whether the processor enters the fault handler, and whether a fault handler can preempt
another fault handler. as described in “Exception model” on page 77.
In some situations, a fault with configurable priority is treated as a hard fault. This is called priority escalation, and the fault is described as escalated to hard fault. Escalation to hard fault
occurs when:
• A fault handler causes the same kind of fault as the one it is servicing. This escalation to hard
fault occurs because a fault handler cannot preempt itself because it must have the same
priority as the current priority level.
• A fault handler causes a fault with the same or lower priority as the fault it is servicing. This is
because the handler for the new fault cannot preempt the currently executing fault handler.
• An exception handler causes a fault for which the priority is the same as or lower than the
currently executing exception.
• A fault occurs and the handler for that fault is not enabled.
If a bus fault occurs during a stack push when entering a bus fault handler, the bus fault does not
escalate to a hard fault. This means that if a corrupted stack causes a fault, the fault handler
executes even though the stack push for the handler failed. The fault handler operates but the
stack contents are corrupted.
Only Reset and NMI can preempt the fixed priority hard fault. A hard fault can preempt any
exception other than Reset, NMI, or another hard fault.

13.6.3

Fault status registers and fault address registers
The fault status registers indicate the cause of a fault. For bus faults and memory management
faults, the fault address register indicates the address accessed by the operation that caused
the fault, as shown in Table 13-12.
Table 13-12. Fault status and fault address registers

13.6.4

Handler

Status register
name

Address register
name

Hard fault

HFSR

“Hard Fault Status Register” on page
201

Memory
management fault

MMFSR

MMFAR

“Memory Management Fault Status
Register” on page 196
“Memory Management Fault Address
Register” on page 202

Bus fault

BFSR

BFAR

“Bus Fault Status Register” on page 197
“Bus Fault Address Register” on page
203

Usage fault

UFSR

“Usage Fault Status Register” on page
199

Lockup
The processor enters a lockup state if a hard fault occurs when executing the hard fault handlers. When the processor is in lockup state it does not execute any instructions. The processor
remains in lockup state until:
• it is reset

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.7

Power management
The Cortex-M3 processor sleep modes reduce power consumption:
• Backup Mode
• Wait Mode
• Sleep Mode
The SLEEPDEEP bit of the SCR selects which sleep mode is used, see “System Control Register” on page 187. For more information about the behavior of the sleep modes see “Low Power
Modes” in the PMC section of the datasheet.
This section describes the mechanisms for entering sleep mode, and the conditions for waking
up from sleep mode.

13.7.1

Entering sleep mode
This section describes the mechanisms software can use to put the processor into sleep mode.
The system can generate spurious wakeup events, for example a debug operation wakes up the
processor. Therefore software must be able to put the processor back into sleep mode after
such an event. A program might have an idle loop to put the processor back to sleep mode.

13.7.1.1

Wait for interrupt
The wait for interrupt instruction, WFI, causes immediate entry to sleep mode. When the processor executes a WFI instruction it stops executing instructions and enters sleep mode. See “WFI”
on page 163 for more information.

13.7.1.2

Wait for event
The wait for event instruction, WFE, causes entry to sleep mode conditional on the value of an
one-bit event register. When the processor executes a WFE instruction, it checks this register:
• if the register is 0 the processor stops executing instructions and enters sleep mode
• if the register is 1 the processor clears the register to 0 and continues executing instructions
without entering sleep mode.
See “WFE” on page 162 for more information.

13.7.1.3

13.7.2

13.7.2.1

Sleep-on-exit
If the SLEEPONEXIT bit of the SCR is set to 1, when the processor completes the execution of
an exception handler it returns to Thread mode and immediately enters sleep mode. Use this
mechanism in applications that only require the processor to run when an exception occurs.
Wakeup from sleep mode
The conditions for the processor to wakeup depend on the mechanism that cause it to enter
sleep mode.
Wakeup from WFI or sleep-on-exit
Normally, the processor wakes up only when it detects an exception with sufficient priority to
cause exception entry.
Some embedded systems might have to execute system restore tasks after the processor
wakes up, and before it executes an interrupt handler. To achieve this set the PRIMASK bit to 1
and the FAULTMASK bit to 0. If an interrupt arrives that is enabled and has a higher priority than
current exception priority, the processor wakes up but does not execute the interrupt handler

87
6430F–ATARM–21-Feb-12

until the processor sets PRIMASK to zero. For more information about PRIMASK and FAULTMASK see “Exception mask registers” on page 62.
13.7.2.2

Wakeup from WFE
The processor wakes up if:
• it detects an exception with sufficient priority to cause exception entry
In addition, if the SEVONPEND bit in the SCR is set to 1, any new pending interrupt triggers an
event and wakes up the processor, even if the interrupt is disabled or has insufficient priority to
cause exception entry. For more information about the SCR see “System Control Register” on
page 187.

13.7.3

Power management programming hints
ANSI C cannot directly generate the WFI and WFE instructions. The CMSIS provides the following intrinsic functions for these instructions:
void __WFE(void) // Wait for Event
void __WFE(void) // Wait for Interrupt

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.8

Instruction set summary
The processor implements a version of the Thumb instruction set. Table 13-13 lists the supported instructions.
In Table 13-13:
• angle brackets, <>, enclose alternative forms of the operand
• braces, {}, enclose optional operands
• the Operands column is not exhaustive
• Op2 is a flexible second operand that can be either a register or a constant
• most instructions can use an optional condition code suffix.
For more information on the instructions and operands, see the instruction descriptions.
Table 13-13. Cortex-M3 instructions
Mnemonic

Operands

Brief description

Flags

Page

ADC, ADCS

{Rd,} Rn, Op2

Add with Carry

N,Z,C,V

page 119

ADD, ADDS

{Rd,} Rn, Op2

Add

N,Z,C,V

page 119

ADD, ADDW

{Rd,} Rn, #imm12

Add

N,Z,C,V

page 119

ADR

Rd, label

Load PC-relative address

page 102

AND, ANDS

{Rd,} Rn, Op2

Logical AND

N,Z,C

page 122

ASR, ASRS

Rd, Rm,

Arithmetic Shift Right

N,Z,C

page 124

label

Branch

page 144

BFC

Rd, #lsb, #width

Bit Field Clear

page 140

BFI

Rd, Rn, #lsb, #width

Bit Field Insert

page 140

BIC, BICS

{Rd,} Rn, Op2

Bit Clear

N,Z,C

page 122

BKPT

#imm

Breakpoint

page 152

label

Branch with Link

page 144

BLX

Branch indirect with Link

page 144

Branch indirect

page 144

CBNZ

Rn, label

Compare and Branch if Non Zero

page 146

CBZ

Rn, label

Compare and Branch if Zero

page 146

CLREX

Clear Exclusive

page 116

CLZ

Rd, Rm

Count leading zeros

page 126

CMN, CMNS

Rn, Op2

Compare Negative

N,Z,C,V

page 127

CMP, CMPS

Rn, Op2

Compare

N,Z,C,V

page 127

CPSID

iflags

Change Processor State, Disable
Interrupts

page 153

CPSIE

iflags

Change Processor State, Enable
Interrupts

page 153

DMB

Data Memory Barrier

page 154

DSB

Data Synchronization Barrier

page 155

EOR, EORS

{Rd,} Rn, Op2

Exclusive OR

N,Z,C

page 122

89
6430F–ATARM–21-Feb-12

Table 13-13. Cortex-M3 instructions (Continued)

Mnemonic

Operands

Brief description

Flags

Page

ISB

Instruction Synchronization Barrier

page 156

If-Then condition block

page 147

LDM

Rn{!}, reglist

Load Multiple registers, increment after -

page 111

LDMDB,
LDMEA

Rn{!}, reglist

Load Multiple registers, decrement
before

page 111

LDMFD,
LDMIA

Rn{!}, reglist

Load Multiple registers, increment after -

page 111

LDR

Rt, [Rn, #offset]

Load Register with word

page 106

LDRB,
LDRBT

Rt, [Rn, #offset]

Load Register with byte

page 106

LDRD

Rt, Rt2, [Rn, #offset]

Load Register with two bytes

page 106

LDREX

Rt, [Rn, #offset]

Load Register Exclusive

page 106

LDREXB

Rt, [Rn]

Load Register Exclusive with byte

page 106

LDREXH

Rt, [Rn]

Load Register Exclusive with halfword

page 106

LDRH,
LDRHT

Rt, [Rn, #offset]

Load Register with halfword

page 106

LDRSB,
LDRSBT

Rt, [Rn, #offset]

Load Register with signed byte

page 106

LDRSH,
LDRSHT

Rt, [Rn, #offset]

Load Register with signed halfword

page 106

LDRT

Rt, [Rn, #offset]

Load Register with word

page 106

LSL, LSLS

Rd, Rm,

Logical Shift Left

N,Z,C

page 124

LSR, LSRS

Rd, Rm,

Logical Shift Right

N,Z,C

page 124

MLA

Rd, Rn, Rm, Ra

Multiply with Accumulate, 32-bit result

page 134

MLS

Rd, Rn, Rm, Ra

Multiply and Subtract, 32-bit result

page 134

MOV, MOVS

Rd, Op2

Move

N,Z,C

page 128

MOVT

Rd, #imm16

Move Top

page 130

MOVW, MOV

Rd, #imm16

Move 16-bit constant

N,Z,C

page 128

MRS

Rd, spec_reg

Move from special register to general
register

page 157

MSR

spec_reg, Rm

Move from general register to special
register

N,Z,C,V

page 158

MUL, MULS

{Rd,} Rn, Rm

Multiply, 32-bit result

N,Z

page 134

MVN, MVNS

Rd, Op2

Move NOT

N,Z,C

page 128

NOP

No Operation

page 159

ORN, ORNS

{Rd,} Rn, Op2

Logical OR NOT

N,Z,C

page 122

ORR, ORRS

{Rd,} Rn, Op2

Logical OR

N,Z,C

page 122

POP

reglist

Pop registers from stack

page 113

PUSH

reglist

Push registers onto stack

page 113

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 13-13. Cortex-M3 instructions (Continued)
Mnemonic

Operands

Brief description

Flags

Page

RBIT

Rd, Rn

Reverse Bits

page 131

REV

Rd, Rn

Reverse byte order in a word

page 131

REV16

Rd, Rn

Reverse byte order in each halfword

page 131

REVSH

Rd, Rn

Reverse byte order in bottom halfword
and sign extend

page 131

ROR, RORS

Rd, Rm,

Rotate Right

N,Z,C

page 124

RRX, RRXS

Rd, Rm

Rotate Right with Extend

N,Z,C

page 124

RSB, RSBS

{Rd,} Rn, Op2

Reverse Subtract

N,Z,C,V

page 119

SBC, SBCS

{Rd,} Rn, Op2

Subtract with Carry

N,Z,C,V

page 119

SBFX

Rd, Rn, #lsb, #width

Signed Bit Field Extract

page 141

SDIV

{Rd,} Rn, Rm

Signed Divide

page 136

SEV

Send Event

page 160

SMLAL

RdLo, RdHi, Rn, Rm

Signed Multiply with Accumulate (32 x
32 + 64), 64-bit result

page 135

SMULL

RdLo, RdHi, Rn, Rm

Signed Multiply (32 x 32), 64-bit result

page 135

SSAT

Rd, #n, Rm {,shift #s}

Signed Saturate

page 137

STM

Rn{!}, reglist

Store Multiple registers, increment after -

page 111

STMDB,
STMEA

Rn{!}, reglist

Store Multiple registers, decrement
before

page 111

STMFD,
STMIA

Rn{!}, reglist

Store Multiple registers, increment after -

page 111

STR

Rt, [Rn, #offset]

Store Register word

page 106

STRB,
STRBT

Rt, [Rn, #offset]

Store Register byte

page 106

STRD

Rt, Rt2, [Rn, #offset]

Store Register two words

page 106

STREX

Rd, Rt, [Rn, #offset]

Store Register Exclusive

page 114

STREXB

Rd, Rt, [Rn]

Store Register Exclusive byte

page 114

STREXH

Rd, Rt, [Rn]

Store Register Exclusive halfword

page 114

STRH,
STRHT

Rt, [Rn, #offset]

Store Register halfword

page 106

STRT

Rt, [Rn, #offset]

Store Register word

page 106

SUB, SUBS

{Rd,} Rn, Op2

Subtract

N,Z,C,V

page 119

SUB, SUBW

{Rd,} Rn, #imm12

Subtract

N,Z,C,V

page 119

SVC

#imm

Supervisor Call

page 161

SXTB

{Rd,} Rm {,ROR #n}

Sign extend a byte

page 142

SXTH

{Rd,} Rm {,ROR #n}

Sign extend a halfword

page 142

TBB

[Rn, Rm]

Table Branch Byte

page 149

TBH

[Rn, Rm, LSL #1]

Table Branch Halfword

page 149

91
6430F–ATARM–21-Feb-12

Table 13-13. Cortex-M3 instructions (Continued)

13.9

Mnemonic

Operands

Brief description

Flags

Page

TEQ

Rn, Op2

Test Equivalence

N,Z,C

page 132

TST

Rn, Op2

Test

N,Z,C

page 132

UBFX

Rd, Rn, #lsb, #width

Unsigned Bit Field Extract

page 141

UDIV

{Rd,} Rn, Rm

Unsigned Divide

page 136

UMLAL

RdLo, RdHi, Rn, Rm

Unsigned Multiply with Accumulate
(32 x 32 + 64), 64-bit result

page 135

UMULL

RdLo, RdHi, Rn, Rm

Unsigned Multiply (32 x 32), 64-bit
result

page 135

USAT

Rd, #n, Rm {,shift #s}

Unsigned Saturate

page 137

UXTB

{Rd,} Rm {,ROR #n}

Zero extend a byte

page 142

UXTH

{Rd,} Rm {,ROR #n}

Zero extend a halfword

page 142

WFE

Wait For Event

page 162

WFI

Wait For Interrupt

page 163

Intrinsic functions
ANSI cannot directly access some Cortex-M3 instructions. This section describes intrinsic functions that can generate these instructions, provided by the CMIS and that might be provided by a
C compiler. If a C compiler does not support an appropriate intrinsic function, you might have to
use inline assembler to access some instructions.
The CMSIS provides the following intrinsic functions to generate instructions that ANSI cannot
directly access:
Table 13-14. CMSIS intrinsic functions to generate some Cortex-M3 instructions

Instruction

CMSIS intrinsic function

CPSIE I

void __enable_irq(void)

CPSID I

void __disable_irq(void)

CPSIE F

void __enable_fault_irq(void)

CPSID F

void __disable_fault_irq(void)

ISB

void __ISB(void)

DSB

void __DSB(void)

DMB

void __DMB(void)

REV

uint32_t __REV(uint32_t int value)

REV16

uint32_t __REV16(uint32_t int value)

REVSH

uint32_t __REVSH(uint32_t int value)

RBIT

uint32_t __RBIT(uint32_t int value)

SEV

void __SEV(void)

WFE

void __WFE(void)

WFI

void __WFI(void)

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The CMSIS also provides a number of functions for accessing the special registers using MRS
and MSR instructions:
Table 13-15. CMSIS intrinsic functions to access the special registers
Special register

Access

CMSIS function

Read

uint32_t __get_PRIMASK (void)

Write

void __set_PRIMASK (uint32_t value)

Read

uint32_t __get_FAULTMASK (void)

Write

void __set_FAULTMASK (uint32_t value)

Read

uint32_t __get_BASEPRI (void)

Write

void __set_BASEPRI (uint32_t value)

Read

uint32_t __get_CONTROL (void)

Write

void __set_CONTROL (uint32_t value)

Read

uint32_t __get_MSP (void)

Write

void __set_MSP (uint32_t TopOfMainStack)

Read

uint32_t __get_PSP (void)

Write

void __set_PSP (uint32_t TopOfProcStack)

PRIMASK

FAULTMASK

BASEPRI

CONTROL

MSP

PSP

13.10 About the instruction descriptions
The following sections give more information about using the instructions:
• “Operands” on page 93
• “Restrictions when using PC or SP” on page 93
• “Flexible second operand” on page 94
• “Shift Operations” on page 95
• “Address alignment” on page 97
• “PC-relative expressions” on page 98
• “Conditional execution” on page 98
• “Instruction width selection” on page 100.
13.10.1

Operands
An instruction operand can be an ARM register, a constant, or another instruction-specific
parameter. Instructions act on the operands and often store the result in a destination register.
When there is a destination register in the instruction, it is usually specified before the operands.
Operands in some instructions are flexible in that they can either be a register or a constant. See
“Flexible second operand”.

13.10.2

Restrictions when using PC or SP
Many instructions have restrictions on whether you can use the Program Counter (PC) or Stack
Pointer (SP) for the operands or destination register. See instruction descriptions for more
information.

93
6430F–ATARM–21-Feb-12

Bit[0] of any address you write to the PC with a BX, BLX, LDM, LDR, or POP instruction must be
1 for correct execution, because this bit indicates the required instruction set, and the Cortex-M3
processor only supports Thumb instructions.

13.10.3

Flexible second operand
Many general data processing instructions have a flexible second operand. This is shown as
Operand2 in the descriptions of the syntax of each instruction.
Operand2 can be a:
• “Constant”
• “Register with optional shift” on page 94

13.10.3.1

Constant
You specify an Operand2 constant in the form:
#constant

where constant can be:
• any constant that can be produced by shifting an 8-bit value left by any number of bits within
a 32-bit word
• any constant of the form 0x00XY00XY
• any constant of the form 0xXY00XY00
• any constant of the form 0xXYXYXYXY.
In the constants shown above, X and Y are hexadecimal digits.

In addition, in a small number of instructions, constant can take a wider range of values.
These are described in the individual instruction descriptions.
When an Operand2 constant is used with the instructions MOVS, MVNS, ANDS, ORRS, ORNS,
EORS, BICS, TEQ or TST, the carry flag is updated to bit[31] of the constant, if the constant is
greater than 255 and can be produced by shifting an 8-bit value. These instructions do not affect
the carry flag if Operand2 is any other constant.
13.10.3.2

Instruction substitution
Your assembler might be able to produce an equivalent instruction in cases where you specify a
constant that is not permitted. For example, an assembler might assemble the instruction CMP
Rd, #0xFFFFFFFE as the equivalent instruction CMN Rd, #0x2.

13.10.3.3

is the register holding the data for the second operand.

shift

is an optional shift to be applied to Rm. It can be one of:
ASR #n

arithmetic shift right n bits, 1 ≤n ≤32.

LSL #n

logical shift left n bits, 1 ≤n ≤31.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
LSR #n

logical shift right n bits, 1 ≤n ≤32.

ROR #n

rotate right n bits, 1 ≤n ≤31.

RRX

rotate right one bit, with extend.

if omitted, no shift occurs, equivalent to LSL #0.

If you omit the shift, or specify LSL #0, the instruction uses the value in Rm.
If you specify a shift, the shift is applied to the value in Rm, and the resulting 32-bit value is used
by the instruction. However, the contents in the register Rm remains unchanged. Specifying a
register with shift also updates the carry flag when used with certain instructions. For information
on the shift operations and how they affect the carry flag, see “Shift Operations”
13.10.4

Shift Operations
Register shift operations move the bits in a register left or right by a specified number of bits, the
shift length. Register shift can be performed:
• directly by the instructions ASR, LSR, LSL, ROR, and RRX, and the result is written to a
destination register
• during the calculation of Operand2 by the instructions that specify the second operand as a
register with shift, see “Flexible second operand” on page 94. The result is used by the
instruction.
The permitted shift lengths depend on the shift type and the instruction, see the individual
instruction description or “Flexible second operand” on page 94. If the shift length is 0, no shift
occurs. Register shift operations update the carry flag except when the specified shift length is 0.
The following sub-sections describe the various shift operations and how they affect the carry
flag. In these descriptions, Rm is the register containing the value to be shifted, and n is the shift
length.

13.10.4.1

ASR
Arithmetic shift right by n bits moves the left-hand 32-n bits of the register Rm, to the right by n
places, into the right-hand 32-n bits of the result. And it copies the original bit[31] of the register
into the left-hand n bits of the result. See Figure 13-4 on page 95.
You can use the ASR #n operation to divide the value in the register Rm by 2n, with the result
being rounded towards negative-infinity.
When the instruction is ASRS or when ASR #n is used in Operand2 with the instructions MOVS,
MVNS, ANDS, ORRS, ORNS, EORS, BICS, TEQ or TST, the carry flag is updated to the last bit
shifted out, bit[n-1], of the register Rm.
• If n is 32 or more, then all the bits in the result are set to the value of bit[31] of Rm.
• If n is 32 or more and the carry flag is updated, it is updated to the value of bit[31] of Rm.
Figure 13-4. ASR #3
Carry
Flag
31

5 4 3 2 1 0

...

95
6430F–ATARM–21-Feb-12

13.10.4.2

LSR
Logical shift right by n bits moves the left-hand 32-n bits of the register Rm, to the right by n
places, into the right-hand 32-n bits of the result. And it sets the left-hand n bits of the result to 0.
See Figure 13-5.
You can use the LSR #n operation to divide the value in the register Rm by 2n, if the value is
regarded as an unsigned integer.
When the instruction is LSRS or when LSR #n is used in Operand2 with the instructions MOVS,
MVNS, ANDS, ORRS, ORNS, EORS, BICS, TEQ or TST, the carry flag is updated to the last bit
shifted out, bit[n-1], of the register Rm.
• If n is 32 or more, then all the bits in the result are cleared to 0.
• If n is 33 or more and the carry flag is updated, it is updated to 0.
Figure 13-5. LSR #3
0

Carry
Flag

5 4 3 2 1 0

...
13.10.4.3

LSL
Logical shift left by n bits moves the right-hand 32-n bits of the register Rm, to the left by n
places, into the left-hand 32-n bits of the result. And it sets the right-hand n bits of the result to 0.
See Figure 13-6 on page 96.
You can use he LSL #n operation to multiply the value in the register Rm by 2n, if the value is
regarded as an unsigned integer or a two’s complement signed integer. Overflow can occur
without warning.
When the instruction is LSLS or when LSL #n, with non-zero n, is used in Operand2 with the
instructions MOVS, MVNS, ANDS, ORRS, ORNS, EORS, BICS, TEQ or TST, the carry flag is
updated to the last bit shifted out, bit[32-n], of the register Rm. These instructions do not affect
the carry flag when used with LSL #0.
• If n is 32 or more, then all the bits in the result are cleared to 0.
• If n is 33 or more and the carry flag is updated, it is updated to 0.
Figure 13-6. LSL #3
0

31
Carry
Flag

5 4 3 2 1 0

...

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.10.4.4

ROR
Rotate right by n bits moves the left-hand 32-n bits of the register Rm, to the right by n places,
into the right-hand 32-n bits of the result. And it moves the right-hand n bits of the register into
the left-hand n bits of the result. See Figure 13-7.
When the instruction is RORS or when ROR #n is used in Operand2 with the instructions MOVS,
MVNS, ANDS, ORRS, ORNS, EORS, BICS, TEQ or TST, the carry flag is updated to the last bit
rotation, bit[n-1], of the register Rm.
• If n is 32, then the value of the result is same as the value in Rm, and if the carry flag is
updated, it is updated to bit[31] of Rm.
• ROR with shift length, n, more than 32 is the same as ROR with shift length n-32.
Figure 13-7. ROR #3
Carry
Flag
31

5 4 3 2

1 0

...
13.10.4.5

RRX
Rotate right with extend moves the bits of the register Rm to the right by one bit. And it copies
the carry flag into bit[31] of the result. See Figure 13-8 on page 97.
When the instruction is RRXS or when RRX is used in Operand2 with the instructions MOVS,
MVNS, ANDS, ORRS, ORNS, EORS, BICS, TEQ or TST, the carry flag is updated to bit[0] of
the register Rm.
Figure 13-8. RRX
Carry
Flag
31 30

1 0

...
13.10.5

...

Address alignment
An aligned access is an operation where a word-aligned address is used for a word, dual word,
or multiple word access, or where a halfword-aligned address is used for a halfword access.
Byte accesses are always aligned.
The Cortex-M3 processor supports unaligned access only for the following instructions:
• LDR, LDRT
• LDRH, LDRHT
• LDRSH, LDRSHT
• STR, STRT
• STRH, STRHT

97
6430F–ATARM–21-Feb-12

All other load and store instructions generate a usage fault exception if they perform an
unaligned access, and therefore their accesses must be address aligned. For more information
about usage faults see “Fault handling” on page 84.
Unaligned accesses are usually slower than aligned accesses. In addition, some memory
regions might not support unaligned accesses. Therefore, ARM recommends that programmers
ensure that accesses are aligned. To avoid accidental generation of unaligned accesses, use
the UNALIGN_TRP bit in the Configuration and Control Register to trap all unaligned accesses,
see “Configuration and Control Register” on page 188.
13.10.6

PC-relative expressions
A PC-relative expression or label is a symbol that represents the address of an instruction or literal data. It is represented in the instruction as the PC value plus or minus a numeric offset. The
assembler calculates the required offset from the label and the address of the current instruction. If the offset is too big, the assembler produces an error.
• For B, BL, CBNZ, and CBZ instructions, the value of the PC is the address of the current
instruction plus 4 bytes.
• For all other instructions that use labels, the value of the PC is the address of the current
instruction plus 4 bytes, with bit[1] of the result cleared to 0 to make it word-aligned.
• Your assembler might permit other syntaxes for PC-relative expressions, such as a label plus
or minus a number, or an expression of the form [PC, #number].

13.10.7

Conditional execution
Most data processing instructions can optionally update the condition flags in the Application
Program Status Register (APSR) according to the result of the operation, see “Application Program Status Register” on page 60. Some instructions update all flags, and some only update a
subset. If a flag is not updated, the original value is preserved. See the instruction descriptions
for the flags they affect.
You can execute an instruction conditionally, based on the condition flags set in another instruction, either:
• immediately after the instruction that updated the flags
• after any number of intervening instructions that have not updated the flags.
Conditional execution is available by using conditional branches or by adding condition code
suffixes to instructions. See Table 13-16 on page 99 for a list of the suffixes to add to instructions
to make them conditional instructions. The condition code suffix enables the processor to test a
condition based on the flags. If the condition test of a conditional instruction fails, the instruction:
• does not execute
• does not write any value to its destination register
• does not affect any of the flags
• does not generate any exception.
Conditional instructions, except for conditional branches, must be inside an If-Then instruction
block. See “IT” on page 147 for more information and restrictions when using the IT instruction.
Depending on the vendor, the assembler might automatically insert an IT instruction if you have
conditional instructions outside the IT block.
Use the CBZ and CBNZ instructions to compare the value of a register against zero and branch
on the result.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
This section describes:
• “The condition flags”
• “Condition code suffixes”.
13.10.7.1

The condition flags
The APSR contains the following condition flags:
N

Set to 1 when the result of the operation was negative, cleared to 0 otherwise.

Set to 1 when the result of the operation was zero, cleared to 0 otherwise.

Set to 1 when the operation resulted in a carry, cleared to 0 otherwise.

Set to 1 when the operation caused overflow, cleared to 0 otherwise.

For more information about the APSR see “Program Status Register” on page 59.
A carry occurs:
• if the result of an addition is greater than or equal to 232
• if the result of a subtraction is positive or zero
• as the result of an inline barrel shifter operation in a move or logical instruction.
Overflow occurs if the result of an add, subtract, or compare is greater than or equal to 231, or
less than –231.
Most instructions update the status flags only if the S suffix is specified. See the instruction
descriptions for more information.

13.10.7.2

Condition code suffixes
The instructions that can be conditional have an optional condition code, shown in syntax
descriptions as {cond}. Conditional execution requires a preceding IT instruction. An instruction
with a condition code is only executed if the condition code flags in the APSR meet the specified
condition. Table 13-16 shows the condition codes to use.
You can use conditional execution with the IT instruction to reduce the number of branch instructions in code.
Table 13-16 also shows the relationship between condition code suffixes and the N, Z, C, and V
flags.
Table 13-16. Condition code suffixes
Suffix

Flags

Meaning

Z=1

Equal

Z=0

Not equal

CS or
HS

C=1

Higher or same, unsigned ≥

CC or
LO

C=0

Lower, unsigned <

N=1

Negative

N=0

Positive or zero

V=1

Overflow

99
6430F–ATARM–21-Feb-12

Table 13-16. Condition code suffixes (Continued)
Suffix

Flags

Meaning

V=0

No overflow

C = 1 and Z = 0

Higher, unsigned >

C = 0 or Z = 1

Lower or same, unsigned ≤

N=V

Greater than or equal, signed ≥

N! =V

Less than, signed <

Z = 0 and N = V

Greater than, signed >

Z = 1 and N ! = V

Less than or equal, signed ≤

Can have any
value

Always. This is the default when no suffix is
specified.

13.10.7.3
Absolute value
The example below shows the use of a conditional instruction to find the absolute value of a number. R0 = ABS(R1).
MOVS
IT
RSBMI

R0, R1
MI
R0, R1, #0

; R0 = R1, setting flags
; IT instruction for the negative condition
; If negative, R0 = -R1

13.10.7.4
Compare and update value
The example below shows the use of conditional instructions to update the value of R4 if the signed values R0 is greater
than R1 and R2 is greater than R3.
CMP
ITT
CMPGT
MOVGT

13.10.8

R0, R1
GT
R2, R3
R4, R5

;
;
;
;

Compare R0 and R1, setting flags
IT instruction for the two GT conditions
If 'greater than', compare R2 and R3, setting flags
If still 'greater than', do R4 = R5

Instruction width selection
There are many instructions that can generate either a 16-bit encoding or a 32-bit encoding
depending on the operands and destination register specified. For some of these instructions,
you can force a specific instruction size by using an instruction width suffix. The .W suffix forces
a 32-bit instruction encoding. The .N suffix forces a 16-bit instruction encoding.
If you specify an instruction width suffix and the assembler cannot generate an instruction
encoding of the requested width, it generates an error.
In some cases it might be necessary to specify the .W suffix, for example if the operand is the
label of an instruction or literal data, as in the case of branch instructions. This is because the
assembler might not automatically generate the right size encoding.

13.10.8.1
Instruction width selection
To use an instruction width suffix, place it immediately after the instruction mnemonic and condition code, if any. The example below shows instructions with the instruction width suffix.
BCS.W

label

; creates a 32-bit instruction even for a short branch

ADDS.W R0, R0, R1 ; creates a 32-bit instruction even though the same
; operation can be done by a 16-bit instruction

100

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.11 Memory access instructions
Table 13-17 shows the memory access instructions:
Table 13-17. Memory access instructions
Mnemonic

Brief description

See

ADR

Load PC-relative address

“ADR” on page 102

CLREX

Clear Exclusive

“CLREX” on page 116

LDM{mode}

Load Multiple registers

“LDM and STM” on page 111

LDR{type}

Load Register using immediate
offset

“LDR and STR, immediate offset” on
page 103

LDR{type}

Load Register using register offset

“LDR and STR, register offset” on page
106

LDR{type}T

Load Register with unprivileged
access

“LDR and STR, unprivileged” on page
108

LDR

Load Register using PC-relative
address

“LDR, PC-relative” on page 109

LDREX{type}

Load Register Exclusive

“LDREX and STREX” on page 114

POP

Pop registers from stack

“PUSH and POP” on page 113

PUSH

Push registers onto stack

“PUSH and POP” on page 113

STM{mode}

Store Multiple registers

“LDM and STM” on page 111

STR{type}

Store Register using immediate
offset

“LDR and STR, immediate offset” on
page 103

STR{type}

Store Register using register offset

“LDR and STR, register offset” on page
106

STR{type}T

Store Register with unprivileged
access

“LDR and STR, unprivileged” on page
108

STREX{type}

Store Register Exclusive

“LDREX and STREX” on page 114

101
6430F–ATARM–21-Feb-12

13.11.1

ADR
Load PC-relative address.

13.11.1.1

Syntax
ADR{cond} Rd, label

where:

13.11.1.2

cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register.

label

is a PC-relative expression. See “PC-relative expressions” on page 98.

Operation
ADR determines the address by adding an immediate value to the PC, and writes the result to
the destination register.
ADR produces position-independent code, because the address is PC-relative.
If you use ADR to generate a target address for a BX or BLX instruction, you must ensure that
bit[0] of the address you generate is set to1 for correct execution.
Values of label must be within the range of −4095 to +4095 from the address in the PC.
You might have to use the .W suffix to get the maximum offset range or to generate addresses
that are not word-aligned. See “Instruction width selection” on page 100.

13.11.1.3

Restrictions
Rd must not be SP and must not be PC.

13.11.1.4

Condition flags
This instruction does not change the flags.

13.11.1.5

Examples

ADR

102

R1, TextMessage

; Write address value of a location labelled as
; TextMessage to R1

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.11.2

13.11.2.1

LDR and STR, immediate offset
Load and Store with immediate offset, pre-indexed immediate offset, or post-indexed immediate
offset.
Syntax
op{type}{cond} Rt, [Rn {, #offset}]

; immediate offset

op{type}{cond} Rt, [Rn, #offset]!

; pre-indexed

op{type}{cond} Rt, [Rn], #offset

; post-indexed

opD{cond} Rt, Rt2, [Rn {, #offset}]

; immediate offset, two words

opD{cond} Rt, Rt2, [Rn, #offset]!

; pre-indexed, two words

opD{cond} Rt, Rt2, [Rn], #offset

; post-indexed, two words

where:
op

is one of:
LDR

Load Register.

STR

Store Register.

type

13.11.2.2

is one of:
B

unsigned byte, zero extend to 32 bits on loads.

signed byte, sign extend to 32 bits (LDR only).

unsigned halfword, zero extend to 32 bits on loads.

signed halfword, sign extend to 32 bits (LDR only).

omit, for word.

cond

is an optional condition code, see “Conditional execution” on page 98.

is the register to load or store.

is the register on which the memory address is based.

offset

is an offset from Rn. If offset is omitted, the address is the contents of Rn.

Rt2

is the additional register to load or store for two-word operations.

Operation
LDR instructions load one or two registers with a value from memory.
STR instructions store one or two register values to memory.
Load and store instructions with immediate offset can use the following addressing modes:

13.11.2.3

Offset addressing
The offset value is added to or subtracted from the address obtained from the register Rn. The
result is used as the address for the memory access. The register Rn is unaltered. The assembly language syntax for this mode is:
[Rn, #offset]

13.11.2.4

Pre-indexed addressing
The offset value is added to or subtracted from the address obtained from the register Rn. The
result is used as the address for the memory access and written back into the register Rn. The
assembly language syntax for this mode is:
103

6430F–ATARM–21-Feb-12

[Rn, #offset]!

13.11.2.5

Post-indexed addressing
The address obtained from the register Rn is used as the address for the memory access. The
offset value is added to or subtracted from the address, and written back into the register Rn.
The assembly language syntax for this mode is:
[Rn], #offset

The value to load or store can be a byte, halfword, word, or two words. Bytes and halfwords can
either be signed or unsigned. See “Address alignment” on page 97.
Table 13-18 shows the ranges of offset for immediate, pre-indexed and post-indexed forms.
Table 13-18. Offset ranges

13.11.2.6

Instruction type

Immediate offset

Pre-indexed

Post-indexed

Word, halfword, signed
halfword, byte, or signed
byte

−255 to 4095

−255 to 255

Two words

multiple of 4 in the
range −1020 to
1020

Restrictions
For load instructions:
• Rt can be SP or PC for word loads only
• Rt must be different from Rt2 for two-word loads
• Rn must be different from Rt and Rt2 in the pre-indexed or post-indexed forms.
When Rt is PC in a word load instruction:
• bit[0] of the loaded value must be 1 for correct execution
• a branch occurs to the address created by changing bit[0] of the loaded value to 0
• if the instruction is conditional, it must be the last instruction in the IT block.
For store instructions:
• Rt can be SP for word stores only
• Rt must not be PC
• Rn must not be PC
• Rn must be different from Rt and Rt2 in the pre-indexed or post-indexed forms.

13.11.2.7

104

Condition flags
These instructions do not change the flags.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.11.2.8

Examples

LDR
LDRNE

R8, [R10]
R2, [R5, #960]!

STR

R2, [R9,#const-struc]

STRH

R3, [R4], #4

LDRD

R8, R9, [R3, #0x20]

STRD

R0, R1, [R8], #-16

;
;
;
;
;
;
;
;
;
;
;
;
;
;

Loads R8 from the address in R10.
Loads (conditionally) R2 from a word
960 bytes above the address in R5, and
increments R5 by 960.
const-struc is an expression evaluating
to a constant in the range 0-4095.
Store R3 as halfword data into address in
R4, then increment R4 by 4
Load R8 from a word 32 bytes above the
address in R3, and load R9 from a word 36
bytes above the address in R3
Store R0 to address in R8, and store R1 to
a word 4 bytes above the address in R8,
and then decrement R8 by 16.

105
6430F–ATARM–21-Feb-12

13.11.3

13.11.3.1

LDR and STR, register offset
Load and Store with register offset.
Syntax
op{type}{cond} Rt, [Rn, Rm {, LSL #n}]

where:
op

is one of:
LDR

Load Register.

STR

Store Register.

type

13.11.3.2

is one of:
B

unsigned byte, zero extend to 32 bits on loads.

signed byte, sign extend to 32 bits (LDR only).

unsigned halfword, zero extend to 32 bits on loads.

signed halfword, sign extend to 32 bits (LDR only).

omit, for word.

cond

is an optional condition code, see “Conditional execution” on page 98.

is the register to load or store.

is the register on which the memory address is based.

is a register containing a value to be used as the offset.

LSL #n

is an optional shift, with n in the range 0 to 3.

Operation
LDR instructions load a register with a value from memory.
STR instructions store a register value into memory.
The memory address to load from or store to is at an offset from the register Rn. The offset is
specified by the register Rm and can be shifted left by up to 3 bits using LSL.
The value to load or store can be a byte, halfword, or word. For load instructions, bytes and halfwords can either be signed or unsigned. See “Address alignment” on page 97.

13.11.3.3

Restrictions
In these instructions:
• Rn must not be PC
• Rm must not be SP and must not be PC
• Rt can be SP only for word loads and word stores
• Rt can be PC only for word loads.
When Rt is PC in a word load instruction:
• bit[0] of the loaded value must be 1 for correct execution, and a branch occurs to this
halfword-aligned address
• if the instruction is conditional, it must be the last instruction in the IT block.

106

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.11.3.4

Condition flags
These instructions do not change the flags.

13.11.3.5

Examples

STR
LDRSB

STR

R0, [R5, R1]

;
;
R0, [R5, R1, LSL #1] ;
;
;
R0, [R1, R2, LSL #2] ;
;

Store value of R0 into an address equal to
sum of R5 and R1
Read byte value from an address equal to
sum of R5 and two times R1, sign extended it
to a word value and put it in R0
Stores R0 to an address equal to sum of R1
and four times R2

107
6430F–ATARM–21-Feb-12

13.11.4

LDR and STR, unprivileged
Load and Store with unprivileged access.

13.11.4.1

Syntax
op{type}T{cond} Rt, [Rn {, #offset}]

; immediate offset

where:
op

is one of:
LDR

Load Register.

STR

Store Register.

type is one of:
B

unsigned byte, zero extend to 32 bits on loads.

signed byte, sign extend to 32 bits (LDR only).

unsigned halfword, zero extend to 32 bits on loads.

signed halfword, sign extend to 32 bits (LDR only).

omit, for word.

cond

is an optional condition code, see “Conditional execution” on page 98.

is the register to load or store.

is the register on which the memory address is based.

offset

is an offset from Rn and can be 0 to 255.
If offset is omitted, the address is the value in Rn.

13.11.4.2

Operation
These load and store instructions perform the same function as the memory access instructions
with immediate offset, see “LDR and STR, immediate offset” on page 103. The difference is that
these instructions have only unprivileged access even when used in privileged software.
When used in unprivileged software, these instructions behave in exactly the same way as normal memory access instructions with immediate offset.

13.11.4.3

Restrictions
In these instructions:
• Rn must not be PC
• Rt must not be SP and must not be PC.

13.11.4.4

Condition flags
These instructions do not change the flags.

13.11.4.5

Examples

108

STRBTEQ

R4, [R7]

LDRHT

R2, [R2, #8]

;
;
;
;

Conditionally store least significant byte in
R4 to an address in R7, with unprivileged access
Load halfword value from an address equal to
sum of R2 and 8 into R2, with unprivileged access

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.11.5

13.11.5.1

LDR, PC-relative
Load register from memory.
Syntax
LDR{type}{cond} Rt, label
LDRD{cond} Rt, Rt2, label

; Load two words

where:
type

13.11.5.2

is one of:
B

unsigned byte, zero extend to 32 bits.

signed byte, sign extend to 32 bits.

unsigned halfword, zero extend to 32 bits.

signed halfword, sign extend to 32 bits.

omit, for word.

cond

is an optional condition code, see “Conditional execution” on page 98.

is the register to load or store.

Rt2

is the second register to load or store.

label

is a PC-relative expression. See “PC-relative expressions” on page 98.

Operation
LDR loads a register with a value from a PC-relative memory address. The memory address is
specified by a label or by an offset from the PC.
The value to load or store can be a byte, halfword, or word. For load instructions, bytes and halfwords can either be signed or unsigned. See “Address alignment” on page 97.
label must be within a limited range of the current instruction. Table 13-19 shows the possible
offsets between label and the PC.
Table 13-19. Offset ranges
Instruction type

Offset range

Word, halfword, signed halfword, byte, signed
byte

−4095 to 4095

Two words

−1020 to 1020

You might have to use the .W suffix to get the maximum offset range. See “Instruction width
selection” on page 100.
13.11.5.3

Restrictions
In these instructions:
• Rt can be SP or PC only for word loads
• Rt2 must not be SP and must not be PC
• Rt must be different from Rt2.
When Rt is PC in a word load instruction:

109
6430F–ATARM–21-Feb-12

• bit[0] of the loaded value must be 1 for correct execution, and a branch occurs to this
halfword-aligned address
• if the instruction is conditional, it must be the last instruction in the IT block.
13.11.5.4

Condition flags
These instructions do not change the flags.

13.11.5.5

Examples

110

LDR

R0, LookUpTable

LDRSB

R7, localdata

;
;
;
;
;

Load R0 with a word of data from an address
labelled as LookUpTable
Load a byte value from an address labelled
as localdata, sign extend it to a word
value, and put it in R7

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.11.6

LDM and STM
Load and Store Multiple registers.

13.11.6.1

Syntax
op{addr_mode}{cond} Rn{!}, reglist

where:
op

is one of:
LDM

Load Multiple registers.

STM

Store Multiple registers.

addr_mode

is any one of the following:

Increment address After each access. This is the default.

Decrement address Before each access.

cond

is an optional condition code, see “Conditional execution” on page 98.

is the register on which the memory addresses are based.

is an optional writeback suffix.
If ! is present the final address, that is loaded from or stored to, is written back into Rn.

reglist
is a list of one or more registers to be loaded or stored, enclosed in braces. It can
contain register ranges. It must be comma separated if it contains more than one register or register range, see “Examples” on page 112.
LDM and LDMFD are synonyms for LDMIA. LDMFD refers to its use for popping data from Full
Descending stacks.
LDMEA is a synonym for LDMDB, and refers to its use for popping data from Empty Ascending
stacks.
STM and STMEA are synonyms for STMIA. STMEA refers to its use for pushing data onto
Empty Ascending stacks.
STMFD is s synonym for STMDB, and refers to its use for pushing data onto Full Descending
stacks
13.11.6.2

Operation
LDM instructions load the registers in reglist with word values from memory addresses based on
Rn.
STM instructions store the word values in the registers in reglist to memory addresses based on
Rn.
For LDM, LDMIA, LDMFD, STM, STMIA, and STMEA the memory addresses used for the
accesses are at 4-byte intervals ranging from Rn to Rn + 4 * (n-1), where n is the number of registers in reglist. The accesses happens in order of increasing register numbers, with the lowest
numbered register using the lowest memory address and the highest number register using the
highest memory address. If the writeback suffix is specified, the value of Rn + 4 * (n-1) is written
back to Rn.
For LDMDB, LDMEA, STMDB, and STMFD the memory addresses used for the accesses are at
4-byte intervals ranging from Rn to Rn - 4 * (n-1), where n is the number of registers in reglist.

111
6430F–ATARM–21-Feb-12

The accesses happen in order of decreasing register numbers, with the highest numbered register using the highest memory address and the lowest number register using the lowest memory
address. If the writeback suffix is specified, the value of Rn - 4 * (n-1) is written back to Rn.
The PUSH and POP instructions can be expressed in this form. See “PUSH and POP” on page
113 for details.
13.11.6.3

Restrictions
In these instructions:
• Rn must not be PC
• reglist must not contain SP
• in any STM instruction, reglist must not contain PC
• in any LDM instruction, reglist must not contain PC if it contains LR
• reglist must not contain Rn if you specify the writeback suffix.
When PC is in reglist in an LDM instruction:
• bit[0] of the value loaded to the PC must be 1 for correct execution, and a branch occurs to
this halfword-aligned address
• if the instruction is conditional, it must be the last instruction in the IT block.

13.11.6.4

Condition flags
These instructions do not change the flags.

13.11.6.5

Examples

LDM
STMDB

13.11.6.6

Incorrect examples

STM
LDM

112

R8,{R0,R2,R9}
; LDMIA is a synonym for LDM
R1!,{R3-R6,R11,R12}

R5!,{R5,R4,R9} ; Value stored for R5 is unpredictable
R2, {}
; There must be at least one register in the list

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.11.7

PUSH and POP
Push registers onto, and pop registers off a full-descending stack.

13.11.7.1

Syntax
PUSH{cond} reglist
POP{cond} reglist

where:
cond

is an optional condition code, see “Conditional execution” on page 98.

reglist
is a non-empty list of registers, enclosed in braces. It can contain register ranges.
It must be comma separated if it contains more than one register or register range.
PUSH and POP are synonyms for STMDB and LDM (or LDMIA) with the memory addresses for
the access based on SP, and with the final address for the access written back to the SP. PUSH
and POP are the preferred mnemonics in these cases.
13.11.7.2

Operation
PUSH stores registers on the stack in order of decreasing the register numbers, with the highest
numbered register using the highest memory address and the lowest numbered register using
the lowest memory address.
POP loads registers from the stack in order of increasing register numbers, with the lowest numbered register using the lowest memory address and the highest numbered register using the
highest memory address.
See “LDM and STM” on page 111 for more information.

13.11.7.3

Restrictions
In these instructions:
• reglist must not contain SP
• for the PUSH instruction, reglist must not contain PC
• for the POP instruction, reglist must not contain PC if it contains LR.
When PC is in reglist in a POP instruction:
• bit[0] of the value loaded to the PC must be 1 for correct execution, and a branch occurs to
this halfword-aligned address
• if the instruction is conditional, it must be the last instruction in the IT block.

13.11.7.4

Condition flags
These instructions do not change the flags.

13.11.7.5

Examples

PUSH
PUSH
POP

{R0,R4-R7}
{R2,LR}
{R0,R10,PC}

113
6430F–ATARM–21-Feb-12

13.11.8

13.11.8.1

LDREX and STREX
Load and Store Register Exclusive.
Syntax
LDREX{cond} Rt, [Rn {, #offset}]
STREX{cond} Rd, Rt, [Rn {, #offset}]
LDREXB{cond} Rt, [Rn]
STREXB{cond} Rd, Rt, [Rn]
LDREXH{cond} Rt, [Rn]
STREXH{cond} Rd, Rt, [Rn]

where:
cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register for the returned status.

is the register to load or store.

is the register on which the memory address is based.

offset

is an optional offset applied to the value in Rn.
If offset is omitted, the address is the value in Rn.

13.11.8.2

Operation
LDREX, LDREXB, and LDREXH load a word, byte, and halfword respectively from a memory
address.
STREX, STREXB, and STREXH attempt to store a word, byte, and halfword respectively to a
memory address. The address used in any Store-Exclusive instruction must be the same as the
address in the most recently executed Load-exclusive instruction. The value stored by the StoreExclusive instruction must also have the same data size as the value loaded by the preceding
Load-exclusive instruction. This means software must always use a Load-exclusive instruction
and a matching Store-Exclusive instruction to perform a synchronization operation, see “Synchronization primitives” on page 75
If an Store-Exclusive instruction performs the store, it writes 0 to its destination register. If it does
not perform the store, it writes 1 to its destination register. If the Store-Exclusive instruction
writes 0 to the destination register, it is guaranteed that no other process in the system has
accessed the memory location between the Load-exclusive and Store-Exclusive instructions.
For reasons of performance, keep the number of instructions between corresponding LoadExclusive and Store-Exclusive instruction to a minimum.
The result of executing a Store-Exclusive instruction to an address that is different from that
used in the preceding Load-Exclusive instruction is unpredictable.

13.11.8.3

Restrictions
In these instructions:
• do not use PC
• do not use SP for Rd and Rt
• for STREX, Rd must be different from both Rt and Rn
• the value of offset must be a multiple of four in the range 0-1020.

114

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.11.8.4

Condition flags
These instructions do not change the flags.

13.11.8.5

Examples

MOV

R1, #0x1

; Initialize the ‘lock taken’ value

LDREX
CMP
ITT
STREXEQ
CMPEQ
BNE
....

R0,
R0,
EQ
R0,
R0,
try

;
;
;
;
;
;
;

try
[LockAddr]
#0
R1, [LockAddr]
#0

Load the lock value
Is the lock free?
IT instruction for STREXEQ and CMPEQ
Try and claim the lock
Did this succeed?
No – try again
Yes – we have the lock

115
6430F–ATARM–21-Feb-12

13.11.9

CLREX
Clear Exclusive.

13.11.9.1

Syntax
CLREX{cond}

where:
cond
13.11.9.2

is an optional condition code, see “Conditional execution” on page 98.

Operation
Use CLREX to make the next STREX, STREXB, or STREXH instruction write 1 to its destination
register and fail to perform the store. It is useful in exception handler code to force the failure of
the store exclusive if the exception occurs between a load exclusive instruction and the matching store exclusive instruction in a synchronization operation.
See “Synchronization primitives” on page 75 for more information.

13.11.9.3

Condition flags
These instructions do not change the flags.

13.11.9.4

Examples

CLREX

116

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.12 General data processing instructions
Table 13-20 shows the data processing instructions:
Table 13-20. Data processing instructions
Mnemonic

Brief description

See

ADC

Add with Carry

“ADD, ADC, SUB, SBC, and RSB” on
page 119

ADD

Add

“ADD, ADC, SUB, SBC, and RSB” on
page 119

ADDW

Add

“ADD, ADC, SUB, SBC, and RSB” on
page 119

AND

Logical AND

“AND, ORR, EOR, BIC, and ORN” on
page 122

ASR

Arithmetic Shift Right

“ASR, LSL, LSR, ROR, and RRX” on page
124

BIC

Bit Clear

“AND, ORR, EOR, BIC, and ORN” on
page 122

CLZ

Count leading zeros

“CLZ” on page 126

CMN

Compare Negative

“CMP and CMN” on page 127

CMP

Compare

“CMP and CMN” on page 127

EOR

Exclusive OR

“AND, ORR, EOR, BIC, and ORN” on
page 122

LSL

Logical Shift Left

“ASR, LSL, LSR, ROR, and RRX” on page
124

LSR

Logical Shift Right

“ASR, LSL, LSR, ROR, and RRX” on page
124

MOV

Move

“MOV and MVN” on page 128

MOVT

Move Top

“MOVT” on page 130

MOVW

Move 16-bit constant

“MOV and MVN” on page 128

MVN

Move NOT

“MOV and MVN” on page 128

ORN

Logical OR NOT

“AND, ORR, EOR, BIC, and ORN” on
page 122

ORR

Logical OR

“AND, ORR, EOR, BIC, and ORN” on
page 122

RBIT

Reverse Bits

“REV, REV16, REVSH, and RBIT” on
page 131

REV

Reverse byte order in a word

“REV, REV16, REVSH, and RBIT” on
page 131

REV16

Reverse byte order in each halfword

“REV, REV16, REVSH, and RBIT” on
page 131

REVSH

Reverse byte order in bottom halfword and
sign extend

“REV, REV16, REVSH, and RBIT” on
page 131

ROR

Rotate Right

“ASR, LSL, LSR, ROR, and RRX” on page
124

117
6430F–ATARM–21-Feb-12

Table 13-20. Data processing instructions (Continued)

118

Mnemonic

Brief description

See

RRX

Rotate Right with Extend

“ASR, LSL, LSR, ROR, and RRX” on page
124

RSB

Reverse Subtract

“ADD, ADC, SUB, SBC, and RSB” on
page 119

SBC

Subtract with Carry

“ADD, ADC, SUB, SBC, and RSB” on
page 119

SUB

Subtract

“ADD, ADC, SUB, SBC, and RSB” on
page 119

SUBW

Subtract

“ADD, ADC, SUB, SBC, and RSB” on
page 119

TEQ

Test Equivalence

“TST and TEQ” on page 132

TST

Test

“TST and TEQ” on page 132

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.12.1

13.12.1.1

ADD, ADC, SUB, SBC, and RSB
Add, Add with carry, Subtract, Subtract with carry, and Reverse Subtract.
Syntax
op{S}{cond} {Rd,} Rn, Operand2
op{cond} {Rd,} Rn, #imm12

; ADD and SUB only

where:
op

is one of:
ADD

Add.

ADC

Add with Carry.

SUB

Subtract.

SBC

Subtract with Carry.

RSB

Reverse Subtract.

S
is an optional suffix. If S is specified, the condition code flags are updated on the
result of the operation, see “Conditional execution” on page 98.
cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register. If Rd is omitted, the destination register is Rn.

is the register holding the first operand.

Operand2

is a flexible second operand.
See “Flexible second operand” on page 94 for details of the options.

imm12
13.12.1.2

is any value in the range 0-4095.

Operation
The ADD instruction adds the value of Operand2 or imm12 to the value in Rn.
The ADC instruction adds the values in Rn and Operand2, together with the carry flag.
The SUB instruction subtracts the value of Operand2 or imm12 from the value in Rn.
The SBC instruction subtracts the value of Operand2 from the value in Rn. If the carry flag is
clear, the result is reduced by one.
The RSB instruction subtracts the value in Rn from the value of Operand2. This is useful
because of the wide range of options for Operand2.
Use ADC and SBC to synthesize multiword arithmetic, see “Multiword arithmetic examples” on
page 121.
See also “ADR” on page 102.
ADDW is equivalent to the ADD syntax that uses the imm12 operand. SUBW is equivalent to the
SUB syntax that uses the imm12 operand.

13.12.1.3

Restrictions
In these instructions:
• Operand2 must not be SP and must not be PC

119
6430F–ATARM–21-Feb-12

• Rd can be SP only in ADD and SUB, and only with the additional restrictions:
– Rn must also be SP
– any shift in Operand2 must be limited to a maximum of 3 bits using LSL
• Rn can be SP only in ADD and SUB
• Rd can be PC only in the ADD{cond} PC, PC, Rm instruction where:
– you must not specify the S suffix
– Rm must not be PC and must not be SP
– if the instruction is conditional, it must be the last instruction in the IT block
• with the exception of the ADD{cond} PC, PC, Rm instruction, Rn can be PC only in ADD and
SUB, and only with the additional restrictions:
– you must not specify the S suffix
– the second operand must be a constant in the range 0 to 4095.
–
– When using the PC for an addition or a subtraction, bits[1:0] of the PC are rounded
to b00 before performing the calculation, making the base address for the calculation
word-aligned.
– If you want to generate the address of an instruction, you have to adjust the constant
based on the value of the PC. ARM recommends that you use the ADR instruction
instead of ADD or SUB with Rn equal to the PC, because your assembler
automatically calculates the correct constant for the ADR instruction.
When Rd is PC in the ADD{cond} PC, PC, Rm instruction:
• bit[0] of the value written to the PC is ignored
• a branch occurs to the address created by forcing bit[0] of that value to 0.
13.12.1.4

Condition flags
If S is specified, these instructions update the N, Z, C and V flags according to the result.

13.12.1.5

Examples

ADD
SUBS
RSB
ADCHI

120

R2, R1, R3
R8, R6, #240
R4, R4, #1280
R11, R0, R3

;
;
;
;

Sets the flags on the result
Subtracts contents of R4 from 1280
Only executed if C flag set and Z
flag clear

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.12.1.6

Multiword arithmetic examples

13.12.1.7
64-bit addition
The example below shows two instructions that add a 64-bit integer contained in R2 and R3 to another 64-bit integer contained in R0 and R1, and place the result in R4 and R5.
ADDS
ADC

R4, R0, R2
R5, R1, R3

; add the least significant words
; add the most significant words with carry

13.12.1.8
96-bit subtraction
Multiword values do not have to use consecutive registers. The example below shows instructions that subtract a 96-bit
integer contained in R9, R1, and R11 from another contained in R6, R2, and R8. The example stores the result in R6, R9,
and R2.
SUBS
SBCS
SBC

R6, R6, R9
R9, R2, R1
R2, R8, R11

; subtract the least significant words
; subtract the middle words with carry
; subtract the most significant words with carry

121
6430F–ATARM–21-Feb-12

13.12.2

13.12.2.1

AND, ORR, EOR, BIC, and ORN
Logical AND, OR, Exclusive OR, Bit Clear, and OR NOT.
Syntax
op{S}{cond} {Rd,} Rn, Operand2

where:
op

is one of:
AND

logical AND.

ORR

logical OR, or bit set.

EOR

logical Exclusive OR.

BIC

logical AND NOT, or bit clear.

ORN

logical OR NOT.

S
is an optional suffix. If S is specified, the condition code flags are updated on the
result of the operation, see “Conditional execution” on page 98.
cond

is an optional condition code, see See “Conditional execution” on page 98..

is the destination register.

is the register holding the first operand.

Operand2
is a flexible second operand. See “Flexible second operand” on page 94 for
details of the options.
13.12.2.2

Operation
The AND, EOR, and ORR instructions perform bitwise AND, Exclusive OR, and OR operations
on the values in Rn and Operand2.
The BIC instruction performs an AND operation on the bits in Rn with the complements of the
corresponding bits in the value of Operand2.
The ORN instruction performs an OR operation on the bits in Rn with the complements of the
corresponding bits in the value of Operand2.

13.12.2.3

Restrictions
Do not use SP and do not use PC.

13.12.2.4

Condition flags
If S is specified, these instructions:
• update the N and Z flags according to the result
• can update the C flag during the calculation of Operand2, see “Flexible second operand” on
page 94
• do not affect the V flag.

122

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.12.2.5

Examples

AND
ORREQ
ANDS
EORS
BIC
ORN
ORNS

R9,
R2,
R9,
R7,
R0,
R7,
R7,

R2, #0xFF00
R0, R5
R8, #0x19
R11, #0x18181818
R1, #0xab
R11, R14, ROR #4
R11, R14, ASR #32

123
6430F–ATARM–21-Feb-12

13.12.3

13.12.3.1

ASR, LSL, LSR, ROR, and RRX
Arithmetic Shift Right, Logical Shift Left, Logical Shift Right, Rotate Right, and Rotate Right with
Extend.
Syntax
op{S}{cond} Rd, Rm, Rs
op{S}{cond} Rd, Rm, #n
RRX{S}{cond} Rd, Rm

where:
op

is one of:
ASR

Arithmetic Shift Right.

LSL

Logical Shift Left.

LSR

Logical Shift Right.

ROR

Rotate Right.

S
is an optional suffix. If S is specified, the condition code flags are updated on the
result of the operation, see “Conditional execution” on page 98.
Rd

is the destination register.

is the register holding the value to be shifted.

Rs
is the register holding the shift length to apply to the value in Rm. Only the least
significant byte is used and can be in the range 0 to 255.
n

is the shift length. The range of shift length depends on the instruction:
ASR

shift length from 1 to 32

LSL

shift length from 0 to 31

LSR

shift length from 1 to 32

ROR

shift length from 1 to 31.

MOV{S}{cond} Rd, Rm is the preferred syntax for LSL{S}{cond} Rd, Rm, #0.

13.12.3.2

Operation
ASR, LSL, LSR, and ROR move the bits in the register Rm to the left or right by the number of
places specified by constant n or register Rs.
RRX moves the bits in register Rm to the right by 1.
In all these instructions, the result is written to Rd, but the value in register Rm remains
unchanged. For details on what result is generated by the different instructions, see “Shift Operations” on page 95.

13.12.3.3

Restrictions
Do not use SP and do not use PC.

13.12.3.4

Condition flags
If S is specified:
• these instructions update the N and Z flags according to the result

124

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• the C flag is updated to the last bit shifted out, except when the shift length is 0, see “Shift
Operations” on page 95.
13.12.3.5

Examples

ASR
LSLS
LSR
ROR
RRX

R7,
R1,
R4,
R4,
R4,

R8,
R2,
R5,
R5,
R5

#9
#3
#6
R6

;
;
;
;
;

Arithmetic shift right by 9 bits
Logical shift left by 3 bits with flag update
Logical shift right by 6 bits
Rotate right by the value in the bottom byte of R6
Rotate right with extend

125
6430F–ATARM–21-Feb-12

13.12.4

CLZ
Count Leading Zeros.

13.12.4.1

Syntax
CLZ{cond} Rd, Rm

where:

13.12.4.2

cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register.

is the operand register.

Operation
The CLZ instruction counts the number of leading zeros in the value in Rm and returns the result
in Rd. The result value is 32 if no bits are set in the source register, and zero if bit[31] is set.

13.12.4.3

Restrictions
Do not use SP and do not use PC.

13.12.4.4

Condition flags
This instruction does not change the flags.

13.12.4.5

Examples

CLZ
CLZNE

126

R4,R9
R2,R3

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.12.5

CMP and CMN
Compare and Compare Negative.

13.12.5.1

Syntax
CMP{cond} Rn, Operand2
CMN{cond} Rn, Operand2

where:
cond

is an optional condition code, see “Conditional execution” on page 98.

is the register holding the first operand.

Operand2
is a flexible second operand. See “Flexible second operand” on page 94 for
details of the options.
13.12.5.2

Operation
These instructions compare the value in a register with Operand2. They update the condition
flags on the result, but do not write the result to a register.
The CMP instruction subtracts the value of Operand2 from the value in Rn. This is the same as
a SUBS instruction, except that the result is discarded.
The CMN instruction adds the value of Operand2 to the value in Rn. This is the same as an
ADDS instruction, except that the result is discarded.

13.12.5.3

Restrictions
In these instructions:
• do not use PC
• Operand2 must not be SP.

13.12.5.4

Condition flags
These instructions update the N, Z, C and V flags according to the result.

13.12.5.5

Examples

CMP
CMN
CMPGT

R2, R9
R0, #6400
SP, R7, LSL #2

127
6430F–ATARM–21-Feb-12

13.12.6

MOV and MVN
Move and Move NOT.

13.12.6.1

Syntax
MOV{S}{cond} Rd, Operand2
MOV{cond} Rd, #imm16
MVN{S}{cond} Rd, Operand2

where:
S
is an optional suffix. If S is specified, the condition code flags are updated on the
result of the operation, see “Conditional execution” on page 98.
cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register.

Operand2
is a flexible second operand. See “Flexible second operand” on page 94 for
details of the options.
imm16
13.12.6.2

is any value in the range 0-65535.

Operation
The MOV instruction copies the value of Operand2 into Rd.
When Operand2 in a MOV instruction is a register with a shift other than LSL #0, the preferred
syntax is the corresponding shift instruction:
• ASR{S}{cond} Rd, Rm, #n is the preferred syntax for MOV{S}{cond} Rd, Rm, ASR #n
• LSL{S}{cond} Rd, Rm, #n is the preferred syntax for MOV{S}{cond} Rd, Rm, LSL #n if n != 0
• LSR{S}{cond} Rd, Rm, #n is the preferred syntax for MOV{S}{cond} Rd, Rm, LSR #n
• ROR{S}{cond} Rd, Rm, #n is the preferred syntax for MOV{S}{cond} Rd, Rm, ROR #n
• RRX{S}{cond} Rd, Rm is the preferred syntax for MOV{S}{cond} Rd, Rm, RRX.
Also, the MOV instruction permits additional forms of Operand2 as synonyms for shift
instructions:
• MOV{S}{cond} Rd, Rm, ASR Rs is a synonym for ASR{S}{cond} Rd, Rm, Rs
• MOV{S}{cond} Rd, Rm, LSL Rs is a synonym for LSL{S}{cond} Rd, Rm, Rs
• MOV{S}{cond} Rd, Rm, LSR Rs is a synonym for LSR{S}{cond} Rd, Rm, Rs
• MOV{S}{cond} Rd, Rm, ROR Rs is a synonym for ROR{S}{cond} Rd, Rm, Rs
See “ASR, LSL, LSR, ROR, and RRX” on page 124.
The MVN instruction takes the value of Operand2, performs a bitwise logical NOT operation on
the value, and places the result into Rd.
The MOVW instruction provides the same function as MOV, but is restricted to using the imm16
operand.

13.12.6.3

Restrictions
You can use SP and PC only in the MOV instruction, with the following restrictions:
• the second operand must be a register without shift
• you must not specify the S suffix.
When Rd is PC in a MOV instruction:

128

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• bit[0] of the value written to the PC is ignored
• a branch occurs to the address created by forcing bit[0] of that value to 0.
Though it is possible to use MOV as a branch instruction, ARM strongly recommends the use of
a BX or BLX instruction to branch for software portability to the ARM instruction set.

13.12.6.4

13.12.6.5

Example

MOVS
MOV
MOVS
MOV
MOV
MVNS

R11, #0x000B
R1, #0xFA05
R10, R12
R3, #23
R8, SP
R2, #0xF

;
;
;
;
;
;
;

Write value of 0x000B to R11, flags get updated
Write value of 0xFA05 to R1, flags are not updated
Write value in R12 to R10, flags get updated
Write value of 23 to R3
Write value of stack pointer to R8
Write value of 0xFFFFFFF0 (bitwise inverse of 0xF)
to the R2 and update flags

129
6430F–ATARM–21-Feb-12

13.12.7

MOVT
Move Top.

13.12.7.1

Syntax
MOVT{cond} Rd, #imm16

where:

13.12.7.2

cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register.

imm16

is a 16-bit immediate constant.

Operation
MOVT writes a 16-bit immediate value, imm16, to the top halfword, Rd[31:16], of its destination
register. The write does not affect Rd[15:0].
The MOV, MOVT instruction pair enables you to generate any 32-bit constant.

13.12.7.3

Restrictions
Rd must not be SP and must not be PC.

13.12.7.4

Condition flags
This instruction does not change the flags.

13.12.7.5

Examples

MOVT

130

R3, #0xF123 ; Write 0xF123 to upper halfword of R3, lower halfword
; and APSR are unchanged

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.12.8

REV, REV16, REVSH, and RBIT
Reverse bytes and Reverse bits.

13.12.8.1

Syntax
op{cond} Rd, Rn

where:
op

is any of:
REV

Reverse byte order in a word.

REV16 Reverse byte order in each halfword independently.
REVSH Reverse byte order in the bottom halfword, and sign extend to 32 bits.
RBIT

13.12.8.2

Reverse the bit order in a 32-bit word.

cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register.

is the register holding the operand.

Operation
Use these instructions to change endianness of data:
REV
converts 32-bit big-endian data into little-endian data or 32-bit little-endian data
into big-endian data.
REV16
converts 16-bit big-endian data into little-endian data or 16-bit little-endian data
into big-endian data.
REVSH

converts either:

16-bit signed big-endian data into 32-bit signed little-endian data
16-bit signed little-endian data into 32-bit signed big-endian data.
13.12.8.3

Restrictions
Do not use SP and do not use PC.

13.12.8.4

Condition flags
These instructions do not change the flags.

13.12.8.5

Examples

REV
REV16
REVSH
REVHS
RBIT

R3,
R0,
R0,
R3,
R7,

R7
R0
R5
R7
R8

;
;
;
;
;

Reverse
Reverse
Reverse
Reverse
Reverse

byte order of value in R7 and write it to R3
byte order of each 16-bit halfword in R0
Signed Halfword
with Higher or Same condition
bit order of value in R8 and write the result to R7

131
6430F–ATARM–21-Feb-12

13.12.9

TST and TEQ
Test bits and Test Equivalence.

13.12.9.1

Syntax
TST{cond} Rn, Operand2
TEQ{cond} Rn, Operand2

where:
cond

is an optional condition code, see “Conditional execution” on page 98.

is the register holding the first operand.

Operand2
is a flexible second operand. See “Flexible second operand” on page 94 for
details of the options.
13.12.9.2

Operation
These instructions test the value in a register against Operand2. They update the condition flags
based on the result, but do not write the result to a register.
The TST instruction performs a bitwise AND operation on the value in Rn and the value of
Operand2. This is the same as the ANDS instruction, except that it discards the result.
To test whether a bit of Rn is 0 or 1, use the TST instruction with an Operand2 constant that has
that bit set to 1 and all other bits cleared to 0.
The TEQ instruction performs a bitwise Exclusive OR operation on the value in Rn and the value
of Operand2. This is the same as the EORS instruction, except that it discards the result.
Use the TEQ instruction to test if two values are equal without affecting the V or C flags.
TEQ is also useful for testing the sign of a value. After the comparison, the N flag is the logical
Exclusive OR of the sign bits of the two operands.

13.12.9.3

Restrictions
Do not use SP and do not use PC.

13.12.9.4

Condition flags
These instructions:
• update the N and Z flags according to the result
• can update the C flag during the calculation of Operand2, see “Flexible second operand” on
page 94
• do not affect the V flag.

13.12.9.5

132

Examples

TST

R0, #0x3F8

TEQEQ

R10, R9

;
;
;
;

Perform bitwise AND of R0 value to 0x3F8,
APSR is updated but result is discarded
Conditionally test if value in R10 is equal to
value in R9, APSR is updated but result is discarded

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.13 Multiply and divide instructions
Table 13-21 shows the multiply and divide instructions:
Table 13-21. Multiply and divide instructions
Mnemonic

Brief description

See

MLA

Multiply with Accumulate, 32-bit result

“MUL, MLA, and MLS” on page 134

MLS

Multiply and Subtract, 32-bit result

“MUL, MLA, and MLS” on page 134

MUL

Multiply, 32-bit result

“MUL, MLA, and MLS” on page 134

SDIV

Signed Divide

“SDIV and UDIV” on page 136

SMLAL

Signed Multiply with Accumulate
(32x32+64), 64-bit result

“UMULL, UMLAL, SMULL, and SMLAL” on
page 135

SMULL

Signed Multiply (32x32), 64-bit result

“UMULL, UMLAL, SMULL, and SMLAL” on
page 135

UDIV

Unsigned Divide

“SDIV and UDIV” on page 136

UMLAL

Unsigned Multiply with Accumulate
(32x32+64), 64-bit result

“UMULL, UMLAL, SMULL, and SMLAL” on
page 135

UMULL

Unsigned Multiply (32x32), 64-bit
result

“UMULL, UMLAL, SMULL, and SMLAL” on
page 135

133
6430F–ATARM–21-Feb-12

13.13.1

MUL, MLA, and MLS
Multiply, Multiply with Accumulate, and Multiply with Subtract, using 32-bit operands, and producing a 32-bit result.

13.13.1.1

Syntax
MUL{S}{cond} {Rd,} Rn, Rm ; Multiply
MLA{cond} Rd, Rn, Rm, Ra

; Multiply with accumulate

MLS{cond} Rd, Rn, Rm, Ra

; Multiply with subtract

where:
cond

is an optional condition code, see “Conditional execution” on page 98.

S
is an optional suffix. If S is specified, the condition code flags are updated on the
result of the operation, see “Conditional execution” on page 98.

13.13.1.2

is the destination register. If Rd is omitted, the destination register is Rn.

Rn, Rm

are registers holding the values to be multiplied.

is a register holding the value to be added or subtracted from.

Operation
The MUL instruction multiplies the values from Rn and Rm, and places the least significant 32
bits of the result in Rd.
The MLA instruction multiplies the values from Rn and Rm, adds the value from Ra, and places
the least significant 32 bits of the result in Rd.
The MLS instruction multiplies the values from Rn and Rm, subtracts the product from the value
from Ra, and places the least significant 32 bits of the result in Rd.
The results of these instructions do not depend on whether the operands are signed or
unsigned.

13.13.1.3

Restrictions
In these instructions, do not use SP and do not use PC.
If you use the S suffix with the MUL instruction:
• Rd, Rn, and Rm must all be in the range R0 to R7
• Rd must be the same as Rm
• you must not use the cond suffix.

13.13.1.4

Condition flags
If S is specified, the MUL instruction:
• updates the N and Z flags according to the result
• does not affect the C and V flags.

13.13.1.5

Examples

MUL
MLA
MULS
MULLT
MLS

134

R10, R2, R5
R10, R2, R1, R5
R0, R2, R2
R2, R3, R2
R4, R5, R6, R7

;
;
;
;
;

Multiply, R10
Multiply with
Multiply with
Conditionally
Multiply with

= R2 x R5
accumulate, R10 =
flag update, R0 =
multiply, R2 = R3
subtract, R4 = R7

(R2 x R1) + R5
R2 x R2
x R2
- (R5 x R6)

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.13.2

UMULL, UMLAL, SMULL, and SMLAL
Signed and Unsigned Long Multiply, with optional Accumulate, using 32-bit operands and producing a 64-bit result.

13.13.2.1

Syntax
op{cond} RdLo, RdHi, Rn, Rm

where:
op

is one of:
UMULL Unsigned Long Multiply.
UMLAL Unsigned Long Multiply, with Accumulate.
SMULL Signed Long Multiply.
SMLAL Signed Long Multiply, with Accumulate.

cond

is an optional condition code, see “Conditional execution” on page 98.

RdHi, RdLo

are the destination registers.
For UMLAL and SMLAL they also hold the accumulating value.

Rn, Rm
13.13.2.2

are registers holding the operands.

Operation
The UMULL instruction interprets the values from Rn and Rm as unsigned integers. It multiplies
these integers and places the least significant 32 bits of the result in RdLo, and the most significant 32 bits of the result in RdHi.
The UMLAL instruction interprets the values from Rn and Rm as unsigned integers. It multiplies
these integers, adds the 64-bit result to the 64-bit unsigned integer contained in RdHi and RdLo,
and writes the result back to RdHi and RdLo.
The SMULL instruction interprets the values from Rn and Rm as two’s complement signed integers. It multiplies these integers and places the least significant 32 bits of the result in RdLo, and
the most significant 32 bits of the result in RdHi.
The SMLAL instruction interprets the values from Rn and Rm as two’s complement signed integers. It multiplies these integers, adds the 64-bit result to the 64-bit signed integer contained in
RdHi and RdLo, and writes the result back to RdHi and RdLo.

13.13.2.3

Restrictions
In these instructions:
• do not use SP and do not use PC
• RdHi and RdLo must be different registers.

13.13.2.4

Condition flags
These instructions do not affect the condition code flags.

13.13.2.5

Examples

UMULL
SMLAL

R0, R4, R5, R6
R4, R5, R3, R8

; Unsigned (R4,R0) = R5 x R6
; Signed (R5,R4) = (R5,R4) + R3 x R8

135
6430F–ATARM–21-Feb-12

13.13.3

SDIV and UDIV
Signed Divide and Unsigned Divide.

13.13.3.1

Syntax
SDIV{cond} {Rd,} Rn, Rm
UDIV{cond} {Rd,} Rn, Rm

where:

13.13.3.2

cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register. If Rd is omitted, the destination register is Rn.

is the register holding the value to be divided.

is a register holding the divisor.

Operation
SDIV performs a signed integer division of the value in Rn by the value in Rm.
UDIV performs an unsigned integer division of the value in Rn by the value in Rm.
For both instructions, if the value in Rn is not divisible by the value in Rm, the result is rounded
towards zero.

13.13.3.3

Restrictions
Do not use SP and do not use PC.

13.13.3.4

Condition flags
These instructions do not change the flags.

13.13.3.5

Examples

SDIV
UDIV

136

R0, R2, R4
R8, R8, R1

; Signed divide, R0 = R2/R4
; Unsigned divide, R8 = R8/R1

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.14 Saturating instructions
This section describes the saturating instructions, SSAT and USAT.
13.14.1

13.14.1.1

SSAT and USAT
Signed Saturate and Unsigned Saturate to any bit position, with optional shift before saturating.
Syntax
op{cond} Rd, #n, Rm {, shift #s}

where:
op

is one of:
SSAT

Saturates a signed value to a signed range.

USAT

Saturates a signed value to an unsigned range.

cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register.

specifies the bit position to saturate to:
n ranges from 1 to 32 for SSAT
n ranges from 0 to 31 for USAT.

is the register containing the value to saturate.

shift #s

is an optional shift applied to Rm before saturating. It must be one of the following:

ASR #s where s is in the range 1 to 31
LSL #s where s is in the range 0 to 31.
13.14.1.2

Operation
These instructions saturate to a signed or unsigned n-bit value.
The SSAT instruction applies the specified shift, then saturates to the signed range −
2n–1 ≤x ≤2n–1−1.
The USAT instruction applies the specified shift, then saturates to the unsigned range 0 ≤x ≤2n−1.
For signed n-bit saturation using SSAT, this means that:
• if the value to be saturated is less than −2n−1, the result returned is −2n-1
• if the value to be saturated is greater than 2n−1−1, the result returned is 2n-1−1
• otherwise, the result returned is the same as the value to be saturated.
For unsigned n-bit saturation using USAT, this means that:
• if the value to be saturated is less than 0, the result returned is 0
• if the value to be saturated is greater than 2n−1, the result returned is 2n−1
• otherwise, the result returned is the same as the value to be saturated.
If the returned result is different from the value to be saturated, it is called saturation. If saturation occurs, the instruction sets the Q flag to 1 in the APSR. Otherwise, it leaves the Q flag
unchanged. To clear the Q flag to 0, you must use the MSR instruction, see “MSR” on page 158.
To read the state of the Q flag, use the MRS instruction, see “MRS” on page 157.

137
6430F–ATARM–21-Feb-12

13.14.1.3

Restrictions
Do not use SP and do not use PC.

13.14.1.4

Condition flags
These instructions do not affect the condition code flags.
If saturation occurs, these instructions set the Q flag to 1.

13.14.1.5

138

Examples

SSAT

R7, #16, R7, LSL #4

USATNE

R0, #7, R5

;
;
;
;
;

Logical shift left value in R7 by 4, then
saturate it as a signed 16-bit value and
write it back to R7
Conditionally saturate value in R5 as an
unsigned 7 bit value and write it to R0

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.15 Bitfield instructions
Table 13-22 shows the instructions that operate on adjacent sets of bits in registers or bitfields:
Table 13-22. Packing and unpacking instructions
Mnemonic

Brief description

See

BFC

Bit Field Clear

“BFC and BFI” on page 140

BFI

Bit Field Insert

“BFC and BFI” on page 140

SBFX

Signed Bit Field Extract

“SBFX and UBFX” on page 141

SXTB

Sign extend a byte

“SXT and UXT” on page 142

SXTH

Sign extend a halfword

“SXT and UXT” on page 142

UBFX

Unsigned Bit Field Extract

“SBFX and UBFX” on page 141

UXTB

Zero extend a byte

“SXT and UXT” on page 142

UXTH

Zero extend a halfword

“SXT and UXT” on page 142

139
6430F–ATARM–21-Feb-12

13.15.1

BFC and BFI
Bit Field Clear and Bit Field Insert.

13.15.1.1

Syntax
BFC{cond} Rd, #lsb, #width
BFI{cond} Rd, Rn, #lsb, #width

where:
cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register.

is the source register.

lsb

is the position of the least significant bit of the bitfield.
lsb must be in the range 0 to 31.

width
13.15.1.2

is the width of the bitfield and must be in the range 1 to 32−lsb.

Operation
BFC clears a bitfield in a register. It clears width bits in Rd, starting at the low bit position lsb.
Other bits in Rd are unchanged.
BFI copies a bitfield into one register from another register. It replaces width bits in Rd starting at
the low bit position lsb, with width bits from Rn starting at bit[0]. Other bits in Rd are unchanged.

13.15.1.3

Restrictions
Do not use SP and do not use PC.

13.15.1.4

Condition flags
These instructions do not affect the flags.

13.15.1.5

Examples

BFC
BFI

140

R4, #8, #12
R9, R2, #8, #12

; Clear bit 8 to bit 19 (12 bits) of R4 to 0
; Replace bit 8 to bit 19 (12 bits) of R9 with
; bit 0 to bit 11 from R2

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.15.2

SBFX and UBFX
Signed Bit Field Extract and Unsigned Bit Field Extract.

13.15.2.1

Syntax
SBFX{cond} Rd, Rn, #lsb, #width
UBFX{cond} Rd, Rn, #lsb, #width

where:
cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register.

is the source register.

lsb

is the position of the least significant bit of the bitfield.
lsb must be in the range 0 to 31.

width
13.15.2.2

is the width of the bitfield and must be in the range 1 to 32−lsb.

Operation
SBFX extracts a bitfield from one register, sign extends it to 32 bits, and writes the result to the
destination register.
UBFX extracts a bitfield from one register, zero extends it to 32 bits, and writes the result to the
destination register.

13.15.2.3

Restrictions
Do not use SP and do not use PC.

13.15.2.4

Condition flags
These instructions do not affect the flags.

13.15.2.5

Examples

SBFX
UBFX

R0, R1, #20, #4

;
;
R8, R11, #9, #10 ;
;

Extract bit 20 to bit 23 (4 bits) from R1 and sign
extend to 32 bits and then write the result to R0.
Extract bit 9 to bit 18 (10 bits) from R11 and zero
extend to 32 bits and then write the result to R8

141
6430F–ATARM–21-Feb-12

13.15.3

SXT and UXT
Sign extend and Zero extend.

13.15.3.1

Syntax
SXTextend{cond} {Rd,} Rm {, ROR #n}
UXTextend{cond} {Rd}, Rm {, ROR #n}

where:
extend

is one of:

Extends an 8-bit value to a 32-bit value.

Extends a 16-bit value to a 32-bit value.

cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register.

is the register holding the value to extend.

ROR #n

is one of:

ROR #8 Value from Rm is rotated right 8 bits.
ROR #16 Value from Rm is rotated right 16 bits.
ROR #24 Value from Rm is rotated right 24 bits.
If ROR #n is omitted, no rotation is performed.
13.15.3.2

Operation
These instructions do the following:
• Rotate the value from Rm right by 0, 8, 16 or 24 bits.
• Extract bits from the resulting value:
SXTB extracts bits[7:0] and sign extends to 32 bits.
UXTB extracts bits[7:0] and zero extends to 32 bits.
SXTH extracts bits[15:0] and sign extends to 32 bits.
UXTH extracts bits[15:0] and zero extends to 32 bits.

13.15.3.3

Restrictions
Do not use SP and do not use PC.

13.15.3.4

Condition flags
These instructions do not affect the flags.

13.15.3.5

Examples

142

SXTH

R4, R6, ROR #16

UXTB

R3, R10

;
;
;
;
;

Rotate R6 right by 16 bits, then obtain the lower
halfword of the result and then sign extend to
32 bits and write the result to R4.
Extract lowest byte of the value in R10 and zero
extend it, and write the result to R3

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.16 Branch and control instructions
Table 13-23 shows the branch and control instructions:
Table 13-23. Branch and control instructions
Mnemonic

Brief description

See

Branch

“B, BL, BX, and BLX” on page 144

Branch with Link

“B, BL, BX, and BLX” on page 144

BLX

Branch indirect with Link

“B, BL, BX, and BLX” on page 144

Branch indirect

“B, BL, BX, and BLX” on page 144

CBNZ

Compare and Branch if Non Zero

“CBZ and CBNZ” on page 146

CBZ

Compare and Branch if Non Zero

“CBZ and CBNZ” on page 146

If-Then

“IT” on page 147

TBB

Table Branch Byte

“TBB and TBH” on page 149

TBH

Table Branch Halfword

“TBB and TBH” on page 149

143
6430F–ATARM–21-Feb-12

13.16.1

13.16.1.1

B, BL, BX, and BLX
Branch instructions.
Syntax
B{cond} label
BL{cond} label
BX{cond} Rm
BLX{cond} Rm

where:
B

is branch (immediate).

is branch with link (immediate).

is branch indirect (register).

BLX

is branch indirect with link (register).

cond

is an optional condition code, see “Conditional execution” on page 98.

label

is a PC-relative expression. See “PC-relative expressions” on page 98.

Rm
is a register that indicates an address to branch to. Bit[0] of the value in Rm must
be 1, but the address to branch to is created by changing bit[0] to 0.
13.16.1.2

Operation
All these instructions cause a branch to label, or to the address indicated in Rm. In addition:
• The BL and BLX instructions write the address of the next instruction to LR (the link register,
R14).
• The BX and BLX instructions cause a UsageFault exception if bit[0] of Rm is 0.
Bcond label is the only conditional instruction that can be either inside or outside an IT block. All
other branch instructions must be conditional inside an IT block, and must be unconditional outside the IT block, see “IT” on page 147.
Table 13-24 shows the ranges for the various branch instructions.
Table 13-24. Branch ranges
Instruction

Branch range

B label

−16 MB to +16 MB

Bcond label (outside IT block)

−1 MB to +1 MB

Bcond label (inside IT block)

−16 MB to +16 MB

BL{cond} label

−16 MB to +16 MB

BX{cond} Rm

Any value in register

BLX{cond} Rm

Any value in register

You might have to use the .W suffix to get the maximum branch range. See “Instruction width
selection” on page 100.

13.16.1.3

Restrictions
The restrictions are:

144

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• do not use PC in the BLX instruction
• for BX and BLX, bit[0] of Rm must be 1 for correct execution but a branch occurs to the target
address created by changing bit[0] to 0
• when any of these instructions is inside an IT block, it must be the last instruction of the IT
block.
Bcond is the only conditional instruction that is not required to be inside an IT block. However, it
has a longer branch range when it is inside an IT block.

13.16.1.4

Condition flags
These instructions do not change the flags.

13.16.1.5

Examples

B
BLE
B.W
BEQ
BEQ.W
BL

loopA
ng
target
target
target
funC

BX
BXNE
BLX

LR
R0
R0

;
;
;
;
;
;
;
;
;
;
;

Branch to loopA
Conditionally branch to label ng
Branch to target within 16MB range
Conditionally branch to target
Conditionally branch to target within 1MB
Branch with link (Call) to function funC, return address
stored in LR
Return from function call
Conditionally branch to address stored in R0
Branch with link and exchange (Call) to a address stored
in R0

145
6430F–ATARM–21-Feb-12

13.16.2

CBZ and CBNZ
Compare and Branch on Zero, Compare and Branch on Non-Zero.

13.16.2.1

Syntax
CBZ Rn, label
CBNZ Rn, label

where:

13.16.2.2

is the register holding the operand.

label

is the branch destination.

Operation
Use the CBZ or CBNZ instructions to avoid changing the condition code flags and to reduce the
number of instructions.
CBZ Rn, label does not change condition flags but is otherwise equivalent to:
CMP
BEQ

Rn, #0
label

CBNZ Rn, label does not change condition flags but is otherwise equivalent to:
CMP
BNE

13.16.2.3

Rn, #0
label

Restrictions
The restrictions are:
• Rn must be in the range of R0 to R7
• the branch destination must be within 4 to 130 bytes after the instruction
• these instructions must not be used inside an IT block.

13.16.2.4

Condition flags
These instructions do not change the flags.

13.16.2.5

Examples

CBZ
CBNZ

146

R5, target ; Forward branch if R5 is zero
R0, target ; Forward branch if R0 is not zero

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.16.3

IT
If-Then condition instruction.

13.16.3.1

Syntax
IT{x{y{z}}} cond

where:
x

specifies the condition switch for the second instruction in the IT block.

specifies the condition switch for the third instruction in the IT block.

specifies the condition switch for the fourth instruction in the IT block.

cond

specifies the condition for the first instruction in the IT block.

The condition switch for the second, third and fourth instruction in the IT block can be either:
T

Then. Applies the condition cond to the instruction.

Else. Applies the inverse condition of cond to the instruction.

It is possible to use AL (the always condition) for cond in an IT instruction. If this is done, all of
the instructions in the IT block must be unconditional, and each of x, y, and z must be T or omitted but not E.

13.16.3.2

Operation
The IT instruction makes up to four following instructions conditional. The conditions can be all
the same, or some of them can be the logical inverse of the others. The conditional instructions
following the IT instruction form the IT block.
The instructions in the IT block, including any branches, must specify the condition in the {cond}
part of their syntax.
Your assembler might be able to generate the required IT instructions for conditional instructions
automatically, so that you do not need to write them yourself. See your assembler documentation for details.
A BKPT instruction in an IT block is always executed, even if its condition fails.
Exceptions can be taken between an IT instruction and the corresponding IT block, or within an
IT block. Such an exception results in entry to the appropriate exception handler, with suitable
return information in LR and stacked PSR.
Instructions designed for use for exception returns can be used as normal to return from the
exception, and execution of the IT block resumes correctly. This is the only way that a PC-modifying instruction is permitted to branch to an instruction in an IT block.

13.16.3.3

Restrictions
The following instructions are not permitted in an IT block:
• IT
• CBZ and CBNZ
• CPSID and CPSIE.
Other restrictions when using an IT block are:

147
6430F–ATARM–21-Feb-12

• a branch or any instruction that modifies the PC must either be outside an IT block or must be
the last instruction inside the IT block. These are:
– ADD PC, PC, Rm
– MOV PC, Rm
– B, BL, BX, BLX
– any LDM, LDR, or POP instruction that writes to the PC
– TBB and TBH
• do not branch to any instruction inside an IT block, except when returning from an exception
handler
• all conditional instructions except Bcond must be inside an IT block. Bcond can be either
outside or inside an IT block but has a larger branch range if it is inside one
• each instruction inside the IT block must specify a condition code suffix that is either the
same or logical inverse as for the other instructions in the block.
Your assembler might place extra restrictions on the use of IT blocks, such as prohibiting the
use of assembler directives within them.

13.16.3.4

Condition flags
This instruction does not change the flags.

13.16.3.5

Example

148

ITTE
ANDNE
ADDSNE
MOVEQ

NE
R0, R0, R1
R2, R2, #1
R2, R3

;
;
;
;

Next 3 instructions are conditional
ANDNE does not update condition flags
ADDSNE updates condition flags
Conditional move

CMP

R0, #9

ITE
ADDGT
ADDLE

GT
R1, R0, #55
R1, R0, #48

;
;
;
;
;

Convert R0 hex value (0 to 15) into ASCII
('0'-'9', 'A'-'F')
Next 2 instructions are conditional
Convert 0xA -> 'A'
Convert 0x0 -> '0'

; IT block with only one conditional instruction

ADDGT

R1, R1, #1

; Increment R1 conditionally

ITTEE
MOVEQ
ADDEQ
ANDNE
BNE.W

EQ
R0, R1
R2, R2, #10
R3, R3, #1
dloop

;
;
;
;
;
;

IT
ADD

NE
R0, R0, R1

; Next instruction is conditional
; Syntax error: no condition code used in IT block

Next 4 instructions are conditional
Conditional move
Conditional add
Conditional AND
Branch instruction can only be used in the last
instruction of an IT block

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.16.4

TBB and TBH
Table Branch Byte and Table Branch Halfword.

13.16.4.1

Syntax
TBB [Rn, Rm]
TBH [Rn, Rm, LSL #1]

where:
Rn
is the register containing the address of the table of branch lengths. If Rn is PC,
then the address of the table is the address of the byte immediately following the TBB or TBH
instruction.
Rm
is the index register. This contains an index into the table. For halfword tables,
LSL #1 doubles the value in Rm to form the right offset into the table.
13.16.4.2

Operation
These instructions cause a PC-relative forward branch using a table of single byte offsets for
TBB, or halfword offsets for TBH. Rn provides a pointer to the table, and Rm supplies an index
into the table. For TBB the branch offset is twice the unsigned value of the byte returned from
the table. and for TBH the branch offset is twice the unsigned value of the halfword returned
from the table. The branch occurs to the address at that offset from the address of the byte
immediately after the TBB or TBH instruction.

13.16.4.3

Restrictions
The restrictions are:
• Rn must not be SP
• Rm must not be SP and must not be PC
• when any of these instructions is used inside an IT block, it must be the last instruction of the
IT block.

13.16.4.4

Condition flags
These instructions do not change the flags.

149
6430F–ATARM–21-Feb-12

13.16.4.5

Examples

ADR.W R0, BranchTable_Byte
TBB
[R0, R1]
; R1 is the index, R0 is the base address of the
; branch table
Case1
; an instruction sequence follows
Case2
; an instruction sequence follows
Case3
; an instruction sequence follows
BranchTable_Byte
DCB
0
; Case1 offset calculation
DCB
((Case2-Case1)/2) ; Case2 offset calculation
DCB
((Case3-Case1)/2) ; Case3 offset calculation
TBH
[PC, R1, LSL #1]
; R1 is the index, PC is used as base of the
; branch table
BranchTable_H
DCI
((CaseA - BranchTable_H)/2) ; CaseA offset calculation
DCI
((CaseB - BranchTable_H)/2) ; CaseB offset calculation
DCI
((CaseC - BranchTable_H)/2) ; CaseC offset calculation
CaseA
; an instruction sequence follows
CaseB
; an instruction sequence follows
CaseC
; an instruction sequence follows

150

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.17 Miscellaneous instructions
Table 13-25 shows the remaining Cortex-M3 instructions:
Table 13-25. Miscellaneous instructions
Mnemonic

Brief description

See

BKPT

Breakpoint

“BKPT” on page 152

CPSID

Change Processor State, Disable
Interrupts

“CPS” on page 153

CPSIE

Change Processor State, Enable
Interrupts

“CPS” on page 153

DMB

Data Memory Barrier

“DMB” on page 154

DSB

Data Synchronization Barrier

“DSB” on page 155

ISB

Instruction Synchronization Barrier

“ISB” on page 156

MRS

Move from special register to register

“MRS” on page 157

MSR

Move from register to special register

“MSR” on page 158

NOP

No Operation

“NOP” on page 159

SEV

Send Event

“SEV” on page 160

SVC

Supervisor Call

“SVC” on page 161

WFE

Wait For Event

“WFE” on page 162

WFI

Wait For Interrupt

“WFI” on page 163

151
6430F–ATARM–21-Feb-12

13.17.1

BKPT
Breakpoint.

13.17.1.1

Syntax
BKPT #imm

where:
imm
13.17.1.2

is an expression evaluating to an integer in the range 0-255 (8-bit value).

Operation
The BKPT instruction causes the processor to enter Debug state. Debug tools can use this to
investigate system state when the instruction at a particular address is reached.
imm is ignored by the processor. If required, a debugger can use it to store additional information about the breakpoint.
The BKPT instruction can be placed inside an IT block, but it executes unconditionally, unaffected by the condition specified by the IT instruction.

13.17.1.3

Condition flags
This instruction does not change the flags.

13.17.1.4

Examples

BKPT 0xAB

152

; Breakpoint with immediate value set to 0xAB (debugger can
; extract the immediate value by locating it using the PC)

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.17.2

CPS
Change Processor State.

13.17.2.1

Syntax
CPSeffect iflags

where:
effect

is one of:
IE

Clears the special purpose register.

Sets the special purpose register.

iflags

13.17.2.2

is a sequence of one or more flags:
i

Set or clear PRIMASK.

Set or clear FAULTMASK.

Operation
CPS changes the PRIMASK and FAULTMASK special register values. See “Exception mask
registers” on page 62 for more information about these registers.

13.17.2.3

Restrictions
The restrictions are:
• use CPS only from privileged software, it has no effect if used in unprivileged software
• CPS cannot be conditional and so must not be used inside an IT block.

13.17.2.4

Condition flags
This instruction does not change the condition flags.

13.17.2.5

Examples

CPSID
CPSID
CPSIE
CPSIE

i
f
i
f

;
;
;
;

Disable interrupts and configurable fault handlers (set PRIMASK)
Disable interrupts and all fault handlers (set FAULTMASK)
Enable interrupts and configurable fault handlers (clear PRIMASK)
Enable interrupts and fault handlers (clear FAULTMASK)

153
6430F–ATARM–21-Feb-12

13.17.3

DMB
Data Memory Barrier.

13.17.3.1

Syntax
DMB{cond}

where:
cond
13.17.3.2

is an optional condition code, see “Conditional execution” on page 98.

Operation
DMB acts as a data memory barrier. It ensures that all explicit memory accesses that appear, in
program order, before the DMB instruction are completed before any explicit memory accesses
that appear, in program order, after the DMB instruction. DMB does not affect the ordering or
execution of instructions that do not access memory.

13.17.3.3

Condition flags
This instruction does not change the flags.

13.17.3.4

Examples

DMB

154

; Data Memory Barrier

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.17.4

DSB
Data Synchronization Barrier.

13.17.4.1

Syntax
DSB{cond}

where:
cond
13.17.4.2

is an optional condition code, see “Conditional execution” on page 98.

Operation
DSB acts as a special data synchronization memory barrier. Instructions that come after the
DSB, in program order, do not execute until the DSB instruction completes. The DSB instruction
completes when all explicit memory accesses before it complete.

13.17.4.3

Condition flags
This instruction does not change the flags.

13.17.4.4

Examples

DSB ; Data Synchronisation Barrier

155
6430F–ATARM–21-Feb-12

13.17.5

ISB
Instruction Synchronization Barrier.

13.17.5.1

Syntax
ISB{cond}

where:
cond
13.17.5.2

is an optional condition code, see “Conditional execution” on page 98.

Operation
ISB acts as an instruction synchronization barrier. It flushes the pipeline of the processor, so that
all instructions following the ISB are fetched from memory again, after the ISB instruction has
been completed.

13.17.5.3

Condition flags
This instruction does not change the flags.

13.17.5.4

Examples

ISB

156

; Instruction Synchronisation Barrier

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.17.6

MRS
Move the contents of a special register to a general-purpose register.

13.17.6.1

Syntax
MRS{cond} Rd, spec_reg

where:
cond

is an optional condition code, see “Conditional execution” on page 98.

is the destination register.

spec_reg
can be any of: APSR, IPSR, EPSR, IEPSR, IAPSR, EAPSR, PSR, MSP, PSP,
PRIMASK, BASEPRI, BASEPRI_MAX, FAULTMASK, or CONTROL.
13.17.6.2

Operation
Use MRS in combination with MSR as part of a read-modify-write sequence for updating a PSR,
for example to clear the Q flag.
In process swap code, the programmers model state of the process being swapped out must be
saved, including relevant PSR contents. Similarly, the state of the process being swapped in
must also be restored. These operations use MRS in the state-saving instruction sequence and
MSR in the state-restoring instruction sequence.
BASEPRI_MAX is an alias of BASEPRI when used with the MRS instruction.
See “MSR” on page 158.

13.17.6.3

Restrictions
Rd must not be SP and must not be PC.

13.17.6.4

Condition flags
This instruction does not change the flags.

13.17.6.5

Examples

MRS

R0, PRIMASK ; Read PRIMASK value and write it to R0

157
6430F–ATARM–21-Feb-12

13.17.7

MSR
Move the contents of a general-purpose register into the specified special register.

13.17.7.1

Syntax
MSR{cond} spec_reg, Rn

where:
cond

is an optional condition code, see “Conditional execution” on page 98.

is the source register.

spec_reg
can be any of: APSR, IPSR, EPSR, IEPSR, IAPSR, EAPSR, PSR, MSP, PSP,
PRIMASK, BASEPRI, BASEPRI_MAX, FAULTMASK, or CONTROL.
13.17.7.2

Operation
The register access operation in MSR depends on the privilege level. Unprivileged software can
only access the APSR, see “Application Program Status Register” on page 60. Privileged software can access all special registers.
In unprivileged software writes to unallocated or execution state bits in the PSR are ignored.
When you write to BASEPRI_MAX, the instruction writes to BASEPRI only if either:
• Rn is non-zero and the current BASEPRI value is 0
• Rn is non-zero and less than the current BASEPRI value.

See “MRS” on page 157.
13.17.7.3

Restrictions
Rn must not be SP and must not be PC.

13.17.7.4

13.17.7.5
MSR

158

Condition flags
This instruction updates the flags explicitly based on the value in Rn.
Examples
CONTROL, R1 ; Read R1 value and write it to the CONTROL register

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.17.8

NOP
No Operation.

13.17.8.1

Syntax
NOP{cond}

where:
cond
13.17.8.2

is an optional condition code, see “Conditional execution” on page 98.

Operation
NOP does nothing. NOP is not necessarily a time-consuming NOP. The processor might
remove it from the pipeline before it reaches the execution stage.
Use NOP for padding, for example to place the following instruction on a 64-bit boundary.

13.17.8.3

Condition flags
This instruction does not change the flags.

13.17.8.4

Examples

NOP

; No operation

159
6430F–ATARM–21-Feb-12

13.17.9

SEV
Send Event.

13.17.9.1

Syntax
SEV{cond}

where:
cond
13.17.9.2

is an optional condition code, see “Conditional execution” on page 98.

Operation
SEV is a hint instruction that causes an event to be signaled to all processors within a multiprocessor system. It also sets the local event register to 1, see “Power management” on page 87.

13.17.9.3

Condition flags
This instruction does not change the flags.

13.17.9.4

Examples

SEV ; Send Event

160

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.17.10 SVC
Supervisor Call.
13.17.10.1

Syntax
SVC{cond} #imm

where:

13.17.10.2

cond

is an optional condition code, see “Conditional execution” on page 98.

imm

is an expression evaluating to an integer in the range 0-255 (8-bit value).

Operation
The SVC instruction causes the SVC exception.
imm is ignored by the processor. If required, it can be retrieved by the exception handler to
determine what service is being requested.

13.17.10.3

Condition flags
This instruction does not change the flags.

13.17.10.4

Examples

SVC

0x32

; Supervisor Call (SVC handler can extract the immediate value
; by locating it via the stacked PC)

161
6430F–ATARM–21-Feb-12

13.17.11 WFE
Wait For Event.
13.17.11.1

Syntax
WFE{cond}

where:
cond
13.17.11.2

is an optional condition code, see “Conditional execution” on page 98.

Operation
WFE is a hint instruction.
If the event register is 0, WFE suspends execution until one of the following events occurs:
• an exception, unless masked by the exception mask registers or the current priority level
• an exception enters the Pending state, if SEVONPEND in the System Control Register is set
• a Debug Entry request, if Debug is enabled
• an event signaled by a peripheral or another processor in a multiprocessor system using the
SEV instruction.
If the event register is 1, WFE clears it to 0 and returns immediately.
For more information see “Power management” on page 87.

13.17.11.3

Condition flags
This instruction does not change the flags.

13.17.11.4

Examples

WFE

162

; Wait for event

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.17.12 WFI
Wait for Interrupt.
13.17.12.1

Syntax
WFI{cond}

where:
cond
13.17.12.2

is an optional condition code, see “Conditional execution” on page 98.

Operation
WFI is a hint instruction that suspends execution until one of the following events occurs:
• an exception
• a Debug Entry request, regardless of whether Debug is enabled.

13.17.12.3

Condition flags
This instruction does not change the flags.

13.17.12.4

Examples

WFI ; Wait for interrupt

163
6430F–ATARM–21-Feb-12

13.18 About the Cortex-M3 peripherals
The address map of the Private peripheral bus (PPB) is:
Table 13-26. Core peripheral register regions
Address

Core peripheral

Description

0xE000E0080xE000E00F

System control block

Table 13-30 on page 178

0xE000E0100xE000E01F

System timer

Table 13-33 on page 205

0xE000E1000xE000E4EF

Nested Vectored Interrupt
Controller

Table 13-27 on page 165

0xE000ED000xE000ED3F

System control block

Table 13-30 on page 178

0xE000ED900xE000EDB8

Memory protection unit

Table 13-35 on page 211

0xE000EF000xE000EF03

Nested Vectored Interrupt
Controller

Table 13-27 on page 165

In register descriptions:
• the register type is described as follows:
RW

Read and write.

Read-only.

Write-only.

• the required privilege gives the privilege level required to access the register, as follows:

164

Privileged

Only privileged software can access the register.

Unprivileged

Both unprivileged and privileged software can access the register.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.19 Nested Vectored Interrupt Controller
This section describes the Nested Vectored Interrupt Controller (NVIC) and the registers it uses.
The NVIC supports:
• 1 to 30 interrupts.
• A programmable priority level of 0-15 for each interrupt. A higher level corresponds to a lower
priority, so level 0 is the highest interrupt priority.
• Level detection of interrupt signals.
• Dynamic reprioritization of interrupts.
• Grouping of priority values into group priority and subpriority fields.
• Interrupt tail-chaining.
The processor automatically stacks its state on exception entry and unstacks this state on
exception exit, with no instruction overhead. This provides low latency exception handling. The
hardware implementation of the NVIC registers is:
Table 13-27. NVIC register summary
Address
0xE000E100
0xE000E180
0xE000E200

Name

Type

Required
privilege

Reset
value

Description

ISER0

Privileged

0x00000000

“Interrupt Set-enable Registers” on page 167

Privileged

0x00000000

“Interrupt Clear-enable Registers” on page 168

Privileged

0x00000000

“Interrupt Set-pending Registers” on page 169

Privileged

0x00000000

“Interrupt Clear-pending Registers” on page 170

Privileged

0x00000000

“Interrupt Active Bit Registers” on page 171

Privileged

0x00000000

“Interrupt Priority Registers” on page 172

0x00000000

“Software Trigger Interrupt Register” on page
175

ICER0
ISPR0

0xE000E280

ICPR0

0xE000E300

IABR0

0xE000E4000xE000E41C

IPR0IPR7

0xE000EF00

STIR

13.19.1

Configurable
(1)

See the register description for more information.

The CMSIS mapping of the Cortex-M3 NVIC registers
To improve software efficiency, the CMSIS simplifies the NVIC register presentation. In the
CMSIS:
• the Set-enable, Clear-enable, Set-pending, Clear-pending and Active Bit registers map to
arrays of 32-bit integers, so that:
– the array ISER[0] corresponds to the registers ISER0
– the array ICER[0] corresponds to the registers ICER0
– the array ISPR[0] corresponds to the registers ISPR0
– the array ICPR[0] corresponds to the registers ICPR0
– the array IABR[0] corresponds to the registers IABR0

165
6430F–ATARM–21-Feb-12

• the 4-bit fields of the Interrupt Priority Registers map to an array of 4-bit integers, so that the
array IP[0] to IP[29] corresponds to the registers IPR0-IPR7, and the array entry IP[n] holds
the interrupt priority for interrupt n.
The CMSIS provides thread-safe code that gives atomic access to the Interrupt Priority Registers. For more information see the description of the NVIC_SetPriority function in “NVIC
programming hints” on page 177. Table 13-28 shows how the interrupts, or IRQ numbers, map
onto the interrupt registers and corresponding CMSIS variables that have one bit per interrupt.
Table 13-28. Mapping of interrupts to the interrupt variables
CMSIS array elements (1)
Interrupts

Set-enable

Clear-enable

Set-pending

Clear-pending

Active Bit

0-29

ISER[0]

ICER[0]

ISPR[0]

ICPR[0]

IABR[0]

166

Each array element corresponds to a single NVIC register, for example the element
ICER[0] corresponds to the ICER0 register.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.19.2

Interrupt Set-enable Registers
The ISER0 register enables interrupts, and show which interrupts are enabled. See:
• the register summary in Table 13-27 on page 165 for the register attributes
• Table 13-28 on page 166 for which interrupts are controlled by each register.
The bit assignments are:

SETENA bits
23

SETENA bits
15

SETENA bits
7

SETENA bits

• SETENA
Interrupt set-enable bits.
Write:
0 = no effect
1 = enable interrupt.
Read:
0 = interrupt disabled
1 = interrupt enabled.
If a pending interrupt is enabled, the NVIC activates the interrupt based on its priority. If an interrupt is not enabled, asserting its interrupt signal changes the interrupt state to pending, but the NVIC never activates the interrupt, regardless of its
priority.

167
6430F–ATARM–21-Feb-12

13.19.3

Interrupt Clear-enable Registers
The ICER0 register disables interrupts, and shows which interrupts are enabled. See:
• the register summary in Table 13-27 on page 165 for the register attributes
• Table 13-28 on page 166 for which interrupts are controlled by each register
The bit assignments are:

CLRENA
23

CLRENA
15

CLRENA
7

CLRENA

• CLRENA
Interrupt clear-enable bits.
Write:
0 = no effect
1 = disable interrupt.
Read:
0 = interrupt disabled
1 = interrupt enabled.

168

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.19.4

Interrupt Set-pending Registers
The ISPR0 register forces interrupts into the pending state, and shows which interrupts are
pending. See:
• the register summary in Table 13-27 on page 165 for the register attributes
• Table 13-28 on page 166 for which interrupts are controlled by each register.
The bit assignments are:

SETPEND
23

SETPEND
15

SETPEND
7

SETPEND

• SETPEND
Interrupt set-pending bits.
Write:
0 = no effect.
1 = changes interrupt state to pending.
Read:
0 = interrupt is not pending.
1 = interrupt is pending.
Writing 1 to the ISPR bit corresponding to:
• an interrupt that is pending has no effect
• a disabled interrupt sets the state of that interrupt to pending

169
6430F–ATARM–21-Feb-12

13.19.5

Interrupt Clear-pending Registers
The ICPR0 register removes the pending state from interrupts, and show which interrupts are
pending. See:
• the register summary in Table 13-27 on page 165 for the register attributes
• Table 13-28 on page 166 for which interrupts are controlled by each register.
The bit assignments are:

CLRPEND
23

CLRPEND
15

CLRPEND
7

CLRPEND

• CLRPEND
Interrupt clear-pending bits.
Write:
0 = no effect.
1 = removes pending state an interrupt.
Read:
0 = interrupt is not pending.
1 = interrupt is pending.
Writing 1 to an ICPR bit does not affect the active state of the corresponding interrupt.

170

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.19.6

Interrupt Active Bit Registers
The IABR0 register indicates which interrupts are active. See:
• the register summary in Table 13-27 on page 165 for the register attributes
• Table 13-28 on page 166 for which interrupts are controlled by each register.
The bit assignments are:

ACTIVE
23

ACTIVE
15

ACTIVE
7

ACTIVE

• ACTIVE
Interrupt active flags:
0 = interrupt not active
1 = interrupt active.
A bit reads as one if the status of the corresponding interrupt is active or active and pending.

171
6430F–ATARM–21-Feb-12

13.19.7

13.19.7.1

Interrupt Priority Registers
The IPR0-IPR7 registers provide a 4-bit priority field for each interrupt (See the “Peripheral Identifiers” section of the datasheet for more details). These registers are byte-accessible. See the
register summary in Table 13-27 on page 165 for their attributes. Each register holds four priority
fields, that map up to four elements in the CMSIS interrupt priority array IP[0] to IP[29], as shown:
IPRm

IP[4m+3]
23

IP[4m+2]
15

IP[4m+1]
7

IP[4m]

13.19.7.2

IPR4

IP[19]
23

IP[18]
15

IP[17]
7

IP[16]

13.19.7.3
31

IPR3
30

IP[15]
23

IP[14]
15

IP[13]
7

IP[12]

172

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.19.7.4

IPR2

IP[11]
23

IP[10]
15

IP[9]
7

IP[8]

13.19.7.5

IPR1

IP[7]
23

IP[6]
15

IP[5]
7

IP[4]

13.19.7.6

IPR0

IP[3]
23

IP[2]
15

IP[1]
7

IP[0]

• Priority, byte offset 3
• Priority, byte offset 2
• Priority, byte offset 1
• Priority, byte offset 0
Each priority field holds a priority value, 0-15. The lower the value, the greater the priority of the corresponding interrupt.
The processor implements only bits[7:4] of each field, bits[3:0] read as zero and ignore writes.
See “The CMSIS mapping of the Cortex-M3 NVIC registers” on page 165 for more information about the IP[0] to IP[29]
interrupt priority array, that provides the software view of the interrupt priorities.

173
6430F–ATARM–21-Feb-12

Find the IPR number and byte offset for interrupt N as follows:
• the corresponding IPR number, M, is given by M = N DIV 4
• the byte offset of the required Priority field in this register is N MOD 4, where:
– byte offset 0 refers to register bits[7:0]
– byte offset 1 refers to register bits[15:8]
– byte offset 2 refers to register bits[23:16]
– byte offset 3 refers to register bits[31:24].

174

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.19.8

Software Trigger Interrupt Register
Write to the STIR to generate a Software Generated Interrupt (SGI). See the register summary
in Table 13-27 on page 165 for the STIR attributes.
When the USERSETMPEND bit in the SCR is set to 1, unprivileged software can access the
STIR, see “System Control Register” on page 187.
Only privileged software can enable unprivileged access to the STIR.
The bit assignments are:

Reserved
23

Reserved
15

Reserved
7

INTID

• INTID
Interrupt ID of the required SGI, in the range 0-239. For example, a value of b000000011 specifies interrupt IRQ3.

175
6430F–ATARM–21-Feb-12

13.19.9

Level-sensitive interrupts
The processor supports level-sensitive interrupts.
A level-sensitive interrupt is held asserted until the peripheral deasserts the interrupt signal. Typically this happens because the ISR accesses the peripheral, causing it to clear the interrupt
request.
When the processor enters the ISR, it automatically removes the pending state from the interrupt, see “Hardware and software control of interrupts”. For a level-sensitive interrupt, if the
signal is not deasserted before the processor returns from the ISR, the interrupt becomes pending again, and the processor must execute its ISR again. This means that the peripheral can
hold the interrupt signal asserted until it no longer needs servicing.

13.19.9.1

Hardware and software control of interrupts
The Cortex-M3 latches all interrupts. A peripheral interrupt becomes pending for one of the following reasons:
• the NVIC detects that the interrupt signal is HIGH and the interrupt is not active
• the NVIC detects a rising edge on the interrupt signal
• software writes to the corresponding interrupt set-pending register bit, see “Interrupt Setpending Registers” on page 169, or to the STIR to make an SGI pending, see “Software
Trigger Interrupt Register” on page 175.
A pending interrupt remains pending until one of the following:
The processor enters the ISR for the interrupt. This changes the state of the interrupt from pending to active. Then:
– For a level-sensitive interrupt, when the processor returns from the ISR, the NVIC
samples the interrupt signal. If the signal is asserted, the state of the interrupt
changes to pending, which might cause the processor to immediately re-enter the
ISR. Otherwise, the state of the interrupt changes to inactive.
– If the interrupt signal is not pulsed while the processor is in the ISR, when the
processor returns from the ISR the state of the interrupt changes to inactive.
• Software writes to the corresponding interrupt clear-pending register bit.
For a level-sensitive interrupt, if the interrupt signal is still asserted, the state of the interrupt
does not change. Otherwise, the state of the interrupt changes to inactive.

176

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.19.10 NVIC design hints and tips
Ensure software uses correctly aligned register accesses. The processor does not support
unaligned accesses to NVIC registers. See the individual register descriptions for the supported
access sizes.
A interrupt can enter pending state even it is disabled.
Before programming VTOR to relocate the vector table, ensure the vector table entries of the
new vector table are setup for fault handlers and all enabled exception like interrupts. For more
information see “Vector Table Offset Register” on page 184.
13.19.10.1

NVIC programming hints
Software uses the CPSIE I and CPSID I instructions to enable and disable interrupts. The
CMSIS provides the following intrinsic functions for these instructions:
void __disable_irq(void) // Disable Interrupts
void __enable_irq(void) // Enable Interrupts

In addition, the CMSIS provides a number of functions for NVIC control, including:
Table 13-29. CMSIS functions for NVIC control
CMSIS interrupt control function

Description

void NVIC_SetPriorityGrouping(uint32_t
priority_grouping)

Set the priority grouping

void NVIC_EnableIRQ(IRQn_t IRQn)

Enable IRQn

void NVIC_DisableIRQ(IRQn_t IRQn)

Disable IRQn

uint32_t NVIC_GetPendingIRQ (IRQn_t IRQn)

Return true if IRQn is pending

void NVIC_SetPendingIRQ (IRQn_t IRQn)

Set IRQn pending

void NVIC_ClearPendingIRQ (IRQn_t IRQn)

Clear IRQn pending status

uint32_t NVIC_GetActive (IRQn_t IRQn)

Return the IRQ number of the active
interrupt

void NVIC_SetPriority (IRQn_t IRQn, uint32_t priority)

Set priority for IRQn

uint32_t NVIC_GetPriority (IRQn_t IRQn)

Read priority of IRQn

void NVIC_SystemReset (void)

Reset the system

For more information about these functions see the CMSIS documentation.

177
6430F–ATARM–21-Feb-12

13.20 System control block
The System control block (SCB) provides system implementation information, and system control. This includes configuration, control, and reporting of the system exceptions. The system
control block registers are:
Table 13-30. Summary of the system control block registers
Address

Name

Type

Required
privilege

Reset
value

Description

0xE000E008

ACTLR

Privileged

0x00000000

“Auxiliary Control Register” on page 179

0xE000ED00

CPUID

Privileged

0x412FC230

“CPUID Base Register” on page 180

Privileged

0x00000000

“Interrupt Control and State Register” on page 181

(1)

0xE000ED04

ICSR

0xE000ED08

VTOR

Privileged

0x00000000

“Vector Table Offset Register” on page 184

0xE000ED0C

AIRCR

RW (1)

Privileged

0xFA050000

“Application Interrupt and Reset Control Register” on page
185

0xE000ED10

SCR

Privileged

0x00000000

“System Control Register” on page 187

0xE000ED14

CCR

Privileged

0x00000200

“Configuration and Control Register” on page 188

0xE000ED18

SHPR1

Privileged

0x00000000

“System Handler Priority Register 1” on page 191

0xE000ED1C

SHPR2

Privileged

0x00000000

“System Handler Priority Register 2” on page 192

0xE000ED20

SHPR3

Privileged

0x00000000

“System Handler Priority Register 3” on page 192

0xE000ED24

SHCRS

Privileged

0x00000000

“System Handler Control and State Register” on page 193

0xE000ED28

CFSR

Privileged

0x00000000

“Configurable Fault Status Register” on page 195

0xE000ED28

MMSR(2)

Privileged

0x00

“Memory Management Fault Address Register” on page
202

0xE000ED29

BFSR (2)

Privileged

0x00

“Bus Fault Status Register” on page 197

0xE000ED2A

UFSR (2)

Privileged

0x0000

“Usage Fault Status Register” on page 199

0xE000ED2C

HFSR

Privileged

0x00000000

“Hard Fault Status Register” on page 201

0xE000ED34

MMAR

Privileged

Unknown

“Memory Management Fault Address Register” on page
202

0xE000ED38

BFAR

Privileged

Unknown

“Bus Fault Address Register” on page 203

0xE000ED3C

AFSR

Privileged

0x00000000

“Auxiliary Fault Status Register” on page 204

Notes:

1. See the register description for more information.
2. A subregister of the CFSR.

13.20.1

178

The CMSIS mapping of the Cortex-M3 SCB registers
To improve software efficiency, the CMSIS simplifies the SCB register presentation. In the
CMSIS, the byte array SHP[0] to SHP[12] corresponds to the registers SHPR1-SHPR3.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.2

Auxiliary Control Register
The ACTLR provides disable bits for the following processor functions:
• IT folding
• write buffer use for accesses to the default memory map
• interruption of multi-cycle instructions.
See the register summary in Table 13-30 on page 178 for the ACTLR attributes. The bit assignments are:

2
DISFOLD

1
DISDEFWBUF

0
DISMCYCINT

Reserved
23

Reserved
15

Reserved
7

5
Reserved

• DISFOLD
When set to 1, disables IT folding. see “About IT folding” on page 179 for more information.
• DISDEFWBUF
When set to 1, disables write buffer use during default memory map accesses. This causes all bus faults to be precise bus
faults but decreases performance because any store to memory must complete before the processor can execute the next
instruction.
This bit only affects write buffers implemented in the Cortex-M3 processor.
• DISMCYCINT
When set to 1, disables interruption of load multiple and store multiple instructions. This increases the interrupt latency of
the processor because any LDM or STM must complete before the processor can stack the current state and enter the
interrupt handler.
13.20.2.1

About IT folding
In some situations, the processor can start executing the first instruction in an IT block while it is
still executing the IT instruction. This behavior is called IT folding, and improves performance,
However, IT folding can cause jitter in looping. If a task must avoid jitter, set the DISFOLD bit to
1 before executing the task, to disable IT folding.

179
6430F–ATARM–21-Feb-12

13.20.3

CPUID Base Register
The CPUID register contains the processor part number, version, and implementation information. See the register summary in Table 13-30 on page 178 for its attributes. The bit assignments
are:

Implementer
23

Variant
15

Constant
13

PartNo
7

PartNo

Revision

• Implementer
Implementer code:
0x41 = ARM
• Variant
Variant number, the r value in the rnpn product revision identifier:
0x2 = r2p0
• Constant
Reads as 0xF
• PartNo
Part number of the processor:
0xC23 = Cortex-M3
• Revision
Revision number, the p value in the rnpn product revision identifier:
0x0 = r2p0

180

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.4

Interrupt Control and State Register
The ICSR:
• provides:
– set-pending and clear-pending bits for the PendSV and SysTick exceptions
• indicates:
– the exception number of the exception being processed
– whether there are preempted active exceptions
– the exception number of the highest priority pending exception
– whether any interrupts are pending.
See the register summary in Table 13-30 on page 178, and the Type descriptions in Table 13-33
on page 205, for the ICSR attributes. The bit assignments are:

Reserved

Reserved for
Debug

ISRPENDING

PENDSVSET

PENDSVCLR

PENDSTSET

PENDSTCLR

Reserved

VECTPENDING
13

VECTPENDING
7

RETTOBASE
4

Reserved
2

VECTACTIVE
1

VECTACTIVE

• PENDSVSET
RW
PendSV set-pending bit.
Write:
0 = no effect
1 = changes PendSV exception state to pending.
Read:
0 = PendSV exception is not pending
1 = PendSV exception is pending.
Writing 1 to this bit is the only way to set the PendSV exception state to pending.
• PENDSVCLR
WO
PendSV clear-pending bit.
Write:
0 = no effect
1 = removes the pending state from the PendSV exception.

181
6430F–ATARM–21-Feb-12

• PENDSTSET
RW
SysTick exception set-pending bit.
Write:
0 = no effect
1 = changes SysTick exception state to pending.
Read:
0 = SysTick exception is not pending
1 = SysTick exception is pending.
• PENDSTCLR
WO
SysTick exception clear-pending bit.
Write:
0 = no effect
1 = removes the pending state from the SysTick exception.
This bit is WO. On a register read its value is Unknown.
• Reserved for Debug use
RO
This bit is reserved for Debug use and reads-as-zero when the processor is not in Debug.
• ISRPENDING
RO
Interrupt pending flag, excluding Faults:
0 = interrupt not pending
1 = interrupt pending.
• VECTPENDING
RO
Indicates the exception number of the highest priority pending enabled exception:
0 = no pending exceptions
Nonzero = the exception number of the highest priority pending enabled exception.
The value indicated by this field includes the effect of the BASEPRI and FAULTMASK registers, but not any effect of the
PRIMASK register.

182

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• RETTOBASE
RO
Indicates whether there are preempted active exceptions:
0 = there are preempted active exceptions to execute
1 = there are no active exceptions, or the currently-executing exception is the only active exception.
• VECTACTIVE
RO
Contains the active exception number:
0 = Thread mode
Nonzero = The exception number (1) of the currently active exception.
Subtract 16 from this value to obtain the IRQ number required to index into the Interrupt Clear-Enable, Set-Enable, ClearPending, Set-Pending, or Priority Registers, see “Interrupt Program Status Register” on page 61.
When you write to the ICSR, the effect is Unpredictable if you:
• write 1 to the PENDSVSET bit and write 1 to the PENDSVCLR bit
• write 1 to the PENDSTSET bit and write 1 to the PENDSTCLR bit.

Note:

1. This is the same value as IPSR bits [8:0] see “Interrupt Program Status Register” on page 61.

183
6430F–ATARM–21-Feb-12

13.20.5

Vector Table Offset Register
The VTOR indicates the offset of the vector table base address from memory address
0x00000000. See the register summary in Table 13-30 on page 178 for its attributes.
The bit assignments are:

Reserved
23

TBLOFF
22

TBLOFF
15

TBLOFF
7

TBLOFF

Reserved

• TBLOFF
Vector table base offset field. It contains bits[29:7] of the offset of the table base from the bottom of the memory map.
Bit[29] determines whether the vector table is in the code or SRAM memory region:
0 = code
1 = SRAM.
Bit[29] is sometimes called the TBLBASE bit.
When setting TBLOFF, you must align the offset to the number of exception entries in the vector table. The minimum alignment is 32 words, enough for up to 16 interrupts. For more interrupts, adjust the alignment by rounding up to the next power
of two. For example, if you require 21 interrupts, the alignment must be on a 64-word boundary because the required table
size is 37 words, and the next power of two is 64.
Table alignment requirements mean that bits[6:0] of the table offset are always zero.

184

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.6

Application Interrupt and Reset Control Register
The AIRCR provides priority grouping control for the exception model, endian status for data
accesses, and reset control of the system. See the register summary in Table 13-30 on page
178 and Table 13-33 on page 205 for its attributes.
To write to this register, you must write 0x05FA to the VECTKEY field, otherwise the processor
ignores the write.
The bit assignments are:

On Read: VECTKEYSTAT, On Write: VECTKEY
23

On Read: VECTKEYSTAT, On Write: VECTKEY
15

ENDIANESS
7

Reserved
6

PRIGROUP
4

Reserved

SYSRESETREQ

VECTCLRACTIVE

VECTRESET

• VECTKEYSTAT
Register Key:
Reads as 0xFA05
• VECTKEY
Register key:
On writes, write 0x5FA to VECTKEY, otherwise the write is ignored.
• ENDIANESS
RO
Data endianness bit:
0 = Little-endian
ENDIANESS is set from the BIGEND configuration signal during reset.
• PRIGROUP
R/W
Interrupt priority grouping field. This field determines the split of group priority from subpriority, see “Binary point” on page
186.
• SYSRESETREQ
WO
System reset request:
0 = no effect
1 = asserts a proc_reset_signal.
This is intended to force a large system reset of all major components except for debug.
This bit reads as 0.
185
6430F–ATARM–21-Feb-12

• VECTCLRACTIVE
WO
Reserved for Debug use. This bit reads as 0. When writing to the register you must write 0 to this bit, otherwise behavior is
Unpredictable.
• VECTRESET
WO
Reserved for Debug use. This bit reads as 0. When writing to the register you must write 0 to this bit, otherwise behavior is
Unpredictable.
13.20.6.1

Binary point
The PRIGROUP field indicates the position of the binary point that splits the PRI_n fields in the
Interrupt Priority Registers into separate group priority and subpriority fields. Table 13-31 shows
how the PRIGROUP value controls this split.

Table 13-31. Priority grouping
Interrupt priority level value, PRI_N[7:0]

Number of

PRIGROUP

Binary
point (1)

Group priority
bits

Subpriority
bits

Group
priorities

Subpriorities

b011

bxxxx.0000

[7:4]

None

b100

bxxx.y0000

[7:5]

[4]

b101

bxx.yy0000

[7:6]

[5:4]

b110

bx.yyy0000

[7]

[6:4]

b111

b.yyyy0000

None

[7:4]

PRI_n[7:0] field showing the binary point. x denotes a group priority field bit, and y denotes a subpriority field bit.

Determining preemption of an exception uses only the group priority field, see “Interrupt priority
grouping” on page 82.

186

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.7

System Control Register
The SCR controls features of entry to and exit from low power state. See the register summary
in Table 13-30 on page 178 for its attributes. The bit assignments are:

Reserved
23

Reserved
15

Reserved
7

Reserved

SEVONPEND

Reserved

SLEEPDEEP

SLEEONEXIT

Reserved

• SEVONPEND
Send Event on Pending bit:
0 = only enabled interrupts or events can wakeup the processor, disabled interrupts are excluded
1 = enabled events and all interrupts, including disabled interrupts, can wakeup the processor.
When an event or interrupt enters pending state, the event signal wakes up the processor from WFE. If the processor is not
waiting for an event, the event is registered and affects the next WFE.
The processor also wakes up on execution of an SEV instruction or an external event.
• SLEEPDEEP
Controls whether the processor uses sleep or deep sleep as its low power mode:
0 = sleep
1 = deep sleep.
• SLEEPONEXIT
Indicates sleep-on-exit when returning from Handler mode to Thread mode:
0 = do not sleep when returning to Thread mode.
1 = enter sleep, or deep sleep, on return from an ISR.
Setting this bit to 1 enables an interrupt driven application to avoid returning to an empty main application.

187
6430F–ATARM–21-Feb-12

13.20.8

Configuration and Control Register
The CCR controls entry to Thread mode and enables:
• the handlers for hard fault and faults escalated by FAULTMASK to ignore bus faults
• trapping of divide by zero and unaligned accesses
• access to the STIR by unprivileged software, see “Software Trigger Interrupt Register” on
page 175.
See the register summary in Table 13-30 on page 178 for the CCR attributes.
The bit assignments are:

Reserved
23

Reserved
15

Reserved
7

Reserved

DIV_0_TRP

UNALIGN_T
RP

STKALIGN

BFHFNMIGN

Reserved

USERSETM
PEND

NONBASET
HRDENA

• STKALIGN
Indicates stack alignment on exception entry:
0 = 4-byte aligned
1 = 8-byte aligned.
On exception entry, the processor uses bit[9] of the stacked PSR to indicate the stack alignment. On return from the exception it uses this stacked bit to restore the correct stack alignment.
• BFHFNMIGN
Enables handlers with priority -1 or -2 to ignore data bus faults caused by load and store instructions. This applies to the
hard fault and FAULTMASK escalated handlers:
0 = data bus faults caused by load and store instructions cause a lock-up
1 = handlers running at priority -1 and -2 ignore data bus faults caused by load and store instructions.
Set this bit to 1 only when the handler and its data are in absolutely safe memory. The normal use of this bit is to probe system devices and bridges to detect control path problems and fix them.
• DIV_0_TRP
Enables faulting or halting when the processor executes an SDIV or UDIV instruction with a divisor of 0:
0 = do not trap divide by 0
1 = trap divide by 0.
When this bit is set to 0,a divide by zero returns a quotient of 0.
• UNALIGN_TRP
Enables unaligned access traps:
0 = do not trap unaligned halfword and word accesses

188

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
1 = trap unaligned halfword and word accesses.
If this bit is set to 1, an unaligned access generates a usage fault.
Unaligned LDM, STM, LDRD, and STRD instructions always fault irrespective of whether UNALIGN_TRP is set to 1.
• USERSETMPEND
Enables unprivileged software access to the STIR, see “Software Trigger Interrupt Register” on page 175:
0 = disable
1 = enable.
• NONEBASETHRDENA
Indicates how the processor enters Thread mode:
0 = processor can enter Thread mode only when no exception is active.
1 = processor can enter Thread mode from any level under the control of an EXC_RETURN value, see “Exception return”
on page 84.

189
6430F–ATARM–21-Feb-12

13.20.9

System Handler Priority Registers
The SHPR1-SHPR3 registers set the priority level, 0 to 15 of the exception handlers that have
configurable priority.
SHPR1-SHPR3 are byte accessible. See the register summary in Table 13-30 on page 178 for
their attributes.
The system fault handlers and the priority field and register for each handler are:
Table 13-32. System fault handler priority fields
Handler

Field

Memory management
fault

PRI_4

Bus fault

PRI_5

Usage fault

PRI_6

SVCall

PRI_11

PendSV

PRI_14

SysTick

PRI_15

“System Handler Priority Register 1” on page 191

“System Handler Priority Register 2” on page 192
“System Handler Priority Register 3” on page 192

Each PRI_N field is 8 bits wide, but the processor implements only bits[7:4] of each field, and
bits[3:0] read as zero and ignore writes.

190

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.9.1

System Handler Priority Register 1
The bit assignments are:

PRI_7: Reserved
23

PRI_6
15

PRI_5
7

PRI_4

• PRI_7
Reserved
• PRI_6
Priority of system handler 6, usage fault
• PRI_5
Priority of system handler 5, bus fault
• PRI_4
Priority of system handler 4, memory management fault

191
6430F–ATARM–21-Feb-12

13.20.9.2

System Handler Priority Register 2
The bit assignments are:
30

PRI_11
23

Reserved
15

Reserved
7

Reserved

• PRI_11
Priority of system handler 11, SVCall
13.20.9.3

System Handler Priority Register 3
The bit assignments are:
30

PRI_15
23

PRI_14
15

Reserved
7

Reserved

• PRI_15
Priority of system handler 15, SysTick exception
• PRI_14
Priority of system handler 14, PendSV

192

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.10 System Handler Control and State Register
The SHCSR enables the system handlers, and indicates:
• the pending status of the bus fault, memory management fault, and SVC exceptions
• the active status of the system handlers.
See the register summary in Table 13-30 on page 178 for the SHCSR attributes. The bit assignments are:
31

Reserved
23

Reserved

USGFAULTENA

BUSFAULTENA

MEMFAULTENA

SVCALLPENDE
D

BUSFAULTPEND
ED

MEMFAULTPEN
DED

USGFAULTPEND
ED

SYSTICKACT

PENDSVACT

Reserved

MONITORACT

SVCALLAVCT

Reserved

USGFAULTACT

Reserved

BUSFAULTACT

MEMFAULTACT

• USGFAULTENA
Usage fault enable bit, set to 1 to enable (1)
• BUSFAULTENA
Bus fault enable bit, set to 1 to enable (3)
• MEMFAULTENA
Memory management fault enable bit, set to 1 to enable (3)
• SVCALLPENDED
SVC call pending bit, reads as 1 if exception is pending (2)
• BUSFAULTPENDED
Bus fault exception pending bit, reads as 1 if exception is pending (2)
• MEMFAULTPENDED
Memory management fault exception pending bit, reads as 1 if exception is pending (2)
• USGFAULTPENDED
Usage fault exception pending bit, reads as 1 if exception is pending (2)
• SYSTICKACT
SysTick exception active bit, reads as 1 if exception is active (3)
• PENDSVACT
PendSV exception active bit, reads as 1 if exception is active
1.

Enable bits, set to 1 to enable the exception, or set to 0 to disable the exception.

Pending bits, read as 1 if the exception is pending, or as 0 if it is not pending. You can write to these bits to change the pending
status of the exceptions.

Active bits, read as 1 if the exception is active, or as 0 if it is not active. You can write to these bits to change the active status of
the exceptions, but see the Caution in this section.

193
6430F–ATARM–21-Feb-12

• MONITORACT
Debug monitor active bit, reads as 1 if Debug monitor is active
• SVCALLACT
SVC call active bit, reads as 1 if SVC call is active
• USGFAULTACT
Usage fault exception active bit, reads as 1 if exception is active
• BUSFAULTACT
Bus fault exception active bit, reads as 1 if exception is active
• MEMFAULTACT
Memory management fault exception active bit, reads as 1 if exception is active

If you disable a system handler and the corresponding fault occurs, the processor treats the fault as a hard fault.
You can write to this register to change the pending or active status of system exceptions. An OS kernel can write to the
active bits to perform a context switch that changes the current exception type.
• Software that changes the value of an active bit in this register without correct adjustment to the stacked content can
cause the processor to generate a fault exception. Ensure software that writes to this register retains and subsequently
restores the current active status.
• After you have enabled the system handlers, if you have to change the value of a bit in this register you must use a readmodify-write procedure to ensure that you change only the required bit.

194

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.11 Configurable Fault Status Register
The CFSR indicates the cause of a memory management fault, bus fault, or usage fault. See the
register summary in Table 13-30 on page 178 for its attributes. The bit assignments are:
31

Usage Fault Status Register: UFSR
23

Usage Fault Status Register: UFSR
15

Bus Fault Status Register: BFSR
7

Memory Management Fault Status Register: MMFSR

The following subsections describe the subregisters that make up the CFSR:
• “Memory Management Fault Status Register” on page 196
• “Bus Fault Status Register” on page 197
• “Usage Fault Status Register” on page 199.
The CFSR is byte accessible. You can access the CFSR or its subregisters as follows:
• access the complete CFSR with a word access to 0xE000ED28
• access the MMFSR with a byte access to 0xE000ED28
• access the MMFSR and BFSR with a halfword access to 0xE000ED28
• access the BFSR with a byte access to 0xE000ED29
• access the UFSR with a halfword access to 0xE000ED2A.

195
6430F–ATARM–21-Feb-12

13.20.11.1

Memory Management Fault Status Register
The flags in the MMFSR indicate the cause of memory access faults. The bit assignments are:

MMARVALID

Reserved

MSTKERR

MUNSTKERR

Reserved

DACCVIOL

IACCVIOL

• MMARVALID
Memory Management Fault Address Register (MMAR) valid flag:
0 = value in MMAR is not a valid fault address
1 = MMAR holds a valid fault address.
If a memory management fault occurs and is escalated to a hard fault because of priority, the hard fault handler must set
this bit to 0. This prevents problems on return to a stacked active memory management fault handler whose MMAR value
has been overwritten.
• MSTKERR
Memory manager fault on stacking for exception entry:
0 = no stacking fault
1 = stacking for an exception entry has caused one or more access violations.
When this bit is 1, the SP is still adjusted but the values in the context area on the stack might be incorrect. The processor
has not written a fault address to the MMAR.
• MUNSTKERR
Memory manager fault on unstacking for a return from exception:
0 = no unstacking fault
1 = unstack for an exception return has caused one or more access violations.
This fault is chained to the handler. This means that when this bit is 1, the original return stack is still present. The processor has not adjusted the SP from the failing return, and has not performed a new save. The processor has not written a
fault address to the MMAR.
• DACCVIOL
Data access violation flag:
0 = no data access violation fault
1 = the processor attempted a load or store at a location that does not permit the operation.
When this bit is 1, the PC value stacked for the exception return points to the faulting instruction. The processor has loaded
the MMAR with the address of the attempted access.
• IACCVIOL
Instruction access violation flag:
0 = no instruction access violation fault
1 = the processor attempted an instruction fetch from a location that does not permit execution.
This fault occurs on any access to an XN region, even when the MPU is disabled or not present.
When this bit is 1, the PC value stacked for the exception return points to the faulting instruction. The processor has not
written a fault address to the MMAR.

196

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.11.2

Bus Fault Status Register
The flags in the BFSR indicate the cause of a bus access fault. The bit assignments are:

BFRVALID

Reserved

STKERR

UNSTKERR

IMPRECISERR

PRECISERR

IBUSERR

• BFARVALID
Bus Fault Address Register (BFAR) valid flag:
0 = value in BFAR is not a valid fault address
1 = BFAR holds a valid fault address.
The processor sets this bit to 1 after a bus fault where the address is known. Other faults can set this bit to 0, such as a
memory management fault occurring later.
If a bus fault occurs and is escalated to a hard fault because of priority, the hard fault handler must set this bit to 0. This prevents problems if returning to a stacked active bus fault handler whose BFAR value has been overwritten.
• STKERR
Bus fault on stacking for exception entry:
0 = no stacking fault
1 = stacking for an exception entry has caused one or more bus faults.
When the processor sets this bit to 1, the SP is still adjusted but the values in the context area on the stack might be incorrect. The processor does not write a fault address to the BFAR.
• UNSTKERR
Bus fault on unstacking for a return from exception:
0 = no unstacking fault
1 = unstack for an exception return has caused one or more bus faults.
This fault is chained to the handler. This means that when the processor sets this bit to 1, the original return stack is still
present. The processor does not adjust the SP from the failing return, does not performed a new save, and does not write
a fault address to the BFAR.
• IMPRECISERR
Imprecise data bus error:
0 = no imprecise data bus error
1 = a data bus error has occurred, but the return address in the stack frame is not related to the instruction that caused the
error.
When the processor sets this bit to 1, it does not write a fault address to the BFAR.
This is an asynchronous fault. Therefore, if it is detected when the priority of the current process is higher than the bus fault
priority, the bus fault becomes pending and becomes active only when the processor returns from all higher priority processes. If a precise fault occurs before the processor enters the handler for the imprecise bus fault, the handler detects
both IMPRECISERR set to 1 and one of the precise fault status bits set to 1.

197
6430F–ATARM–21-Feb-12

• PRECISERR
Precise data bus error:
0 = no precise data bus error
1 = a data bus error has occurred, and the PC value stacked for the exception return points to the instruction that caused
the fault.
When the processor sets this bit is 1, it writes the faulting address to the BFAR.
• IBUSERR
Instruction bus error:
0 = no instruction bus error
1 = instruction bus error.
The processor detects the instruction bus error on prefetching an instruction, but it sets the IBUSERR flag to 1 only if it
attempts to issue the faulting instruction.
When the processor sets this bit is 1, it does not write a fault address to the BFAR.

198

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.11.3

Usage Fault Status Register
The UFSR indicates the cause of a usage fault. The bit assignments are:

Reserved
7

Reserved

DIVBYZERO

UNALIGNED

NOCP

INVPC

INVSTATE

UNDEFINSTR

• DIVBYZERO
Divide by zero usage fault:
0 = no divide by zero fault, or divide by zero trapping not enabled
1 = the processor has executed an SDIV or UDIV instruction with a divisor of 0.
When the processor sets this bit to 1, the PC value stacked for the exception return points to the instruction that performed
the divide by zero.
Enable trapping of divide by zero by setting the DIV_0_TRP bit in the CCR to 1, see “Configuration and Control Register”
on page 188.
• UNALIGNED
Unaligned access usage fault:
0 = no unaligned access fault, or unaligned access trapping not enabled
1 = the processor has made an unaligned memory access.
Enable trapping of unaligned accesses by setting the UNALIGN_TRP bit in the CCR to 1, see “Configuration and Control
Register” on page 188.
Unaligned LDM, STM, LDRD, and STRD instructions always fault irrespective of the setting of UNALIGN_TRP.
• NOCP
No coprocessor usage fault. The processor does not support coprocessor instructions:
0 = no usage fault caused by attempting to access a coprocessor
1 = the processor has attempted to access a coprocessor.
• INVPC
Invalid PC load usage fault, caused by an invalid PC load by EXC_RETURN:
0 = no invalid PC load usage fault
1 = the processor has attempted an illegal load of EXC_RETURN to the PC, as a result of an invalid context, or an invalid
EXC_RETURN value.
When this bit is set to 1, the PC value stacked for the exception return points to the instruction that tried to perform the illegal load of the PC.
• INVSTATE
Invalid state usage fault:
0 = no invalid state usage fault
1 = the processor has attempted to execute an instruction that makes illegal use of the EPSR.

199
6430F–ATARM–21-Feb-12

When this bit is set to 1, the PC value stacked for the exception return points to the instruction that attempted the illegal use
of the EPSR.
This bit is not set to 1 if an undefined instruction uses the EPSR.
• UNDEFINSTR
Undefined instruction usage fault:
0 = no undefined instruction usage fault
1 = the processor has attempted to execute an undefined instruction.
When this bit is set to 1, the PC value stacked for the exception return points to the undefined instruction.
An undefined instruction is an instruction that the processor cannot decode.

The UFSR bits are sticky. This means as one or more fault occurs, the associated bits are set to 1. A bit that is set to 1 is
cleared to 0 only by writing 1 to that bit, or by a reset.

200

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.12 Hard Fault Status Register
The HFSR gives information about events that activate the hard fault handler. See the register
summary in Table 13-30 on page 178 for its attributes.
This register is read, write to clear. This means that bits in the register read normally, but writing
1 to any bit clears that bit to 0. The bit assignments are:
31

DEBUGEVT

FORCED

Reserved
21

Reserved
15

Reserved
7

Reserved

VECTTBL

Reserved

• DEBUGEVT
Reserved for Debug use. When writing to the register you must write 0 to this bit, otherwise behavior is Unpredictable.
• FORCED
Indicates a forced hard fault, generated by escalation of a fault with configurable priority that cannot be handles, either
because of priority or because it is disabled:
0 = no forced hard fault
1 = forced hard fault.
When this bit is set to 1, the hard fault handler must read the other fault status registers to find the cause of the fault.
• VECTTBL
Indicates a bus fault on a vector table read during exception processing:
0 = no bus fault on vector table read
1 = bus fault on vector table read.
This error is always handled by the hard fault handler.
When this bit is set to 1, the PC value stacked for the exception return points to the instruction that was preempted by the
exception.

The HFSR bits are sticky. This means as one or more fault occurs, the associated bits are set to 1. A bit that is set to 1 is
cleared to 0 only by writing 1 to that bit, or by a reset.

201
6430F–ATARM–21-Feb-12

13.20.13 Memory Management Fault Address Register
The MMFAR contains the address of the location that generated a memory management fault.
See the register summary in Table 13-30 on page 178 for its attributes. The bit assignments are:
31

ADDRESS
23

ADDRESS
15

ADDRESS
7

ADDRESS

• ADDRESS
When the MMARVALID bit of the MMFSR is set to 1, this field holds the address of the location that generated the memory
management fault
When an unaligned access faults, the address is the actual address that faulted. Because a single read or write instruction
can be split into multiple aligned accesses, the fault address can be any address in the range of the requested access size.
Flags in the MMFSR indicate the cause of the fault, and whether the value in the MMFAR is valid. See “Memory Management Fault Status Register” on page 196.

202

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.20.14 Bus Fault Address Register
The BFAR contains the address of the location that generated a bus fault. See the register summary in Table 13-30 on page 178 for its attributes. The bit assignments are:
31

ADDRESS
23

ADDRESS
15

ADDRESS
7

ADDRESS

• ADDRESS
When the BFARVALID bit of the BFSR is set to 1, this field holds the address of the location that generated the bus fault
When an unaligned access faults the address in the BFAR is the one requested by the instruction, even if it is not the
address of the fault.
Flags in the BFSR indicate the cause of the fault, and whether the value in the BFAR is valid. See “Bus Fault Status Register” on page 197.

203
6430F–ATARM–21-Feb-12

13.20.15 Auxiliary Fault Status Register
The AFSR contains additional system fault information. See the register summary in Table 1330 on page 178 for its attributes.
This register is read, write to clear. This means that bits in the register read normally, but writing
1 to any bit clears that bit to 0.
The bit assignments are:
31

IMPDEF
23

IMPDEF
15

IMPDEF
7

IMPDEF

• IMPDEF
Implementation defined. The bits map to the AUXFAULT input signals.
Each AFSR bit maps directly to an AUXFAULT input of the processor, and a single-cycle HIGH signal on the input sets the
corresponding AFSR bit to one. It remains set to 1 until you write 1 to the bit to clear it to zero.
When an AFSR bit is latched as one, an exception does not occur. Use an interrupt if an exception is required.
13.20.16 System control block design hints and tips
Ensure software uses aligned accesses of the correct size to access the system control block
registers:
• except for the CFSR and SHPR1-SHPR3, it must use aligned word accesses
• for the CFSR and SHPR1-SHPR3 it can use byte or aligned halfword or word accesses.
The processor does not support unaligned accesses to system control block registers.
In a fault handler. to determine the true faulting address:
• Read and save the MMFAR or BFAR value.
• Read the MMARVALID bit in the MMFSR, or the BFARVALID bit in the BFSR. The MMFAR or
BFAR address is valid only if this bit is 1.
Software must follow this sequence because another higher priority exception might change the
MMFAR or BFAR value. For example, if a higher priority handler preempts the current fault handler, the other fault might change the MMFAR or BFAR value.

204

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.21 System timer, SysTick
The processor has a 24-bit system timer, SysTick, that counts down from the reload value to
zero, reloads (wraps to) the value in the LOAD register on the next clock edge, then counts
down on subsequent clocks.
When the processor is halted for debugging the counter does not decrement.
The system timer registers are:
Table 13-33. System timer registers summary
Address

Name

Type

Required
privilege

Reset
value

Description

0xE000E010

CTRL

Privileged

0x00000004

“SysTick Control and Status Register” on page 206

0xE000E014

LOAD

Privileged

0x00000000

“SysTick Reload Value Register” on page 207

0xE000E018

VAL

Privileged

0x00000000

“SysTick Current Value Register” on page 208

0xE000E01C

CALIB

Privileged

0x0002904 (1)

“SysTick Calibration Value Register” on page 209

SysTick calibration value.

205
6430F–ATARM–21-Feb-12

13.21.1

SysTick Control and Status Register
The SysTick CTRL register enables the SysTick features. See the register summary in Table 1333 on page 205 for its attributes. The bit assignments are:

Reserved
23

Reserved
15

COUNTFLAG
11

Reserved
7

Reserved

CLKSOURCE

TICKINT

ENABLE

• COUNTFLAG
Returns 1 if timer counted to 0 since last time this was read.
• CLKSOURCE
Indicates the clock source:
0 = MCK/8
1 = MCK
• TICKINT
Enables SysTick exception request:
0 = counting down to zero does not assert the SysTick exception request
1 = counting down to zero to asserts the SysTick exception request.
Software can use COUNTFLAG to determine if SysTick has ever counted to zero.
• ENABLE
Enables the counter:
0 = counter disabled
1 = counter enabled.
When ENABLE is set to 1, the counter loads the RELOAD value from the LOAD register and then counts down. On reaching 0, it sets the COUNTFLAG to 1 and optionally asserts the SysTick depending on the value of TICKINT. It then loads the
RELOAD value again, and begins counting.

206

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.21.2

SysTick Reload Value Register
The LOAD register specifies the start value to load into the VAL register. See the register summary in Table 13-33 on page 205 for its attributes. The bit assignments are:

Reserved
23

RELOAD
15

RELOAD
7

-RELOAD

• RELOAD
Value to load into the VAL register when the counter is enabled and when it reaches 0, see “Calculating the RELOAD
value”.
13.21.2.1

Calculating the RELOAD value
The RELOAD value can be any value in the range 0x00000001-0x00FFFFFF. A start value of 0
is possible, but has no effect because the SysTick exception request and COUNTFLAG are activated when counting from 1 to 0.
The RELOAD value is calculated according to its use:
• To generate a multi-shot timer with a period of N processor clock cycles, use a RELOAD
value of N-1. For example, if the SysTick interrupt is required every 100 clock pulses, set
RELOAD to 99.
• To deliver a single SysTick interrupt after a delay of N processor clock cycles, use a RELOAD
of value N. For example, if a SysTick interrupt is required after 400 clock pulses, set RELOAD
to 400.

207
6430F–ATARM–21-Feb-12

13.21.3

SysTick Current Value Register
The VAL register contains the current value of the SysTick counter. See the register summary in
Table 13-33 on page 205 for its attributes. The bit assignments are:
30

Reserved
23

CURRENT
15

CURRENT
7

CURRENT

• CURRENT
Reads return the current value of the SysTick counter.
A write of any value clears the field to 0, and also clears the SysTick CTRL.COUNTFLAG bit to 0.

208

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.21.4

SysTick Calibration Value Register
The CALIB register indicates the SysTick calibration properties. See the register summary in
Table 13-33 on page 205 for its attributes. The bit assignments are:

NOREF

SKEW

Reserved
21

TENMS
15

TENMS
7

TENMS

• NOREF
Reads as zero.
• SKEW
Reads as zero
• TENMS
Read as 0x0002904. The SysTick calibration value is fixed at 0x0002904 (10500), which allows the generation of a time
base of 1 ms with SysTick clock at 10.5 MHz (84/8 = 10.5 MHz)
13.21.5

SysTick design hints and tips
The SysTick counter runs on the processor clock. If this clock signal is stopped for low power
mode, the SysTick counter stops.
Ensure software uses aligned word accesses to access the SysTick registers.

209
6430F–ATARM–21-Feb-12

13.22 Memory protection unit
This section describes the Memory protection unit (MPU).
The MPU divides the memory map into a number of regions, and defines the location, size,
access permissions, and memory attributes of each region. It supports:
• independent attribute settings for each region
• overlapping regions
• export of memory attributes to the system.
The memory attributes affect the behavior of memory accesses to the region. The Cortex-M3
MPU defines:
• eight separate memory regions, 0-7
• a background region.
When memory regions overlap, a memory access is affected by the attributes of the region with
the highest number. For example, the attributes for region 7 take precedence over the attributes
of any region that overlaps region 7.
The background region has the same memory access attributes as the default memory map, but
is accessible from privileged software only.
The Cortex-M3 MPU memory map is unified. This means instruction accesses and data
accesses have same region settings.
If a program accesses a memory location that is prohibited by the MPU, the processor generates
a memory management fault. This causes a fault exception, and might cause termination of the
process in an OS environment.
In an OS environment, the kernel can update the MPU region setting dynamically based on the
process to be executed. Typically, an embedded OS uses the MPU for memory protection.
Configuration of MPU regions is based on memory types, see “Memory regions, types and attributes” on page 68.
Table 13-34 shows the possible MPU region attributes. These include Share ability and cache
behavior attributes that are not relevant to most microcontroller implementations. See “MPU
configuration for a microcontroller” on page 223 for guidelines for programming such an
implementation.
Table 13-34. Memory attributes summary
Memory
type

210

Shareability

Other attributes

Description

Stronglyordered

All accesses to Strongly-ordered memory occur
in program order. All Strongly-ordered regions
are assumed to be shared.

Device

Shared

Memory-mapped peripherals that several
processors share.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 13-34. Memory attributes summary (Continued)
Memory
type

Normal

Shareability

Other attributes

Description

Non-shared

Memory-mapped peripherals that only a single
processor uses.

Shared

Normal memory that is shared between several
processors.

Non-shared

Normal memory that only a single processor
uses.

Use the MPU registers to define the MPU regions and their attributes. The MPU registers are:
Table 13-35. MPU registers summary
Address

Name

Type

Required
privilege

Reset
value

Description

0xE000ED90

TYPE

Privileged

0x00000800

“MPU Type Register” on page 212

0xE000ED94

CTRL

Privileged

0x00000000

“MPU Control Register” on page 213

0xE000ED98

RNR

Privileged

0x00000000

“MPU Region Number Register” on page 215

0xE000ED9C

RBAR

Privileged

0x00000000

“MPU Region Base Address Register” on page 216

0xE000EDA0

RASR

Privileged

0x00000000

“MPU Region Attribute and Size Register” on page 217

0xE000EDA4

RBAR_A1

Privileged

0x00000000

Alias of RBAR, see “MPU Region Base Address
Register” on page 216

0xE000EDA8

RASR_A1

Privileged

0x00000000

Alias of RASR, see “MPU Region Attribute and Size
Register” on page 217

0xE000EDAC

RBAR_A2

Privileged

0x00000000

Alias of RBAR, see “MPU Region Base Address
Register” on page 216

0xE000EDB0

RASR_A2

Privileged

0x00000000

Alias of RASR, see “MPU Region Attribute and Size
Register” on page 217

0xE000EDB4

RBAR_A3

Privileged

0x00000000

Alias of RBAR, see “MPU Region Base Address
Register” on page 216

0xE000EDB8

RASR_A3

Privileged

0x00000000

Alias of RASR, see “MPU Region Attribute and Size
Register” on page 217

211
6430F–ATARM–21-Feb-12

13.22.1

MPU Type Register
The TYPE register indicates whether the MPU is present, and if so, how many regions it supports. See the register summary in Table 13-35 on page 211 for its attributes. The bit
assignments are:

Reserved
23

IREGION
15

DREGION
7

Reserved

SEPARATE

• IREGION
Indicates the number of supported MPU instruction regions.
Always contains 0x00. The MPU memory map is unified and is described by the DREGION field.
• DREGION
Indicates the number of supported MPU data regions:
0x08 = Eight MPU regions.
• SEPARATE
Indicates support for unified or separate instruction and date memory maps:
0 = unified.

212

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.22.2

MPU Control Register
The MPU CTRL register:
• enables the MPU
• enables the default memory map background region
• enables use of the MPU when in the hard fault, Non-maskable Interrupt (NMI), and
FAULTMASK escalated handlers.
See the register summary in Table 13-35 on page 211 for the MPU CTRL attributes. The bit
assignments are:

Reserved
23

Reserved
15

Reserved
7

Reserved

PRIVDEFENA

HFNMIENA

ENABLE

• PRIVDEFENA
Enables privileged software access to the default memory map:
0 = If the MPU is enabled, disables use of the default memory map. Any memory access to a location not covered by any
enabled region causes a fault.
1 = If the MPU is enabled, enables use of the default memory map as a background region for privileged software
accesses.
When enabled, the background region acts as if it is region number -1. Any region that is defined and enabled has priority
over this default map.
If the MPU is disabled, the processor ignores this bit.
• HFNMIENA
Enables the operation of MPU during hard fault, NMI, and FAULTMASK handlers.
When the MPU is enabled:
0 = MPU is disabled during hard fault, NMI, and FAULTMASK handlers, regardless of the value of the ENABLE bit
1 = the MPU is enabled during hard fault, NMI, and FAULTMASK handlers.
When the MPU is disabled, if this bit is set to 1 the behavior is Unpredictable.
• ENABLE
Enables the MPU:
0 = MPU disabled
1 = MPU enabled.
When ENABLE and PRIVDEFENA are both set to 1:
For privileged accesses, the default memory map is as described in “Memory model” on page 68. Any access by privileged
software that does not address an enabled memory region behaves as defined by the default memory map.

213
6430F–ATARM–21-Feb-12

Any access by unprivileged software that does not address an enabled memory region causes a memory management
fault.
XN and Strongly-ordered rules always apply to the System Control Space regardless of the value of the ENABLE bit.
When the ENABLE bit is set to 1, at least one region of the memory map must be enabled for the system to function unless
the PRIVDEFENA bit is set to 1. If the PRIVDEFENA bit is set to 1 and no regions are enabled, then only privileged software can operate.
When the ENABLE bit is set to 0, the system uses the default memory map. This has the same memory attributes as if the
MPU is not implemented, see Table 13-34 on page 210. The default memory map applies to accesses from both privileged
and unprivileged software.
When the MPU is enabled, accesses to the System Control Space and vector table are always permitted. Other areas are
accessible based on regions and whether PRIVDEFENA is set to 1.
Unless HFNMIENA is set to 1, the MPU is not enabled when the processor is executing the handler for an exception with
priority –1 or –2. These priorities are only possible when handling a hard fault or NMI exception, or when FAULTMASK is
enabled. Setting the HFNMIENA bit to 1 enables the MPU when operating with these two priorities.

214

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.22.3

MPU Region Number Register
The RNR selects which memory region is referenced by the RBAR and RASR registers. See the
register summary in Table 13-35 on page 211 for its attributes. The bit assignments are:

Reserved
23

Reserved
15

Reserved
7

REGION

• REGION
Indicates the MPU region referenced by the RBAR and RASR registers.
The MPU supports 8 memory regions, so the permitted values of this field are 0-7.
Normally, you write the required region number to this register before accessing the RBAR or RASR. However you can
change the region number by writing to the RBAR with the VALID bit set to 1, see “MPU Region Base Address Register” on
page 216. This write updates the value of the REGION field.

215
6430F–ATARM–21-Feb-12

13.22.4

MPU Region Base Address Register
The RBAR defines the base address of the MPU region selected by the RNR, and can update
the value of the RNR. See the register summary in Table 13-35 on page 211 for its attributes.
Write RBAR with the VALID bit set to 1 to change the current region number and update the
RNR. The bit assignments are:

ADDR
23

ADDR
15

ADDR
N-1

Reserved

VALID

REGION

• ADDR
Region base address field. The value of N depends on the region size. For more information see “The ADDR field”.
• VALID
MPU Region Number valid bit:
Write:
0 = RNR not changed, and the processor:
updates the base address for the region specified in the RNR
ignores the value of the REGION field
1 = the processor:
updates the value of the RNR to the value of the REGION field
updates the base address for the region specified in the REGION field.
Always reads as zero.
• REGION
MPU region field:
For the behavior on writes, see the description of the VALID field.
On reads, returns the current region number, as specified by the RNR.
13.22.4.1

The ADDR field
The ADDR field is bits[31:N] of the RBAR. The region size, as specified by the SIZE field in the
RASR, defines the value of N:
N = Log2(Region size in bytes),

If the region size is configured to 4GB, in the RASR, there is no valid ADDR field. In this case,
the region occupies the complete memory map, and the base address is 0x00000000.
The base address is aligned to the size of the region. For example, a 64KB region must be
aligned on a multiple of 64KB, for example, at 0x00010000 or 0x00020000.

216

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.22.5

MPU Region Attribute and Size Register
The RASR defines the region size and memory attributes of the MPU region specified by the
RNR, and enables that region and any subregions. See the register summary in Table 13-35 on
page 211 for its attributes.
RASR is accessible using word or halfword accesses:
• the most significant halfword holds the region attributes
• the least significant halfword holds the region size and the region and subregion enable bits.
The bit assignments are:

Reserved
23

Reserved

Reserved
15

AP
18

TEX
14

SRD
7

Reserved

SIZE

ENABLE

• XN
Instruction access disable bit:
0 = instruction fetches enabled
1 = instruction fetches disabled.
• AP
Access permission field, see Table 13-39 on page 219.
• TEX, C, B
Memory access attributes, see Table 13-37 on page 218.
• S
Shareable bit, see Table 13-36 on page 218.
• SRD
Subregion disable bits. For each bit in this field:
0 = corresponding sub-region is enabled
1 = corresponding sub-region is disabled
See “Subregions” on page 222 for more information.
Region sizes of 128 bytes and less do not support subregions. When writing the attributes for such a region, write the SRD
field as 0x00.
• SIZE
Specifies the size of the MPU protection region. The minimum permitted value is 3 (b00010), see See “SIZE field values”
on page 218 for more information.
• ENABLE

217
6430F–ATARM–21-Feb-12

Region enable bit.
For information about access permission, see “MPU access permission attributes”.
13.22.5.1

SIZE field values
The SIZE field defines the size of the MPU memory region specified by the RNR. as follows:
(Region size in bytes) = 2(SIZE+1)

The smallest permitted region size is 32B, corresponding to a SIZE value of 4. Table 13-36 gives
example SIZE values, with the corresponding region size and value of N in the RBAR.
Table 13-36. Example SIZE field values
SIZE value

Region size

Value of
N (1)

b00100 (4)

32B

Minimum permitted
size

b01001 (9)

1KB

b10011 (19)

1MB

b11101 (29)

1GB

b11111 (31)

4GB

b01100

Maximum possible
size

13.22.6

Note

In the RBAR, see “MPU Region Base Address Register” on
page 216.

MPU access permission attributes
This section describes the MPU access permission attributes. The access permission bits, TEX,
C, B, S, AP, and XN, of the RASR, control access to the corresponding memory region. If an
access is made to an area of memory without the required permissions, then the MPU generates
a permission fault.
Table 13-37 shows the encodings for the TEX, C, B, and S access permission bits.
Table 13-37. TEX, C, B, and S encoding
TEX

Memory type

Shareability

Other attributes

x (1)

Stronglyordered

Shareable

x (1)

Device

Shareable

Normal

Not
shareable

0
0

b000

Shareable

Not
shareable

Outer and inner write-through. No write
allocate.

1
1

Normal
1

218

Outer and inner write-back. No write
allocate.

Shareable

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 13-37. TEX, C, B, and S encoding (Continued)
TEX

Memory type

Shareability

Normal

Not
shareable

0
1

Shareable
(1)

x (1)

b001

Reserved encoding

Implementation defined
attributes.

0
1

Normal
1

Device

x (1)

Reserved encoding

(1)

Reserved encoding

(1)

Normal
1

Not
shareable

x (1)

Outer and inner write-back. Write and
read allocate.

b1B
B

Not
shareable
Shareable

0
b010

Other attributes

Nonshared Device.

Not
shareable
Shareable

The MPU ignores the value of this bit.

Table 13-38 shows the cache policy for memory attribute encodings with a TEX value is in the
range 4-7.
Table 13-38. Cache policy for memory attribute encoding
Encoding, AA or BB

Corresponding cache policy

Non-cacheable

Write back, write and read allocate

Write through, no write allocate

Write back, no write allocate

Table 13-39 shows the AP encodings that define the access permissions for privileged and
unprivileged software.
Table 13-39. AP encoding
AP[2:0]

Privileged
permissions

Unprivileged
permissions

Description

000

No access

All accesses generate a permission fault

001

No access

Access from privileged software only

010

Writes by unprivileged software generate a permission
fault

011

Full access

100

Unpredictable

Reserved

219
6430F–ATARM–21-Feb-12

Table 13-39. AP encoding (Continued)
AP[2:0]

Privileged
permissions

Unprivileged
permissions

Description

101

No access

Reads by privileged software only

110

Read only, by privileged or unprivileged software

111

Read only, by privileged or unprivileged software

13.22.7

MPU mismatch
When an access violates the MPU permissions, the processor generates a memory management fault, see “Exceptions and interrupts” on page 66. The MMFSR indicates the cause of the
fault. See “Memory Management Fault Status Register” on page 196 for more information.

13.22.8

Updating an MPU region
To update the attributes for an MPU region, update the RNR, RBAR and RASR registers. You
can program each register separately, or use a multiple-word write to program all of these registers. You can use the RBAR and RASR aliases to program up to four regions simultaneously
using an STM instruction.

13.22.8.1

Updating an MPU region using separate words
Simple code to configure one region:
; R1 = region number
; R2 = size/enable
; R3 = attributes
; R4 = address
LDR R0,=MPU_RNR

; 0xE000ED98, MPU region number register

STR R1, [R0, #0x0]

; Region Number

STR R4, [R0, #0x4]

; Region Base Address

STRH R2, [R0, #0x8]

; Region Size and Enable

STRH R3, [R0, #0xA]

; Region Attribute

Disable a region before writing new region settings to the MPU if you have previously enabled
the region being changed. For example:
; R1 = region number
; R2 = size/enable
; R3 = attributes
; R4 = address
LDR R0,=MPU_RNR

; 0xE000ED98, MPU region number register

STR R1, [R0, #0x0]

; Region Number

BIC R2, R2, #1

; Disable

STRH R2, [R0, #0x8]

; Region Size and Enable

STR R4, [R0, #0x4]

; Region Base Address

STRH R3, [R0, #0xA]

; Region Attribute

ORR R2, #1

; Enable

STRH R2, [R0, #0x8]

; Region Size and Enable

Software must use memory barrier instructions:

220

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• before MPU setup if there might be outstanding memory transfers, such as buffered writes,
that might be affected by the change in MPU settings
• after MPU setup if it includes memory transfers that must use the new MPU settings.
However, memory barrier instructions are not required if the MPU setup process starts by entering an exception handler, or is followed by an exception return, because the exception entry and
exception return mechanism cause memory barrier behavior.
Software does not need any memory barrier instructions during MPU setup, because it accesses
the MPU through the PPB, which is a Strongly-Ordered memory region.
For example, if you want all of the memory access behavior to take effect immediately after the
programming sequence, use a DSB instruction and an ISB instruction. A DSB is required after
changing MPU settings, such as at the end of context switch. An ISB is required if the code that
programs the MPU region or regions is entered using a branch or call. If the programming
sequence is entered using a return from exception, or by taking an exception, then you do not
require an ISB.
13.22.8.2

Updating an MPU region using multi-word writes
You can program directly using multi-word writes, depending on how the information is divided.
Consider the following reprogramming:
; R1 = region number
; R2 = address
; R3 = size, attributes in one
LDR R0, =MPU_RNR

; 0xE000ED98, MPU region number register

STR R1, [R0, #0x0]

; Region Number

STR R2, [R0, #0x4]

; Region Base Address

STR R3, [R0, #0x8]

; Region Attribute, Size and Enable

Use an STM instruction to optimize this:
; R1 = region number
; R2 = address
; R3 = size, attributes in one
LDR R0, =MPU_RNR

; 0xE000ED98, MPU region number register

STM R0, {R1-R3}

; Region Number, address, attribute, size and enable

You can do this in two words for pre-packed information. This means that the RBAR contains the
required region number and had the VALID bit set to 1, see “MPU Region Base Address Register” on page 216. Use this when the data is statically packed, for example in a boot loader:
; R1 = address and region number in one
; R2 = size and attributes in one
LDR R0, =MPU_RBAR

; 0xE000ED9C, MPU Region Base register

STR R1, [R0, #0x0]

; Region base address and
; region number combined with VALID (bit 4) set to 1

STR R2, [R0, #0x4]

; Region Attribute, Size and Enable

Use an STM instruction to optimize this:
; R1 = address and region number in one
; R2 = size and attributes in one
LDR R0,=MPU_RBAR

; 0xE000ED9C, MPU Region Base register

STM R0, {R1-R2}

; Region base address, region number and VALID bit,

221
6430F–ATARM–21-Feb-12

; and Region Attribute, Size and Enable

13.22.8.3

Subregions
Regions of 256 bytes or more are divided into eight equal-sized subregions. Set the corresponding bit in the SRD field of the RASR to disable a subregion, see “MPU Region Attribute and Size
Register” on page 217. The least significant bit of SRD controls the first subregion, and the most
significant bit controls the last subregion. Disabling a subregion means another region overlapping the disabled range matches instead. If no other enabled region overlaps the disabled
subregion the MPU issues a fault.
Regions of 32, 64, and 128 bytes do not support subregions, With regions of these sizes, you
must set the SRD field to 0x00, otherwise the MPU behavior is Unpredictable.

13.22.8.4

Example of SRD use
Two regions with the same base address overlap. Region one is 128KB, and region two is
512KB. To ensure the attributes from region one apply to the first128KB region, set the SRD
field for region two to b00000011 to disable the first two subregions, as Figure 13-9 shows
Figure 13-9. SRD use
Region 2, with
subregions

Region 1

Base address of both regions

13.22.9

Offset from
base address
512KB
448KB
384KB
320KB
256KB
192KB
128KB
Disabled subregion
64KB
Disabled subregion
0

MPU design hints and tips
To avoid unexpected behavior, disable the interrupts before updating the attributes of a region
that the interrupt handlers might access.
Ensure software uses aligned accesses of the correct size to access MPU registers:
• except for the RASR, it must use aligned word accesses
• for the RASR it can use byte or aligned halfword or word accesses.
The processor does not support unaligned accesses to MPU registers.
When setting up the MPU, and if the MPU has previously been programmed, disable unused
regions to prevent any previous region settings from affecting the new MPU setup.

222

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.22.9.1

MPU configuration for a microcontroller
Usually, a microcontroller system has only a single processor and no caches. In such a system,
program the MPU as follows:
Table 13-40. Memory region attributes for a microcontroller
Memory region

TEX

Memory type and attributes

Flash memory

b000

Normal memory, Non-shareable, write-through

Internal SRAM

b000

Normal memory, Shareable, write-through

External SRAM

b000

Normal memory, Shareable, write-back, write-allocate

Peripherals

b000

Device memory, Shareable

In most microcontroller implementations, the share ability and cache policy attributes do not
affect the system behavior. However, using these settings for the MPU regions can make the
application code more portable. The values given are for typical situations. In special systems,
such as multiprocessor designs or designs with a separate DMA engine, the share ability attribute might be important. In these cases refer to the recommendations of the memory device
manufacturer.

223
6430F–ATARM–21-Feb-12

13.23 Glossary
This glossary describes some of the terms used in technical documents from ARM.
Abort
A mechanism that indicates to a processor that the value associated with a memory access is
invalid. An abort can be caused by the external or internal memory system as a result of
attempting to access invalid instruction or data memory.
Aligned
A data item stored at an address that is divisible by the number of bytes that defines the data
size is said to be aligned. Aligned words and halfwords have addresses that are divisible by four
and two respectively. The terms word-aligned and halfword-aligned therefore stipulate
addresses that are divisible by four and two respectively.
Banked register
A register that has multiple physical copies, where the state of the processor determines which
copy is used. The Stack Pointer, SP (R13) is a banked register.
Base register
In instruction descriptions, a register specified by a load or store instruction that is used to hold
the base value for the instruction’s address calculation. Depending on the instruction and its
addressing mode, an offset can be added to or subtracted from the base register value to form
the address that is sent to memory.
See also “Index register”
Breakpoint
A breakpoint is a mechanism provided by debuggers to identify an instruction at which program
execution is to be halted. Breakpoints are inserted by the programmer to enable inspection of
register contents, memory locations, variable values at fixed points in the program execution to
test that the program is operating correctly. Breakpoints are removed after the program is successfully tested.
Condition field
A four-bit field in an instruction that specifies a condition under which the instruction can
execute.
Conditional execution
If the condition code flags indicate that the corresponding condition is true when the instruction
starts executing, it executes normally. Otherwise, the instruction does nothing.
Context
The environment that each process operates in for a multitasking operating system. In ARM processors, this is limited to mean the physical address range that it can access in memory and the
associated memory access permissions.
Coprocessor
A processor that supplements the main processor. Cortex-M3 does not support any
coprocessors.
Debugger

224

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
A debugging system that includes a program, used to detect, locate, and correct software faults,
together with custom hardware that supports software debugging.
Direct Memory Access (DMA)
An operation that accesses main memory directly, without the processor performing any
accesses to the data concerned.
Doubleword
A 64-bit data item. The contents are taken as being an unsigned integer unless otherwise
stated.
Doubleword-aligned
A data item having a memory address that is divisible by eight.
Endianness
Byte ordering. The scheme that determines the order that successive bytes of a data word are
stored in memory. An aspect of the system’s memory mapping.
See also “Little-endian (LE)”
Exception
An event that interrupts program execution. When an exception occurs, the processor suspends
the normal program flow and starts execution at the address indicated by the corresponding
exception vector. The indicated address contains the first instruction of the handler for the
exception.
An exception can be an interrupt request, a fault, or a software-generated system exception.
Faults include attempting an invalid memory access, attempting to execute an instruction in an
invalid processor state, and attempting to execute an undefined instruction.
Exception service routine
See “Interrupt handler”.
Exception vector
See “Interrupt vector”.
Flat address mapping
A system of organizing memory in which each physical address in the memory space is the
same as the corresponding virtual address.
Halfword
A 16-bit data item.
Illegal instruction
An instruction that is architecturally Undefined.
Implementation-defined
The behavior is not architecturally defined, but is defined and documented by individual
implementations.
Implementation-specific

225
6430F–ATARM–21-Feb-12

The behavior is not architecturally defined, and does not have to be documented by individual
implementations. Used when there are a number of implementation options available and the
option chosen does not affect software compatibility.
Index register
In some load and store instruction descriptions, the value of this register is used as an offset to
be added to or subtracted from the base register value to form the address that is sent to memory. Some addressing modes optionally enable the index register value to be shifted prior to the
addition or subtraction.
See also “Base register”
Instruction cycle count
The number of cycles that an instruction occupies the Execute stage of the pipeline.
Interrupt handler
A program that control of the processor is passed to when an interrupt occurs.
Interrupt vector
One of a number of fixed addresses in low memory, or in high memory if high vectors are configured, that contains the first instruction of the corresponding interrupt handler.
Little-endian (LE)
Byte ordering scheme in which bytes of increasing significance in a data word are stored at
increasing addresses in memory.
See also “Condition field”, “Endianness”.
Little-endian memory
Memory in which:
a byte or halfword at a word-aligned address is the least significant byte or halfword within the
word at that address
a byte at a halfword-aligned address is the least significant byte within the halfword at that
address.
Load/store architecture
A processor architecture where data-processing operations only operate on register contents,
not directly on memory contents.
Memory Protection Unit (MPU)
Hardware that controls access permissions to blocks of memory. An MPU does not perform any
address translation.
Prefetching
In pipelined processors, the process of fetching instructions from memory to fill up the pipeline
before the preceding instructions have finished executing. Prefetching an instruction does not
mean that the instruction has to be executed.
Read
Reads are defined as memory operations that have the semantics of a load. Reads include the
Thumb instructions LDM, LDR, LDRSH, LDRH, LDRSB, LDRB, and POP.
226

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Region
A partition of memory space.
Reserved
A field in a control register or instruction format is reserved if the field is to be defined by the
implementation, or produces Unpredictable results if the contents of the field are not zero. These
fields are reserved for use in future extensions of the architecture or are implementation-specific.
All reserved bits not used by the implementation must be written as 0 and read as 0.
Should Be One (SBO)
Write as 1, or all 1s for bit fields, by software. Writing as 0 produces Unpredictable results.
Should Be Zero (SBZ)
Write as 0, or all 0s for bit fields, by software. Writing as 1 produces Unpredictable results.
Should Be Zero or Preserved (SBZP)
Write as 0, or all 0s for bit fields, by software, or preserved by writing the same value back that
has been previously read from the same field on the same processor.
Thread-safe
In a multi-tasking environment, thread-safe functions use safeguard mechanisms when accessing shared resources, to ensure correct operation without the risk of shared access conflicts.
Thumb instruction
One or two halfwords that specify an operation for a processor to perform. Thumb instructions
must be halfword-aligned.
Unaligned
A data item stored at an address that is not divisible by the number of bytes that defines the data
size is said to be unaligned. For example, a word stored at an address that is not divisible by
four.
Undefined
Indicates an instruction that generates an Undefined instruction exception.
Unpredictable (UNP)
You cannot rely on the behavior. Unpredictable behavior must not represent security holes.
Unpredictable behavior must not halt or hang the processor, or any parts of the system.
Warm reset
Also known as a core reset. Initializes the majority of the processor excluding the debug controller and debug logic. This type of reset is useful if you are using the debugging features of a
processor.
Word
A 32-bit data item.
Write
Writes are defined as operations that have the semantics of a store. Writes include the Thumb
instructions STM, STR, STRH, STRB, and PUSH.

227
6430F–ATARM–21-Feb-12

228

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
14. Debug and Test Features
14.1

Overview
The SAM3U Series Microcontrollers feature a number of complementary debug and test
capabilities. The Serial Wire/JTAG Debug Port (SWJ-DP) combining a Serial Wire Debug Port
(SW-DP) and JTAG Debug (JTAG-DP) port is used for standard debugging functions, such as
downloading code and single-stepping through programs. It also embeds a serial wire trace.
Figure 14-1. Debug and Test Block Diagram
TMS
TCK/SWCLK
TDI

Boundary
TAP

JTAGSEL

SWJ-DP

TDO/TRACESWO

Reset
and
Test

POR
TST

229
6430F–ATARM–21-Feb-12

14.2
14.2.1

Application Examples
Debug Environment
Figure 14-2 shows a complete debug environment example. The SWJ-DP interface is used for
standard debugging functions, such as downloading code and single-stepping through the program and viewing core and peripheral registers.
Figure 14-2. Application Debug Environment Example

Host Debugger
PC
SWJ-DP
Emulator/Probe

SWJ-DP
Connector

SAM3

SAM3-based Application Board

14.2.2

230

Test Environment
Figure 14-3 shows a test environment example (JTAG Boundary scan). Test vectors are sent
and interpreted by the tester. In this example, the “board in test” is designed using a number of
JTAG-compliant devices. These devices can be connected to form a single scan chain.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 14-3. Application Test Environment Example
Test Adaptor

Tester

JTAG
Probe

JTAG
Connector

Chip n

SAM3

Chip 2

Chip 1

SAM3-based Application Board In Test

14.3

Debug and Test Pin Description
Table 14-1.

Debug and Test Signal List

Signal Name

Function

Type

Active Level

Input/Output

Low

Reset/Test
NRST

Microcontroller Reset

TST

Test Select

Input
SWD/JTAG

TCK/SWCLK

Test Clock/Serial Wire Clock

Input

TDI

Test Data In

Input

TDO/TRACESWO

Test Data Out/Trace Asynchronous
Data Out

TMS/SWDIO

Test Mode Select/Serial Wire
Input/Output

Input

JTAGSEL

JTAG Selection

Input

Note:

14.4
14.4.1

Output(1)

High

1. TDO pin is set in input mode when the Cortex-M3 Core is not in debug mode. Thus an external
pull-up (100 kΩ) must be added to avoid current consumption due to floating input.

Functional Description
Test Pin
One dedicated pin, TST, is used to define the device operating mode. When this pin is at low
level during power-up, the device is in normal operating mode. When at high level, the device is
in test mode or FFPI mode. The TST pin integrates a permanent pull-down resistor of about
15 kΩ, so that it can be left unconnected for normal operation. Note that when setting the TST pin

231
6430F–ATARM–21-Feb-12

to low or high level at power up, it must remain in the same state during the duration of the whole
operation.
14.4.2

Debug Architecture
Figure 14-4 shows the Debug Architecture used in the SAM3. The Cortex-M3 embeds four functional units for debug:
• SWJ-DP (Serial Wire/JTAG Debug Port)
• FPB (Flash Patch Breakpoint)
• DWT (Data Watchpoint and Trace)
• ITM (Instrumentation Trace Macrocell)
• TPIU (Trace Port Interface Unit)
The debug architecture information that follows is mainly dedicated to developers of SWJ-DP
Emulators/Probes and debugging tool vendors for Cortex M3-based microcontrollers. For further
details on SWJ-DP see the Cortex M3 technical reference manual.

Figure 14-4. Debug Architecture
DWT
4 watchpoints
FPB

SWJ-DP

PC sampler
6 breakpoints
data address sampler

SWD/JTAG
ITM

data sampler

software trace
32 channels
interrupt trace

SWO trace
TPIU

time stamping
CPU statistics

14.4.3

Serial Wire/JTAG Debug Port (SWJ-DP)
The Cortex-M3 embeds a SWJ-DP Debug port which is the standard CoreSight™ debug port. It
combines Serial Wire Debug Port (SW-DP), from 2 to 3 pins and JTAG debug Port (JTAG-DP),
5 pins.
By default, the JTAG Debug Port is active. If the host debugger wants to switch to the Serial
Wire Debug Port, it must provide a dedicated JTAG sequence on TMS/SWDIO and
TCK/SWCLK which disables JTAG-DP and enables SW-DP.

232

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
When the Serial Wire Debug Port is active, TDO/TRACESWO can be used for trace. The asynchronous TRACE output (TRACESWO) is multiplexed with TDO. So the asynchronous trace
can only be used with SW-DP, not JTAG-DP.
Table 14-2.

SWJ-DP Pin List

Pin Name

JTAG Port

Serial Wire Debug Port

TMS/SWDIO

TMS

SWDIO

TCK/SWCLK

TCK

SWCLK

TDI

TDO/TRACESWO

TDO

TRACESWO (optional: trace)

SW-DP or JTAG-DP mode is selected when JTAGSEL is low. It is not possible to switch directly
between SWJ-DP and JTAG boundary scan operations. A chip reset must be performed after
JTAGSEL is changed.
14.4.3.1

SW-DP and JTAG-DP Selection Mechanism
Debug port selection mechanism is done by sending specific SWDIOTMS sequence. The JTAGDP is selected by default after reset.
• Switch from JTAG-DP to SW-DP. The sequence is:
– Send more than 50 SWCLKTCK cycles with SWDIOTMS = 1
– Send the 16-bit sequence on SWDIOTMS = 0111100111100111 (0x79E7 MSB first)
– Send more than 50 SWCLKTCK cycles with SWDIOTMS = 1
• Switch from SWD to JTAG. The sequence is:
– Send more than 50 SWCLKTCK cycles with SWDIOTMS = 1
– Send the 16-bit sequence on SWDIOTMS = 0011110011100111 (0x3CE7 MSB first)
– Send more than 50 SWCLKTCK cycles with SWDIOTMS = 1

14.4.4

FPB (Flash Patch Breakpoint)
The FPB:
• Implements hardware breakpoints
• Patches code and data from code space to system space.
The FPB unit contains:
• Two literal comparators for matching against literal loads from Code space, and remapping to
a corresponding area in System space.
• Six instruction comparators for matching against instruction fetches from Code space and
remapping to a corresponding area in System space.
• Alternatively, comparators can also be configured to generate a Breakpoint instruction to the
processor core on a match.

14.4.5

DWT (Data Watchpoint and Trace)
The DWT contains four comparators which can be configured to generate the following:
• PC sampling packets at set intervals
• PC or Data watchpoint packets

233
6430F–ATARM–21-Feb-12

• Watchpoint event to halt core
The DWT contains counters for the items that follow:
• Clock cycle (CYCCNT)
• Folded instructions
• Load Store Unit (LSU) operations
• Sleep Cycles
• CPI (all instruction cycles except for the first cycle)
• Interrupt overhead
14.4.6

ITM (Instrumentation Trace Macrocell)
The ITM is an application driven trace source that supports printf style debugging to trace Operating System (OS) and application events, and emits diagnostic system information. The ITM
emits trace information as packets which can be generated by three different sources with several priority levels:
• Software trace: Software can write directly to ITM stimulus registers. This can be done
thanks to the “printf” function. For more information, refer to Section 14.4.6.1 “How to
Configure the ITM”.
• Hardware trace: The ITM emits packets generated by the DWT.
• Time stamping: Timestamps are emitted relative to packets. The ITM contains a 21-bit
counter to generate the timestamp.

14.4.6.1

How to Configure the ITM
The following example describes how to output trace data in asynchronous trace mode.
• Configure the TPIU for asynchronous trace mode (refer to Section 14.4.6.3 “5.4.3. How to
Configure the TPIU”)
• Enable the write accesses into the ITM registers by writing “0xC5ACCE55” into the
Lock Access Register (Address: 0xE0000FB0)
• Write 0x00010015 into the Trace Control Register:
– Enable ITM
– Enable Synchronization packets
– Enable SWO behavior
– Fix the ATB ID to 1
• Write 0x1 into the Trace Enable Register:
– Enable the Stimulus port 0
• Write 0x1 into the Trace Privilege Register:
– Stimulus port 0 only accessed in privileged mode (Clearing a bit in this register will
result in the corresponding stimulus port being accessible in user mode.)
• Write into the Stimulus port 0 register: TPIU (Trace Port Interface Unit)
The TPIU acts as a bridge between the on-chip trace data and the Instruction Trace Macrocell (ITM).
The TPIU formats and transmits trace data off-chip at frequencies asynchronous to the core.

234

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
14.4.6.2

Asynchronous Mode
The TPIU is configured in asynchronous mode, trace data are output using the single TRACESWO pin. The TRACESWO signal is multiplexed with the TDO signal of the JTAG Debug Port.
As a consequence, asynchronous trace mode is only available when the Serial Wire Debug
mode is selected since TDO signal is used in JTAG debug mode.
Two encoding formats are available for the single pin output:
• Manchester encoded stream. This is the reset value.
• NRZ_based UART byte structure

14.4.6.3

5.4.3. How to Configure the TPIU
This example only concerns the asynchronous trace mode.
• Set the TRCENA bit to 1 into the Debug Exception and Monitor Register (0xE000EDFC) to
enable the use of trace and debug blocks.
• Write 0x2 into the Selected Pin Protocol Register
– Select the Serial Wire Output – NRZ
• Write 0x100 into the Formatter and Flush Control Register
• Set the suitable clock prescaler value into the Async Clock Prescaler Register to scale the
baud rate of the asynchronous output (this can be done automatically by the debugging tool).

14.4.7

IEEE 1149.1 JTAG Boundary Scan
IEEE 1149.1 JTAG Boundary Scan allows pin-level access independent of the device packaging
technology.
IEEE 1149.1 JTAG Boundary Scan is enabled when FWUP, NRSTB and JTAGSEL are high
while TST is tied low during power-up and must be kept in this state during the whole boundary
scan operation. The SAMPLE, EXTEST and BYPASS functions are implemented. In
SWD/JTAG debug mode, the ARM processor responds with a non-JTAG chip ID that identifies
the processor. This is not IEEE 1149.1 JTAG-compliant.
It is not possible to switch directly between JTAG Boundary Scan and SWJ Debug Port operations. A chip reset must be performed after JTAGSEL is changed.
A Boundary-scan Descriptor Language (BSDL) file is provided on Atmel’s web site to set up the
test.

14.4.7.1

JTAG Boundary-scan Register
The Boundary-scan Register (BSR) contains a number of bits which correspond to active pins
and associated control signals.
Each SAM3 input/output pin corresponds to a 3-bit register in the BSR. The OUTPUT bit contains data that can be forced on the pad. The INPUT bit facilitates the observability of data
applied to the pad. The CONTROL bit selects the direction of the pad.
For more information, please refer to BDSL files available for the SAM3U Series.

235
6430F–ATARM–21-Feb-12

14.4.8
ID Code Register
Access: Read-only
31

VERSION
23

PART NUMBER
21

PART NUMBER
15

PART NUMBER
7

MANUFACTURER IDENTITY

• VERSION[31:28]: Product Version Number
Set to 0x0.
• PART NUMBER[27:12]: Product Part Number
Chip Name

Chip ID

SAM3U

0x5B2A

• MANUFACTURER IDENTITY[11:1]
Set to 0x01F.

• Bit[0] Required by IEEE Std. 1149.1
Set to 0x1.
Chip Name
SAM3U

236

JTAG ID Code
05B2_A03F

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
15. Watchdog Timer (WDT)
15.1

Description
The Watchdog Timer can be used to prevent system lock-up if the software becomes trapped in
a deadlock. It features a 12-bit down counter that allows a watchdog period of up to 16 seconds
(slow clock at 32.768 kHz). It can generate a general reset or a processor reset only. In addition,
it can be stopped while the processor is in debug mode or idle mode.

15.2

Block Diagram

Figure 15-1. Watchdog Timer Block Diagram
write WDT_MR
WDT_MR
WDV

WDT_CR
WDRSTT

reload
1

12-bit Down
Counter
WDT_MR
WDD

reload
Current
Value

1/128

SLCK

<= WDD
WDT_MR
WDRSTEN
= 0
wdt_fault
(to Reset Controller)

set
set
read WDT_SR
or
reset

WDERR
reset

WDUNF
reset

wdt_int

WDFIEN
WDT_MR

237
6430F–ATARM–21-Feb-12

15.3

Functional Description
The Watchdog Timer can be used to prevent system lock-up if the software becomes trapped in
a deadlock. It is supplied with VDDCORE. It restarts with initial values on processor reset.
The Watchdog is built around a 12-bit down counter, which is loaded with the value defined in
the field WDV of the Mode Register (WDT_MR). The Watchdog Timer uses the Slow Clock
divided by 128 to establish the maximum Watchdog period to be 16 seconds (with a typical Slow
Clock of 32.768 kHz).
After a Processor Reset, the value of WDV is 0xFFF, corresponding to the maximum value of
the counter with the external reset generation enabled (field WDRSTEN at 1 after a Backup
Reset). This means that a default Watchdog is running at reset, i.e., at power-up. The user must
either disable it (by setting the WDDIS bit in WDT_MR) if he does not expect to use it or must
reprogram it to meet the maximum Watchdog period the application requires.
The Watchdog Mode Register (WDT_MR) can be written only once. Only a processor reset
resets it. Writing the WDT_MR register reloads the timer with the newly programmed mode
parameters.
In normal operation, the user reloads the Watchdog at regular intervals before the timer underflow occurs, by writing the Control Register (WDT_CR) with the bit WDRSTT to 1. The
Watchdog counter is then immediately reloaded from WDT_MR and restarted, and the Slow
Clock 128 divider is reset and restarted. The WDT_CR register is write-protected. As a result,
writing WDT_CR without the correct hard-coded key has no effect. If an underflow does occur,
the “wdt_fault” signal to the Reset Controller is asserted if the bit WDRSTEN is set in the Mode
Register (WDT_MR). Moreover, the bit WDUNF is set in the Watchdog Status Register
(WDT_SR).
To prevent a software deadlock that continuously triggers the Watchdog, the reload of the
Watchdog must occur while the Watchdog counter is within a window between 0 and WDD,
WDD is defined in the WatchDog Mode Register WDT_MR.
Any attempt to restart the Watchdog while the Watchdog counter is between WDV and WDD
results in a Watchdog error, even if the Watchdog is disabled. The bit WDERR is updated in the
WDT_SR and the “wdt_fault” signal to the Reset Controller is asserted.
Note that this feature can be disabled by programming a WDD value greater than or equal to the
WDV value. In such a configuration, restarting the Watchdog Timer is permitted in the whole
range [0; WDV] and does not generate an error. This is the default configuration on reset (the
WDD and WDV values are equal).
The status bits WDUNF (Watchdog Underflow) and WDERR (Watchdog Error) trigger an interrupt, provided the bit WDFIEN is set in the mode register. The signal “wdt_fault” to the reset
controller causes a Watchdog reset if the WDRSTEN bit is set as already explained in the reset
controller programmer Datasheet. In that case, the processor and the Watchdog Timer are
reset, and the WDERR and WDUNF flags are reset.
If a reset is generated or if WDT_SR is read, the status bits are reset, the interrupt is cleared,
and the “wdt_fault” signal to the reset controller is deasserted.
Writing the WDT_MR reloads and restarts the down counter.
While the processor is in debug state or in idle mode, the counter may be stopped depending on
the value programmed for the bits WDIDLEHLT and WDDBGHLT in the WDT_MR.

238

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 15-2. Watchdog Behavior
Watchdog Error

Watchdog Underflow
if WDRSTEN is 1

FFF
Normal behavior

if WDRSTEN is 0

WDV
Forbidden
Window
WDD
Permitted
Window
0

Watchdog
Fault

WDT_CR = WDRSTT

239
6430F–ATARM–21-Feb-12

15.4

Watchdog Timer (WDT) User Interface

Table 15-1.

Offset

Name

0x00

Control Register

0x04
0x08

240

Access

Reset

WDT_CR

Write-only

Mode Register

WDT_MR

Read-write Once

0x3FFF_2FFF

Status Register

WDT_SR

Read-only

0x0000_0000

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
15.4.1
Watchdog Timer Control Register
Register Name:
WDT_CR
Address:

0x400E1250

Access Type:
31

Write-only
30

KEY
23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
–

0
WDRSTT

• WDRSTT: Watchdog Restart
0: No effect.
1: Restarts the Watchdog.
• KEY: Password
Should be written at value 0xA5. Writing any other value in this field aborts the write operation.

241
6430F–ATARM–21-Feb-12

15.4.2
Watchdog Timer Mode Register
Register Name:
WDT_MR
Address:

0x400E1254

Access Type:
31

Read-write Once
30

29
WDIDLEHLT

28
WDDBGHLT

WDD

WDD
15
WDDIS

WDRPROC

WDRSTEN

WDFIEN

WDV
3

WDV

• WDV: Watchdog Counter Value
Defines the value loaded in the 12-bit Watchdog Counter.
• WDFIEN: Watchdog Fault Interrupt Enable
0: A Watchdog fault (underflow or error) has no effect on interrupt.
1: A Watchdog fault (underflow or error) asserts interrupt.
• WDRSTEN: Watchdog Reset Enable
0: A Watchdog fault (underflow or error) has no effect on the resets.
1: A Watchdog fault (underflow or error) triggers a Watchdog reset.
• WDRPROC: Watchdog Reset Processor
0: If WDRSTEN is 1, a Watchdog fault (underflow or error) activates all resets.
1: If WDRSTEN is 1, a Watchdog fault (underflow or error) activates the processor reset.
• WDD: Watchdog Delta Value
Defines the permitted range for reloading the Watchdog Timer.
If the Watchdog Timer value is less than or equal to WDD, writing WDT_CR with WDRSTT = 1 restarts the timer.
If the Watchdog Timer value is greater than WDD, writing WDT_CR with WDRSTT = 1 causes a Watchdog error.
• WDDBGHLT: Watchdog Debug Halt
0: The Watchdog runs when the processor is in debug state.
1: The Watchdog stops when the processor is in debug state.
• WDIDLEHLT: Watchdog Idle Halt
0: The Watchdog runs when the system is in idle mode.
1: The Watchdog stops when the system is in idle state.
• WDDIS: Watchdog Disable
0: Enables the Watchdog Timer.
1: Disables the Watchdog Timer.
242

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
15.4.3
Watchdog Timer Status Register
Register Name:
WDT_SR
Address:

0x400E1258

Access Type:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
WDERR

0
WDUNF

• WDUNF: Watchdog Underflow
0: No Watchdog underflow occurred since the last read of WDT_SR.
1: At least one Watchdog underflow occurred since the last read of WDT_SR.
• WDERR: Watchdog Error
0: No Watchdog error occurred since the last read of WDT_SR.
1: At least one Watchdog error occurred since the last read of WDT_SR.

243
6430F–ATARM–21-Feb-12

244

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
16. Reset Controller (RSTC)
16.1

Overview
The Reset Controller (RSTC), based on power-on reset cells, handles all the resets of the system without any external components. It reports which reset occurred last.
The Reset Controller also drives independently or simultaneously the external reset and the
peripheral and processor resets.

16.2

Block Diagram
Figure 16-1. Reset Controller Block Diagram

Reset Controller
core_backup_reset
rstc_irq
vddcore_nreset

user_reset

NRST
nrst_out

NRST
Manager

Reset
State
Manager

proc_nreset

periph_nreset
exter_nreset

WDRPROC
wd_fault

SLCK

16.3
16.3.1

Functional Description
Reset Controller Overview
The Reset Controller is made up of an NRST Manager and a Reset State Manager. It runs at
Slow Clock and generates the following reset signals:
• proc_nreset: Processor reset line. It also resets the Watchdog Timer.
• periph_nreset: Affects the whole set of embedded peripherals.
• nrst_out: Drives the NRST pin.
These reset signals are asserted by the Reset Controller, either on external events or on software action. The Reset State Manager controls the generation of reset signals and provides a
signal to the NRST Manager when an assertion of the NRST pin is required.
The NRST Manager shapes the NRST assertion during a programmable time, thus controlling
external device resets.

245
6430F–ATARM–21-Feb-12

The Reset Controller Mode Register (RSTC_MR), allowing the configuration of the Reset Controller, is powered with VDDBU, so that its configuration is saved as long as VDDBU is on.
16.3.2

NRST Manager
The NRST Manager samples the NRST input pin and drives this pin low when required by the
Reset State Manager. Figure 16-2 shows the block diagram of the NRST Manager.
Figure 16-2. NRST Manager
RSTC_MR

URSTIEN

RSTC_SR

URSTS
NRSTL

rstc_irq
RSTC_MR

URSTEN

Other
interrupt
sources
user_reset

NRST

RSTC_MR

ERSTL
nrst_out

16.3.2.1

External Reset Timer

exter_nreset

NRST Signal or Interrupt
The NRST Manager samples the NRST pin at Slow Clock speed. When the line is detected low,
a User Reset is reported to the Reset State Manager.
However, the NRST Manager can be programmed to not trigger a reset when an assertion of
NRST occurs. Writing the bit URSTEN at 0 in RSTC_MR disables the User Reset trigger.
The level of the pin NRST can be read at any time in the bit NRSTL (NRST level) in RSTC_SR.
As soon as the pin NRST is asserted, the bit URSTS in RSTC_SR is set. This bit clears only
when RSTC_SR is read.
The Reset Controller can also be programmed to generate an interrupt instead of generating a
reset. To do so, the bit URSTIEN in RSTC_MR must be written at 1.

16.3.2.2

NRST External Reset Control
The Reset State Manager asserts the signal ext_nreset to assert the NRST pin. When this
occurs, the “nrst_out” signal is driven low by the NRST Manager for a time programmed by the
field ERSTL in RSTC_MR. This assertion duration, named EXTERNAL_RESET_LENGTH, lasts
2(ERSTL+1) Slow Clock cycles. This gives the approximate duration of an assertion between 60 μs
and 2 seconds. Note that ERSTL at 0 defines a two-cycle duration for the NRST pulse.
This feature allows the Reset Controller to shape the NRST pin level, and thus to guarantee that
the NRST line is driven low for a time compliant with potential external devices connected on the
system reset.
As the ERSTL field is within RSTC_MR register, which is backed-up, it can be used to shape the
system power-up reset for devices requiring a longer startup time than the Slow Clock Oscillator.

246

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Please note that the NRST output is in high impedance state when the chip is in OFF mode.
16.3.3

Brownout Manager
The Brownout manager is embedded within the Supply Controller, please refer to the Supply
Controller section for a detailed description.

16.3.4

Reset States
The Reset State Manager handles the different reset sources and generates the internal reset
signals. It reports the reset status in the field RSTTYP of the Status Register (RSTC_SR). The
update of the field RSTTYP is performed when the processor reset is released.

16.3.4.1

General Reset
A general reset occurs when a Power-on-reset is detected, an Asynchronous Master Reset
(NRSTB pin) is requested, a Brownout or a Voltage regulation loss is detected by the Supply
controller. The vddcore_nreset signal is asserted by the Supply Controller when a general reset
occurs.
All the reset signals are released and the field RSTTYP in RSTC_SR reports a General Reset.
As the RSTC_MR is reset, the NRST line rises 2 cycles after the vddcore_nreset, as ERSTL
defaults at value 0x0.
Figure 16-3 shows how the General Reset affects the reset signals.

Figure 16-3. General Reset State

SLCK
Any
Freq.

MCK

backup_nreset
Processor Startup
= 2 cycles

proc_nreset
RSTTYP

XXX

0x0 = General Reset

XXX

periph_nreset

NRST
(nrst_out)
EXTERNAL RESET LENGTH
= 2 cycles

247
6430F–ATARM–21-Feb-12

16.3.4.2

Backup Reset
A Backup reset occurs when the chip returns from Backup mode. The vddcore_nreset signal is
asserted by the Supply Controller when a Backup reset occurs.
The field RSTTYP in RSTC_SR is updated to report a Backup Reset.

16.3.4.3

User Reset
The User Reset is entered when a low level is detected on the NRST pin and the bit URSTEN in
RSTC_MR is at 1. The NRST input signal is resynchronized with SLCK to insure proper behavior of the system.
The User Reset is entered as soon as a low level is detected on NRST. The Processor Reset
and the Peripheral Reset are asserted.
The User Reset is left when NRST rises, after a two-cycle resynchronization time and a 3-cycle
processor startup. The processor clock is re-enabled as soon as NRST is confirmed high.
When the processor reset signal is released, the RSTTYP field of the Status Register
(RSTC_SR) is loaded with the value 0x4, indicating a User Reset.
The NRST Manager guarantees that the NRST line is asserted for
EXTERNAL_RESET_LENGTH Slow Clock cycles, as programmed in the field ERSTL. However, if NRST does not rise after EXTERNAL_RESET_LENGTH because it is driven low
externally, the internal reset lines remain asserted until NRST actually rises.

Figure 16-4. User Reset State

SLCK
MCK

Any
Freq.

NRST
Resynch.
2 cycles

Resynch.
2 cycles

Processor Startup
= 2 cycles

proc_nreset
RSTTYP

Any

XXX

0x4 = User Reset

periph_nreset

NRST
(nrst_out)
>= EXTERNAL RESET LENGTH

248

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
16.3.4.4

Software Reset
The Reset Controller offers several commands used to assert the different reset signals. These
commands are performed by writing the Control Register (RSTC_CR) with the following bits at
1:
• PROCRST: Writing PROCRST at 1 resets the processor and the watchdog timer.
• PERRST: Writing PERRST at 1 resets all the embedded peripherals, including the memory
system, and, in particular, the Remap Command. The Peripheral Reset is generally used for
debug purposes.
• EXTRST: Writing EXTRST at 1 asserts low the NRST pin during a time defined by the field
ERSTL in the Mode Register (RSTC_MR).
The software reset is entered if at least one of these bits is set by the software. All these commands can be performed independently or simultaneously. The software reset lasts 3 Slow
Clock cycles.
The internal reset signals are asserted as soon as the register write is performed. This is
detected on the Master Clock (MCK). They are released when the software reset is left, i.e.; synchronously to SLCK.
If EXTRST is set, the nrst_out signal is asserted depending on the programming of the field
ERSTL. However, the resulting falling edge on NRST does not lead to a User Reset.
If and only if the PROCRST bit is set, the Reset Controller reports the software status in the field
RSTTYP of the Status Register (RSTC_SR). Other Software Resets are not reported in
RSTTYP.
As soon as a software operation is detected, the bit SRCMP (Software Reset Command in Progress) is set in the Status Register (RSTC_SR). It is cleared as soon as the software reset is left.
No other software reset can be performed while the SRCMP bit is set, and writing any value in
RSTC_CR has no effect.

249
6430F–ATARM–21-Feb-12

Figure 16-5. Software Reset

SLCK
MCK

Any
Freq.

Write RSTC_CR
Resynch. Processor Startup
1 cycle
= 2 cycles

proc_nreset
if PROCRST=1
RSTTYP

Any

XXX

0x3 = Software Reset

periph_nreset
if PERRST=1
NRST
(nrst_out)
if EXTRST=1
EXTERNAL RESET LENGTH
8 cycles (ERSTL=2)

SRCMP in RSTC_SR

16.3.4.5

Watchdog Reset
The Watchdog Reset is entered when a watchdog fault occurs. This state lasts 3 Slow Clock
cycles.
When in Watchdog Reset, assertion of the reset signals depends on the WDRPROC bit in
WDT_MR:
• If WDRPROC is 0, the Processor Reset and the Peripheral Reset are asserted. The NRST
line is also asserted, depending on the programming of the field ERSTL. However, the
resulting low level on NRST does not result in a User Reset state.
• If WDRPROC = 1, only the processor reset is asserted.
The Watchdog Timer is reset by the proc_nreset signal. As the watchdog fault always causes a
processor reset if WDRSTEN is set, the Watchdog Timer is always reset after a Watchdog
Reset, and the Watchdog is enabled by default and with a period set to a maximum.
When the WDRSTEN in WDT_MR bit is reset, the watchdog fault has no impact on the reset
controller.

250

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 16-6. Watchdog Reset

SLCK
MCK

Any
Freq.

wd_fault
Processor Startup
= 2 cycles

proc_nreset
RSTTYP

Any

XXX

0x2 = Watchdog Reset

periph_nreset
Only if
WDRPROC = 0

NRST
(nrst_out)
EXTERNAL RESET LENGTH
8 cycles (ERSTL=2)

16.3.5

Reset State Priorities
The Reset State Manager manages the following priorities between the different reset sources,
given in descending order:
• General Reset
• Backup Reset
• Watchdog Reset
• Software Reset
• User Reset
Particular cases are listed below:
• When in User Reset:
– A watchdog event is impossible because the Watchdog Timer is being reset by the
proc_nreset signal.
– A software reset is impossible, since the processor reset is being activated.
• When in Software Reset:
– A watchdog event has priority over the current state.
– The NRST has no effect.
• When in Watchdog Reset:
– The processor reset is active and so a Software Reset cannot be programmed.
– A User Reset cannot be entered.

16.3.6

Reset Controller Status Register
The Reset Controller status register (RSTC_SR) provides several status fields:
• RSTTYP field: This field gives the type of the last reset, as explained in previous sections.
251

6430F–ATARM–21-Feb-12

• SRCMP bit: This field indicates that a Software Reset Command is in progress and that no
further software reset should be performed until the end of the current one. This bit is
automatically cleared at the end of the current software reset.
• NRSTL bit: The NRSTL bit of the Status Register gives the level of the NRST pin sampled on
each MCK rising edge.
• URSTS bit: A high-to-low transition of the NRST pin sets the URSTS bit of the RSTC_SR
register. This transition is also detected on the Master Clock (MCK) rising edge (see Figure
16-7). If the User Reset is disabled (URSTEN = 0) and if the interruption is enabled by the
URSTIEN bit in the RSTC_MR register, the URSTS bit triggers an interrupt. Reading the
RSTC_SR status register resets the URSTS bit and clears the interrupt.
Figure 16-7.

Reset Controller Status and Interrupt
MCK
read
RSTC_SR

Peripheral Access

2 cycle
resynchronization

NRST
NRSTL

URSTS
rstc_irq
if (URSTEN = 0) and
(URSTIEN = 1)

252

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
16.4

Reset Controller (RSTC) User Interface

Table 16-1.

Offset

Name

Access

Reset

0x00

Control Register

RSTC_CR

Write-only

0x04

Status Register

RSTC_SR

Read-only

0x0000_0000

0x08

Mode Register

RSTC_MR

Read-write

0x0000_0000

253
6430F–ATARM–21-Feb-12

16.4.1
Name:

Reset Controller Control Register
RSTC_CR

Address:

0x400E1200

Access Type:

Write-only

KEY
23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

8
–

7
–

6
–

5
–

4
–

3
EXTRST

2
PERRST

1
–

0
PROCRST

• PROCRST: Processor Reset
0 = No effect.
1 = If KEY is correct, resets the processor.
• PERRST: Peripheral Reset
0 = No effect.
1 = If KEY is correct, resets the peripherals.
• EXTRST: External Reset
0 = No effect.
1 = If KEY is correct, asserts the NRST pin.
• KEY: Password
Should be written at value 0xA5. Writing any other value in this field aborts the write operation.

254

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
16.4.2
Name:

Reset Controller Status Register
RSTC_SR

Address:

0x400E1204

Access Type:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
SRCMP

16
NRSTL

15
–

14
–

13
–

12
–

11
–

9
RSTTYP

7
–

6
–

5
–

4
–

3
–

2
–

1
–

0
URSTS

• URSTS: User Reset Status
0 = No high-to-low edge on NRST happened since the last read of RSTC_SR.
1 = At least one high-to-low transition of NRST has been detected since the last read of RSTC_SR.
• RSTTYP: Reset Type
Reports the cause of the last processor reset. Reading this RSTC_SR does not reset this field.
RSTTYP

Reset Type

Comments

General Reset

First power-up Reset

Backup Reset

Return from Backup mode

Watchdog Reset

Watchdog fault occurred

Software Reset

Processor reset required by the software

User Reset

NRST pin detected low

• NRSTL: NRST Pin Level
Registers the NRST Pin Level at Master Clock (MCK).
• SRCMP: Software Reset Command in Progress
0 = No software command is being performed by the reset controller. The reset controller is ready for a software command.
1 = A software reset command is being performed by the reset controller. The reset controller is busy.

255
6430F–ATARM–21-Feb-12

16.4.3
Name:

Reset Controller Mode Register
RSTC_MR

Address:

0x400E1208

Access Type:

Read-write

17
–

16
–

1
–

0
URSTEN

KEY
23
–

22
–

21
–

20
–

19
–

18
–

15
–

14
–

13
–

12
–

7
–

6
–

4
URSTIEN

3
–

ERSTL
2
–

• URSTEN: User Reset Enable
0 = The detection of a low level on the pin NRST does not generate a User Reset.
1 = The detection of a low level on the pin NRST triggers a User Reset.
• URSTIEN: User Reset Interrupt Enable
0 = USRTS bit in RSTC_SR at 1 has no effect on rstc_irq.
1 = USRTS bit in RSTC_SR at 1 asserts rstc_irq if URSTEN = 0.
• ERSTL: External Reset Length
This field defines the external reset length. The external reset is asserted during a time of 2(ERSTL+1) Slow Clock cycles. This
allows assertion duration to be programmed between 60 μs and 2 seconds.
• KEY: Password
Should be written at value 0xA5. Writing any other value in this field aborts the write operation.

256

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
17. Real-time Timer (RTT)
17.1

Description
The Real-time Timer is built around a 32-bit counter used to count roll-over events of the programmable 16-bit prescaler which enables counting elapsed seconds from a 32 kHz slow clock
source. It generates a periodic interrupt and/or triggers an alarm on a programmed value.

17.2

Embedded Characteristics
• 32-bit Free-running Counter on prescaled slow clock
• 16-bit Configurable Prescaler
• Interrupt on Alarm

17.3

Block Diagram

Figure 17-1. Real-time Timer
RTT_MR
RTTRST

RTT_MR
RTPRES
RTT_MR

SLCK

RTTINCIEN

reload
16-bit
Divider

set

0
RTT_MR
RTTRST

RTTINC

RTT_SR
1

reset

rtt_int
32-bit
Counter

read
RTT_SR

RTT_MR
ALMIEN

RTT_VR

reset

CRTV
RTT_SR

ALMS
set

rtt_alarm

=
RTT_AR

ALMV

257
6430F–ATARM–21-Feb-12

17.4

Functional Description
The Real-time Timer can be used to count elapsed seconds. It is built around a 32-bit counter
fed by Slow Clock divided by a programmable 16-bit value. The value can be programmed in the
field RTPRES of the Real-time Mode Register (RTT_MR).
Programming RTPRES at 0x00008000 corresponds to feeding the real-time counter with a 1 Hz
signal (if the Slow Clock is 32.768 kHz). The 32-bit counter can count up to 232 seconds, corresponding to more than 136 years, then roll over to 0.
The Real-time Timer can also be used as a free-running timer with a lower time-base. The best
accuracy is achieved by writing RTPRES to 3. Programming RTPRES to 1 or 2 is possible, but
may result in losing status events because the status register is cleared two Slow Clock cycles
after read. Thus if the RTT is configured to trigger an interrupt, the interrupt occurs during 2 Slow
Clock cycles after reading RTT_SR. To prevent several executions of the interrupt handler, the
interrupt must be disabled in the interrupt handler and re-enabled when the status register is
clear.
The Real-time Timer value (CRTV) can be read at any time in the register RTT_VR (Real-time
Value Register). As this value can be updated asynchronously from the Master Clock, it is advisable to read this register twice at the same value to improve accuracy of the returned value.
The current value of the counter is compared with the value written in the alarm register
RTT_AR (Real-time Alarm Register). If the counter value matches the alarm, the bit ALMS in
RTT_SR is set. The alarm register is set to its maximum value, corresponding to 0xFFFF_FFFF,
after a reset.
The bit RTTINC in RTT_SR is set each time the Real-time Timer counter is incremented. This bit
can be used to start a periodic interrupt, the period being one second when the RTPRES is programmed with 0x8000 and Slow Clock equal to 32.768 Hz.
Reading the RTT_SR status register resets the RTTINC and ALMS fields.
Writing the bit RTTRST in RTT_MR immediately reloads and restarts the clock divider with the
new programmed value. This also resets the 32-bit counter.
Note:

258

Because of the asynchronism between the Slow Clock (SCLK) and the System Clock (MCK):
1) The restart of the counter and the reset of the RTT_VR current value register is effective only 2
slow clock cycles after the write of the RTTRST bit in the RTT_MR register.
2) The status register flags reset is taken into account only 2 slow clock cycles after the read of the
RTT_SR (Status Register).

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 17-2. RTT Counting
APB cycle

APB cycle

SCLK

RTPRES - 1
Prescaler
0

RTT

...

ALMV-1

ALMV

ALMV+1

ALMV+2

ALMV+3

RTTINC (RTT_SR)
ALMS (RTT_SR)
APB Interface

read RTT_SR

259
6430F–ATARM–21-Feb-12

17.5

Real-time Timer (RTT) User Interface

Table 17-1.

Offset

Name

Access

Reset

0x00

Mode Register

RTT_MR

Read-write

0x0000_8000

0x04

Alarm Register

RTT_AR

Read-write

0xFFFF_FFFF

0x08

Value Register

RTT_VR

Read-only

0x0000_0000

0x0C

Status Register

RTT_SR

Read-only

0x0000_0000

260

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
17.5.1
Name:

Real-time Timer Mode Register
RTT_MR

Address:

0x400E1230

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
RTTRST

17
RTTINCIEN

16
ALMIEN

RTPRES
7

4
RTPRES

• RTPRES: Real-time Timer Prescaler Value
Defines the number of SLCK periods required to increment the Real-time timer. RTPRES is defined as follows:
RTPRES = 0: The prescaler period is equal to 216 * SCLK period.
RTPRES ≠ 0: The prescaler period is equal to RTPRES * SCLK period.
• ALMIEN: Alarm Interrupt Enable
0 = The bit ALMS in RTT_SR has no effect on interrupt.
1 = The bit ALMS in RTT_SR asserts interrupt.
• RTTINCIEN: Real-time Timer Increment Interrupt Enable
0 = The bit RTTINC in RTT_SR has no effect on interrupt.
1 = The bit RTTINC in RTT_SR asserts interrupt.
• RTTRST: Real-time Timer Restart
0 = No effect.
1 = Reloads and restarts the clock divider with the new programmed value. This also resets the 32-bit counter.

261
6430F–ATARM–21-Feb-12

17.5.2
Name:

Real-time Timer Alarm Register
RTT_AR

Address:

0x400E1234

Access:

Read-write

ALMV
23

20
ALMV

12
ALMV

4
ALMV

• ALMV: Alarm Value
Defines the alarm value (ALMV+1) compared with the Real-time Timer.

262

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
17.5.3
Name:

Real-time Timer Value Register
RTT_VR

Address:

0x400E1238

Access:

Read-only

CRTV
23

20
CRTV

12
CRTV

4
CRTV

• CRTV: Current Real-time Value
Returns the current value of the Real-time Timer.

263
6430F–ATARM–21-Feb-12

SAM3U Series
17.5.4
Name:

Real-time Timer Status Register
RTT_SR

Address:

0x400E123C

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
RTTINC

0
ALMS

• ALMS: Real-time Alarm Status
0 = The Real-time Alarm has not occurred since the last read of RTT_SR.
1 = The Real-time Alarm occurred since the last read of RTT_SR.
• RTTINC: Real-time Timer Increment
0 = The Real-time Timer has not been incremented since the last read of the RTT_SR.
1 = The Real-time Timer has been incremented since the last read of the RTT_SR.

264
6430F–ATARM–21-Feb-12

SAM3U Series
18. Real Time Clock (RTC)
18.1

Description
The Real-time Clock (RTC) peripheral is designed for very low power consumption.
It combines a complete time-of-day clock with alarm and a two-hundred-year Gregorian calendar, complemented by a programmable periodic interrupt. The alarm and calendar registers are
accessed by a 32-bit data bus.
The time and calendar values are coded in binary-coded decimal (BCD) format. The time format
can be 24-hour mode or 12-hour mode with an AM/PM indicator.
Updating time and calendar fields and configuring the alarm fields are performed by a parallel
capture on the 32-bit data bus. An entry control is performed to avoid loading registers with
incompatible BCD format data or with an incompatible date according to the current
month/year/century.

18.2

Block Diagram

Figure 18-1. RTC Block Diagram

18.3

Slow Clock: SLCK

32768 Divider

Bus Interface

Time

Date

Entry
Control

Interrupt
Control

RTC Interrupt

Product Dependencies

18.3.1

Power Management
The Real-time Clock is continuously clocked at 32768 Hz. The Power Management Controller
has no effect on RTC behavior.

18.3.2

Interrupt
RTC interrupt line is connected on one of the internal sources of the interrupt controller. RTC
interrupt requires the interrupt controller to be programmed first.

265
6430F–ATARM–21-Feb-12

18.4

Functional Description
The RTC provides a full binary-coded decimal (BCD) clock that includes century (19/20), year
(with leap years), month, date, day, hours, minutes and seconds.
The valid year range is 1900 to 2099 in Gregorian mode, a two-hundred-year calendar.
The RTC can operate in 24-hour mode or in 12-hour mode with an AM/PM indicator.
Corrections for leap years are included (all years divisible by 4 being leap years). This is correct
up to the year 2099.

18.4.1

Reference Clock
The reference clock is Slow Clock (SLCK). It can be driven internally or by an external 32.768
kHz crystal.
During low power modes of the processor, the oscillator runs and power consumption is critical.
The crystal selection has to take into account the current consumption for power saving and the
frequency drift due to temperature effect on the circuit for time accuracy.

18.4.2

Timing
The RTC is updated in real time at one-second intervals in normal mode for the counters of seconds, at one-minute intervals for the counter of minutes and so on.
Due to the asynchronous operation of the RTC with respect to the rest of the chip, to be certain
that the value read in the RTC registers (century, year, month, date, day, hours, minutes, seconds) are valid and stable, it is necessary to read these registers twice. If the data is the same
both times, then it is valid. Therefore, a minimum of two and a maximum of three accesses are
required.

18.4.3

Alarm
The RTC has five programmable fields: month, date, hours, minutes and seconds.
Each of these fields can be enabled or disabled to match the alarm condition:
• If all the fields are enabled, an alarm flag is generated (the corresponding flag is asserted
and an interrupt generated if enabled) at a given month, date, hour/minute/second.
• If only the “seconds” field is enabled, then an alarm is generated every minute.
Depending on the combination of fields enabled, a large number of possibilities are available to
the user ranging from minutes to 365/366 days.

18.4.4

Error Checking
Verification on user interface data is performed when accessing the century, year, month, date,
day, hours, minutes, seconds and alarms. A check is performed on illegal BCD entries such as
illegal date of the month with regard to the year and century configured.
If one of the time fields is not correct, the data is not loaded into the register/counter and a flag is
set in the validity register. The user can not reset this flag. It is reset as soon as an acceptable
value is programmed. This avoids any further side effects in the hardware. The same procedure
is done for the alarm.
The following checks are performed:
1. Century (check if it is in range 19 - 20)
2. Year (BCD entry check)

266

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
3. Date (check range 01 - 31)
4. Month (check if it is in BCD range 01 - 12, check validity regarding “date”)
5. Day (check range 1 - 7)
6. Hour (BCD checks: in 24-hour mode, check range 00 - 23 and check that AM/PM flag is
not set if RTC is set in 24-hour mode; in 12-hour mode check range 01 - 12)
7. Minute (check BCD and range 00 - 59)
8. Second (check BCD and range 00 - 59)
Note:

18.4.5

If the 12-hour mode is selected by means of the RTC_MODE register, a 12-hour value can be programmed and the returned value on RTC_TIME will be the corresponding 24-hour value. The
entry control checks the value of the AM/PM indicator (bit 22 of RTC_TIME register) to determine
the range to be checked.

Updating Time/Calendar
To update any of the time/calendar fields, the user must first stop the RTC by setting the corresponding field in the Control Register. Bit UPDTIM must be set to update time fields (hour,
minute, second) and bit UPDCAL must be set to update calendar fields (century, year, month,
date, day).
Then the user must poll or wait for the interrupt (if enabled) of bit ACKUPD in the Status Register. Once the bit reads 1, it is mandatory to clear this flag by writing the corresponding bit in
RTC_SCCR. The user can now write to the appropriate Time and Calendar register.
Once the update is finished, the user must reset (0) UPDTIM and/or UPDCAL in the Control
When entering programming mode of the calendar fields, the time fields remain enabled. When
entering the programming mode of the time fields, both time and calendar fields are stopped.
This is due to the location of the calendar logic circuity (downstream for low-power considerations). It is highly recommended to prepare all the fields to be updated before entering
programming mode. In successive update operations, the user must wait at least one second
after resetting the UPDTIM/UPDCAL bit in the RTC_CR (Control Register) before setting these
bits again. This is done by waiting for the SEC flag in the Status Register before setting
UPDTIM/UPDCAL bit. After resetting UPDTIM/UPDCAL, the SEC flag must also be cleared.

267
6430F–ATARM–21-Feb-12

Figure 18-2. Update Sequence

Begin

Prepare TIme or Calendar Fields

Set UPDTIM and/or UPDCAL
bit(s) in RTC_CR

Read RTC_SR
Polling or
IRQ (if enabled)

ACKUPD
=1?

Yes

Clear ACKUPD bit in RTC_SCCR

Update Time and/or Calendar values in
RTC_TIMR/RTC_CALR

Clear UPDTIM and/or UPDCAL bit in
RTC_CR

End

268

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
18.5

Real Time Clock (RTC) User Interface

Table 18-1.

Offset

Name

Access

Reset

0x00

Control Register

RTC_CR

Read-write

0x0

0x04

Mode Register

RTC_MR

Read-write

0x0

0x08

Time Register

RTC_TIMR

Read-write

0x0

0x0C

Calendar Register

RTC_CALR

Read-write

0x01210720

0x10

Time Alarm Register

RTC_TIMALR

Read-write

0x0

0x14

Calendar Alarm Register

RTC_CALALR

Read-write

0x01010000

0x18

Status Register

RTC_SR

Read-only

0x0

0x1C

Status Clear Command Register

RTC_SCCR

Write-only

–

0x20

Interrupt Enable Register

RTC_IER

Write-only

–

0x24

Interrupt Disable Register

RTC_IDR

Write-only

–

0x28

Interrupt Mask Register

RTC_IMR

Read-only

0x0

0x2C

Valid Entry Register

RTC_VER

Read-only

0x0

0x30–0xE0

Reserved Register

–

RTC_WPMR

Read-write

0x00000000

0xE4

Write Protect Mode Register

0xE8–0xF8

Reserved Register

–

0xFC

Reserved Register

–

Note:

if an offset is not listed in the table it must be considered as reserved.

269
6430F–ATARM–21-Feb-12

18.5.1
Name:

RTC Control Register
RTC_CR

Address:

0x400E1260

Access:

Read-write

–

CALEVSEL
9

TIMEVSEL

–

UPDCAL

UPDTIM

This register can only be written if the WPEN bit is cleared in “RTC Write Protect Mode Register” on page 282.
• UPDTIM: Update Request Time Register
0 = No effect.
1 = Stops the RTC time counting.
Time counting consists of second, minute and hour counters. Time counters can be programmed once this bit is set and
acknowledged by the bit ACKUPD of the Status Register.
• UPDCAL: Update Request Calendar Register
0 = No effect.
1 = Stops the RTC calendar counting.
Calendar counting consists of day, date, month, year and century counters. Calendar counters can be programmed once
this bit is set.
• TIMEVSEL: Time Event Selection
The event that generates the flag TIMEV in RTC_SR (Status Register) depends on the value of TIMEVSEL.
Value

Name

Description

MINUTE

Minute change

HOUR

Hour change

MIDNIGHT

Every day at midnight

NOON

Every day at noon

• CALEVSEL: Calendar Event Selection
The event that generates the flag CALEV in RTC_SR depends on the value of CALEVSEL

270

Value

Name

Description

WEEK

Week change (every Monday at time 00:00:00)

MONTH

Month change (every 01 of each month at time 00:00:00)

YEAR

Year change (every January 1 at time 00:00:00)

–

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
18.5.2
Name:

RTC Mode Register
RTC_MR

Address:

0x400E1264

Access:

Read-write

–

HRMOD

• HRMOD: 12-/24-hour Mode
0 = 24-hour mode is selected.
1 = 12-hour mode is selected.

All non-significant bits read zero.

271
6430F–ATARM–21-Feb-12

18.5.3
Name:

RTC Time Register
RTC_TIMR

Address:

0x400E1268

Access:

Read-write

–

AMPM

HOUR
13

–
7

MIN
6

–

SEC

• SEC: Current Second
The range that can be set is 0 - 59 (BCD).
The lowest four bits encode the units. The higher bits encode the tens.
• MIN: Current Minute
The range that can be set is 0 - 59 (BCD).
The lowest four bits encode the units. The higher bits encode the tens.
• HOUR: Current Hour
The range that can be set is 1 - 12 (BCD) in 12-hour mode or 0 - 23 (BCD) in 24-hour mode.
• AMPM: Ante Meridiem Post Meridiem Indicator
This bit is the AM/PM indicator in 12-hour mode.
0 = AM.
1 = PM.

All non-significant bits read zero.

272

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
18.5.4
Name:

RTC Calendar Register
RTC_CALR

Address:

0x400E126C

Access:

Read-write

–

DAY
15

DATE

MONTH
13

YEAR
7

–

CENT

• CENT: Current Century
The range that can be set is 19 - 20 (BCD).
The lowest four bits encode the units. The higher bits encode the tens.
• YEAR: Current Year
The range that can be set is 00 - 99 (BCD).
The lowest four bits encode the units. The higher bits encode the tens.
• MONTH: Current Month
The range that can be set is 01 - 12 (BCD).
The lowest four bits encode the units. The higher bits encode the tens.
• DAY: Current Day in Current Week
The range that can be set is 1 - 7 (BCD).
The coding of the number (which number represents which day) is user-defined as it has no effect on the date counter.
• DATE: Current Day in Current Month
The range that can be set is 01 - 31 (BCD).
The lowest four bits encode the units. The higher bits encode the tens.

All non-significant bits read zero.

273
6430F–ATARM–21-Feb-12

18.5.5
Name:

RTC Time Alarm Register
RTC_TIMALR

Address:

0x400E1270

Access:

Read-write

–

HOUREN

AMPM

HOUR
13

MINEN
7

MIN
6

SECEN

SEC

This register can only be written if the WPEN bit is cleared in “RTC Write Protect Mode Register” on page 282.
• SEC: Second Alarm
This field is the alarm field corresponding to the BCD-coded second counter.
• SECEN: Second Alarm Enable
0 = The second-matching alarm is disabled.
1 = The second-matching alarm is enabled.
• MIN: Minute Alarm
This field is the alarm field corresponding to the BCD-coded minute counter.
• MINEN: Minute Alarm Enable
0 = The minute-matching alarm is disabled.
1 = The minute-matching alarm is enabled.
• HOUR: Hour Alarm
This field is the alarm field corresponding to the BCD-coded hour counter.
• AMPM: AM/PM Indicator
This field is the alarm field corresponding to the BCD-coded hour counter.
• HOUREN: Hour Alarm Enable
0 = The hour-matching alarm is disabled.
1 = The hour-matching alarm is enabled.

274

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
18.5.6
Name:

RTC Calendar Alarm Register
RTC_CALALR

Address:

0x400E1274

Access:

Read-write

DATEEN

–

DATE

MTHEN

–

MONTH

–

This register can only be written if the WPEN bit is cleared in “RTC Write Protect Mode Register” on page 282.
• MONTH: Month Alarm
This field is the alarm field corresponding to the BCD-coded month counter.
• MTHEN: Month Alarm Enable
0 = The month-matching alarm is disabled.
1 = The month-matching alarm is enabled.
• DATE: Date Alarm
This field is the alarm field corresponding to the BCD-coded date counter.
• DATEEN: Date Alarm Enable
0 = The date-matching alarm is disabled.
1 = The date-matching alarm is enabled.

275
6430F–ATARM–21-Feb-12

18.5.7
Name:

RTC Status Register
RTC_SR

Address:

0x400E1278

Access:

Read-only

–

CALEV

TIMEV

SEC

ALARM

ACKUPD

• ACKUPD: Acknowledge for Update
0 = Time and calendar registers cannot be updated.
1 = Time and calendar registers can be updated.
• ALARM: Alarm Flag
0 = No alarm matching condition occurred.
1 = An alarm matching condition has occurred.
• SEC: Second Event
0 = No second event has occurred since the last clear.
1 = At least one second event has occurred since the last clear.
• TIMEV: Time Event
0 = No time event has occurred since the last clear.
1 = At least one time event has occurred since the last clear.
The time event is selected in the TIMEVSEL field in RTC_CR (Control Register) and can be any one of the following
events: minute change, hour change, noon, midnight (day change).
• CALEV: Calendar Event
0 = No calendar event has occurred since the last clear.
1 = At least one calendar event has occurred since the last clear.
The calendar event is selected in the CALEVSEL field in RTC_CR and can be any one of the following events: week
change, month change and year change.

276

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
18.5.8
Name:

RTC Status Clear Command Register
RTC_SCCR

Address:

0x400E127C

Access:

Write-only

–

CALCLR

TIMCLR

SECCLR

ALRCLR

ACKCLR

• ACKCLR: Acknowledge Clear
0 = No effect.
1 = Clears corresponding status flag in the Status Register (RTC_SR).
• ALRCLR: Alarm Clear
0 = No effect.
1 = Clears corresponding status flag in the Status Register (RTC_SR).
• SECCLR: Second Clear
0 = No effect.
1 = Clears corresponding status flag in the Status Register (RTC_SR).
• TIMCLR: Time Clear
0 = No effect.
1 = Clears corresponding status flag in the Status Register (RTC_SR).
• CALCLR: Calendar Clear
0 = No effect.
1 = Clears corresponding status flag in the Status Register (RTC_SR).

277
6430F–ATARM–21-Feb-12

18.5.9
Name:

RTC Interrupt Enable Register
RTC_IER

Address:

0x400E1280

Access:

Write-only

–

CALEN

TIMEN

SECEN

ALREN

ACKEN

• ACKEN: Acknowledge Update Interrupt Enable
0 = No effect.
1 = The acknowledge for update interrupt is enabled.
• ALREN: Alarm Interrupt Enable
0 = No effect.
1 = The alarm interrupt is enabled.
• SECEN: Second Event Interrupt Enable
0 = No effect.
1 = The second periodic interrupt is enabled.
• TIMEN: Time Event Interrupt Enable
0 = No effect.
1 = The selected time event interrupt is enabled.
• CALEN: Calendar Event Interrupt Enable
0 = No effect.
• 1 = The selected calendar event interrupt is enabled.

278

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
18.5.10
Name:

RTC Interrupt Disable Register
RTC_IDR

Address:

0x400E1284

Access:

Write-only

–

CALDIS

TIMDIS

SECDIS

ALRDIS

ACKDIS

• ACKDIS: Acknowledge Update Interrupt Disable
0 = No effect.
1 = The acknowledge for update interrupt is disabled.
• ALRDIS: Alarm Interrupt Disable
0 = No effect.
1 = The alarm interrupt is disabled.
• SECDIS: Second Event Interrupt Disable
0 = No effect.
1 = The second periodic interrupt is disabled.
• TIMDIS: Time Event Interrupt Disable
0 = No effect.
1 = The selected time event interrupt is disabled.
• CALDIS: Calendar Event Interrupt Disable
0 = No effect.
1 = The selected calendar event interrupt is disabled.

279
6430F–ATARM–21-Feb-12

18.5.11
Name:
Address:
Access:

RTC Interrupt Mask Register
RTC_IMR
0x400E1288
Read-only

–

CAL

TIM

SEC

ALR

ACK

• ACK: Acknowledge Update Interrupt Mask
0 = The acknowledge for update interrupt is disabled.
1 = The acknowledge for update interrupt is enabled.
• ALR: Alarm Interrupt Mask
0 = The alarm interrupt is disabled.
1 = The alarm interrupt is enabled.
• SEC: Second Event Interrupt Mask
0 = The second periodic interrupt is disabled.
1 = The second periodic interrupt is enabled.
• TIM: Time Event Interrupt Mask
0 = The selected time event interrupt is disabled.
1 = The selected time event interrupt is enabled.
• CAL: Calendar Event Interrupt Mask
0 = The selected calendar event interrupt is disabled.
1 = The selected calendar event interrupt is enabled.

280

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
18.5.12
Name:

RTC Valid Entry Register
RTC_VER

Address:

0x400E128C

Access:

Read-only

–

NVCALALR

NVTIMALR

NVCAL

NVTIM

• NVTIM: Non-valid Time
0 = No invalid data has been detected in RTC_TIMR (Time Register).
1 = RTC_TIMR has contained invalid data since it was last programmed.
• NVCAL: Non-valid Calendar
0 = No invalid data has been detected in RTC_CALR (Calendar Register).
1 = RTC_CALR has contained invalid data since it was last programmed.
• NVTIMALR: Non-valid Time Alarm
0 = No invalid data has been detected in RTC_TIMALR (Time Alarm Register).
1 = RTC_TIMALR has contained invalid data since it was last programmed.
• NVCALALR: Non-valid Calendar Alarm
0 = No invalid data has been detected in RTC_CALALR (Calendar Alarm Register).
1 = RTC_CALALR has contained invalid data since it was last programmed.

281
6430F–ATARM–21-Feb-12

18.5.13
Name:

RTC Write Protect Mode Register
RTC_WPMR

Address:

0x400E1344

Access:

Read-write

WPKEY
23

20
WPKEY

12
WPKEY

—

WPEN

• WPEN: Write Protect Enable
0 = Disables the Write Protect if WPKEY corresponds to 0x525443 (“RTC” in ASCII).
1 = Enables the Write Protect if WPKEY corresponds to 0x525443 (“RTC” in ASCII).
Protects the registers:
“RTC Mode Register” on page 271
“RTC Time Alarm Register” on page 274
“RTC Calendar Alarm Register” on page 275

282

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
19. Supply Controller (SUPC)
19.1

Description
The Supply Controller (SUPC) controls the supply voltage of the Core of the system and manages the Backup Low Power Mode. In this mode, the current consumption is reduced to a few
microamps for Backup power retention. Exit from this mode is possible on multiple wake-up
sources including events on FWUP or WKUP pins, or a Clock alarm. The SUPC also generates
the Slow Clock by selecting either the Low Power RC oscillator or the Low Power Crystal
oscillator.

19.2

Embedded Characteristics
• Manages the Core Power Supply VDDCORE and the Backup Low Power Mode by
Controlling the Embedded Voltage Regulator
• Generates the Slow Clock SLCK, by Selecting Either the 22-42 kHz Low Power RC Oscillator
or the 32 kHz Low Power Crystal Oscillator
• Supports Multiple Wake Up Sources, for Exit from Backup Low Power Mode
– Force Wake Up Pin, with Programmable Debouncing
– 16 Wake Up Inputs, with Programmable Debouncing
– Real Time Clock Alarm
– Real Time Timer Alarm
– Supply Monitor Detection on VDDUTMI, with Programmable Scan Period and
Voltage Threshold
• A Supply Monitor Detection on VDDUTMI or a Brownout Detection on VDDCORE can Trigger
a Core Reset
• Embeds:
– One 22 to 42 kHz Low Power RC Oscillator
– One 32 kHz Low Power Crystal Oscillator
– One Zero-Power Power-On Reset Cell
– One Software Programmable Supply Monitor, on VDDUTMI Located in Backup
Section
– One Brownout Detector on VDDCORE Located in the Core

283
6430F–ATARM–21-Feb-12

19.3

Block Diagram

Figure 19-1. Supply Controller Block Diagram
VDDBU

VDDIN

vr_standby
Software Controlled
Voltage Regulator

vr_vdd

FWUP

vr_deep

VDDOUT

SHDN
WKUP0 - WKUP15
NRSTB

Supply
Controller

VDDIO
PIOA/B/C
Input / Output Buffers

Zero-Power
Power-on Reset

PIOx
VDDANA

General Purpose
Backup Registers

ADVREF

ADC (front-end)

ADx
SLCK

RTC

rtc_alarm
sm_in

SLCK

RTT

VDDUTMI

Supply
Monitor

sm_on
rtt_alarm

USBx

USB

osc32k_xtal_en
VDDCORE

vddcore_nreset
XIN32
XOUT32

XTALSEL

Xtal 32 kHz
Oscillator
Embedded
32 kHz RC
Oscillator

Slow Clock
SLCK

bodcore_on

Brownout
Detector

bodcore_in
supc_interrupt

osc32k_rc_en

SRAM

Backup Power Supply
Peripherals
vddcore_nreset

proc_nreset
periph_nreset
ice_nreset

Reset
Controller

NRST

Cortex-M3

Matrix
Peripheral
Bridge

FSTT0 - FSTT15(1)
Embedded
12 / 8 / 4 MHz
RC
Oscillator

XIN
XOUT

Main Clock
MAINCK

3 - 20 MHz
XTAL Oscillator
MAINCK

PLLACK
PLLA

MAINCK

Flash

SLCK

Power
Management
Controller

Master Clock
MCK

SLCK

Watchdog
Timer

UPLLCK
UPLL

Core Power Supply

FSTT0 - FSTT15 are possible Fast Startup Sources, generated by WKUP0-WKUP15 Pins,
but are not physical pins.

284

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
19.4
19.4.1

Supply Controller Functional Description
Supply Controller Overview
The device can be divided into two power supply areas:
• The Backup VDDBU Power Supply: including the Supply Controller, a part of the Reset
Controller, the Slow Clock switch, the General Purpose Backup Registers, the Supply
Monitor and the Clock which includes the Real Time Timer and the Real Time Clock
• The Core Power Supply: including the other part of the Reset Controller, the Brownout
Detector, the Processor, the SRAM memory, the FLASH memory and the Peripherals
The Supply Controller (SUPC) controls the supply voltage of the core power supply. The SUPC
intervenes when the VDDUTMI power supply rises (when the system is starting) or when the
Backup Low Power Mode is entered.
The SUPC also integrates the Slow Clock generator which is based on a 32 kHz crystal oscillator and an embedded 32 kHz RC oscillator. The Slow Clock defaults to the RC oscillator, but the
software can enable the crystal oscillator and select it as the Slow Clock source.
The Supply Controller and the VDDUTMI power supply have a reset circuitry based on the
NRSTB pin and a zero-power power-on reset cell. The zero-power power-on reset allows the
SUPC to start properly as soon as the VDDUTMI voltage becomes valid. The NRSTB pin allows
to reset the system from outside.
At startup of the system, once the backup voltage VDDUTMI is valid and the reset pin NRSTB is
not driven low and the embedded 32 kHz RC oscillator is stabilized, the SUPC starts up the core
by sequentially enabling the internal Voltage Regulator, waiting that the core voltage VDDCORE
is valid, then releasing the reset signal of the core “vddcore_nreset” signal.
Once the system has started, the user can program a supply monitor and/or a brownout detector. If the supply monitor detects a voltage on VDDUTMI that is too low, the SUPC can assert the
reset signal of the core “vddcore_nreset” signal until VDDUTMI is valid. Likewise, if the brownout
detector detects a core voltage VDDCORE that is too low, the SUPC can assert the reset signal
“vddcore_nreset” until VDDCORE is valid.
When the Backup Low Power Mode is entered, the SUPC sequentially asserts the reset signal
of the core power supply “vddcore_nreset” and disables the voltage regulator, in order to supply
only the VDDUTMI power supply. In this mode the current consumption is reduced to a few
microamps for Backup part retention. Exit from this mode is possible on multiple wake-up
sources including an event on FWUP pin or WKUP pins, or a Clock alarm. To exit this mode, the
SUPC operates in the same way as system startup.

285
6430F–ATARM–21-Feb-12

19.4.2

Slow Clock Generator
The Supply Controller embeds a slow clock generator that is supplied with the VDDUTMI power
supply. As soon as the VDDUTMI is supplied, both the crystal oscillator and the embedded RC
oscillator are powered up, but only the embedded RC oscillator is enabled. This allows the slow
clock to be valid in a short time (about 100 μs).
The user can select the crystal oscillator to be the source of the slow clock, as it provides a more
accurate frequency. The command is made by writing the Supply Controller Control Register
(SUPC_CR) with the XTALSEL bit at 1. This results in a sequence which first enables the crystal
oscillator, then waits for 32,768 slow clock cycles, then switches the slow clock on the output of
the crystal oscillator and then disables the RC oscillator to save power. The switch of the slow
clock source is glitch free. The OSCSEL bit of the Supply Controller Status Register (SUPC_SR)
allows knowing when the switch sequence is done.
Coming back on the RC oscillator is only possible by shutting down the VDDUTMI power supply.
If the user does not need the crystal oscillator, the XIN32 and XOUT32 pins should be left
unconnected.
The user can also set the crystal oscillator in bypass mode instead of connecting a crystal. In
this case, the user has to provide the external clock signal on XIN32. The input characteristics of
the XIN32 pin are given in the product electrical characteristics section. In order to set the
bypass mode, the OSCBYPASS bit of the Supply Controller Mode Register (SUPC_MR) needs
to be set at 1.

19.4.3

Voltage Regulator Control/Backup Low Power Mode
The Supply Controller can be used to control the embedded 1.8V voltage regulator.
The voltage regulator automatically adapts its quiescent current depending on the required load
current. Please refer to the electrical characteristics section.
The programmer can switch off the voltage regulator, and thus put the device in Backup mode,
by writing the Supply Controller Control Register (SUPC_CR) with the VROFF bit at 1.
This can be done also by using WFE (Wait for Event) Cortex-M3 instruction with the deep mode
bit set to 1.
The Backup mode can also be entered by executing the WFI (Wait for Interrupt) or WFE (Wait for
Event) Cortex-M3 instructions. To select the Backup mode entry mechanism, two options are
available, depending on the SLEEPONEXIT bit in the Cortex-M3 System Control register:
• Sleep-now: if the SLEEPONEXIT bit is cleared, the device enters Backup mode as soon as
the WFI or WFE instruction is executed.
• Sleep-on-exit: if the SLEEPONEXIT bit is set when the WFI instruction is executed, the
device enters Backup mode as soon as it exits the lowest priority ISR.
This asserts the vddcore_nreset signal after the write resynchronization time which lasts, in the
worse case, two slow clock cycles. Once the vddcore_nreset signal is asserted, the processor
and the peripherals are stopped one slow clock cycle before the core power supply shuts off.
When the user does not use the internal voltage regulator and wants to supply VDDCORE by an
external supply, it is possible to disable the voltage regulator. Note that it is different from the
Backup mode. Depending on the application, disabling the voltage regulator can reduce power
consumption as the voltage regulator input (VDDIN) is shared with the ADC and DAC. This is
done through ONREG bit in SUPC_MR.

286

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
19.4.4

Using Backup Batteries/Backup Supply
The product can be used with or without backup batteries, or more generally a backup supply.
When a backup supply is used (See Figure 19-2), only VDDBU voltage is present in Backup
mode and no other external supply is applied on the chip. In this case the user needs to clear
VDDIORDY bit in the Supply Controller Mode Register (SUPC_MR) at least two slow clock periods before VDDIO voltage is removed. When waking up from Backup mode, the programmer
needs to set VDDIORDY.
Figure 19-2. Separated Backup Supply Powering Scheme
FWUP

SHDN

Backup Batteries VDDBU

VDDUTMI

VDDANA

VDDIO

VDDIN
Voltage
Regulator

Main Supply (1.8V-3.6V)
VDDOUT

VDDCORE

VDDPLL

Note:

Restrictions: With Main Supply < 3V, some peripherals such as USB and ADC might not be operational. Refer to the DC Characteristics of the product for actual possible ranges for such
peripherals.

When a separated backup supply for VDDBU is not used (See Figure 19-3), since the external
voltage applied on VDDIO is kept, all of the I/O configurations (i.e. WKUP pin configuration) are
kept during backup mode. When not using backup batteries, VDDIORDY is set so the user does
not need to program it.

287
6430F–ATARM–21-Feb-12

Figure 19-3. No Separated Backup Supply Powering Scheme

VDDBU

VDDUTMI

VDDANA

VDDIO

Main Supply (1.8V-3.6V)

VDDIN
Voltage
Regulator
VDDOUT

VDDCORE

VDDPLL

Note:

19.4.5

Restrictions: With Main Supply < 3V, some peripherals such as USB and ADC might not be operational. Refer to the DC Characteristics of the product for actual possible ranges for such
peripherals.

Supply Monitor
The Supply Controller embeds a supply monitor which is located in the VDDBU Backup Power
Supply and which monitors VDDUTMI power supply.
The supply monitor can be used to prevent the processor from falling into an unpredictable state
if the Main power supply drops below a certain level.
The threshold of the supply monitor is programmable. It can be selected from 1.9V to 3.4V by
steps of 100 mV. This threshold is programmed in the SMTH field of the Supply Controller Supply Monitor Mode Register (SUPC_SMMR).
The supply monitor can also be enabled during one slow clock period on every one of either 32,
256 or 2048 slow clock periods, according to the choice of the user. This can be configured by
programming the SMSMPL field in SUPC_SMMR.
Enabling the supply monitor for such reduced times allows to divide the typical supply monitor
power consumption respectively by factors of 32, 256 or 2048, if the user does not need a continuous monitoring of the VDDUTMI power supply.

288

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
A supply monitor detection can either generate a reset of the core power supply or a wake up of
the core power supply. Generating a core reset when a supply monitor detection occurs is
enabled by writing the SMRSTEN bit to 1 in SUPC_SMMR.
Waking up the core power supply when a supply monitor detection occurs can be enabled by
programming the SMEN bit to 1 in the Supply Controller Wake Up Mode Register
(SUPC_WUMR).
The Supply Controller provides two status bits in the Supply Controller Status Register for the
supply monitor which allows to determine whether the last wake up was due to the supply
monitor:
• The SMOS bit provides real time information, which is updated at each measurement cycle
or updated at each Slow Clock cycle, if the measurement is continuous.
• The SMS bit provides saved information and shows a supply monitor detection has occurred
since the last read of SUPC_SR.
The SMS bit can generate an interrupt if the SMIEN bit is set to 1 in the Supply Controller Supply
Monitor Mode Register (SUPC_SMMR).
Figure 19-4. Supply Monitor Status Bit and Associated Interrupt
Continuous Sampling (SMSMPL = 1)
Periodic Sampling

Supply Monitor ON

3.3 V
Threshold

0V
Read SUPC_SR

SMS and SUPC interrupt

289
6430F–ATARM–21-Feb-12

19.4.6

Backup Power Supply Reset

19.4.6.1

Raising the Backup Power Supply
As soon as the backup voltage VDDUTMI rises, the RC oscillator is powered up and the zeropower power-on reset cell maintains its output low as long as VDDUTMI has not reached its target voltage. During this time, the Supply Controller is entirely reset. When the VDDUTMI voltage
becomes valid and zero-power power-on reset signal is released, a counter is started for 5 slow
clock cycles. This is the time it takes for the 32 kHz RC oscillator to stabilize.
After this time, the SHDN pin is asserted and the voltage regulator is enabled. The core power
supply rises and the brownout detector provides the bodcore_in signal as soon as the core voltage VDDCORE is valid. This results in releasing the vddcore_nreset signal to the Reset
Controller after the bodcore_in signal has been confirmed as being valid for at least one slow
clock cycle.

Figure 19-5. Raising the VDDUTMI Power Supply
7 x Slow Clock Cycles

Backup Power Supply

TON Voltage
Regulator

3 x Slow Clock
Cycles

6.5 x Slow Clock
Cycles

Zero-Power POR

Zero-Power Power-On
Reset Cell output
22 - 42 kHz RC
Oscillator output
SHDN / vr_on

Core Power Supply

Fast RC
Oscillator output
bodcore_in

vddcore_nreset
NRST
periph_nreset

proc_nreset

Note: After “proc_nreset” rising, the core starts fecthing instructions from Flash at 4 MHz.

290

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
19.4.6.2

NRSTB Asynchronous Reset Pin
The NRSTB pin is an asynchronous reset input, which acts exactly like the zero-power power-on
reset cell.
As soon as NRSTB is tied to GND, the supply controller is reset generating in turn, a reset of the
whole system.
When NRSTB is released, the system can start as described in Section 19.4.6.1 ”Raising the
Backup Power Supply”.
The NRSTB pin does not need to be driven during power-up phase to allow a reset of the system, it is done by the zero-power power-on cell.

Figure 19-6. NRSTB Reset
30 Slow Clock Cycles = about 1ms

between 2 and 3 Slow Clock Cycles

NRSTB
32 kHz Low Power Crystal
Oscillator output
SHDN / vr_standby

bodcore_in

vddcore_nreset

Note:

periph_nreset, ice_reset and proc_nreset are not shown, but are asserted low thanks to the vddcore_nreset signal controlling
the Reset controller.

19.4.6.3

19.4.7

SHDN output pin
As shown in Figure 19-6, the SHDN pin acts like the vr_standby signal making it possible to use
the SHDN pin to control external voltage regulator with shutdown capabilities.
Core Reset
The Supply Controller manages the vddcore_nreset signal to the Reset Controller, as described
previously in Section 19.4.6 ”Backup Power Supply Reset”. The vddcore_nreset signal is normally asserted before shutting down the core power supply and released as soon as the core
power supply is correctly regulated.
There are two additional sources which can be programmed to activate vddcore_nreset:
• a supply monitor detection
• a brownout detection

19.4.7.1

Supply Monitor Reset
The supply monitor is capable of generating a reset of the system. This can be enabled by setting the SMRSTEN bit in the Supply Controller Supply Monitor Mode Register (SUPC_SMMR).
If SMRSTEN is set and if a supply monitor detection occurs, the vddcore_nreset signal is immediately activated for a minimum of 1 slow clock cycle.

291
6430F–ATARM–21-Feb-12

19.4.7.2

Brownout Detector Reset
The brownout detector provides the bodcore_in signal to the SUPC which indicates that the voltage regulation is operating as programmed. If this signal is lost for longer than 1 slow clock
period while the voltage regulator is enabled, the Supply Controller can assert vddcore_nreset.
This feature is enabled by writing the bit, BODRSTEN (Brownout Detector Reset Enable) to 1 in
the Supply Controller Mode Register (SUPC_MR).
If BODRSTEN is set and the voltage regulation is lost (output voltage of the regulator too low),
the vddcore_nreset signal is asserted for a minimum of 1 slow clock cycle and then released if
bodcore_in has been reactivated. The BODRSTS bit is set in the Supply Controller Status Register (SUPC_SR) so that the user can know the source of the last reset.
Until bodcore_in is deactivated, the vddcore_nreset signal remains active.

19.4.8

Wake Up Sources
The wake up events allow the device to exit backup mode. When a wake up event is detected,
the Supply Controller performs a sequence which automatically reenables the core power
supply.

Figure 19-7. Wake Up Sources
SMEN
sm_int
RTCEN
rtc_alarm
Core
Supply
Restart

RTTEN
rtt_alarm
FWUPDBC
SLCK
FWUPEN

FWUP

WKUPT0

WKUP0

292

WKUPIS0

WKUPDBC
WKUPEN1

WKUPIS1

WKUPS

SLCK
Debouncer

Falling/Rising
Edge
Detector
WKUPT15

WKUP15

WKUPEN0

Falling/Rising
Edge
Detector
WKUPT1

WKUP1

FWUP
Debouncer

Falling
Edge
Detector

WKUPEN15

WKUPIS15

Falling/Rising
Edge
Detector

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
19.4.8.1

Force Wake Up
The FWUP pin is enabled as a wake up source by writing the FWUPEN bit to 1 in the Supply
Controller Wake Up Mode Register (SUPC_WUMR). Then, the FWUPDBC field in the same
register selects the debouncing period, which can be selected between 3, 32, 512, 4,096 or
32,768 slow clock cycles. This corresponds respectively to about 100 μs, about 1 ms, about 16
ms, about 128 ms and about 1 second (for a typical slow clock frequency of 32 kHz). Programming FWUPDBC to 0x0 selects an immediate wake up, i.e., the FWUP must be low during a
minimum of one slow clock period to wake up the core power supply.
If the FWUP pin is asserted for a time longer than the debouncing period, a wake up of the core
power supply is started and the FWUP bit in the Supply Controller Status Register (SUPC_SR)
is set and remains high until the register is read.

19.4.8.2

Wake Up Inputs
The wake up inputs, WKUP0 to WKUP15, can be programmed to perform a wake up of the core
power supply. Each input can be enabled by writing to 1 the corresponding bit, WKUPEN0 to
WKUPEN 15, in the Wake Up Inputs Register (SUPC_WUIR). The wake up level can be
selected with the corresponding polarity bit, WKUPPL0 to WKUPPL15, also located in
SUPC_WUIR.
All the resulting signals are wired-ORed to trigger a debounce counter, which can be programmed with the WKUPDBC field in the Supply Controller Wake Up Mode Register
(SUPC_WUMR). The WKUPDBC field can select a debouncing period of 3, 32, 512, 4,096 or
32,768 slow clock cycles. This corresponds respectively to about 100 μs, about 1 ms, about
16 ms, about 128 ms and about 1 second (for a typical slow clock frequency of 32 kHz). Programming WKUPDBC to 0x0 selects an immediate wake up, i.e., an enabled WKUP pin must be
active according to its polarity during a minimum of one slow clock period to wake up the core
power supply.
If an enabled WKUP pin is asserted for a time longer than the debouncing period, a wake up of
the core power supply is started and the signals, WKUP0 to WKUP15 as shown in Figure 19-7,
are latched in the Supply Controller Status Register (SUPC_SR). This allows the user to identify
the source of the wake up, however, if a new wake up condition occurs, the primary information
is lost. No new wake up can be detected since the primary wake up condition has disappeared.

19.4.8.3

Clock Alarms
The RTC and the RTT alarms can generate a wake up of the core power supply. This can be
enabled by writing respectively, the bits RTCEN and RTTEN to 1 in the Supply Controller Wake
Up Mode Register (SUPC_WUMR).
The Supply Controller does not provide any status as the information is available in the User
Interface of either the Real Time Timer or the Real Time Clock.

19.4.8.4

Supply Monitor Detection
The supply monitor can generate a wakeup of the core power supply. See Section 19.4.5 ”Supply Monitor”.

293
6430F–ATARM–21-Feb-12

19.5

Supply Controller (SUPC) User Interface
The User Interface of the Supply Controller is part of the System Controller User Interface.

19.5.1

System Controller (SYSC) User Interface

Table 19-1.

System Controller Registers

Offset

System Controller Peripheral

Name

0x00-0x0c

Reset Controller

RSTC

0x10-0x2C

Supply Controller

SUPC

0x30-0x3C

Real Time Timer

RTT

0x50-0x5C

Watchdog Tiler

WDT

0x60-0x7C

Real Time Clock

RTC

0x90-0xDC

General Purpose Backup Register

GPBR

19.5.2

Supply Controller (SUPC) User Interface

Table 19-2.

Offset

Name

Access

Reset

0x00

Supply Controller Control Register

SUPC_CR

Write-only

N/A

0x04

Supply Controller Supply Monitor Mode Register

SUPC_SMMR

Read-write

0x0000_0000

0x08

Supply Controller Mode Register

SUPC_MR

Read-write

0x0000_5A00

0x0C

Supply Controller Wake Up Mode Register

SUPC_WUMR

Read-write

0x0000_0000

0x10

Supply Controller Wake Up Inputs Register

SUPC_WUIR

Read-write

0x0000_0000

0x14

Supply Controller Status Register

SUPC_SR

Read-only

0x0000_0800

0x18

Reserved

294

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
19.5.3
Name:

Supply Controller Control Register
SUPC_CR

Address:

0x400E1210

Access:

Write-only

KEY
23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

8
–

7
–

6
–

5
–

4
–

3
XTALSEL

2
VROFF

1
–

0
–

• VROFF: Voltage Regulator Off
0 (NO_EFFECT) = no effect.
1 (STOP_VREG) = if KEY is correct, asserts vddcore_nreset and stops the voltage regulator.
• XTALSEL: Crystal Oscillator Select
0 (NO_EFFECT) = no effect.
1 (CRYSTAL_SEL) = if KEY is correct, switches the slow clock on the crystal oscillator output.
• KEY: Password
Should be written to value 0xA5. Writing any other value in this field aborts the write operation.

295
6430F–ATARM–21-Feb-12

19.5.4
Name:

Supply Controller Supply Monitor Mode Register
SUPC_SMMR

Address:

0x400E1214

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
SMIEN

12
SMRSTEN

11
–

9
SMSMPL

7
–

6
–

5
–

4
–

SMTH

• SMTH: Supply Monitor Threshold

296

Value

Name

Description

0x0

1_9V

1.9 V

0x1

2_0V

2.0 V

0x2

2_1V

2.1 V

0x3

2_2V

2.2 V

0x4

2_3V

2.3 V

0x5

2_4V

2.4 V

0x6

2_5V

2.5 V

0x7

2_6V

2.6 V

0x8

2_7V

2.7 V

0x9

2_8V

2.8 V

0xA

2_9V

2.9 V

0xB

3_0V

3.0 V

0xC

3_1V

3.1 V

0xD

3_2V

3.2 V

0xE

3_3V

3.3 V

0xF

3_4V

3.4 V

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• SMSMPL: Supply Monitor Sampling Period
Value

Name

Description

0x0

SMD

Supply Monitor disabled

0x1

CSM

Continuous Supply Monitor

0x2

32SLCK

Supply Monitor enabled one SLCK period every 32 SLCK periods

0x3

256SLCK

Supply Monitor enabled one SLCK period every 256 SLCK periods

0x4

2048SLCK

Supply Monitor enabled one SLCK period every 2,048 SLCK periods

0x5-0x7

Reserved

• SMRSTEN: Supply Monitor Reset Enable
0 (NOT_ENABLE) = the core reset signal “vddcore_nreset” is not affected when a supply monitor detection occurs.
1 (ENABLE) = the core reset signal, vddcore_nreset is asserted when a supply monitor detection occurs.
• SMIEN: Supply Monitor Interrupt Enable
0 (NOT_ENABLE) = the SUPC interrupt signal is not affected when a supply monitor detection occurs.
1 (ENABLE) = the SUPC interrupt signal is asserted when a supply monitor detection occurs.

297
6430F–ATARM–21-Feb-12

19.5.5
Name:

Supply Controller Mode Register
SUPC_MR

Address:

0x400E1218

Access:

Read-write

KEY
23
–

22
–

21
–

20
OSCBYPASS

19
–

18
–

17
–

16
–

14
VDDIORDY
ONREG

BODDIS

BODRSTEN

–

5
–

4
–

3
–

2
–

1
–

0
–

–

7
–

6
–

• BODRSTEN: Brownout Detector Reset Enable
0 (NOT_ENABLE) = the core reset signal “vddcore_nreset” is not affected when a brownout detection occurs.
1 (ENABLE) = the core reset signal, vddcore_nreset is asserted when a brownout detection occurs.
• BODDIS: Brownout Detector Disable
0 (ENABLE) = the core brownout detector is enabled.
1 (DISABLE) = the core brownout detector is disabled.
• VDDIORDY: VDDIO Ready
0 (VDDIO_REMOVED) = VDDIO is removed (used before going to backup mode when backup batteries are used)
1 (VDDIO_PRESENT) = VDDIO is present (used before going to backup mode when backup batteries are used)
If the backup batteries are not used, VDDIORDY must be kept set to 1.
• ONREG: Voltage Regulator enable
0 (ONREG_UNUSED) = Voltage Regulator is not used
1 (ONREG_USED) = Voltage Regulator is used
• OSCBYPASS: Oscillator Bypass
0 (NO_EFFECT) = no effect. Clock selection depends on XTALSEL value.
1 (BYPASS) = the 32-KHz XTAL oscillator is selected and is put in bypass mode.
• KEY: Password Key
Should be written to value 0xA5. Writing any other value in this field aborts the write operation.

298

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
19.5.6
Name:

Supply Controller Wake Up Mode Register
SUPC_WUMR

Address:

0x400E121C

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

13
WKUPDBC

11
–

9
FWUPDBC

7
–

6
–

5
–

4
–

3
RTCEN

2
RTTEN

1
SMEN

0
FWUPEN

• FWUPEN: Force Wake Up Enable
0 (NOT_ENABLE) = the Force Wake Up pin has no wake up effect.
1 (ENABLE) = the Force Wake Up pin low forces the wake up of the core power supply.
• SMEN: Supply Monitor Wake Up Enable
0 (NOT_ENABLE) = the supply monitor detection has no wake up effect.
1 (ENABLE) = the supply monitor detection forces the wake up of the core power supply.
• RTTEN: Real Time Timer Wake Up Enable
0 (NOT_ENABLE) = the RTT alarm signal has no wake up effect.
1 (ENABLE) = the RTT alarm signal forces the wake up of the core power supply.
• RTCEN: Real Time Clock Wake Up Enable
0 (NOT_ENABLE) = the RTC alarm signal has no wake up effect.
1 (ENABLE) = the RTC alarm signal forces the wake up of the core power supply.
• FWUPDBC: Force Wake Up Debouncer Period
Value

Name

Description

IMMEDIATE

3_SCLK

FWUP shall be low for at least 3 SLCK periods

32_SCLK

FWUP shall be low for at least 32 SLCK periods

512_SCLK

FWUP shall be low for at least 512 SLCK periods

4096_SCLK

FWUP shall be low for at least 4,096 SLCK periods

32768_SCLK

FWUP shall be low for at least 32,768 SLCK periods

Reserved

Immediate, no debouncing, detected active at least on one Slow Clock edge.

299
6430F–ATARM–21-Feb-12

• WKUPDBC: Wake Up Inputs Debouncer Period

300

Value

Name

Description

IMMEDIATE

3_SCLK

WKUPx shall be in its active state for at least 3 SLCK periods

32_SCLK

WKUPx shall be in its active state for at least 32 SLCK periods

512_SCLK

WKUPx shall be in its active state for at least 512 SLCK periods

4096_SCLK

WKUPx shall be in its active state for at least 4,096 SLCK periods

32768_SCLK

WKUPx shall be in its active state for at least 32,768 SLCK periods

Reserved

Immediate, no debouncing, detected active at least on one Slow Clock edge.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
19.5.7
Name:

System Controller Wake Up Inputs Register
SUPC_WUIR

Address:

0x400E1220

Access:

Read-write

31
WKUPT15

30
WKUPT14

29
WKUPT13

28
WKUPT12

27
WKUPT11

26
WKUPT10

25
WKUPT9

24
WKUPT8

23
WKUPT7

22
WKUPT6

21
WKUPT5

20
WKUPT4

19
WKUPT3

18
WKUPT2

17
WKUPT1

16
WKUPT0

15
WKUPEN15

14
WKUPEN14

13
WKUPEN13

12
WKUPEN12

11
WKUPEN11

10
WKUPEN10

9
WKUPEN9

8
WKUPEN8

7
WKUPEN7

6
WKUPEN6

5
WKUPEN5

4
WKUPEN4

3
WKUPEN3

2
WKUPEN2

1
WKUPEN1

0
WKUPEN0

• WKUPEN0 - WKUPEN15: Wake Up Input Enable 0 to 15
0 (NOT_ENABLE) = the corresponding wake-up input has no wake up effect.
1 (ENABLE) = the corresponding wake-up input forces the wake up of the core power supply.
• WKUPT0 - WKUPT15: Wake Up Input Transition 0 to 15
0 (HIGH_TO_LOW) = a high to low level transition on the corresponding wake-up input forces the wake up of the core
power supply.
1 (LOW_TO_HIGH) = a low to high level transition on the corresponding wake-up input forces the wake up of the core
power supply.

301
6430F–ATARM–21-Feb-12

19.5.8
Name:

Supply Controller Status Register
SUPC_SR

Address:

0x400E1224

Access:

Read-write

31
WKUPIS15

30
WKUPIS14

29
WKUPIS13

28
WKUPIS12

27
WKUPIS11

26
WKUPIS10

25
WKUPIS9

24
WKUPIS8

23
WKUPIS7

22
WKUPIS6

21
WKUPIS5

20
WKUPIS4

19
WKUPIS3

18
WKUPIS2

17
WKUPIS1

16
WKUPIS0

15
–

14
–

13
–

12
FWUPIS

11
–

10
–

9
–

8
–

7
OSCSEL

6
SMOS

5
SMS

4
SMRSTS

3
BODRSTS

2
SMWS

1
WKUPS

0
FWUPS

Note:

Because of the asynchronism between the Slow Clock (SCLK) and the System Clock (MCK), the status register flag reset is
taken into account only 2 slow clock cycles after the read of the SUPC_SR.

• FWUPS: FWUP Wake Up Status
0 (NO) = no wake up due to the assertion of the FWUP pin has occurred since the last read of SUPC_SR.
1 (PRESENT) = at least one wake up due to the assertion of the FWUP pin has occurred since the last read of SUPC_SR.
• WKUPS: WKUP Wake Up Status
0 (NO) = no wake up due to the assertion of the WKUP pins has occurred since the last read of SUPC_SR.
1 (PRESENT) = at least one wake up due to the assertion of the WKUP pins has occurred since the last read of SUPC_SR.
• SMWS: Supply Monitor Detection Wake Up Status
0 (NO) = no wake up due to a supply monitor detection has occurred since the last read of SUPC_SR.
1 (PRESENT) = at least one wake up due to a supply monitor detection has occurred since the last read of SUPC_SR.
• BODRSTS: Brownout Detector Reset Status
0 (NO) = no core brownout rising edge event has been detected since the last read of the SUPC_SR.
1 (PRESENT) = at least one brownout output rising edge event has been detected since the last read of the SUPC_SR.
When the voltage remains below the defined threshold, there is no rising edge event at the output of the brownout detection cell. The rising edge event occurs only when there is a voltage transition below the threshold.
• SMRSTS: Supply Monitor Reset Status
0 (NO) = no supply monitor detection has generated a core reset since the last read of the SUPC_SR.
1 (PRESENT) = at least one supply monitor detection has generated a core reset since the last read of the SUPC_SR.
• SMS: Supply Monitor Status
0 (NO) = no supply monitor detection since the last read of SUPC_SR.
1 (PRESENT) = at least one supply monitor detection since the last read of SUPC_SR.

302

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• SMOS: Supply Monitor Output Status
0 (HIGH) = the supply monitor detected VDDUTMI higher than its threshold at its last measurement.
1 (LOW) = the supply monitor detected VDDUTMI lower than its threshold at its last measurement.
• OSCSEL: 32-kHz Oscillator Selection Status
0 (RC) = the slow clock, SLCK is generated by the embedded 32-kHz RC oscillator.
1 (CRYST) = the slow clock, SLCK is generated by the 32-kHz crystal oscillator.
• FWUPIS: FWUP Input Status
0 (LOW) = FWUP input is tied low.
1 (HIGH) = FWUP input is tied high.
• WKUPIS0-WKUPIS15: WKUP Input Status 0 to 15
0 (DIS) = the corresponding wake-up input is disabled, or was inactive at the time the debouncer triggered a wake up
event.
1 (EN) = the corresponding wake-up input was active at the time the debouncer triggered a wake up event.

303
6430F–ATARM–21-Feb-12

304

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
20. General Purpose Backup Registers (GPBR)
20.1

Embedded Characteristics
• Four 32-bit General Purpose Backup Registers

20.2

Description
The System Controller embeds Four general-purpose backup registers.

20.2.1

Power Management Controller (PMC) User Interface

Table 20-1.

Offset
0x0
...
0xc

Name

General Purpose Backup Register 0

SYS_GPBR0

...

General Purpose Backup Register 3

SYS_GPBR3

Access

Reset

Read-write

–

...

Read-write

–

305
6430F–ATARM–21-Feb-12

20.2.1.1
Name:

General Purpose Backup Register x
SYS_GPBRx

Address:

0x400E1290 [0] .. 0x400E129C [3]

Access:

Read-write

GPBR_VALUEx
23

GPBR_VALUEx
15

GPBR_VALUEx
7

GPBR_VALUEx

• GPBR_VALUEx: Value of GPBR x

306

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
21. Enhanced Embedded Flash Controller (EEFC)
21.1

Description
The Enhanced Embedded Flash Controller (EEFC) ensures the interface of the Flash block with
the 32-bit internal bus.
Its 128-bit or 64-bit wide memory interface increases performance. It also manages the programming, erasing, locking and unlocking sequences of the Flash using a full set of commands.
One of the commands returns the embedded Flash descriptor definition that informs the system
about the Flash organization, thus making the software generic.

21.2

Embedded Characteristics
• Interface of the Flash Block with the 32-bit Internal Bus
• Increases Performance in Thumb2 Mode with 128-bit or -64 bit Wide Memory Interface up to
24 MHz
• 32 Lock Bits, Each Protecting a Lock Region
• GPNVMx General-purpose GPNVM Bits
• One-by-one Lock Bit Programming
• Commands Protected by a Keyword
• Erases the Entire Flash
• Erases by Plane
• Possibility of Erasing before Programming
• Locking and Unlocking Operations
• Consecutive Programming and Locking Operations

21.3

Product Dependencies

21.3.1

Power Management
The Enhanced Embedded Flash Controller (EEFC) is continuously clocked. The Power Management Controller has no effect on its behavior.

21.3.2

Interrupt Sources
The Enhanced Embedded Flash Controller (EEFC) interrupt line is connected to the Nested
Vectored Interrupt Controller (NVIC). Using the Enhanced Embedded Flash Controller (EEFC)
interrupt requires the NVIC to be programmed first. The EEFC interrupt is generated only on
FRDY bit rising.

21.4
21.4.1

Functional Description
Embedded Flash Organization
The embedded Flash interfaces directly with the 32-bit internal bus. The embedded Flash is
composed of:
• One memory plane organized in several pages of the same size.
• Two 128-bit or 64-bit read buffers used for code read optimization.
• One 128-bit or 64-bit read buffer used for data read optimization.

307
6430F–ATARM–21-Feb-12

SAM3U Series
• One write buffer that manages page programming. The write buffer size is equal to the page
size. This buffer is write-only and accessible all along the 1 MByte address space, so that
each word can be written to its final address.
• Several lock bits used to protect write/erase operation on several pages (lock region). A lock
bit is associated with a lock region composed of several pages in the memory plane.
• Several bits that may be set and cleared through the Enhanced Embedded Flash Controller
(EEFC) interface, called General Purpose Non Volatile Memory bits (GPNVM bits).
The embedded Flash size, the page size, the lock regions organization and GPNVM bits definition are described in the product definition section. The Enhanced Embedded Flash Controller
(EEFC) returns a descriptor of the Flash controlled after a get descriptor command issued by the
application (see “Getting Embedded Flash Descriptor” on page 312).
Figure 21-1. Embedded Flash Organization
Memory Plane

Start Address

Page 0

Lock Region 0

Lock Bit 0

Lock Region 1

Lock Bit 1

Lock Region (n-1)

Lock Bit (n-1)

Page (m-1)

Start Address + Flash size -1

Page (n*m-1)

308
6430F–ATARM–21-Feb-12

SAM3U Series
21.4.2

Read Operations
An optimized controller manages embedded Flash reads, thus increasing performance when the
processor is running in Thumb2 mode by means of the 128- or 64- bit wide memory interface.
The Flash memory is accessible through 8-, 16- and 32-bit reads.
As the Flash block size is smaller than the address space reserved for the internal memory area,
the embedded Flash wraps around the address space and appears to be repeated within it.
The read operations can be performed with or without wait states. Wait states must be programmed in the field FWS (Flash Read Wait State) in the Flash Mode Register (EEFC_FMR).
Defining FWS to be 0 enables the single-cycle access of the embedded Flash. Refer to the Electrical Characteristics for more details.

21.4.2.1

128-bit or 64-bit Access Mode
By default the read accesses of the Flash are performed through a 128-bit wide memory interface. It enables better system performance especially when 2 or 3 wait state needed.
For systems requiring only 1 wait state, or to privilege current consumption rather than performance, the user can select a 64-bit wide memory access via the FAM bit in the Flash Mode
Register (EEFC_FMR)
Please refer to the electrical characteristics section of the product datasheet for more details.

21.4.2.2

Code Read Optimization
A system of 2 x 128-bit or 2 x 64-bit buffers is added in order to optimize sequential Code Fetch.
Note:

Immediate consecutive code read accesses are not mandatory to benefit from this optimization.

Figure 21-2. Code Read Optimization for FWS = 0
Master Clock

ARM Request
(32-bit)
@Byte 0

Flash Access

Buffer 0 (128bits)

Buffer 1 (128bits)

Data To ARM XXX

Note:

@Byte 4

@Byte 8

Bytes 0-15

Bytes 16-31

XXX

@Byte 12

@Byte 16

@Byte 20

@Byte 24

@Byte 32

Bytes 32-47

Bytes 0-15

XXX

Bytes 0-3

@Byte 28

Bytes 16-31

Bytes 4-7

Bytes 8-11

Bytes 12-15

Bytes 16-19

Bytes 20-23

Bytes 24-27

Bytes 28-31

When FWS is equal to 0, all the accesses are performed in a single-cycle access.

309
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 21-3. Code Read Optimization for FWS = 3
Master Clock

ARM Request
(32-bit)
@Byte 0

Flash Access

Bytes 0-15

@12 @16

@20

@24

Bytes 16-31

XXX

Buffer 0 (128bits)

@28 @32

@36 @40

Bytes 32-47

XXX

@48 @52

Bytes 48-63

Bytes 32-47

XXX

Data To ARM

@44

Bytes 0-15

Buffer 1 (128bits)

Note:

Bytes 16-31

0-3

4-7

8-11

12-15

16-19

20-23

24-27

28-31 32-35

36-39

40-43

44-47

48-51

When FWS is included between 1 and 3, in case of sequential reads, the first access takes (FWS+1) cycles, the other ones only
1 cycle.

21.4.2.3

Data Read Optimization
The organization of the Flash in 128 bits (or 64 bits) is associated with two 128-bit (or 64-bit)
prefetch buffers and one 128-bit (or 64-bit) data read buffer, thus providing maximum system
performance. This buffer is added in order to store the requested data plus all the data contained
in the 128-bit (64-bit) aligned data. This speeds up sequential data reads if, for example, FWS is
equal to 1 (see Figure 21-4).
Note:

No consecutive data read accesses are mandatory to benefit from this optimization.

Figure 21-4. Data Read Optimization for FWS = 1
Master Clock
ARM Request
(32-bit)
@Byte 0

Flash Access XXX

Buffer (128bits)

Data To ARM

@ 12

@ 16

Bytes 0-15

@ 24

@ 28

4-7

8-11

12-15

@ 36
Bytes 32-47

Bytes 0-15

Bytes 0-3

@ 32

Bytes 16-31

XXX

@ 20

Bytes 16-31

16-19

20-23

24-27

28-31

32-35

310
6430F–ATARM–21-Feb-12

SAM3U Series
21.4.3

Flash Commands
The Enhanced Embedded Flash Controller (EEFC) offers a set of commands such as programming the memory Flash, locking and unlocking lock regions, consecutive programming and
locking and full Flash erasing, etc.

Table 21-1.

Set of Commands

Command

Value

Mnemonic

Get Flash Descriptor

0x00

GETD

Write page

0x01

Write page and lock

0x02

WPL

Erase page and write page

0x03

EWP

Erase page and write page then lock

0x04

EWPL

Erase all

0x05

Set Lock Bit

0x08

SLB

Clear Lock Bit

0x09

CLB

Get Lock Bit

0x0A

GLB

Set GPNVM Bit

0x0B

SGPB

Clear GPNVM Bit

0x0C

CGPB

Get GPNVM Bit

0x0D

GGPB

Start Read Unique Identifier

0x0E

STUI

Stop Read Unique Identifier

0x0F

SPUI

In order to perform one of these commands, the Flash Command Register (EEFC_FCR) has to
be written with the correct command using the FCMD field. As soon as the EEFC_FCR register
is written, the FRDY flag and the FVALUE field in the EEFC_FRR register are automatically
cleared. Once the current command is achieved, then the FRDY flag is automatically set. If an
interrupt has been enabled by setting the FRDY bit in EEFC_FMR, the corresponding interrupt
line of the NVIC is activated. (Note that this is true for all commands except for the STUI Command. The FRDY flag is not set when the STUI command is achieved.)
All the commands are protected by the same keyword, which has to be written in the 8 highest
bits of the EEFC_FCR register.
Writing EEFC_FCR with data that does not contain the correct key and/or with an invalid command has no effect on the whole memory plane, but the FCMDE flag is set in the EEFC_FSR
register. This flag is automatically cleared by a read access to the EEFC_FSR register.
When the current command writes or erases a page in a locked region, the command has no
effect on the whole memory plane, but the FLOCKE flag is set in the EEFC_FSR register. This
flag is automatically cleared by a read access to the EEFC_FSR register.

311
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 21-5. Command State Chart
Read Status: MC_FSR

No
Check if FRDY flag Set

Yes

Write FCMD and PAGENB in Flash Command Register

Read Status: MC_FSR

No
Check if FRDY flag Set

Yes

Check if FLOCKE flag Set

Yes

Locking region violation

Check if FCMDE flag Set

Yes

Bad keyword violation

No
Command Successfull

21.4.3.1

Getting Embedded Flash Descriptor
This command allows the system to learn about the Flash organization. The system can take full
advantage of this information. For instance, a device could be replaced by one with more Flash
capacity, and so the software is able to adapt itself to the new configuration.
To get the embedded Flash descriptor, the application writes the GETD command in the
EEFC_FCR register. The first word of the descriptor can be read by the software application in
the EEFC_FRR register as soon as the FRDY flag in the EEFC_FSR register rises. The next
reads of the EEFC_FRR register provide the following word of the descriptor. If extra read oper-

312
6430F–ATARM–21-Feb-12

SAM3U Series
ations to the EEFC_FRR register are done after the last word of the descriptor has been
returned, then the EEFC_FRR register value is 0 until the next valid command.
Table 21-2.

Flash Descriptor Definition

Symbol

Word Index

Description

FL_ID

Flash Interface Description

FL_SIZE

Flash size in bytes

FL_PAGE_SIZE

Page size in bytes

FL_NB_PLANE

Number of planes.

FL_PLANE[0]

Number of bytes in the first plane.

4 + FL_NB_PLANE - 1

Number of bytes in the last plane.

FL_NB_LOCK

4 + FL_NB_PLANE

Number of lock bits. A bit is associated
with a lock region. A lock bit is used to
prevent write or erase operations in the
lock region.

FL_LOCK[0]

4 + FL_NB_PLANE + 1

Number of bytes in the first lock region.

...
FL_PLANE[FL_NB_PLANE-1]

...

21.4.3.2

Write Commands
Several commands can be used to program the Flash.
Flash technology requires that an erase be done before programming. The full memory plane
can be erased at the same time, or several pages can be erased at the same time (refer to Figure 21-6, "Example of Partial Page Programming", and the paragraph below the figure.). Also, a
page erase can be automatically done before a page write using EWP or EWPL commands.
After programming, the page (the whole lock region) can be locked to prevent miscellaneous
write or erase sequences. The lock bit can be automatically set after page programming using
WPL or EWPL commands.
Data to be written are stored in an internal latch buffer. The size of the latch buffer corresponds
to the page size. The latch buffer wraps around within the internal memory area address space
and is repeated as many times as the number of pages within this address space.
Note:

Writing of 8-bit and 16-bit data is not allowed and may lead to unpredictable data corruption.

Write operations are performed in a number of wait states equal to the number of wait states for
read operations.
Data are written to the latch buffer before the programming command is written to the Flash
Command Register EEFC_FCR. The sequence is as follows:
• Write the full page, at any page address, within the internal memory area address space.
• Programming starts as soon as the page number and the programming command are written
to the Flash Command Register. The FRDY bit in the Flash Programming Status Register
(EEFC_FSR) is automatically cleared.
• When programming is completed, the FRDY bit in the Flash Programming Status Register
(EEFC_FSR) rises. If an interrupt has been enabled by setting the bit FRDY in EEFC_FMR,
the corresponding interrupt line of the NVIC is activated.
Two errors can be detected in the EEFC_FSR register after a programming sequence:
313
6430F–ATARM–21-Feb-12

SAM3U Series
• Command Error: a bad keyword has been written in the EEFC_FCR register.
• Lock Error: the page to be programmed belongs to a locked region. A command must be
previously run to unlock the corresponding region.
By using the WP command, a page can be programmed in several steps if it has been erased
before (see Figure 21-6 below).
Figure 21-6. Example of Partial Page Programming
32-bit wide

X words

FF
FF

FF
FF
FF
FF

X words

FF
FF
FF

X words

FF
FF

32-bit wide

FF
FF

FF FF
...
FF FF
FF FF

CA FE

FF
FF

CA FE
CA FE

FF FF
...
FF FF
FF FF

FF
FF

FF FF
...
FF FF
FF FF

FF
FF

...

Step 1.
Erase All Flash
So Page Y erased

...

32-bit wide

FF
FF

CA FE

CA
CA

FE
FE

CA FE
CA FE

DE CA

FF
FF

DE CA
DE CA

FF
FF

Step 2.
Programming of the second part of Page Y

FF
...
FF FF
FF FF

FF
FF
FF

CA
CA

FE
FE

...

DE CA
DE CA
DE CA

...

FF
FF

Step 3.
Programming of the third part of Page Y

The Partial Programming mode works only with 128-bit (or higher) boundaries. It cannot be used
with boundaries lower than 128 bits (8, 16 or 32-bit for example).
21.4.3.3

Erase Commands
Erase commands are allowed only on unlocked regions. Depending on the Flash memory, several commands can be used to erase the Flash:
• Erase all memory (EA): all memory is erased. The processor must not fetch code from the
Flash memory.
The erase sequence is:
• Erase starts as soon as one of the erase commands and the FARG field are written in the
Flash Command Register.
• When the programming completes, the FRDY bit in the Flash Programming Status Register
(EEFC_FSR) rises. If an interrupt has been enabled by setting the FRDY bit in EEFC_FMR,
the interrupt line of the NVIC is activated.
Two errors can be detected in the EEFC_FSR register after a programming sequence:
• Command Error: a bad keyword has been written in the EEFC_FCR register.
• Lock Error: at least one page to be erased belongs to a locked region. The erase command
has been refused, no page has been erased. A command must be run previously to unlock
the corresponding region.

314
6430F–ATARM–21-Feb-12

SAM3U Series
21.4.3.4

Lock Bit Protection
Lock bits are associated with several pages in the embedded Flash memory plane. This defines
lock regions in the embedded Flash memory plane. They prevent writing/erasing protected
pages.
The lock sequence is:
• The Set Lock command (SLB) and a page number to be protected are written in the Flash
Command Register.
• When the locking completes, the FRDY bit in the Flash Programming Status Register
(EEFC_FSR) rises. If an interrupt has been enabled by setting the FRDY bit in EEFC_FMR,
the interrupt line of the NVIC is activated.
• If the lock bit number is greater than the total number of lock bits, then the command has no
effect. The result of the SLB command can be checked running a GLB (Get Lock Bit)
command.
One error can be detected in the EEFC_FSR register after a programming sequence:
• Command Error: a bad keyword has been written in the EEFC_FCR register.
It is possible to clear lock bits previously set. Then the locked region can be erased or programmed. The unlock sequence is:
• The Clear Lock command (CLB) and a page number to be unprotected are written in the
Flash Command Register.
• When the unlock completes, the FRDY bit in the Flash Programming Status Register
(EEFC_FSR) rises. If an interrupt has been enabled by setting the FRDY bit in EEFC_FMR,
the interrupt line of the NVIC is activated.
• If the lock bit number is greater than the total number of lock bits, then the command has no
effect.
One error can be detected in the EEFC_FSR register after a programming sequence:
• Command Error: a bad keyword has been written in the EEFC_FCR register.
The status of lock bits can be returned by the Enhanced Embedded Flash Controller (EEFC).
The Get Lock Bit status sequence is:
• The Get Lock Bit command (GLB) is written in the Flash Command Register, FARG field is
meaningless.
• Lock bits can be read by the software application in the EEFC_FRR register. The first word
read corresponds to the 32 first lock bits, next reads providing the next 32 lock bits as long as
it is meaningful. Extra reads to the EEFC_FRR register return 0.
For example, if the third bit of the first word read in the EEFC_FRR is set, then the third lock
region is locked.
One error can be detected in the EEFC_FSR register after a programming sequence:
• Command Error: a bad keyword has been written in the EEFC_FCR register.
Note:

21.4.3.5

Access to the Flash in read is permitted when a set, clear or get lock bit command is performed.

GPNVM Bit
GPNVM bits do not interfere with the embedded Flash memory plane. Refer to the product definition section for information on the GPNVM Bit Action.

315
6430F–ATARM–21-Feb-12

SAM3U Series
The set GPNVM bit sequence is:
• Start the Set GPNVM Bit command (SGPB) by writing the Flash Command Register with the
SGPB command and the number of the GPNVM bit to be set.
• When the GPVNM bit is set, the bit FRDY in the Flash Programming Status Register
(EEFC_FSR) rises. If an interrupt was enabled by setting the FRDY bit in EEFC_FMR, the
interrupt line of the NVIC is activated.
• If the GPNVM bit number is greater than the total number of GPNVM bits, then the command
has no effect. The result of the SGPB command can be checked by running a GGPB (Get
GPNVM Bit) command.
One error can be detected in the EEFC_FSR register after a programming sequence:
• Command Error: a bad keyword has been written in the EEFC_FCR register.
It is possible to clear GPNVM bits previously set. The clear GPNVM bit sequence is:
• Start the Clear GPNVM Bit command (CGPB) by writing the Flash Command Register with
CGPB and the number of the GPNVM bit to be cleared.
• When the clear completes, the FRDY bit in the Flash Programming Status Register
(EEFC_FSR) rises. If an interrupt has been enabled by setting the FRDY bit in EEFC_FMR,
the interrupt line of the NVIC is activated.
• If the GPNVM bit number is greater than the total number of GPNVM bits, then the command
has no effect.
One error can be detected in the EEFC_FSR register after a programming sequence:
• Command Error: a bad keyword has been written in the EEFC_FCR register.
The status of GPNVM bits can be returned by the Enhanced Embedded Flash Controller
(EEFC). The sequence is:
• Start the Get GPNVM bit command by writing the Flash Command Register with GGPB. The
FARG field is meaningless.
• GPNVM bits can be read by the software application in the EEFC_FRR register. The first
word read corresponds to the 32 first GPNVM bits, following reads provide the next 32
GPNVM bits as long as it is meaningful. Extra reads to the EEFC_FRR register return 0.
For example, if the third bit of the first word read in the EEFC_FRR is set, then the third GPNVM
bit is active.
One error can be detected in the EEFC_FSR register after a programming sequence:
• Command Error: a bad keyword has been written in the EEFC_FCR register.
Note:

21.4.3.6

Access to the Flash in read is permitted when a set, clear or get GPNVM bit command is
performed.

Security Bit Protection
When the security is enabled, access to the Flash, either through the JTAG/SWD interface or
through the Fast Flash Programming Interface, is forbidden. This ensures the confidentiality of
the code programmed in the Flash.
The security bit is GPNVM0.
Disabling the security bit can only be achieved by asserting the ERASE pin at 1, and after a full
Flash erase is performed. When the security bit is deactivated, all accesses to the Flash are
permitted.

316
6430F–ATARM–21-Feb-12

SAM3U Series
21.4.3.7

Unique Identifier
Each part is programmed with a 128-bit Unique Identifier. It can be used to generate keys for
example.
To read the Unique Identifier the sequence is:
• Send the Start Read unique Identifier command (STUI) by writing the Flash Command
Register with the STUI command.
• When the Unique Identifier is ready to be read, the FRDY bit in the Flash Programming
Status Register (EEFC_FSR) falls.
• The Unique Identifier is located in the first 128 bits of the Flash memory mapping, thus, at the
address 0x80000-0x8000F.
• To stop the Unique Identifier mode, the user needs to send the Stop Read unique Identifier
command (SPUI) by writing the Flash Command Register with the SPUI command.
• When the Stop read Unique Identifier command (SPUI) has been performed, the FRDY bit in
the Flash Programming Status Register (EEFC_FSR) rises. If an interrupt was enabled by
setting the FRDY bit in EEFC_FMR, the interrupt line of the NVIC is activated.
Note that during the sequence, the software can not run out of Flash (or the second plane in
case of dual plane).

21.5

Enhanced Embedded Flash Controller (EEFC) User Interface

The User Interface of the Enhanced Embedded Flash Controller (EEFC) is integrated within the System Controller with
base address 0x400E0800.

Table 21-3.

Offset

Name

Access

Reset State

0x00

EEFC Flash Mode Register

EEFC_FMR

Read-write

0x0

0x04

EEFC Flash Command Register

EEFC_FCR

Write-only

–

0x08

EEFC Flash Status Register

EEFC_FSR

Read-only

0x00000001

0x0C

EEFC Flash Result Register

EEFC_FRR

Read-only

0x0

0x10

Reserved

–

317
6430F–ATARM–21-Feb-12

SAM3U Series
21.5.1
Name:

EEFC Flash Mode Register
EEFC_FMR

Address:

0x400E0800 (0), 0x400E0A00 (1)

Access:

Read-write

Offset:

0x00

–

FAM

–

SCOD

–

FRDY

–

FWS

• FRDY: Ready Interrupt Enable
0: Flash Ready does not generate an interrupt.
1: Flash Ready (to accept a new command) generates an interrupt.
• FWS: Flash Wait State
This field defines the number of wait states for read and write operations:
Number of cycles for Read/Write operations = FWS+1
• SCOD: Sequential Code Optimization Disable
0: The sequential code optimization is enabled.
1: The sequential code optimization is disabled.
No Flash read should be done during change of this register.
• FAM: Flash Access Mode
0: 128-bit access in read Mode only, to enhance access speed.
1: 64-bit access in read Mode only, to enhance power consumption.
No Flash read should be done during change of this register.

318
6430F–ATARM–21-Feb-12

SAM3U Series
21.5.2
Name:

EEFC Flash Command Register
EEFC_FCR

Address:

0x400E0804 (0), 0x400E0A04 (1)

Access:

Write-only

Offset:

0x04

FKEY
23

20
FARG

12
FARG

4
FCMD

• FCMD: Flash Command
This field defines the Flash commands. Refer to “Flash Commands” on page 311.
• FARG: Flash Command Argument
Erase all command

Field is meaningless.

Programming command

FARG defines the page number to be programmed.

Lock command

FARG defines the page number to be locked.

• FKEY: Flash Writing Protection Key
This field should be written with the value 0x5A to enable the command defined by the bits of the register. If the field is written with a different value, the write is not performed and no action is started.

319
6430F–ATARM–21-Feb-12

SAM3U Series
21.5.3
Name:

EEFC Flash Status Register
EEFC_FSR

Address:

0x400E0808 (0), 0x400E0A08 (1)

Access:

Read-only

Offset:

0x08

–

FLOCKE

FCMDE

FRDY

• FRDY: Flash Ready Status
0: The Enhanced Embedded Flash Controller (EEFC) is busy.
1: The Enhanced Embedded Flash Controller (EEFC) is ready to start a new command.
When it is set, this flags triggers an interrupt if the FRDY flag is set in the EEFC_FMR register.
This flag is automatically cleared when the Enhanced Embedded Flash Controller (EEFC) is busy.
• FCMDE: Flash Command Error Status
0: No invalid commands and no bad keywords were written in the Flash Mode Register EEFC_FMR.
1: An invalid command and/or a bad keyword was/were written in the Flash Mode Register EEFC_FMR.
This flag is automatically cleared when EEFC_FSR is read or EEFC_FCR is written.
• FLOCKE: Flash Lock Error Status
0: No programming/erase of at least one locked region has happened since the last read of EEFC_FSR.
1: Programming/erase of at least one locked region has happened since the last read of EEFC_FSR.
This flag is automatically cleared when EEFC_FSR is read or EEFC_FCR is written.

320
6430F–ATARM–21-Feb-12

SAM3U Series
21.5.4
Name:

EEFC Flash Result Register
EEFC_FRR

Address:

0x400E080C (0), 0x400E0A0C (1)

Access:

Read-only

Offset:

0x0C

FVALUE
23

20
FVALUE

12
FVALUE

4
FVALUE

• FVALUE: Flash Result Value
The result of a Flash command is returned in this register. If the size of the result is greater than 32 bits, then the next
resulting value is accessible at the next register read.

321
6430F–ATARM–21-Feb-12

SAM3U Series

322
6430F–ATARM–21-Feb-12

SAM3U Series
22. Fast Flash Programming Interface (FFPI)
22.1

Overview
The Fast Flash Programming Interface provides solutions for high-volume programming using a
standard gang programmer. The parallel interface is fully handshaked and the device is considered to be a standard EEPROM. Additionally, the parallel protocol offers an optimized access to
all the embedded Flash functionalities.
Although the Fast Flash Programming Mode is a dedicated mode for high volume programming,
this mode is not designed for in-situ programming.

22.2
22.2.1

Parallel Fast Flash Programming
Device Configuration
In Fast Flash Programming Mode, the device is in a specific test mode. Only a certain set of pins
is significant. Other pins must be left unconnected.
Figure 22-1. Parallel Programming Interface
VDDBU

TST

VDDBU

NRSTB

VDDBU

FWUP

VDDBU
VDDIO
VDDIN
VDDANA

PGMNCMD

VDDUTMI

RDY

PGMRDY

VDDPLL

NOE

PGMNOE

NCMD

NVALID

PGMNVALID

VDDCORE
GND
GNDBU

MODE[3:0]

PGMM[3:0]

GNDANA

DATA[15:0]

PGMD[15:0]

GNDPLL

XIN

GNDUTMI

0 - 50MHz (VDDCORE)

323
6430F–ATARM–21-Feb-12

Table 22-1.
Signal Name

Signal Description List
Function

Type

Active
Level

Comments

Power
VDDIO

I/O Lines Power Supply

Power

Apply external 3.0V-3.6V

VDDBU

Backup I/O Lines Power Supply

Power

Apply external 3.0V-3.6V

VDDUTMI

UTMI+ Interface Power Supply

Power

Apply external 3.0V-3.6V

VDDANA

ADC Analog Power Supply

Power

Apply external 3.0V-3.6V

VDDIN

Voltage Regulator Input

Power

Apply external 3.0V-3.6V

VDDCORE

Core Power Supply

Power

Apply external 1.65V-1.95V

VDDPLL

PLLs and Oscillator Power Supply

Power

Apply external 1.65V-1.95V

GND

Ground

GNDPLL

Ground

GNDBU

Ground

GNDANA

Ground

GNDUTMI

Ground

Ground
Clocks

XIN

Clock Input

0 to 50MHz (0-VDDCORE square
wave)

Input
Test

TST

Test Mode Select

Input

High

Must be connected to VDDIO

NRSTB

Asynchronous Microcontroller Reset

Input

High

Must be connected to VDDIO

FWUP

Wake-up pin

Input

High

Must be connected to VDDIO

Input

Low

Pulled-up input at reset

Output

High

Pulled-up input at reset

Input

Low

Pulled-up input at reset

Output

Low

Pulled-up input at reset

PIO
PGMNCMD

Valid command available

PGMRDY

0: Device is busy
1: Device is ready for a new command

PGMNOE

Output Enable (active high)

PGMNVALID

0: DATA[15:0] is in input mode
1: DATA[15:0] is in output mode

PGMM[3:0]

Specifies DATA type (See Table 22-2)

PGMD[15:0]

Bi-directional data bus

324

Input

Pulled-up input at reset

Input/Output

Pulled-up input at reset

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
22.2.2

Signal Names
Depending on the MODE settings, DATA is latched in different internal registers.
Table 22-2.

Mode Coding

MODE[3:0]

Symbol

Data

0000

CMDE

Command Register

0001

ADDR0

Address Register LSBs

0010

ADDR1

Address Register MSBs

0101

DATA

Data Register

Default

IDLE

No register

When MODE is equal to CMDE, then a new command (strobed on DATA[15:0] signals) is stored
in the command register.

Table 22-3.
DATA[15:0]

Symbol

Command Executed

0x0011

READ

Read Flash

0x0012

Write Page Flash

0x0022

WPL

Write Page and Lock Flash

0x0032

EWP

Erase Page and Write Page

0x0042

EWPL

Erase Page and Write Page then Lock

0x0013

Erase All

0x0014

SLB

Set Lock Bit

0x0024

CLB

Clear Lock Bit

0x0015

GLB

Get Lock Bit

0x0034

SGPB

Set General Purpose NVM bit

0x0044

CGPB

Clear General Purpose NVM bit

0x0025

GGPB

Get General Purpose NVM bit

0x0054

SSE

Set Security Bit

0x0035

GSE

Get Security Bit

0x001F

WRAM

Write Memory

0x0016

SEFC

Select EEFC Controller(1)

0x001E

GVE

Get Version

Note:

22.2.3

Command Bit Coding

1. Applies to 256 kbytes Flash version (dual EEFC)

Entering Programming Mode
The following algorithm puts the device in Parallel Programming Mode:
• Apply GND, TST, NRTSB, FWUP and the supplies as described in Table 22-1, “Signal
Description List,” on page 324.

325
6430F–ATARM–21-Feb-12

• Apply XIN clock
• Wait for 20 ms
• Start a read or write handshaking.
22.2.4

Programmer Handshaking
A handshake is defined for read and write operations. When the device is ready to start a new
operation (RDY signal set), the programmer starts the handshake by clearing the NCMD signal.
The handshaking is achieved once NCMD signal is high and RDY is high.

22.2.4.1

Write Handshaking
For details on the write handshaking sequence, refer to Figure 22-2 and Table 22-4.
Figure 22-2. Parallel Programming Timing, Write Sequence

NCMD

4
3

RDY

NOE
NVALID

DATA[15:0]
1
MODE[3:0]

Table 22-4.

Write Handshake

Step

Programmer Action

Device Action

Data I/O

Sets MODE and DATA signals

Waits for NCMD low

Input

Clears NCMD signal

Latches MODE and DATA

Input

Waits for RDY low

Clears RDY signal

Input

Releases MODE and DATA signals

Executes command and polls NCMD high

Input

Sets NCMD signal

Executes command and polls NCMD high

Input

Waits for RDY high

Sets RDY

Input

22.2.4.2

326

Read Handshaking
For details on the read handshaking sequence, refer to Figure 22-3 and Table 22-5.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 22-3.

Parallel Programming Timing, Read Sequence

NCMD

2
3

RDY

NOE

NVALID
4
DATA[15:0]

Adress IN

6
Z

Data OUT

1
MODE[3:0]

Table 22-5.

ADDR

Read Handshake

Step

Programmer Action

Device Action

DATA I/O

Sets MODE and DATA signals

Waits for NCMD low

Input

Clears NCMD signal

Latch MODE and DATA

Input

Waits for RDY low

Clears RDY signal

Input

Sets DATA signal in tristate

Waits for NOE Low

Input

Clears NOE signal

Waits for NVALID low

Tristate
Sets DATA bus in output mode and outputs
the flash contents.

Output

Clears NVALID signal

Output

Waits for NOE high

Output

Reads value on DATA Bus

Sets NOE signal

Waits for NVALID high

Sets DATA bus in input mode

Sets DATA in output mode

Sets NVALID signal

Input

Sets NCMD signal

Waits for NCMD high

Input

Waits for RDY high

Sets RDY signal

Input

22.2.5

Output

Device Operations
Several commands on the Flash memory are available. These commands are summarized in
Table 22-3 on page 325. Each command is driven by the programmer through the parallel interface running several read/write handshaking sequences.
When a new command is executed, the previous one is automatically achieved. Thus, chaining
a read command after a write automatically flushes the load buffer in the Flash.

327
6430F–ATARM–21-Feb-12

22.2.5.1

Flash Read Command
This command is used to read the contents of the Flash memory. The read command can start
at any valid address in the memory plane and is optimized for consecutive reads. Read handshaking can be chained; an internal address buffer is automatically increased.
Table 22-6.

22.2.5.2

Read Command

Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

READ

Write handshaking

ADDR0

Memory Address LSB

Write handshaking

ADDR1

Memory Address

Read handshaking

DATA

*Memory Address++

Read handshaking

DATA

*Memory Address++

...

Write handshaking

ADDR0

Memory Address LSB

n+1

Write handshaking

ADDR1

Memory Address

n+2

Read handshaking

DATA

*Memory Address++

n+3

Read handshaking

DATA

*Memory Address++

...

Flash Write Command
This command is used to write the Flash contents.
The Flash memory plane is organized into several pages. Data to be written are stored in a load
buffer that corresponds to a Flash memory page. The load buffer is automatically flushed to the
Flash:
• before access to any page other than the current one
• when a new command is validated (MODE = CMDE)
The Write Page command (WP) is optimized for consecutive writes. Write handshaking can be
chained; an internal address buffer is automatically increased.
Table 22-7.

328

Write Command

Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

WP or WPL or EWP or EWPL

Write handshaking

ADDR0

Memory Address LSB

Write handshaking

ADDR1

Memory Address

Write handshaking

DATA

*Memory Address++

Write handshaking

DATA

*Memory Address++

...

Write handshaking

ADDR0

Memory Address LSB

n+1

Write handshaking

ADDR1

Memory Address

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 22-7.

Write Command (Continued)

Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

n+2

Write handshaking

DATA

*Memory Address++

n+3

Write handshaking

DATA

*Memory Address++

...

The Flash command Write Page and Lock (WPL) is equivalent to the Flash Write Command.
However, the lock bit is automatically set at the end of the Flash write operation. As a lock region
is composed of several pages, the programmer writes to the first pages of the lock region using
Flash write commands and writes to the last page of the lock region using a Flash write and lock
command.
The Flash command Erase Page and Write (EWP) is equivalent to the Flash Write Command.
However, before programming the load buffer, the page is erased.
The Flash command Erase Page and Write the Lock (EWPL) combines EWP and WPL
commands.
22.2.5.3

Flash Full Erase Command
This command is used to erase the Flash memory planes.
All lock regions must be unlocked before the Full Erase command by using the CLB command.
Otherwise, the erase command is aborted and no page is erased.
Table 22-8.

22.2.5.4

Full Erase Command

Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

Write handshaking

DATA

Flash Lock Commands
Lock bits can be set using WPL or EWPL commands. They can also be set by using the Set
Lock command (SLB). With this command, several lock bits can be activated. A Bit Mask is provided as argument to the command. When bit 0 of the bit mask is set, then the first lock bit is
activated.
Likewise, the Clear Lock command (CLB) is used to clear lock bits.
Table 22-9.

Set and Clear Lock Bit Command

Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

SLB or CLB

Write handshaking

DATA

Bit Mask

329
6430F–ATARM–21-Feb-12

Lock bits can be read using Get Lock Bit command (GLB). The nth lock bit is active when the bit
n of the bit mask is set..
Table 22-10. Get Lock Bit Command

22.2.5.5

Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

GLB

Read handshaking

DATA

Lock Bit Mask Status
0 = Lock bit is cleared
1 = Lock bit is set

Flash General-purpose NVM Commands
General-purpose NVM bits (GP NVM bits) can be set using the Set GPNVM command (SGPB).
This command also activates GP NVM bits. A bit mask is provided as argument to the command. When bit 0 of the bit mask is set, then the first GP NVM bit is activated.
Likewise, the Clear GPNVM command (CGPB) is used to clear general-purpose NVM bits. All
the general-purpose NVM bits are also cleared by the EA command. The general-purpose NVM
bit is deactivated when the corresponding bit in the pattern value is set to 1.
Table 22-11. Set/Clear GP NVM Command
Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

SGPB or CGPB

Write handshaking

DATA

GP NVM bit pattern value

General-purpose NVM bits can be read using the Get GPNVM Bit command (GGPB). The nth
GP NVM bit is active when bit n of the bit mask is set..
Table 22-12. Get GP NVM Bit Command

22.2.5.6

Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

GGPB

Read handshaking

DATA

GP NVM Bit Mask Status
0 = GP NVM bit is cleared
1 = GP NVM bit is set

Flash Security Bit Command
A security bit can be set using the Set Security Bit command (SSE). Once the security bit is
active, the Fast Flash programming is disabled. No other command can be run. An event on the
Erase pin can erase the security bit once the contents of the Flash have been erased.
The AT9SAM3U256 security bit is controlled by the EEFC0. To use the Set Security Bit command, the EEFC0 must be selected using the Select EFC command.

330

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series

Table 22-13. Set Security Bit Command
Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

SSE

Write handshaking

DATA

Once the security bit is set, it is not possible to access FFPI. The only way to erase the security
bit is to erase the Flash.
In order to erase the Flash, the user must perform the following:
• Power-off the chip
• Power-on the chip with TST = 0
• Assert Erase during a period of more than 220 ms
• Power-off the chip
Then it is possible to return to FFPI mode and check that Flash is erased.
22.2.5.7

SAM3U 256 Kbytes Flash Select EEFC Command
The commands WPx, EA, xLB, xFB are executed using the current EFC controller. The default
EEFC controller is EEFC0. The Select EEFC command (SEFC) allows selection of the current
EEFC controller.
Table 22-14. Select EFC Command

22.2.5.8

Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

SEFC

Write handshaking

DATA

0 = Select EEFC0
1 = Select EEFC1

Memory Write Command
This command is used to perform a write access to any memory location.
The Memory Write command (WRAM) is optimized for consecutive writes. Write handshaking
can be chained; an internal address buffer is automatically increased.
Table 22-15. Write Command
Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

WRAM

Write handshaking

ADDR0

Memory Address LSB

Write handshaking

ADDR1

Memory Address

Write handshaking

DATA

*Memory Address++

Write handshaking

DATA

*Memory Address++

...

Write handshaking

ADDR0

Memory Address LSB

n+1

Write handshaking

ADDR1

Memory Address

331
6430F–ATARM–21-Feb-12

Table 22-15. Write Command (Continued)

22.2.5.9

Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

n+2

Write handshaking

DATA

*Memory Address++

n+3

Write handshaking

DATA

*Memory Address++

...

Get Version Command
The Get Version (GVE) command retrieves the version of the FFPI interface.
Table 22-16. Get Version Command

332

Step

Handshake Sequence

MODE[3:0]

DATA[15:0]

Write handshaking

CMDE

GVE

Write handshaking

DATA

Version

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
23. SAM3U4/2/1 Boot Program
23.1

Description
The SAM-BA® Boot Program integrates an array of programs permitting download and/or upload
into the different memories of the product.

23.2

Flow Diagram
The Boot Program implements the algorithm in Figure 23-1.
Figure 23-1. Boot Program Algorithm Flow Diagram
No

Device
Setup

No
USB Enumeration
Successful ?

Character # received
from UART?

Yes
Run SAM-BA Monitor

The SAM-BA Boot program seeks to detect a source clock either from the embedded main oscillator with external crystal (main osccillator enabled) or from a 12 MHz signal applied to the XIN
pin (Main oscillator in bypass mode).
If a clock is found from the two possible sources above, the boot program checks to verify that
the frequency is 12 MHz (taking into account the frequency range of the 32 kHz RC oscillator). If
the frequency is 12 MHz, USB activation is allowed, else (no clock or frequency other than
12MHz), the internal 12 MHz RC oscilator is used as main clock and USB clock is not allowed
due to frequency drift of the 12 MHz RC oscillator.

23.3

Device Initialization
Initialization follows the steps described below:
1. Stack setup
2. Setup the Embedded Flash Controller
3. External Clock detection (Quartz or external clock on XIN)
4. If Quartz or external clock is 12.000 MHz, allow USB activation
5. Else, does not allow USB activation and use internal RC 12 MHz
6. Main oscillator frequency detection if no external clock detected
7. Switch Master Clock on Main Oscillator
8. C variable initialization
9. PLLA setup: PLLA is initialized to generate a 48 MHz clock
10. UPLL setup in case of USB activation allowed
11. Disable of the Watchdog
12. Initialization of the UART (115200 bauds, 8, N, 1)
13. Initialization of the USB Device Port (in case of USB activation allowed)
14. Wait for one of the following events

333
6430F–ATARM–21-Feb-12

a. check if USB device enumeration has occured
b.

check if characters have been received in the UART

15. Jump to SAM-BA Monitor (see Section 23.4 ”SAM-BA Monitor”)

23.4

SAM-BA Monitor
The SAM-BA boot principle:
Once the communication interface is identified, to run in an infinite loop waiting for different commands as shown in Table 23-1.
Table 23-1.

Commands Available through the SAM-BA Boot

Command

Action

Argument(s)

Example

write a byte

Address, Value#

O200001,CA#

read a byte

Address,#

o200001,#

write a half word

Address, Value#

H200002,CAFE#

read a half word

Address,#

h200002,#

write a word

Address, Value#

W200000,CAFEDECA#

read a word

Address,#

w200000,#

send a file

Address,#

S200000,#

receive a file

Address, NbOfBytes#

R200000,1234#

Address#

G200200#

display version

No argument

• Write commands: Write a byte (O), a halfword (H) or a word (W) to the target.
– Address: Address in hexadecimal.
– Value: Byte, halfword or word to write in hexadecimal.
– Output: ‘>’.
• Read commands: Read a byte (o), a halfword (h) or a word (w) from the target.
– Address: Address in hexadecimal
– Output: The byte, halfword or word read in hexadecimal following by ‘>’
• Send a file (S): Send a file to a specified address
– Address: Address in hexadecimal
– Output: ‘>’.
Note:

There is a time-out on this command which is reached when the prompt ‘>’ appears before the
end of the command execution.

• Receive a file (R): Receive data into a file from a specified address
– Address: Address in hexadecimal
– NbOfBytes: Number of bytes in hexadecimal to receive
– Output: ‘>’
• Go (G): Jump to a specified address and execute the code
– Address: Address to jump in hexadecimal
– Output: ‘>’
• Get Version (V): Return the SAM-BA boot version
– Output: ‘>’

334

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
23.4.1

UART Serial Port
Communication is performed through the UART initialized to 115200 Baud, 8, n, 1.
The Send and Receive File commands use the Xmodem protocol to communicate. Any terminal
performing this protocol can be used to send the application file to the target. The size of the
binary file to send depends on the SRAM size embedded in the product. In all cases, the size of
the binary file must be lower than the SRAM size because the Xmodem protocol requires some
SRAM memory to work. See, Section 23.5 ”Hardware and Software Constraints”

23.4.2

Xmodem Protocol
The Xmodem protocol supported is the 128-byte length block. This protocol uses a two-character CRC-16 to guarantee detection of a maximum bit error.
Xmodem protocol with CRC is accurate provided both sender and receiver report successful
transmission. Each block of the transfer looks like:
<255-blk #><--128 data bytes--> in which:
– = 01 hex
– = binary number, starts at 01, increments by 1, and wraps 0FFH to 00H (not
to 01)
– <255-blk #> = 1’s complement of the blk#.
– = 2 bytes CRC16
Figure 23-2 shows a transmission using this protocol.
Figure 23-2. Xmodem Transfer Example
Host

Device
C
SOH 01 FE Data[128] CRC CRC
ACK
SOH 02 FD Data[128] CRC CRC
ACK
SOH 03 FC Data[100] CRC CRC
ACK
EOT
ACK

335
6430F–ATARM–21-Feb-12

23.4.3

USB Device Port
A 12.000 MHz Crystal (or 12.000 MHz external clock on XIN) is necessary to use the USB
Device port.
The device uses the USB communication device class (CDC) drivers to take advantage of the
installed PC RS-232 software to talk over the USB. The CDC class is implemented in all
releases of Windows®, from Windows 98SE® to Windows XP®. The CDC document, available at
www.usb.org, describes a way to implement devices such as ISDN modems and virtual COM
ports.
The Vendor ID (VID) is Atmel’s vendor ID 0x03EB. The product ID (PID) is 0x6124. These references are used by the host operating system to mount the correct driver. On Windows systems,
the INF files contain the correspondence between vendor ID and product ID.
For More details about VID/PID for End Product/Systems, please refer to the Vendor ID form
available from the USB Implementers Forum:
http://www.usb.org/developers/vendor/VID_Only_Form_withCCAuth_102407b.pdf
"Unauthorized use of assigned or unassigned USB Vendor ID Numbers and associated Product
ID Numbers is strictly prohibited."
Atmel provides an INF example to see the device as a new serial port and also provides another
custom driver used by the SAM-BA application: atm6124.sys. Refer to the document “USB Basic
Application”, literature number 6123, for more details.

23.4.3.1

Enumeration Process
The USB protocol is a master/slave protocol. This is the host that starts the enumeration sending requests to the device through the control endpoint. The device handles standard requests
as defined in the USB Specification.
Table 23-2.

336

Handled Standard Requests

Request

Definition

GET_DESCRIPTOR

Returns the current device configuration value.

SET_ADDRESS

Sets the device address for all future device access.

SET_CONFIGURATION

Sets the device configuration.

GET_CONFIGURATION

Returns the current device configuration value.

GET_STATUS

Returns status for the specified recipient.

SET_FEATURE

Set or Enable a specific feature.

CLEAR_FEATURE

Clear or Disable a specific feature.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The device also handles some class requests defined in the CDC class.
Table 23-3.

Handled Class Requests

Request

Definition

SET_LINE_CODING

Configures DTE rate, stop bits, parity and number of
character bits.

GET_LINE_CODING

Requests current DTE rate, stop bits, parity and number of
character bits.

SET_CONTROL_LINE_STATE

RS-232 signal used to tell the DCE device the DTE device
is now present.

Unhandled requests are STALLed.
23.4.3.2

Communication Endpoints
There are two communication endpoints and endpoint 0 is used for the enumeration process.
Endpoint 1 is a 64-byte Bulk OUT endpoint and endpoint 2 is a 64-byte Bulk IN endpoint. SAMBA Boot commands are sent by the host through endpoint 1. If required, the message is split by
the host into several data payloads by the host driver.
If the command requires a response, the host can send IN transactions to pick up the response.

23.4.4

In Application Programming (IAP) Feature
The IAP feature is a function located in ROM that can be called by any software application.
When called, this function sends the desired FLASH command to the EEFC and waits for the
Flash to be ready (looping while the FRDY bit is not set in the MC_FSR register).
Since this function is executed from ROM, this allows Flash programming (such as sector write)
to be done by code running in Flash.
The IAP function entry point is retrieved by reading the NMI vector in ROM (0x00180008).
This function takes one argument in parameter: the command to be sent to the EEFC.
This function returns the value of the MC_FSR register.
IAP software code example:
(unsigned int) (*IAP_Function)(unsigned long);
void main (void){
unsigned long FlashSectorNum = 200; //
unsigned long flash_cmd = 0;
unsigned long flash_status = 0;
unsigned long EFCIndex = 0; // 0:EEFC0, 1: EEFC1
/* Initialize the function pointer (retrieve function address from NMI
vector) */
IAP_Function = ((unsigned long) (*)(unsigned long)) 0x00180008;
/* Send your data to the sector here */

337
6430F–ATARM–21-Feb-12

/* build the command to send to EEFC */
flash_cmd =

(0x5A << 24) | (FlashSectorNum << 8) | AT91C_MC_FCMD_EWP;

/* Call the IAP function with appropriate command */
flash_status = IAP_Function (EFCIndex, flash_cmd);
}

23.5

Hardware and Software Constraints
• SAM-BA Boot uses the first 2048 bytes of the SRAM for variables and stacks. The remaining
available size can be used for user's code.
• USB requirements:
– 12.000 MHz Quartz or 12.000 MHz external clock on XIN. 12 MHz must be
±500 ppm and 1.8V Square Wave Signal.
Table 23-4.

338

Pins Driven during Boot Program Execution

Peripheral

PIO Line

UART

URXD

PA11

UART

UTXD

PA12

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
24. Bus Matrix (MATRIX)
24.1

Description
Bus Matrix implements a multi-layer AHB, based on AHB-Lite protocol, that enables parallel
access paths between multiple AHB masters and slaves in a system, which increases the overall bandwidth. Bus Matrix interconnects 5 AHB Masters to 10 AHB Slaves. The normal latency to
connect a master to a slave is one cycle except for the default master of the accessed slave
which is connected directly (zero cycle latency).
The Bus Matrix user interface is compliant with ARM® Advance Peripheral Bus and provides a
Chip Configuration User Interface with Registers that allow the Bus Matrix to support application
specific features.

24.2

Memory Mapping
Bus Matrix provides one decoder for every AHB Master Interface. The decoder offers each AHB
Master several memory mappings. In fact, depending on the product, each memory area may be
assigned to several slaves. Booting at the same address while using different AHB slaves (i.e.
internal ROM or internal Flash ) becomes possible.
The Bus Matrix user interface provides Master Remap Control Register (MATRIX_MRCR) that
allows to perform remap action for every master independently.

24.3

Special Bus Granting Techniques
The Bus Matrix provides some speculative bus granting techniques in order to anticipate access
requests from some masters. This mechanism allows to reduce latency at first accesses of a
burst or single transfer. The bus granting mechanism allows to set a default master for every
slave.
At the end of the current access, if no other request is pending, the slave remains connected to
its associated default master. A slave can be associated with three kinds of default masters: no
default master, last access master and fixed default master.

24.3.1

No Default Master
At the end of the current access, if no other request is pending, the slave is disconnected from
all masters. No Default Master suits low power mode.

24.3.2

Last Access Master
At the end of the current access, if no other request is pending, the slave remains connected to
the last master that performed an access request.

24.3.3

Fixed Default Master
At the end of the current access, if no other request is pending, the slave connects to its fixed
default master. Unlike last access master, the fixed master doesn’t change unless the user modifies it by a software action (field FIXED_DEFMSTR of the related MATRIX_SCFG).
To change from one kind of default master to another, the Bus Matrix user interface provides the
Slave Configuration Registers, one for each slave, that allow to set a default master for each
slave. The Slave Configuration Register contains two fields:
DEFMSTR_TYPE and FIXED_DEFMSTR. The 2-bit DEFMSTR_TYPE field allows to choose
the default master type (no default, last access master, fixed default master) whereas the 4-bit

339
6430F–ATARM–21-Feb-12

FIXED_DEFMSTR field allows to choose a fixed default master provided that DEFMSTR_TYPE
is set to fixed default master. Please refer to the Bus Matrix user interface description.

24.4

Arbitration
The Bus Matrix provides an arbitration mechanism that allows to reduce latency when conflict
cases occur, basically when two or more masters try to access the same slave at the same time.
One arbiter per AHB slave is provided, allowing to arbitrate each slave differently.
The Bus Matrix provides to the user the possibility to choose between 2 arbitration types, and
this for each slave:
1. Round-Robin Arbitration (the default)
2. Fixed Priority Arbitration
This choice is given through the field ARBT of the Slave Configuration Registers
(MATRIX_SCFG).
Each algorithm may be complemented by selecting a default master configuration for each
slave.
When a re-arbitration has to be done, it is realized only under some specific conditions detailed
in the following paragraph.

24.4.1

Arbitration Rules
Each arbiter has the ability to arbitrate between two or more different master’s requests. In order
to avoid burst breaking and also to provide the maximum throughput for slave interfaces, arbitration may only take place during the following cycles:
1. Idle Cycles: when a slave is not connected to any master or is connected to a master which is
not currently accessing it.
2. Single Cycles: when a slave is currently doing a single access.
3. End of Burst Cycles: when the current cycle is the last cycle of a burst transfer. For defined
length burst, predicted end of burst matches the size of the transfer but is managed differently
for undefined length burst (See Section 24.4.1.1 “Undefined Length Burst Arbitration” on page
340“).
4. Slot Cycle Limit: when the slot cycle counter has reached the limit value indicating that the
current master access is too long and must be broken (See Section 24.4.1.2 “Slot Cycle Limit
Arbitration” on page 341).

24.4.1.1

Undefined Length Burst Arbitration
In order to avoid too long slave handling during undefined length bursts (INCR), the Bus Matrix
provides specific logic in order to re-arbitrate before the end of the INCR transfer.
A predicted end of burst is used as for defined length burst transfer, which is selected between
the following:
1. Infinite: no predicted end of burst is generated and therefore INCR burst transfer will never be
broken.
2. Four beat bursts: predicted end of burst is generated at the end of each four beat boundary
inside INCR transfer.

340

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
3. Eight beat bursts: predicted end of burst is generated at the end of each eight beat boundary
inside INCR transfer.
4. Sixteen beat bursts: predicted end of burst is generated at the end of each sixteen beat
boundary inside INCR transfer.
This selection can be done through the field ULBT of the Master Configuration Registers
(MATRIX_MCFG).
24.4.1.2

24.4.2

Slot Cycle Limit Arbitration
The Bus Matrix contains specific logic to break too long accesses such as very long bursts on a
very slow slave (e.g. an external low speed memory). At the beginning of the burst access, a
counter is loaded with the value previously written in the SLOT_CYCLE field of the related Slave
Configuration Register (MATRIX_SCFG) and decreased at each clock cycle. When the counter
reaches zero, the arbiter has the ability to re-arbitrate at the end of the current byte, half word or
word transfer.
Round-Robin Arbitration
This algorithm allows the Bus Matrix arbiters to dispatch the requests from different masters to
the same slave in a round-robin manner. If two or more master’s requests arise at the same
time, the master with the lowest number is first serviced then the others are serviced in a roundrobin manner.
There are three round-robin algorithm implemented:
• Round-Robin arbitration without default master
• Round-Robin arbitration with last access master
• Round-Robin arbitration with fixed default master

24.4.2.1

Round-Robin arbitration without default master
This is the main algorithm used by Bus Matrix arbiters. It allows the Bus Matrix to dispatch
requests from different masters to the same slave in a pure round-robin manner. At the end of
the current access, if no other request is pending, the slave is disconnected from all masters.
This configuration incurs one latency cycle for the first access of a burst. Arbitration without
default master can be used for masters that perform significant bursts.

24.4.2.2

Round-Robin arbitration with last access master
This is a biased round-robin algorithm used by Bus Matrix arbiters. It allows the Bus Matrix to
remove the one latency cycle for the last master that accessed the slave. In fact, at the end of
the current transfer, if no other master request is pending, the slave remains connected to the
last master that performs the access. Other non privileged masters will still get one latency cycle
if they want to access the same slave. This technique can be used for masters that mainly perform single accesses.

24.4.2.3

Round-Robin arbitration with fixed default master
This is another biased round-robin algorithm, it allows the Bus Matrix arbiters to remove the one
latency cycle for the fixed default master per slave. At the end of the current access, the slave
remains connected to its fixed default master. Every request attempted by this fixed default master will not cause any latency whereas other non privileged masters will still get one latency
cycle. This technique can be used for masters that mainly perform single accesses.

341
6430F–ATARM–21-Feb-12

24.4.3

Fixed Priority Arbitration
This algorithm allows the Bus Matrix arbiters to dispatch the requests from different masters to
the same slave by using the fixed priority defined by the user. If two or more master’s requests
are active at the same time, the master with the highest priority number is serviced first. If two or
more master’s requests with the same priority are active at the same time, the master with the
highest number is serviced first.
For each slave, the priority of each master may be defined through the Priority Registers for
Slaves (MATRIX_PRAS and MATRIX_PRBS).

24.5

Write Protect Registers
To prevent any single software error that may corrupt MATRIX behavior, the entire MATRIX
address space from address offset 0x000 to 0x1FC can be write-protected by setting the WPEN
bit in the MATRIX Write Protect Mode Register (MATRIX_WPMR).
If a write access to anywhere in the MATRIX address space from address offset 0x000 to 0x1FC
is detected, then the WPVS flag in the MATRIX Write Protect Status Register (MATRIX_WPSR)
is set and the field WPVSRC indicates in which register the write access has been attempted.
The WPVS flag is reset by writing the MATRIX Write Protect Mode Register (MATRIX_WPMR)
with the appropriate access key WPKEY.
The protected registers are:
“Bus Matrix Master Configuration Registers”
“Bus Matrix Slave Configuration Registers”
“Bus Matrix Priority Registers For Slaves”
“Bus Matrix Master Remap Control Register”
“Bus Matrix Master Remap Control Register”

342

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
24.6

Bus Matrix (MATRIX) User Interface

Table 24-1.

Offset

Name

Access

Reset

0x0000

Master Configuration Register 0

MATRIX_MCFG0

Read-write

0x00000000

0x0004

Master Configuration Register 1

MATRIX_MCFG1

Read-write

0x00000000

0x0008

Master Configuration Register 2

MATRIX_MCFG2

Read-write

0x00000000

0x000C

Master Configuration Register 3

MATRIX_MCFG3

Read-write

0x00000000

0x0010

Master Configuration Register 4

MATRIX_MCFG4

Read-write

0x00000000

–

0x0014 - 0x003C

Reserved

0x0040

Slave Configuration Register 0

MATRIX_SCFG0

Read-write

0x00010010

0x0044

Slave Configuration Register 1

MATRIX_SCFG1

Read-write

0x00050010

0x0048

Slave Configuration Register 2

MATRIX_SCFG2

Read-write

0x00000010

0x004C

Slave Configuration Register 3

MATRIX_SCFG3

Read-write

0x00000010

0x0050

Slave Configuration Register 4

MATRIX_SCFG4

Read-write

0x00000010

0x0054

Slave Configuration Register 5

MATRIX_SCFG5

Read-write

0x00000010

0x0058

Slave Configuration Register 6

MATRIX_SCFG6

Read-write

0x00000010

0x005C

Slave Configuration Register 7

MATRIX_SCFG7

Read-write

0x00000010

0x0060

Slave Configuration Register 8

MATRIX_SCFG8

Read-write

0x00000010

0x0064

Slave Configuration Register 9

MATRIX_SCFG9

Read-write

0x00000010

–

MATRIX_PRAS0

Read-write

0x00000000

–

MATRIX_PRAS1

Read-write

0x00000000

–

MATRIX_PRAS2

Read-write

0x00000000

–

MATRIX_PRAS3

Read-write

0x00000000

–

MATRIX_PRAS4

Read-write

0x00000000

–

MATRIX_PRAS5

Read-write

0x00000000

–

MATRIX_PRAS6

Read-write

0x00000000

–

MATRIX_PRAS7

Read-write

0x00000000

–

MATRIX_PRAS8

Read-write

0x00000000

0x0068 - 0x007C

Reserved

0x0080

Priority Register A for Slave 0

0x0084

Reserved

0x0088

Priority Register A for Slave 1

0x008C

Reserved

0x0090

Priority Register A for Slave 2

0x0094

Reserved

0x0098

Priority Register A for Slave 3

0x009C

Reserved

0x00A0

Priority Register A for Slave 4

0x00A4

Reserved

0x00A8

Priority Register A for Slave 5

0x00AC

Reserved

0x00B0

Priority Register A for Slave 6

0x00B4

Reserved

0x00B8

Priority Register A for Slave 7

0x00BC

Reserved

0x00C0

Priority Register A for Slave 8

343
6430F–ATARM–21-Feb-12

SAM3U Series
Table 24-1.

Offset

Name

Access

Reset

0x00C4

Reserved

–

0x00C8

Priority Register A for Slave 9

MATRIX_PRAS9

Read-write

0x00000000

–

MATRIX_MRCR

Read-write

0x00000000

–

0x00CC- 0x00FC
0x0100
0x0104 - 0x010C

Reserved
Master Remap Control Register
Reserved

0x1E4

Write Protect Mode Register

MATRIX_WPMR

Read-write

0x0

0x1E8

Write Protect Status Register

MATRIX_WPSR

Read-only

0x0

–

0x0110 - 0x01FC

Reserved

344
6430F–ATARM–21-Feb-12

SAM3U Series
24.6.1
Name:

Bus Matrix Master Configuration Registers
MATRIX_MCFG0..MATRIX_MCFG4

Address:

0x400E0200

Access:

Read-write

–

ULBT

This register can only be written if the WPEN bit is cleared in the “Write Protect Mode Register”.
• ULBT: Undefined Length Burst Type
0: Infinite Length Burst
No predicted end of burst is generated and therefore INCR bursts coming from this master cannot be broken.
1: Single Access
The undefined length burst is treated as a succession of single access allowing rearbitration at each beat of the INCR
burst.
2: Four Beat Burst
The undefined length burst is split into 4 beats burst allowing rearbitration at each 4 beats burst end.
3: Eight Beat Burst
The undefined length burst is split into 8 beats burst allowing rearbitration at each 8 beats burst end.
4: Sixteen Beat Burst
The undefined length burst is split into 16 beats burst allowing rearbitration at each 16 beats burst end.

345
6430F–ATARM–21-Feb-12

SAM3U Series
24.6.2
Name:

Bus Matrix Slave Configuration Registers
MATRIX_SCFG0..MATRIX_SCFG9

Address:

0x400E0240

Access:

Read-write

–

FIXED_DEFMSTR

24
ARBT

DEFMSTR_TYPE

SLOT_CYCLE

This register can only be written if the WPEN bit is cleared in the “Write Protect Mode Register”.
• SLOT_CYCLE: Maximum Number of Allowed Cycles for a Burst
When the SLOT_CYCLE limit is reach for a burst it may be broken by another master trying to access this slave.
This limit has been placed to avoid locking very slow slave by when very long burst are used.
This limit should not be very small though. Unreasonable small value will break every burst and Bus Matrix will spend its
time to arbitrate without performing any data transfer. 16 cycles is a reasonable value for SLOT_CYCLE.
• DEFMSTR_TYPE: Default Master Type
0: No Default Master
At the end of current slave access, if no other master request is pending, the slave is disconnected from all masters.
This results in having a one cycle latency for the first acccess of a burst transfer or for a single access.
1: Last Default Master
At the end of current slave access, if no other master request is pending, the slave stay connected with the last master having accessed it.
This results in not having the one cycle latency when the last master re-tries access on the slave again.
2: Fixed Default Master
At the end of the current slave access, if no other master request is pending, the slave connects to the fixed master which
number has been written in the FIXED_DEFMSTR field.
This results in not having the one cycle latency when the fixed master re-tries access on the slave again.
• FIXED_DEFMSTR: Fixed Default Master
This is the number of the Default Master for this slave. Only used if DEFMSTR_TYPE is 2. Specifying the number of a master which is not connected to the selected slave is equivalent to setting DEFMSTR_TYPE to 0.

346
6430F–ATARM–21-Feb-12

SAM3U Series
• ARBT: Arbitration Type
0: Round-Robin Arbitration
1: Fixed Priority Arbitration
2: Reserved
3: Reserved

347
6430F–ATARM–21-Feb-12

SAM3U Series
24.6.3
Name:

Bus Matrix Priority Registers For Slaves
MATRIX_PRAS0..MATRIX_PRAS9

Addresses: 0x400E0280 [0], 0x400E0288 [1], 0x400E0290 [2], 0x400E0298 [3], 0x400E02A0 [4], 0x400E02A8 [5],
0x400E02B0 [6], 0x400E02B8 [7], 0x400E02C0 [8], 0x400E02C8 [9]
Access:

Read-write

–

M3PR
5

4
M1PR

16
M4PR

8
M2PR

0
M0PR

This register can only be written if the WPEN bit is cleared in the “Write Protect Mode Register”.
• MxPR: Master x Priority
Fixed prority of Master x for accessing to the selected slave.The higher the number, the higher the priority.

348
6430F–ATARM–21-Feb-12

SAM3U Series
24.6.4
Name:

Bus Matrix Master Remap Control Register
MATRIX_MRCR

Address:

0x400E0300

Access:

Read-write

Reset:

0x0000_0000

–

RCB4

RCB3

RCB2

RCB1

RCB0

This register can only be written if the WPEN bit is cleared in the “Write Protect Mode Register”.
• RCBx: Remap Command Bit for AHB Master x
0: Disable remaped address decoding for the selected Master
1: Enable remaped address decoding for the selected Master

349
6430F–ATARM–21-Feb-12

24.6.5
Name:

Write Protect Mode Register
MATRIX_WPMR

Address:

0x400E03E4

Access:

Read-write

WPKEY
23

20
WPKEY

12
WPKEY

–

WPEN

For more details on MATRIX_WPMR, refer to Section 24.5 “Write Protect Registers” on page 342.
• WPEN: Write Protect ENable
0 = Disables the Write Protect if WPKEY corresponds to 0x4D4154 (“MAT” in ASCII).
1 = Enables the Write Protect if WPKEY corresponds to 0x4D4154 (“MAT” in ASCII).
Protects the entire MATRIX address space from address offset 0x000 to 0x1FC.
• WPKEY: Write Protect KEY (Write-only)
Should be written at value 0x4D4154 (“MAT”in ASCII). Writing any other value in this field aborts the write operation of the
WPEN bit. Always reads as 0.

350

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
24.6.6
Name:

Write Protect Status Register
MATRIX_WPSR

Address:

0x400E03E8

Access:

Read-only

–

WPVSRC
15

12
WPVSRC

–

WPVS

For more details on MATRIX_WPSR, refer to Section 24.5 “Write Protect Registers” on page 342.
• WPVS: Write Protect Violation Status
0: No Write Protect Violation has occurred since the last write of MATRIX_WPMR.
1: At least one Write Protect Violation has occurred since the last write of MATRIX_WPMR.
• WPVSRC: Write Protect Violation Source
Should be written at value 0x4D4154 (“MAT”in ASCII). Writing any other value in this field aborts the write operation of the
WPEN bit. Always reads as 0.
The protected registers are:
• “Bus Matrix Master Configuration Registers”
• “Bus Matrix Slave Configuration Registers”
• “Bus Matrix Priority Registers For Slaves”
• “Bus Matrix Master Remap Control Register”
• “Bus Matrix Master Remap Control Register”

351
6430F–ATARM–21-Feb-12

352

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25. Static Memory Controller (SMC)
25.1

Description
The External Bus Interface is designed to ensure the successful data transfer between several
external devices and the Cortex-M3 based device. The External Bus Interface of the SAM3U
consists of a Static Memory Controller (SMC).
This SMC is capable of handling several types of external memory and peripheral devices, such
as SRAM, PSRAM, PROM, EPROM, EEPROM, LCD Module, NOR Flash and NAND Flash.
The SMC generates the signals that control the access to external memory devices or peripheral
devices. It has 4 Chip Selects and a 24-bit address bus. The 16-bit data bus can be configured
to interface with 8- or 16-bit external devices. Separate read and write control signals allow for
direct memory and peripheral interfacing. Read and write signal waveforms are fully
parametrizable.
The SMC can manage wait requests from external devices to extend the current access. The
SMC is provided with an automatic slow clock mode. In slow clock mode, it switches from userprogrammed waveforms to slow-rate specific waveforms on read and write signals.
The SMC embeds a NAND Flash Controller (NFC). The NFC can handle automatic transfers,
sending the commands and address cycles to the NAND Flash and transferring the contents of
the page (for read and write) to the NFC SRAM. It minimizes the CPU overhead.
The SMC includes programmable hardware error correcting code with one bit error correction
capability and supports two bits error detection. In order to improve overall system performance
the DATA phase of the transfer can be DMA assisted.
The External Data Bus can be scrambled/unscrambled by means of user keys.

25.2

Embedded Characteristics
• 16-Mbyte Address Space per Chip Select
• 8- or 16-bit Data Bus
• Word, Halfword, Byte Transfers
• Byte Write or Byte Select Lines
• Programmable Setup, Pulse and Hold Time for Read Signals per Chip Select
• Programmable Setup, Pulse and Hold Time for Write Signals per Chip Select
• Programmable Data Float Time per Chip Select
• External Data Bus Scrambling/Unscrambling Function
• External Wait Request
• Automatic Switch to Slow Clock Mode
• NAND Flash Controller Supporting NAND Flash with Multiplexed Data/Address Buses
• Supports SLC NAND Flash Technology
• Hardware Configurable Number of Chip Selects from 1 to 4
• Programmable Timing on a per Chip Select Basis
• AHB Slave Interface
• Atmel APB Configuration Interface
• Programmable Flash Data Width 8 Bits or 16 Bits

353
6430F–ATARM–21-Feb-12

• Supports Hardware Error Correcting Code (ECC), 1-bit Error Correction, 2-bit Error Detection
• Detection and Correction by Software
• Supports NAND Flash and SmartMedia™ Devices with 8- or 16-bit Data Path
• Supports NAND Flash/SmartMedia with Page Sizes of 528, 1056, 2112 and 4224 Bytes,
Specified by Software
• Supports 1-bit Correction for a Page of 512, 1024, 2048 and 4096 Bytes with 8- or 16-bit
Data Path
• Supports 1-bit Correction per 512 Bytes of Data for a Page Size of 512, 2048 and 4096 Bytes
with 8-bit Data Path
• Supports 1-bit Correction per 256 Bytes of Data for a Page Size of 512, 2048 and 4096 Bytes
with 8-bit Data Path

25.3

Block Diagram

SMC
AHB
Interface

NAND Flash
Controller (NFC)

AHB
arbiter

Scrambler

Figure 25-1. Block Diagram

D[15:0]
A[0]/NBS0
A[20:1]
A21/NANDALE
A22/NANDCLE
A23
NCS[3:0]

SMC
Interface

ECC

SRAM
AHB
Interface

354

User Interface
NFC (4 K bytes)
Internal SRAM

Control & Status
Registers

NRD
NWR0/NWE
NWR1/NBS1
NANDOE
NANDWE
NANDRDY
NWAIT

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.4

I/O Lines Description

Table 25-1.

I/O Line Description

Name

Description

Type

Active Level

NCS[3:0]

Static Memory Controller Chip Select Lines

Output

Low

NRD

Read Signal

Output

Low

NWR0/NWE

Write 0/Write Enable Signal

Output

Low

A0/NBS0

Address Bit 0/Byte 0 Select Signal

Output

Low

NWR1/NBS1

Write 1/Byte 1 Select Signal

Output

Low

Address Bit 1

Output

Low

A[23:2]

Address Bus

Output

D[15:0]

Data Bus

NWAIT

External Wait Signal

Input

NANDRDY

NAND Flash Ready/Busy

Input

NANDWE

NAND Flash Write Enable

Output

Low

NANDOE

NAND Flash Output Enable

Output

Low

NANDALE

NAND Flash Address Latch Enable

Output

NANDCLE

NAND Flash Command Latch Enable

Output

25.5

I/O
Low

Multiplexed Signals

Table 25-2.

Static Memory Controller (SMC) Multiplexed Signals

Multiplexed Signals

Related Function

NWR0

NWE

Byte-write or byte-select access, see Figure 25-4 "Memory Connection for an 8-bit Data
Bus" and Figure 25-5 "Memory Connection for a 16-bit Data Bus"

NBS0

8-bit or 16-bit data bus, see Section 25.9.1 ”Data Bus Width”

A22

NANDCLE

NAND Flash Command Latch Enable

A21

NANDALE

NAND Flash Address Latch Enable

NWR1
A1

NBS1

Byte-write or byte-select access, see Figure 25-4 and Figure 25-5

–

8-/16-bit data bus, see Section 25.9.1 ”Data Bus Width”
Byte-write or byte-select access, see Figure 25-4 and Figure 25-5

355
6430F–ATARM–21-Feb-12

25.6

Application Example

25.6.1

Implementation Examples
For Hardware implementation examples, refer to ATSAM3U-EK schematics which show examples of connection to an LCD module, PSRAM and NAND Flash.

25.6.2

Hardware Interface

Figure 25-2. SMC Connections to Static Memory Devices
D0-D15

A0/NBS0
NWR0/NWE
NWR1/NBS1
A1

D0 - D7

128K x 8
SRAM

D8-D15

D0 - D7
CS

NRD
NWR0/NWE

OE
WE

D0-D7
CS

A0 - A16

NCS0
NCS1
NCS2
NCS3

128K x 8
SRAM

A2 - A18

A0 - A16
NRD
NWR1/NBS1

A2 - A18

OE
WE

A2 - A23

Static Memory
Controller

25.7
25.7.1

Product Dependencies
I/O Lines
The pins used for interfacing the Static Memory Controller are multiplexed with the PIO lines.
The programmer must first program the PIO controller to assign the Static Memory Controller
pins to their peripheral function. If I/O Lines of the SMC are not used by the application, they can
be used for other purposes by the PIO Controller.

25.7.2

356

Power Management
The SMC is clocked through the Power Management Controller (PMC), thus the programmer
must first configure the PMC to enable the SMC clock.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.7.3

Interrupt
The SMC has an interrupt line connected to the Nested Vector Interrupt Controller (NVIC). Handling the SMC interrupt requires programming the NVIC before configuring the SMC.
Table 25-3.

25.8

Peripheral IDs

Instance

SMC

External Memory Mapping
Table 25-4.

External Memory Mapping

Address

Use

Access

0x60000000-0x60FFFFFF

Chip Select 0 (16 MB)

Read-write

0x61000000-0x61FFFFFF

Chip Select 1

Read-write

0x62000000-0x62FFFFFF

Chip Select 2

Read-write

0x63000000-0x63FFFFFF

Chip Select 3

Read-write

0x04000000-0x07FFFFFF

Undefined Area

0x68000000-0x6FFFFFFF

NFC Command Registers(1)

Note:

Read-write

1. See Section 25.16.2 ”NFC Control Registers”, i.e., CMD_ADDR description.

The SMC provides up to 24 address lines, A[23:0]. This allows each chip select line to address
up to 16 Mbytes of memory.
If the physical memory device connected on one chip select is smaller than 16 Mbytes, it wraps
around and appears to be repeated within this space. The SMC correctly handles any valid
access to the memory device within the page (see Figure ).
A[23:0] is only significant for 8-bit memory, A[23:1] is used for 16-bit memory.
Figure 25-3. Memory Connections for External Devices
NCS[0] - NCS[3]
NRD
SMC

NWE

NCS3

A[23:0]

NCS2

D[15:0]

NCS1
NCS0

Memory Enable
Memory Enable

Memory Enable

Memory Enable
Output Enable
Write Enable
A[25:0]

8 or 16

D[15:0] or D[7:0]

357
6430F–ATARM–21-Feb-12

25.9

Connection to External Devices

25.9.1

Data Bus Width
A data bus width of 8 or 16 bits can be selected for each chip select. This option is controlled by
the field DBW in SMC_MODE (Mode Register) for the corresponding chip select.
Figure 25-4 shows how to connect a 512K x 8-bit memory on NCS2. Figure 25-5 shows how to
connect a 512K x 16-bit memory on NCS2.

25.9.2

Byte Write or Byte Select Access
Each chip select with a 16-bit data bus can operate with one of two different types of write
access: byte write or byte select access. This is controlled by the BAT field of the SMC_MODE
register for the corresponding chip select.

Figure 25-4.

Memory Connection for an 8-bit Data Bus

SMC

D[7:0]

A[18:2]

NWE

Write Enable

NRD

Output Enable

NCS[2]

Figure 25-5.

Memory Connection for a 16-bit Data Bus
D[15:0]

D[15:0]

A[19:2]

A[18:1]

A1
SMC

A[0]

NBS0

Low Byte Enable

NBS1

High Byte Enable

NWE

Write Enable

NRD

Output Enable

NCS[2]

25.9.2.1

Memory Enable

Byte Write Access
Byte write access supports one byte write signal per byte of the data bus and a single read
signal.
Note that the SMC does not allow boot in Byte Write Access mode.
• For 16-bit devices: the SMC provides NWR0 and NWR1 write signals for respectively, byte0
(lower byte) and byte1 (upper byte) of a 16-bit bus. One single read signal (NRD) is provided.
Byte Write Access is used to connect 2 x 8-bit devices as a 16-bit memory.

358

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.9.2.2

Byte Select Access
In this mode, read/write operations can be enabled/disabled at byte level. One byte-select line
per byte of the data bus is provided. One NRD and one NWE signal control read and write.
• For 16-bit devices: the SMC provides NBS0 and NBS1 selection signals for respectively
byte0 (lower byte) and byte1 (upper byte) of a 16-bit bus.
Byte Select Access is used to connect one 16-bit device.

Figure 25-6. Connection of 2 x 8-bit Devices on a 16-bit Bus: Byte Write Option
D[7:0]

D[7:0]

D[15:8]
A[24:2]

SMC

A[23:1]
A[0]

A1
NWR0

Write Enable

NWR1
NRD

Read Enable
Memory Enable

NCS[3]

D[15:8]
A[23:1]
A[0]
Write Enable
Read Enable
Memory Enable

25.9.2.3

Signal Multiplexing
Depending on the byte access type (BAT), only the write signals or the byte select signals are
used. To save IOs at the external bus interface, control signals at the SMC interface are multiplexed. Table 25-5 shows signal multiplexing depending on the data bus width and the byte
access type.
For 16-bit devices, bit A0 of address is unused. When Byte Select Option is selected, NWR1 is
unused. When Byte Write option is selected, NBS0 is unused.
Table 25-5.

SMC Multiplexed Signal Translation

Signal Name
Device Type
Byte Access Type (BAT)

16-bit Bus

8-bit Bus

1x16-bit

2 x 8-bit

Byte Select

Byte Write

NBS0_A0

NBS0

NWE_NWR0

NWE

NWR0

NBS1_NWR1

NBS1

NWR1

1 x 8-bit

A0
NWE

359
6430F–ATARM–21-Feb-12

25.10 Standard Read and Write Protocols
In the following sections, the byte access type is not considered. Byte select lines (NBS0 to
NBS1) always have the same timing as the A address bus. NWE represents either the NWE signal in byte select access type or one of the byte write lines (NWR0 to NWR1) in byte write
access type. NWR0 to NWR3 have the same timings and protocol as NWE. In the same way,
NCS represents one of the NCS[0..3] chip select lines.
25.10.1

Read Waveforms
The read cycle is shown on Figure 25-7.
The read cycle starts with the address setting on the memory address bus, i.e.:
{A[23:2], A1, A0} for 8-bit devices
{A[23:2], A1} for 16-bit devices

Figure 25-7. Standard Read Cycle
MCK

A[23:2]

NBS0,NBS1,
A0, A1

NRD

NCS

D[15:0]
NRD_SETUP

NCS_RD_SETUP

NRD_PULSE

NCS_RD_PULSE

NRD_HOLD

NCS_RD_HOLD

NRD_CYCLE

25.10.1.1

NRD Waveform
The NRD signal is characterized by a setup timing, a pulse width and a hold timing.
1. NRD_SETUP: the NRD setup time is defined as the setup of address before the NRD
falling edge.
2. NRD_PULSE: the NRD pulse length is the time between NRD falling edge and NRD
rising edge.
3. NRD_HOLD: the NRD hold time is defined as the hold time of address after the NRD
rising edge.

360

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.10.1.2

NCS Waveform
Similarly, the NCS signal can be divided into a setup time, pulse length and hold time:
1. NCS_RD_SETUP: the NCS setup time is defined as the setup time of address before
the NCS falling edge.
2. NCS_RD_PULSE: the NCS pulse length is the time between NCS falling edge and
NCS rising edge.
3. NCS_RD_HOLD: the NCS hold time is defined as the hold time of address after the
NCS rising edge.

25.10.1.3

Read Cycle
The NRD_CYCLE time is defined as the total duration of the read cycle, i.e., from the time where
address is set on the address bus to the point where address may change. The total read cycle
time is equal to:
NRD_CYCLE = NRD_SETUP + NRD_PULSE + NRD_HOLD
= NCS_RD_SETUP + NCS_RD_PULSE + NCS_RD_HOLD
All NRD and NCS timings are defined separately for each chip select as an integer number of
Master Clock cycles. To ensure that the NRD and NCS timings are coherent, the user must
define the total read cycle instead of the hold timing. NRD_CYCLE implicitly defines the NRD
hold time and NCS hold time as:
NRD_HOLD = NRD_CYCLE - NRD SETUP - NRD PULSE
NCS_RD_HOLD = NRD_CYCLE - NCS_RD_SETUP - NCS_RD_PULSE

25.10.2

Read Mode
As NCS and NRD waveforms are defined independently of one other, the SMC needs to know
when the read data is available on the data bus. The SMC does not compare NCS and NRD timings to know which signal rises first. The READ_MODE parameter in the SMC_MODE register
of the corresponding chip select indicates which signal of NRD and NCS controls the read
operation.

25.10.2.1

Read is Controlled by NRD (READ_MODE = 1):
Figure 25-8 shows the waveforms of a read operation of a typical asynchronous RAM. The read
data is available tPACC after the falling edge of NRD, and turns to ‘Z’ after the rising edge of NRD.
In this case, the READ_MODE must be set to 1 (read is controlled by NRD), to indicate that data
is available with the rising edge of NRD. The SMC samples the read data internally on the rising
edge of Master Clock that generates the rising edge of NRD, whatever the programmed waveform of NCS may be.

361
6430F–ATARM–21-Feb-12

Figure 25-8. READ_MODE = 1: Data is sampled by SMC before the rising edge of NRD
MCK

A[23:2]

NBS0,NBS1,
A0, A1

NRD

NCS
tPACC
D[15:0]

Data Sampling

25.10.2.2

Read is Controlled by NCS (READ_MODE = 0)
Figure 25-9 shows the typical read cycle. The read data is valid tPACC after the falling edge of the
NCS signal and remains valid until the rising edge of NCS. Data must be sampled when NCS is
raised. In that case, the READ_MODE must be set to 0 (read is controlled by NCS): the SMC
internally samples the data on the rising edge of Master Clock that generates the rising edge of
NCS, whatever the programmed waveform of NRD may be.

Figure 25-9. READ_MODE = 0: Data is sampled by SMC before the rising edge of NCS
MCK

A[23:2]

NBS0,NBS1,
A0, A1

NRD

NCS
tPACC
D[15:0]

Data Sampling

362

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.10.3

25.10.3.1

Write Waveforms
The write protocol is similar to the read protocol. It is depicted in Figure 25-10. The write cycle
starts with the address setting on the memory address bus.
NWE Waveforms
The NWE signal is characterized by a setup timing, a pulse width and a hold timing.
1. NWE_SETUP: the NWE setup time is defined as the setup of address and data before
the NWE falling edge.
2. NWE_PULSE: The NWE pulse length is the time between NWE falling edge and NWE
rising edge.
3. NWE_HOLD: The NWE hold time is defined as the hold time of address and data after
the NWE rising edge.
The NWE waveforms apply to all byte-write lines in Byte Write access mode: NWR0 to NWR3.

25.10.3.2

NCS Waveforms
The NCS signal waveforms in write operation are not the same as those applied in read operations, but are separately defined:
1. NCS_WR_SETUP: the NCS setup time is defined as the setup time of address before
the NCS falling edge.
2. NCS_WR_PULSE: the NCS pulse length is the time between NCS falling edge and
NCS rising edge.
3. NCS_WR_HOLD: the NCS hold time is defined as the hold time of address after the
NCS rising edge.

Figure 25-10. Write Cycle
MCK

A[23:2]

NBS0, NBS1,
A0, A1

NWE

NCS

NWE_SETUP

NCS_WR_SETUP

NWE_PULSE

NCS_WR_PULSE

NWE_HOLD

NCS_WR_HOLD

NWE_CYCLE

363
6430F–ATARM–21-Feb-12

25.10.3.3

Write Cycle
The write cycle time is defined as the total duration of the write cycle, that is, from the time where
address is set on the address bus to the point where address may change. The total write cycle
time is equal to:
NWE_CYCLE = NWE_SETUP + NWE_PULSE + NWE_HOLD
= NCS_WR_SETUP + NCS_WR_PULSE + NCS_WR_HOLD
All NWE and NCS (write) timings are defined separately for each chip select as an integer number of Master Clock cycles. To ensure that the NWE and NCS timings are coherent, the user
must define the total write cycle instead of the hold timing. This implicitly defines the NWE hold
time and NCS (write) hold times as:
NWE_HOLD = NWE_CYCLE - NWE_SETUP - NWE_PULSE
NCS_WR_HOLD = NWE_CYCLE - NCS_WR_SETUP - NCS_WR_PULSE

25.10.4

Write Mode
The WRITE_MODE parameter in the SMC_MODE register of the corresponding chip select indicates which signal controls the write operation.

25.10.4.1

Write is Controlled by NWE (WRITE_MODE = 1)
Figure 25-11 shows the waveforms of a write operation with WRITE_MODE set to 1. The data is
put on the bus during the pulse and hold steps of the NWE signal. The internal data buffers are
turned out after the NWE_SETUP time, and until the end of the write cycle, regardless of the
programmed waveform on NCS.

Figure 25-11. WRITE_MODE = 1. The write operation is controlled by NWE
MCK

A[23:2]

NBS0, NBS1,
A0, A1
NWE,
NWR0, NWR1

NCS
D[15:0]

25.10.4.2

364

Write is Controlled by NCS (WRITE_MODE = 0)
Figure 25-12 shows the waveforms of a write operation with WRITE_MODE set to 0. The data is
put on the bus during the pulse and hold steps of the NCS signal. The internal data buffers are
turned out after the NCS_WR_SETUP time, and until the end of the write cycle, regardless of
the programmed waveform on NWE.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 25-12. WRITE_MODE = 0. The write operation is controlled by NCS
MCK

A[23:2]

NBS0, NBS1,
A0, A1
NWE,
NWR0, NWR1

NCS

D[15:0]

25.10.5

Coding Timing Parameters
All timing parameters are defined for one chip select and are grouped together in one
SMC_REGISTER according to their type.
• The SMC_SETUP register groups the definition of all setup parameters: NRD_SETUP,
NCS_RD_SETUP, NWE_SETUP, NCS_WR_SETUP
• The SMC_PULSE register groups the definition of all pulse parameters: NRD_PULSE,
NCS_RD_PULSE, NWE_PULSE, NCS_WR_PULSE
• The SMC_CYCLE register groups the definition of all cycle parameters: NRD_CYCLE,
NWE_CYCLE
Table 25-6 shows how the timing parameters are coded and their permitted range.

Table 25-6.

Coding and Range of Timing Parameters
Permitted Range

Coded Value

Number of Bits

Effective Value

Coded Value

Effective Value

setup [5:0]

128 x setup[5] + setup[4:0]

0 ≤setup ≤31

128..(128+31)

pulse [6:0]

256 x pulse[6] + pulse[5:0]

0 ≤pulse ≤63

256..(256+63)

0 ≤cycle ≤127

256.. (256+127)
512.. (512+127)
768..(768+127)

cycle [8:0]

256 x cycle[8:7] + cycle[6:0]

365
6430F–ATARM–21-Feb-12

25.10.6

Reset Values of Timing Parameters
Table 25-7 gives the default value of timing parameters at reset.
Table 25-7.

25.10.7

Reset Values of Timing Parameters

Reset Value

Description

SMC_SETUP

0x01010101

All setup timings are set to 1

SMC_PULSE

0x01010101

All pulse timings are set to 1

SMC_CYCLE

0x00030003

The read and write operation last 3 Master Clock cycles
and provide one hold cycle

WRITE_MODE

Write is controlled with NWE

READ_MODE

Read is controlled with NRD

Usage Restriction
The SMC does not check the validity of the user-programmed parameters. If the sum of SETUP
and PULSE parameters is larger than the corresponding CYCLE parameter, this leads to unpredictable behavior of the SMC.

25.10.7.1

For Read Operations
Null but positive setup and hold of address and NRD and/or NCS can not be guaranteed at the
memory interface because of the propagation delay of theses signals through external logic and
pads. If positive setup and hold values must be verified, then it is strictly recommended to program non-null values so as to cover possible skews between address, NCS and NRD signals.

25.10.7.2

For Write Operations
If a null hold value is programmed on NWE, the SMC can guarantee a positive hold of address,
byte select lines, and NCS signal after the rising edge of NWE. This is true for WRITE_MODE =
1 only. See “Early Read Wait State” on page 368.
For read and write operations: a null value for pulse parameters is forbidden and may lead to
unpredictable behavior.
In read and write cycles, the setup and hold time parameters are defined in reference to the
address bus. For external devices that require setup and hold time between NCS and NRD signals (read), or between NCS and NWE signals (write), these setup and hold times must be
converted into setup and hold times in reference to the address bus.

25.11 Scrambling/Unscrambling Function
The external data bus D[15:0] can be scrambled in order to prevent intellectual property data
located in off-chip memories from being easily recovered by analyzing data at the package pin
level of either microcontroller or memory device.
The scrambling and unscrambling are performed on-the-fly without additional wait states.
The scrambling method depends on two user-configurable key registers, SMC_KEY1 and
SMC_KEY2. These key registers are only accessible in write mode.
The key must be securely stored in a reliable non-volatile memory in order to recover data from
the off-chip memory. Any data scrambled with a given key cannot be recovered if the key is lost.
The scrambling/unscrambling function can be enabled or disabled by programming the
SMC_OCMS register.

366

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
One bit is dedicated to enable/disable NAND Flash scrambling and one bit is dedicated
enable/disable scrambling the off chip SRAM. When at least one external SRAM is scrambled,
the SMSC field must be set in the SMC_OCMS register.
When multiple chip selects (external SRAM) are handled, it is possible to configure the scrambling function per chip select using the OCMS field in the SMC_TIMINGS registers.
To scramble the NAND Flash contents, the SRSE field must be set in the SMC_OCMS register.
When NAND Flash memory content is scrambled, the on-chip SRAM page buffer associated for
the transfer is also scrambled.

25.12 Automatic Wait States
Under certain circumstances, the SMC automatically inserts idle cycles between accesses to
avoid bus contention or operation conflict.
25.12.1

Chip Select Wait States
The SMC always inserts an idle cycle between 2 transfers on separate chip selects. This idle
cycle ensures that there is no bus contention between the de-activation of one device and the
activation of the next one.
During chip select wait state, all control lines are turned inactive: NBS0 to NBS1, NWR0 to
NWR1, NCS[0..3], and NRD lines. They are all set to 1.
Figure 25-13 illustrates a chip select wait state between access on Chip Select 0 and Chip
Select 2.

Figure 25-13. Chip Select Wait State between a Read Access on NCS0 and a Write Access on NCS2
MCK

A[23:2]

NBS0, NBS1,
A0,A1
NRD
NWE

NCS0

NCS2
NRD_CYCLE

NWE_CYCLE

D[15:0]

Read to Write Chip Select
Wait State
Wait State

367
6430F–ATARM–21-Feb-12

25.12.2

Early Read Wait State
In some cases, the SMC inserts a wait state cycle between a write access and a read access to
allow time for the write cycle to end before the subsequent read cycle begins. This wait state is
not generated in addition to a chip select wait state. The early read cycle thus only occurs
between a write and read access to the same memory device (same chip select).
An early read wait state is automatically inserted if at least one of the following conditions is
valid:
• if the write controlling signal has no hold time and the read controlling signal has no setup
time (Figure 25-14).
• in NCS write controlled mode (WRITE_MODE = 0), if there is no hold timing on the NCS
signal and the NCS_RD_SETUP parameter is set to 0, regardless of the read mode (Figure
25-15). The write operation must end with a NCS rising edge. Without an Early Read Wait
State, the write operation could not complete properly.
• in NWE controlled mode (WRITE_MODE = 1) and if there is no hold timing (NWE_HOLD =
0), the feedback of the write control signal is used to control address, data, chip select and
byte select lines. If the external write control signal is not inactivated as expected due to load
capacitances, an Early Read Wait State is inserted and address, data and control signals are
maintained one more cycle. See Figure 25-16.

Figure 25-14. Early Read Wait State: Write with No Hold Followed by Read with No Setup
MCK

A[23:2]

NBS0, NBS1,
A0, A1

NWE

NRD
no hold
no setup
D[15:0]

write cycle

368

Early Read
wait state

read cycle

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 25-15. Early Read Wait State: NCS Controlled Write with No Hold Followed by a Read with No NCS Setup
MCK

A[23:2]

NBS0, NBS1,
A0,A1

NCS

NRD
no hold

no setup

D[15:0]

write cycle
(WRITE_MODE = 0)

Early Read
wait state

read cycle
(READ_MODE = 0 or READ_MODE = 1)

Figure 25-16. Early Read Wait State: NWE-controlled Write with No Hold Followed by a Read with one Set-up Cycle
MCK

A[23:2]

NBS0, NBS1,
A0, A1
internal write controlling signal
external write controlling signal
(NWE)
no hold

read setup = 1

NRD

D[15:0]

write cycle
(WRITE_MODE = 1)

Early Read
wait state

read cycle
(READ_MODE = 0 or READ_MODE = 1)

369
6430F–ATARM–21-Feb-12

25.12.3

Reload User Configuration Wait State
The user may change any of the configuration parameters by writing the SMC user interface.
When detecting that a new user configuration has been written in the user interface, the SMC
inserts a wait state before starting the next access. The so called “Reload User Configuration
Wait State” is used by the SMC to load the new set of parameters to apply to next accesses.
The Reload Configuration Wait State is not applied in addition to the Chip Select Wait State. If
accesses before and after re-programming the user interface are made to different devices
(Chip Selects), then one single Chip Select Wait State is applied.
On the other hand, if accesses before and after writing the user interface are made to the same
device, a Reload Configuration Wait State is inserted, even if the change does not concern the
current Chip Select.

25.12.3.1

User Procedure
To insert a Reload Configuration Wait State, the SMC detects a write access to any
SMC_MODE register of the user interface. If only the timing registers are modified
(SMC_SETUP, SMC_PULSE, SMC_CYCLE registers) in the user interface, the user must validate the modification by writing the SMC_MODE register, even if no change was made on the
mode parameters.

25.12.3.2

Slow Clock Mode Transition
A Reload Configuration Wait State is also inserted when the Slow Clock Mode is entered or
exited, after the end of the current transfer (see “Slow Clock Mode” on page 381).

25.12.4

Read to Write Wait State
Due to an internal mechanism, a wait cycle is always inserted between consecutive read and
write SMC accesses.
This wait cycle is referred to as a read to write wait state in this document.
This wait cycle is applied in addition to chip select and reload user configuration wait states
when they are to be inserted. See Figure 25-13 on page 367.

25.13 Data Float Wait States
Some memory devices are slow to release the external bus. For such devices, it is necessary to
add wait states (data float wait states) after a read access:
• before starting a read access to a different external memory,
• before starting a write access to the same device or to a different external one.
The Data Float Output Time (t DF ) for each external memory device is programmed in the
TDF_CYCLES field of the SMC_MODE register for the corresponding chip select. The value of
TDF_CYCLES indicates the number of data float wait cycles (between 0 and 15) before the
external device releases the bus, and represents the time allowed for the data output to go to
high impedance after the memory is disabled.
Data float wait states do not delay internal memory accesses. Hence, a single access to an
external memory with long t DF will not slow down the execution of a program from internal
memory.
The data float wait states management depends on the READ_MODE and the TDF_MODE
fields of the SMC_MODE register for the corresponding chip select.

370

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.13.1

READ_MODE
Setting READ_MODE to 1 indicates to the SMC that the NRD signal is responsible for turning off
the tri-state buffers of the external memory device. The Data Float Period then begins after the
rising edge of the NRD signal and lasts TDF_CYCLES MCK cycles.
When the read operation is controlled by the NCS signal (READ_MODE = 0), the TDF field gives
the number of MCK cycles during which the data bus remains busy after the rising edge of NCS.
Figure 25-17 illustrates the Data Float Period in NRD-controlled mode (READ_MODE =1),
assuming a data float period of 2 cycles (TDF_CYCLES = 2). Figure 25-18 shows the read operation when controlled by NCS (READ_MODE = 0) and the TDF_CYCLES parameter equals 3.

Figure 25-17. TDF Period in NRD Controlled Read Access (TDF = 2)
MCK

A[23:2]

NBS0, NBS1,
A0, A1

NRD

NCS
tpacc
D[15:0]
TDF = 2 clock cycles

NRD controlled read operation

371
6430F–ATARM–21-Feb-12

Figure 25-18. TDF Period in NCS Controlled Read Operation (TDF = 3)
MCK

A[23:2]

NBS0, NBS1,
A0,A1

NRD

NCS
tpacc
D[15:0]

TDF = 3 clock cycles
NCS controlled read operation

25.13.2

TDF Optimization Enabled (TDF_MODE = 1)
When the TDF_MODE of the SMC_MODE register is set to 1 (TDF optimization is enabled), the
SMC takes advantage of the setup period of the next access to optimize the number of wait
states cycle to insert.
Figure 25-19 shows a read access controlled by NRD, followed by a write access controlled by
NWE, on Chip Select 0. Chip Select 0 has been programmed with:
NRD_HOLD = 4; READ_MODE = 1 (NRD controlled)
NWE_SETUP = 3; WRITE_MODE = 1 (NWE controlled)
TDF_CYCLES = 6; TDF_MODE = 1 (optimization enabled).

372

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 25-19. TDF Optimization: No TDF wait states are inserted if the TDF period is over when the next access begins
MCK

A[23:2]

NRD
NRD_HOLD= 4
NWE

NWE_SETUP= 3
NCS0

TDF_CYCLES = 6

D[15:0]

read access on NCS0 (NRD controlled)

25.13.3

Read to Write
Wait State

write access on NCS0 (NWE controlled)

TDF Optimization Disabled (TDF_MODE = 0)
When optimization is disabled, tdf wait states are inserted at the end of the read transfer, so that
the data float period ends when the second access begins. If the hold period of the read1 controlling signal overlaps the data float period, no additional tdf wait states will be inserted.
Figure 25-20, Figure 25-21 and Figure 25-22 illustrate the cases:
• read access followed by a read access on another chip select,
• read access followed by a write access on another chip select,
• read access followed by a write access on the same chip select,
with no TDF optimization.

373
6430F–ATARM–21-Feb-12

Figure 25-20. TDF Optimization Disabled (TDF Mode = 0). TDF wait states between 2 read accesses on different chip
selects
MCK

A[23:2]

NBS0, NBS1,
A0, A1
read1 controlling signal
(NRD)

read1 hold = 1

read2 controlling signal
(NRD)

read2 setup = 1

TDF_CYCLES = 6

D[15:0]

5 TDF WAIT STATES
read 2 cycle
TDF_MODE = 0
(optimization disabled)

read1 cycle
TDF_CYCLES = 6
Chip Select Wait State

Figure 25-21. TDF Mode = 0: TDF wait states between a read and a write access on different chip selects
MCK

A[23:2]

NBS0, NBS1,
A0, A1
read1 controlling signal
(NRD)

read1 hold = 1

write2 controlling signal
(NWE)

write2 setup = 1

TDF_CYCLES = 4

D[15:0]

2 TDF WAIT STATES

read1 cycle
TDF_CYCLES = 4
Read to Write Chip Select
Wait State Wait State

374

write2 cycle
TDF_MODE = 0
(optimization disabled)

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 25-22. TDF Mode = 0: TDF wait states between read and write accesses on the same chip select
MCK

A[23:2]
NBS0, NBS1,
A0, A1
read1 controlling signal
(NRD)

write2 setup = 1

read1 hold = 1

write2 controlling signal
(NWE)

TDF_CYCLES = 5

D[15:0]

4 TDF WAIT STATES
read1 cycle
TDF_CYCLES = 5

Read to Write
Wait State

write2 cycle
TDF_MODE = 0
(optimization disabled)

25.14 External Wait
Any access can be extended by an external device using the NWAIT input signal of the SMC.
The EXNW_MODE field of the SMC_MODE register on the corresponding chip select must be
set to either to “10” (frozen mode) or “11” (ready mode). When the EXNW_MODE is set to “00”
(disabled), the NWAIT signal is simply ignored on the corresponding chip select. The NWAIT
signal delays the read or write operation in regards to the read or write controlling signal,
depending on the read and write modes of the corresponding chip select.
25.14.1

Restriction
When one of the EXNW_MODE is enabled, it is mandatory to program at least one hold cycle
for the read/write controlling signal. For that reason, the NWAIT signal cannot be used in Slow
Clock Mode (“Slow Clock Mode” on page 381).
The NWAIT signal is assumed to be a response of the external device to the read/write request
of the SMC. Then NWAIT is examined by the SMC only in the pulse state of the read or write
controlling signal. The assertion of the NWAIT signal outside the expected period has no impact
on SMC behavior.

25.14.2

Frozen Mode
When the external device asserts the NWAIT signal (active low), and after internal synchronization of this signal, the SMC state is frozen, i.e., SMC internal counters are frozen, and all control
signals remain unchanged. When the resynchronized NWAIT signal is deasserted, the SMC
completes the access, resuming the access from the point where it was stopped. See Figure 2523. This mode must be selected when the external device uses the NWAIT signal to delay the
access and to freeze the SMC.

375
6430F–ATARM–21-Feb-12

The assertion of the NWAIT signal outside the expected period is ignored as illustrated in Figure
25-24.
Figure 25-23. Write Access with NWAIT Assertion in Frozen Mode (EXNW_MODE = 10)
MCK

A[23:2]

NBS0, NBS1,
A0,A1
FROZEN STATE
4

NWE
6

NCS

D[15:0]

NWAIT

internally synchronized
NWAIT signal

Write cycle
EXNW_MODE = 10 (Frozen)
WRITE_MODE = 1 (NWE_controlled)
NWE_PULSE = 5
NCS_WR_PULSE = 7

376

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 25-24. Read Access with NWAIT Assertion in Frozen Mode (EXNW_MODE = 10)
MCK

A[23:2]

NBS0, NBS1,
A0,A1
FROZEN STATE
NCS

1
NRD

0
2

0
5

NWAIT

internally synchronized
NWAIT signal

Read cycle
EXNW_MODE = 10 (Frozen)
READ_MODE = 0 (NCS_controlled)
NRD_PULSE = 2, NRD_HOLD = 6
NCS_RD_PULSE =5, NCS_RD_HOLD =3

Assertion is ignored

377
6430F–ATARM–21-Feb-12

25.14.3

Ready Mode
In Ready mode (EXNW_MODE = 11), the SMC behaves differently. Normally, the SMC begins
the access by down counting the setup and pulse counters of the read/write controlling signal. In
the last cycle of the pulse phase, the resynchronized NWAIT signal is examined.
If asserted, the SMC suspends the access as shown in Figure 25-25 and Figure 25-26. After
deassertion, the access is completed: the hold step of the access is performed.
This mode must be selected when the external device uses deassertion of the NWAIT signal to
indicate its ability to complete the read or write operation.
If the NWAIT signal is deasserted before the end of the pulse, or asserted after the end of the
pulse of the controlling read/write signal, it has no impact on the access length as shown in Figure 25-26.

Figure 25-25. NWAIT Assertion in Write Access: Ready Mode (EXNW_MODE = 11)
MCK

A[23:2]

NBS0, NBS1,
A0,A1
Wait STATE
4

NWE
6

NCS

D[15:0]

NWAIT

internally synchronized
NWAIT signal

Write cycle
EXNW_MODE = 11 (Ready mode)
WRITE_MODE = 1 (NWE_controlled)
NWE_PULSE = 5
NCS_WR_PULSE = 7

378

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 25-26. NWAIT Assertion in Read Access: Ready Mode (EXNW_MODE = 11)
MCK

A[23:2]

NBS0, NBS1,
A0,A1
Wait STATE
6

NCS

NRD

NWAIT

internally synchronized
NWAIT signal

Read cycle
EXNW_MODE = 11(Ready mode)
READ_MODE = 0 (NCS_controlled)
Assertion is ignored

Assertion is ignored
NRD_PULSE = 7
NCS_RD_PULSE =7

379
6430F–ATARM–21-Feb-12

25.14.4

NWAIT Latency and Read/Write Timings
There may be a latency between the assertion of the read/write controlling signal and the assertion of the NWAIT signal by the device. The programmed pulse length of the read/write
controlling signal must be at least equal to this latency plus the 2 cycles of resynchronization + 1
cycle. Otherwise, the SMC may enter the hold state of the access without detecting the NWAIT
signal assertion. This is true in frozen mode as well as in ready mode. This is illustrated on Figure 25-27.
When EXNW_MODE is enabled (ready or frozen), the user must program a pulse length of the
read and write controlling signal of at least:
minimal pulse length = NWAIT latency + 2 resynchronization cycles + 1 cycle

Figure 25-27. NWAIT Latency
MCK
A[23:2]

NBS0, NBS1,
A0,A1
WAIT STATE
4

NRD
minimal pulse length

NWAIT
intenally synchronized
NWAIT signal

NWAIT latency 2 cycle resynchronization

Read cycle
EXNW_MODE = 10 or 11
READ_MODE = 1 (NRD_controlled)
NRD_PULSE = 5

380

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.15 Slow Clock Mode
The SMC is able to automatically apply a set of “slow clock mode” read/write waveforms when
an internal signal driven by the Power Management Controller is asserted because MCK has
been turned to a very slow clock rate (typically 32 kHz clock rate). In this mode, the user-programmed waveforms are ignored and the slow clock mode waveforms are applied. This mode is
provided so as to avoid reprogramming the User Interface with appropriate waveforms at very
slow clock rate. When activated, the slow mode is active on all chip selects.
25.15.1

Slow Clock Mode Waveforms
Figure 25-28 illustrates the read and write operations in slow clock mode. They are valid on all
chip selects. Table 25-8 indicates the value of read and write parameters in slow clock mode.

Figure 25-28. Read/Write Cycles in Slow Clock Mode
MCK

MCK

A[23:2]

NBS0, NBS1,
A0,A1

NWE

NRD

1
NCS

NCS

NRD_CYCLE = 2

NWE_CYCLE = 3
SLOW CLOCK MODE WRITE

Table 25-8.

SLOW CLOCK MODE READ

Read and Write Timing Parameters in Slow Clock Mode

Read Parameters

Duration (cycles)

Write Parameters

Duration (cycles)

NRD_SETUP

NWE_SETUP

NRD_PULSE

NWE_PULSE

NCS_RD_SETUP

NCS_WR_SETUP

NCS_RD_PULSE

NCS_WR_PULSE

NRD_CYCLE

NWE_CYCLE

381
6430F–ATARM–21-Feb-12

25.15.2

Switching from (to) Slow Clock Mode to (from) Normal Mode
When switching from slow clock mode to normal mode, the current slow clock mode transfer is
completed at high clock rate, with the set of slow clock mode parameters. See Figure 25-29. The
external device may not be fast enough to support such timings.
Figure 25-30 illustrates the recommended procedure to properly switch from one mode to the
other.

Figure 25-29. Clock Rate Transition Occurs while the SMC is Performing a Write Operation
Slow Clock Mode
internal signal from PMC
MCK
A[23:2]
NBS0, NBS1,
A0, A1
NWE

NCS

NWE_CYCLE = 3

NWE_CYCLE = 7

SLOW CLOCK MODE WRITE SLOW CLOCK MODE WRITE

This write cycle finishes with the slow clock mode set
of parameters after the clock rate transition

382

NORMAL MODE WRITE

Slow clock mode transition is detected:
Reload Configuration Wait State

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 25-30. Recommended Procedure to Switch from Slow Clock Mode to Normal Mode or from Normal Mode to Slow
Clock Mode
Slow Clock Mode
internal signal from PMC

MCK

A[23:2]

NBS0, NBS1,
A0, A1

NWE
1

NCS

SLOW CLOCK MODE WRITE

IDLE STATE

NORMAL MODE WRITE

Reload Configuration
Wait State

383
6430F–ATARM–21-Feb-12

25.16 NAND Flash Controller Operations
25.16.1

NFC Overview
The NFC can handle automatic transfers, sending the commands and address to the NAND
Flash and transferring the contents of the page (for read and write) to the NFC SRAM. It minimizes the CPU overhead.

25.16.2

NFC Control Registers
NAND Flash Read and NAND Flash Program operations can be performed through the NFC
Command Registers. In order to minimize CPU intervention and latency, commands are posted
in a command buffer. This buffer provides zero wait state latency. The detailed description of the
command encoding scheme is explained below.
The NFC handles automatic transfer between the external NAND Flash and the chip via the
NFC SRAM. It is done via NFC Command Registers.
The NFC Command Registers are very efficient to use. When writing to these registers:
• the address of the register (NFCADDR_CMD) contains the command used,
• the data of the register (NFCDATA_ADDT) contains the address to be sent to the NAND
Flash.
So, in one single access the command is sent and immediately executed by the NFC. Even two
commands can be programmed within a single access (CMD1, CMD2) depending on the
VCMD2 value.
The NFC can send up to 5 Address cycles.
Figure 25-31 below shows a typical NAND Flash Page Read Command of a NAND Flash Memory and correspondence with NFC Address Command Register.

Figure 25-31. NFC/NAND Flash Access Example
00h

Col. Add1

Col. Add2

Column Address

CMD1

Row Add1 Row Add2 Row Add3

30h

Row Address

ADD cycles (0 to 5)

Depends on ACYCLE value

CMD2

If VCMD2 = 1

For more details refer to “NFC Address Command” on page 386.
The NFC Command Registers can be found at address 0x68000000 - 0x6FFFFFFF. (See
Table 25-4, “External Memory Mapping”.)
Reading the NFC command register (to any address) will give the status of the NFC. Especially
useful to know if the NFC is busy, for example.

384

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.16.2.1

Building NFC Address Command Example
The base address is made of address 0x60000000 + NFCCMD bit set = 0x68000000.
Page read operation example:
// Build the Address Command (NFCADDR_CMD)
AddressCommand = (0x60000000

NFCCMD=1 | // NFC Command Enable
NFCWR=0

|// NFC Read Data from NAND Flash

NFCEN=1

CSID=1

// NFC Enable.
|

ACYCLE= 5 |

// Chip Select ID = 1

// Number of address cycle.

VCMD2=1

CMD2=0x30

CMD1=0x0)

// CMD2 is sent after Address Cycles
// CMD2 = 30h
// CMD1 = Read Command = 00h

// Set the Address for Cycle 0
SMC_ADDR = Col. Add1
// Write command with the Address Command built above
*AddressCommand = (Col. Add2

|// ADDR_CYCLE1

Row Add1

| // ADDR_CYCLE2

Row Add2

|// ADDR_CYCLE3

Row Add3

)// ADDR_CYCLE4

385
6430F–ATARM–21-Feb-12

25.16.2.2

NFC Address Command

• Name: NFCADDR_CMD
• Access: Read-write
• Reset: 0x00000000
31
–

30
–

29
–

28
–

27
NFCCMD

26
NFCWR

25
NFCEN

24
CSID

20
ACYCLE

18
VCMD2

CSID
15

CMD2
8

CMD2
7

CMD1
4

CMD1

1
–

0
–

• CMD1: Command Register Value for Cycle 1
If NFCCMD is set, when a read or write access occurs, the NFC sends this command.
• CMD2: Command Register Value for Cycle 2
If NFCCMD and VCMD2 field are set to one, the NFC sends this command after CMD1.
• VCMD2: Valid Cycle 2 Command
When set to true, the CMD2 field is issued after the address cycle.
• ACYCLE: Number of Address required for the current command
When ACYCLE field is different from zero, ACYCLE Address cycles are performed after Command Cycle 1. The maximum
number of cycles is 5.
• CSID: Chip Select Identifier
Chip select used
• NFCEN: NFC Enable
When set to true, the NFC will automatically read or write data after the command.
• NFCWR: NFC Write Enable
0: The NFC reads data from the NAND Flash.
1: The NFC writes data into the NAND Flash.
• NFCCMD: NFC Command Enable
If set to true, CMD indicates that the NFC shall execute the command encoded in the NFCADDR_CMD.

386

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.16.2.3

NFC Data Address

• Name: NFCDATA_ADDT
• Access: Write
• Reset: 0x00000000
31

28
27
ADDR_CYCLE4

20
19
ADDR_CYCLE3

12
11
ADDR_CYCLE2

4
3
ADDR_CYCLE1

• ADDR_CYCLE1: NAND Flash Array Address Cycle 1
When less than 5 address cycles are used, ADDR_CYCLE1 is the first byte written to NAND Flash
When 5 address cycles are used, ADDR_CYCLE1 is the second byte written to NAND Flash
• ADDR_CYCLE2: NAND Flash Array Address Cycle 2
When less than 5 address cycles are used, ADDR_CYCLE2 is the second byte written to NAND Flash
When 5 address cycles are used, ADDR_CYCLE2 is the third byte written to NAND Flash
• ADDR_CYCLE3: NAND Flash Array Address Cycle 3
When less than 5 address cycles are used, ADDR_CYCLE3 is the third byte written to NAND Flash
When 5 address cycles are used, ADDR_CYCLE3 is the fourth byte written to NAND Flash
• ADDR_CYCLE4: NAND Flash Array Address Cycle 4
When less than 5 address cycles are used, ADDR_CYCLE4 is the fourth byte written to NAND Flash
When 5 address cycles are used, ADDR_CYCLE4 is the fifth byte written to NAND Flash

Note:

If 5 address cycles are used, the first address cycle is ADDR_CYCLE0. Refer to SMC_ADDR register.

387
6430F–ATARM–21-Feb-12

25.16.2.4

NFC DATA Status

• Name: NFCDATA_Status
• Access: Read
• Reset: 0x00000000
31
–

30
–

29
–

28
–

27
NFCBUSY

26
NFCWR

25
NFCEN

24
CSID

20
ACYCLE

18
VCMD2

CSID
15

CMD2

CMD2
7

8
CMD1

4
CMD1

1
–

0
–

• CMD1: Command Register Value for Cycle 1
When a Read or Write Access occurs, the Physical Memory Interface drives the IO bus with CMD1 field during the Command Latch cycle 1.
• CMD2: Command Register Value for Cycle 2
When VCMD2 field is set to true, the Physical Memory Interface drives the IO bus with CMD2 field during the Command
Latch cycle 2.
• VCMD2: Valid Cycle 2 Command
When set to true, the CMD2 field is issued after addressing cycle.
• ACYCLE: Number of Address required for the current command
When ACYCLE field is different from zero, ACYCLE Address cycles are performed after Command Cycle 1.
• CSID: Chip Select Identifier
Chip select used
• NFCEN: NFC Enable
When set to true, The NFC is enabled.
• NFCWR: NFC Write Enable
0: The NFC is in read mode.
1: The NFC is in write mode.
• NFCBUSY: NFC Busy Status Flag
If set to true, it indicates that the NFC is busy.

388

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.16.3

NFC Initialization
Prior to any Command and Data Transfer, the SMC User Interface must be configured to meet
the device timing requirements.
• Write enable Configuration
Use NWE_SETUP, NWE_PULSE and NWE_CYCLE to define the write enable waveform
according to the device datasheet.
Use TADL field in the SMC_TIMINGS register to configure the timing between the last address
latch cycle and the first rising edge of WEN for data input.

Figure 25-32. Write Enable Timing Configuration
mck

wen

tWEN_SETUP

tWEN_PULSE

tWEN_HOLD

tWEN_CYCLES

Figure 25-33. Write Enable Timing for NAND Flash Device Data Input Mode.
mck
ale

wen

tADL

• Read Enable Configuration
Use NRD_SETUP, NRD_PULSE and NRD_CYCLE to define the read enable waveform according to the device datasheet.
Use TAR field in the SMC_TIMINGS register to configure the timings between address latch
enable falling edge to read enable falling edge.
Use TCLR field in the SMC_TIMINGS register to configure the timings between the command
latch enable falling edge to the read enable falling edge.

389
6430F–ATARM–21-Feb-12

Figure 25-34. Read Enable Timing Configuration Working with NAND Flash Device
mck

cen

ale

cle

ren

tCLR
tAR

tREN_SETUP

tREN_PULSE

tREH

tREN_CYCLE

• Ready/Busy Signal Timing configuration working with a NAND Flash device
Use TWB field in SMC_TIMINGS register to configure the maximum elapsed time between the
rising edge of wen signal and the falling edge of rbn signal. Use TRR field in the SMC_TIMINGS
register to program the number of clock cycle between the rising edge of the rbn signal and the
falling edge of ren signal.
Figure 25-35. Ready/Busy Timing Configuration
mck

rbn

ren

wen

tWB

390

busy

tRR

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.16.3.1

NAND Flash Controller Timing Engine
When the NFC Command register is written, the NFC issues a NAND Flash Command and
optionally performs a data transfer between the NFC SRAM and the NAND Flash device. The
NAND Flash Controller Timing Engine guarantees valid NAND Flash timings, depending on the
set of parameters decoded from the address bus. These timings are defined in the
SMC_TIMINGS register.
For information on the timing used depending on the command, see Figure 25-36:
Figure 25-36. NAND Flash Controller Timing Engine
Timing Check Engine

NFCEN=1 NFCWR=1 TADL =1
Wait TADL
NFCEN=1 NFCWR=0 TWB != 0
Wait TWB
NFCEN=0 VCMD2=1 TCLR!= 0
Wait TCLR
!NFCEN=1 VCMD2=0 ACYCLE!=0 NFCWR=1 TADL != 0
Wait TADL
!NFCEN=1 VCMD2=0 ACYCLE!=0 NFCWR=0 TAR!= 0
Wait TAR
!NFCEN=1 VCMD2=0 ACYCLE!=0 TCLR!= 0
Wait TCLR

See ”NFC Address Command” register description and ”SMC Timings Register”.

391
6430F–ATARM–21-Feb-12

25.16.4
25.16.4.1

NFC SRAM
NFC SRAM Mapping
If the NFC is used to read and write Data from and to the NAND Flash, the configuration
depends on the page size. See Table 25-9 and Table 25-10 for detailed mapping.
The NFC SRAM size is 4 Kbytes. The NFC can handle NAND Flash with a page size of 4 Kbytes
or of a lower size (such as 2 Kbytes for example). In case of a 2-KByte or lower page size, the
NFC SRAM can be split into several banks. The SMC_BANK field enables to select the bank
used.
Note that a “ping-pong” mode (write or read to a bank while the NFC writes or reads to another
bank) is not accessible with the NFC (using 2 different banks).
If the NFC is not used, the NFC SRAM can be used for a general purpose by the application.
Table 25-9.

NFC SRAM Mapping with NAND Flash Page Size of 2 Kbytes + 64 bytes

Offset

Use

Access

0x00000000-0x000001FF

Bank 0 Main Area Buffer 0

Read-write

0x00000200-0x000003FF

Bank 0 Main Area Buffer 1

Read-write

0x00000400-0x000005FF

Bank 0 Main Area Buffer 2

Read-write

0x00000600-0x000007FF

Bank 0 Main Area Buffer 3

Read-write

0x00000800-0x0000080F

Bank 0 Spare Area 0

Read-write

0x00000810-0x0000081F

Bank 0 Spare Area 1

Read-write

0x00000820-0x0000082F

Bank 0 Spare Area 2

Read-write

0x00000830-0x0000083F

Bank 0 Spare Area 3

Read-write

0x00000840-0x00000A3F

Bank 1 Main Area Buffer 0

Read-write

0x00000A40-0x00000C3F

Bank 1 Main Area Buffer 1

Read-write

0x00000C40-0x00000E3F

Bank 1 Main Area Buffer 2

Read-write

0x00000E40-0x0000103F

Bank 1 Main Area Buffer 3

Read-write

0x00001040-0x0000104F

Bank 1 Spare Area 0

Read-write

0x00001050-0x0000105F

Bank 1 Spare Area 1

Read-write

0x00001060-0x0000106F

Bank 1 Spare Area 2

Read-write

0x00001070-0x0000107F

Bank 1 Spare Area 3

Read-write

0x00001080-0x00001FFF

Reserved

–

Table 25-10. NFC SRAM Mapping with NAND Flash Page Size of 4 Kbytes + 128 bytes

392

Offset

Use

Access

0x00000000-0x000001FF

Bank 0 Main Area Buffer 0

Read-write

0x00000200-0x000003FF

Bank 0 Main Area Buffer 1

Read-write

0x00000400-0x000005FF

Bank 0 Main Area Buffer 2

Read-write

0x00000600-0x000007FF

Bank 0 Main Area Buffer 3

Read-write

0x00000800-0x000009FF

Bank 0 Main Area Buffer 4

Read-write

0x00000A00-0x00000BFF

Bank 0 Main Area Buffer 5

Read-write

0x00000C00-0x00000DFF

Bank 0 Main Area Buffer 6

Read-write

0x00000E00-0x00000FFF

Bank 0 Main Area Buffer 7

Read-write

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 25-10. NFC SRAM Mapping with NAND Flash Page Size of 4 Kbytes + 128 bytes

25.16.4.2

25.16.5

Offset

Use

Access

0x00001000-0x0000100F

Bank 0 Spare Area 0

Read-write

0x00001010-0x0000101F

Bank 0 Spare Area 1

Read-write

0x00001020-0x0000102F

Bank 0 Spare Area 2

Read-write

0x00001030-0x0000103F

Bank 0 Spare Area 3

Read-write

0x00001040-0x0000104F

Bank 0 Spare Area 4

Read-write

0x00001050-0x0000105F

Bank 0 Spare Area 5

Read-write

0x00001060-0x0000106F

Bank 0 Spare Area 6

Read-write

0x00001070-0x0000107F

Bank 0 Spare Area 7

Read-write

0x00001080-0x00001FFF

Reserved

–

NFC SRAM Access Prioritization Algorithm
When the NAND Flash Controller (NFC) is reading from or writing to the NFC SRAM, the internal
memory is no longer accessible. If an NFC SRAM access occurs when the NFC performs a read
or write operation then the access is discarded. The write operation is not performed. The read
operation returns undefined data. If this situation is encountered, the status flag AWB located in
the NFC status Register is raised and indicates that a shared resource access violation has
occurred.
NAND Flash Operations
This section describes the software operations needed to issue commands to the NAND Flash
device and perform data transfers using NFC.

393
6430F–ATARM–21-Feb-12

25.16.5.1

Page Read

Figure 25-37. Page Read Flow Chart

Configure Device,
writing in theUser Interface

Using NFC

Write the NFC
Command registers

Enable XFRDONE
interrupt (SMC_IER)

Wait for Interrupt

Copy the data from NFC
SRAM to application
memory (via DMA for example)

Check Error
Correcting Codes

Note that instead of using the interrupt one can poll the NFCBUSY Flag.
For more information on the NFC Control Register, see Section 25.16.2.2 ”NFC Address
Command”.

394

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.16.5.2

Program Page

Figure 25-38. Program Page Flow Chart

Con gure Device,
writing in the User
interface

Write Data in the NFC
SRAM (CPU or DMA)
Enable XFRDONE

Write the Command
Register through the
AHB interface

Wait for interrupt

Write ECC

wait for Ready/Busy
interrupt

Writing the ECC can not be done using the NFC so it needs to be done “manually”.
Note that instead of using the interrupt one can poll the NFCBUSY Flag.
For more information on the NFC Control Register, see Section 25.16.2.2 ”NFC Address
Command”.

395
6430F–ATARM–21-Feb-12

25.17 SMC Error Correcting Code Functional Description
A page in NAND Flash and SmartMedia memories contains an area for main data and an additional area used for redundancy (ECC). The page is organized in 8-bit or 16-bit words. The page
size corresponds to the number of words in the main area plus the number of words in the extra
area used for redundancy.
Over time, some memory locations may fail to program or erase properly. In order to ensure that
data is stored properly over the life of the NAND Flash device, NAND Flash providers recommend to utilize either 1 ECC per 256 bytes of data, 1 ECC per 512 bytes of data or 1 ECC for all
of the page.
The only configurations required for ECC are the NAND Flash or the SmartMedia page size
(528/2112/4224) and the type of correction wanted (1 ECC for all the page/1 ECC per 256 bytes
of data /1 ECC per 512 bytes of data). Page size is configured setting the PAGESIZE field in the
ECC Mode Register (ECC_MR). Type of correction is configured setting the TYPCORRECT
field in the ECC Mode Register (ECC_MR).
Note:

There is a limitation when using a 16-bit NAND Flash: only 1 ECC for a whole page is possible.
There is no limitation when using an 8-bit NAND Flash.

ECC is automatically computed as soon as a read (00h)/write (80h) command to the NAND
Flash or the SmartMedia is detected. Read and write access must start at a page boundary.
ECC results are available as soon as the counter reaches the end of the main area. Values in
the ECC Parity Registers (ECC_PR0 to ECC_PR15) are then valid and locked until a new start
condition occurs (read/write command followed by address cycles).
25.17.1

Write Access
Once the Flash memory page is written, the computed ECC codes are available in the ECC Parity (ECC_PR0 to ECC_PR15) registers. The ECC code values must be written by the software
application in the extra area used for redundancy. The number of write accesses in the extra
area is a function of the value of the type of correction field. For example, for 1 ECC per 256
bytes of data for a page of 512 bytes, only the values of ECC_PR0 and ECC_PR1 must be written by the software application. Other registers are meaningless.

25.17.2

Read Access
After reading the whole data in the main area, the application must perform read accesses to the
extra area where ECC code has been previously stored. Error detection is automatically performed by the ECC controller.
Note:

It is mandatory to read consecutively the entire main area and the locations where Parity and
NParity values have been previously stored, to let the ECC controller perform error detection.

The application can check the ECC Status Registers (ECC_SR1/ECC_SR2) for any detected
errors. It is up to the application to correct any detected error. ECC computation can detect four
different circumstances:
• No error: XOR between the ECC computation and the ECC code stored at the end of the
NAND Flash or SmartMedia page is equal to 0. No error flags in the ECC Status Registers
(ECC_SR1/ECC_SR2).
• Recoverable error: Only the RECERR flags in the ECC Status registers
(ECC_SR1/ECC_SR2) are set. The corrupted word offset in the read page is defined by the
WORDADDR field in the ECC Parity Registers (ECC_PR0 to ECC_PR15). The corrupted bit
position in the concerned word is defined in the BITADDR field in the ECC Parity Registers
(ECC_PR0 to ECC_PR15).
396

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• ECC error: The ECCERR flag in the ECC Status Registers (ECC_SR1/ECC_SR2) is set. An
error has been detected in the ECC code stored in the Flash memory. The position of the
corrupted bit can be found by the application performing an XOR between the Parity and the
NParity contained in the ECC code stored in the Flash memory.
• Non correctable error: The MULERR flag in the ECC Status Registers
(ECC_SR1/ECC_SR2) is set. Several unrecoverable errors have been detected in the Flash
memory page.
ECC Status Registers, ECC Parity Registers are cleared when a read/write command is
detected or a software reset is performed.
For Single-bit Error Correction and Double-bit Error Detection (SEC-DED) Hsiao code is used.
24-bit ECC is generated in order to perform one bit correction per 256 or 512 bytes for pages of
512/2048/4096 8-bit words. 32-bit ECC is generated in order to perform one bit correction per
512/1024/2048/4096 8- or 16-bit words. They are generated according to the schemes shown in
Figure 25-39 and Figure 25-40.
Figure 25-39. Parity Generation for 512/1024/2048/4096 8-bit Words
1st byte
2nd byte
3rd byte
4 th byte

(page size -3 )th byte
(page size -2 )th byte
(page size -1 )th byte
Page size th byte

Bit7

Bit6

Bit5

Bit7

Bit6

Bit5

Bit3

Bit2

Bit1

Bit0

Bit3

Bit1

Bit0

P8
P8'

P16

Bit4

Bit2

Bit7
Bit7

Bit6
Bit6

Bit5
Bit5

Bit4
Bit4

Bit3
Bit3

Bit2
Bit2

Bit1
Bit1

Bit0
Bit0

P8
P8'

P16'

Bit7
Bit7

Bit6
Bit6

Bit5
Bit5

Bit4
Bit4

Bit3
Bit3

Bit2
Bit2

Bit1
Bit1

Bit0
Bit0

P8
P8'

P16

Bit7
Bit7

Bit6
Bit6

Bit5
Bit5

Bit4
Bit4

Bit3
Bit3

Bit2
Bit2

Bit1
Bit1

Bit0
Bit0

P8
P8'

P16'

P1'

Bit4

P1'
P2'

Page size
Page size
Page size
Page size

= 512
= 1024
= 2048
= 4096

Px = 2048
Px = 4096
Px = 8192
Px = 16384

P32

PX'

P2'
P4'

P1=bit7(+)bit5(+)bit3(+)bit1(+)P1
P2=bit7(+)bit6(+)bit3(+)bit2(+)P2
P4=bit7(+)bit6(+)bit5(+)bit4(+)P4
P1'=bit6(+)bit4(+)bit2(+)bit0(+)P1'
P2'=bit5(+)bit4(+)bit1(+)bit0(+)P2'
P4'=bit7(+)bit6(+)bit5(+)bit4(+)P4'

To calculate P8’ to PX’ and P8 to PX, apply the algorithm that follows.
Page size = 2n
for i =0 to n

397
6430F–ATARM–21-Feb-12

begin
for (j = 0 to page_size_byte)
begin
if(j[i] ==1)
P[2i+3]=bit7(+)bit6(+)bit5(+)bit4(+)bit3(+)
bit2(+)bit1(+)bit0(+)P[2i+3]
else
P[2i+3]’=bit7(+)bit6(+)bit5(+)bit4(+)bit3(+)
bit2(+)bit1(+)bit0(+)P[2i+3]'
end
end

398

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 25-40. Parity Generation for 512/1024/2048/4096 16-bit Words
1st word
2nd word
3rd word
4th word

(Page size -3 )th word
(Page size -2 )th word
(Page size -1 )th word
Page size th word

(+)

To calculate P8’ to PX’ and P8 to PX, apply the algorithm that follows.
Page size = 2n
for i =0 to n
begin
for (j = 0 to page_size_word)
begin
if(j[i] ==1)
P[2i+3]= bit15(+)bit14(+)bit13(+)bit12(+)
bit11(+)bit10(+)bit9(+)bit8(+)
bit7(+)bit6(+)bit5(+)bit4(+)bit3(+)
bit2(+)bit1(+)bit0(+)P[2n+3]
else
P[2i+3]’=bit15(+)bit14(+)bit13(+)bit12(+)
bit11(+)bit10(+)bit9(+)bit8(+)
bit7(+)bit6(+)bit5(+)bit4(+)bit3(+)
bit2(+)bit1(+)bit0(+)P[2i+3]'
end
end

399
6430F–ATARM–21-Feb-12

25.18 Power Management Controller (PMC) User Interface
The SMC is programmed using the fields listed in Table 25-11. For each chip, select a set of 4
registers is used to program the parameters of the external device. In Table 25-11,
“CS_number” denotes the chip select number. 16 Bytes per chip select are required.
Table 25-11. Register Mapping
Offset

Name

Access

Reset

0x000

SMC NFC Configuration Register

SMC_CFG

Read-write

0x0

0x004

SMC NFC Control Register

SMC_CTRL

Write-only

0x0

0x008

SMC NFC Status Register

SMC_SR

Read-only

0x0

0x00C

SMC NFC Interrupt Enable Register

SMC_IER

Write-only

0x0

0x010

SMC NFC Interrupt Disable Register

SMC_IDR

Write-only

0x0

0x014

SMC NFC Interrupt Mask Register

SMC_IMR

Read-only

0x0

0x018

SMC NFC Address Cycle Zero Register

SMC_ADDR

Read-write

0x0

0x01C

SMC Bank Address Register

SMC_BANK

Read-write

0x0

0x020

SMC ECC Control Register

SMC_ECC_CTRL

Write-only

0x0

0x024

SMC ECC Mode Register

SMC_ECC_MD

Read-write

0x0

0x028

SMC ECC Status 1 Register

SMC_ECC_SR1

Read-only

0x0

0x02C

SMC ECC Parity 0 Register

SMC_ECC_PR0

Read-only

0x0

0x030

SMC ECC parity 1 Register

SMC_ECC_PR1

Read-only

0x0

0x034

SMC ECC status 2 Register

SMC_ECC_SR2

Read-only

0x0

0x038

SMC ECC parity 2 Register

SMC_ECC_PR2

Read-only

0x0

0x03C

SMC ECC parity 3 Register

SMC_ECC_PR3

Read-only

0x0

0x040

SMC ECC parity 4 Register

SMC_ECC_PR4

Read-only

0x0

0x044

SMC ECC parity 5 Register

SMC_ECC_PR5

Read-only

0x0

0x048

SMC ECC parity 6 Register

SMC_ECC_PR6

Read-only

0x0

0x04C

SMC ECC parity 7 Register

SMC_ECC_PR7

Read-only

0x0

0x050

SMC ECC parity 8 Register

SMC_ECC_PR8

Read-only

0x0

0x054

SMC ECC parity 9 Register

SMC_ECC_PR9

Read-only

0x0

0x058

SMC ECC parity 10 Register

SMC_ECC_PR10

Read-only

0x0

0x05C

SMC ECC parity 11 Register

SMC_ECC_PR11

Read-only

0x0

0x060

SMC ECC parity 12 Register

SMC_ECC_PR12

Read-only

0x0

0x064

SMC ECC parity 13 Register

SMC_ECC_PR13

Read-only

0x0

0x068

SMC ECC parity 14 Register

SMC_ECC_PR14

Read-only

0x0

0x06C

SMC ECC parity 15 Register

SMC_ECC_PR15

Read-only

0x0

0x14*CS_number+0x070

SMC Setup Register

SMC_SETUP

Read-write

0x01010101

0x14*CS_number+0x074

SMC Pulse Register

SMC_PULSE

Read-write

0x01010101

0x14*CS_number+0x078

SMC Cycle Register

SMC_CYCLE

Read-write

0x00030003

0x14*CS_number+0x7C

SMC Timings Register

SMC_TIMINGS

Read-write

0x00000000

400

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 25-11. Register Mapping (Continued)
Offset

Name

Access

Reset

0x14*CS_number+0x80

SMC Mode Register

SMC_MODE

Read-write

0x10000003

0x110

SMC OCMS Register

SMC_OCMS

Read-write

0x0

0x114

SMC OCMS KEY1 Register

SMC_KEY1

Write-once

0x0

0x118

SMC OCMS KEY2 Register

SMC_KEY2

Write-once

0x0

0x1E4

Write Protection Control Register

SMC_WPCR

Write-only

0x0

0x1E8

Write Protection Status Register

SMC_WPSR

Read-only

0x0

0x1FC

Reserved

–

401
6430F–ATARM–21-Feb-12

25.18.1
Name:

SMC NFC Configuration Register
SMC_CFG

Address:

0x400E0000

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

21
DTOMUL

15
–

14
–

13
RBEDGE

12
EDGECTRL

11
–

10
–

9
RSPARE

8
WSPARE

7
–

6
–

5
–

4
–

3
–

2
–

DTOCYC

PAGESIZE

• PAGESIZE
This field defines the page size of the NAND Flash device.
Value

Name

Description

PS512_16

Main area 512 Bytes + Spare area 16 Bytes = 528 Bytes

PS1024_32

Main area 1024 Bytes + Spare area 32 Bytes = 1056 Bytes

PS2048_64

Main area 2048 Bytes + Spare area 64 Bytes = 2112 Bytes

PS4096_128

Main area 4096 Bytes + Spare area 128 Bytes = 4224 Bytes

• WSPARE: Write Spare Area
0: The NFC skips the spare area in write mode.
1: The NFC writes both main area and spare area in write mode.
• RSPARE: Read Spare Area
0: The NFC skips the spare area in read mode.
1: The NFC reads both main area and spare area in read mode.
• EDGECTRL: Rising/Falling Edge Detection Control
0: Rising edge is detected.
1: Falling edge is detected.
• RBEDGE: Ready/Busy Signal Edge Detection
0: When set to zero, RB_EDGE fields indicate the level of the Ready/Busy lines.
1: When set to one, RB_EDGE fields indicate only transition on Ready/Busy lines.
• DTOCYC: Data Timeout Cycle Number

402

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• DTOMUL: Data Timeout Multiplier
These fields determine the maximum number of Master Clock cycles that the SMC waits until the detection of a rising edge
on Ready/Busy signal.
Data Timeout Multiplier is defined by DTOMUL as shown in the following table:
Value

Name

Description

DTOCYC

X16

DTOCYC x 16

X128

DTOCYC x 128

X256

DTOCYC x 256

X1024

DTOCYC x 1024

X4096

DTOCYC x 4096

X65536

DTOCYC x 65536

X1048576

DTOCYC x 1048576

If the data timeout set by DTOCYC and DTOMUL has been exceeded, the Data Timeout Error flag (DTOE) in the SMC Status Register (SMC_SR) raises.

403
6430F–ATARM–21-Feb-12

25.18.2
Name:

SMC NFC Control Register
SMC_CTRL

Address:

0x400E0004

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
NFCDIS

0
NFCEN

• NFCEN: NAND Flash Controller Enable
0: no effect.
1: Enable the NAND Flash controller.
• NFCDIS: NAND Flash Controller Disable
0: no effect
1: Disable the NAND Flash controller.

404

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.3
Name:

SMC NFC Status Register
SMC_SR

Address:

0x400E0008

Access:

Read-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
RB_EDGE0

23
NFCASE

22
AWB

21
UNDEF

20
DTOE

19
–

18
–

17
CMDDONE

16
XFRDONE

15
–

13
NFCSID

11
NFCWR

10
–

9
–

8
NFCBUSY

7
–

6
–

5
RB_FALL

4
RB_RISE

3
–

2
–

1
–

0
SMCSTS

• SMCSTS: NAND Flash Controller status (this field cannot be reset)
0: NAND Flash Controller is disabled.
1: NAND Flash Controller is enabled.
• RB_RISE: Selected Ready Busy Rising Edge Detected
When set to one, this flag indicates that a rising edge on Ready/Busy Line has been detected. This flag is reset after a status read operation. The Ready/Busy line selected is the decoding of the set NFCCSID, RBNSEL fields.
• RB_FALL: Selected Ready Busy Falling Edge Detected
When set to one, this flag indicates that a falling edge on Ready/Busy Line has been detected. This flag is reset after a status read operation. The Ready/Busy line is selected through the decoding of the set NFCSID, RBNSEL fields.
• NFCBUSY: NFC Busy (this field cannot be reset)
When set to one this flag indicates that the Controller is activated and accesses the memory device.
• NFCWR: NFC Write/Read Operation (this field cannot be reset)
When a command is issued, this field indicates the current Read or Write Operation. This field can be manually updated
with the use of the SMC_CTRL register.
• NFCSID: NFC Chip Select ID (this field cannot be reset)
When a command is issued, this field indicates the value of the targeted chip select. This field can be manually updated
with the use of the SMC_CTRL register.
• XFRDONE: NFC Data Transfer Terminated
When set to one, this flag indicates that the NFC has terminated the Data Transfer. This flag is reset after a status read
operation.
• CMDDONE: Command Done
When set to one, this flag indicates that the NFC has terminated the Command. This flag is reset after a status read
operation.

405
6430F–ATARM–21-Feb-12

• DTOE: Data Timeout Error
When set to one this flag indicates that the Data timeout set be by DTOMUL and DTOCYC has been exceeded. This flag is
reset after a status read operation.
• UNDEF: Undefined Area Error
When set to one this flag indicates that the processor performed an access in an undefined memory area. This flag is reset
after a status read operation.
• AWB: Accessing While Busy
If set to one this flag indicates that an AHB master has performed an access during the busy phase. This flag is reset after
a status read operation.
• NFCASE: NFC Access Size Error
If set to one, this flag indicates that an illegal access has been detected in the NFC Memory Area. Only Word Access is
allowed within the NFC memory area. This flag is reset after a status read operation.
• RB_EDGEx: Ready/Busy Line x Edge Detected
If set to one, this flag indicates that an edge has been detected on the Ready/Busy Line x. Depending on the EDGE CTRL
field located in the SMC_MODE register, only rising or falling edge is detected. This flag is reset after a status read
operation.

406

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.4
Name:

SMC NFC Interrupt Enable Register
SMC_IER

Address:

0x400E000C

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
RB_EDGE0

23
NFCASE

22
AWB

21
UNDEF

20
DTOE

19
–

18
–

17
CMDDONE

16
XFRDONE

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
RB_FALL

4
RB_RISE

3
–

2
–

1
–

0
–

• RB_RISE: Ready Busy Rising Edge Detection Interrupt Enable
0: No effect.
1: Interrupt source is enabled.
• RB_FALL: Ready Busy Falling Edge Detection Interrupt Enable
0: No effect.
1: Interrupt source is enabled.
• XFRDONE: Transfer Done Interrupt Enable
0: No effect.
1: Interrupt source is enabled.
• CMDDONE: Command Done Interrupt Enable
0: No effect.
1: Interrupt source is enabled.
• DTOE: Data Timeout Error Interrupt Enable
0: No effect.
1: Interrupt source is enabled.
• UNDEF: Undefined Area Access Interrupt Enable
0: No effect.
1: Interrupt source is enabled.
• AWB: Accessing While Busy Interrupt Enable
0: No effect.
1: Interrupt source is enabled.

407
6430F–ATARM–21-Feb-12

• NFCASE: NFC Access Size Error Interrupt Enable
0: No effect.
1: Interrupt source is enabled.
• RB_EDGEx: Ready/Busy Line x Interrupt Enable
0: No effect.
1: Interrupt source is enabled.

408

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.5
Name:

SMC NFC Interrupt Disable Register
SMC_IDR

Address:

0x400E0010

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
RB_EDGE0

23
NFCASE

22
AWB

21
UNDEF

20
DTOE

19
–

18
–

17
CMDDONE

16
XFRDONE

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
RB_FALL

4
RB_RISE

3
–

2
–

1
–

0
–

• RB_RISE: Ready Busy Rising Edge Detection Interrupt Disable
0: No effect.
1: Interrupt source is disabled.
• RB_FALL: Ready Busy Falling Edge Detection Interrupt Disable
0: No effect.
1: Interrupt source is disabled.
• XFRDONE: Transfer Done Interrupt Disable
0: No effect.
1: Interrupt source is disabled.
• CMDDONE: Command Done Interrupt Disable
0: No effect.
1: Interrupt source is disabled.
• DTOE: Data Timeout Error Interrupt Disable
0: No effect.
1: Interrupt source is disabled.
• UNDEF: Undefined Area Access Interrupt Disable
0: No effect.
1: Interrupt source is disabled.
• AWB: Accessing While Busy Interrupt Disable
0: No effect.
1: Interrupt source is disabled.

409
6430F–ATARM–21-Feb-12

• NFCASE: NFC Access Size Error Interrupt Disable
0: No effect.
1: Interrupt source is disabled.
• RB_EDGEx: Ready/Busy Line x Interrupt Disable
0: No effect.
1: Interrupt source is disabled.

410

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.6
Name:

SMC NFC Interrupt Mask Register
SMC_IMR

Address:

0x400E0014

Access:

Read-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
RB_EDGE0

23
NFCASE

22
AWB

21
UNDEF

20
DTOE

19
–

18
–

17
CMDDONE

16
XFRDONE

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
RB_FALL

4
RB_RISE

3
–

2
–

1
–

0
–

• RB_RISE: Ready Busy Rising Edge Detection Interrupt Mask
0: Interrupt source is disabled.
1: Interrupt source is enabled.
• RB_FALL: Ready Busy Falling Edge Detection Interrupt Mask
0: Interrupt source is disabled.
1: Interrupt source is enabled.
• XFRDONE: Transfer Done Interrupt Mask
0: Interrupt source is disabled.
1: Interrupt source is enabled.
• CMDDONE: Command Done Interrupt Mask
0: Interrupt source is disabled.
1: Interrupt source is enabled.
• DTOE: Data Timeout Error Interrupt Mask
0: Interrupt source is disabled.
1: Interrupt source is enabled.
• UNDEF: Undefined Area Access Interrupt Mask5
0: Interrupt source is disabled.
1: Interrupt source is enabled.
• AWB: Accessing While Busy Interrupt Mask
0: Interrupt source is disabled.
1: Interrupt source is enabled.

411
6430F–ATARM–21-Feb-12

• NFCASE: NFC Access Size Error Interrupt Mask
0: Interrupt source is disabled.
1: Interrupt source is enabled.
• RB_EDGEx: Ready/Busy Line x Interrupt Mask
0: Interrupt source is disabled.
1: Interrupt source is enabled.

412

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.7
Name:

SMC NFC Address Cycle Zero Register
SMC_ADDR

Address:

0x400E0018

Access:

Read-Write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

4
3
ADDR_CYCLE0

• ADDR_CYCLE0: NAND Flash Array Address cycle 0
When 5 address cycles are used, ADDR_CYCLE0 is the first byte written to the NAND Flash (used by the NFC).

413
6430F–ATARM–21-Feb-12

25.18.8
Name:

SMC NFC Bank Register
SMC_BANK

Address:

0x400E001C

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

1
BANK

• BANK: Bank Identifier
Number of the bank used

414

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.9
Name:

SMC ECC Control Register
SMC_ECC_CTRL

Address:

0x400E0020

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
SWRST

0
RST

• RST: Reset ECC
0: No effect.
1: Reset ECC parity registers.
• SWRST: Software Reset
0: No effect.
1: Reset all registers.

415
6430F–ATARM–21-Feb-12

25.18.10 SMC ECC MODE Register
Name:
SMC_ECC_MD
Address:

0x400E0024

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

3
–

2
–

1
0
ECC_PAGESIZE

TYPCORREC

• ECC_PAGESIZE: ECC Page Size
This field defines the page size of the NAND Flash device.
Value

Name

Description

PS512_16

Main area 512 Bytes + Spare area 16 Bytes = 528 Bytes

PS1024_32

Main area 1024 Bytes + Spare area 32 Bytes = 1056 Bytes

PS2048_64

Main area 2048 Bytes + Spare area 64 Bytes = 2112 Bytes

PS4096_128

Main area 4096 Bytes + Spare area 128 Bytes = 4224 Bytes

• TYPCORREC: Type of Correction
Value

Name

Description

CPAGE

1 bit correction for a page of 512/1024/2048/4096 Bytes
(for 8 or 16-bit NAND Flash)

C256B

1 bit correction for 256 Bytes of data for a page of 512/2048/4096 bytes
(for 8-bit NAND Flash only)

C512B

1 bit correction for 512 Bytes of data for a page of 512/2048/4096 bytes
(for 8-bit NAND Flash only)

416

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.11 SMC ECC Status Register 1
Name:
SMC_ECC_SR1
Address:

0x400E0028

Access:

Read-only

Reset:

0x00000000

31
–
23
–
15
–
7
–

30
ECCERR7
22
ECCERR5
14
MULERR3
6
MULERR1

29
ECCERR7
21
ECCERR5
13
ECCERR3
5
ECCERR1

28
RECERR7
20
RECERR5
12
RECERR3
4
RECERR1

27
–
19
–
11
–
3
–

26
ECCERR6
18
ECCERR4
10
MULERR2
2
ECCERR0

25
ECCERR6
17
ECCERR4
9
ECCERR2
1
ECCERR0

24
RECERR6
16
RECERR4
8
RECERR2
0
RECERR0

• RECERR0: Recoverable Error
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise multiple uncorrected errors
were detected.
• ECCERR0: ECC Error
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
If TYPECORRECT = 0, read both ECC Parity 0 and ECC Parity 1 registers, the error occurred at the location which contains a 1 in the least significant 16 bits; else read ECC Parity 0 register, the error occurred at the location which contains a
1 in the least significant 24 bits.
• MULERR0: Multiple Error
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR1: Recoverable Error in the page between the 256th and the 511th bytes or the 512nd and the 1023rd
bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise multiple uncorrected errors
were detected.
• ECCERR1: ECC Error in the page between the 256th and the 511th bytes or between the 512nd and the 1023rd
bytes
Fixed to 0 if TYPECORREC = 0
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 1 register, the error occurred at the location which contains a 1 in the least significant 24 bits.

417
6430F–ATARM–21-Feb-12

• MULERR1: Multiple Error in the page between the 256th and the 511th bytes or between the 512nd and the
1023rd bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR2: Recoverable Error in the page between the 512nd and the 767th bytes or between the 1024th and the
1535th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise, multiple uncorrected errors
were detected.
• ECCERR2: ECC Error in the page between the 512nd and the 767th bytes or between the 1024th and the 1535th
bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 2 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR2: Multiple Error in the page between the 512nd and the 767th bytes or between the 1024th and the
1535th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR3: Recoverable Error in the page between the 768th and the 1023rd bytes or between the 1536th and the
2047th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise multiple uncorrected errors
were detected.
• ECCERR3: ECC Error in the page between the 768th and the 1023rd bytes or between the 1536th and the 2047th
bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 3 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR3: Multiple Error in the page between the 768th and the 1023rd bytes or between the 1536th and the
2047th bytes
Fixed to 0 if TYPECORREC = 0.

418

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR4: Recoverable Error in the page between the 1024th and the 1279th bytes or between the 2048th and
the 2559th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise multiple uncorrected errors
were detected.
• ECCERR4: ECC Error in the page between the 1024th and the 1279th bytes or between the 2048th and the 2559th
bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 4 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR4: Multiple Error in the page between the 1024th and the 1279th bytes or between the 2048th and the
2559th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR5: Recoverable Error in the page between the 1280th and the 1535th bytes or between the 2560th and
the 3071st bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise multiple uncorrected errors
were detected
• ECCERR5: ECC Error in the page between the 1280th and the 1535th bytes or between the 2560th and the 3071st
bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 5 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR5: Multiple Error in the page between the 1280th and the 1535th bytes or between the 2560th and the
3071st bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.

419
6430F–ATARM–21-Feb-12

• RECERR6: Recoverable Error in the page between the 1536th and the 1791st bytes or between the 3072nd and
the 3583rd bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise multiple uncorrected errors
were detected.
• ECCERR6: ECC Error in the page between the 1536th and the 1791st bytes or between the 3072nd and the
3583rd bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 6 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR6: Multiple Error in the page between the 1536th and the 1791st bytes or between the 3072nd and the
3583rd bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR7: Recoverable Error in the page between the 1792nd and the 2047th bytes or between the 3584th and
the 4095th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise, multiple uncorrected errors
were detected.
• ECCERR7: ECC Error in the page between the 1792nd and the 2047th bytes or between the 3584th and the
4095th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 7 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR7: Multiple Error in the page between the 1792nd and the 2047th bytes or between the 3584th and the
4095th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.

420

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.12 SMC ECC Status Register 2
Name:
SMC_ECC_SR2
Address:

0x400E0034

Access:

Read-only

Reset:

0x00000000

31
–
23
–
15
–
7
–

30
ECCERR15
22
ECCERR13
14
MULERR11
6
MULERR9

29
ECCERR15
21
ECCERR13
13
ECCERR11
5
ECCERR9

28
RECERR15
20
RECERR13
12
RECERR11
4
RECERR9

27
–
19
–
11
–
3
–

26
ECCERR14
18
ECCERR12
10
MULERR10
2
ECCERR8

25
ECCERR14
17
ECCERR12
9
ECCERR10
1
ECCERR8

24
RECERR14
16
RECERR12
8
RECERR10
0
RECERR8

• RECERR8: Recoverable Error in the page between the 2048th and the 2303rd bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise multiple uncorrected errors
were detected
• ECCERR8: ECC Error in the page between the 2048th and the 2303rd bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 8 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR8: Multiple Error in the page between the 2048th and the 2303rd bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR9: Recoverable Error in the page between the 2304th and the 2559th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise multiple uncorrected errors
were detected.
• ECCERR9: ECC Error in the page between the 2304th and the 2559th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 9 register, the error occurred at the location which contains a 1 in the least significant 24 bits.

421
6430F–ATARM–21-Feb-12

• MULERR9: Multiple Error in the page between the 2304th and the 2559th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR10: Recoverable Error in the page between the 2560th and the 2815th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise, multiple uncorrected errors
were detected.
• ECCERR10: ECC Error in the page between the 2560th and the 2815th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 10 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR10: Multiple Error in the page between the 2560th and the 2815th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR11: Recoverable Error in the page between the 2816th and the 3071st bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise, multiple uncorrected errors
were detected
• ECCERR11: ECC Error in the page between the 2816th and the 3071st bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 11 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR11: Multiple Error in the page between the 2816th and the 3071st bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR12: Recoverable Error in the page between the 3072nd and the 3327th bytes
Fixed to 0 if TYPECORREC = 0
0: No Errors Detected
422

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise multiple uncorrected errors
were detected
• ECCERR12: ECC Error in the page between the 3072nd and the 3327th bytes
Fixed to 0 if TYPECORREC = 0
0: No Errors Detected
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 12 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR12: Multiple Error in the page between the 3072nd and the 3327th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR13: Recoverable Error in the page between the 3328th and the 3583rd bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise multiple uncorrected errors
were detected.
• ECCERR13: ECC Error in the page between the 3328th and the 3583rd bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 13 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR13: Multiple Error in the page between the 3328th and the 3583rd bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR14: Recoverable Error in the page between the 3584th and the 3839th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise, multiple uncorrected errors
were detected.
• ECCERR14: ECC Error in the page between the 3584th and the 3839th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 14 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
423
6430F–ATARM–21-Feb-12

• MULERR14: Multiple Error in the page between the 3584th and the 3839th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.
• RECERR15: Recoverable Error in the page between the 3840th and the 4095th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: Errors Detected. If MUL_ERROR is 0, a single correctable error was detected. Otherwise, multiple uncorrected errors
were detected.
• ECCERR15: ECC Error in the page between the 3840th and the 4095th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Errors Detected.
1: A single bit error occurred in the ECC bytes.
Read ECC Parity 15 register, the error occurred at the location which contains a 1 in the least significant 24 bits.
• MULERR15: Multiple Error in the page between the 3840th and the 4095th bytes
Fixed to 0 if TYPECORREC = 0.
0: No Multiple Errors Detected.
1: Multiple Errors Detected.

424

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.13 SMC ECC Parity Register 0 for a Page of 512/1024/2048/4096 Bytes
Name:
SMC_ECC_PR0
Address:

0x400E002C

Access:

Read-only

Reset:

0x00000000

31
–
23
–
15

30
–
22
–
14

29
–
21
–
13

28
–
20
–
12

27
–
19
–
11

26
–
18
–
10

25
–
17
–
9

24
–
16
–
8

WORDADDR
WORDADDR

BITADDR

Once the entire main area of a page is written with data, the register content must be stored at any free location of the
spare area.
• BITADDR: Bit Address
During a page read, this value contains the corrupted bit offset where an error occurred, if a single error was detected. If
multiple errors were detected, this value is meaningless.
• WORDADDR: Word Address
During a page read, this value contains the word address (8-bit or 16-bit word depending on the memory plane organization).where an error occurred, if a single error was detected. If multiple errors were detected, this value is meaningless.

425
6430F–ATARM–21-Feb-12

25.18.14 SMC ECC Parity Register 1 for a Page of 512/1024/2048/4096 Bytes
Name:
SMC_ECC_PR1
Address:

0x400E0030

Access:

Read-only

Reset:

0x00000000

31
–
23
–
15

30
–
22
–
14

29
–
21
–
13

28
–
20
–
12

27
–
19
–
11

26
–
18
–
10

25
–
17
–
9

24
–
16
–
8

NPARITY
NPARITY

• NPARITY: Parity N

426

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.15 SMC ECC Parity Registers for 1 ECC per 512 Bytes for a Page of 512/2048/4096 Bytes, 9-bit Word
Name:
SMC_ECC_PRx [x=0..7] (W9BIT)
Address:

0x400E0038 [2] .. 0x400E006C [15]

Access:

Read-only

Reset:

0x00000000
31
–
23

30
–
22

29
–
21

28
–
20
NPARITY
12

27
–
19

26
–
18

25
–
17

24
–
16

1
BITADDR

NPARITY
5
WORDADDR

WORDADDR

Once the entire main area of a page is written with data, the register content must be stored at any free location of the
spare area.
• BITADDR: Corrupted Bit Address in the Page between (i x 512) and ((i + 1) x 512) - 1) Bytes
During a page read, this value contains the corrupted bit offset where an error occurred, if a single error was detected. If
multiple errors were detected, this value is meaningless.
• WORDADDR: Corrupted Word Address in the Page between (i x 512) and ((i + 1) x 512) - 1) Bytes
During a page read, this value contains the word address (9-bit word) where an error occurred, if a single error was
detected. If multiple errors were detected, this value is meaningless.
• NPARITY: Parity N

427
6430F–ATARM–21-Feb-12

25.18.16 SMC ECC Parity Registers for 1 ECC per 256 Bytes for a Page of 512/2048/4096 Bytes, 8-bit Word
Name:
SMC_ECC_PRx [x=0..15] (W8BIT)
Address:

0x400E0038 [2] .. 0x400E006C [15]

Access:

Read-only

Reset:

0x00000000
31
–
23
0
15

30
–
22

29
–
21

28
–
20

NPARITY
5
WORDADDR

27
–
19
NPARITY
11
0
3

26
–
18

25
–
17

24
–
16

9
WORDADDR
1
BITADDR

Once the entire main area of a page is written with data, the register content must be stored at any free location of the
spare area.
• BITADDR: Corrupted Bit Address in the Page between (i x 256) and ((i + 1) x 512) - 1) Bytes
• During a page read, this value contains the corrupted bit offset where an error occurred, if a single error was
detected. If multiple errors were detected, this value is meaningless.
• WORDADDR: Corrupted Word Address in the Page between (i x 256) and ((i + 1) x 512) - 1) Bytes
During a page read, this value contains the word address (8-bit word) where an error occurred, if a single error was
detected. If multiple errors were detected, this value is meaningless.
• NPARITY: Parity N

428

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.17 SMC Setup Register
Name:
SMC_SETUPx [x=0..3]
Address:

0x400E0070 [0], 0x400E0084 [1], 0x400E0098 [2], 0x400E00AC [3]

Access:

Read-write

Reset:

0x01010101

31
–

30
–

27
26
NCS_RD_SETUP

23
–

22
–

15
–

14
–

11
10
NCS_WR_SETUP

7
–

6
–

NRD_SETUP

2
NWE_SETUP

• NWE_SETUP: NWE Setup Length
The NWE signal setup length is defined as:
NWE setup length = (128 * NWE_SETUP[5] + NWE_SETUP[4:0]) clock cycles.
• NCS_WR_SETUP: NCS Setup Length in Write Access
In write access, the NCS signal setup length is defined as:
NCS setup length = (128 * NCS_WR_SETUP[5] + NCS_WR_SETUP[4:0]) clock cycles.
• NRD_SETUP: NRD Setup Length
The NRD signal setup length is defined as:
NRD setup length = (128 * NRD_SETUP[5] + NRD_SETUP[4:0]) clock cycles.
• NCS_RD_SETUP: NCS Setup Length in Read Access
In Read access, the NCS signal setup length is defined as:
NCS setup length = (128 * NCS_RD_SETUP[5] + NCS_RD_SETUP[4:0]) clock cycles.

429
6430F–ATARM–21-Feb-12

25.18.18 SMC Pulse Register
Name:
SMC_PULSEx [x=0..3]
Address:

0x400E0074 [0], 0x400E0088 [1], 0x400E009C [2], 0x400E00B0 [3]

Access:

Read-write

Reset:

0x01010101

31
–

30
–

27
26
NCS_RD_PULSE

23
–

22
–

15
–

14
–

11
10
NCS_WR_PULSE

7
–

6
–

NRD_PULSE

2
NWE_PULSE

• NWE_PULSE: NWE Pulse Length
The NWE signal pulse length is defined as:
NWE pulse length = (256 * NWE_PULSE[6]+NWE_PULSE[5:0]) clock cycles.
The NWE pulse must be at least one clock cycle.
• NCS_WR_PULSE: NCS Pulse Length in WRITE Access
In Write access, The NCS signal pulse length is defined as:
NCS pulse length = (256 * NCS_WR_PULSE[6] + NCS_WR_PULSE[5:0]) clock cycles.
the NCS pulse must be at least one clock cycle.
• NRD_PULSE: NRD Pulse Length
The NRD signal pulse length is defined as:
NRD pulse length = (256 * NRD_PULSE[6] + NRD_PULSE[5:0]) clock cycles.
The NRD pulse width must be as least 1 clock cycle.
• NCS_RD_PULSE: NCS Pulse Length in READ Access
In READ mode, The NCS signal pulse length is defined as:
NCS pulse length = (256 * NCS_RD_PULSE[6] + NCS_RD_PULSE[5:0]) clock cycles.

430

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.19 SMC Cycle Register
Name:
SMC_CYCLEx [x=0..3]
Address:

0x400E0078 [0], 0x400E008C [1], 0x400E00A0 [2], 0x400E00B4 [3]

Access:

Read-write

Reset:

0x00030003

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
NRD_CYCLE

NRD_CYCLE
15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
NWE_CYCLE

NWE_CYCLE

• NWE_CYCLE: Total Write Cycle Length
The total write cycle length is the total duration in clock cycles of the write cycle. It is equal to the sum of the setup, pulse
and hold steps of the NWE and NCS signals. It is defined as:
Write cycle length = (NWE_CYCLE[8:7] * 256) + NWE_CYCLE[6:0] clock cycles.
• NRD_CYCLE: Total Read Cycle Length
The total read cycle length is the total duration in clock cycles of the read cycle. It is equal to the sum of the setup, pulse
and hold steps of the NRD and NCS signals. It is defined as:
Read cycle length = (NRD_CYCLE[8:7] * 256) + NRD_CYCLE[6:0] clock cycles.

431
6430F–ATARM–21-Feb-12

25.18.20 SMC Timings Register
Name:
SMC_TIMINGSx [x=0..3]
Address:

0x400E007C [0], 0x400E0090 [1], 0x400E00A4 [2], 0x400E00B8 [3]

Access:

Read-write

Reset:

0x00000000

31
NFSEL

29
RBNSEL

23
–

22
–

21
–

20
–

15
–

14
–

13
–

12
OCMS

TWB
18
TRR
10
TAR
2

TADL

TCLR

• TCLR: CLE to REN Low Delay
Command Latch Enable falling edge to Read Enable falling edge timing.
Latch Enable Falling to Read Enable Falling = (TCLR[3] * 64) + TCLR[2:0] clock cycles.
• TADL: ALE to Data Start
Last address latch cycle to the first rising edge of WEN for data input.
Last address latch to first rising edge of WEN = (TADL[3] * 64) + TADL[2:0] clock cycles.
• TAR: ALE to REN Low Delay
Address Latch Enable falling edge to Read Enable falling edge timing.
Address Latch Enable to Read Enable = (TAR[3] * 64) + TAR[2:0] clock cycles.
• OCMS: Off Chip Memory Scrambling Enable
When set to one, the memory scrambling is activated.
• TRR: Ready to REN Low Delay
Ready/Busy signal to Read Enable falling edge timing.
Read to REN = (TRR[3] * 64) + TRR[2:0] clock cycles.
• TWB: WEN High to REN to Busy
Write Enable rising edge to Ready/Busy falling edge timing.
Write Enable to Read/Busy = (TWB[3] * 64) + TWB[2:0] clock cycles.
• RBNSEL: Ready/Busy Line Selection
This field indicates the selected Ready/Busy Line from the RBN bundle.
• NFSEL: NAND Flash Selection
If this bit is set to one, the chip select is assigned to NAND Flash write enable and read enable lines drive the Error Correcting Code module.

432

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.21 SMC Mode Register
Name:
SMC_MODEx [x=0..3]
Address:

0x400E0080 [0], 0x400E0094 [1], 0x400E00A8 [2], 0x400E00BC [3]

Access:

Read-write

Reset:

0x10000003

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
TDF_MODE

18
17
TDF_CYCLES

15
–

14
–

13
–

12
DBW

11
–

10
–

7
–

6
–

3
–

2
–

EXNW_MODE

9
–

8
BAT

1
0
WRITE_MODE READ_MODE

• READ_MODE
1 (NRD_CTRL): The Read operation is controlled by the NRD signal.
0 (NCS_CTRL): The Read operation is controlled by the NCS signal.
• WRITE_MODE
1 (NWE_CTRL): The Write operation is controlled by the NWE signal.
0 (NCS_CTRL): The Write operation is controller by the NCS signal.
• EXNW_MODE: NWAIT Mode
The NWAIT signal is used to extend the current read or write signal. It is only taken into account during the pulse phase
Read and Write controlling signal. When the use of NWAIT is enabled, at least one cycle hold duration must be programmed for the read and write controlling signal
Value

Name

DISABLED

Description
Disabled
Reserved

FROZEN

Frozen Mode

READY

Ready Mode

• Disabled: The NWAIT input signal is ignored on the corresponding Chip Select.
• Frozen Mode: If asserted, the NWAIT signal freezes the current read or write cycle. After deassertion, the read/write
cycle is resumed from the point where it was stopped.
• Ready Mode: The NWAIT signal indicates the availability of the external device at the end of the pulse of the controlling
read or write signal, to complete the access. If high, the access normally completes. If low, the access is extended until
NWAIT returns high.
• BAT: Byte Access Type
This field is used only if DBW defines a 16-bit data bus.

433
6430F–ATARM–21-Feb-12

1 (BYTE_WRITE): Byte write access type:
– Write operation is controlled using NCS, NWR0, NWR1.
– Read operation is controlled using NCS and NRD.

0 (BYTE_SELECT): Byte select access type:
– Write operation is controlled using NCS, NWE, NBS0, NBS1.
– Read operation is controlled using NCS, NRD, NBS0, NBS1.
• DBW: Data Bus Width
Value

Name

Description

BIT_8

8-bit bus

BIT_16

16-bit bus

• TDF_CYCLES: Data Float Time
This field gives the integer number of clock cycles required by the external device to release the data after the rising edge
of the read controlling signal. The SMC always provide one full cycle of bus turnaround after the TDF_CYCLES period. The
external bus cannot be used by another chip select during TDF_CYCLES + 1 cycles. From 0 up to 15 TDF_CYCLES can
be set.
• TDF_MODE: TDF Optimization
1: TDF optimization is enabled.
– The number of TDF wait states is optimized using the setup period of the next read/write access.

0: TDF optimization is disabled.
– The number of TDF wait states is inserted before the next access begins.

434

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.22 SMC OCMS Register
Name:
SMC_OCMS
Address:

0x400E0110

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
SRSE

0
SMSE

• SMSE: Static Memory Controller Scrambling Enable
0: Disable “Off Chip” Scrambling for SMC access.
1: Enable “Off Chip” Scrambling for SMC access. (If OCMS field is set to 1 in the relevant SMC_TIMINGS register.)
• SRSE: SRAM Scrambling Enable
0: Disable SRAM Scrambling for SRAM access.
1: Enable SRAM Scrambling for SRAM access. (If OCMS field is set to 1 in the relevant SMC_TIMINGS register.)

435
6430F–ATARM–21-Feb-12

25.18.23 SMC OCMS Key1 Register
Name:
SMC_KEY1
Address:

0x400E0114

Access:

Write-once

Reset:

0x00000000

KEY1
23

20
KEY1

12
KEY1

4
KEY1

• KEY1: Off Chip Memory Scrambling (OCMS) Key Part 1
When Off Chip Memory Scrambling is enabled by setting the SMC_OMCS and SMC_TIMINGS registers in accordance,
the data scrambling depends on KEY1 and KEY2 values.

436

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.24 SMC OCMS Key2 Register
Name:
SMC_KEY2
Address:

0x400E0118

Access:

Write-once

Reset:

0x00000000

KEY2
23

20
KEY2

12
KEY2

4
KEY2

• KEY2: Off Chip Memory Scrambling (OCMS) Key Part 2
When Off Chip Memory Scrambling is enabled by setting the SMC_OMCS and SMC_TIMINGS registers in accordance,
the data scrambling depends on KEY2 and KEY1 values.

437
6430F–ATARM–21-Feb-12

25.18.25 SMC Write Protection Control
Name:
SMC_WPCR
Address:

0x400E01E4

Access:

Write-only

0
WP_EN

WP_KEY
23

20
WP_KEY

12
WP_KEY

• WP_EN: Write Protection Enable
0: Disables the Write Protection if WP_KEY corresponds.
1: Enables the Write Protection if WP_KEY corresponds.
• WP_KEY: Write Protection KEY password
Should be written at value 0x534D43 (ASCII code for “SMC”). Writing any other value in this field has no effect.

438

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
25.18.26 SMC Write Protection Status
Name:
SMC_WPSR
Address:

0x400E01E8

Access:

Read-only

31
-

30
-

29
-

28
-

27
-

26
-

25
-

24
-

WP_VSRC
15

12
WP_VSRC

7
-

6
-

5
-

4
-

WP_VS

• WP_VS: Write Protection Violation Status
0: No Write Protect Violation has occurred since the last read of the SMC_WPSR register.
1: A Write Protect Violation has occurred since the last read of the SMC_WPSR register. If this violation is an unauthorized
attempt to write a protected register, the associated violation is reported into field WP_VSRC.
• WP_VSRC: Write Protection Violation Source
WP_VSRC field Indicates the Register offset where the last violation occurred.

439
6430F–ATARM–21-Feb-12

440

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
26. Peripheral DMA Controller (PDC)
This document describes the AHB Peripheral DMA Controller (AHB PDC) version 1.0.0.

26.1

Description
The Peripheral DMA Controller (PDC) transfers data between on-chip serial peripherals and the
on- and/or off-chip memories. The link between the PDC and a serial peripheral is operated by
the AHB to APB bridge.
The user interface of each PDC channel is integrated into the user interface of the peripheral it
serves. The user interface of mono directional channels (receive only or transmit only), contains
two 32-bit memory pointers and two 16-bit counters, one set (pointer, counter) for current transfer and one set (pointer, counter) for next transfer. The bi-directional channel user interface
contains four 32-bit memory pointers and four 16-bit counters. Each set (pointer, counter) is
used by current transmit, next transmit, current receive and next receive.
Using the PDC removes processor overhead by reducing its intervention during the transfer.
This significantly reduces the number of clock cycles required for a data transfer, which
improves microcontroller performance.
To launch a transfer, the peripheral triggers its associated PDC channels by using transmit and
receive signals. When the programmed data is transferred, an end of transfer interrupt is generated by the peripheral itself.

26.2

Embedded Characteristics
• AMBA™ Advanced High-performance Bus (AHB Lite) Compliant Master
• Performs Transfers to/from APB Communication Serial Peripherals
• Supports Half-duplex and Full-duplex Peripherals

441
6430F–ATARM–21-Feb-12

26.3

Block Diagram

Figure 26-1. Block Diagram
FULL DUPLEX
PERIPHERAL

PDC

THR

PDC Channel A

RHR

PDC Channel B

Control

Status & Control

HALF DUPLEX
PERIPHERAL

Control

THR
PDC Channel C
RHR

Control

Status & Control

RECEIVE or TRANSMIT
PERIPHERAL
RHR or THR

Control

26.4
26.4.1

PDC Channel D

Status & Control

Functional Description
Configuration
The PDC channel user interface enables the user to configure and control data transfers for
each channel. The user interface of each PDC channel is integrated into the associated peripheral user interface.
The user interface of a serial peripheral, whether it is full or half duplex, contains four 32-bit
pointers (RPR, RNPR, TPR, TNPR) and four 16-bit counter registers (RCR, RNCR, TCR,
TNCR). However, the transmit and receive parts of each type are programmed differently: the

442

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
transmit and receive parts of a full duplex peripheral can be programmed at the same time,
whereas only one part (transmit or receive) of a half duplex peripheral can be programmed at a
time.
32-bit pointers define the access location in memory for current and next transfer, whether it is
for read (transmit) or write (receive). 16-bit counters define the size of current and next transfers.
It is possible, at any moment, to read the number of transfers left for each channel.
The PDC has dedicated status registers which indicate if the transfer is enabled or disabled for
each channel. The status for each channel is located in the associated peripheral status register.
Transfers can be enabled and/or disabled by setting TXTEN/TXTDIS and RXTEN/RXTDIS in
the peripheral’s Transfer Control Register.
At the end of a transfer, the PDC channel sends status flags to its associated peripheral. These
flags are visible in the peripheral status register (ENDRX, ENDTX, RXBUFF, and TXBUFE).
Refer to Section 26.4.3 and to the associated peripheral user interface.
26.4.2

Memory Pointers
Each full duplex peripheral is connected to the PDC by a receive channel and a transmit channel. Both channels have 32-bit memory pointers that point respectively to a receive area and to
a transmit area in on- and/or off-chip memory.
Each half duplex peripheral is connected to the PDC by a bidirectional channel. This channel
has two 32-bit memory pointers, one for current transfer and the other for next transfer. These
pointers point to transmit or receive data depending on the operating mode of the peripheral.
Depending on the type of transfer (byte, half-word or word), the memory pointer is incremented
respectively by 1, 2 or 4 bytes.
If a memory pointer address changes in the middle of a transfer, the PDC channel continues
operating using the new address.

26.4.3

Transfer Counters
Each channel has two 16-bit counters, one for current transfer and the other one for next transfer. These counters define the size of data to be transferred by the channel. The current transfer
counter is decremented first as the data addressed by current memory pointer starts to be transferred. When the current transfer counter reaches zero, the channel checks its next transfer
counter. If the value of next counter is zero, the channel stops transferring data and sets the
appropriate flag. But if the next counter value is greater then zero, the values of the next
pointer/next counter are copied into the current pointer/current counter and the channel resumes
the transfer whereas next pointer/next counter get zero/zero as values. At the end of this transfer the PDC channel sets the appropriate flags in the Peripheral Status Register.
The following list gives an overview of how status register flags behave depending on the counters’ values:
• ENDRX flag is set when the PERIPH_RCR register reaches zero.
• RXBUFF flag is set when both PERIPH_RCR and PERIPH_RNCR reach zero.
• ENDTX flag is set when the PERIPH_TCR register reaches zero.
• TXBUFE flag is set when both PERIPH_TCR and PERIPH_TNCR reach zero.
These status flags are described in the Peripheral Status Register.

443
6430F–ATARM–21-Feb-12

26.4.4

Data Transfers
The serial peripheral triggers its associated PDC channels’ transfers using transmit enable
(TXEN) and receive enable (RXEN) flags in the transfer control register integrated in the peripheral’s user interface.
When the peripheral receives an external data, it sends a Receive Ready signal to its PDC
receive channel which then requests access to the Matrix. When access is granted, the PDC
receive channel starts reading the peripheral Receive Holding Register (RHR). The read data
are stored in an internal buffer and then written to memory.
When the peripheral is about to send data, it sends a Transmit Ready to its PDC transmit channel which then requests access to the Matrix. When access is granted, the PDC transmit
channel reads data from memory and puts them to Transmit Holding Register (THR) of its associated peripheral. The same peripheral sends data according to its mechanism.

26.4.5

PDC Flags and Peripheral Status Register
Each peripheral connected to the PDC sends out receive ready and transmit ready flags and the
PDC sends back flags to the peripheral. All these flags are only visible in the Peripheral Status
Register.
Depending on the type of peripheral, half or full duplex, the flags belong to either one single
channel or two different channels.

26.4.5.1

Receive Transfer End
This flag is set when PERIPH_RCR register reaches zero and the last data has been transferred
to memory.
It is reset by writing a non zero value in PERIPH_RCR or PERIPH_RNCR.

26.4.5.2

Transmit Transfer End
This flag is set when PERIPH_TCR register reaches zero and the last data has been written into
peripheral THR.
It is reset by writing a non zero value in PERIPH_TCR or PERIPH_TNCR.

26.4.5.3

Receive Buffer Full
This flag is set when PERIPH_RCR register reaches zero with PERIPH_RNCR also set to zero
and the last data has been transferred to memory.
It is reset by writing a non zero value in PERIPH_TCR or PERIPH_TNCR.

26.4.5.4

Transmit Buffer Empty
This flag is set when PERIPH_TCR register reaches zero with PERIPH_TNCR also set to zero
and the last data has been written into peripheral THR.
It is reset by writing a non zero value in PERIPH_TCR or PERIPH_TNCR.

444

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
26.5

Peripheral DMA Controller (PDC) User Interface

Table 26-1.

Offset

Name
(1)

Access

Reset

0x100

Receive Pointer Register

PERIPH _RPR

Read-write

0x104

Receive Counter Register

PERIPH_RCR

Read-write

0x108

Transmit Pointer Register

PERIPH_TPR

Read-write

0x10C

Transmit Counter Register

PERIPH_TCR

Read-write

0x110

Receive Next Pointer Register

PERIPH_RNPR

Read-write

0x114

Receive Next Counter Register

PERIPH_RNCR

Read-write

0x118

Transmit Next Pointer Register

PERIPH_TNPR

Read-write

0x11C

Transmit Next Counter Register

PERIPH_TNCR

Read-write

0x120

Transfer Control Register

PERIPH_PTCR

Write-only

0x124

Transfer Status Register

PERIPH_PTSR

Read-only

Note:

1. PERIPH: Ten registers are mapped in the peripheral memory space at the same offset. These can be defined by the user
according to the function and the desired peripheral.)

445
6430F–ATARM–21-Feb-12

26.5.1
Name:

Receive Pointer Register
PERIPH_RPR

Access:
31

Read-write
30

RXPTR
23

20
RXPTR

12
RXPTR

4
RXPTR

• RXPTR: Receive Pointer Register
RXPTR must be set to receive buffer address.
When a half duplex peripheral is connected to the PDC, RXPTR = TXPTR.

446

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
26.5.2
Name:

Receive Counter Register
PERIPH_RCR

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

RXCTR
7

4
RXCTR

• RXCTR: Receive Counter Register
RXCTR must be set to receive buffer size.
When a half duplex peripheral is connected to the PDC, RXCTR = TXCTR.
0 = Stops peripheral data transfer to the receiver
1 - 65535 = Starts peripheral data transfer if corresponding channel is active

447
6430F–ATARM–21-Feb-12

26.5.3
Name:

Transmit Pointer Register
PERIPH_TPR

Access:
31

Read-write
30

TXPTR
23

20
TXPTR

12
TXPTR

4
TXPTR

• TXPTR: Transmit Counter Register
TXPTR must be set to transmit buffer address.
When a half duplex peripheral is connected to the PDC, RXPTR = TXPTR.

448

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
26.5.4
Name:

Transmit Counter Register
PERIPH_TCR

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

TXCTR
7

4
TXCTR

• TXCTR: Transmit Counter Register
TXCTR must be set to transmit buffer size.
When a half duplex peripheral is connected to the PDC, RXCTR = TXCTR.
0 = Stops peripheral data transfer to the transmitter
1- 65535 = Starts peripheral data transfer if corresponding channel is active

449
6430F–ATARM–21-Feb-12

26.5.5
Name:

Receive Next Pointer Register
PERIPH_RNPR

Access:
31

Read-write
30

RXNPTR
23

20
RXNPTR

12
RXNPTR

4
RXNPTR

• RXNPTR: Receive Next Pointer
RXNPTR contains next receive buffer address.
When a half duplex peripheral is connected to the PDC, RXNPTR = TXNPTR.

450

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
26.5.6
Name:

Receive Next Counter Register
PERIPH_RNCR

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

RXNCTR
7

4
RXNCTR

• RXNCTR: Receive Next Counter
RXNCTR contains next receive buffer size.
When a half duplex peripheral is connected to the PDC, RXNCTR = TXNCTR.

451
6430F–ATARM–21-Feb-12

26.5.7
Name:

Transmit Next Pointer Register
PERIPH_TNPR

Access:
31

Read-write
30

TXNPTR
23

20
TXNPTR

12
TXNPTR

4
TXNPTR

• TXNPTR: Transmit Next Pointer
TXNPTR contains next transmit buffer address.
When a half duplex peripheral is connected to the PDC, RXNPTR = TXNPTR.

452

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
26.5.8
Name:

Transmit Next Counter Register
PERIPH_TNCR

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

TXNCTR
7

4
TXNCTR

• TXNCTR: Transmit Counter Next
TXNCTR contains next transmit buffer size.
When a half duplex peripheral is connected to the PDC, RXNCTR = TXNCTR.

453
6430F–ATARM–21-Feb-12

26.5.9
Name:

Transfer Control Register
PERIPH_PTCR

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
TXTDIS

8
TXTEN

7
–

6
–

5
–

4
–

3
–

2
–

1
RXTDIS

0
RXTEN

• RXTEN: Receiver Transfer Enable
0 = No effect.
1 = Enables PDC receiver channel requests if RXTDIS is not set.
When a half duplex peripheral is connected to the PDC, enabling the receiver channel requests automatically disables the
transmitter channel requests. It is forbidden to set both TXTEN and RXTEN for a half duplex peripheral.
• RXTDIS: Receiver Transfer Disable
0 = No effect.
1 = Disables the PDC receiver channel requests.
When a half duplex peripheral is connected to the PDC, disabling the receiver channel requests also disables the transmitter channel requests.
• TXTEN: Transmitter Transfer Enable
0 = No effect.
1 = Enables the PDC transmitter channel requests.
When a half duplex peripheral is connected to the PDC, it enables the transmitter channel requests only if RXTEN is not
set. It is forbidden to set both TXTEN and RXTEN for a half duplex peripheral.
• TXTDIS: Transmitter Transfer Disable
0 = No effect.
1 = Disables the PDC transmitter channel requests.
When a half duplex peripheral is connected to the PDC, disabling the transmitter channel requests disables the receiver
channel requests.

454

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
26.5.10
Name:

Transfer Status Register
PERIPH_PTSR

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
TXTEN

7
–

6
–

5
–

4
–

3
–

2
–

1
–

0
RXTEN

• RXTEN: Receiver Transfer Enable
0 = PDC Receiver channel requests are disabled.
1 = PDC Receiver channel requests are enabled.
• TXTEN: Transmitter Transfer Enable
0 = PDC Transmitter channel requests are disabled.
1 = PDC Transmitter channel requests are enabled.

455
6430F–ATARM–21-Feb-12

456

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
27. Clock Generator
27.1

Description
The Clock Generator is made up of:
• A Low Power 32,768 Hz Slow Clock Oscillator with bypass mode
• A Low Power RC Oscillator
• A 3 to 20 MHz Crystal Oscillator (12 MHz needed in case of USB), which can be bypassed
• A factory programmed Fast RC Oscillator, 3 output frequencies can be selected: 4, 8 or
12 MHz. By default 4 MHz is selected.
• A 480 MHz UTMI PLL providing a clock for the USB High Speed Device Controller
• A 96 to 192 MHz programmable PLL (input from 8 to 16 MHz), capable of providing the clock
MCK to the processor and to the peripherals.
It provides the following clocks:
• SLCK, the Slow Clock, which is the only permanent clock within the system
• MAINCK is the output of the Main Clock Oscillator selection: either Crystal Oscillator or
4/8/12 MHz Fast RC Oscillator
• PLLACK is the output of the Divider and 96 to 192 MHz programmable PLL (PLLA)
• UPLLCK is the output of the 480 MHz UTMI PLL (UPLL)
The Clock Generator User Interface is embedded within the Power Management Controller and
is described in Section 28.14 ”Power Management Controller (PMC) User Interface”. However,
the Clock Generator registers are named CKGR_.

457
6430F–ATARM–21-Feb-12

27.2

Block Diagram
Figure 27-1. Clock Generator Block Diagram
Clock Generator
XTALSEL
(Supply Controller)

Embedded
32 kHz RC
Oscillator

0
Slow Clock
SLCK

XIN32
XOUT32

32768 Hz
Crystal
Oscillator

MOSCSEL
Embedded
12/8/4 MHz
Fast
RC Oscillator

0
Main Clock
MAINCK

XIN
XOUT

3-20 MHz
Crystal
Oscillator

Status

PLLA and
Divider

PLLA Clock
PLLACK

USB UTMI
PLL

UPLL Clock
UPLLCK

Control

Power
Management
Controller

458

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
27.3

Slow Clock
The Slow Clock is generated either by the Slow Clock Crystal Oscillator or by the Slow Clock RC
Oscillator.
The selection is made by writing the XTALSEL bit in the Supply Controller Control Register
(SUPC_CR).
By default, the RC Oscillator is selected.

27.3.1

Slow Clock RC Oscillator
By default, the Slow Clock RC Oscillator is enabled and selected. The user has to take into
account the possible drifts of the RC Oscillator. More details are given in the section “DC Characteristics” of the product datasheet.
It can be disabled via the XTALSEL bit in the Supply Controller Control Register (SUPC_CR).

27.3.2

Slow Clock Crystal Oscillator
The Clock Generator integrates a 32,768 Hz low-power oscillator.The XIN and XOUT pins must
be connected to a 32,768 Hz crystal. Two external capacitors must be wired as shown in Figure
27-2.More details are given in the section “DC Characteristics” of the product datasheet.
Note that the user is not obliged to use the Slow Clock Crystal and can use the RC Oscillator
instead. In this case, XIN and XOUT can be left unconnected.
Figure 27-2. Typical Slow Clock Crystal Oscillator Connection
XIN32

XOUT32

GNDBU

32,768 Hz
Crystal

The user can set the Slow Clock Crystal Oscillator in bypass mode instead of connecting a crystal. In this case, the user has to provide the external clock signal on XIN32. The input
characteristics of the XIN pin under these conditions are given in the product electrical characteristics section.
The programmer has to be sure to set the OSCBYPASS bit in the Supply Controller Mode Register (SUPC_MR) and XTALSEL bit in the Supply Controller Control Register (SUPC_CR).

459
6430F–ATARM–21-Feb-12

27.4

Main Clock
Figure 27-3 shows the Main Clock block diagram.
Figure 27-3. Main Clock Block Diagram
MOSCRCEN

MOSCRCF
MOSCRCS

4/8/12 MHz
Fast RC
Oscillator

MOSCSEL

MOSCSELS

1
MAINCK
Main Clock

MOSCXTEN

0
XIN

3-20 MHz
Crystal
Oscillator

XOUT

MOSCXTCNT

3-20 MHz Crystal
Oscillator
Counter

SLCK
Slow Clock

MOSCXTS

MOSCRCEN
MOSCXTEN
MOSCSEL

Main Clock
Frequency
Counter

MAINF
MAINRDY

The Main Clock has two sources:
• 4/8/12 MHz Fast RC Oscillator which starts very quickly and is used at startup
• 3 to 20 MHz Crystal Oscillator, which can be bypassed
27.4.1

4/8/12 MHz Fast RC Oscillator
After reset, the 4/8/12 MHz Fast RC Oscillator is enabled with the 4 MHz frequency selected and
it is selected as the source of MAINCK. MAINCK is the default clock selected to start up the
system.
The Fast RC Oscillator 8 and 12 MHz frequencies are calibrated in production. Note that is not
the case for the 4 MHz frequency.
Please refer to the “DC Characteristics” section of the product datasheet.
The software can disable or enable the 4/8/12 MHz Fast RC Oscillator with the MOSCRCEN bit
in the Clock Generator Main Oscillator Register (CKGR_MOR).
The user can also select the output frequency of the Fast RC Oscillator: either 4 MHz, 8 MHz or
12 MHz are available. It can be done through MOSCRCF bits in CKGR_MOR. When changing

460

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
this frequency selection, the MOSCRCS bit in the Power Management Controller Status Register (PMC_SR) is automatically cleared and MAINCK is stopped until the oscillator is stabilized.
Once the oscillator is stabilized, MAINCK restarts and MOSCRCS is set.
When disabling the Main Clock by clearing the MOSCRCEN bit in CKGR_MOR, the MOSCRCS
bit in the Power Management Controller Status Register (PMC_SR) is automatically cleared,
indicating the Main Clock is off.
Setting the MOSCRCS bit in the Power Management Controller Interrupt Enable Register
(PMC_IER) can trigger an interrupt to the processor.
27.4.2

3 to 20 MHz Crystal Oscillator
After reset, the 3 to 20 MHz Crystal Oscillator is disabled and it is not selected as the source of
MAINCK.
The user can select the 3 to 20 MHz crystal oscillator to be the source of MAINCK, as it provides
a more accurate frequency. The software enables or disables the main oscillator so as to reduce
power consumption by clearing the MOSCXTEN bit in the Main Oscillator Register
(CKGR_MOR).
When disabling the main oscillator by clearing the MOSCXTEN bit in CKGR_MOR, the
MOSCXTS bit in PMC_SR is automatically cleared, indicating the Main Clock is off.
When enabling the main oscillator, the user must initiate the main oscillator counter with a value
corresponding to the startup time of the oscillator. This startup time depends on the crystal frequency connected to the oscillator.
When the MOSCXTEN bit and the MOSCXTCNT are written in CKGR_MOR to enable the main
oscillator, the MOSCXTS bit in the Power Management Controller Status Register (PMC_SR) is
cleared and the counter starts counting down on the slow clock divided by 8 from the MOSCXTCNT value. Since the MOSCXTCNT value is coded with 8 bits, the maximum startup time is
about 62 ms.
When the counter reaches 0, the MOSCXTS bit is set, indicating that the main clock is valid.
Setting the MOSCXTS bit in PMC_IMR can trigger an interrupt to the processor.

27.4.3

Main Clock Oscillator Selection
The user can select either the 4/8/12 MHz Fast RC Oscillator or the 3 to 20 MHz Crystal Oscillator to be the source of Main Clock.
The advantage of the 4/8/12 MHz Fast RC Oscillator is to have fast startup time, this is why it is
selected by default (to start up the system) and when entering in Wait Mode.
The advantage of the 3 to 20 MHz Crystal Oscillator is that it is very accurate.
The selection is made by writing the MOSCSEL bit in the Main Oscillator Register
(CKGR_MOR). The switch of the Main Clock source is glitch free, so there is no need to run out
of SLCK, PLLACK or UPLLCK in order to change the selection. The MOSCSELS bit of the
Power Management Controller Status Register (PMC_SR) allows knowing when the switch
sequence is done.
Setting the MOSCSELS bit in PMC_IMR can trigger an interrupt to the processor.

461
6430F–ATARM–21-Feb-12

27.4.4

Main Clock Frequency Counter
The device features a Main Clock frequency counter that provides the frequency of the Main
Clock.
The Main Clock frequency counter is reset and starts incrementing at the Main Clock speed after
the next rising edge of the Slow Clock in the following cases:
• when the 4/8/12 MHz Fast RC Oscillator clock is selected as the source of Main Clock and
when this oscillator becomes stable (i.e., when the MOSCRCS bit is set)
• when the 3 to 20 MHz Crystal Oscillator is selected as the source of Main Clock and when
this oscillator becomes stable (i.e., when the MOSCXTS bit is set)
• when the Main Clock Oscillator selection is modified
Then, at the 16th falling edge of Slow Clock, the MAINFRDY bit in the Clock Generator Main
Clock Frequency Register (CKGR_MCFR) is set and the counter stops counting. Its value can
be read in the MAINF field of CKGR_MCFR and gives the number of Main Clock cycles during
16 periods of Slow Clock, so that the frequency of the 4/8/12 MHz Fast RC Oscillator or 3 to 20
MHz Crystal Oscillator can be determined.

462

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
27.5

Divider and PLLA Block
The PLLA embeds an input divider to increase the accuracy of the resulting clock signals. However, the user must respect the PLLA minimum input frequency when programming the divider.
Figure 27-4 shows the block diagram of the divider and PLLA block.
Figure 27-4. Divider and PLLA Block Diagram
DIVA

MAINCK

Divider

MULA

OUTA

PLLA

PLLACK

PLLACOUNT
SLCK

27.5.1

PLLA
Counter

LOCKA

Divider and Phase Lock Loop Programming
The divider can be set between 1 and 255 in steps of 1. When a divider field (DIV) is set to 0, the
output of the corresponding divider and the PLL output is a continuous signal at level 0. On
reset, each DIV field is set to 0, thus the corresponding PLL input clock is set to 0.
The PLLA allows multiplication of the divider’s outputs. The PLLA clock signal has a frequency
that depends on the respective source signal frequency and on the parameters DIVA and
MULA. The factor applied to the source signal frequency is (MULA + 1)/DIVA. When MULA is
written to 0, the PLLA is disabled and its power consumption is saved. Re-enabling the PLLA
can be performed by writing a value higher than 0 in the MULA field.
Whenever the PLLA is re-enabled or one of its parameters is changed, the LOCKA bit in
PMC_SR is automatically cleared. The values written in the PLLACOUNT field in CKGR_PLLAR
are loaded in the PLLA counter. The PLLA counter then decrements at the speed of the Slow
Clock until it reaches 0. At this time, the LOCK bit is set in PMC_SR and can trigger an interrupt
to the processor. The user has to load the number of Slow Clock cycles required to cover the
PLLA transient time into the PLLACOUNT field.

463
6430F–ATARM–21-Feb-12

27.6

UTMI Phase Lock Loop Programming
The clock source of the UTMI PLL is the 3-20 MHz Crystal Oscillator. When the 4/8/12 MHz Fast
RC Oscillator is selected as the source of MAINCK, the 12 MHz frequency must also be selected
because the UTMI PLL multiplier contains a built-in multiplier of x 40 to obtain the USB High
Speed 480 MHz.
A 12 MHz crystal is needed to use the USB.
Figure 27-5. UTMI PLL Block Diagram
UPLLEN

3-20 MHz Crystal

UTMI PLL

UPLLCK

UPLLCOUNT
SLCK

UTMI PLL
Counter

LOCKU

Whenever the UTMI PLL is enabled by writing UPLLEN in CKGR_UCKR, the LOCKU bit in
PMC_SR is automatically cleared. The values written in the PLLCOUNT field in CKGR_UCKR
are loaded in the UTMI PLL counter. The UTMI PLL counter then decrements at the speed of
the Slow Clock divided by 8 until it reaches 0. At this time, the LOCKU bit is set in PMC_SR and
can trigger an interrupt to the processor. The user has to load the number of Slow Clock cycles
required to cover the UTMI PLL transient time into the PLLCOUNT field.

464

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28. Power Management Controller (PMC)
28.1

Description
The Power Management Controller (PMC) optimizes power consumption by controlling all system and user peripheral clocks. The PMC enables/disables the clock inputs to many of the
peripherals and the Cortex-M3 Processor.
The Power Management Controller provides the following clocks:
• MCK, the Master Clock, programmable from a few hundred Hz to the maximum operating
frequency of the device. It is available to the modules running permanently, such as the
Enhanced Embedded Flash Controller.
• Processor Clock (HCLK) is automatically switched off when the processor enters Sleep
Mode.
• Free running processor Clock (FCLK)
• the Cortex-M3 SysTick external clock
• the USB Device HS Clock (UDPCK)
• Peripheral Clocks, typically MCK, provided to the embedded peripherals (USART, PMC, SPI,
TWI, TC, HSMCI, etc.) and independently controllable. In order to reduce the number of clock
names in a product, the Peripheral Clocks are named MCK in the product datasheet.
• Programmable Clock Outputs can be selected from the clocks provided by the clock
generator and driven on the PCKx pins.

465
6430F–ATARM–21-Feb-12

28.2

Block Diagram

Figure 28-1. General Clock Block Diagram
Clock Generator
Processor
Clock
Controller

XTALSEL
(Supply Controller)

Processor Clock
HCLK

Sleep Mode
Embedded
32 kHz RC
Oscillator

0
Divider
/8

Slow Clock
SLCK
XIN32
XOUT32

32768 Hz
Crystal
Oscillator

SLCK

Master Clock Controller

UPLLCK/2
PLLACK

Embedded
12/8/4 MHz
Fast
RC Oscillator

Free Running Clock
FCLK
Master Clock
MCK

Prescaler
/1,/2,/3,/4,...,/64
Peripherals
Clock Controller

0
Main Clock
MAINCK

XOUT

SysTick

MAINCK

MOSCSEL

XIN

int

3-20 MHz
Crystal
Oscillator

periph_clk[..]

ON/OFF

1
MCK

Programmable Clock Controller

SLCK
MAINCK
UPLLCK/2

PLLA and
Divider

PLLACK

PLLA Clock
PLLACK

ON/OFF
Prescaler
/1,/2,/4,...,/64

pck[..]

Divider
/2
USB Clock
USB UTMI
PLL

Status

UPLL Clock
UPLLCK

UDPCK

Control

Power
Management
Controller

28.3

Master Clock Controller
The Master Clock Controller provides selection and division of the Master Clock (MCK). MCK is
the clock provided to all the peripherals and the memory controller.
The Master Clock is selected from one of the clocks provided by the Clock Generator. Selecting
the Slow Clock provides a Slow Clock signal to the whole device. Selecting the Main Clock
saves power consumption of the PLLs.
The Master Clock Controller is made up of a clock selector and a prescaler.
The Master Clock selection is made by writing the CSS field (Clock Source Selection) in
PMC_MCKR (Master Clock Register). The prescaler supports the division by a power of 2 of the
selected clock between 1 and 64, and the division by 3. The PRES field in PMC_MCKR programs the prescaler.
Each time PMC_MCKR is written to define a new Master Clock, the MCKRDY bit is cleared in
PMC_SR. It reads 0 until the Master Clock is established. Then, the MCKRDY bit is set and can
trigger an interrupt to the processor. This feature is useful when switching from a high-speed
clock to a lower one to inform the software when the change is actually done.

466

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 28-2. Master Clock Controller
PMC_MCKR
CSS

PMC_MCKR
PRES

SLCK
MAINCK
PLLACK

Master Clock
Prescaler

MCK

UPLLCK
To the Processor
Clock Controller (PCK)

28.4

Processor Clock Controller
The PMC features a Processor Clock Controller (HCLK) that implements the Processor Sleep
Mode. The Processor Clock can be disabled by executing the WFI (WaitForInterrupt) or the
WFE (WaitForEvent) processor instruction while the LPM bit is at 0 in the PMC Fast Startup
Mode Register (PMC_FSMR).
The Processor Clock HCLK is enabled after a reset and is automatically re-enabled by any
enabled interrupt. The Processor Sleep Mode is achieved by disabling the Processor Clock,
which is automatically re-enabled by any enabled fast or normal interrupt, or by the reset of the
product.
When Processor Sleep Mode is entered, the current instruction is finished before the clock is
stopped, but this does not prevent data transfers from other masters of the system bus.

28.5

SysTick Clock
The SysTick calibration value is fixed to 10500 which allows the generation of a time base of
1 ms with SysTick clock at 10.5 MHz (max HCLK/8).

28.6

Peripheral Clock Controller
The Power Management Controller controls the clocks of each embedded peripheral by the way
of the Peripheral Clock Controller. The user can individually enable and disable the Master
Clock on the peripherals by writing into the Peripheral Clock Enable (PMC_PCER) and Peripheral Clock Disable (PMC_PCDR) registers. The status of the peripheral clock activity can be
read in the Peripheral Clock Status Register (PMC_PCSR).
When a peripheral clock is disabled, the clock is immediately stopped. The peripheral clocks are
automatically disabled after a reset.
In order to stop a peripheral, it is recommended that the system software wait until the peripheral
has executed its last programmed operation before disabling the clock. This is to avoid data corruption or erroneous behavior of the system.
The bit number within the Peripheral Clock Control registers (PMC_PCER, PMC_PCDR, and
PMC_PCSR) is the Peripheral Identifier defined at the product level. Generally, the bit number
corresponds to the interrupt source number assigned to the peripheral.

28.7

Free Running Processor Clock
The free running processor clock (FCLK) used for sampling interrupts and clocking debug blocks
ensures that interrupts can be sampled, and sleep events can be traced while the processor is
sleeping. It is connected to Master Clock (MCK).

467
6430F–ATARM–21-Feb-12

28.8

Programmable Clock Output Controller
The PMC controls 3 signals to be output on external pins, PCKx. Each signal can be independently programmed via the PMC_PCKx registers.
PCKx can be independently selected between the Slow Clock (SLCK), the Main Clock
(MAINCK), the PLLA Clock (PLLACK), the UTMI PLL Clock (UPLLCK/2) and the Master Clock
(MCK) by writing the CSS field in PMC_PCKx. Each output signal can also be divided by a
power of 2 between 1 and 64 by writing the PRES (Prescaler) field in PMC_PCKx.
Each output signal can be enabled and disabled by writing 1 in the corresponding bit, PCKx of
PMC_SCER and PMC_SCDR, respectively. Status of the active programmable output clocks
are given in the PCKx bits of PMC_SCSR (System Clock Status Register).
Moreover, like the PCK, a status bit in PMC_SR indicates that the Programmable Clock is actually what has been programmed in the Programmable Clock registers.
As the Programmable Clock Controller does not manage with glitch prevention when switching
clocks, it is strongly recommended to disable the Programmable Clock before any configuration
change and to re-enable it after the change is actually performed.

28.9

Fast Startup
The SAM3U device allows the processor to restart in less than six microseconds while the
device is in Wait mode. The system enters Wait mode either by writing the WAITMODE bit at 1
in the PMC Clock Generator Main Oscillator Register (CKGR_MOR), of by executing the WaitForEvent (WFE) instruction of the processor while the LPM bit is at 1 in the PMC Fast Startup
Mode Register (PMC_FSMR).
Important: Prior to asserting any WFE instruction to the processor, the internal sources of
wakeup provided by RTT, RTC and USB must be cleared and verified too, that none of the
enabled external wakeup inputs (WKUP) hold an active polarity.
A Fast Startup is enabled upon the detection of a programmed level on one of the 19 wake-up
inputs (WKUP) or upon an active alarm form the RTC, RTT and USB High Speed Device Controller. The polarity of the 16 wake-up inputs is programmable by writing the PMC Fast Startup
Polarity Register (PMC_FSPR).
The Fast Restart circuitry, as shown in Figure 28-3, is fully asynchronous and provides a fast
startup signal to the Power Management Controller. As soon as the fast startup signal is
asserted, this automatically restarts the embedded 4/8/12 MHz Fast RC oscillator.

468

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 28-3. Fast Startup Circuitry
FSTT0
WKUP0

FSTP0

FSTT1

WKUP1

FSTP1
FSTT15
WKUP15

fast_restart

FSTP15

RTTAL

RTT Alarm
RTCAL

RTC Alarm
USBAL

USB Alarm

Each wake-up input pin and alarm can be enabled to generate a Fast Startup event by writing at
1 the corresponding bit in the Fast Startup Mode Register PMC_FSMR.
The user interface does not provide any status for Fast Startup, but the user can easily recover
this information by reading the PIO Controller, and the status registers of the RTC, RTT and
USB High Speed Device Controller.

469
6430F–ATARM–21-Feb-12

28.10 Clock Failure Detector
The clock failure detector allows to monitor the 3 to 20 MHz Crystal Oscillator and to detect an
eventual defect of this oscillator (for example if the crystal is disconnected).
The clock failure detector can be enabled or disabled by means of the CFDEN bit in the PMC
Clock Generator Main Oscillator Register (CKGR_MOR). After reset, the detector is disabled.
However, if the 3 to 20 MHz Crystal Oscillator is disabled, the clock failure detector is disabled
too.
A failure is detected by means of a counter incrementing on the 3 to 20 MHzCrystal oscillator or
Ceramic Resonator-based oscillator clock edge and timing logic clocked on the slow clock RC
oscillator controlling the counter. The counter is cleared when the slow clock RC oscillator signal
is low and enabled when the slow clock RC oscillator is high. Thus the failure detection time is 1
slow clock RC oscillator clock period. If, during the high level period of slow clock RC oscillator,
less than 8 fast crystal clock periods have been counted, then a failure is declared.
If a failure of the 3 to 20 MHz Crystal Oscillator clock is detected, the CFDEV flag is set in the
PMC Status Register (PMC_SR), and can generate an interrupt if it is not masked. The interrupt
remains active until a read operation in the PMC_SR register. The user can know the status of
the clock failure detector at any time by reading the CFDS bit in the PMC_SR register.
If the 3 to 20 MHz Crystal Oscillator clock is selected as the source clock of MAINCK (MOSCSEL = 1), and if the Master Clock Source is PLLACK or UPLLCK (CSS = 2 or 3), then a clock
failure detection switches automatically the Master Clock on MAINCK. Then whatever the PMC
configuration is, a clock failure detection switches automatically MAINCK on the 4/8/12 MHz
Fast RC Oscillator clock. If the Fast RC Oscillator is disabled when a clock failure detection
occurs, it is automatically re-enabled by the clock failure detection mechanism.
A clock failure detection activates a fault output that is connected to the Pulse Width Modulator
(PWM) Controller. With this connection, the PWM controller is able to force its outputs and to
protect the driven device, if a clock failure is detected. This fault output remains active until the
defect is detected and until it is not cleared by the bit FOCLR in the PMC Fault Output Clear
Register (PMC_FOCR).
It takes 2 slow clock RC oscillator cycles to detect and switch from the 3 to 20 MHz Crystal or
Ceramic Resonator-based oscillator to the 4/8/12 MHz Fast RC Oscillator if the Master Clock
source is Main Clock, or 3 slow clock RC oscillator cycles if the Master Clock source is PLL.
The user can know the status of the fault output at any time by reading the FOS bit in the
PMC_SR register.

470

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.11 Programming Sequence
1. Enabling the Main Oscillator:
The main oscillator is enabled by setting the MOSCXTEN field in the CKGR_MOR register.
The user can define a start-up time. This can be achieved by writing a value in the
MOSCXTST field in the CKGR_MOR register.
Once this register has been correctly configured, the user must wait for MOSCXTS field in
the PMC_SR register to be set. This can be done either by polling the status register or by
waiting the interrupt line to be raised if the associated interrupt to MOSCXTS has been
enabled in the PMC_IER register.
Start Up Time = 8 * MOSCXTST / SLCK = 56 Slow Clock Cycles.
So, the main oscillator will be enabled (MOSCXTS bit set) after 56 Slow Clock Cycles.
2. Checking the Main Oscillator Frequency (Optional):
In some situations the user may need an accurate measure of the main clock frequency.
This measure can be accomplished via the CKGR_MCFR register.
Once the MAINFRDY field is set in CKGR_MCFR register, the user may read the MAINF
field in CKGR_MCFR register. This provides the number of main clock cycles within sixteen
slow clock cycles.
3. Setting PLL and Divider:
All parameters needed to configure PLLA and the divider are located in the CKGR_PLLAR
register.
The DIVA field is used to control the divider itself. It must be set to 1 when PLLA is used. By
default, DIVA parameter is set to 0 which means that the divider is turned off.
The MULA field is the PLLA multiplier factor. This parameter can be programmed between 0
and 2047. If MULA is set to 0, PLLA will be turned off, otherwise the PLLA output frequency
is PLLA input frequency multiplied by (MULA + 1).
The PLLACOUNT field specifies the number of slow clock cycles before LOCKA bit is set in
the PMC_SR register after CKGR_PLLAR register has been written.
Once the CKGR_PLLA register has been written, the user must wait for the LOCKA bit to be
set in the PMC_SR register. This can be done either by polling the status register or by waiting the interrupt line to be raised if the associated interrupt to LOCKA has been enabled in
the PMC_IER register. All parameters in CKGR_PLLAR can be programmed in a single
write operation. If at some stage one of the following parameters, MULA, DIVA is modified,
LOCKA bit will go low to indicate that PLLA is not ready yet. When PLLA is locked, LOCKA
will be set again. The user is constrained to wait for LOCKA bit to be set before using the
PLLA output clock.
4. Selection of Master Clock and Processor Clock
The Master Clock and the Processor Clock are configurable via the PMC_MCKR register.
The CSS field is used to select the Master Clock divider source. By default, the selected
clock source is main clock.
The PRES field is used to control the Master Clock prescaler. The user can choose between
different values. Master Clock output is prescaler input divided by PRES parameter. By
default, PRES parameter is set to 1 which means that master clock is equal to main clock.

471
6430F–ATARM–21-Feb-12

Once the PMC_MCKR register has been written, the user must wait for the MCKRDY bit to
be set in the PMC_SR register. This can be done either by polling the status register or by
waiting for the interrupt line to be raised if the associated interrupt to MCKRDY has been
enabled in the PMC_IER register.
The PMC_MCKR register must not be programmed in a single write operation. The preferred programming sequence for the PMC_MCKR register is as follows:
• If a new value for CSS field corresponds to PLL Clock,
– Program the PRES field in the PMC_MCKR register.
– Wait for the MCKRDY bit to be set in the PMC_SR register.
– Program the CSS field in the PMC_MCKR register.
– Wait for the MCKRDY bit to be set in the PMC_SR register.
• If a new value for CSS field corresponds to Main Clock or Slow Clock,
– Program the CSS field in the PMC_MCKR register.
– Wait for the MCKRDY bit to be set in the PMC_SR register.
– Program the PRES field in the PMC_MCKR register.
– Wait for the MCKRDY bit to be set in the PMC_SR register.
If at some stage one of the following parameters, CSS or PRES, is modified, the MCKRDY
bit will go low to indicate that the Master Clock and the Processor Clock are not ready yet.
The user must wait for MCKRDY bit to be set again before using the Master and Processor
Clocks.
Note:

IF PLLx clock was selected as the Master Clock and the user decides to modify it by writing in
CKGR_PLLR, the MCKRDY flag will go low while PLL is unlocked. Once PLL is locked again,
LOCK goes high and MCKRDY is set.
While PLL is unlocked, the Master Clock selection is automatically changed to Slow Clock. For further information, see Section 28.12.2 “Clock Switching Waveforms” on page 475.

Code Example:
write_register(PMC_MCKR,0x00000001)
wait (MCKRDY=1)
write_register(PMC_MCKR,0x00000011)
wait (MCKRDY=1)

The Master Clock is main clock divided by 16.
The Processor Clock is the Master Clock.
5. Selection of Programmable Clocks
Programmable clocks are controlled via registers; PMC_SCER, PMC_SCDR and
PMC_SCSR.
Programmable clocks can be enabled and/or disabled via the PMC_SCER and PMC_SCDR
registers. 3 Programmable clocks can be enabled or disabled. The PMC_SCSR provides a
clear indication as to which Programmable clock is enabled. By default all Programmable
clocks are disabled.
PMC_PCKx registers are used to configure Programmable clocks.

472

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The CSS field is used to select the Programmable clock divider source. Four clock options
are available: main clock, slow clock, PLLACK and UPLLCK. By default, the clock source
selected is main clock.
The PRES field is used to control the Programmable clock prescaler. It is possible to choose
between different values (1, 2, 4, 8, 16, 32, 64). Programmable clock output is prescaler
input divided by PRES parameter. By default, the PRES parameter is set to 0 which means
that master clock is equal to slow clock.
Once the PMC_PCKx register has been programmed, The corresponding Programmable
clock must be enabled and the user is constrained to wait for the PCKRDYx bit to be set in
the PMC_SR register. This can be done either by polling the status register or by waiting the
interrupt line to be raised if the associated interrupt to PCKRDYx has been enabled in the
PMC_IER register. All parameters in PMC_PCKx can be programmed in a single write
operation.
If the CSS and PRES parameters are to be modified, the corresponding Programmable
clock must be disabled first. The parameters can then be modified. Once this has been
done, the user must re-enable the Programmable clock and wait for the PCKRDYx bit to be
set.

Code Example:
write_register(PMC_PCK0,0x00000015)

Programmable clock 0 is main clock divided by 32.
6. Enabling Peripheral Clocks
Once all of the previous steps have been completed, the peripheral clocks can be enabled
and/or disabled via registers PMC_PCER and PMC_PCDR.
15 peripheral clocks can be enabled or disabled. The PMC_PCSR provides a clear view as
to which peripheral clock is enabled.
Note:

Each enabled peripheral clock corresponds to Master Clock.

Code Examples:
write_register(PMC_PCER,0x00000110)

Peripheral clocks 4 and 8 are enabled.

write_register(PMC_PCDR,0x00000010)

Peripheral clock 4 is disabled.

473
6430F–ATARM–21-Feb-12

28.12 Clock Switching Details
28.12.1

Master Clock Switching Timings
Table 28-1 and Table 28-2 give the worst case timings required for the Master Clock to switch
from one selected clock to another one. This is in the event that the prescaler is de-activated.
When the prescaler is activated, an additional time of 64 clock cycles of the new selected clock
has to be added.
Table 28-1.

Clock Switching Timings (Worst Case)
From

Main Clock

SLCK

PLL Clock

–

4 x SLCK +
2.5 x Main Clock

3 x PLL Clock +
4 x SLCK +
1 x Main Clock

0.5 x Main Clock +
4.5 x SLCK

–

3 x PLL Clock +
5 x SLCK

0.5 x Main Clock +
4 x SLCK +
PLLCOUNT x SLCK +
2.5 x PLLx Clock

2.5 x PLL Clock +
5 x SLCK +
PLLCOUNT x SLCK

2.5 x PLL Clock +
4 x SLCK +
PLLCOUNT x SLCK

To
Main Clock

SLCK

PLL Clock

Notes:

1. PLL designates either the PLLA or the UPLL Clock.
2. PLLCOUNT designates either PLLACOUNT or UPLLCOUNT.

Table 28-2.

Clock Switching Timings between Two PLLs (Worst Case)
From

PLLA Clock

UPLL Clock

PLLA Clock

2.5 x PLLA Clock +
4 x SLCK +
PLLACOUNT x SLCK

3 x PLLA Clock +
4 x SLCK +
1.5 x PLLA Clock

UPLL Clock

3 x UPLL Clock +
4 x SLCK +
1.5 x UPLL Clock

2.5 x UPLL Clock +
4 x SLCK +
UPLLCOUNT x SLCK

474

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.12.2

Clock Switching Waveforms
Figure 28-4. Switch Master Clock from Slow Clock to PLL Clock
Slow Clock

PLL Clock

LOCK

MCKRDY

Master Clock

Write PMC_MCKR

Figure 28-5. Switch Master Clock from Main Clock to Slow Clock
Slow Clock

Main Clock

MCKRDY

Master Clock

Write PMC_MCKR

475
6430F–ATARM–21-Feb-12

Figure 28-6. Change PLLA Programming
Slow Clock

PLLA Clock

LOCKA

MCKRDY

Master Clock
Slow Clock
Write CKGR_PLLAR

Figure 28-7. Programmable Clock Output Programming
PLL Clock

PCKRDY

PCKx Output

Write PMC_PCKx

Write PMC_SCER

Write PMC_SCDR

476

PLL Clock is selected

PCKx is enabled

PCKx is disabled

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.13 Write Protection Registers
To prevent any single software error that may corrupt PMC behavior, certain address spaces
can be write-protected by setting the WPEN bit in the “PMC Write Protect Mode Register”
(PMC_WPMR).
If a write access to the protected registers is detected, then the WPVS flag in the PMC Write
Protect Status Register (US_WPSR) is set and the field WPVSRC indicates in which register the
write access has been attempted.
The WPVS flag is reset by writing the PMC Write Protect Mode Register (PMC_WPMR) with the
appropriate access key, WPKEY.
The protected registers are:
• “PMC System Clock Enable Register” on page 479
• “PMC System Clock Disable Register” on page 480
• “PMC Peripheral Clock Enable Register” on page 481
• “PMC Peripheral Clock Disable Register” on page 482
• “PMC UTMI Clock Configuration Register” on page 484
• “PMC Clock Generator Main Oscillator Register” on page 485
• “PMC Clock Generator PLLA Register” on page 488
• “PMC Programmable Clock Register” on page 490
• “PMC Fast Startup Mode Register” on page 496
• “PMC Fast Startup Polarity Register” on page 497

477
6430F–ATARM–21-Feb-12

28.14 Power Management Controller (PMC) User Interface
Table 28-3.

Offset

Name

Access

Reset

0x0000

System Clock Enable Register

PMC_SCER

Write-only

–

0x0004

System Clock Disable Register

PMC_SCDR

Write-only

–

0x0008

System Clock Status Register

PMC _SCSR

Read-only

0x0000_0001

0x000C

Reserved

–

0x0010

Peripheral Clock Enable Register

PMC _PCER

Write-only

N.A.

0x0014

Peripheral Clock Disable Register

PMC_PCDR

Write-only

–

0x0018

Peripheral Clock Status Register

PMC_PCSR

Read-only

0x0000_0000

0x001C

UTMI Clock Register

CKGR_UCKR

Read-write

0x1020_0800

0x0020

Main Oscillator Register

CKGR_MOR

Read-write

0x0000_0001

0x0024

Main Clock Frequency Register

CKGR_MCFR

Read-only

0x0000_0000

0x0028

PLLA Register

CKGR_PLLAR

Read-write

0x0000_3F00

0x002C

Reserved

–

0x0030

Master Clock Register

Read-write

0x0000_0001

–

0x0034 - 0x003C

–
PMC_MCKR

Reserved

–

0x0040

Programmable Clock 0 Register

PMC_PCK0

Read-write

0x0000_0000

0x0044

Programmable Clock 1 Register

PMC_PCK1

Read-write

0x0000_0000

0x0048

Programmable Clock 2 Register

PMC_PCK2

Read-write

0x0000_0000

–

0x004C - 0x005C

Reserved

–

0x0060

Interrupt Enable Register

PMC_IER

Write-only

–

0x0064

Interrupt Disable Register

PMC_IDR

Write-only

–

0x0068

Status Register

PMC_SR

Read-only

0x0001_0008

0x006C

Interrupt Mask Register

PMC_IMR

Read-only

0x0000_0000

0x0070

Fast Startup Mode Register

PMC_FSMR

Read-write

0x0000_0000

0x0074

Fast Startup Polarity Register

PMC_FSPR

Read-write

0x0000_0000

0x0078

Fault Output Clear Register

PMC_FOCR

Write-only

–

0x007C- 0x00FC

478

–

Reserved

–

0xE4

Write Protect Mode Register

PMC_WPMR

Read-write

0x0

0xE8

Write Protect Status Register

PMC_WPSR

Read-only

0x0

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.1
Name:

PMC System Clock Enable Register
PMC_SCER

Address:

0x400E0400

Access:

Write-only

–

PCK2

PCK1

PCK0

–

• PCKx: Programmable Clock x Output Enable
0 = No effect.
1 = Enables the corresponding Programmable Clock output.
This register can only be written if the WPEN bit is cleared in the “PMC Write Protect Mode Register”.

479
6430F–ATARM–21-Feb-12

28.14.2
Name:

PMC System Clock Disable Register
PMC_SCDR

Address:

0x400E0404

Access:

Write-only

–

PCK2

PCK1

PCK0

–

• PCKx: Programmable Clock x Output Disable
0 = No effect.
1 = Disables the corresponding Programmable Clock output.
This register can only be written if the WPEN bit is cleared in the “PMC Write Protect Mode Register”.

28.14.3
Name:

PMC System Clock Status Register
PMC_SCSR

Address:

0x400E0408

Access:

Read-only

–

PCK2

PCK1

PCK0

–

• PCKx: Programmable Clock x Output Status
0 = The corresponding Programmable Clock output is disabled.
1 = The corresponding Programmable Clock output is enabled.

480

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.4
Name:

PMC Peripheral Clock Enable Register
PMC_PCER

Address:

0x400E0410

Access:

Write-only

PID31

PID30

PID29

PID28

PID27

PID26

PID25

PID24

PID23

PID22

PID21

PID20

PID19

PID18

PID17

PID16

PID15

PID14

PID13

PID12

PID11

PID10

PID9

PID8

PID7

PID6

PID5

PID4

PID3

PID2

This register can only be written if the WPEN bit is cleared in the “PMC Write Protect Mode Register”.
• PIDx: Peripheral Clock x Enable
0 = No effect.
1 = Enables the corresponding peripheral clock.
Note:

PID2 to PID31 refer to identifiers as defined in the section “Peripheral Identifiers” in the product datasheet.

Note:

Programming the control bits of the Peripheral ID that are not implemented has no effect on the behavior of the PMC.

481
6430F–ATARM–21-Feb-12

28.14.5
Name:

PMC Peripheral Clock Disable Register
PMC_PCDR

Address:

0x400E0414

Access:

Write-only

PID31

PID30

PID29

PID28

PID27

PID26

PID25

PID24

PID23

PID22

PID21

PID20

PID19

PID18

PID17

PID16

PID15

PID14

PID13

PID12

PID11

PID10

PID9

PID8

PID7

PID6

PID5

PID4

PID3

PID2

This register can only be written if the WPEN bit is cleared in the “PMC Write Protect Mode Register”.
• PIDx: Peripheral Clock x Disable
0 = No effect.
1 = Disables the corresponding peripheral clock.
Note:

482

PID2 to PID31 refer to identifiers as defined in the section “Peripheral Identifiers” in the product datasheet.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.6
Name:

PMC Peripheral Clock Status Register
PMC_PCSR

Address:

0x400E0418

Access:

Read-only

PID31

PID30

PID29

PID28

PID27

PID26

PID25

PID24

PID23

PID22

PID21

PID20

PID19

PID18

PID17

PID16

PID15

PID14

PID13

PID12

PID11

PID10

PID9

PID8

PID7

PID6

PID5

PID4

PID3

PID2

–

• PIDx: Peripheral Clock x Status
0 = The corresponding peripheral clock is disabled.
1 = The corresponding peripheral clock is enabled.
Note:

PID2 to PID31 refer to identifiers as defined in the section “Peripheral Identifiers” in the product datasheet.

483
6430F–ATARM–21-Feb-12

28.14.7
Name:

PMC UTMI Clock Configuration Register
CKGR_UCKR

Address:

0x400E041C

Access:

Read-write

31
–

30
–

29
–

27
–

26
–

25
–

24
–

19
–

18
–

17
–

16
UPLLEN

UPLLCOUNT
15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
–

0
–

This register can only be written if the WPEN bit is cleared in the “PMC Write Protect Mode Register”.
• UPLLEN: UTMI PLL Enable
0 = The UTMI PLL is disabled.
1 = The UTMI PLL is enabled.
When UPLLEN is set, the LOCKU flag is set once the UTMI PLL startup time is achieved.
• UPLLCOUNT: UTMI PLL Start-up Time
Specifies the number of Slow Clock cycles multiplied by 8 for the UTMI PLL start-up time.

484

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.8
Name:

PMC Clock Generator Main Oscillator Register
CKGR_MOR

Address:

0x400E0420

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
CFDEN

24
MOSCSEL

3
MOSCRCEN

2
WAITMODE

1
MOSCXTBY

0
MOSCXTEN

KEY
15

12
MOSCXTST

7
–

5
MOSCRCF

This register can only be written if the WPEN bit is cleared in the “PMC Write Protect Mode Register”.
• KEY: Password
Should be written at value 0x37. Writing any other value in this field aborts the write operation.
• MOSCXTEN: Main Crystal Oscillator Enable
A crystal must be connected between XIN and XOUT.
0 = The Main Crystal Oscillator is disabled.
1 = The Main Crystal Oscillator is enabled. MOSCXTBY must be set to 0.
When MOSCXTEN is set, the MOSCXTS flag is set once the Main Crystal Oscillator startup time is achieved.
• MOSCXTBY: Main Crystal Oscillator Bypass
0 = No effect.
1 = The Main Crystal Oscillator is bypassed. MOSCXTEN must be set to 0. An external clock must be connected on XIN.
When MOSCXTBY is set, the MOSCXTS flag in PMC_SR is automatically set.
Clearing MOSCXTEN and MOSCXTBY bits allows resetting the MOSCXTS flag.
• WAITMODE: Wait Mode Command
0 = No effect.
1 = Enters the device in Wait mode.
Note:

The bit WAITMODE is write-only

• MOSCRCEN: Main On-Chip RC Oscillator Enable
0 = The Main On-Chip RC Oscillator is disabled.
1 = The Main On-Chip RC Oscillator is enabled.
When MOSCRCEN is set, the MOSCRCS flag is set once the Main On-Chip RC Oscillator startup time is achieved.

485
6430F–ATARM–21-Feb-12

• MOSCRCF: Main On-Chip RC Oscillator Frequency Selection
At start-up, the Main On-Chip RC Oscillator frequency is 4 MHz.
0 = The Fast RC Oscillator Frequency is at 4 MHz (default).
1 = The Fast RC Oscillator Frequency is at 8 MHz.
2 = The Fast RC Oscillator Frequency is at 12 MHz.
3 = Reserved.
• MOSCXTST: Main Crystal Oscillator Start-up Time
Specifies the number of Slow Clock cycles multiplied by 8 for the Main Crystal Oscillator start-up time.
• MOSCSEL: Main Oscillator Selection
0 = The Main On-Chip RC Oscillator is selected.
1 = The Main Crystal Oscillator is selected.
• CFDEN: Clock Failure Detector Enable
0 = The Clock Failure Detector is disabled.
1 = The Clock Failure Detector is enabled.

486

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.9
Name:

PMC Clock Generator Main Clock Frequency Register
CKGR_MCFR

Address:

0x400E0424

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
MAINFRDY

MAINF
7

4
MAINF

• MAINF: Main Clock Frequency
Gives the number of Main Clock cycles within 16 Slow Clock periods.
• MAINFRDY: Main Clock Ready
0 = MAINF value is not valid or the Main Oscillator is disabled.
1 = The Main Oscillator has been enabled previously and MAINF value is available.

487
6430F–ATARM–21-Feb-12

28.14.10 PMC Clock Generator PLLA Register
Name:
CKGR_PLLAR
Address:

0x400E0428

Access:

Read-write

31
–

30
–

29
1

28
–

27
–

25
MULA

MULA
15
–

11
PLLACOUNT

–
5

3
DIVA

This register can only be written if the WPEN bit is cleared in the “PMC Write Protect Mode Register”.
Possible limitations on PLLA input frequencies and multiplier factors should be checked before using the PMC.
Warning: Bit 29 must always be set to 1 when programming the CKGR_PLLAR register.
• DIVA: Divider
DIVA

Divider Selected

Divider output is 0

Divider is bypassed (DIVA=1)

2 - 255

Divider output is the selected clock divided by DIVA

• PLLACOUNT: PLLA Counter
Specifies the number of Slow Clock cycles x8 before the LOCKA bit is set in PMC_SR after CKGR_PLLAR is written.
• MULA: PLLA Multiplier
0 = The PLLA is deactivated.
1 up to 2047 = The PLLA Clock frequency is the PLLA input frequency multiplied by MULA + 1.

488

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.11 PMC Master Clock Register
Name:
PMC_MCKR
Address:

0x400E0430

Access:

Read-write

–

UPLLDIV

–

PRES

CSS

• CSS: Master Clock Source Selection
Value

Name

Description

SLOW_CLK

Slow Clock is selected

MAIN_CLK

Main Clock is selected

PLLA_CLK

PLLA Clock is selected

UPLL_CLK

UPLL/2 Clock is selected

• PRES: Processor Clock Prescaler
Value

Name

Description

CLK_1

Selected clock

CLK_2

Selected clock divided by 2

CLK_4

Selected clock divided by 4

CLK_8

Selected clock divided by 8

CLK_16

Selected clock divided by 16

CLK_32

Selected clock divided by 32

CLK_64

Selected clock divided by 64

CLK_3

Selected clock divided by 3

• UPLLDIV: UPLL Divider
0: UPLLDIV = 1 (clock is not divided).
1: UPLLDIV = 2 (clock is divided by 2).
UPLLDIV must be set (clock divided by 2) when UPLL is selected as source of clock for MCK or PCK.

489
6430F–ATARM–21-Feb-12

28.14.12 PMC Programmable Clock Register
Name:
PMC_PCKx
Address:

0x400E0440

Access:

Read-write

–

PRES

–

CSS

This register can only be written if the WPEN bit is cleared in the “PMC Write Protect Mode Register”.
• CSS: Master Clock Source Selection
Value

Name

Description

SLOW_CLK

Slow Clock is selected

MAIN_CLK

Main Clock is selected

PLLA_CLK

PLLA Clock is selected

UPLL_CLK

UPLL/2 Clock is selected

MCK

Master Clock is selected

• PRES: Programmable Clock Prescaler
Value

490

Name

Description

CLK_1

Selected clock

CLK_2

Selected clock divided by 2

CLK_4

Selected clock divided by 4

CLK_8

Selected clock divided by 8

CLK_16

Selected clock divided by 16

CLK_32

Selected clock divided by 32

CLK_64

Selected clock divided by 64

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.13 PMC Interrupt Enable Register
Name:
PMC_IER
Address:

0x400E0460

Access:

Write-only

–

CFDEV

MOSCRCS

MOSCSELS

–

PCKRDY2

PCKRDY1

PCKRDY0

–

LOCKU

–

MCKRDY

–

LOCKA

MOSCXTS

0 = No effect.
1 = Enables the corresponding interrupt.
• MOSCXTS: Main Crystal Oscillator Status Interrupt Enable
• LOCKA: PLL A Lock Interrupt Enable
• MCKRDY: Master Clock Ready Interrupt Enable
• LOCKU: UTMI PLL Lock Interrupt Enable
• PCKRDYx: Programmable Clock Ready x Interrupt Enable
• MOSCSELS: Main Oscillator Selection Status Interrupt Enable
• MOSCRCS: Main On-Chip RC Status Interrupt Enable
• CFDEV: Clock Failure Detector Event Interrupt Enable

491
6430F–ATARM–21-Feb-12

28.14.14 PMC Interrupt Disable Register
Name:
PMC_IDR
Address:

0x400E0464

Access:

Write-only

–

CFDEV

MOSCRCS

MOSCSELS

–

PCKRDY2

PCKRDY1

PCKRDY0

–

LOCKU

–

MCKRDY

–

LOCKA

MOSCXTS

0 = No effect.
1 = Disables the corresponding interrupt.
• MOSCXTS: Main Crystal Oscillator Status Interrupt Disable
• LOCKA: PLL A Lock Interrupt Disable
• MCKRDY: Master Clock Ready Interrupt Disable
• LOCKU: UTMI PLL Lock Interrupt Disable
• PCKRDYx: Programmable Clock Ready x Interrupt Disable
• MOSCSELS: Main Oscillator Selection Status Interrupt Disable
• MOSCRCS: Main On-Chip RC Status Interrupt Disable
• CFDEV: Clock Failure Detector Event Interrupt Disable

492

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.15 PMC Status Register
Name:
PMC_SR
Address:

0x400E0468

Access:

Read-only

–

FOS

CFDS

CFDEV

MOSCRCS

MOSCSELS

–

PCKRDY2

PCKRDY1

PCKRDY0

OSCSELS

LOCKU

–

MCKRDY

–

LOCKA

MOSCXTS

• MOSCXTS: Main XTAL Oscillator Status
0 = Main XTAL oscillator is not stabilized.
1 = Main XTAL oscillator is stabilized.
• LOCKA: PLL A Lock Status
0 = PLL A is not locked
1 = PLL A is locked.
• MCKRDY: Master Clock Status
0 = Master Clock is not ready.
1 = Master Clock is ready.
• LOCKU: UTMI PLL Lock Status
0 = UTMI PLL is not locked.
1 = UTMI PLL is locked.
• OSCSELS: Slow Clock Oscillator Selection
0 = Internal slow clock RC oscillator is selected.
1 = External slow clock 32 kHz oscillator is selected.
• PCKRDYx: Programmable Clock Ready Status
0 = Programmable Clock x is not ready.
1 = Programmable Clock x is ready.
• MOSCSELS: Main Oscillator Selection Status
0 = Selection is in progress.
1 = Selection is done.
• MOSCRCS: Main On-Chip RC Oscillator Status
0 = Main on-chip RC oscillator is not stabilized.
1 = Main on-chip RC oscillator is stabilized.

493
6430F–ATARM–21-Feb-12

• CFDEV: Clock Failure Detector Event
0 = No clock failure detection of the main on-chip RC oscillator clock has occurred since the last read of PMC_SR.
1 = At least one clock failure detection of the main on-chip RC oscillator clock has occurred since the last read of PMC_SR.
• CFDS: Clock Failure Detector Status
0 = A clock failure of the main on-chip RC oscillator clock is not detected.
1 = A clock failure of the main on-chip RC oscillator clock is detected.
• FOS: Clock Failure Detector Fault Output Status
0 = The fault output of the clock failure detector is inactive.
0 = The fault output of the clock failure detector is active.

494

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.16 PMC Interrupt Mask Register
Name:
PMC_IMR
Address:

0x400E046C

Access:

Read-only

–

CFDEV

MOSCRCS

MOSCSELS

–

PCKRDY2

PCKRDY1

PCKRDY0

–

LOCKU

–

MCKRDY

–

LOCKA

MOSCXTS

0 = The corresponding interrupt is disabled.
1 = The corresponding interrupt is enabled.
• MOSCXTS: Main Crystal Oscillator Status Interrupt Mask
• LOCKA: PLL A Lock Interrupt Mask
• MCKRDY: Master Clock Ready Interrupt Mask
• LOCKU: UTMI PLL Lock Interrupt Mask
• PCKRDYx: Programmable Clock Ready x Interrupt Mask
• MOSCSELS: Main Oscillator Selection Status Interrupt Mask
• MOSCRCS: Main On-Chip RC Status Interrupt Mask
• CFDEV: Clock Failure Detector Event Interrupt Mask

495
6430F–ATARM–21-Feb-12

28.14.17 PMC Fast Startup Mode Register
Name:
PMC_FSMR
Address:

0x400E0470

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
LPM

19
–

18
USBAL

17
RTCAL

16
RTTAL

15
FSTT15

14
FSTT14

13
FSTT13

12
FSTT12

11
FSTT11

10
FSTT10

9
FSTT9

8
FSTT8

7
FSTT7

6
FSTT6

5
FSTT5

4
FSTT4

3
FSTT3

2
FSTT2

1
FSTT1

0
FSTT0

This register can only be written if the WPEN bit is cleared in the “PMC Write Protect Mode Register”.
• FSTT0 - FSTT15: Fast Startup Input Enable 0 to 15
0 = The corresponding wake up input has no effect on the Power Management Controller.
1 = The corresponding wake up input enables a fast restart signal to the Power Management Controller.
• RTTAL: RTT Alarm Enable
0 = The RTT alarm has no effect on the Power Management Controller.
1 = The RTT alarm enables a fast restart signal to the Power Management Controller.
• RTCAL: RTC Alarm Enable
0 = The RTC alarm has no effect on the Power Management Controller.
1 = The RTC alarm enables a fast restart signal to the Power Management Controller.
• USBAL: USB Alarm Enable
0 = The USB alarm has no effect on the Power Management Controller.
1 = The USB alarm enables a fast restart signal to the Power Management Controller.
• LPM: Low Power Mode
0 = The WaitForInterrupt (WFI) or WaitForEvent (WFE) instruction of the processor makes the processor enter in Idle
Mode.
1 = The WaitForEvent (WFE) instruction of the processor makes the system enter in Wait Mode.

496

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.18 PMC Fast Startup Polarity Register
Name:
PMC_FSPR
Address:

0x400E0474

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
FSTP15

14
FSTP14

13
FSTP13

12
FSTP12

11
FSTP11

10
FSTP10

9
FSTP9

8
FSTP8

7
FSTP7

6
FSTP6

5
FSTP5

4
FSTP4

3
FSTP3

2
FSTP2

1
FSTP1

0
FSTP0

This register can only be written if the WPEN bit is cleared in the “PMC Write Protect Mode Register”.
• FSTPx: Fast Startup Input Polarityx
Defines the active polarity of the corresponding wake up input. If the corresponding wake up input is enabled and at the
FSTP level, it enables a fast restart signal.

497
6430F–ATARM–21-Feb-12

28.14.19 PMC Fault Output Clear Register
Name:
PMC_FOCR
Address:

0x400E0478

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
–

0
FOCLR

• FOCLR: Fault Output Clear
Clears the clock failure detector fault output.

498

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
28.14.20 PMC Write Protect Mode Register
Name:
PMC_WPMR
Address:

0x400E04E4

Access:

Read-write

Reset:

See Table 28-3

WPKEY
23

20
WPKEY

12
WPKEY

–

WPEN

• WPEN: Write Protect Enable
0 = Disables the Write Protect if WPKEY corresponds to 0x504D43 (“PMC” in ASCII).
1 = Enables the Write Protect if WPKEY corresponds to 0x504D43 (“PMC” in ASCII).
Protects the registers:
• “PMC System Clock Enable Register” on page 479
• “PMC System Clock Disable Register” on page 480
• “PMC Peripheral Clock Enable Register” on page 481
• “PMC Peripheral Clock Disable Register” on page 482
• “PMC UTMI Clock Configuration Register” on page 484
• “PMC Clock Generator Main Oscillator Register” on page 485
• “PMC Clock Generator PLLA Register” on page 488
• “PMC Programmable Clock Register” on page 490
• “PMC Fast Startup Mode Register” on page 496
• “PMC Fast Startup Polarity Register” on page 497
• WPKEY: Write Protect KEY
Should be written at value 0x504D43 (“PMC” in ASCII). Writing any other value in this field aborts the write operation of the
WPEN bit. Always reads as 0.

499
6430F–ATARM–21-Feb-12

28.14.21 PMC Write Protect Status Register
Name:
PMC_WPSR
Address:

0x400E04E8

Access:

Read-only

Reset:

See Table 28-3

–

WPVSRC
15

12
WPVSRC

–

WPVS

• WPVS: Write Protect Violation Status
0 = No Write Protect Violation has occurred since the last read of the PMC_WPSR register.
1 = A Write Protect Violation has occurred since the last read of the PMC_WPSR register. If this violation is an unauthorized attempt to write a protected register, the associated violation is reported into field WPVSRC.
• WPVSRC: Write Protect Violation Source
When WPVS is active, this field indicates the write-protected register (through address offset or code) in which a write
access has been attempted.
Reading PMC_WPSR automatically clears all fields.

500

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
29. Chip Identifier (CHIPID)
29.1

Description
Chip Identifier registers permit recognition of the device and its revision. These registers provide
the sizes and types of the on-chip memories, as well as the set of embedded peripherals.
Two chip identifier registers are embedded: CHIPID_CIDR (Chip ID Register) and CHIPID_EXID
(Extension ID). Both registers contain a hard-wired value that is read-only. The first register contains the following fields:
• EXT - shows the use of the extension identifier register
• NVPTYP and NVPSIZ - identifies the type of embedded non-volatile memory and its size
• ARCH - identifies the set of embedded peripherals
• SRAMSIZ - indicates the size of the embedded SRAM
• EPROC - indicates the embedded ARM processor
• VERSION - gives the revision of the silicon
The second register is device-dependent and reads 0 if the bit EXT is 0.

29.2

Embedded Characteristics
• Chip ID Registers
– Identification of the Device Revision, Sizes of the Embedded Memories, Set of
Peripherals, Embedded Processor

Table 29-1.

ATSAM3U Chip IDs Register

Chip Name

CHIPID_CIDR

CHIPID_EXID

AT91SAM3U4C (Rev A)

0x28000961

0x0

AT91SAM3U2C (Rev A)

0x280A0761

0x0

AT91SAM3U1C (Rev A)

0x28090561

0x0

AT91SAM3U4E (Rev A)

0x28100961

0x0

AT91SAM3U2E (Rev A)

0x281A0761

0x0

AT91SAM3U1E (Rev A)

0x28190561

0x0

501
6430F–ATARM–21-Feb-12

29.3

Chip Identifier (CHIPID) User Interface

Table 29-2.
Offset

502

Name

Access

Reset

0x0

Chip ID Register

CHIPID_CIDR

Read-only

–

0x4

Chip ID Extension Register

CHIPID_EXID

Read-only

–

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
29.3.1
Name:

Chip ID Register
CHIPID_CIDR

Address:

0x400E0740

Access:

Read-only

EXT

NVPTYP

ARCH
15

SRAMSIZ
12

NVPSIZ2
7

25
ARCH

NVPSIZ
5

EPROC

2
VERSION

• VERSION: Version of the Device
Current version of the device.
• EPROC: Embedded Processor
Value

Name

Description

ARM946ES

ARM7TDMI

CM3

Cortex-M3

ARM920T

ARM926EJS

CA5

Cortex-A5

CM4

Cortex-M4

• NVPSIZ: Nonvolatile Program Memory Size
Value

Name

Description

NONE

None

8K bytes

16K

16K bytes

32K

32K bytes

4
5

Reserved
64K

6
7

64K bytes
Reserved

128K

128K bytes
Reserved

256K

256K bytes

512K

512K bytes

11
12

Reserved
1024K

1024K bytes

503
6430F–ATARM–21-Feb-12

Value

Name

13
14

Description
Reserved

2048K

2048K bytes
Reserved

• NVPSIZ2: Second Nonvolatile Program Memory Size
Value

Name

Description

NONE

None

8K bytes

16K

16K bytes

32K

32K bytes

4
5

Reserved
64K

6
7

64K bytes
Reserved

128K

128K bytes
Reserved

256K

256K bytes

512K

512K bytes

11
12

Reserved
1024K

13
14

1024K bytes
Reserved

2048K

2048K bytes
Reserved

• SRAMSIZ: Internal SRAM Size
Value

504

Name

Description

48K

48K bytes

1K bytes

2K bytes

6K bytes

24K

24K bytes

4K bytes

80K

80K bytes

160K

160K bytes

8K bytes

16K

16K bytes

32K

32K bytes

64K

64K bytes

128K

128K bytes

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Value

Name

Description

256K

256K bytes

96K

96K bytes

512K

512K bytes

• ARCH: Architecture Identifier
Value

Name

Description

0x19

AT91SAM9xx

AT91SAM9xx Series

0x29

AT91SAM9XExx

AT91SAM9XExx Series

0x34

AT91x34

AT91x34 Series

0x37

CAP7

CAP7 Series

0x39

CAP9

CAP9 Series

0x3B

CAP11

CAP11 Series

0x40

AT91x40

AT91x40 Series

0x42

AT91x42

AT91x42 Series

0x55

AT91x55

AT91x55 Series

0x60

AT91SAM7Axx

AT91SAM7Axx Series

0x61

AT91SAM7AQxx

AT91SAM7AQxx Series

0x63

AT91x63

AT91x63 Series

0x70

AT91SAM7Sxx

AT91SAM7Sxx Series

0x71

AT91SAM7XCxx

AT91SAM7XCxx Series

0x72

AT91SAM7SExx

AT91SAM7SExx Series

0x73

AT91SAM7Lxx

AT91SAM7Lxx Series

0x75

AT91SAM7Xxx

AT91SAM7Xxx Series

0x76

AT91SAM7SLxx

AT91SAM7SLxx Series

0x80

SAM3UxC

SAM3UxC Series (100-pin version)

0x81

SAM3UxE

SAM3UxE Series (144-pin version)

0x83

SAM3AxC

SAM3AxC Series (100-pin version)

0x83

SAM4AxC

SAM4AxC Series (100-pin version)

0x84

SAM3XxC

SAM3XxC Series (100-pin version)

0x84

SAM4XxC

SAM4XxC Series (100-pin version)

0x85

SAM3XxE

SAM3XxE Series (144-pin version)

0x85

SAM4XxE

SAM4XxE Series (144-pin version)

0x86

SAM3XxG

SAM3XxG Series (208/217-pin version)

0x86

SAM4XxG

SAM4XxG Series (208/217-pin version)

0x88

SAM3SxA

SAM3SxASeries (48-pin version)

0x88

SAM4SxA

SAM4SxA Series (48-pin version)

0x89

SAM3SxB

SAM3SxB Series (64-pin version)

505
6430F–ATARM–21-Feb-12

Value

Name

Description

0x89

SAM4SxB

SAM4SxB Series (64-pin version)

0x8A

SAM3SxC

SAM3SxC Series (100-pin version)

0x8A

SAM4SxC

SAM4SxC Series (100-pin version)

0x92

AT91x92

AT91x92 Series

0x93

SAM3NxA

SAM3NxA Series (48-pin version)

0x94

SAM3NxB

SAM3NxB Series (64-pin version)

0x95

SAM3NxC

SAM3NxC Series (100-pin version)

0x99

SAM3SDxB

SAM3SDxB Series (64-pin version)

0x9A

SAM3SDxC

SAM3SDxC Series (100-pin version)

0xA5

SAM5A

0xF0

AT75Cxx

AT75Cxx Series

• NVPTYP: Nonvolatile Program Memory Type
Value

Name

Description

ROM

ROMLESS

ROMless or on-chip Flash

SRAM

SRAM emulating ROM

FLASH

Embedded Flash Memory

ROM_FLASH

ROM and Embedded Flash Memory
NVPSIZ is ROM size
NVPSIZ2 is Flash size

• EXT: Extension Flag
0 = Chip ID has a single register definition without extension
1 = An extended Chip ID exists.

506

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
29.3.2
Name:

Chip ID Extension Register
CHIPID_EXID

Address:

0x400E0744

Access:

Read-only

EXID
23

20
EXID

12
EXID

4
EXID

• EXID: Chip ID Extension
Reads 0 if the bit EXT in CHIPID_CIDR is 0.

507
6430F–ATARM–21-Feb-12

508

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30. Parallel Input/Output Controller (PIO)
30.1

Description
The Parallel Input/Output Controller (PIO) manages up to 32 fully programmable input/output
lines. Each I/O line may be dedicated as a general-purpose I/O or be assigned to a function of
an embedded peripheral. This assures effective optimization of the pins of a product.
Each I/O line is associated with a bit number in all of the 32-bit registers of the 32-bit wide User
Interface.
Each I/O line of the PIO Controller features:
• An input change interrupt enabling level change detection on any I/O line.
• Additional Interrupt modes enabling rising edge, falling edge, low level or high level detection
on any I/O line.
• A glitch filter providing rejection of glitches lower than one-half of system clock cycle.
• A debouncing filter providing rejection of unwanted pulses from key or push button
operations.
• Multi-drive capability similar to an open drain I/O line.
• Control of the pull-up of the I/O line.
• Input visibility and output control.
The PIO Controller also features a synchronous output providing up to 32 bits of data output in a
single write operation.

30.2

Embedded Characteristics
• Up to 32 Programmable I/O Lines
• Fully Programmable through Set/Clear Registers
• Multiplexing of Two Peripheral Functions per I/O Line
• For each I/O Line (Whether Assigned to a Peripheral or Used as General Purpose I/O)
– Input Change Interrupt
– Programmable Glitch Filter
– Programmable Debouncing Filter
– Multi-drive Option Enables Driving in Open Drain
– Programmable Pull Up on Each I/O Line
– Pin Data Status Register, Supplies Visibility of the Level on the Pin at Any Time
– Additional Interrupt Modes on a Programmable Event: Rising Edge, Falling Edge,
Low Level or High Level
– Lock of the Configuration by the Connected Peripheral
• Synchronous Output, Provides Set and Clear of Several I/O lines in a Single Write

509
6430F–ATARM–21-Feb-12

30.3

Block Diagram

Figure 30-1. Block Diagram

PIO Controller
PIO Interrupt

NVIC

PIO Clock

PMC

Data, Enable

Up to 32
peripheral IOs

Embedded
Peripheral

PIN 0
Data, Enable
PIN 1
Up to 32 pins
Embedded
Peripheral

Up to 32
peripheral IOs
PIN 31

APB

Figure 30-2. Application Block Diagram
On-Chip Peripheral Drivers
Keyboard Driver

Control & Command
Driver
On-Chip Peripherals

PIO Controller
Keyboard Driver

510

General Purpose I/Os

External Devices

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.4

Product Dependencies

30.4.1

Pin Multiplexing
Each pin is configurable, according to product definition as either a general-purpose I/O line
only, or as an I/O line multiplexed with one or two peripheral I/Os. As the multiplexing is hardware defined and thus product-dependent, the hardware designer and programmer must
carefully determine the configuration of the PIO controllers required by their application. When
an I/O line is general-purpose only, i.e. not multiplexed with any peripheral I/O, programming of
the PIO Controller regarding the assignment to a peripheral has no effect and only the PIO Controller can control how the pin is driven by the product.

30.4.2

Power Management
The Power Management Controller controls the PIO Controller clock in order to save power.
Writing any of the registers of the user interface does not require the PIO Controller clock to be
enabled. This means that the configuration of the I/O lines does not require the PIO Controller
clock to be enabled.
However, when the clock is disabled, not all of the features of the PIO Controller are available,
including glitch filtering. Note that the Input Change Interrupt, Interrupt Modes on a programmable event and the read of the pin level require the clock to be validated.
After a hardware reset, the PIO clock is disabled by default.
The user must configure the Power Management Controller before any access to the input line
information.

30.4.3

Interrupt Generation
The PIO COntroller is connected on one of the sources of the Nested Vectored Interrupt Controller (NVIC). Using the PIO Controller requires the NVIC to be programmed first.
The PIO Controller interrupt can be generated only if the PIO Controller clock is enabled.

511
6430F–ATARM–21-Feb-12

30.5

Functional Description
The PIO Controller features up to 32 fully-programmable I/O lines. Most of the control logic associated to each I/O is represented in Figure 30-3. In this description each signal shown
represents but one of up to 32 possible indexes.

Figure 30-3. I/O Line Control Logic
PIO_OER[0]
PIO_OSR[0]

PIO_PUER[0]

PIO_ODR[0]

PIO_PUSR[0]
PIO_PUDR[0]

Peripheral A
Output Enable

Peripheral B
Output Enable

0
0

PIO_PER[0]
PIO_ABSR[0]

PIO_PSR[0]
PIO_PDR[0]

Peripheral A
Output

PIO_MDER[0]

PIO_MDSR[0]
0

Peripheral B
Output

PIO_MDDR[0]
0

PIO_SODR[0]
1

PIO_ODSR[0]

Pad

PIO_CODR[0]

Peripheral A
Input
Peripheral B
Input

PIO_PDSR[0]
PIO_ISR[0]
0
D

System Clock

Slow Clock

PIO_SCDR

Clock
Divider

Programmable
Glitch
or
Debouncing
Filter

Q
DFF

EVENT
DETECTOR

(Up to 32 possible inputs)
PIO Interrupt

1
Resynchronization
Stage

PIO_IER[0]
PIO_IMR[0]

PIO_IFER[0]

PIO_IDR[0]

PIO_IFSR[0]
PIO_DCIFSR[0]
PIO_IFSCR[0]
PIO_SCIFSR[0]

PIO_IFDR[0]

PIO_ISR[31]
PIO_IER[31]
PIO_IMR[31]
PIO_IDR[31]

512

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.5.1

Pull-up Resistor Control
Each I/O line is designed with an embedded pull-up resistor. The pull-up resistor can be enabled
or disabled by writing respectively PIO_PUER (Pull-up Enable Register) and PIO_PUDR (Pullup Disable Resistor). Writing in these registers results in setting or clearing the corresponding bit
in PIO_PUSR (Pull-up Status Register). Reading a 1 in PIO_PUSR means the pull-up is disabled and reading a 0 means the pull-up is enabled.
Control of the pull-up resistor is possible regardless of the configuration of the I/O line.
After reset, all of the pull-ups are enabled, i.e. PIO_PUSR resets at the value 0x0.

30.5.2

I/O Line or Peripheral Function Selection
When a pin is multiplexed with one or two peripheral functions, the selection is controlled with
the registers PIO_PER (PIO Enable Register) and PIO_PDR (PIO Disable Register). The register PIO_PSR (PIO Status Register) is the result of the set and clear registers and indicates
whether the pin is controlled by the corresponding peripheral or by the PIO Controller. A value of
0 indicates that the pin is controlled by the corresponding on-chip peripheral selected in the
PIO_ABSR (AB Select Register). A value of 1 indicates the pin is controlled by the PIO
controller.
If a pin is used as a general purpose I/O line (not multiplexed with an on-chip peripheral),
PIO_PER and PIO_PDR have no effect and PIO_PSR returns 1 for the corresponding bit.
After reset, most generally, the I/O lines are controlled by the PIO controller, i.e. PIO_PSR
resets at 1. However, in some events, it is important that PIO lines are controlled by the peripheral (as in the case of memory chip select lines that must be driven inactive after reset or for
address lines that must be driven low for booting out of an external memory). Thus, the reset
value of PIO_PSR is defined at the product level, depending on the multiplexing of the device.

30.5.3

Peripheral A or B Selection
The PIO Controller provides multiplexing of up to two peripheral functions on a single pin. The
selection is performed by writing PIO_ABSR (AB Select Register). For each pin, the corresponding bit at level 0 means peripheral A is selected whereas the corresponding bit at level 1
indicates that peripheral B is selected.
Note that multiplexing of peripheral lines A and B only affects the output line. The peripheral
input lines are always connected to the pin input.
After reset, PIO_ABSR is 0, thus indicating that all the PIO lines are configured on peripheral A.
However, peripheral A generally does not drive the pin as the PIO Controller resets in I/O line
mode.
Writing in PIO_ABSR manages the multiplexing regardless of the configuration of the pin. However, assignment of a pin to a peripheral function requires a write in the peripheral selection
register (PIO_ABSR) in addition to a write in PIO_PDR.

513
6430F–ATARM–21-Feb-12

30.5.4

Output Control
When the I/0 line is assigned to a peripheral function, i.e. the corresponding bit in PIO_PSR is at
0, the drive of the I/O line is controlled by the peripheral. Peripheral A or B depending on the
value in PIO_ABSR (AB Select Register) determines whether the pin is driven or not.
When the I/O line is controlled by the PIO controller, the pin can be configured to be driven. This
is done by writing PIO_OER (Output Enable Register) and PIO_ODR (Output Disable Register).
The results of these write operations are detected in PIO_OSR (Output Status Register). When
a bit in this register is at 0, the corresponding I/O line is used as an input only. When the bit is at
1, the corresponding I/O line is driven by the PIO controller.
The level driven on an I/O line can be determined by writing in PIO_SODR (Set Output Data
Register) and PIO_CODR (Clear Output Data Register). These write operations respectively set
and clear PIO_ODSR (Output Data Status Register), which represents the data driven on the I/O
lines. Writing in PIO_OER and PIO_ODR manages PIO_OSR whether the pin is configured to
be controlled by the PIO controller or assigned to a peripheral function. This enables configuration of the I/O line prior to setting it to be managed by the PIO Controller.
Similarly, writing in PIO_SODR and PIO_CODR effects PIO_ODSR. This is important as it
defines the first level driven on the I/O line.

30.5.5

Synchronous Data Output
Clearing one (or more) PIO line(s) and setting another one (or more) PIO line(s) synchronously
cannot be done by using PIO_SODR and PIO_CODR registers. It requires two successive write
operations into two different registers. To overcome this, the PIO Controller offers a direct control of PIO outputs by single write access to PIO_ODSR (Output Data Status Register).Only bits
unmasked by PIO_OWSR (Output Write Status Register) are written. The mask bits in
PIO_OWSR are set by writing to PIO_OWER (Output Write Enable Register) and cleared by
writing to PIO_OWDR (Output Write Disable Register).
After reset, the synchronous data output is disabled on all the I/O lines as PIO_OWSR resets at
0x0.

30.5.6

Multi Drive Control (Open Drain)
Each I/O can be independently programmed in Open Drain by using the Multi Drive feature. This
feature permits several drivers to be connected on the I/O line which is driven low only by each
device. An external pull-up resistor (or enabling of the internal one) is generally required to guarantee a high level on the line.
The Multi Drive feature is controlled by PIO_MDER (Multi-driver Enable Register) and
PIO_MDDR (Multi-driver Disable Register). The Multi Drive can be selected whether the I/O line
is controlled by the PIO controller or assigned to a peripheral function. PIO_MDSR (Multi-driver
Status Register) indicates the pins that are configured to support external drivers.
After reset, the Multi Drive feature is disabled on all pins, i.e. PIO_MDSR resets at value 0x0.

30.5.7

514

Output Line Timings
Figure 30-4 shows how the outputs are driven either by writing PIO_SODR or PIO_CODR, or by
directly writing PIO_ODSR. This last case is valid only if the corresponding bit in PIO_OWSR is
set. Figure 30-4 also shows when the feedback in PIO_PDSR is available.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 30-4. Output Line Timings
MCK

Write PIO_SODR
Write PIO_ODSR at 1

APB Access

Write PIO_CODR
Write PIO_ODSR at 0

APB Access

PIO_ODSR
2 cycles

2 cycles

PIO_PDSR

30.5.8

Inputs
The level on each I/O line can be read through PIO_PDSR (Pin Data Status Register). This register indicates the level of the I/O lines regardless of their configuration, whether uniquely as an
input or driven by the PIO controller or driven by a peripheral.
Reading the I/O line levels requires the clock of the PIO controller to be enabled, otherwise
PIO_PDSR reads the levels present on the I/O line at the time the clock was disabled.

30.5.9

Input Glitch and Debouncing Filters
Optional input glitch and debouncing filters are independently programmable on each I/O line.
The glitch filter can filter a glitch with a duration of less than 1/2 Master Clock (MCK) and the
debouncing filter can filter a pulse of less than 1/2 Period of a Programmable Divided Slow
Clock.
The selection between glitch filtering or debounce filtering is done by writing in the registers
PIO_SCIFSR (System Clock Glitch Input Filter Select Register) and PIO_DIFSR (Debouncing
Input Filter Select Register). Writing PIO_SCIFSR and PIO_DIFSR respectively, sets and clears
bits in PIO_IFDGSR.
The current selection status can be checked by reading the register PIO_IFDGSR (Glitch or
Debouncing Input Filter Selection Status Register).
• If PIO_IFDGSR[i] = 0: The glitch filter can filter a glitch with a duration of less than 1/2 Period
of Master Clock.
• If PIO_IFDGSR[i] = 1: The debouncing filter can filter a pulse with a duration of less than 1/2
Period of the Programmable Divided Slow Clock.
For the debouncing filter, the Period of the Divided Slow Clock is performed by writing in the DIV
field of the PIO_SCDR (Slow Clock Divider Register)
Tdiv_slclk = ( (DIV+1)*2).Tslow_clock
When the glitch or debouncing filter is enabled, a glitch or pulse with a duration of less than 1/2
Selected Clock Cycle (Selected Clock represents MCK or Divided Slow Clock depending on
PIO_SCIFSR and PIO_DIFSR programming) is automatically rejected, while a pulse with a
duration of 1 Selected Clock (MCK or Divided Slow Clock) cycle or more is accepted. For pulse
durations between 1/2 Selected Clock cycle and 1 Selected Clock cycle the pulse may or may
515

6430F–ATARM–21-Feb-12

not be taken into account, depending on the precise timing of its occurrence. Thus for a pulse to
be visible it must exceed 1 Selected Clock cycle, whereas for a glitch to be reliably filtered out,
its duration must not exceed 1/2 Selected Clock cycle.
The filters also introduce some latencies, this is illustrated in Figure 30-5 and Figure 30-6.
The glitch filters are controlled by the register set: PIO_IFER (Input Filter Enable Register),
PIO_IFDR (Input Filter Disable Register) and PIO_IFSR (Input Filter Status Register). Writing
PIO_IFER and PIO_IFDR respectively sets and clears bits in PIO_IFSR. This last register
enables the glitch filter on the I/O lines.
When the glitch and/or debouncing filter is enabled, it does not modify the behavior of the inputs
on the peripherals. It acts only on the value read in PIO_PDSR and on the input change interrupt
detection. The glitch and debouncing filters require that the PIO Controller clock is enabled.
Figure 30-5. Input Glitch Filter Timing
PIO_IFCSR = 0
MCK
up to 1.5 cycles
Pin Level
1 cycle

1 cycle

PIO_PDSR
if PIO_IFSR = 0
2 cycles

1 cycle

up to 2.5 cycles

PIO_PDSR
if PIO_IFSR = 1

up to 2 cycles

Figure 30-6. Input Debouncing Filter Timing
PIO_IFCSR = 1
Divided Slow Clock

Pin Level
up to 2 cycles Tmck

up to 2 cycles Tmck

PIO_PDSR
if PIO_IFSR = 0

1 cycle Tdiv_slclk
PIO_PDSR
if PIO_IFSR = 1

up to 1.5 cycles Tdiv_slclk
up to 1.5 cycles Tdiv_slclk
up to 2 cycles Tmck

30.5.10

516

1 cycle Tdiv_slclk

up to 2 cycles Tmck

Input Edge/Level Interrupt
The PIO Controller can be programmed to generate an interrupt when it detects an edge or a
level on an I/O line. The Input Edge/Level Interrupt is controlled by writing PIO_IER (Interrupt
Enable Register) and PIO_IDR (Interrupt Disable Register), which respectively enable and disable the input change interrupt by setting and clearing the corresponding bit in PIO_IMR
(Interrupt Mask Register). As Input change detection is possible only by comparing two succes-

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
sive samplings of the input of the I/O line, the PIO Controller clock must be enabled. The Input
Change Interrupt is available, regardless of the configuration of the I/O line, i.e. configured as an
input only, controlled by the PIO Controller or assigned to a peripheral function.
By default, the interrupt can be generated at any time an edge is detected on the input.
Some additional Interrupt modes can be enabled/disabled by writing in the PIO_AIMER (Additional Interrupt Modes Enable Register) and PIO_AIMDR (Additional Interrupt Modes Disable
Register). The current state of this selection can be read through the PIO_AIMMR (Additional
Interrupt Modes Mask Register)
These Additional Modes are:
• Rising Edge Detection
• Falling Edge Detection
• Low Level Detection
• High Level Detection
In order to select an Additional Interrupt Mode:
• The type of event detection (Edge or Level) must be selected by writing in the set of registers;
PIO_ESR (Edge Select Register) and PIO_LSR (Level Select Register) which enable
respectively, the Edge and Level Detection. The current status of this selection is accessible
through the PIO_ELSR (Edge/Level Status Register).
• The Polarity of the event detection (Rising/Falling Edge or High/Low Level) must be selected
by writing in the set of registers; PIO_FELLSR (Falling Edge /Low Level Select Register) and
PIO_REHLSR (Rising Edge/High Level Select Register) which allow to select Falling or
Rising Edge (if Edge is selected in the PIO_ELSR) Edge or High or Low Level Detection (if
Level is selected in the PIO_ELSR). The current status of this selection is accessible through
the PIO_FRLHSR (Fall/Rise - Low/High Status Register).
When an input Edge or Level is detected on an I/O line, the corresponding bit in PIO_ISR (Interrupt Status Register) is set. If the corresponding bit in PIO_IMR is set, the PIO Controller
interrupt line is asserted. The interrupt signals of the thirty-two channels are ORed-wired
together to generate a single interrupt signal to the Nested Vector Interrupt Controller (NVIC).
When the software reads PIO_ISR, all the interrupts are automatically cleared. This signifies that
all the interrupts that are pending when PIO_ISR is read must be handled. When an Interrupt is
enabled on a “Level”, the interrupt is generated as long as the interrupt source is not cleared,
even if some read accesses in PIO_ISR are performed.

517
6430F–ATARM–21-Feb-12

Figure 30-7. Event Detector on Input Lines (Figure represents line 0)
Event Detector
Rising Edge
Detector

Falling Edge
Detector

0
0

PIO_REHLSR[0]
1

PIO_FRLHSR[0]

Event detection on line 0

PIO_FELLSR[0]

Resynchronized input on line 0

High Level
Detector

Low Level
Detector

PIO_LSR[0]
PIO_ELSR[0]
PIO_ESR[0]

PIO_AIMER[0]
PIO_AIMMR[0]
PIO_AIMDR[0]

Edge
Detector

30.5.10.1

Example
If generating an interrupt is required on the following:
• Rising edge on PIO line 0
• Falling edge on PIO line 1
• Rising edge on PIO line 2
• Low Level on PIO line 3
• High Level on PIO line 4
• High Level on PIO line 5
• Falling edge on PIO line 6
• Rising edge on PIO line 7
• Any edge on the other lines
The configuration required is described below.

30.5.10.2

Interrupt Mode Configuration
All the interrupt sources are enabled by writing 32’hFFFF_FFFF in PIO_IER.
Then the Additional Interrupt Mode is enabled for line 0 to 7 by writing 32’h0000_00FF in
PIO_AIMER.

30.5.10.3

Edge or Level Detection Configuration
Lines 3, 4 and 5 are configured in Level detection by writing 32’h0000_0038 in PIO_LSR.
The other lines are configured in Edge detection by default, if they have not been previously configured. Otherwise, lines 0, 1, 2, 6 and 7 must be configured in Edge detection by writing
32’h0000_00C7 in PIO_ESR.

30.5.10.4

518

Falling/Rising Edge or Low/High Level Detection Configuration.
Lines 0, 2, 4, 5 and 7 are configured in Rising Edge or High Level detection by writing
32’h0000_00B5 in PIO_REHLSR.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The other lines are configured in Falling Edge or Low Level detection by default, if they have not
been previously configured. Otherwise, lines 1, 3 and 6 must be configured in Falling Edge/Low
Level detection by writing 32’h0000_004A in PIO_FELLSR.
Figure 30-8. Input Change Interrupt Timings if there are no Additional Interrupt Modes
MCK

Pin Level

PIO_ISR

Read PIO_ISR

30.5.11

APB Access

I/O Lines Lock
When an I/O line is controlled by a peripheral (particularly the Pulse Width Modulation Controller
PWM), it can become locked by the action of this peripheral via an input of the PIO controller.
When an I/O line is locked, the write of the corresponding bit in the registers PIO_PER,
PIO_PDR, PIO_MDER, PIO_MDDR, PIO_PUDR, PIO_PUER and PIO_ABSR is discarded in
order to lock its configuration. The user can know at anytime which I/O line is locked by reading
the PIO Lock Status register PIO_LOCKSR. Once an I/O line is locked, the only way to unlock it
is to apply an hardware reset to the PIO Controller.

519
6430F–ATARM–21-Feb-12

30.6

I/O Lines Programming Example
The programing example as shown in Table 30-1 below is used to obtain the following
configuration.
• 4-bit output port on I/O lines 0 to 3, (should be written in a single write operation), open-drain,
with pull-up resistor
• Four output signals on I/O lines 4 to 7 (to drive LEDs for example), driven high and low, no
pull-up resistor
• Four input signals on I/O lines 8 to 11 (to read push-button states for example), with pull-up
resistors, glitch filters and input change interrupts
• Four input signals on I/O line 12 to 15 to read an external device status (polled, thus no input
change interrupt), no pull-up resistor, no glitch filter
• I/O lines 16 to 19 assigned to peripheral A functions with pull-up resistor
• I/O lines 20 to 23 assigned to peripheral B functions, no pull-up resistor
• I/O line 24 to 27 assigned to peripheral A with Input Change Interrupt and pull-up resistor
Table 30-1.

520

Programming Example
Register

Value to be Written

PIO_PER

0x0000 FFFF

PIO_PDR

0xFFFF 0000

PIO_OER

0x0000 00FF

PIO_ODR

0xFFFF FF00

PIO_IFER

0x0000 0F00

PIO_IFDR

0xFFFF F0FF

PIO_SODR

0x0000 0000

PIO_CODR

0x0FFF FFFF

PIO_IER

0x0F00 0F00

PIO_IDR

0xF0FF F0FF

PIO_MDER

0x0000 000F

PIO_MDDR

0xFFFF FFF0

PIO_PUDR

0xF0F0 00F0

PIO_PUER

0x0F0F FF0F

PIO_ABSR

0x00F0 0000

PIO_OWER

0x0000 000F

PIO_OWDR

0x0FFF FFF0

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.6.1

Write Protection Registers
To prevent any single software error that may corrupt PIO behavior, certain address spaces can
be write-protected by setting the WPEN bit in the “PIO Write Protect Mode Register”
(PIO_WPMR).
If a write access to the protected registers is detected, then the WPVS flag in the PIO Write Protect Status Register (PIO_WPSR) is set and the field WPVSRC indicates in which register the
write access has been attempted.
The WPVS flag is reset by writing the PIO Write Protect Mode Register (PIO_WPMR) with the
appropriate access key, WPKEY.
The protected registers are:
• “PIO Controller PIO Enable Register” on page 524
• “PIO Controller PIO Disable Register” on page 524
• “PIO Controller Output Enable Register” on page 525
• “PIO Controller Output Disable Register” on page 526
• “PIO Controller Input Filter Enable Register” on page 527
• “PIO Controller Input Filter Disable Register” on page 527
• “PIO Multi-driver Enable Register” on page 532
• “PIO Multi-driver Disable Register” on page 533
• “PIO Pull Up Disable Register” on page 534
• “PIO Pull Up Enable Register” on page 534
• “PIO Peripheral AB Select Register” on page 535
• “PIO Output Write Enable Register” on page 538
• “PIO Output Write Disable Register” on page 538

521
6430F–ATARM–21-Feb-12

30.7

Parallel Input/Output Controller (PIO) User Interface
Each I/O line controlled by the PIO Controller is associated with a bit in each of the PIO Controller
User Interface registers. Each register is 32 bits wide. If a parallel I/O line is not defined, writing to the
corresponding bits has no effect. Undefined bits read zero. If the I/O line is not multiplexed with any
peripheral, the I/O line is controlled by the PIO Controller and PIO_PSR returns 1 systematically.

Table 30-2.

Offset

Name

Access

Reset

0x0000

PIO Enable Register

PIO_PER

Write-only

–

0x0004

PIO Disable Register

PIO_PDR

Write-only

–

0x0008

PIO Status Register

PIO_PSR

Read-only

(1)

0x000C

Reserved

0x0010

Output Enable Register

PIO_OER

Write-only

–

0x0014

Output Disable Register

PIO_ODR

Write-only

–

0x0018

Output Status Register

PIO_OSR

Read-only

0x0000 0000

0x001C

Reserved

0x0020

Glitch Input Filter Enable Register

PIO_IFER

Write-only

–

0x0024

Glitch Input Filter Disable Register

PIO_IFDR

Write-only

–

0x0028

Glitch Input Filter Status Register

PIO_IFSR

Read-only

0x0000 0000

0x002C

Reserved

0x0030

Set Output Data Register

PIO_SODR

Write-only

–

0x0034

Clear Output Data Register

PIO_CODR

Write-only

0x0038

Output Data Status Register

PIO_ODSR

Read-only
or(2)
Read-write

–

0x003C

Pin Data Status Register

PIO_PDSR

Read-only

(3)

0x0040

Interrupt Enable Register

PIO_IER

Write-only

–

0x0044

Interrupt Disable Register

PIO_IDR

Write-only

–

0x0048

Interrupt Mask Register

PIO_IMR

Read-only

0x00000000

PIO_ISR

Read-only

0x00000000

(4)

0x004C

Interrupt Status Register

0x0050

Multi-driver Enable Register

PIO_MDER

Write-only

–

0x0054

Multi-driver Disable Register

PIO_MDDR

Write-only

–

0x0058

Multi-driver Status Register

PIO_MDSR

Read-only

0x00000000

0x005C

Reserved

0x0060

Pull-up Disable Register

PIO_PUDR

Write-only

–

0x0064

Pull-up Enable Register

PIO_PUER

Write-only

–

0x0068

Pad Pull-up Status Register

PIO_PUSR

Read-only

0x00000000

0x006C

Reserved

522

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 30-2.
Offset

Name
(5)

Access

Reset

PIO_ABSR

Read-Write

0x00000000

0x0070

Peripheral AB Select Register

0x0074
to
0x007C

Reserved

0x0080

System Clock Glitch Input Filter Select Register

PIO_SCIFSR

Write-Only

–

0x0084

Debouncing Input Filter Select Register

PIO_DIFSR

Write-Only

–

0x0088

Glitch or Debouncing Input Filter Clock Selection Status Register

PIO_IFDGSR

Read-Only

0x00000000

0x008C

Slow Clock Divider Debouncing Register

PIO_SCDR

Read-Write

0x00000000

0x0090
to
0x009C

Reserved

0x00A0

Output Write Enable

PIO_OWER

Write-only

–

0x00A4

Output Write Disable

PIO_OWDR

Write-only

–

0x00A8

Output Write Status Register

PIO_OWSR

Read-only

0x00000000

0x00AC

Reserved

0x00B0

Additional Interrupt Modes Enable Register

PIO_AIMER

Write-Only

–

0x00B4

Additional Interrupt Modes Disables Register

PIO_AIMDR

Write-Only

–

0x00B8

Additional Interrupt Modes Mask Register

PIO_AIMMR

Read-Only

0x00000000

0x00BC

Reserved

0x00C0

Edge Select Register

PIO_ESR

Write-Only

–

0x00C4

Level Select Register

PIO_LSR

Write-Only

–

0x00C8

Edge/Level Status Register

PIO_ELSR

Read-Only

0x00000000

0x00CC

Reserved

0x00D0

Falling Edge/Low Level Select Register

PIO_FELLSR

Write-Only

–

0x00D4

Rising Edge/ High Level Select Register

PIO_REHLSR

Write-Only

–

0x00D8

Fall/Rise - Low/High Status Register

PIO_FRLHSR

Read-Only

0x00000000

0x00DC

Reserved

0x00E0

Lock Status

PIO_LOCKSR

Read-Only

0x00000000

0x00E4

Write Protect Mode Register

PIO_WPMR

Read-write

0x0

0x00E8

Write Protect Status Register

PIO_WPSR

Read-only

0x0

0x00EC
to
0x00F8

Reserved

0x0100
to
0x0144

Reserved

Notes:

1. Reset value of PIO_PSR depends on the product implementation.
2. PIO_ODSR is Read-only or Read/Write depending on PIO_OWSR I/O lines.
3. Reset value of PIO_PDSR depends on the level of the I/O lines. Reading the I/O line levels requires the clock of the PIO
Controller to be enabled, otherwise PIO_PDSR reads the levels present on the I/O line at the time the clock was disabled.
4. PIO_ISR is reset at 0x0. However, the first read of the register may read a different value as input changes may have occurred.
5. Only this set of registers clears the status by writing 1 in the first register and sets the status by writing 1 in the second register.

523
6430F–ATARM–21-Feb-12

30.7.1
Name:

PIO Controller PIO Enable Register
PIO_PER

Address:

0x400E0C00 (PIOA), 0x400E0E00 (PIOB), 0x400E1000 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: PIO Enable
0 = No effect.
1 = Enables the PIO to control the corresponding pin (disables peripheral control of the pin).

30.7.2
Name:

PIO Controller PIO Disable Register
PIO_PDR

Address:

0x400E0C04 (PIOA), 0x400E0E04 (PIOB), 0x400E1004 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: PIO Disable
0 = No effect.
1 = Disables the PIO from controlling the corresponding pin (enables peripheral control of the pin).

524

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.3
Name:

PIO Controller PIO Status Register
PIO_PSR

Address:

0x400E0C08 (PIOA), 0x400E0E08 (PIOB), 0x400E1008 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: PIO Status
0 = PIO is inactive on the corresponding I/O line (peripheral is active).
1 = PIO is active on the corresponding I/O line (peripheral is inactive).

30.7.4
Name:

PIO Controller Output Enable Register
PIO_OER

Address:

0x400E0C10 (PIOA), 0x400E0E10 (PIOB), 0x400E1010 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Output Enable
0 = No effect.
1 = Enables the output on the I/O line.

525
6430F–ATARM–21-Feb-12

30.7.5
Name:

PIO Controller Output Disable Register
PIO_ODR

Address:

0x400E0C14 (PIOA), 0x400E0E14 (PIOB), 0x400E1014 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Output Disable
0 = No effect.
1 = Disables the output on the I/O line.

30.7.6
Name:

PIO Controller Output Status Register
PIO_OSR

Address:

0x400E0C18 (PIOA), 0x400E0E18 (PIOB), 0x400E1018 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Output Status
0 = The I/O line is a pure input.
1 = The I/O line is enabled in output.

526

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.7
Name:

PIO Controller Input Filter Enable Register
PIO_IFER

Address:

0x400E0C20 (PIOA), 0x400E0E20 (PIOB), 0x400E1020 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Input Filter Enable
0 = No effect.
1 = Enables the input glitch filter on the I/O line.

30.7.8
Name:

PIO Controller Input Filter Disable Register
PIO_IFDR

Address:

0x400E0C24 (PIOA), 0x400E0E24 (PIOB), 0x400E1024 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Input Filter Disable
0 = No effect.
1 = Disables the input glitch filter on the I/O line.

527
6430F–ATARM–21-Feb-12

30.7.9
Name:

PIO Controller Input Filter Status Register
PIO_IFSR

Address:

0x400E0C28 (PIOA), 0x400E0E28 (PIOB), 0x400E1028 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Input Filer Status
0 = The input glitch filter is disabled on the I/O line.
1 = The input glitch filter is enabled on the I/O line.

30.7.10
Name:

PIO Controller Set Output Data Register
PIO_SODR

Address:

0x400E0C30 (PIOA), 0x400E0E30 (PIOB), 0x400E1030 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Set Output Data
0 = No effect.
1 = Sets the data to be driven on the I/O line.

528

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.11
Name:

PIO Controller Clear Output Data Register
PIO_CODR

Address:

0x400E0C34 (PIOA), 0x400E0E34 (PIOB), 0x400E1034 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Clear Output Data
0 = No effect.
1 = Clears the data to be driven on the I/O line.

30.7.12
Name:

PIO Controller Output Data Status Register
PIO_ODSR

Address:

0x400E0C38 (PIOA), 0x400E0E38 (PIOB), 0x400E1038 (PIOC)

Access:

Read-only or Read/Write

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Output Data Status
0 = The data to be driven on the I/O line is 0.
1 = The data to be driven on the I/O line is 1.

529
6430F–ATARM–21-Feb-12

30.7.13
Name:

PIO Controller Pin Data Status Register
PIO_PDSR

Address:

0x400E0C3C (PIOA), 0x400E0E3C (PIOB), 0x400E103C (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Output Data Status
0 = The I/O line is at level 0.
1 = The I/O line is at level 1.

30.7.14
Name:

PIO Controller Interrupt Enable Register
PIO_IER

Address:

0x400E0C40 (PIOA), 0x400E0E40 (PIOB), 0x400E1040 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Input Change Interrupt Enable
0 = No effect.
1 = Enables the Input Change Interrupt on the I/O line.

530

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.15
Name:

PIO Controller Interrupt Disable Register
PIO_IDR

Address:

0x400E0C44 (PIOA), 0x400E0E44 (PIOB), 0x400E1044 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Input Change Interrupt Disable
0 = No effect.
1 = Disables the Input Change Interrupt on the I/O line.

30.7.16
Name:

PIO Controller Interrupt Mask Register
PIO_IMR

Address:

0x400E0C48 (PIOA), 0x400E0E48 (PIOB), 0x400E1048 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Input Change Interrupt Mask
0 = Input Change Interrupt is disabled on the I/O line.
1 = Input Change Interrupt is enabled on the I/O line.

531
6430F–ATARM–21-Feb-12

30.7.17
Name:

PIO Controller Interrupt Status Register
PIO_ISR

Address:

0x400E0C4C (PIOA), 0x400E0E4C (PIOB), 0x400E104C (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Input Change Interrupt Status
0 = No Input Change has been detected on the I/O line since PIO_ISR was last read or since reset.
1 = At least one Input Change has been detected on the I/O line since PIO_ISR was last read or since reset.

30.7.18
Name:

PIO Multi-driver Enable Register
PIO_MDER

Address:

0x400E0C50 (PIOA), 0x400E0E50 (PIOB), 0x400E1050 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Multi Drive Enable.
0 = No effect.
1 = Enables Multi Drive on the I/O line.

532

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.19
Name:

PIO Multi-driver Disable Register
PIO_MDDR

Address:

0x400E0C54 (PIOA), 0x400E0E54 (PIOB), 0x400E1054 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Multi Drive Disable.
0 = No effect.
1 = Disables Multi Drive on the I/O line.

30.7.20
Name:

PIO Multi-driver Status Register
PIO_MDSR

Address:

0x400E0C58 (PIOA), 0x400E0E58 (PIOB), 0x400E1058 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Multi Drive Status.
0 = The Multi Drive is disabled on the I/O line. The pin is driven at high and low level.
1 = The Multi Drive is enabled on the I/O line. The pin is driven at low level only.

533
6430F–ATARM–21-Feb-12

30.7.21
Name:

PIO Pull Up Disable Register
PIO_PUDR

Address:

0x400E0C60 (PIOA), 0x400E0E60 (PIOB), 0x400E1060 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Pull Up Disable.
0 = No effect.
1 = Disables the pull up resistor on the I/O line.

30.7.22
Name:

PIO Pull Up Enable Register
PIO_PUER

Address:

0x400E0C64 (PIOA), 0x400E0E64 (PIOB), 0x400E1064 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Pull Up Enable.
0 = No effect.
1 = Enables the pull up resistor on the I/O line.

534

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.23
Name:

PIO Pull Up Status Register
PIO_PUSR

Address:

0x400E0C68 (PIOA), 0x400E0E68 (PIOB), 0x400E1068 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Pull Up Status.
0 = Pull Up resistor is enabled on the I/O line.
1 = Pull Up resistor is disabled on the I/O line.

30.7.24
Name:

PIO Peripheral AB Select Register
PIO_ABSR

Address:

0x400E0C70 (PIOA), 0x400E0E70 (PIOB), 0x400E1070 (PIOC)

Access:

Read-Write

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Peripheral A Select.
0 = Assigns the I/O line to the Peripheral A function.
1 = Assigns the I/O line to the Peripheral B function.

535
6430F–ATARM–21-Feb-12

30.7.25
Name:

PIO System Clock Glitch Input Filtering Select Register
PIO_SCIFSR

Address:

0x400E0C80 (PIOA), 0x400E0E80 (PIOB), 0x400E1080 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: System Clock Glitch Filtering Select.
0 = No Effect.
1 = The Glitch Filter is able to filter glitches with a duration < Tmck/2.

30.7.26
Name:

PIO Debouncing Input Filtering Select Register
PIO_DIFSR

Address:

0x400E0C84 (PIOA), 0x400E0E84 (PIOB), 0x400E1084 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Debouncing Filtering Select.
0 = No Effect.
1 = The Debouncing Filter is able to filter pulses with a duration < Tdiv_slclk/2.

536

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.27
Name:

PIO Glitch or Debouncing Input Filter Selection Status Register
PIO_IFDGSR

Address:

0x400E0C88 (PIOA), 0x400E0E88 (PIOB), 0x400E1088 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Glitch or Debouncing Filter Selection Status
0 = The Glitch Filter is able to filter glitches with a duration < Tmck2.
1 = The Debouncing Filter is able to filter pulses with a duration < Tdiv_slclk/2.

30.7.28
Name:

PIO Slow Clock Divider Debouncing Register
PIO_SCDR

Address:

0x400E0C8C (PIOA), 0x400E0E8C (PIOB), 0x400E108C (PIOC)

Access:

Read-Write

–

DIV
5

DIV

• DIV: Slow Clock Divider Selection for Debouncing
Tdiv_slclk = 2*(DIV+1)*Tslow_clock.

537
6430F–ATARM–21-Feb-12

30.7.29
Name:

PIO Output Write Enable Register
PIO_OWER

Address:

0x400E0CA0 (PIOA), 0x400E0EA0 (PIOB), 0x400E10A0 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Output Write Enable.
0 = No effect.
1 = Enables writing PIO_ODSR for the I/O line.

30.7.30
Name:

PIO Output Write Disable Register
PIO_OWDR

Address:

0x400E0CA4 (PIOA), 0x400E0EA4 (PIOB), 0x400E10A4 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

This register can only be written if the WPEN bit is cleared in “PIO Write Protect Mode Register”.
• P0-P31: Output Write Disable.
0 = No effect.
1 = Disables writing PIO_ODSR for the I/O line.

538

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.31
Name:

PIO Output Write Status Register
PIO_OWSR

Address:

0x400E0CA8 (PIOA), 0x400E0EA8 (PIOB), 0x400E10A8 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Output Write Status.
0 = Writing PIO_ODSR does not affect the I/O line.
1 = Writing PIO_ODSR affects the I/O line.

30.7.32
Name:

Additional Interrupt Modes Enable Register
PIO_AIMER

Address:

0x400E0CB0 (PIOA), 0x400E0EB0 (PIOB), 0x400E10B0 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Additional Interrupt Modes Enable.
0 = No effect.
1 = The interrupt source is the event described in PIO_ELSR and PIO_FRLHSR.

539
6430F–ATARM–21-Feb-12

30.7.33
Name:

Additional Interrupt Modes Disable Register
PIO_AIMDR

Address:

0x400E0CB4 (PIOA), 0x400E0EB4 (PIOB), 0x400E10B4 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Additional Interrupt Modes Disable.
0 = No effect.
1 = The interrupt mode is set to the default interrupt mode (Both Edge detection).

30.7.34
Name:

Additional Interrupt Modes Mask Register
PIO_AIMMR

Address:

0x400E0CB8 (PIOA), 0x400E0EB8 (PIOB), 0x400E10B8 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Peripheral CD Status.
0 = The interrupt source is a Both Edge detection event
1 = The interrupt source is described by the registers PIO_ELSR and PIO_FRLHSR

540

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.35
Name:

Edge Select Register
PIO_ESR

Address:

0x400E0CC0 (PIOA), 0x400E0EC0 (PIOB), 0x400E10C0 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Edge Interrupt Selection.
0 = No effect.
1 = The interrupt source is an Edge detection event.

30.7.36
Name:

Level Select Register
PIO_LSR

Address:

0x400E0CC4 (PIOA), 0x400E0EC4 (PIOB), 0x400E10C4 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Level Interrupt Selection.
0 = No effect.
1 = The interrupt source is a Level detection event.

541
6430F–ATARM–21-Feb-12

30.7.37
Name:

Edge/Level Status Register
PIO_ELSR

Address:

0x400E0CC8 (PIOA), 0x400E0EC8 (PIOB), 0x400E10C8 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Edge/Level Interrupt source selection.
0 = The interrupt source is an Edge detection event.
1 = The interrupt source is a Level detection event.

30.7.38
Name:

Falling Edge/Low Level Select Register
PIO_FELLSR

Address:

0x400E0CD0 (PIOA), 0x400E0ED0 (PIOB), 0x400E10D0 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Falling Edge/Low Level Interrupt Selection.
0 = No effect.
1 = The interrupt source is set to a Falling Edge detection or Low Level detection event, depending on PIO_ELSR.

542

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.39
Name:

Rising Edge/High Level Select Register
PIO_REHLSR

Address:

0x400E0CD4 (PIOA), 0x400E0ED4 (PIOB), 0x400E10D4 (PIOC)

Access:

Write-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Rising Edge /High Level Interrupt Selection.
0 = No effect.
1 = The interrupt source is set to a Rising Edge detection or High Level detection event, depending on PIO_ELSR.

30.7.40
Name:

Fall/Rise - Low/High Status Register
PIO_FRLHSR

Address:

0x400E0CD8 (PIOA), 0x400E0ED8 (PIOB), 0x400E10D8 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Edge /Level Interrupt Source Selection.
0 = The interrupt source is a Falling Edge detection (if PIO_ELSR = 0) or Low Level detection event (if PIO_ELSR = 1).
1 = The interrupt source is a Rising Edge detection (if PIO_ELSR = 0) or High Level detection event (if PIO_ELSR = 1).

543
6430F–ATARM–21-Feb-12

30.7.41
Name:

Lock Status Register
PIO_LOCKSR

Address:

0x400E0CE0 (PIOA), 0x400E0EE0 (PIOB), 0x400E10E0 (PIOC)

Access:

Read-only

P31

P30

P29

P28

P27

P26

P25

P24

P23

P22

P21

P20

P19

P18

P17

P16

P15

P14

P13

P12

P11

P10

• P0-P31: Lock Status.
0 = The I/O line is not locked.
1 = The I/O line is locked.

544

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7.42
Name:

PIO Write Protect Mode Register
PIO_WPMR

Address:

0x400E0CE4 (PIOA), 0x400E0EE4 (PIOB), 0x400E10E4 (PIOC)

Access:

Read-write

Reset:

See Table 30-2

WPKEY
23

20
WPKEY

12
WPKEY

—

WPEN

For more information on Write Protection Registers, refer to Section 30.6.1 ”Write Protection Registers”.
• WPEN: Write Protect Enable
0 = Disables the Write Protect if WPKEY corresponds to 0x50494F (“PIO” in ASCII).
1 = Enables the Write Protect if WPKEY corresponds to 0x50494F (“PIO” in ASCII).
Protects the registers:
• “PIO Controller PIO Enable Register” on page 524
• “PIO Controller PIO Disable Register” on page 524
• “PIO Controller Output Enable Register” on page 525
• “PIO Controller Output Disable Register” on page 526
• “PIO Controller Input Filter Enable Register” on page 527
• “PIO Controller Input Filter Disable Register” on page 527
• “PIO Multi-driver Enable Register” on page 532
• “PIO Multi-driver Disable Register” on page 533
• “PIO Pull Up Disable Register” on page 534
• “PIO Pull Up Enable Register” on page 534
• “PIO Peripheral AB Select Register” on page 535
• “PIO Output Write Enable Register” on page 538
• “PIO Output Write Disable Register” on page 538
• WPKEY: Write Protect KEY
Should be written at value 0x50494F (“PIO” in ASCII). Writing any other value in this field aborts the write operation of the
WPEN bit. Always reads as 0.

545
6430F–ATARM–21-Feb-12

30.7.43
Name:

PIO Write Protect Status Register
PIO_WPSR

Address:

0x400E0CE8 (PIOA), 0x400E0EE8 (PIOB), 0x400E10E8 (PIOC)

Access:

Read-only

Reset:

See Table 30-2

—

WPVSRC
15

12
WPVSRC

—

WPVS

• WPVS: Write Protect Violation Status
0 = No Write Protect Violation has occurred since the last read of the PIO_WPSR register.
1 = A Write Protect Violation has occurred since the last read of the PIO_WPSR register. If this violation is an unauthorized
attempt to write a protected register, the associated violation is reported into field WPVSRC.
• WPVSRC: Write Protect Violation Source
When WPVS is active, this field indicates the write-protected register (through address offset or code) in which a write
access has been attempted.
Note:

546

Reading PIO_WPSR automatically clears all fields.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
31. Synchronous Serial Controller (SSC)
31.1

Description
The Atmel Synchronous Serial Controller (SSC) provides a synchronous communication link
with external devices. It supports many serial synchronous communication protocols generally
used in audio and telecom applications such as I2S, Short Frame Sync, Long Frame Sync, etc.
The SSC contains an independent receiver and transmitter and a common clock divider. The
receiver and the transmitter each interface with three signals: the TD/RD signal for data, the
TK/RK signal for the clock and the TF/RF signal for the Frame Sync. The transfers can be programmed to start automatically or on different events detected on the Frame Sync signal.
The SSC’s high-level of programmability and its use of DMA permit a continuous high bit rate
data transfer without processor intervention.
Featuring connection to the DMA, the SSC permits interfacing with low processor overhead to
the following:
• CODEC’s in master or slave mode
• DAC through dedicated serial interface, particularly I2S
• Magnetic card reader

31.2

Embedded Characteristics
• Provides Serial Synchronous Communication Links Used in Audio and Telecom Applications
• Contains an Independent Receiver and Transmitter and a Common Clock Divider
• Interfaced with the DMA Controller (DMAC) to Reduce Processor Overhead
• Offers a Configurable Frame Sync and Data Length
• Receiver and Transmitter Can be Programmed to Start Automatically or on Detection of
Different Events on the Frame Sync Signal
• Receiver and Transmitter Include a Data Signal, a Clock Signal and a Frame Synchronization
Signal

547
6430F–ATARM–21-Feb-12

31.3

Block Diagram

Figure 31-1. Block Diagram
System
Bus

APB Bridge

DMA
Peripheral
Bus
TF
TK

PMC

MCK
PIO

SSC Interface

RF
RK
Interrupt Control
RD

SSC Interrupt

31.4

Application Block Diagram

Figure 31-2. Application Block Diagram
OS or RTOS Driver

Power
Management

Interrupt
Management

Test
Management

SSC

Serial AUDIO

548

Codec

Time Slot
Management

Frame
Management

Line Interface

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
31.5

Pin Name List

Table 31-1.

I/O Lines Description

Pin Name

Pin Description

Receiver Frame Synchro

Input/Output

Receiver Clock

Input/Output

Receiver Data

Input

Transmitter Frame Synchro

Input/Output

Transmitter Clock

Input/Output

Transmitter Data

Output

31.6
31.6.1

Type

Product Dependencies
I/O Lines
The pins used for interfacing the compliant external devices may be multiplexed with PIO lines.
Before using the SSC receiver, the PIO controller must be configured to dedicate the SSC
receiver I/O lines to the SSC peripheral mode.
Before using the SSC transmitter, the PIO controller must be configured to dedicate the SSC
transmitter I/O lines to the SSC peripheral mode.
Table 31-2.

I/O Lines

Instance

Signal

I/O Line

Peripheral

SSC

PA27

SSC

PA31

SSC

PA29

SSC

PA26

SSC

PA30

SSC

PA28

31.6.2

Power Management
The SSC is not continuously clocked. The SSC interface may be clocked through the Power
Management Controller (PMC), therefore the programmer must first configure the PMC to
enable the SSC clock.

31.6.3

Interrupt
The SSC interface has an interrupt line connected to the Nested Vector Interrupt Controller
(NVIC). Handling interrupts requires programming the NVIC before configuring the SSC.

549
6430F–ATARM–21-Feb-12

All SSC interrupts can be enabled/disabled configuring the SSC Interrupt mask register. Each
Table 31-3.

Peripheral IDs

Instance

SSC

pending and unmasked SSC interrupt will assert the SSC interrupt line. The SSC interrupt service routine can get the interrupt origin by reading the SSC interrupt status register.

550

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
31.7

Functional Description
This chapter contains the functional description of the following: SSC Functional Block, Clock
Management, Data format, Start, Transmitter, Receiver and Frame Sync.
The receiver and transmitter operate separately. However, they can work synchronously by programming the receiver to use the transmit clock and/or to start a data transfer when transmission
starts. Alternatively, this can be done by programming the transmitter to use the receive clock
and/or to start a data transfer when reception starts. The transmitter and the receiver can be programmed to operate with the clock signals provided on either the TK or RK pins. This allows the
SSC to support many slave-mode data transfers. The maximum clock speed allowed on the TK
and RK pins is the master clock divided by 2.

Figure 31-3. SSC Functional Block Diagram
Transmitter

MCK

TK Input

Clock
Divider

Transmit Clock
Controller

RX clock

Clock Output
Controller

Frame Sync
Controller

Data
Controller

Clock Output
Controller

Frame Sync
Controller

Data
Controller

TX clock

TXEN
RX Start Start
Selector
TF

TX Start
Transmit Shift Register

Transmit Holding
Register

APB

Transmit Sync
Holding Register

User
Interface
Receiver

RK Input

Receive Clock RX Clock
Controller

TX Clock
RXEN
TX Start
Start
RF
Selector
RC0R

Interrupt Control

RX Start
Receive Shift Register

Receive Holding
Register

Receive Sync
Holding Register

NVIC

31.7.1

Clock Management
The transmitter clock can be generated by:
• an external clock received on the TK I/O pad
• the receiver clock
• the internal clock divider
551

6430F–ATARM–21-Feb-12

The receiver clock can be generated by:
• an external clock received on the RK I/O pad
• the transmitter clock
• the internal clock divider
Furthermore, the transmitter block can generate an external clock on the TK I/O pad, and the
receiver block can generate an external clock on the RK I/O pad.
This allows the SSC to support many Master and Slave Mode data transfers.
31.7.1.1

Clock Divider
Figure 31-4. Divided Clock Block Diagram
Clock Divider
SSC_CMR
MCK

12-bit Counter

Divided Clock

The Master Clock divider is determined by the 12-bit field DIV counter and comparator (so its
maximal value is 4095) in the Clock Mode Register SSC_CMR, allowing a Master Clock division
by up to 8190. The Divided Clock is provided to both the Receiver and Transmitter. When this
field is programmed to 0, the Clock Divider is not used and remains inactive.
When DIV is set to a value equal to or greater than 1, the Divided Clock has a frequency of Master Clock divided by 2 times DIV. Each level of the Divided Clock has a duration of the Master
Clock multiplied by DIV. This ensures a 50% duty cycle for the Divided Clock regardless of
whether the DIV value is even or odd.
Figure 31-5.

Divided Clock Generation
Master Clock

Divided Clock
DIV = 1
Divided Clock Frequency = MCK/2

Master Clock

Divided Clock
DIV = 3
Divided Clock Frequency = MCK/6

552

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series

Table 31-4.

31.7.1.2

Maximum

Minimum

MCK / 2

MCK / 8190

Transmitter Clock Management
The transmitter clock is generated from the receiver clock or the divider clock or an external
clock scanned on the TK I/O pad. The transmitter clock is selected by the CKS field in
SSC_TCMR (Transmit Clock Mode Register). Transmit Clock can be inverted independently by
the CKI bits in SSC_TCMR.
The transmitter can also drive the TK I/O pad continuously or be limited to the actual data transfer. The clock output is configured by the SSC_TCMR register. The Transmit Clock Inversion
(CKI) bits have no effect on the clock outputs. Programming the TCMR register to select TK pin
(CKS field) and at the same time Continuous Transmit Clock (CKO field) might lead to unpredictable results.
Figure 31-6. Transmitter Clock Management
TK (pin)

Clock
Output

Tri_state
Controller

MUX
Receiver
Clock

Divider
Clock
Data Transfer

CKO

CKS

31.7.1.3

INV
MUX

Tri-state
Controller

CKI

CKG

Transmitter
Clock

Receiver Clock Management
The receiver clock is generated from the transmitter clock or the divider clock or an external
clock scanned on the RK I/O pad. The Receive Clock is selected by the CKS field in
SSC_RCMR (Receive Clock Mode Register). Receive Clocks can be inverted independently by
the CKI bits in SSC_RCMR.
The receiver can also drive the RK I/O pad continuously or be limited to the actual data transfer.
The clock output is configured by the SSC_RCMR register. The Receive Clock Inversion (CKI)
bits have no effect on the clock outputs. Programming the RCMR register to select RK pin (CKS
field) and at the same time Continuous Receive Clock (CKO field) can lead to unpredictable
results.

553
6430F–ATARM–21-Feb-12

Figure 31-7. Receiver Clock Management
RK (pin)
Tri-state
Controller

MUX

Clock
Output

Transmitter
Clock

Divider
Clock
Data Transfer

CKO

CKS

31.7.1.4

INV
MUX

Tri-state
Controller

CKI

CKG

Receiver
Clock

Serial Clock Ratio Considerations
The Transmitter and the Receiver can be programmed to operate with the clock signals provided
on either the TK or RK pins. This allows the SSC to support many slave-mode data transfers. In
this case, the maximum clock speed allowed on the RK pin is:
– Master Clock divided by 2 if Receiver Frame Synchro is input
– Master Clock divided by 3 if Receiver Frame Synchro is output
In addition, the maximum clock speed allowed on the TK pin is:
– Master Clock divided by 6 if Transmit Frame Synchro is input
– Master Clock divided by 2 if Transmit Frame Synchro is output

31.7.2

Transmitter Operations
A transmitted frame is triggered by a start event and can be followed by synchronization data
before data transmission.
The start event is configured by setting the Transmit Clock Mode Register (SSC_TCMR). See
“Start” on page 556.
The frame synchronization is configured setting the Transmit Frame Mode Register
(SSC_TFMR). See “Frame Sync” on page 558.
To transmit data, the transmitter uses a shift register clocked by the transmitter clock signal and
the start mode selected in the SSC_TCMR. Data is written by the application to the SSC_THR
register then transferred to the shift register according to the data format selected.
When both the SSC_THR and the transmit shift register are empty, the status flag TXEMPTY is
set in SSC_SR. When the Transmit Holding register is transferred in the Transmit shift register,
the status flag TXRDY is set in SSC_SR and additional data can be loaded in the holding
register.

554

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 31-8. Transmitter Block Diagram
SSC_CRTXEN
SSC_SRTXEN

TXEN

SSC_CRTXDIS
SSC_TCMR.STTDLY
SSC_TFMR.FSDEN
SSC_RCMR.START SSC_TCMR.START SSC_TFMR.DATNB
SSC_TFMR.DATDEF
SSC_TFMR.MSBF
RXEN
TXEN
TX Start
TX Start
Start
RX Start
Start
RF
Selector
Selector
RF
RC0R

TX Controller
TD

Transmit Shift Register
SSC_TFMR.FSDEN
SSC_TCMR.STTDLY != 0

SSC_TFMR.DATLEN

SSC_THR

Transmitter Clock

SSC_TSHR

SSC_TFMR.FSLEN

TX Controller counter reached STTDLY

31.7.3

Receiver Operations
A received frame is triggered by a start event and can be followed by synchronization data
before data transmission.
The start event is configured setting the Receive Clock Mode Register (SSC_RCMR). See
“Start” on page 556.
The frame synchronization is configured setting the Receive Frame Mode Register
(SSC_RFMR). See “Frame Sync” on page 558.
The receiver uses a shift register clocked by the receiver clock signal and the start mode
selected in the SSC_RCMR. The data is transferred from the shift register depending on the
data format selected.
When the receiver shift register is full, the SSC transfers this data in the holding register, the status flag RXRDY is set in SSC_SR and the data can be read in the receiver holding register. If
another transfer occurs before read of the RHR register, the status flag OVERUN is set in
SSC_SR and the receiver shift register is transferred in the RHR register.

555
6430F–ATARM–21-Feb-12

Figure 31-9. Receiver Block Diagram
SSC_CR.RXEN
SSC_SR.RXEN
SSC_CR.RXDIS
SSC_TCMR.START
SSC_RCMR.START
TXEN
RX Start
RF

Start
Selector

RXEN
RF
RC0R

Start
Selector

SSC_RFMR.MSBF
SSC_RFMR.DATNB
RX Start

RX Controller
RD

Receive Shift Register

SSC_RCMR.STTDLY != 0
load

SSC_RSHR

SSC_RFMR.FSLEN

load

SSC_RHR

Receiver Clock

SSC_RFMR.DATLEN
RX Controller counter reached STTDLY

31.7.4

Start
The transmitter and receiver can both be programmed to start their operations when an event
occurs, respectively in the Transmit Start Selection (START) field of SSC_TCMR and in the
Receive Start Selection (START) field of SSC_RCMR.
Under the following conditions the start event is independently programmable:
• Continuous. In this case, the transmission starts as soon as a word is written in SSC_THR
and the reception starts as soon as the Receiver is enabled.
• Synchronously with the transmitter/receiver
• On detection of a falling/rising edge on TF/RF
• On detection of a low level/high level on TF/RF
• On detection of a level change or an edge on TF/RF
A start can be programmed in the same manner on either side of the Transmit/Receive Clock
Register (RCMR/TCMR). Thus, the start could be on TF (Transmit) or RF (Receive).
Moreover, the Receiver can start when data is detected in the bit stream with the Compare
Functions.
Detection on TF/RF input/output is done by the field FSOS of the Transmit/Receive Frame Mode
Register (TFMR/RFMR).

556

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 31-10. Transmit Start Mode
TK
TF
(Input)

Start = Low Level on TF

Start = Falling Edge on TF

Start = High Level on TF

Start = Rising Edge on TF

Start = Level Change on TF

Start = Any Edge on TF

TD
(Output)
TD
(Output)

STTDLY
BO

B1
STTDLY
BO

TD
(Output)

B1
STTDLY

TD
(Output)

B1
STTDLY

TD
(Output)
TD
(Output)

B1
STTDLY

Figure 31-11. Receive Pulse/Edge Start Modes
RK
RF
(Input)

Start = Low Level on RF

Start = Falling Edge on RF

Start = High Level on RF

Start = Rising Edge on RF

Start = Level Change on RF

Start = Any Edge on RF

RD
(Input)
RD
(Input)

STTDLY
BO

B1
STTDLY
BO

RD
(Input)

B1
STTDLY

RD
(Input)

B1
STTDLY

RD
(Input)
RD
(Input)

B1
STTDLY

557
6430F–ATARM–21-Feb-12

31.7.5

Frame Sync
The Transmitter and Receiver Frame Sync pins, TF and RF, can be programmed to generate
different kinds of frame synchronization signals. The Frame Sync Output Selection (FSOS) field
in the Receive Frame Mode Register (SSC_RFMR) and in the Transmit Frame Mode Register
(SSC_TFMR) are used to select the required waveform.
• Programmable low or high levels during data transfer are supported.
• Programmable high levels before the start of data transfers or toggling are also supported.
If a pulse waveform is selected, the Frame Sync Length (FSLEN) field in SSC_RFMR and
SSC_TFMR programs the length of the pulse, from 1 bit time up to 256 bit time.
The periodicity of the Receive and Transmit Frame Sync pulse output can be programmed
through the Period Divider Selection (PERIOD) field in SSC_RCMR and SSC_TCMR.

31.7.5.1

Frame Sync Data
Frame Sync Data transmits or receives a specific tag during the Frame Sync signal.
During the Frame Sync signal, the Receiver can sample the RD line and store the data in the
Receive Sync Holding Register and the transmitter can transfer Transmit Sync Holding Register
in the Shifter Register. The data length to be sampled/shifted out during the Frame Sync signal
is programmed by the FSLEN field in SSC_RFMR/SSC_TFMR and has a maximum value of 16.
Concerning the Receive Frame Sync Data operation, if the Frame Sync Length is equal to or
lower than the delay between the start event and the actual data reception, the data sampling
operation is performed in the Receive Sync Holding Register through the Receive Shift Register.
The Transmit Frame Sync Operation is performed by the transmitter only if the bit Frame Sync
Data Enable (FSDEN) in SSC_TFMR is set. If the Frame Sync length is equal to or lower than
the delay between the start event and the actual data transmission, the normal transmission has
priority and the data contained in the Transmit Sync Holding Register is transferred in the Transmit Register, then shifted out.

31.7.5.2

31.7.6

Frame Sync Edge Detection
The Frame Sync Edge detection is programmed by the FSEDGE field in
SSC_RFMR/SSC_TFMR. This sets the corresponding flags RXSYN/TXSYN in the SSC Status
Register (SSC_SR) on frame synchro edge detection (signals RF/TF).
Receive Compare Modes
Figure 31-12. Receive Compare Modes
RK

RD
(Input)

CMP0

CMP1

CMP2

CMP3

Ignored

Start

FSLEN
Up to 16 Bits
(4 in This Example)

558

STDLY

DATLEN

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
31.7.6.1

31.7.7

Compare Functions
Length of the comparison patterns (Compare 0, Compare 1) and thus the number of bits they
are compared to is defined by FSLEN, but with a maximum value of 16 bits. Comparison is
always done by comparing the last bits received with the comparison pattern. Compare 0 can be
one start event of the Receiver. In this case, the receiver compares at each new sample the last
bits received at the Compare 0 pattern contained in the Compare 0 Register (SSC_RC0R).
When this start event is selected, the user can program the Receiver to start a new data transfer
either by writing a new Compare 0, or by receiving continuously until Compare 1 occurs. This
selection is done with the bit (STOP) in SSC_RCMR.
Data Format
The data framing format of both the transmitter and the receiver are programmable through the
Transmitter Frame Mode Register (SSC_TFMR) and the Receiver Frame Mode Register
(SSC_RFMR). In either case, the user can independently select:
• the event that starts the data transfer (START)
• the delay in number of bit periods between the start event and the first data bit (STTDLY)
• the length of the data (DATLEN)
• the number of data to be transferred for each start event (DATNB).
• the length of synchronization transferred for each start event (FSLEN)
• the bit sense: most or lowest significant bit first (MSBF)
Additionally, the transmitter can be used to transfer synchronization and select the level driven
on the TD pin while not in data transfer operation. This is done respectively by the Frame Sync
Data Enable (FSDEN) and by the Data Default Value (DATDEF) bits in SSC_TFMR.

559
6430F–ATARM–21-Feb-12

Table 31-5.

Data Frame Registers

Transmitter

Receiver

Field

Length

Comment

SSC_TFMR

SSC_RFMR

DATLEN

Up to 32

Size of word

SSC_TFMR

SSC_RFMR

DATNB

Up to 16

Number of words transmitted in frame

SSC_TFMR

SSC_RFMR

MSBF

SSC_TFMR

SSC_RFMR

FSLEN

Up to 16

Size of Synchro data register

SSC_TFMR

DATDEF

0 or 1

Data default value ended

SSC_TFMR

FSDEN

Most significant bit first

Enable send SSC_TSHR

SSC_TCMR

SSC_RCMR

PERIOD

Up to 512

Frame size

SSC_TCMR

SSC_RCMR

STTDLY

Up to 255

Size of transmit start delay

Figure 31-13. Transmit and Receive Frame Format in Edge/Pulse Start Modes
Start

Start
PERIOD
TF/RF

(1)

FSLEN
TD
(If FSDEN = 1)

TD
(If FSDEN = 0)
RD

Sync Data

Default

From SSC_TSHR FromDATDEF
Default

Data
From SSC_THR

Ignored

STTDLY

Data

To SSC_RHR

DATLEN

Sync Data

FromDATDEF

From SSC_THR

Data

To SSC_RSHR

Default

Data

Data
From SSC_THR

From DATDEF
Sync Data

Data
From SSC_THR

Default
From DATDEF
Ignored

Sync Data

DATNB

Note:

1. Example of input on falling edge of TF/RF.

Figure 31-14. Transmit Frame Format in Continuous Mode
Start

Data
From SSC_THR

Data

Default

From SSC_THR

DATLEN

Start: 1. TXEMPTY set to 1
2. Write into the SSC_THR

560

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Note:

1. STTDLY is set to 0. In this example, SSC_THR is loaded twice. FSDEN value has no effect on
the transmission. SyncData cannot be output in continuous mode.

Figure 31-15. Receive Frame Format in Continuous Mode
Start = Enable Receiver

Data

To SSC_RHR

DATLEN

Note:

31.7.8

1. STTDLY is set to 0.

Loop Mode
The receiver can be programmed to receive transmissions from the transmitter. This is done by
setting the Loop Mode (LOOP) bit in SSC_RFMR. In this case, RD is connected to TD, RF is
connected to TF and RK is connected to TK.

31.7.9

Interrupt
Most bits in SSC_SR have a corresponding bit in interrupt management registers.
The SSC can be programmed to generate an interrupt when it detects an event. The interrupt is
controlled by writing SSC_IER (Interrupt Enable Register) and SSC_IDR (Interrupt Disable Register) These registers enable and disable, respectively, the corresponding interrupt by setting
and clearing the corresponding bit in SSC_IMR (Interrupt Mask Register), which controls the
generation of interrupts by asserting the SSC interrupt line connected to the NVIC.
Figure 31-16. Interrupt Block Diagram
SSC_IMR
SSC_IER

SSC_IDR

Set

Clear

Transmitter
TXRDY
TXEMPTY
TXSYNC
Interrupt
Control

SSC Interrupt

Receiver
RXRDY
OVRUN
RXSYNC

561
6430F–ATARM–21-Feb-12

31.8

SSC Application Examples
The SSC can support several serial communication modes used in audio or high speed serial
links. Some standard applications are shown in the following figures. All serial link applications
supported by the SSC are not listed here.

Figure 31-17. Audio Application Block Diagram
Clock SCK
TK
Word Select WS

I2S
RECEIVER

TF
Data SD
SSC

TD
RD

Clock SCK

Word Select WS

MSB

Data SD

LSB

MSB
Right Channel

Left Channel

Figure 31-18. Codec Application Block Diagram
Serial Data Clock (SCLK)
TK
Frame sync (FSYNC)
TF
Serial Data Out
SSC

CODEC

TD
Serial Data In
RD
RF
RK

Serial Data Clock (SCLK)
Frame sync (FSYNC)

First Time Slot
Dstart

Dend

Serial Data Out

Serial Data In

562

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 31-19. Time Slot Application Block Diagram
SCLK
TK
FSYNC
TF

CODEC
First
Time Slot

Data Out
TD
SSC
RD

Data in

RF
RK

CODEC
Second
Time Slot

Serial Data Clock (SCLK)
Frame sync (FSYNC)

First Time Slot
Dstart

Second Time Slot
Dend

Serial Data Out

Serial Data in

563
6430F–ATARM–21-Feb-12

31.8.1

Write Protection Registers
To prevent any single software error that may corrupt SSC behavior, certain address spaces can
be write-protected by setting the WPEN bit in the “SSC Write Protect Mode Register”
(SSC_WPMR).
If a write access to the protected registers is detected, then the WPVS flag in the SSC Write Protect Status Register (US_WPSR) is set and the field WPVSRC indicates in which register the
write access has been attempted.
The WPVS flag is reset by writing the SSC Write Protect Mode Register (SSC_WPMR) with the
appropriate access key, WPKEY.
The protected registers are:
• “SSC Clock Mode Register” on page 567
• “SSC Receive Clock Mode Register” on page 568
• “SSC Receive Frame Mode Register” on page 570
• “SSC Transmit Clock Mode Register” on page 572
• “SSC Transmit Frame Mode Register” on page 574
• “SSC Receive Compare 0 Register” on page 580
• “SSC Receive Compare 1 Register” on page 581

564

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
31.9

Synchronous Serial Controller (SSC) User Interface

Table 31-6.
Offset

Name

Access

Reset

SSC_CR

Write-only

–

SSC_CMR

Read-write

0x0

Control Register

0x4

Clock Mode Register

0x8

Reserved

–

0xC

Reserved

–

0x10

Receive Clock Mode Register

SSC_RCMR

Read-write

0x0

0x14

Receive Frame Mode Register

SSC_RFMR

Read-write

0x0

0x18

Transmit Clock Mode Register

SSC_TCMR

Read-write

0x0

0x1C

Transmit Frame Mode Register

SSC_TFMR

Read-write

0x0

0x20

Receive Holding Register

SSC_RHR

Read-only

0x0

0x24

Transmit Holding Register

SSC_THR

Write-only

–

0x28

Reserved

–

0x2C

Reserved

–

0x30

Receive Sync. Holding Register

SSC_RSHR

Read-only

0x0

0x34

Transmit Sync. Holding Register

SSC_TSHR

Read-write

0x0

0x38

Receive Compare 0 Register

SSC_RC0R

Read-write

0x0

0x3C

Receive Compare 1 Register

SSC_RC1R

Read-write

0x0

0x40

Status Register

SSC_SR

Read-only

0x000000CC

0x44

Interrupt Enable Register

SSC_IER

Write-only

–

0x48

Interrupt Disable Register

SSC_IDR

Write-only

–

0x4C

Interrupt Mask Register

SSC_IMR

Read-only

0x0

0xE4

Write Protect Mode Register

SSC_WPMR

Read-write

0x0

0xE8

Write Protect Status Register

SSC_WPSR

Read-only

0x0

0x50-0xFC

Reserved

–

0x100- 0x124

Reserved

–

565
6430F–ATARM–21-Feb-12

31.9.1
Name:

SSC Control Register
SSC_CR

Address:

0x40004000

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
SWRST

14
–

13
–

12
–

11
–

10
–

9
TXDIS

8
TXEN

7
–

6
–

5
–

4
–

3
–

2
–

1
RXDIS

0
RXEN

• RXEN: Receive Enable
0 = No effect.
1 = Enables Receive if RXDIS is not set.
• RXDIS: Receive Disable
0 = No effect.
1 = Disables Receive. If a character is currently being received, disables at end of current character reception.
• TXEN: Transmit Enable
0 = No effect.
1 = Enables Transmit if TXDIS is not set.
• TXDIS: Transmit Disable
0 = No effect.
1 = Disables Transmit. If a character is currently being transmitted, disables at end of current character transmission.
• SWRST: Software Reset
0 = No effect.
1 = Performs a software reset. Has priority on any other bit in SSC_CR.

566

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
31.9.2
Name:

SSC Clock Mode Register
SSC_CMR

Address:

0x40004004

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

DIV
3

DIV

This register can only be written if the WPEN bit is cleared in “SSC Write Protect Mode Register” .
• DIV: Clock Divider
0 = The Clock Divider is not active.
Any Other Value: The Divided Clock equals the Master Clock divided by 2 times DIV. The maximum bit rate is MCK/2. The
minimum bit rate is MCK/2 x 4095 = MCK/8190.

567
6430F–ATARM–21-Feb-12

31.9.3
Name:

SSC Receive Clock Mode Register
SSC_RCMR

Address:

0x40004010

Access:

Read-write

PERIOD
23

20
STTDLY

15
–
7

14
–

13
–

12
STOP

5
CKI

3
CKO

CKG

START
2

0
CKS

This register can only be written if the WPEN bit is cleared in “SSC Write Protect Mode Register” .
• CKS: Receive Clock Selection
Value

Name

Description

MCK

Divided Clock

TK Clock signal

RK pin
Reserved

• CKO: Receive Clock Output Mode Selection
Value

Name

Description

RK Pin

NONE

None

Input-only

CONTINUOUS

Continuous Receive Clock

Output

TRANSFER

Receive Clock only during data transfers

Output

3-7

Reserved

• CKI: Receive Clock Inversion
0 = The data inputs (Data and Frame Sync signals) are sampled on Receive Clock falling edge. The Frame Sync signal
output is shifted out on Receive Clock rising edge.
1 = The data inputs (Data and Frame Sync signals) are sampled on Receive Clock rising edge. The Frame Sync signal output is shifted out on Receive Clock falling edge.
CKI affects only the Receive Clock and not the output clock signal.

568

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• CKG: Receive Clock Gating Selection
Value

Name

Description

RK Pin

NONE

None

Input-only

CONTINUOUS

Continuous Receive Clock

Output

TRANSFER

Receive Clock only during data transfers

Output

3-7

Reserved

• START: Receive Start Selection
Value

Name

Description

CONTINUOUS

Continuous, as soon as the receiver is enabled, and
immediately after the end of transfer of the previous data.

TRANSMIT

Transmit start

RF_LOW

Detection of a low level on RF signal

RF_HIGH

Detection of a high level on RF signal

RF_FALLING

Detection of a falling edge on RF signal

RF_RISING

Detection of a rising edge on RF signal

RF_LEVEL

Detection of any level change on RF signal

RF_EDGE

Detection of any edge on RF signal

CMP_0

Compare 0

• STOP: Receive Stop Selection
0 = After completion of a data transfer when starting with a Compare 0, the receiver stops the data transfer and waits for a
new compare 0.
1 = After starting a receive with a Compare 0, the receiver operates in a continuous mode until a Compare 1 is detected.
• STTDLY: Receive Start Delay
If STTDLY is not 0, a delay of STTDLY clock cycles is inserted between the start event and the actual start of reception.
When the Receiver is programmed to start synchronously with the Transmitter, the delay is also applied.
Note: It is very important that STTDLY be set carefully. If STTDLY must be set, it should be done in relation to TAG
(Receive Sync Data) reception.
• PERIOD: Receive Period Divider Selection
This field selects the divider to apply to the selected Receive Clock in order to generate a new Frame Sync Signal. If 0, no
PERIOD signal is generated. If not 0, a PERIOD signal is generated each 2 x (PERIOD+1) Receive Clock.

569
6430F–ATARM–21-Feb-12

31.9.4
Name:

SSC Receive Frame Mode Register
SSC_RFMR

Address:

0x40004014

Access:

Read-write

31
FSLEN_EXT

30
FSLEN_EXT

29
FSLEN_EXT

23
–

15
–
7
MSBF

28
FSLEN_EXT

27
–

26
–

21
FSOS

14
–

13
–

12
–

6
–

5
LOOP

25
–

24
FSEDGE

FSLEN
10
DATNB
2
DATLEN

This register can only be written if the WPEN bit is cleared in “SSC Write Protect Mode Register” .
• DATLEN: Data Length
0 = Forbidden value (1-bit data length not supported).
Any other value: The bit stream contains DATLEN + 1 data bits.
• LOOP: Loop Mode
0 = Normal operating mode.
1 = RD is driven by TD, RF is driven by TF and TK drives RK.
• MSBF: Most Significant Bit First
0 = The lowest significant bit of the data register is sampled first in the bit stream.
1 = The most significant bit of the data register is sampled first in the bit stream.
• DATNB: Data Number per Frame
This field defines the number of data words to be received after each transfer start, which is equal to (DATNB + 1).
• FSLEN: Receive Frame Sync Length
This field defines the number of bits sampled and stored in the Receive Sync Data Register. When this mode is selected by
the START field in the Receive Clock Mode Register, it also determines the length of the sampled data to be compared to
the Compare 0 or Compare 1 register.
This field is used with FSLEN_EXT to determine the pulse length of the Receive Frame Sync signal.
Pulse length is equal to FSLEN + (FSLEN_EXT * 16) + 1 Receive Clock periods.

570

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• FSOS: Receive Frame Sync Output Selection
Value

Name

Description

RF Pin

NONE

None

Input-only

NEGATIVE

Negative Pulse

Output

POSITIVE

Positive Pulse

Output

LOW

Driven Low during data transfer

Output

HIGH

Driven High during data transfer

Output

TOGGLING

Toggling at each start of data transfer

Output

Reserved

Undefined

6-7

• FSEDGE: Frame Sync Edge Detection
Determines which edge on Frame Sync will generate the interrupt RXSYN in the SSC Status Register.
Value

Name

Description

POSITIVE

Positive Edge Detection

NEGATIVE

Negative Edge Detection

• FSLEN_EXT: FSLEN Field Extension
Extends FSLEN field. For details, refer to FSLEN bit description on page 570.

571
6430F–ATARM–21-Feb-12

31.9.5
Name:

SSC Transmit Clock Mode Register
SSC_TCMR

Address:

0x40004018

Access:

Read-write

PERIOD
23

20
STTDLY

15
–
7

14
–

13
–

12
–

5
CKI

3
CKO

CKG

START
2

0
CKS

This register can only be written if the WPEN bit is cleared in “SSC Write Protect Mode Register” .
• CKS: Transmit Clock Selection
Value

Name

Description

MCK

Divided Clock

TK Clock signal

RK pin
Reserved

• CKO: Transmit Clock Output Mode Selection
Value

Name

Description

TK Pin

NONE

None

Input-only

CONTINUOUS

Continuous Receive Clock

Output

TRANSFER

Transmit Clock only during data transfers

Output

3-7

Reserved

• CKI: Transmit Clock Inversion
0 = The data outputs (Data and Frame Sync signals) are shifted out on Transmit Clock falling edge. The Frame sync signal
input is sampled on Transmit clock rising edge.
1 = The data outputs (Data and Frame Sync signals) are shifted out on Transmit Clock rising edge. The Frame sync signal
input is sampled on Transmit clock falling edge.
CKI affects only the Transmit Clock and not the output clock signal.

572

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• CKG: Transmit Clock Gating Selection
Value

Name

Description

NONE

None

CONTINUOUS

Transmit Clock enabled only if TF Low

TRANSFER

Transmit Clock enabled only if TF High

• START: Transmit Start Selection
Value

Name

Description

CONTINUOUS

Continuous, as soon as a word is written in the SSC_THR Register (if Transmit is enabled), and
immediately after the end of transfer of the previous data.

RECEIVE

Receive start

RF_LOW

Detection of a low level on TF signal

RF_HIGH

Detection of a high level on TF signal

RF_FALLING

Detection of a falling edge on TF signal

RF_RISING

Detection of a rising edge on TF signal

RF_LEVEL

Detection of any level change on TF signal

RF_EDGE

Detection of any edge on TF signal

CMP_0

Compare 0

• STTDLY: Transmit Start Delay
If STTDLY is not 0, a delay of STTDLY clock cycles is inserted between the start event and the actual start of transmission
of data. When the Transmitter is programmed to start synchronously with the Receiver, the delay is also applied.
Note: STTDLY must be set carefully. If STTDLY is too short in respect to TAG (Transmit Sync Data) emission, data is emitted instead of the end of TAG.
• PERIOD: Transmit Period Divider Selection
This field selects the divider to apply to the selected Transmit Clock to generate a new Frame Sync Signal. If 0, no period
signal is generated. If not 0, a period signal is generated at each 2 x (PERIOD+1) Transmit Clock.

573
6430F–ATARM–21-Feb-12

31.9.6
Name:

SSC Transmit Frame Mode Register
SSC_TFMR

Address:

0x4000401C

Access:

Read-write

31
FSLEN_EXT

30
FSLEN_EXT

29
FSLEN_EXT

23
FSDEN

15
–
7
MSBF

28
FSLEN_EXT

27
–

26
–

21
FSOS

14
–

13
–

12
–

6
–

5
DATDEF

25
–

24
FSEDGE

FSLEN
10
DATNB
2
DATLEN

This register can only be written if the WPEN bit is cleared in “SSC Write Protect Mode Register” .
• DATLEN: Data Length
0 = Forbidden value (1-bit data length not supported).
Any other value: The bit stream contains DATLEN + 1 data bits.
• DATDEF: Data Default Value
This bit defines the level driven on the TD pin while out of transmission. Note that if the pin is defined as multi-drive by the
PIO Controller, the pin is enabled only if the SCC TD output is 1.
• MSBF: Most Significant Bit First
0 = The lowest significant bit of the data register is shifted out first in the bit stream.
1 = The most significant bit of the data register is shifted out first in the bit stream.
• DATNB: Data Number per frame
This field defines the number of data words to be transferred after each transfer start, which is equal to (DATNB +1).
• FSLEN: Transmit Frame Sync Length
This field defines the length of the Transmit Frame Sync signal and the number of bits shifted out from the Transmit Sync
Data Register if FSDEN is 1.
This field is used with FSLEN_EXT to determine the pulse length of the Transmit Frame Sync signal.
Pulse length is equal to FSLEN + (FSLEN_EXT * 16) + 1 Transmit Clock period.

574

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• FSOS: Transmit Frame Sync Output Selection
Value

Name

Description

RF Pin

NONE

None

Input-only

NEGATIVE

Negative Pulse

Output

POSITIVE

Positive Pulse

Output

LOW

Driven Low during data transfer

Output

HIGH

Driven High during data transfer

Output

TOGGLING

Toggling at each start of data transfer

Output

Reserved

Undefined

6-7

• FSDEN: Frame Sync Data Enable
0 = The TD line is driven with the default value during the Transmit Frame Sync signal.
1 = SSC_TSHR value is shifted out during the transmission of the Transmit Frame Sync signal.
• FSEDGE: Frame Sync Edge Detection
Determines which edge on frame sync will generate the interrupt TXSYN (Status Register).
Value

Name

Description

POSITIVE

Positive Edge Detection

NEGATIVE

Negative Edge Detection

• FSLEN_EXT: FSLEN Field Extension
Extends FSLEN field. For details, refer to FSLEN bit description on page 574.

575
6430F–ATARM–21-Feb-12

31.9.7
Name:

SSC Receive Holding Register
SSC_RHR

Address:

0x40004020

Access:

Read-only

RDAT
23

20
RDAT

12
RDAT

4
RDAT

• RDAT: Receive Data
Right aligned regardless of the number of data bits defined by DATLEN in SSC_RFMR.

576

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
31.9.8
Name:

SSC Transmit Holding Register
SSC_THR

Address:

0x40004024

Access:

Write-only

TDAT
23

20
TDAT

12
TDAT

4
TDAT

• TDAT: Transmit Data
Right aligned regardless of the number of data bits defined by DATLEN in SSC_TFMR.

577
6430F–ATARM–21-Feb-12

31.9.9
Name:

SSC Receive Synchronization Holding Register
SSC_RSHR

Address:

0x40004030

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

RSDAT
7

4
RSDAT

• RSDAT: Receive Synchronization Data

578

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
31.9.10
Name:

SSC Transmit Synchronization Holding Register
SSC_TSHR

Address:

0x40004034

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

TSDAT
7

4
TSDAT

• TSDAT: Transmit Synchronization Data

579
6430F–ATARM–21-Feb-12

31.9.11
Name:

SSC Receive Compare 0 Register
SSC_RC0R

Address:

0x40004038

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

CP0
7

4
CP0

This register can only be written if the WPEN bit is cleared in “SSC Write Protect Mode Register” .
• CP0: Receive Compare Data 0

580

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
31.9.12
Name:

SSC Receive Compare 1 Register
SSC_RC1R

Address:

0x4000403C

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

CP1
7

4
CP1

This register can only be written if the WPEN bit is cleared in “SSC Write Protect Mode Register” .
• CP1: Receive Compare Data 1

581
6430F–ATARM–21-Feb-12

31.9.13
Name:

SSC Status Register
SSC_SR

Address:

0x40004040

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
RXEN

16
TXEN

15
–

14
–

13
–

12
–

11
RXSYN

10
TXSYN

9
CP1

8
CP0

7
–

6
–

5
OVRUN

4
RXRDY

3
–

2
–

1
TXEMPTY

0
TXRDY

• TXRDY: Transmit Ready
0 = Data has been loaded in SSC_THR and is waiting to be loaded in the Transmit Shift Register (TSR).
1 = SSC_THR is empty.
• TXEMPTY: Transmit Empty
0 = Data remains in SSC_THR or is currently transmitted from TSR.
1 = Last data written in SSC_THR has been loaded in TSR and last data loaded in TSR has been transmitted.
• RXRDY: Receive Ready
0 = SSC_RHR is empty.
1 = Data has been received and loaded in SSC_RHR.
• OVRUN: Receive Overrun
0 = No data has been loaded in SSC_RHR while previous data has not been read since the last read of the Status
Register.
1 = Data has been loaded in SSC_RHR while previous data has not yet been read since the last read of the Status
Register.
• CP0: Compare 0
0 = A compare 0 has not occurred since the last read of the Status Register.
1 = A compare 0 has occurred since the last read of the Status Register.
• CP1: Compare 1
0 = A compare 1 has not occurred since the last read of the Status Register.
1 = A compare 1 has occurred since the last read of the Status Register.
• TXSYN: Transmit Sync
0 = A Tx Sync has not occurred since the last read of the Status Register.
1 = A Tx Sync has occurred since the last read of the Status Register.

582

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• RXSYN: Receive Sync
0 = An Rx Sync has not occurred since the last read of the Status Register.
1 = An Rx Sync has occurred since the last read of the Status Register.
• TXEN: Transmit Enable
0 = Transmit is disabled.
1 = Transmit is enabled.
• RXEN: Receive Enable
0 = Receive is disabled.
1 = Receive is enabled.

583
6430F–ATARM–21-Feb-12

31.9.14
Name:

SSC Interrupt Enable Register
SSC_IER

Address:

0x40004044

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
RXSYN

10
TXSYN

9
CP1

8
CP0

7
–

6
–

5
OVRUN

4
RXRDY

3
–

2
–

1
TXEMPTY

0
TXRDY

• TXRDY: Transmit Ready Interrupt Enable
0 = 0 = No effect.
1 = Enables the Transmit Ready Interrupt.
• TXEMPTY: Transmit Empty Interrupt Enable
0 = No effect.
1 = Enables the Transmit Empty Interrupt.
• RXRDY: Receive Ready Interrupt Enable
0 = No effect.
1 = Enables the Receive Ready Interrupt.
• OVRUN: Receive Overrun Interrupt Enable
0 = No effect.
1 = Enables the Receive Overrun Interrupt.
• CP0: Compare 0 Interrupt Enable
0 = No effect.
1 = Enables the Compare 0 Interrupt.
• CP1: Compare 1 Interrupt Enable
0 = No effect.
1 = Enables the Compare 1 Interrupt.
• TXSYN: Tx Sync Interrupt Enable
0 = No effect.
1 = Enables the Tx Sync Interrupt.
• RXSYN: Rx Sync Interrupt Enable

584

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
0 = No effect.
1 = Enables the Rx Sync Interrupt.

585
6430F–ATARM–21-Feb-12

31.9.15
Name:

SSC Interrupt Disable Register
SSC_IDR

Address:

0x40004048

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
RXSYN

10
TXSYN

9
CP1

8
CP0

7
–

6
–

5
OVRUN

4
RXRDY

3
–

2
–

1
TXEMPTY

0
TXRDY

• TXRDY: Transmit Ready Interrupt Disable
0 = No effect.
1 = Disables the Transmit Ready Interrupt.
• TXEMPTY: Transmit Empty Interrupt Disable
0 = No effect.
1 = Disables the Transmit Empty Interrupt.
• RXRDY: Receive Ready Interrupt Disable
0 = No effect.
1 = Disables the Receive Ready Interrupt.
• OVRUN: Receive Overrun Interrupt Disable
0 = No effect.
1 = Disables the Receive Overrun Interrupt.
• CP0: Compare 0 Interrupt Disable
0 = No effect.
1 = Disables the Compare 0 Interrupt.
• CP1: Compare 1 Interrupt Disable
0 = No effect.
1 = Disables the Compare 1 Interrupt.
• TXSYN: Tx Sync Interrupt Enable
0 = No effect.
1 = Disables the Tx Sync Interrupt.

586

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• RXSYN: Rx Sync Interrupt Enable
0 = No effect.
1 = Disables the Rx Sync Interrupt.

587
6430F–ATARM–21-Feb-12

31.9.16
Name:

SSC Interrupt Mask Register
SSC_IMR

Address:

0x4000404C

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
RXSYN

10
TXSYN

9
CP1

8
CP0

7
–

6
–

5
OVRUN

4
RXRDY

3
–

2
–

1
TXEMPTY

0
TXRDY

• TXRDY: Transmit Ready Interrupt Mask
0 = The Transmit Ready Interrupt is disabled.
1 = The Transmit Ready Interrupt is enabled.
• TXEMPTY: Transmit Empty Interrupt Mask
0 = The Transmit Empty Interrupt is disabled.
1 = The Transmit Empty Interrupt is enabled.
• RXRDY: Receive Ready Interrupt Mask
0 = The Receive Ready Interrupt is disabled.
1 = The Receive Ready Interrupt is enabled.
• OVRUN: Receive Overrun Interrupt Mask
0 = The Receive Overrun Interrupt is disabled.
1 = The Receive Overrun Interrupt is enabled.
• CP0: Compare 0 Interrupt Mask
0 = The Compare 0 Interrupt is disabled.
1 = The Compare 0 Interrupt is enabled.
• CP1: Compare 1 Interrupt Mask
0 = The Compare 1 Interrupt is disabled.
1 = The Compare 1 Interrupt is enabled.
• TXSYN: Tx Sync Interrupt Mask
0 = The Tx Sync Interrupt is disabled.
1 = The Tx Sync Interrupt is enabled.

588

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• RXSYN: Rx Sync Interrupt Mask
0 = The Rx Sync Interrupt is disabled.
1 = The Rx Sync Interrupt is enabled.

589
6430F–ATARM–21-Feb-12

31.9.17
Name:

SSC Write Protect Mode Register
SSC_WPMR

Address:

0x400040E4

Access:

Read-write

Reset:

See Table 31-6

WPKEY
23

20
WPKEY

12
WPKEY

—

WPEN

• WPEN: Write Protect Enable
0 = Disables the Write Protect if WPKEY corresponds to 0x535343 (“SSC” in ASCII).
1 = Enables the Write Protect if WPKEY corresponds to 0x535343 (“SSC” in ASCII).
Protects the registers:
• “SSC Clock Mode Register” on page 567
• “SSC Receive Clock Mode Register” on page 568
• “SSC Receive Frame Mode Register” on page 570
• “SSC Transmit Clock Mode Register” on page 572
• “SSC Transmit Frame Mode Register” on page 574
• “SSC Receive Compare 0 Register” on page 580
• “SSC Receive Compare 1 Register” on page 581
• WPKEY: Write Protect KEY
Should be written at value 0x535343 (“SSC” in ASCII). Writing any other value in this field aborts the write operation of the
WPEN bit. Always reads as 0.

590

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
31.9.18
Name:

SSC Write Protect Status Register
SSC_WPSR

Address:

0x400040E8

Access:

Read-only

Reset:

See Table 31-6

—

WPVSRC
15

12
WPVSRC

—

WPVS

• WPVS: Write Protect Violation Status
0 = No Write Protect Violation has occurred since the last read of the SSC_WPSR register.
1 = A Write Protect Violation has occurred since the last read of the SSC_WPSR register. If this violation is an unauthorized
attempt to write a protected register, the associated violation is reported into field WPVSRC.
• WPVSRC: Write Protect Violation Source
When WPVS is active, this field indicates the write-protected register (through address offset or code) in which a write
access has been attempted.
Note:

Reading SSC_WPSR automatically clears all fields.

591
6430F–ATARM–21-Feb-12

592

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32. Serial Peripheral Interface (SPI) Programmer Datasheet
32.1

Description
The Serial Peripheral Interface (SPI) circuit is a synchronous serial data link that provides communication with external devices in Master or Slave Mode. It also enables communication
between processors if an external processor is connected to the system.
The Serial Peripheral Interface is essentially a shift register that serially transmits data bits to
other SPIs. During a data transfer, one SPI system acts as the “master”' which controls the data
flow, while the other devices act as “slaves'' which have data shifted into and out by the master.
Different CPUs can take turn being masters (Multiple Master Protocol opposite to Single Master
Protocol where one CPU is always the master while all of the others are always slaves) and one
master may simultaneously shift data into multiple slaves. However, only one slave may drive its
output to write data back to the master at any given time.
A slave device is selected when the master asserts its NSS signal. If multiple slave devices
exist, the master generates a separate slave select signal for each slave (NPCS).
The SPI system consists of two data lines and two control lines:
• Master Out Slave In (MOSI): This data line supplies the output data from the master shifted
into the input(s) of the slave(s).
• Master In Slave Out (MISO): This data line supplies the output data from a slave to the input
of the master. There may be no more than one slave transmitting data during any particular
transfer.
• Serial Clock (SPCK): This control line is driven by the master and regulates the flow of the
data bits. The master may transmit data at a variety of baud rates; the SPCK line cycles once
for each bit that is transmitted.
• Slave Select (NSS): This control line allows slaves to be turned on and off by hardware.

32.2

Embedded Characteristics
• Supports Communication with Serial External Devices
– Four Chip Selects with External Decoder Support Allow Communication with Up to
15 Peripherals
– Serial Memories, such as DataFlash and 3-wire EEPROMs
– Serial Peripherals, such as ADCs, DACs, LCD Controllers, CAN Controllers and
Sensors
– External Co-processors
• Master or Slave Serial Peripheral Bus Interface
– 8- to 16-bit Programmable Data Length Per Chip Select
– Programmable Phase and Polarity Per Chip Select
– Programmable Transfer Delay Between Consecutive Transfers and Delay Before SPI
Clock per Chip Select
– Programmable Delay Between Chip Selects
– Selectable Mode Fault Detection
• Connection to DMA Channel Capabilities Optimizes Data Transfers
– One channel for the Receiver, One Channel for the Transmitter

593
6430F–ATARM–21-Feb-12

32.3

Block Diagram

Figure 32-1. Block Diagram
AHB Matrix

DMA Ch.

Peripheral Bridge
APB
SPCK
MISO

PMC

MOSI

MCK
SPI Interface

PIO

NPCS0/NSS
NPCS1
NPCS2

Interrupt Control
NPCS3

SPI Interrupt

594

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.4

Application Block Diagram

Figure 32-2. Application Block Diagram: Single Master/Multiple Slave Implementation

SPI Master

SPCK

MISO

MOSI

NPCS0

NSS

Slave 0

SPCK

NPCS1
NPCS2

NPCS3

MISO
Slave 1
MOSI
NSS
SPCK
MISO
Slave 2
MOSI
NSS

32.5

Signal Description

Table 32-1.

Signal Description
Type

Pin Name

Pin Description

Master

Slave

MISO

Master In Slave Out

Input

Output

MOSI

Master Out Slave In

Output

Input

SPCK

Serial Clock

Output

Input

NPCS1-NPCS3

Peripheral Chip Selects

Output

Unused

NPCS0/NSS

Peripheral Chip Select/Slave Select

Output

Input

595
6430F–ATARM–21-Feb-12

32.6
32.6.1

Product Dependencies
I/O Lines
The pins used for interfacing the compliant external devices may be multiplexed with PIO lines.
The programmer must first program the PIO controllers to assign the SPI pins to their peripheral
functions.
Table 32-2.

I/O Lines

Instance

Signal

I/O Line

Peripheral

SPI

MISO

PA13

SPI

MOSI

PA14

SPI

NPCS0

PA16

SPI

NPCS1

PA0

SPI

NPCS1

PC3

SPI

NPCS1

PC19

SPI

NPCS2

PA1

SPI

NPCS2

PC4

SPI

NPCS2

PC14

SPI

NPCS3

PA19

SPI

NPCS3

PC5

SPI

SPCK

PA15

32.6.2

Power Management
The SPI may be clocked through the Power Management Controller (PMC), thus the programmer must first configure the PMC to enable the SPI clock.

32.6.3

Interrupt
The SPI interface has an interrupt line connected to the Interrupt Controller. Handling the SPI
interrupt requires programming the interrupt controller before configuring the SPI.
Table 32-3.

596

Peripheral IDs

Instance

SPI

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.7

Functional Description

32.7.1

Modes of Operation
The SPI operates in Master Mode or in Slave Mode.
Operation in Master Mode is programmed by writing at 1 the MSTR bit in the Mode Register.
The pins NPCS0 to NPCS3 are all configured as outputs, the SPCK pin is driven, the MISO line
is wired on the receiver input and the MOSI line driven as an output by the transmitter.
If the MSTR bit is written at 0, the SPI operates in Slave Mode. The MISO line is driven by the
transmitter output, the MOSI line is wired on the receiver input, the SPCK pin is driven by the
transmitter to synchronize the receiver. The NPCS0 pin becomes an input, and is used as a
Slave Select signal (NSS). The pins NPCS1 to NPCS3 are not driven and can be used for other
purposes.
The data transfers are identically programmable for both modes of operations. The baud rate
generator is activated only in Master Mode.

32.7.2

Data Transfer
Four combinations of polarity and phase are available for data transfers. The clock polarity is
programmed with the CPOL bit in the Chip Select Register. The clock phase is programmed with
the NCPHA bit. These two parameters determine the edges of the clock signal on which data is
driven and sampled. Each of the two parameters has two possible states, resulting in four possible combinations that are incompatible with one another. Thus, a master/slave pair must use the
same parameter pair values to communicate. If multiple slaves are used and fixed in different
configurations, the master must reconfigure itself each time it needs to communicate with a different slave.
Table 32-4 shows the four modes and corresponding parameter settings.

Table 32-4.

SPI Bus Protocol Mode

SPI Mode

CPOL

NCPHA

Shift SPCK Edge

Capture SPCK Edge

SPCK Inactive Level

Falling

Rising

Low

Rising

Falling

Low

Rising

Falling

High

Falling

Rising

High

Figure 32-3 and Figure 32-4 show examples of data transfers.

597
6430F–ATARM–21-Feb-12

Figure 32-3. SPI Transfer Format (NCPHA = 1, 8 bits per transfer)
1

SPCK cycle (for reference)

SPCK
(CPOL = 0)

SPCK
(CPOL = 1)

MOSI
(from master)

MSB

MISO
(from slave)

MSB

LSB

NSS
(to slave)

* Not defined, but normally MSB of previous character received.
Figure 32-4. SPI Transfer Format (NCPHA = 0, 8 bits per transfer)
1

SPCK cycle (for reference)

SPCK
(CPOL = 0)

SPCK
(CPOL = 1)

MOSI
(from master)

MISO
(from slave)

MSB

LSB

NSS
(to slave)

* Not defined but normally LSB of previous character transmitted.

598

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.7.3

Master Mode Operations
When configured in Master Mode, the SPI operates on the clock generated by the internal programmable baud rate generator. It fully controls the data transfers to and from the slave(s)
connected to the SPI bus. The SPI drives the chip select line to the slave and the serial clock
signal (SPCK).
The SPI features two holding registers, the Transmit Data Register and the Receive Data Register, and a single Shift Register. The holding registers maintain the data flow at a constant rate.
After enabling the SPI, a data transfer begins when the processor writes to the SPI_TDR (Transmit Data Register). The written data is immediately transferred in the Shift Register and transfer
on the SPI bus starts. While the data in the Shift Register is shifted on the MOSI line, the MISO
line is sampled and shifted in the Shift Register. Receiving data cannot occur without transmitting data. If receiving mode is not needed, for example when communicating with a slave
receiver only (such as an LCD), the receive status flags in the status register can be discarded.
Before writing the TDR, the PCS field in the SPI_MR register must be set in order to select a
slave.
After enabling the SPI, a data transfer begins when the processor writes to the SPI_TDR (Transmit Data Register). The written data is immediately transferred in the Shift Register and transfer
on the SPI bus starts. While the data in the Shift Register is shifted on the MOSI line, the MISO
line is sampled and shifted in the Shift Register. Transmission cannot occur without reception.
Before writing the TDR, the PCS field must be set in order to select a slave.
If new data is written in SPI_TDR during the transfer, it stays in it until the current transfer is
completed. Then, the received data is transferred from the Shift Register to SPI_RDR, the data
in SPI_TDR is loaded in the Shift Register and a new transfer starts.
The transfer of a data written in SPI_TDR in the Shift Register is indicated by the TDRE bit
(Transmit Data Register Empty) in the Status Register (SPI_SR). When new data is written in
SPI_TDR, this bit is cleared. The TDRE bit is used to trigger the Transmit DMA channel.
The end of transfer is indicated by the TXEMPTY flag in the SPI_SR register. If a transfer delay
(DLYBCT) is greater than 0 for the last transfer, TXEMPTY is set after the completion of said
delay. The master clock (MCK) can be switched off at this time.
The transfer of received data from the Shift Register in SPI_RDR is indicated by the RDRF bit
(Receive Data Register Full) in the Status Register (SPI_SR). When the received data is read,
the RDRF bit is cleared.
If the SPI_RDR (Receive Data Register) has not been read before new data is received, the
Overrun Error bit (OVRES) in SPI_SR is set. As long as this flag is set, data is loaded in
SPI_RDR. The user has to read the status register to clear the OVRES bit.
Figure 32-5, shows a block diagram of the SPI when operating in Master Mode. Figure 32-6 on
page 601 shows a flow chart describing how transfers are handled.

599
6430F–ATARM–21-Feb-12

32.7.3.1

Master Mode Block Diagram

Figure 32-5. Master Mode Block Diagram
SPI_CSR0..3
SCBR
Baud Rate Generator

MCK

SPCK

SPI
Clock
SPI_CSR0..3
BITS
NCPHA
CPOL
LSB

MISO

SPI_RDR

RDRF
OVRES

MSB

Shift Register

MOSI

SPI_TDR
TD

TDRE

SPI_CSR0..3
SPI_RDR
CSAAT

PCS

NPCS3

PCSDEC

SPI_MR
PCS
0

NPCS2

Current
Peripheral

NPCS1
SPI_TDR
NPCS0

PCS
1

MSTR

MODF

NPCS0
MODFDIS

600

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.7.3.2

Master Mode Flow Diagram

Figure 32-6. Master Mode Flow Diagram
SPI Enable

- NPCS defines the current Chip Select
- CSAAT, DLYBS, DLYBCT refer to the fields of the
Chip Select Register corresponding to the Current Chip Select
- When NPCS is 0xF, CSAAT is 0.
1

TDRE ?

0
1
CSAAT ?

PS ?

1
0

Fixed
peripheral

PS ?

Fixed
peripheral

Variable
peripheral

SPI_TDR(PCS)
= NPCS ?
no

NPCS = SPI_TDR(PCS)

NPCS = SPI_MR(PCS)

yes

SPI_MR(PCS)
= NPCS ?
no

NPCS = 0xF

Delay DLYBCS

NPCS = SPI_TDR(PCS)

NPCS = SPI_MR(PCS),
SPI_TDR(PCS)

Delay DLYBS

Serializer = SPI_TDR(TD)
TDRE = 1

Data Transfer

SPI_RDR(RD) = Serializer
RDRF = 1

Delay DLYBCT

0
TDRE ?

1
CSAAT ?

0
NPCS = 0xF

Delay DLYBCS

601
6430F–ATARM–21-Feb-12

Figure 32-7 shows Transmit Data Register Empty (TDRE), Receive Data Register (RDRF) and
Transmission Register Empty (TXEMPTY) status flags behavior within the SPI_SR (Status Register) during an 8-bit data transfer in fixed mode and no Peripheral Data Controller involved.
Figure 32-7.

Status Register Flags Behavior
1

SPCK
NPCS0
MOSI
(from master)

MSB

LSB

TDRE
RDR read
Write in
SPI_TDR
RDRF
MISO
(from slave)

MSB

LSB

TXEMPTY

shift register empty

32.7.3.3

Clock Generation
The SPI Baud rate clock is generated by dividing the Master Clock (MCK), by a value between 1
and 255.
This allows a maximum operating baud rate at up to Master Clock and a minimum operating
baud rate of MCK divided by 255.
Programming the SCBR field at 0 is forbidden. Triggering a transfer while SCBR is at 0 can lead
to unpredictable results.
At reset, SCBR is 0 and the user has to program it at a valid value before performing the first
transfer.
The divisor can be defined independently for each chip select, as it has to be programmed in the
SCBR field of the Chip Select Registers. This allows the SPI to automatically adapt the baud
rate for each interfaced peripheral without reprogramming.

32.7.3.4

Transfer Delays
Figure 32-8 shows a chip select transfer change and consecutive transfers on the same chip
select. Three delays can be programmed to modify the transfer waveforms:
• The delay between chip selects, programmable only once for all the chip selects by writing
the DLYBCS field in the Mode Register. Allows insertion of a delay between release of one
chip select and before assertion of a new one.

602

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• The delay before SPCK, independently programmable for each chip select by writing the field
DLYBS. Allows the start of SPCK to be delayed after the chip select has been asserted.
• The delay between consecutive transfers, independently programmable for each chip select
by writing the DLYBCT field. Allows insertion of a delay between two transfers occurring on
the same chip select
These delays allow the SPI to be adapted to the interfaced peripherals and their speed and bus
release time.
Figure 32-8. Programmable Delays

Chip Select 1

Chip Select 2

SPCK
DLYBCS

32.7.3.5

DLYBS

DLYBCT

Peripheral Selection
The serial peripherals are selected through the assertion of the NPCS0 to NPCS3 signals. By
default, all the NPCS signals are high before and after each transfer.
• Fixed Peripheral Select: SPI exchanges data with only one peripheral
Fixed Peripheral Select is activated by writing the PS bit to zero in SPI_MR (Mode Register). In
this case, the current peripheral is defined by the PCS field in SPI_MR and the PCS field in the
SPI_TDR has no effect.
• Variable Peripheral Select: Data can be exchanged with more than one peripheral without
having to reprogram the NPCS field in the SPI_MR register.
Variable Peripheral Select is activated by setting PS bit to one. The PCS field in SPI_TDR is
used to select the current peripheral. This means that the peripheral selection can be defined for
each new data. The value to write in the SPI_TDR register as the following format.
[xxxxxxx(7-bit) + LASTXFER(1-bit)(1)+ xxxx(4-bit) + PCS (4-bit) + DATA (8 to 16-bit)] with PCS
equals to the chip select to assert as defined in Section 32.8.4 (SPI Transmit Data Register) and
LASTXFER bit at 0 or 1 depending on CSAAT bit.
Note:

1. Optional.

CSAAT, LASTXFER and CSNAAT bits are discussed in Section 32.7.3.9 ”Peripheral Deselection with DMAC” .
If LASTXFER is used, the command must be issued before writing the last character. Instead of
LASTXFER, the user can use the SPIDIS command. After the end of the DMA transfer, wait for
the TXEMPTY flag, then write SPIDIS into the SPI_CR register (this will not change the configuration register values); the NPCS will be deactivated after the last character transfer. Then,
another DMA transfer can be started if the SPIEN was previously written in the SPI_CR register.

603
6430F–ATARM–21-Feb-12

32.7.3.6

SPI Direct Access Memory Controller (DMAC)
In both fixed and variable mode the Direct Memory Access Controller (DMAC) can be used to
reduce processor overhead.
The Fixed Peripheral Selection allows buffer transfers with a single peripheral. Using the DMAC
is an optimal means, as the size of the data transfer between the memory and the SPI is either 8
bits or 16 bits. However, changing the peripheral selection requires the Mode Register to be
reprogrammed.
The Variable Peripheral Selection allows buffer transfers with multiple peripherals without reprogramming the Mode Register. Data written in SPI_TDR is 32 bits wide and defines the real data
to be transmitted and the peripheral it is destined to. Using the DMAC in this mode requires 32bit wide buffers, with the data in the LSBs and the PCS and LASTXFER fields in the MSBs, however the SPI still controls the number of bits (8 to16) to be transferred through MISO and MOSI
lines with the chip select configuration registers. This is not the optimal means in term of memory size for the buffers, but it provides a very effective means to exchange data with several
peripherals without any intervention of the processor.

32.7.3.7

Peripheral Chip Select Decoding
The user can program the SPI to operate with up to 15 peripherals by decoding the four Chip
Select lines, NPCS0 to NPCS3 with 1 of up to 16 decoder/demultiplexer. This can be enabled by
writing the PCSDEC bit at 1 in the Mode Register (SPI_MR).
When operating without decoding, the SPI makes sure that in any case only one chip select line
is activated, i.e., one NPCS line driven low at a time. If two bits are defined low in a PCS field,
only the lowest numbered chip select is driven low.
When operating with decoding, the SPI directly outputs the value defined by the PCS field on
NPCS lines of either the Mode Register or the Transmit Data Register (depending on PS).
As the SPI sets a default value of 0xF on the chip select lines (i.e. all chip select lines at 1) when
not processing any transfer, only 15 peripherals can be decoded.
The SPI has only four Chip Select Registers, not 15. As a result, when decoding is activated,
each chip select defines the characteristics of up to four peripherals. As an example, SPI_CRS0
defines the characteristics of the externally decoded peripherals 0 to 3, corresponding to the
PCS values 0x0 to 0x3. Thus, the user has to make sure to connect compatible peripherals on
the decoded chip select lines 0 to 3, 4 to 7, 8 to 11 and 12 to 14. Figure 32-9 below shows such
an implementation.
If the CSAAT bit is used, with or without the DMAC, the Mode Fault detection for NPCS0 line
must be disabled. This is not needed for all other chip select lines since Mode Fault Detection is
only on NPCS0.

604

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 32-9. Chip Select Decoding Application Block Diagram: Single Master/Multiple Slave Implementation
SPCK
MISO
MOSI
SPCK MISO MOSI

SPCK MISO MOSI

Slave 0

Slave 1

Slave 14

NSS

SPI Master

NSS

NPCS0
NPCS1
NPCS2
NPCS3

1-of-n Decoder/Demultiplexer

32.7.3.8

Peripheral Deselection without DMAC
During a transfer of more than one data on a Chip Select without the DMAC, the SPI_TDR is
loaded by the processor, the flag TDRE rises as soon as the content of the SPI_TDR is transferred into the internal shift register. When this flag is detected high, the SPI_TDR can be
reloaded. If this reload by the processor occurs before the end of the current transfer and if the
next transfer is performed on the same chip select as the current transfer, the Chip Select is not
de-asserted between the two transfers. But depending on the application software handling the
SPI status register flags (by interrupt or polling method) or servicing other interrupts or other
tasks, the processor may not reload the SPI_TDR in time to keep the chip select active (low). A
null Delay Between Consecutive Transfer (DLYBCT) value in the SPI_CSR register, will give
even less time for the processor to reload the SPI_TDR. With some SPI slave peripherals,
requiring the chip select line to remain active (low) during a full set of transfers might lead to
communication errors.
To facilitate interfacing with such devices, the Chip Select Register [CSR0...CSR3] can be programmed with the CSAAT bit (Chip Select Active After Transfer) at 1. This allows the chip select
lines to remain in their current state (low = active) until transfer to another chip select is required.
Even if the SPI_TDR is not reloaded the chip select will remain active. To have the chip select
line to raise at the end of the transfer the Last transfer Bit (LASTXFER) in the SPI_MR register
must be set at 1 before writing the last data to transmit into the SPI_TDR.

605
6430F–ATARM–21-Feb-12

32.7.3.9

Peripheral Deselection with DMAC
When the Direct Memory Access Controller is used, the chip select line will remain low during
the whole transfer since the TDRE flag is managed by the DMAC itself. The reloading of the
SPI_TDR by the DMAC is done as soon as TDRE flag is set to one. In this case the use of
CSAAT bit might not be needed. However, it may happen that when other DMAC channels connected to other peripherals are in use as well, the SPI DMAC might be delayed by another
(DMAC with a higher priority on the bus). Having DMAC buffers in slower memories like flash
memory or SDRAM compared to fast internal SRAM, may lengthen the reload time of the
SPI_TDR by the DMAC as well. This means that the SPI_TDR might not be reloaded in time to
keep the chip select line low. In this case the chip select line may toggle between data transfer
and according to some SPI Slave devices, the communication might get lost. The use of the
CSAAT bit might be needed.
When the CSAAT bit is set at 0, the NPCS does not rise in all cases between two transfers on
the same peripheral. During a transfer on a Chip Select, the flag TDRE rises as soon as the content of the SPI_TDR is transferred into the internal shifter. When this flag is detected the
SPI_TDR can be reloaded. If this reload occurs before the end of the current transfer and if the
next transfer is performed on the same chip select as the current transfer, the Chip Select is not
de-asserted between the two transfers. This might lead to difficulties for interfacing with some
serial peripherals requiring the chip select to be de-asserted after each transfer. To facilitate
interfacing with such devices, the Chip Select Register can be programmed with the CSNAAT bit
(Chip Select Not Active After Transfer) at 1. This allows to de-assert systematically the chip
select lines during a time DLYBCS. (The value of the CSNAAT bit is taken into account only if
the CSAAT bit is set at 0 for the same Chip Select).
Figure 32-10 shows different peripheral deselection cases and the effect of the CSAAT and
CSNAAT bits.

606

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 32-10. Peripheral Deselection
CSAAT = 0 and CSNAAT = 0

TDRE

CSAAT = 1 and CSNAAT= 0 / 1

DLYBCT

NPCS[0..3]

DLYBCT

DLYBCS

DLYBCS
PCS = A

PCS = A

Write SPI_TDR

TDRE

DLYBCT

NPCS[0..3]

DLYBCT

DLYBCS

PCS=A

PCS = A

Write SPI_TDR

TDRE

DLYBCT

NPCS[0..3]

DLYBCT

DLYBCS

PCS = B

Write SPI_TDR

CSAAT = 0 and CSNAAT = 0

CSAAT = 0 and CSNAAT = 1
DLYBCT

DLYBCT
TDRE

NPCS[0..3]

A
DLYBCS

PCS = A

Write SPI_TDR

607
6430F–ATARM–21-Feb-12

32.7.3.10

Mode Fault Detection
A mode fault is detected when the SPI is programmed in Master Mode and a low level is driven
by an external master on the NPCS0/NSS signal. In this case, multi-master configuration,
NPCS0, MOSI, MISO and SPCK pins must be configured in open drain (through the PIO controller). When a mode fault is detected, the MODF bit in the SPI_SR is set until the SPI_SR is read
and the SPI is automatically disabled until re-enabled by writing the SPIEN bit in the SPI_CR
(Control Register) at 1.
By default, the Mode Fault detection circuitry is enabled. The user can disable Mode Fault
detection by setting the MODFDIS bit in the SPI Mode Register (SPI_MR).

32.7.4

SPI Slave Mode
When operating in Slave Mode, the SPI processes data bits on the clock provided on the SPI
clock pin (SPCK).
The SPI waits for NSS to go active before receiving the serial clock from an external master.
When NSS falls, the clock is validated on the serializer, which processes the number of bits
defined by the BITS field of the Chip Select Register 0 (SPI_CSR0). These bits are processed
following a phase and a polarity defined respectively by the NCPHA and CPOL bits of the
SPI_CSR0. Note that BITS, CPOL and NCPHA of the other Chip Select Registers have no
effect when the SPI is programmed in Slave Mode.
The bits are shifted out on the MISO line and sampled on the MOSI line.
(For more information on BITS field, see also, the
“SPI Chip Select Register” on page 622.)

(Note:)

below the register table; Section 32.8.9

When all the bits are processed, the received data is transferred in the Receive Data Register
and the RDRF bit rises. If the SPI_RDR (Receive Data Register) has not been read before new
data is received, the Overrun Error bit (OVRES) in SPI_SR is set. As long as this flag is set, data
is loaded in SPI_RDR. The user has to read the status register to clear the OVRES bit.
When a transfer starts, the data shifted out is the data present in the Shift Register. If no data
has been written in the Transmit Data Register (SPI_TDR), the last data received is transferred.
If no data has been received since the last reset, all bits are transmitted low, as the Shift Register resets at 0.
When a first data is written in SPI_TDR, it is transferred immediately in the Shift Register and the
TDRE bit rises. If new data is written, it remains in SPI_TDR until a transfer occurs, i.e. NSS falls
and there is a valid clock on the SPCK pin. When the transfer occurs, the last data written in
SPI_TDR is transferred in the Shift Register and the TDRE bit rises. This enables frequent
updates of critical variables with single transfers.
Then, a new data is loaded in the Shift Register from the Transmit Data Register. In case no
character is ready to be transmitted, i.e. no character has been written in SPI_TDR since the last
load from SPI_TDR to the Shift Register, the Shift Register is not modified and the last received
character is retransmitted. In this case the Underrun Error Status Flag (UNDES) is set in the
SPI_SR.

Figure 32-11 shows a block diagram of the SPI when operating in Slave Mode.

608

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 32-11. Slave Mode Functional Bloc Diagram
SPCK
NSS

SPI
Clock
SPIEN
SPIENS
SPIDIS
SPI_CSR0
BITS
NCPHA
CPOL

MOSI

LSB

SPI_RDR

RDRF
OVRES

MSB

Shift Register

MISO

SPI_TDR
TD

TDRE

609
6430F–ATARM–21-Feb-12

32.7.5

Write Protected Registers
To prevent any single software error that may corrupt SPI behavior, the registers listed below
can be write-protected by setting the WPEN bit in the SPI Write Protection Mode Register
(SPI_WPMR).
If a write access in a write-protected register is detected, then the WPVS flag in the SPI Write
Protection Status Register (SPI_WPSR) is set and the field WPVSRC indicates in which register
the write access has been attempted.
The WPVS flag is automatically reset after reading the SPI Write Protection Status Register
(SPI_WPSR).
List of the write-protected registers:
Section 32.8.2 ”SPI Mode Register”
Section 32.8.9 ”SPI Chip Select Register”

610

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.8

Serial Peripheral Interface (SPI) User Interface

Table 32-5.

Offset

Name

Access

Reset

0x00

Control Register

SPI_CR

Write-only

---

0x04

Mode Register

SPI_MR

Read-write

0x0

0x08

Receive Data Register

SPI_RDR

Read-only

0x0

0x0C

Transmit Data Register

SPI_TDR

Write-only

---

0x10

Status Register

SPI_SR

Read-only

0x000000F0

0x14

Interrupt Enable Register

SPI_IER

Write-only

---

0x18

Interrupt Disable Register

SPI_IDR

Write-only

---

0x1C

Interrupt Mask Register

SPI_IMR

Read-only

0x0

0x20 - 0x2C

Reserved

0x30

Chip Select Register 0

SPI_CSR0

Read-write

0x0

0x34

Chip Select Register 1

SPI_CSR1

Read-write

0x0

0x38

Chip Select Register 2

SPI_CSR2

Read-write

0x0

0x3C

Chip Select Register 3

SPI_CSR3

Read-write

0x0

–

0x4C - 0xE0

Reserved

0xE4

Write Protection Control Register

SPI_WPMR

Read-write

0x0

0xE8

Write Protection Status Register

SPI_WPSR

Read-only

0x0

0x00E8 - 0x00F8

Reserved

–

0x00FC

Reserved

–

611
6430F–ATARM–21-Feb-12

32.8.1
Name:

SPI Control Register
SPI_CR

Address:

0x40008000

Access:

Write-only

–

LASTXFER

–

SWRST

–

SPIDIS

SPIEN

• SPIEN: SPI Enable
0 = No effect.
1 = Enables the SPI to transfer and receive data.
• SPIDIS: SPI Disable
0 = No effect.
1 = Disables the SPI.
As soon as SPIDIS is set, SPI finishes its transfer.
All pins are set in input mode and no data is received or transmitted.
If a transfer is in progress, the transfer is finished before the SPI is disabled.
If both SPIEN and SPIDIS are equal to one when the control register is written, the SPI is disabled.
• SWRST: SPI Software Reset
0 = No effect.
1 = Reset the SPI. A software-triggered hardware reset of the SPI interface is performed.
The SPI is in slave mode after software reset.
DMAC channels are not affected by software reset.
• LASTXFER: Last Transfer
0 = No effect.
1 = The current NPCS will be deasserted after the character written in TD has been transferred. When CSAAT is set, this
allows to close the communication with the current serial peripheral by raising the corresponding NPCS line as soon as TD
transfer has completed.
Refer to Section 32.7.3.5 ”Peripheral Selection” for more details.

612

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.8.2
Name:

SPI Mode Register
SPI_MR

Address:

0x40008004

Access:

Read-write

DLYBCS
23

–

PCS

LLB

–

WDRBT

MODFDIS

–

PCSDEC

MSTR

This register can only be written if the WPEN bit is cleared in ”SPI Write Protection Mode Register”.
• MSTR: Master/Slave Mode
0 = SPI is in Slave mode.
1 = SPI is in Master mode.
• PS: Peripheral Select
0 = Fixed Peripheral Select.
1 = Variable Peripheral Select.
• PCSDEC: Chip Select Decode
0 = The chip selects are directly connected to a peripheral device.
1 = The four chip select lines are connected to a 4- to 16-bit decoder.
When PCSDEC equals one, up to 15 Chip Select signals can be generated with the four lines using an external 4- to 16-bit
decoder. The Chip Select Registers define the characteristics of the 15 chip selects according to the following rules:
SPI_CSR0 defines peripheral chip select signals 0 to 3.
SPI_CSR1 defines peripheral chip select signals 4 to 7.
SPI_CSR2 defines peripheral chip select signals 8 to 11.
SPI_CSR3 defines peripheral chip select signals 12 to 14.
• MODFDIS: Mode Fault Detection
0 = Mode fault detection is enabled.
1 = Mode fault detection is disabled.
• WDRBT: Wait Data Read Before Transfer
0 = No Effect. In master mode, a transfer can be initiated whatever the state of the Receive Data Register is.
1 = In Master Mode, a transfer can start only if the Receive Data Register is empty, i.e. does not contain any unread data.
This mode prevents overrun error in reception.
• LLB: Local Loopback Enable
613
6430F–ATARM–21-Feb-12

0 = Local loopback path disabled.
1 = Local loopback path enabled
LLB controls the local loopback on the data serializer for testing in Master Mode only. (MISO is internally connected on
MOSI.)
• PCS: Peripheral Chip Select
This field is only used if Fixed Peripheral Select is active (PS = 0).
If PCSDEC = 0:
PCS = xxx0

NPCS[3:0] = 1110

PCS = xx01

NPCS[3:0] = 1101

PCS = x011

NPCS[3:0] = 1011

PCS = 0111

NPCS[3:0] = 0111

PCS = 1111

forbidden (no peripheral is selected)

(x = don’t care)
If PCSDEC = 1:
NPCS[3:0] output signals = PCS.
• DLYBCS: Delay Between Chip Selects
This field defines the delay from NPCS inactive to the activation of another NPCS. The DLYBCS time guarantees non-overlapping chip selects and solves bus contentions in case of peripherals having long data float times.
If DLYBCS is less than or equal to six, six MCK periods will be inserted by default.
Otherwise, the following equation determines the delay:
Delay Between Chip Selects = DLYBCS
----------------------MCK

614

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.8.3
Name:

SPI Receive Data Register
SPI_RDR

Address:

0x40008008

Access:

Read-only

–

PCS
11

RD
7

• RD: Receive Data
Data received by the SPI Interface is stored in this register right-justified. Unused bits read zero.
• PCS: Peripheral Chip Select
In Master Mode only, these bits indicate the value on the NPCS pins at the end of a transfer. Otherwise, these bits read
zero.
Note:

When using variable peripheral select mode (PS = 1 in SPI_MR) it is mandatory to also set the WDRBT field to 1 if the
SPI_RDR PCS field is to be processed.

615
6430F–ATARM–21-Feb-12

32.8.4
Name:

SPI Transmit Data Register
SPI_TDR

Address:

0x4000800C

Access:

Write-only

–

LASTXFER

–

PCS
11

TD
7

• TD: Transmit Data
Data to be transmitted by the SPI Interface is stored in this register. Information to be transmitted must be written to the
transmit data register in a right-justified format.
• PCS: Peripheral Chip Select
This field is only used if Variable Peripheral Select is active (PS = 1).
If PCSDEC = 0:
PCS = xxx0

NPCS[3:0] = 1110

PCS = xx01

NPCS[3:0] = 1101

PCS = x011

NPCS[3:0] = 1011

PCS = 0111

NPCS[3:0] = 0111

PCS = 1111

forbidden (no peripheral is selected)

(x = don’t care)
If PCSDEC = 1:
NPCS[3:0] output signals = PCS
• LASTXFER: Last Transfer
0 = No effect.
1 = The current NPCS will be deasserted after the character written in TD has been transferred. When CSAAT is set, this
allows to close the communication with the current serial peripheral by raising the corresponding NPCS line as soon as TD
transfer has completed.
This field is only used if Variable Peripheral Select is active (PS = 1).

616

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.8.5
Name:

SPI Status Register
SPI_SR

Address:

0x40008010

Access:

Read-only

–

SPIENS

–

UNDES

TXEMPTY

NSSR

–

OVRES

MODF

TDRE

RDRF

• RDRF: Receive Data Register Full
0 = No data has been received since the last read of SPI_RDR
1 = Data has been received and the received data has been transferred from the serializer to SPI_RDR since the last read
of SPI_RDR.
• TDRE: Transmit Data Register Empty
0 = Data has been written to SPI_TDR and not yet transferred to the serializer.
1 = The last data written in the Transmit Data Register has been transferred to the serializer.
TDRE equals zero when the SPI is disabled or at reset. The SPI enable command sets this bit to one.
• MODF: Mode Fault Error
0 = No Mode Fault has been detected since the last read of SPI_SR.
1 = A Mode Fault occurred since the last read of the SPI_SR.
• OVRES: Overrun Error Status
0 = No overrun has been detected since the last read of SPI_SR.
1 = An overrun has occurred since the last read of SPI_SR.
An overrun occurs when SPI_RDR is loaded at least twice from the serializer since the last read of the SPI_RDR.
• NSSR: NSS Rising
0 = No rising edge detected on NSS pin since last read.
1 = A rising edge occurred on NSS pin since last read.
• TXEMPTY: Transmission Registers Empty
0 = As soon as data is written in SPI_TDR.
1 = SPI_TDR and internal shifter are empty. If a transfer delay has been defined, TXEMPTY is set after the completion of
such delay.
• UNDES: Underrun Error Status (Slave Mode Only)
0 = No underrun has been detected since the last read of SPI_SR.

617
6430F–ATARM–21-Feb-12

1 = A transfer begins whereas no data has been loaded in the Transmit Data Register.
• SPIENS: SPI Enable Status
0 = SPI is disabled.
1 = SPI is enabled.

618

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.8.6
Name:

SPI Interrupt Enable Register
SPI_IER

Address:

0x40008014

Access:

Write-only

–

UNDES

TXEMPTY

NSSR

–

OVRES

MODF

TDRE

RDRF

0 = No effect.
1 = Enables the corresponding interrupt.
• RDRF: Receive Data Register Full Interrupt Enable
• TDRE: SPI Transmit Data Register Empty Interrupt Enable
• MODF: Mode Fault Error Interrupt Enable
• OVRES: Overrun Error Interrupt Enable
• NSSR: NSS Rising Interrupt Enable
• TXEMPTY: Transmission Registers Empty Enable
• UNDES: Underrun Error Interrupt Enable

619
6430F–ATARM–21-Feb-12

32.8.7
Name:

SPI Interrupt Disable Register
SPI_IDR

Address:

0x40008018

Access:

Write-only

–

UNDES

TXEMPTY

NSSR

–

OVRES

MODF

TDRE

RDRF

0 = No effect.
1 = Disables the corresponding interrupt.
• RDRF: Receive Data Register Full Interrupt Disable
• TDRE: SPI Transmit Data Register Empty Interrupt Disable
• MODF: Mode Fault Error Interrupt Disable
• OVRES: Overrun Error Interrupt Disable
• NSSR: NSS Rising Interrupt Disable
• TXEMPTY: Transmission Registers Empty Disable
• UNDES: Underrun Error Interrupt Disable

620

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.8.8
Name:

SPI Interrupt Mask Register
SPI_IMR

Address:

0x4000801C

Access:

Read-only

–

UNDES

TXEMPTY

NSSR

–

OVRES

MODF

TDRE

RDRF

0 = The corresponding interrupt is not enabled.
1 = The corresponding interrupt is enabled.
• RDRF: Receive Data Register Full Interrupt Mask
• TDRE: SPI Transmit Data Register Empty Interrupt Mask
• MODF: Mode Fault Error Interrupt Mask
• OVRES: Overrun Error Interrupt Mask
• NSSR: NSS Rising Interrupt Mask
• TXEMPTY: Transmission Registers Empty Mask
• UNDES: Underrun Error Interrupt Mask

621
6430F–ATARM–21-Feb-12

32.8.9
Name:

SPI Chip Select Register
SPI_CSRx[x=0..3]

Address:

0x40008030

Access:

Read/Write

DLYBCT
23

DLYBS
15

SCBR
7

BITS

CSAAT

CSNAAT

NCPHA

CPOL

This register can only be written if the WPEN bit is cleared in ”SPI Write Protection Mode Register”.
Note:

SPI_CSRx registers must be written even if the user wants to use the defaults. The BITS field will not be updated with the translated value unless the register is written.

• CPOL: Clock Polarity
0 = The inactive state value of SPCK is logic level zero.
1 = The inactive state value of SPCK is logic level one.
CPOL is used to determine the inactive state value of the serial clock (SPCK). It is used with NCPHA to produce the
required clock/data relationship between master and slave devices.
• NCPHA: Clock Phase
0 = Data is changed on the leading edge of SPCK and captured on the following edge of SPCK.
1 = Data is captured on the leading edge of SPCK and changed on the following edge of SPCK.
NCPHA determines which edge of SPCK causes data to change and which edge causes data to be captured. NCPHA is
used with CPOL to produce the required clock/data relationship between master and slave devices.
• CSNAAT: Chip Select Not Active After Transfer (Ignored if CSAAT = 1)
0 = The Peripheral Chip Select does not rise between two transfers if the SPI_TDR is reloaded before the end of the first
transfer and if the two transfers occur on the same Chip Select.
1 = The Peripheral Chip Select rises systematically after each transfer performed on the same slave. It remains active after
the end of transfer for a minimal duration of:
DLYBCT
– ----------------------- (if DLYBCT field is different from 0)
MCK
DLYBCT + 1
– --------------------------------- (if DLYBCT field equals 0)
MCK

• CSAAT: Chip Select Active After Transfer
0 = The Peripheral Chip Select Line rises as soon as the last transfer is achieved.
1 = The Peripheral Chip Select does not rise after the last transfer is achieved. It remains active until a new transfer is
requested on a different chip select.

622

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series

• BITS: Bits Per Transfer
(See the (Note:) below the register table; Section 32.8.9 “SPI Chip Select Register” on page 622.)
The BITS field determines the number of data bits transferred. Reserved values should not be used.
Value
0
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

Name
8_BIT
9_BIT
10_BIT
11_BIT
12_BIT
13_BIT
14_BIT
15_BIT
16_BIT
–
–
–
–
–
–
–

Description
8 bits for transfer
9 bits for transfer
10 bits for transfer
11 bits for transfer
12 bits for transfer
13 bits for transfer
14 bits for transfer
15 bits for transfer
16 bits for transfer
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved
Reserved

• SCBR: Serial Clock Baud Rate
In Master Mode, the SPI Interface uses a modulus counter to derive the SPCK baud rate from the Master Clock MCK. The
Baud rate is selected by writing a value from 1 to 255 in the SCBR field. The following equations determine the SPCK baud
rate:
MCK
SPCK Baudrate = --------------SCBR
Programming the SCBR field at 0 is forbidden. Triggering a transfer while SCBR is at 0 can lead to unpredictable results.
At reset, SCBR is 0 and the user has to program it at a valid value before performing the first transfer.
Note:

If one of the SCBR fields inSPI_CSRx is set to 1, the other SCBR fields in SPI_CSRx must be set to 1 as well, if they are
required to process transfers. If they are not used to transfer data, they can be set at any value.

• DLYBS: Delay Before SPCK
This field defines the delay from NPCS valid to the first valid SPCK transition.
When DLYBS equals zero, the NPCS valid to SPCK transition is 1/2 the SPCK clock period.
Otherwise, the following equations determine the delay:

DLYBS
Delay Before SPCK = ------------------MCK

623
6430F–ATARM–21-Feb-12

• DLYBCT: Delay Between Consecutive Transfers
This field defines the delay between two consecutive transfers with the same peripheral without removing the chip select.
The delay is always inserted after each transfer and before removing the chip select if needed.
When DLYBCT equals zero, no delay between consecutive transfers is inserted and the clock keeps its duty cycle over the
character transfers.
Otherwise, the following equation determines the delay:
32 × DLYBCT
Delay Between Consecutive Transfers = ------------------------------------MCK

624

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
32.8.10
Name:
Address:

SPI Write Protection Mode Register
SPI_WPMR
0x400080E4

Access:
31

Read-write
30

WPKEY
23

WPKEY
15

WPKEY
7

WPEN

• WPEN: Write Protection Enable
0: The Write Protection is Disabled
1: The Write Protection is Enabled
• WPKEY: Write Protection Key Password
If a value is written in WPEN, the value is taken into account only if WPKEY is written with “SPI” (SPI written in ASCII Code,
ie 0x535049 in hexadecimal).
List of the write-protected registers:
Section 32.8.2 ”SPI Mode Register”
Section 32.8.9 ”SPI Chip Select Register”

625
6430F–ATARM–21-Feb-12

32.8.11
Name:

SPI Write Protection Status Register
SPI_WPSR

Address:

0x400080E8

Access:

Read-only

–

WPVSRC
7

–

WPVS

• WPVS: Write Protection Violation Status
0 = No Write Protect Violation has occurred since the last read of the SPI_WPSR register.
1 = A Write Protect Violation has occurred since the last read of the SPI_WPSR register. If this violation is an unauthorized
attempt to write a protected register, the associated violation is reported into field WPVSRC.
• WPVSRC: Write Protection Violation Source
This Field indicates the APB Offset of the register concerned by the violation (SPI_MR or SPI_CSRx)

626

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33. Two-wire Interface (TWI)
33.1

Description
The Atmel Two-wire Interface (TWI) interconnects components on a unique two-wire bus, made
up of one clock line and one data line with speeds of up to 400 Kbits per second, based on a
byte-oriented transfer format. It can be used with any Atmel Two-wire Interface bus Serial
EEPROM and I²C compatible device such as Real Time Clock (RTC), Dot Matrix/Graphic LCD
Controllers and Temperature Sensor, to name but a few. The TWI is programmable as a master
or a slave with sequential or single-byte access. Multiple master capability is supported. 20
Arbitration of the bus is performed internally and puts the TWI in slave mode automatically if the
bus arbitration is lost.
A configurable baud rate generator permits the output data rate to be adapted to a wide range of
core clock frequencies.
Below, Table 33-1 lists the compatibility level of the Atmel Two-wire Interface in Master Mode and
a full I2C compatible device.
Atmel TWI compatibility with I2C Standard

Table 33-1.
I2C Standard

Atmel TWI

Standard Mode Speed (100 KHz)

Supported

Fast Mode Speed (400 KHz)

Supported

7 or 10 bits Slave Addressing

Supported

(1)

START BYTE

Not Supported

Repeated Start (Sr) Condition

Supported

ACK and NACK Management

Supported

Slope control and input filtering (Fast mode)

Not Supported

Clock stretching

Supported

Multi Master Capability

Supported

Note:

33.2

1. START + b000000001 + Ack + Sr

Embedded Characteristics
• Compatible with Atmel Two-wire Interface Serial Memory and I²C Compatible Devices(1)
• One, Two or Three Bytes for Slave Address
• Sequential Read-write Operations
• Master, Multi-master and Slave Mode Operation
• Bit Rate: Up to 400 Kbits
• General Call Supported in Slave mode
• SMBUS Quick Command Supported in Master Mode
• Connection to Peripheral DMA Controller (PDC) Channel Capabilities Optimizes Data
Transfers in Master Mode Only
– One Channel for the Receiver, One Channel for the Transmitter
– Next Buffer Support
Note:

1. See Table 33-1 for details on compatibility with I²C Standard.

627
6430F–ATARM–21-Feb-12

33.3

List of Abbreviations
Table 33-2.

33.4

Abbreviations

Abbreviation

Description

TWI

Two-wire Interface

Acknowledge

Non Acknowledge

Stop

Start

Repeated Start

SADR

Slave Address

ADR

Any address except SADR

Read

Write

Block Diagram
Figure 33-1. Block Diagram
APB Bridge

TWCK
PIO

PMC

MCK

TWD

Two-wire
Interface

TWI
Interrupt

628

Interrupt
Controller

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33.5

Application Block Diagram
Figure 33-2. Application Block Diagram
VDD
Rp

Host with
TWI
Interface

TWD
TWCK

Atmel TWI
Serial EEPROM
Slave 1

I²C RTC

I²C LCD
Controller

I²C Temp.
Sensor

Slave 2

Slave 3

Slave 4

Rp: Pull up value as given by the I²C Standard

33.5.1

I/O Lines Description

Table 33-3.

I/O Lines Description

Pin Name

Pin Description

TWD

Two-wire Serial Data

Input/Output

TWCK

Two-wire Serial Clock

Input/Output

33.6
33.6.1

Type

Product Dependencies
I/O Lines
Both TWD and TWCK are bidirectional lines, connected to a positive supply voltage via a current
source or pull-up resistor (see Figure 33-2 on page 629). When the bus is free, both lines are
high. The output stages of devices connected to the bus must have an open-drain or open-collector to perform the wired-AND function.
TWD and TWCK pins may be multiplexed with PIO lines. To enable the TWI, the programmer
must perform the following step:
• Program the PIO controller to dedicate TWD and TWCK as peripheral lines.
The user must not program TWD and TWCK as open-drain. It is already done by the hardware.
Table 33-4.

I/O Lines

Instance

Signal

I/O Line

Peripheral

TWI0

TWCK0

PA10

TWI0

TWD0

PA9

TWI1

TWCK1

PA25

TWI1

TWD1

PA24

629
6430F–ATARM–21-Feb-12

33.6.2

Power Management
• Enable the peripheral clock.
The TWI interface may be clocked through the Power Management Controller (PMC), thus the
programmer must first configure the PMC to enable the TWI clock.

33.6.3

Interrupt
The TWI interface has an interrupt line connected to the Interrupt Controller. In order to handle
interrupts, the Interrupt Controller must be programmed before configuring the TWI.

33.7
33.7.1

Functional Description
Transfer Format
The data put on the TWD line must be 8 bits long. Data is transferred MSB first; each byte must
be followed by an acknowledgement. The number of bytes per transfer is unlimited (see Figure
33-4).
Each transfer begins with a START condition and terminates with a STOP condition (see Figure
33-3).
• A high-to-low transition on the TWD line while TWCK is high defines the START condition.
• A low-to-high transition on the TWD line while TWCK is high defines a STOP condition.
Figure 33-3.

START and STOP Conditions
TWD

TWCK
Start

Stop

Figure 33-4. Transfer Format

TWD

TWCK

Start

33.7.2

Address

R/W

Ack

Data

Ack

Data

Ack

Stop

Modes of Operation
The TWI has different modes of operations:
• Master transmitter mode
• Master receiver mode
• Multi-master transmitter mode
• Multi-master receiver mode
• Slave transmitter mode
• Slave receiver mode

630

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
These modes are described in the following chapters.

33.8

Master Mode

33.8.1

Definition
The Master is the device that starts a transfer, generates a clock and stops it.

33.8.2

Application Block Diagram
Figure 33-5. Master Mode Typical Application Block Diagram
VDD
Rp

Host with
TWI
Interface

TWD
TWCK

Atmel TWI
Serial EEPROM
Slave 1

I²C RTC

I²C LCD
Controller

I²C Temp.
Sensor

Slave 2

Slave 3

Slave 4

Rp: Pull up value as given by the I²C Standard

33.8.3

Programming Master Mode
The following registers have to be programmed before entering Master mode:
1. DADR (+ IADRSZ + IADR if a 10 bit device is addressed): The device address is used
to access slave devices in read or write mode.
2. CKDIV + CHDIV + CLDIV: Clock Waveform.
3. SVDIS: Disable the slave mode.
4. MSEN: Enable the master mode.

33.8.4

Master Transmitter Mode
After the master initiates a Start condition when writing into the Transmit Holding Register,
TWI_THR, it sends a 7-bit slave address, configured in the Master Mode register (DADR in
TWI_MMR), to notify the slave device. The bit following the slave address indicates the transfer
direction, 0 in this case (MREAD = 0 in TWI_MMR).
The TWI transfers require the slave to acknowledge each received byte. During the acknowledge clock pulse (9th pulse), the master releases the data line (HIGH), enabling the slave to pull
it down in order to generate the acknowledge. The master polls the data line during this clock
pulse and sets the Not Acknowledge bit (NACK) in the status register if the slave does not
acknowledge the byte. As with the other status bits, an interrupt can be generated if enabled in
the interrupt enable register (TWI_IER). If the slave acknowledges the byte, the data written in
the TWI_THR, is then shifted in the internal shifter and transferred. When an acknowledge is
detected, the TXRDY bit is set until a new write in the TWI_THR.
TXRDY is used as Transmit Ready for the PDC transmit channel.

631
6430F–ATARM–21-Feb-12

While no new data is written in the TWI_THR, the Serial Clock Line is tied low. When new data is
written in the TWI_THR, the SCL is released and the data is sent. To generate a STOP event,
the STOP command must be performed by writing in the STOP field of TWI_CR.
After a Master Write transfer, the Serial Clock line is stretched (tied low) while no new data is
written in the TWI_THR or until a STOP command is performed.
See Figure 33-6, Figure 33-7, and Figure 33-8.
Figure 33-6. Master Write with One Data Byte
STOP Command sent (write in TWI_CR)

TWD

DADR

DATA

TXCOMP

TXRDY
Write THR (DATA)

Figure 33-7. Master Write with Multiple Data Bytes
STOP command performed
(by writing in the TWI_CR)

TWD

DADR

DATA n

DATA n+1

DATA n+2

TWCK

TXCOMP

TXRDY
Write THR (Data n)
Write THR (Data n+1)

632

Write THR (Data n+2)
Last data sent

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 33-8. Master Write with One Byte Internal Address and Multiple Data Bytes
STOP command performed
(by writing in the TWI_CR)

TWD

DADR

IADR

DATA n

DATA n+1

DATA n+2

TWCK

TXCOMP

TXRDY
Write THR (Data n)
Write THR (Data n+1)

33.8.5

Write THR (Data n+2)
Last data sent

Master Receiver Mode
The read sequence begins by setting the START bit. After the start condition has been sent, the
master sends a 7-bit slave address to notify the slave device. The bit following the slave address
indicates the transfer direction, 1 in this case (MREAD = 1 in TWI_MMR). During the acknowledge clock pulse (9th pulse), the master releases the data line (HIGH), enabling the slave to pull
it down in order to generate the acknowledge. The master polls the data line during this clock
pulse and sets the NACK bit in the status register if the slave does not acknowledge the byte.
If an acknowledge is received, the master is then ready to receive data from the slave. After data
has been received, the master sends an acknowledge condition to notify the slave that the data
has been received except for the last data, after the stop condition. See Figure 33-9. When the
RXRDY bit is set in the status register, a character has been received in the receive-holding register (TWI_RHR). The RXRDY bit is reset when reading the TWI_RHR.
When a single data byte read is performed, with or without internal address (IADR), the START
and STOP bits must be set at the same time. See Figure 33-9. When a multiple data byte read is
performed, with or without internal address (IADR), the STOP bit must be set after the next-tolast data received. See Figure 33-10. For Internal Address usage see Section 33.8.6.

Figure 33-9. Master Read with One Data Byte
TWD

DADR

DATA

TXCOMP
Write START &
STOP Bit
RXRDY
Read RHR

633
6430F–ATARM–21-Feb-12

Figure 33-10. Master Read with Multiple Data Bytes
TWD

DADR

DATA n

DATA (n+1)

DATA (n+m)-1

DATA (n+m)

TXCOMP
Write START Bit
RXRDY
Read RHR
DATA n

Read RHR
DATA (n+1)

Read RHR
DATA (n+m)-1

Read RHR
DATA (n+m)

Write STOP Bit
after next-to-last data read

RXRDY is used as Receive Ready for the PDC receive channel.

33.8.6

33.8.6.1

Internal Address
The TWI interface can perform various transfer formats: Transfers with 7-bit slave address
devices and 10-bit slave address devices.
7-bit Slave Addressing
When Addressing 7-bit slave devices, the internal address bytes are used to perform random
address (read or write) accesses to reach one or more data bytes, within a memory page location in a serial memory, for example. When performing read operations with an internal address,
the TWI performs a write operation to set the internal address into the slave device, and then
switch to Master Receiver mode. Note that the second start condition (after sending the IADR) is
sometimes called “repeated start” (Sr) in I2C fully-compatible devices. See Figure 33-12. See
Figure 33-11 and Figure 33-13 for Master Write operation with internal address.
The three internal address bytes are configurable through the Master Mode register
(TWI_MMR).
If the slave device supports only a 7-bit address, i.e. no internal address, IADRSZ must be set to
0.
In the figures below the following abbreviations are used:

634

•S

Start

• Sr

Repeated Start

•P

Stop

•W

Write

•R

Read

•A

Acknowledge

•N

Not Acknowledge

• DADR

Device Address

• IADR

Internal Address

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 33-11. Master Write with One, Two or Three Bytes Internal Address and One Data Byte
Three bytes internal address
S

TWD

DADR

IADR(23:16)

IADR(15:8)

IADR(7:0)

IADR(15:8)

IADR(7:0)

DATA

IADR(7:0)

DATA

Two bytes internal address
S

TWD

DADR

One byte internal address
S

TWD

DADR

Figure 33-12. Master Read with One, Two or Three Bytes Internal Address and One Data Byte
Three bytes internal address
S

TWD

DADR

IADR(23:16)

IADR(15:8)

IADR(7:0)

DADR

DATA

Two bytes internal address
S

TWD

DADR

IADR(15:8)

IADR(7:0)

DADR

DATA

One byte internal address
TWD

33.8.6.2

DADR

DATA

10-bit Slave Addressing
For a slave address higher than 7 bits, the user must configure the address size (IADRSZ) and
set the other slave address bits in the internal address register (TWI_IADR). The two remaining
Internal address bytes, IADR[15:8] and IADR[23:16] can be used the same as in 7-bit Slave
Addressing.
Example: Address a 10-bit device (10-bit device address is b1 b2 b3 b4 b5 b6 b7 b8 b9 b10)
1. Program IADRSZ = 1,
2. Program DADR with 1 1 1 1 0 b1 b2 (b1 is the MSB of the 10-bit address, b2, etc.)
3. Program TWI_IADR with b3 b4 b5 b6 b7 b8 b9 b10 (b10 is the LSB of the 10-bit
address)
Figure 33-13 below shows a byte write to an Atmel AT24LC512 EEPROM. This demonstrates
the use of internal addresses to access the device.
Figure 33-13. Internal Address Usage
S
T
A
R
T

Device
Address

W
R
I
T
E

FIRST
WORD ADDRESS

SECOND
WORD ADDRESS

S
T
O
P

DATA

0
M
S
B

LR A
S / C
BW K

M
S
B

A
C
K

LA
SC
BK

A
C
K

635
6430F–ATARM–21-Feb-12

33.8.7

Using the Peripheral DMA Controller (PDC)
The use of the PDC significantly reduces the CPU load.
To assure correct implementation, respect the following programming sequences:

33.8.7.1

Data Transmit with the PDC
1. Initialize the transmit PDC (memory pointers, transfer size).
2. Configure the master mode.
3. Start the transfer by setting the PDC TXTEN bit.
4. Wait for the PDC ENDTX Flag either by using the polling method or ENDTX interrupt.
5. Disable the PDC by setting the PDC TXDIS bit.

33.8.7.2

Data Receive with the PDC
1. Initialize the receive PDC (memory pointers, transfer size - 1).
2. Configure the master mode (DADR, CKDIV, etc.).
3. Start the transfer by setting the PDC RXTEN bit.
4. Wait for the PDC ENDRX Flag either by using polling method or ENDRX interrupt.
5. Disable the PDC by setting the PDC RXDIS bit.

33.8.8

SMBUS Quick Command (Master Mode Only)
The TWI interface can perform a Quick Command:
1. Configure the master mode (DADR, CKDIV, etc.).
2. Write the MREAD bit in the TWI_MMR register at the value of the one-bit command to
be sent.
3. Start the transfer by setting the QUICK bit in the TWI_CR.
Figure 33-14. SMBUS Quick Command
TWD

DADR

R/W

TXCOMP

TXRDY
Write QUICK command in TWI_CR

33.8.9

636

Read-write Flowcharts
The following flowcharts shown in Figure 33-16 on page 638, Figure 33-17 on page 639, Figure
33-18 on page 640, Figure 33-19 on page 641 and Figure 33-20 on page 642 give examples for
read and write operations. A polling or interrupt method can be used to check the status bits.
The interrupt method requires that the interrupt enable register (TWI_IER) be configured first.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 33-15. TWI Write Operation with Single Data Byte without Internal Address
BEGIN

Set TWI clock
(CLDIV, CHDIV, CKDIV) in TWI_CWGR
(Needed only once)

Set the Control register:
- Master enable
TWI_CR = MSEN + SVDIS

Set the Master Mode register:
- Device slave address (DADR)
- Transfer direction bit
Write ==> bit MREAD = 0

Load Transmit register
TWI_THR = Data to send

Write STOP Command
TWI_CR = STOP

Read Status register

No
TXRDY = 1?
Yes
Read Status register

No
TXCOMP = 1?
Yes
Transfer finished

637
6430F–ATARM–21-Feb-12

Figure 33-16. TWI Write Operation with Single Data Byte and Internal Address
BEGIN

Set TWI clock
(CLDIV, CHDIV, CKDIV) in TWI_CWGR
(Needed only once)

Set the Control register:
- Master enable
TWI_CR = MSEN + SVDIS

Set the Master Mode register:
- Device slave address (DADR)
- Internal address size (IADRSZ)
- Transfer direction bit
Write ==> bit MREAD = 0

Set the internal address
TWI_IADR = address

Load transmit register
TWI_THR = Data to send

Write STOP command
TWI_CR = STOP

Read Status register

No
TXRDY = 1?
Yes
Read Status register

TXCOMP = 1?
No
Yes
Transfer finished

638

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 33-17. TWI Write Operation with Multiple Data Bytes with or without Internal Address
BEGIN

Set TWI clock
(CLDIV, CHDIV, CKDIV) in TWI_CWGR
(Needed only once)

Set the Control register:
- Master enable
TWI_CR = MSEN + SVDIS

Set the Master Mode register:
- Device slave address
- Internal address size (if IADR used)
- Transfer direction bit
Write ==> bit MREAD = 0

No
Internal address size = 0?
Set the internal address
TWI_IADR = address

Yes

Load Transmit register
TWI_THR = Data to send

Read Status register

TWI_THR = data to send

No
TXRDY = 1?
Yes
Data to send?
Yes

Write STOP Command
TWI_CR = STOP

Read Status register
Yes
No
TXCOMP = 1?

END

639
6430F–ATARM–21-Feb-12

Figure 33-18. TWI Read Operation with Single Data Byte without Internal Address
BEGIN

Set TWI clock
(CLDIV, CHDIV, CKDIV) in TWI_CWGR
(Needed only once)

Set the Control register:
- Master enable
TWI_CR = MSEN + SVDIS

Set the Master Mode register:
- Device slave address
- Transfer direction bit
Read ==> bit MREAD = 1

Start the transfer
TWI_CR = START | STOP

Read status register

RXRDY = 1?

Yes
Read Receive Holding Register

Read Status register

No
TXCOMP = 1?
Yes
END

640

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 33-19. TWI Read Operation with Single Data Byte and Internal Address
BEGIN

Set TWI clock
(CLDIV, CHDIV, CKDIV) in TWI_CWGR
(Needed only once)

Set the Control register:
- Master enable
TWI_CR = MSEN + SVDIS

Set the Master Mode register:
- Device slave address
- Internal address size (IADRSZ)
- Transfer direction bit
Read ==> bit MREAD = 1

Set the internal address
TWI_IADR = address

Start the transfer
TWI_CR = START | STOP

Read Status register

No
RXRDY = 1?
Yes
Read Receive Holding register

Read Status register

No
TXCOMP = 1?
Yes
END

641
6430F–ATARM–21-Feb-12

Figure 33-20. TWI Read Operation with Multiple Data Bytes with or without Internal Address
BEGIN

Set TWI clock
(CLDIV, CHDIV, CKDIV) in TWI_CWGR
(Needed only once)

Set the Control register:
- Master enable
TWI_CR = MSEN + SVDIS

Set the Master Mode register:
- Device slave address
- Internal address size (if IADR used)
- Transfer direction bit
Read ==> bit MREAD = 1

Internal address size = 0?
Set the internal address
TWI_IADR = address

Yes
Start the transfer
TWI_CR = START

Read Status register

RXRDY = 1?

Yes
Read Receive Holding register (TWI_RHR)

Last data to read
but one?
Yes
Stop the transfer
TWI_CR = STOP

Read Status register

No
RXRDY = 1?
Yes
Read Receive Holding register (TWI_RHR)

Read status register

TXCOMP = 1?

Yes
END

642

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33.9

Multi-master Mode

33.9.1

Definition
More than one master may handle the bus at the same time without data corruption by using
arbitration.
Arbitration starts as soon as two or more masters place information on the bus at the same time,
and stops (arbitration is lost) for the master that intends to send a logical one while the other
master sends a logical zero.
As soon as arbitration is lost by a master, it stops sending data and listens to the bus in order to
detect a stop. When the stop is detected, the master who has lost arbitration may put its data on
the bus by respecting arbitration.
Arbitration is illustrated in Figure 33-22 on page 644.

33.9.2

Different Multi-master Modes
Two multi-master modes may be distinguished:
1. TWI is considered as a Master only and will never be addressed.
2. TWI may be either a Master or a Slave and may be addressed.
Note:

33.9.2.1

In both Multi-master modes arbitration is supported.

TWI as Master Only
In this mode, TWI is considered as a Master only (MSEN is always at one) and must be driven
like a Master with the ARBLST (ARBitration Lost) flag in addition.
If arbitration is lost (ARBLST = 1), the programmer must reinitiate the data transfer.
If the user starts a transfer (ex.: DADR + START + W + Write in THR) and if the bus is busy, the
TWI automatically waits for a STOP condition on the bus to initiate the transfer (see Figure 3321 on page 644).
Note:

33.9.2.2

The state of the bus (busy or free) is not indicated in the user interface.

TWI as Master or Slave
The automatic reversal from Master to Slave is not supported in case of a lost arbitration.
Then, in the case where TWI may be either a Master or a Slave, the programmer must manage
the pseudo Multi-master mode described in the steps below.
1. Program TWI in Slave mode (SADR + MSDIS + SVEN) and perform Slave Access (if
TWI is addressed).
2. If TWI has to be set in Master mode, wait until TXCOMP flag is at 1.
3. Program Master mode (DADR + SVDIS + MSEN) and start the transfer (ex: START +
Write in THR).
4. As soon as the Master mode is enabled, TWI scans the bus in order to detect if it is
busy or free. When the bus is considered as free, TWI initiates the transfer.
5. As soon as the transfer is initiated and until a STOP condition is sent, the arbitration
becomes relevant and the user must monitor the ARBLST flag.
6. If the arbitration is lost (ARBLST is set to 1), the user must program the TWI in Slave
mode in the case where the Master that won the arbitration wanted to access the TWI.
7. If TWI has to be set in Slave mode, wait until TXCOMP flag is at 1 and then program the
Slave mode.

643
6430F–ATARM–21-Feb-12

Note:

In the case where the arbitration is lost and TWI is addressed, TWI will not acknowledge even if it
is programmed in Slave mode as soon as ARBLST is set to 1. Then, the Master must repeat
SADR.

Figure 33-21. Programmer Sends Data While the Bus is Busy
TWCK
START sent by the TWI

STOP sent by the master
DATA sent by a master

TWD

DATA sent by the TWI

Bus is busy
Bus is free
Transfer is kept

TWI DATA transfer

A transfer is programmed
(DADR + W + START + Write THR)

Bus is considered as free
Transfer is initiated

Figure 33-22. Arbitration Cases
TWCK
TWD

TWCK
Data from a Master

0 0 1 1

Data from TWI

TWD

0 0

Arbitration is lost
TWI stops sending data

1 1

Data from the master

Arbitration is lost

0 0 1

The master stops sending data

0 1

Data from the TWI

ARBLST
Bus is busy

Bus is free

Transfer is kept

TWI DATA transfer
A transfer is programmed
(DADR + W + START + Write THR)

Transfer is stopped
Transfer is programmed again
(DADR + W + START + Write THR)

Bus is considered as free
Transfer is initiated

The flowchart shown in Figure 33-23 on page 645 gives an example of read and write operations
in Multi-master mode.

644

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 33-23. Multi-master Flowchart
START

Programm the SLAVE mode:
SADR + MSDIS + SVEN

Read Status Register

SVACC = 1 ?

Yes

GACC = 1 ?

No
No

SVREAD = 1 ?

EOSACC = 1 ?

TXRDY= 1 ?

Yes

Yes
No

Write in TWI_THR
TXCOMP = 1 ?

RXRDY= 1 ?

Yes
No

Yes
Read TWI_RHR

Need to perform
a master access ?

GENERAL CALL TREATMENT
Yes
Decoding of the
programming sequence
No

Prog seq
OK ?
Change SADR
Program the Master mode
DADR + SVDIS + MSEN + CLK + R / W

Read Status Register

Yes

ARBLST = 1 ?

Yes
Yes

MREAD = 1 ?

RXRDY= 0 ?

TXRDY= 0 ?
No

No
Read TWI_RHR

Yes

Data to read?

Data to send ?

Yes

Write in TWI_THR

No
Stop Transfer
TWI_CR = STOP

Read Status Register
Yes

TXCOMP = 0 ?

645
6430F–ATARM–21-Feb-12

33.10 Slave Mode
33.10.1

Definition
The Slave Mode is defined as a mode where the device receives the clock and the address from
another device called the master.
In this mode, the device never initiates and never completes the transmission (START,
REPEATED_START and STOP conditions are always provided by the master).

33.10.2

Application Block Diagram
Figure 33-24. Slave Mode Typical Application Block Diagram
VDD
R

Master

Host with
TWI
Interface

33.10.3

TWD
TWCK

Host with TWI
Interface

LCD Controller

Slave 1

Slave 2

Slave 3

Programming Slave Mode
The following fields must be programmed before entering Slave mode:
1. SADR (TWI_SMR): The slave device address is used in order to be accessed by master devices in read or write mode.
2. MSDIS (TWI_CR): Disable the master mode.
3. SVEN (TWI_CR): Enable the slave mode.
As the device receives the clock, values written in TWI_CWGR are not taken into account.

33.10.4

Receiving Data
After a Start or Repeated Start condition is detected and if the address sent by the Master
matches with the Slave address programmed in the SADR (Slave ADdress) field, SVACC (Slave
ACCess) flag is set and SVREAD (Slave READ) indicates the direction of the transfer.
SVACC remains high until a STOP condition or a repeated START is detected. When such a
condition is detected, EOSACC (End Of Slave ACCess) flag is set.

33.10.4.1

Read Sequence
In the case of a Read sequence (SVREAD is high), TWI transfers data written in the TWI_THR
(TWI Transmit Holding Register) until a STOP condition or a REPEATED_START + an address
different from SADR is detected. Note that at the end of the read sequence TXCOMP (Transmission Complete) flag is set and SVACC reset.
As soon as data is written in the TWI_THR, TXRDY (Transmit Holding Register Ready) flag is
reset, and it is set when the shift register is empty and the sent data acknowledged or not. If the
data is not acknowledged, the NACK flag is set.

646

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Note that a STOP or a repeated START always follows a NACK.
See Figure 33-25 on page 648.
33.10.4.2

Write Sequence
In the case of a Write sequence (SVREAD is low), the RXRDY (Receive Holding Register
Ready) flag is set as soon as a character has been received in the TWI_RHR (TWI Receive
Holding Register). RXRDY is reset when reading the TWI_RHR.
TWI continues receiving data until a STOP condition or a REPEATED_START + an address different from SADR is detected. Note that at the end of the write sequence TXCOMP flag is set
and SVACC reset.
See Figure 33-26 on page 648.

33.10.4.3

Clock Synchronization Sequence
In the case where TWI_THR or TWI_RHR is not written/read in time, TWI performs a clock
synchronization.
Clock stretching information is given by the SCLWS (Clock Wait state) bit.
See Figure 33-28 on page 650 and Figure 33-29 on page 651.

33.10.4.4

General Call
In the case where a GENERAL CALL is performed, GACC (General Call ACCess) flag is set.
After GACC is set, it is up to the programmer to interpret the meaning of the GENERAL CALL
and to decode the new address programming sequence.
See Figure 33-27 on page 649.

33.10.5
33.10.5.1

Data Transfer
Read Operation
The read mode is defined as a data requirement from the master.
After a START or a REPEATED START condition is detected, the decoding of the address
starts. If the slave address (SADR) is decoded, SVACC is set and SVREAD indicates the direction of the transfer.
Until a STOP or REPEATED START condition is detected, TWI continues sending data loaded
in the TWI_THR register.
If a STOP condition or a REPEATED START + an address different from SADR is detected,
SVACC is reset.
Figure 33-25 on page 648 describes the write operation.

647
6430F–ATARM–21-Feb-12

Figure 33-25. Read Access Ordered by a MASTER
SADR matches,
TWI answers with an ACK

SADR does not match,
TWI answers with a NACK

TWD

ADR

DATA

P/S/Sr

SADR R

DATA

ACK/NACK from the Master
A

DATA

S/Sr

TXRDY
Read RHR

Write THR

NACK
SVACC
SVREAD

SVREAD has to be taken into account only while SVACC is active

EOSVACC

Notes:

1. When SVACC is low, the state of SVREAD becomes irrelevant.
2. TXRDY is reset when data has been transmitted from TWI_THR to the shift register and set when this data has been
acknowledged or non acknowledged.

33.10.5.2

Write Operation
The write mode is defined as a data transmission from the master.
After a START or a REPEATED START, the decoding of the address starts. If the slave address
is decoded, SVACC is set and SVREAD indicates the direction of the transfer (SVREAD is low in
this case).
Until a STOP or REPEATED START condition is detected, TWI stores the received data in the
TWI_RHR register.
If a STOP condition or a REPEATED START + an address different from SADR is detected,
SVACC is reset.
Figure 33-26 on page 648 describes the Write operation.

Figure 33-26. Write Access Ordered by a Master
SADR does not match,
TWI answers with a NACK

TWD

ADR

DATA

SADR matches,
TWI answers with an ACK

P/S/Sr

SADR W

DATA

Read RHR

DATA

S/Sr

RXRDY
SVACC
SVREAD

SVREAD has to be taken into account only while SVACC is active

EOSVACC
Notes:

1. When SVACC is low, the state of SVREAD becomes irrelevant.
2. RXRDY is set when data has been transmitted from the shift register to the TWI_RHR and reset when this data is read.

648

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33.10.5.3

General Call
The general call is performed in order to change the address of the slave.
If a GENERAL CALL is detected, GACC is set.
After the detection of General Call, it is up to the programmer to decode the commands which
come afterwards.
In case of a WRITE command, the programmer has to decode the programming sequence and
program a new SADR if the programming sequence matches.
Figure 33-27 on page 649 describes the General Call access.

Figure 33-27. Master Performs a General Call
0000000 + W

TXD

GENERAL CALL

RESET command = 00000110X
WRITE command = 00000100X

Reset or write DADD

DATA1

DATA2

New SADR

New SADR
Programming sequence
GCACC

Reset after read
SVACC

Note:

This method allows the user to create an own programming sequence by choosing the programming bytes and the number of them. The programming sequence has to be provided to the
master.

649
6430F–ATARM–21-Feb-12

33.10.5.4

Clock Synchronization
In both read and write modes, it may happen that TWI_THR/TWI_RHR buffer is not filled /emptied before the emission/reception of a new character. In this case, to avoid sending/receiving
undesired data, a clock stretching mechanism is implemented.

Clock Synchronization in Read Mode
The clock is tied low if the shift register is empty and if a STOP or REPEATED START condition
was not detected. It is tied low until the shift register is loaded.
Figure 33-28 on page 650 describes the clock synchronization in Read mode.
Figure 33-28. Clock Synchronization in Read Mode
TWI_THR

DATA0

SADR

DATA1

DATA0

DATA1

DATA2

XXXXXXX

DATA2

TWCK
Write THR

CLOCK is tied low by the TWI
as long as THR is empty

SCLWS
TXRDY
SVACC
SVREAD
As soon as a START is detected

TXCOMP

TWI_THR is transmitted to the shift register

Notes:

Ack or Nack from the master

The data is memorized in TWI_THR until a new value is written

The clock is stretched after the ACK, the state of TWD is undefined during clock stretching

1. TXRDY is reset when data has been written in the TWI_THR to the shift register and set when this data has been acknowledged or non acknowledged.
2. At the end of the read sequence, TXCOMP is set after a STOP or after a REPEATED_START + an address different from
SADR.
3. SCLWS is automatically set when the clock synchronization mechanism is started.

650

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Clock Synchronization in Write Mode
The c lock is tied lo w if the shift register and the TWI_RHR is full. If a STOP or
REPEATED_START condition was not detected, it is tied low until TWI_RHR is read.
Figure 33-29 on page 651 describes the clock synchronization in Read mode.
Figure 33-29. Clock Synchronization in Write Mode
TWCK
CLOCK is tied low by the TWI as long as RHR is full
TWD

SADR

DATA0

TWI_RHR

DATA1

DATA0 is not read in the RHR

DATA2

DATA1

ADR

DATA2

SCLWS
SCL is stretched on the last bit of DATA1

RXRDY
Rd DATA0

Rd DATA1

Rd DATA2

SVACC
SVREAD
TXCOMP

Notes:

As soon as a START is detected

1. At the end of the read sequence, TXCOMP is set after a STOP or after a REPEATED_START + an address different from
SADR.
2. SCLWS is automatically set when the clock synchronization mechanism is started and automatically reset when the mechanism is finished.

651
6430F–ATARM–21-Feb-12

33.10.5.5
Reversal after a Repeated Start
Reversal of Read to Write
The master initiates the communication by a read command and finishes it by a write command.
Figure 33-30 on page 652 describes the repeated start + reversal from Read to Write mode.
Figure 33-30. Repeated Start + Reversal from Read to Write Mode
TWI_THR

TWD

DATA0

SADR

DATA0

DATA1

SADR

DATA2

DATA3

DATA2

TWI_RHR

DATA3

SVACC
SVREAD
TXRDY
RXRDY
EOSACC

Cleared after read
As soon as a START is detected

TXCOMP

1. TXCOMP is only set at the end of the transmission because after the repeated start, SADR is detected again.

Reversal of Write to Read
The master initiates the communication by a write command and finishes it by a read command.
Figure 33-31 on page 652 describes the repeated start + reversal from Write to Read mode.
Figure 33-31. Repeated Start + Reversal from Write to Read Mode
DATA2

TWI_THR

TWD

SADR

DATA0

TWI_RHR

DATA1

DATA0

SADR

DATA3

DATA2

DATA3

DATA1

SVACC
SVREAD
TXRDY
RXRDY
EOSACC
TXCOMP

Notes:

Read TWI_RHR

Cleared after read

As soon as a START is detected

1. In this case, if TWI_THR has not been written at the end of the read command, the clock is automatically stretched before
the ACK.
2. TXCOMP is only set at the end of the transmission because after the repeated start, SADR is detected again.

652

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33.10.6

Read Write Flowcharts
The flowchart shown in Figure 33-32 on page 653 gives an example of read and write operations
in Slave mode. A polling or interrupt method can be used to check the status bits. The interrupt
method requires that the interrupt enable register (TWI_IER) be configured first.

Figure 33-32. Read Write Flowchart in Slave Mode

Set the SLAVE mode:
SADR + MSDIS + SVEN

Read Status Register

SVACC = 1 ?
No
No

EOSACC = 1 ?

GACC = 1 ?

SVREAD = 0 ?
TXRDY= 1 ?

Write in TWI_THR

TXCOMP = 1 ?
RXRDY= 0 ?

END
Read TWI_RHR

GENERAL CALL TREATMENT

Decoding of the
programming sequence

Prog seq
OK ?

Change SADR

653
6430F–ATARM–21-Feb-12

33.11 Two-wire Interface (TWI) User Interface
Table 33-5.

Offset

Name

Access

Reset

0x00

Control Register

TWI_CR

Write-only

N/A

0x04

Master Mode Register

TWI_MMR

Read-write

0x00000000

0x08

Slave Mode Register

TWI_SMR

Read-write

0x00000000

0x0C

Internal Address Register

TWI_IADR

Read-write

0x00000000

0x10

Clock Waveform Generator Register

TWI_CWGR

Read-write

0x00000000

0x14 - 0x1C

Reserved

–

0x20

Status Register

TWI_SR

Read-only

0x0000F009

0x24

Interrupt Enable Register

TWI_IER

Write-only

N/A

0x28

Interrupt Disable Register

TWI_IDR

Write-only

N/A

0x2C

Interrupt Mask Register

TWI_IMR

Read-only

0x00000000

0x30

Receive Holding Register

TWI_RHR

Read-only

0x00000000

Transmit Holding Register

TWI_THR

Write-only

0x00000000

0x34
(1)

0xEC - 0xFC

Reserved

–

0x100 - 0x124

Reserved for the PDC

–

Note:

654

1. All unlisted offset values are considered as “reserved”.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33.11.1
Name:

TWI Control Register
TWI_CR

Address:

0x40084000 (0), 0x40088000 (1)

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
SWRST

6
QUICK

5
SVDIS

4
SVEN

3
MSDIS

2
MSEN

1
STOP

0
START

• START: Send a START Condition
0 = No effect.
1 = A frame beginning with a START bit is transmitted according to the features defined in the mode register.
This action is necessary when the TWI peripheral wants to read data from a slave. When configured in Master Mode with a
write operation, a frame is sent as soon as the user writes a character in the Transmit Holding Register (TWI_THR).
• STOP: Send a STOP Condition
0 = No effect.
1 = STOP Condition is sent just after completing the current byte transmission in master read mode.
– In single data byte master read, the START and STOP must both be set.
– In multiple data bytes master read, the STOP must be set after the last data received but one.
– In master read mode, if a NACK bit is received, the STOP is automatically performed.
– In master data write operation, a STOP condition will be sent after the transmission of the current data is
finished.
• MSEN: TWI Master Mode Enabled
0 = No effect.
1 = If MSDIS = 0, the master mode is enabled.
Note:

Switching from Slave to Master mode is only permitted when TXCOMP = 1.

• MSDIS: TWI Master Mode Disabled
0 = No effect.
1 = The master mode is disabled, all pending data is transmitted. The shifter and holding characters (if it contains data) are
transmitted in case of write operation. In read operation, the character being transferred must be completely received
before disabling.

655
6430F–ATARM–21-Feb-12

• SVEN: TWI Slave Mode Enabled
0 = No effect.
1 = If SVDIS = 0, the slave mode is enabled.
Note:

Switching from Master to Slave mode is only permitted when TXCOMP = 1.

• SVDIS: TWI Slave Mode Disabled
0 = No effect.
1 = The slave mode is disabled. The shifter and holding characters (if it contains data) are transmitted in case of read operation. In write operation, the character being transferred must be completely received before disabling.
• QUICK: SMBUS Quick Command
0 = No effect.
1 = If Master mode is enabled, a SMBUS Quick Command is sent.
• SWRST: Software Reset
0 = No effect.
1 = Equivalent to a system reset.

656

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33.11.2
Name:

TWI Master Mode Register
TWI_MMR

Address:

0x40084004 (0), 0x40088004 (1)

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

19
DADR

15
–

14
–

13
–

12
MREAD

11
–

10
–

7
–

6
–

5
–

4
–

3
–

2
–

1
–

8
IADRSZ
0
–

• IADRSZ: Internal Device Address Size
Value

Name

Description

NONE

No internal device address

1_BYTE

One-byte internal device address

2_BYTE

Two-byte internal device address

3_BYTE

Three-byte internal device address

• MREAD: Master Read Direction
0 = Master write direction.
1 = Master read direction.
• DADR: Device Address
The device address is used to access slave devices in read or write mode. Those bits are only used in Master mode.

657
6430F–ATARM–21-Feb-12

33.11.3
Name:

TWI Slave Mode Register
TWI_SMR

Address:

0x40084008 (0), 0x40088008 (1)

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

19
SADR

15
–

14
–

13
–

12
–

11
–

10
–

7
–

6
–

5
–

4
–

3
–

2
–

1
–

0
–

• SADR: Slave Address
The slave device address is used in Slave mode in order to be accessed by master devices in read or write mode.
SADR must be programmed before enabling the Slave mode or after a general call. Writes at other times have no effect.

658

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33.11.4
Name:

TWI Internal Address Register
TWI_IADR

Address:

0x4008400C (0), 0x4008800C (1)

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

IADR
15

12
IADR

4
IADR

• IADR: Internal Address
0, 1, 2 or 3 bytes depending on IADRSZ.

659
6430F–ATARM–21-Feb-12

33.11.5
Name:

TWI Clock Waveform Generator Register
TWI_CWGR

Address:

0x40084010 (0), 0x40088010 (1)

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

17
CKDIV

CHDIV
7

4
CLDIV

TWI_CWGR is only used in Master mode.
• CLDIV: Clock Low Divider
The SCL low period is defined as follows:
T low = ( ( CLDIV × 2

CKDIV

) + 4 ) × T MCK

• CHDIV: Clock High Divider
The SCL high period is defined as follows:
T high = ( ( CHDIV × 2

CKDIV

) + 4 ) × T MCK

• CKDIV: Clock Divider
The CKDIV is used to increase both SCL high and low periods.

660

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33.11.6
Name:

TWI Status Register
TWI_SR

Address:

0x40084020 (0), 0x40088020 (1)

Access:

Read-only

Reset:

0x0000F009

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
TXBUFE

14
RXBUFF

13
ENDTX

12
ENDRX

11
EOSACC

10
SCLWS

9
ARBLST

8
NACK

7
–

6
OVRE

5
GACC

4
SVACC

3
SVREAD

2
TXRDY

1
RXRDY

0
TXCOMP

• TXCOMP: Transmission Completed (automatically set / reset)
TXCOMP used in Master mode:
0 = During the length of the current frame.
1 = When both holding and shifter registers are empty and STOP condition has been sent.
TXCOMP behavior in Master mode can be seen in Figure 33-8 on page 633 and in Figure 33-10 on page 634.
TXCOMP used in Slave mode:
0 = As soon as a Start is detected.
1 = After a Stop or a Repeated Start + an address different from SADR is detected.
TXCOMP behavior in Slave mode can be seen in Figure 33-28 on page 650, Figure 33-29 on page 651, Figure 33-30 on
page 652 and Figure 33-31 on page 652.
• RXRDY: Receive Holding Register Ready (automatically set / reset)
0 = No character has been received since the last TWI_RHR read operation.
1 = A byte has been received in the TWI_RHR since the last read.
RXRDY behavior in Master mode can be seen in Figure 33-10 on page 634.
RXRDY behavior in Slave mode can be seen in Figure 33-26 on page 648, Figure 33-29 on page 651, Figure 33-30 on
page 652 and Figure 33-31 on page 652.
• TXRDY: Transmit Holding Register Ready (automatically set / reset)
TXRDY used in Master mode:
0 = The transmit holding register has not been transferred into shift register. Set to 0 when writing into TWI_THR register.
1 = As soon as a data byte is transferred from TWI_THR to internal shifter or if a NACK error is detected, TXRDY is set at
the same time as TXCOMP and NACK. TXRDY is also set when MSEN is set (enable TWI).
TXRDY behavior in Master mode can be seen in Figure 33-8 on page 633.

661
6430F–ATARM–21-Feb-12

TXRDY used in Slave mode:
0 = As soon as data is written in the TWI_THR, until this data has been transmitted and acknowledged (ACK or NACK).
1 = It indicates that the TWI_THR is empty and that data has been transmitted and acknowledged.
If TXRDY is high and if a NACK has been detected, the transmission will be stopped. Thus when TRDY = NACK = 1, the
programmer must not fill TWI_THR to avoid losing it.
TXRDY behavior in Slave mode can be seen in Figure 33-25 on page 648, Figure 33-28 on page 650, Figure 33-30 on
page 652 and Figure 33-31 on page 652.
• SVREAD: Slave Read (automatically set / reset)
This bit is only used in Slave mode. When SVACC is low (no Slave access has been detected) SVREAD is irrelevant.
0 = Indicates that a write access is performed by a Master.
1 = Indicates that a read access is performed by a Master.
SVREAD behavior can be seen in Figure 33-25 on page 648, Figure 33-26 on page 648, Figure 33-30 on page 652 and
Figure 33-31 on page 652.
• SVACC: Slave Access (automatically set / reset)
This bit is only used in Slave mode.
0 = TWI is not addressed. SVACC is automatically cleared after a NACK or a STOP condition is detected.
1 = Indicates that the address decoding sequence has matched (A Master has sent SADR). SVACC remains high until a
NACK or a STOP condition is detected.
SVACC behavior can be seen in Figure 33-25 on page 648, Figure 33-26 on page 648, Figure 33-30 on page 652 and Figure 33-31 on page 652.
• GACC: General Call Access (clear on read)
This bit is only used in Slave mode.
0 = No General Call has been detected.
1 = A General Call has been detected. After the detection of General Call, if need be, the programmer may acknowledge
this access and decode the following bytes and respond according to the value of the bytes.
GACC behavior can be seen in Figure 33-27 on page 649.
• OVRE: Overrun Error (clear on read)
This bit is only used in Master mode.
0 = TWI_RHR has not been loaded while RXRDY was set
1 = TWI_RHR has been loaded while RXRDY was set. Reset by read in TWI_SR when TXCOMP is set.
• NACK: Not Acknowledged (clear on read)
NACK used in Master mode:
0 = Each data byte has been correctly received by the far-end side TWI slave component.
1 = A data byte has not been acknowledged by the slave component. Set at the same time as TXCOMP.

662

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
NACK used in Slave Read mode:
0 = Each data byte has been correctly received by the Master.
1 = In read mode, a data byte has not been acknowledged by the Master. When NACK is set the programmer must not fill
TWI_THR even if TXRDY is set, because it means that the Master will stop the data transfer or re initiate it.
Note that in Slave Write mode all data are acknowledged by the TWI.
• ARBLST: Arbitration Lost (clear on read)
This bit is only used in Master mode.
0: Arbitration won.
1: Arbitration lost. Another master of the TWI bus has won the multi-master arbitration. TXCOMP is set at the same time.
• SCLWS: Clock Wait State (automatically set / reset)
This bit is only used in Slave mode.
0 = The clock is not stretched.
1 = The clock is stretched. TWI_THR / TWI_RHR buffer is not filled / emptied before the emission / reception of a new
character.
SCLWS behavior can be seen in Figure 33-28 on page 650 and Figure 33-29 on page 651.
• EOSACC: End Of Slave Access (clear on read)
This bit is only used in Slave mode.
0 = A slave access is being performing.
1 = The Slave Access is finished. End Of Slave Access is automatically set as soon as SVACC is reset.
EOSACC behavior can be seen in Figure 33-30 on page 652 and Figure 33-31 on page 652
• ENDRX: End of RX buffer
0 = The Receive Counter Register has not reached 0 since the last write in TWI_RCR or TWI_RNCR.
1 = The Receive Counter Register has reached 0 since the last write in TWI_RCR or TWI_RNCR.
• ENDTX: End of TX buffer
0 = The Transmit Counter Register has not reached 0 since the last write in TWI_TCR or TWI_TNCR.
1 = The Transmit Counter Register has reached 0 since the last write in TWI_TCR or TWI_TNCR.
• RXBUFF: RX Buffer Full
0 = TWI_RCR or TWI_RNCR have a value other than 0.
1 = Both TWI_RCR and TWI_RNCR have a value of 0.
• TXBUFE: TX Buffer Empty
0 = TWI_TCR or TWI_TNCR have a value other than 0.
1 = Both TWI_TCR and TWI_TNCR have a value of 0.

663
6430F–ATARM–21-Feb-12

33.11.7
Name:

TWI Interrupt Enable Register
TWI_IER

Address:

0x40084024 (0), 0x40088024 (1)

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
TXBUFE

14
RXBUFF

13
ENDTX

12
ENDRX

11
EOSACC

10
SCL_WS

9
ARBLST

8
NACK

7
–

6
OVRE

5
GACC

4
SVACC

3
–

2
TXRDY

1
RXRDY

0
TXCOMP

• TXCOMP: Transmission Completed Interrupt Enable
• RXRDY: Receive Holding Register Ready Interrupt Enable
• TXRDY: Transmit Holding Register Ready Interrupt Enable
• SVACC: Slave Access Interrupt Enable
• GACC: General Call Access Interrupt Enable
• OVRE: Overrun Error Interrupt Enable
• NACK: Not Acknowledge Interrupt Enable
• ARBLST: Arbitration Lost Interrupt Enable
• SCL_WS: Clock Wait State Interrupt Enable
• EOSACC: End Of Slave Access Interrupt Enable
• ENDRX: End of Receive Buffer Interrupt Enable
• ENDTX: End of Transmit Buffer Interrupt Enable
• RXBUFF: Receive Buffer Full Interrupt Enable
• TXBUFE: Transmit Buffer Empty Interrupt Enable
0 = No effect.
1 = Enables the corresponding interrupt.

664

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33.11.8
Name:

TWI Interrupt Disable Register
TWI_IDR

Address:

0x40084028 (0), 0x40088028 (1)

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
TXBUFE

14
RXBUFF

13
ENDTX

12
ENDRX

11
EOSACC

10
SCL_WS

9
ARBLST

8
NACK

7
–

6
OVRE

5
GACC

4
SVACC

3
–

2
TXRDY

1
RXRDY

0
TXCOMP

• TXCOMP: Transmission Completed Interrupt Disable
• RXRDY: Receive Holding Register Ready Interrupt Disable
• TXRDY: Transmit Holding Register Ready Interrupt Disable
• SVACC: Slave Access Interrupt Disable
• GACC: General Call Access Interrupt Disable
• OVRE: Overrun Error Interrupt Disable
• NACK: Not Acknowledge Interrupt Disable
• ARBLST: Arbitration Lost Interrupt Disable
• SCL_WS: Clock Wait State Interrupt Disable
• EOSACC: End Of Slave Access Interrupt Disable
• ENDRX: End of Receive Buffer Interrupt Disable
• ENDTX: End of Transmit Buffer Interrupt Disable
• RXBUFF: Receive Buffer Full Interrupt Disable
• TXBUFE: Transmit Buffer Empty Interrupt Disable
0 = No effect.
1 = Disables the corresponding interrupt.

665
6430F–ATARM–21-Feb-12

33.11.9
Name:

TWI Interrupt Mask Register
TWI_IMR

Address:

0x4008402C (0), 0x4008802C (1)

Access:

Read-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
TXBUFE

14
RXBUFF

13
ENDTX

12
ENDRX

11
EOSACC

10
SCL_WS

9
ARBLST

8
NACK

7
–

6
OVRE

5
GACC

4
SVACC

3
–

2
TXRDY

1
RXRDY

0
TXCOMP

• TXCOMP: Transmission Completed Interrupt Mask
• RXRDY: Receive Holding Register Ready Interrupt Mask
• TXRDY: Transmit Holding Register Ready Interrupt Mask
• SVACC: Slave Access Interrupt Mask
• GACC: General Call Access Interrupt Mask
• OVRE: Overrun Error Interrupt Mask
• NACK: Not Acknowledge Interrupt Mask
• ARBLST: Arbitration Lost Interrupt Mask
• SCL_WS: Clock Wait State Interrupt Mask
• EOSACC: End Of Slave Access Interrupt Mask
• ENDRX: End of Receive Buffer Interrupt Mask
• ENDTX: End of Transmit Buffer Interrupt Mask
• RXBUFF: Receive Buffer Full Interrupt Mask
• TXBUFE: Transmit Buffer Empty Interrupt Mask
0 = The corresponding interrupt is disabled.
1 = The corresponding interrupt is enabled.

666

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
33.11.10 TWI Receive Holding Register
Name:
TWI_RHR
Address:

0x40084030 (0), 0x40088030 (1)

Access:

Read-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

RXDATA

• RXDATA: Master or Slave Receive Holding Data

667
6430F–ATARM–21-Feb-12

33.11.11 TWI Transmit Holding Register
Name:
TWI_THR
Address:

0x40084034 (0), 0x40088034 (1)

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

TXDATA

• TXDATA: Master or Slave Transmit Holding Data

668

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
34. Universal Asynchronous Receiver Transceiver (UART)
34.1

Description
The Universal Asynchronous Receiver Transmitter features a two-pin UART that can be used for
communication and trace purposes and offers an ideal medium for in-situ programming solutions. Moreover, the association with two peripheral DMA controller (PDC) channels permits
packet handling for these tasks with processor time reduced to a minimum.

34.2

Embedded Characteristics
• Two-pin UART
– Implemented Features are USART Compatible
– Independent Receiver and Transmitter with a Common Programmable Baud Rate
Generator
– Even, Odd, Mark or Space Parity Generation
– Parity, Framing and Overrun Error Detection
– Automatic Echo, Local Loopback and Remote Loopback Channel Modes
– Interrupt Generation
– Support for Two PDC Channels with Connection to Receiver and Transmitter

669
6430F–ATARM–21-Feb-12

34.3

Block Diagram

Figure 34-1. UART Functional Block Diagram
Peripheral
Bridge
Peripheral DMA Controller

APB

UART
UTXD

Transmit
Power
Management
Controller

Parallel
Input/
Output

Baud Rate
Generator

MCK

Receive
URXD

Interrupt
Control

Table 34-1.

UART Pin Description

Pin Name

Description

Type

URXD

UART Receive Data

Input

UTXD

UART Transmit Data

Output

34.4
34.4.1

uart_irq

Product Dependencies
I/O Lines
The UART pins are multiplexed with PIO lines. The programmer must first configure the corresponding PIO Controller to enable I/O line operations of the UART.
Table 34-2.

34.4.2

670

I/O Lines

Instance

Signal

I/O Line

Peripheral

UART

URXD

PA11

UART

UTXD

PA12

Power Management
The UART clock is controllable through the Power Management Controller. In this case, the programmer must first configure the PMC to enable the UART clock. Usually, the peripheral
identifier used for this purpose is 1.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
34.4.3

34.5

Interrupt Source
The UART interrupt line is connected to one of the interrupt sources of the Nested Vectored
Interrupt Controller (NVIC). Interrupt handling requires programming of the NVIC before configuring the UART.

UART Operations
The UART operates in asynchronous mode only and supports only 8-bit character handling (with
parity). It has no clock pin.
The UART is made up of a receiver and a transmitter that operate independently, and a common
baud rate generator. Receiver timeout and transmitter time guard are not implemented. However, all the implemented features are compatible with those of a standard USART.

34.5.1

Baud Rate Generator
The baud rate generator provides the bit period clock named baud rate clock to both the receiver
and the transmitter.
The baud rate clock is the master clock divided by 16 times the value (CD) written in
UART_BRGR (Baud Rate Generator Register). If UART_BRGR is set to 0, the baud rate clock is
disabled and the UART remains inactive. The maximum allowable baud rate is Master Clock
divided by 16. The minimum allowable baud rate is Master Clock divided by (16 x 65536).
MCK
Baud Rate = -----------------------16 × CD
Figure 34-2. Baud Rate Generator
CD
CD
MCK

16-bit Counter

OUT

>1
1
0

Divide
by 16

Baud Rate
Clock

0
Receiver
Sampling Clock

34.5.2
34.5.2.1

Receiver
Receiver Reset, Enable and Disable
After device reset, the UART receiver is disabled and must be enabled before being used. The
receiver can be enabled by writing the control register UART_CR with the bit RXEN at 1. At this
command, the receiver starts looking for a start bit.
The programmer can disable the receiver by writing UART_CR with the bit RXDIS at 1. If the
receiver is waiting for a start bit, it is immediately stopped. However, if the receiver has already
detected a start bit and is receiving the data, it waits for the stop bit before actually stopping its
operation.

671
6430F–ATARM–21-Feb-12

The programmer can also put the receiver in its reset state by writing UART_CR with the bit
RSTRX at 1. In doing so, the receiver immediately stops its current operations and is disabled,
whatever its current state. If RSTRX is applied when data is being processed, this data is lost.
34.5.2.2

Start Detection and Data Sampling
The UART only supports asynchronous operations, and this affects only its receiver. The UART
receiver detects the start of a received character by sampling the URXD signal until it detects a
valid start bit. A low level (space) on URXD is interpreted as a valid start bit if it is detected for
more than 7 cycles of the sampling clock, which is 16 times the baud rate. Hence, a space that is
longer than 7/16 of the bit period is detected as a valid start bit. A space which is 7/16 of a bit
period or shorter is ignored and the receiver continues to wait for a valid start bit.
When a valid start bit has been detected, the receiver samples the URXD at the theoretical midpoint of each bit. It is assumed that each bit lasts 16 cycles of the sampling clock (1-bit period)
so the bit sampling point is eight cycles (0.5-bit period) after the start of the bit. The first sampling
point is therefore 24 cycles (1.5-bit periods) after the falling edge of the start bit was detected.
Each subsequent bit is sampled 16 cycles (1-bit period) after the previous one.
Figure 34-3. Start Bit Detection
Sampling Clock

URXD

True Start
Detection

Baud Rate
Clock

Figure 34-4. Character Reception
Example: 8-bit, parity enabled 1 stop
0.5 bit
period

1 bit
period

URXD

Sampling

34.5.2.3

672

D0
D1
True Start Detection

Stop Bit
Parity Bit

Receiver Ready
When a complete character is received, it is transferred to the UART_RHR and the RXRDY status bit in UART_SR (Status Register) is set. The bit RXRDY is automatically cleared when the
receive holding register UART_RHR is read.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 34-5. Receiver Ready
S

URXD

RXRDY

Read UART_RHR

34.5.2.4

Receiver Overrun
If UART_RHR has not been read by the software (or the Peripheral Data Controller or DMA
Controller) since the last transfer, the RXRDY bit is still set and a new character is received, the
OVRE status bit in UART_SR is set. OVRE is cleared when the software writes the control register UART_CR with the bit RSTSTA (Reset Status) at 1.

Figure 34-6. Receiver Overrun
URXD

stop

RXRDY
OVRE

RSTSTA

34.5.2.5

Parity Error
Each time a character is received, the receiver calculates the parity of the received data bits, in
accordance with the field PAR in UART_MR. It then compares the result with the received parity
bit. If different, the parity error bit PARE in UART_SR is set at the same time the RXRDY is set.
The parity bit is cleared when the control register UART_CR is written with the bit RSTSTA
(Reset Status) at 1. If a new character is received before the reset status command is written,
the PARE bit remains at 1.

Figure 34-7. Parity Error
URXD

stop

RXRDY
PARE

Wrong Parity Bit

34.5.2.6

RSTSTA

Receiver Framing Error
When a start bit is detected, it generates a character reception when all the data bits have been
sampled. The stop bit is also sampled and when it is detected at 0, the FRAME (Framing Error)
bit in UART_SR is set at the same time the RXRDY bit is set. The FRAME bit remains high until
the control register UART_CR is written with the bit RSTSTA at 1.

673
6430F–ATARM–21-Feb-12

Figure 34-8. Receiver Framing Error
URXD

stop

RXRDY
FRAME

Stop Bit
Detected at 0

34.5.3
34.5.3.1

RSTSTA

Transmitter
Transmitter Reset, Enable and Disable
After device reset, the UART transmitter is disabled and it must be enabled before being used.
The transmitter is enabled by writing the control register UART_CR with the bit TXEN at 1. From
this command, the transmitter waits for a character to be written in the Transmit Holding Register
(UART_THR) before actually starting the transmission.
The programmer can disable the transmitter by writing UART_CR with the bit TXDIS at 1. If the
transmitter is not operating, it is immediately stopped. However, if a character is being processed into the Shift Register and/or a character has been written in the Transmit Holding
Register, the characters are completed before the transmitter is actually stopped.
The programmer can also put the transmitter in its reset state by writing the UART_CR with the
bit RSTTX at 1. This immediately stops the transmitter, whether or not it is processing
characters.

34.5.3.2

Transmit Format
The UART transmitter drives the pin UTXD at the baud rate clock speed. The line is driven
depending on the format defined in the Mode Register and the data stored in the Shift Register.
One start bit at level 0, then the 8 data bits, from the lowest to the highest bit, one optional parity
bit and one stop bit at 1 are consecutively shifted out as shown in the following figure. The field
PARE in the mode register UART_MR defines whether or not a parity bit is shifted out. When a
parity bit is enabled, it can be selected between an odd parity, an even parity, or a fixed space or
mark bit.
Figure 34-9. Character Transmission
Example: Parity enabled
Baud Rate
Clock
UTXD

Start
Bit

34.5.3.3

674

Parity
Bit

Stop
Bit

Transmitter Control
When the transmitter is enabled, the bit TXRDY (Transmitter Ready) is set in the status register
UART_SR. The transmission starts when the programmer writes in the Transmit Holding Register (UART_THR), and after the written character is transferred from UART_THR to the Shift

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Register. The TXRDY bit remains high until a second character is written in UART_THR. As
soon as the first character is completed, the last character written in UART_THR is transferred
into the shift register and TXRDY rises again, showing that the holding register is empty.
When both the Shift Register and UART_THR are empty, i.e., all the characters written in
UART_THR have been processed, the TXEMPTY bit rises after the last stop bit has been
completed.
Figure 34-10. Transmitter Control
UART_THR

Data 0

Data 1

Shift Register

UTXD

Data 0

Data 1

stop

Data 1

stop

TXRDY
TXEMPTY

Write Data 0
in UART_THR

34.5.4

Write Data 1
in UART_THR

Peripheral DMA Controller
Both the receiver and the transmitter of the UART are connected to a Peripheral DMA Controller
(PDC) channel.
The peripheral data controller channels are programmed via registers that are mapped within
the UART user interface from the offset 0x100. The status bits are reported in the UART status
register (UART_SR) and can generate an interrupt.
The RXRDY bit triggers the PDC channel data transfer of the receiver. This results in a read of
the data in UART_RHR. The TXRDY bit triggers the PDC channel data transfer of the transmitter. This results in a write of data in UART_THR.

34.5.5

Test Modes
The UART supports three test modes. These modes of operation are programmed by using the
field CHMODE (Channel Mode) in the mode register (UART_MR).
The Automatic Echo mode allows bit-by-bit retransmission. When a bit is received on the URXD
line, it is sent to the UTXD line. The transmitter operates normally, but has no effect on the
UTXD line.
The Local Loopback mode allows the transmitted characters to be received. UTXD and URXD
pins are not used and the output of the transmitter is internally connected to the input of the
receiver. The URXD pin level has no effect and the UTXD line is held high, as in idle state.
The Remote Loopback mode directly connects the URXD pin to the UTXD line. The transmitter
and the receiver are disabled and have no effect. This mode allows a bit-by-bit retransmission.

675
6430F–ATARM–21-Feb-12

Figure 34-11. Test Modes
Automatic Echo
RXD

Receiver

Transmitter

Disabled

TXD

Local Loopback
Disabled

Receiver

RXD

VDD
Disabled

Transmitter

Remote Loopback

TXD

VDD
Disabled

RXD

Receiver

Disabled

Transmitter

676

TXD

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
34.6

Universal Asynchronous Receiver Transceiver (UART) User Interface

Table 34-3.

Offset

Name

Access

Reset

0x0000

Control Register

UART_CR

Write-only

–

0x0004

Mode Register

UART_MR

Read-write

0x0

0x0008

Interrupt Enable Register

UART_IER

Write-only

–

0x000C

Interrupt Disable Register

UART_IDR

Write-only

–

0x0010

Interrupt Mask Register

UART_IMR

Read-only

0x0

0x0014

Status Register

UART_SR

Read-only

–

0x0018

Receive Holding Register

UART_RHR

Read-only

0x0

0x001C

Transmit Holding Register

UART_THR

Write-only

–

0x0020

Baud Rate Generator Register

UART_BRGR

Read-write

0x0

0x0024 - 0x003C

Reserved

–

0x004C - 0x00FC

Reserved

–

0x0100 - 0x0124

PDC Area

–

677
6430F–ATARM–21-Feb-12

34.6.1
Name:

UART Control Register
UART_CR

Address:

0x400E0600

Access:

Write-only

–

RSTSTA

TXDIS

TXEN

RXDIS

RXEN

RSTTX

RSTRX

–

• RSTRX: Reset Receiver
0 = No effect.
1 = The receiver logic is reset and disabled. If a character is being received, the reception is aborted.
• RSTTX: Reset Transmitter
0 = No effect.
1 = The transmitter logic is reset and disabled. If a character is being transmitted, the transmission is aborted.
• RXEN: Receiver Enable
0 = No effect.
1 = The receiver is enabled if RXDIS is 0.
• RXDIS: Receiver Disable
0 = No effect.
1 = The receiver is disabled. If a character is being processed and RSTRX is not set, the character is completed before the
receiver is stopped.
• TXEN: Transmitter Enable
0 = No effect.
1 = The transmitter is enabled if TXDIS is 0.
• TXDIS: Transmitter Disable
0 = No effect.
1 = The transmitter is disabled. If a character is being processed and a character has been written in the UART_THR and
RSTTX is not set, both characters are completed before the transmitter is stopped.
• RSTSTA: Reset Status Bits
0 = No effect.
1 = Resets the status bits PARE, FRAME and OVRE in the UART_SR.

678

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
34.6.2
Name:

UART Mode Register
UART_MR

Address:

0x400E0604

Access:

Read-write

–

15
CHMODE

–

PAR

–

• PAR: Parity Type
Value

Name

Description

EVEN

Even parity

ODD

Odd parity

SPACE

Space: parity forced to 0

MARK

Mark: parity forced to 1

No parity

• CHMODE: Channel Mode
Value

Name

Description

NORMAL

Normal Mode

AUTOMATIC

Automatic Echo

LOCAL_LOOPBACK

Local Loopback

REMOTE_LOOPBACK

Remote Loopback

679
6430F–ATARM–21-Feb-12

34.6.3
Name:

UART Interrupt Enable Register
UART_IER

Address:

0x400E0608

Access:

Write-only

–

RXBUFF

TXBUFE

–

TXEMPTY

–

PARE

FRAME

OVRE

ENDTX

ENDRX

–

TXRDY

RXRDY

• RXRDY: Enable RXRDY Interrupt
• TXRDY: Enable TXRDY Interrupt
• ENDRX: Enable End of Receive Transfer Interrupt
• ENDTX: Enable End of Transmit Interrupt
• OVRE: Enable Overrun Error Interrupt
• FRAME: Enable Framing Error Interrupt
• PARE: Enable Parity Error Interrupt
• TXEMPTY: Enable TXEMPTY Interrupt
• TXBUFE: Enable Buffer Empty Interrupt
• RXBUFF: Enable Buffer Full Interrupt
0 = No effect.
1 = Enables the corresponding interrupt.

680

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
34.6.4
Name:

UART Interrupt Disable Register
UART_IDR

Address:

0x400E060C

Access:

Write-only

–

RXBUFF

TXBUFE

–

TXEMPTY

–

PARE

FRAME

OVRE

ENDTX

ENDRX

–

TXRDY

RXRDY

• RXRDY: Disable RXRDY Interrupt
• TXRDY: Disable TXRDY Interrupt
• ENDRX: Disable End of Receive Transfer Interrupt
• ENDTX: Disable End of Transmit Interrupt
• OVRE: Disable Overrun Error Interrupt
• FRAME: Disable Framing Error Interrupt
• PARE: Disable Parity Error Interrupt
• TXEMPTY: Disable TXEMPTY Interrupt
• TXBUFE: Disable Buffer Empty Interrupt
• RXBUFF: Disable Buffer Full Interrupt
0 = No effect.
1 = Disables the corresponding interrupt.

681
6430F–ATARM–21-Feb-12

34.6.5
Name:

UART Interrupt Mask Register
UART_IMR

Address:

0x400E0610

Access:

Read-only

–

RXBUFF

TXBUFE

–

TXEMPTY

–

PARE

FRAME

OVRE

ENDTX

ENDRX

–

TXRDY

RXRDY

• RXRDY: Mask RXRDY Interrupt
• TXRDY: Disable TXRDY Interrupt
• ENDRX: Mask End of Receive Transfer Interrupt
• ENDTX: Mask End of Transmit Interrupt
• OVRE: Mask Overrun Error Interrupt
• FRAME: Mask Framing Error Interrupt
• PARE: Mask Parity Error Interrupt
• TXEMPTY: Mask TXEMPTY Interrupt
• TXBUFE: Mask TXBUFE Interrupt
• RXBUFF: Mask RXBUFF Interrupt
0 = The corresponding interrupt is disabled.
1 = The corresponding interrupt is enabled.

682

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
34.6.6
Name:

UART Status Register
UART_SR

Address:

0x400E0614

Access:

Read-only

–

RXBUFF

TXBUFE

–

TXEMPTY

–

PARE

FRAME

OVRE

ENDTX

ENDRX

–

TXRDY

RXRDY

• RXRDY: Receiver Ready
0 = No character has been received since the last read of the UART_RHR or the receiver is disabled.
1 = At least one complete character has been received, transferred to UART_RHR and not yet read.
• TXRDY: Transmitter Ready
0 = A character has been written to UART_THR and not yet transferred to the Shift Register, or the transmitter is disabled.
1 = There is no character written to UART_THR not yet transferred to the Shift Register.
• ENDRX: End of Receiver Transfer
0 = The End of Transfer signal from the receiver Peripheral Data Controller channel is inactive.
1 = The End of Transfer signal from the receiver Peripheral Data Controller channel is active.
• ENDTX: End of Transmitter Transfer
0 = The End of Transfer signal from the transmitter Peripheral Data Controller channel is inactive.
1 = The End of Transfer signal from the transmitter Peripheral Data Controller channel is active.
• OVRE: Overrun Error
0 = No overrun error has occurred since the last RSTSTA.
1 = At least one overrun error has occurred since the last RSTSTA.
• FRAME: Framing Error
0 = No framing error has occurred since the last RSTSTA.
1 = At least one framing error has occurred since the last RSTSTA.
• PARE: Parity Error
0 = No parity error has occurred since the last RSTSTA.
1 = At least one parity error has occurred since the last RSTSTA.

683
6430F–ATARM–21-Feb-12

• TXEMPTY: Transmitter Empty
0 = There are characters in UART_THR, or characters being processed by the transmitter, or the transmitter is disabled.
1 = There are no characters in UART_THR and there are no characters being processed by the transmitter.
• TXBUFE: Transmission Buffer Empty
0 = The buffer empty signal from the transmitter PDC channel is inactive.
1 = The buffer empty signal from the transmitter PDC channel is active.
• RXBUFF: Receive Buffer Full
0 = The buffer full signal from the receiver PDC channel is inactive.
1 = The buffer full signal from the receiver PDC channel is active.

684

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
34.6.7
Name:

UART Receiver Holding Register
UART_RHR

Address:

0x400E0618

Access:

Read-only

–

RXCHR

• RXCHR: Received Character
Last received character if RXRDY is set.

685
6430F–ATARM–21-Feb-12

34.6.8
Name:

UART Transmit Holding Register
UART_THR

Address:

0x400E061C

Access:

Write-only

–

TXCHR

• TXCHR: Character to be Transmitted
Next character to be transmitted after the current character if TXRDY is not set.

686

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
34.6.9
Name:

UART Baud Rate Generator Register
UART_BRGR

Address:

0x400E0620

Access:

Read-write

–

CD
7

4
CD

• CD: Clock Divisor
0 = Baud Rate Clock is disabled
1 to 65,535 = MCK / (CD x 16)

687
6430F–ATARM–21-Feb-12

688

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35. Universal Synchronous Asynchronous Receiver Transmitter (USART)
35.1

Description
The Universal Synchronous Asynchronous Receiver Transceiver (USART) provides one full
duplex universal synchronous asynchronous serial link. Data frame format is widely programmable (data length, parity, number of stop bits) to support a maximum of standards. The receiver
implements parity error, framing error and overrun error detection. The receiver time-out enables
handling variable-length frames and the transmitter timeguard facilitates communications with
slow remote devices. Multidrop communications are also supported through address bit handling in reception and transmission.
The USART features three test modes: remote loopback, local loopback and automatic echo.
The USART supports specific operating modes providing interfaces on RS485 and SPI buses,
with ISO7816 T = 0 or T = 1 smart card slots, infrared transceivers and connection to modem
ports. The hardware handshaking feature enables an out-of-band flow control by automatic
management of the pins RTS and CTS.
The USART supports the connection to the Peripheral DMA Controller, which enables data
transfers to the transmitter and from the receiver. The PDC provides chained buffer management without any intervention of the processor.

35.2

Embedded Characteristics
• Programmable Baud Rate Generator
• 5- to 9-bit Full-duplex Synchronous or Asynchronous Serial Communications
– 1, 1.5 or 2 Stop Bits in Asynchronous Mode or 1 or 2 Stop Bits in Synchronous Mode
– Parity Generation and Error Detection
– Framing Error Detection, Overrun Error Detection
– MSB- or LSB-first
– Optional Break Generation and Detection
– By 8 or by 16 Over-sampling Receiver Frequency
– Optional Hardware Handshaking RTS-CTS
– Optional Modem Signal Management DTR-DSR-DCD-RI
– Receiver Time-out and Transmitter Timeguard
– Optional Multidrop Mode with Address Generation and Detection
• RS485 with Driver Control Signal
• ISO7816, T = 0 or T = 1 Protocols for Interfacing with Smart Cards
– NACK Handling, Error Counter with Repetition and Iteration Limit
• IrDA Modulation and Demodulation
– Communication at up to 115.2 Kbps
• SPI Mode
– Master or Slave
– Serial Clock Programmable Phase and Polarity
– SPI Serial Clock (SCK) Frequency up to Internal Clock Frequency MCK/6

689
6430F–ATARM–21-Feb-12

• Test Modes
– Remote Loopback, Local Loopback, Automatic Echo
• Supports Connection of:
– Two Peripheral DMA Controller Channels (PDC)
• Offers Buffer Transfer without Processor Intervention

35.3

Block Diagram
Figure 35-1. USART Block Diagram
(Peripheral) DMA
Controller
Channel

Channel

PIO
Controller

USART

RXD
Receiver
RTS
Interrupt
Controller

USART
Interrupt

TXD
Transmitter
CTS
DTR

PMC

Modem
Signals
Control

MCK

DIV

DSR
DCD

MCK/DIV

RI
SLCK

Baud Rate
Generator

SCK

User Interface

APB

690

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series

Table 35-1.

35.4

SPI Operating Mode

USART

SPI Slave

SPI Master

RXD

MOSI

MISO

TXD

MISO

MOSI

RTS

–

CTS

–

Application Block Diagram
Figure 35-2. Application Block Diagram
IrLAP

PPP
Modem
Driver

Serial
Driver

Field Bus
Driver

EMV
Driver

IrDA
Driver

SPI
Driver

USART

RS232
Drivers

RS485
Drivers

Serial
Port

Differential
Bus

Smart
Card
Slot

IrDA
Transceivers

SPI
Transceiver

Modem
PSTN

691
6430F–ATARM–21-Feb-12

35.5

I/O Lines Description

Table 35-2.

I/O Line Description

Name

Description

Type

SCK

Serial Clock

I/O

TXD

Transmit Serial Data
or Master Out Slave In (MOSI) in SPI Master Mode
or Master In Slave Out (MISO) in SPI Slave Mode

I/O

RXD

Receive Serial Data
or Master In Slave Out (MISO) in SPI Master Mode
or Master Out Slave In (MOSI) in SPI Slave Mode

Input

Ring Indicator

Input

Low

DSR

Data Set Ready

Input

Low

DCD

Data Carrier Detect

Input

Low

DTR

Data Terminal Ready

Output

Low

CTS

Clear to Send
or Slave Select (NSS) in SPI Slave Mode

Input

Low

RTS

Request to Send
or Slave Select (NSS) in SPI Master Mode

Output

Low

692

Active Level

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.6
35.6.1

Product Dependencies
I/O Lines
The pins used for interfacing the USART may be multiplexed with the PIO lines. The programmer must first program the PIO controller to assign the desired USART pins to their peripheral
function. If I/O lines of the USART are not used by the application, they can be used for other
purposes by the PIO Controller.
To prevent the TXD line from falling when the USART is disabled, the use of an internal pull up
is mandatory. If the hardware handshaking feature or Modem mode is used, the internal pull up
on TXD must also be enabled.
All the pins of the modems may or may not be implemented on the USART. On USARTs not
equipped with the corresponding pin, the associated control bits and statuses have no effect on
the behavior of the USART.
Table 35-3.

I/O Lines

Instance

Signal

I/O Line

Peripheral

USART0

CTS0

PB8

USART0

DCD0

PB11

USART0

DSR0

PB10

USART0

DTR0

PB9

USART0

RI0

PB12

USART0

RTS0

PB7

USART0

RXD0

PA19

USART0

SCK0

PA17

USART0

TXD0

PA18

USART1

CTS1

PA23

USART1

RTS1

PA22

USART1

RXD1

PA21

USART1

SCK1

PA24

USART1

TXD1

PA20

USART2

CTS2

PB22

USART2

RTS2

PB21

USART2

RXD2

PA23

USART2

SCK2

PA25

USART2

TXD2

PA22

USART3

CTS3

PC10

USART3

RTS3

PC11

USART3

RXD3

PC13

USART3

SCK3

PC19

USART3

TXD3

PC12

693
6430F–ATARM–21-Feb-12

35.6.2

Power Management
The USART is not continuously clocked. The programmer must first enable the USART Clock in
the Power Management Controller (PMC) before using the USART. However, if the application
does not require USART operations, the USART clock can be stopped when not needed and be
restarted later. In this case, the USART will resume its operations where it left off.
Configuring the USART does not require the USART clock to be enabled.

35.6.3

Interrupt
The USART interrupt line is connected on one of the internal sources of the Interrupt Controller.
Using the USART interrupt requires the Interrupt Controller to be programmed first. Note that it is
not recommended to use the USART interrupt line in edge sensitive mode.

694

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.7

Functional Description
The USART is capable of managing several types of serial synchronous or asynchronous
communications.
It supports the following communication modes:
• 5- to 9-bit full-duplex asynchronous serial communication
– MSB- or LSB-first
– 1, 1.5 or 2 stop bits
– Parity even, odd, marked, space or none
– By 8 or by 16 over-sampling receiver frequency
– Optional hardware handshaking
– Optional modem signals management
– Optional break management
– Optional multidrop serial communication
• High-speed 5- to 9-bit full-duplex synchronous serial communication
– MSB- or LSB-first
– 1 or 2 stop bits
– Parity even, odd, marked, space or none
– By 8 or by 16 over-sampling frequency
– Optional hardware handshaking
– Optional modem signals management
– Optional break management
– Optional multidrop serial communication
• RS485 with driver control signal
• ISO7816, T0 or T1 protocols for interfacing with smart cards
– NACK handling, error counter with repetition and iteration limit, inverted data.
• InfraRed IrDA Modulation and Demodulation
• SPI Mode
– Master or Slave
– Serial Clock Programmable Phase and Polarity
– SPI Serial Clock (SCK) Frequency up to Internal Clock Frequency MCK/6
• Test modes
– Remote loopback, local loopback, automatic echo

695
6430F–ATARM–21-Feb-12

35.7.1

Baud Rate Generator
The Baud Rate Generator provides the bit period clock named the Baud Rate Clock to both the
receiver and the transmitter.
The Baud Rate Generator clock source can be selected by setting the USCLKS field in the Mode
Register (US_MR) between:
• the Master Clock MCK
• a division of the Master Clock, the divider being product dependent, but generally set to 8
• the external clock, available on the SCK pin
The Baud Rate Generator is based upon a 16-bit divider, which is programmed with the CD field
of the Baud Rate Generator Register (US_BRGR). If CD is programmed to 0, the Baud Rate
Generator does not generate any clock. If CD is programmed to 1, the divider is bypassed and
becomes inactive.
If the external SCK clock is selected, the duration of the low and high levels of the signal provided on the SCK pin must be longer than a Master Clock (MCK) period. The frequency of the
signal provided on SCK must be at least 3 times lower than MCK in USART mode, or 6 in SPI
mode.
Figure 35-3. Baud Rate Generator
USCLKS
MCK
MCK/DIV
SCK

Reserved

CD
CD

SCK

0
1
2

16-bit Counter
FIDI

1
0

SYNC

OVER

Sampling
Divider

0
Baud Rate
Clock

1
1
SYNC
USCLKS = 3

35.7.1.1

Sampling
Clock

Baud Rate in Asynchronous Mode
If the USART is programmed to operate in asynchronous mode, the selected clock is first
divided by CD, which is field programmed in the Baud Rate Generator Register (US_BRGR).
The resulting clock is provided to the receiver as a sampling clock and then divided by 16 or 8,
depending on the programming of the OVER bit in US_MR.
If OVER is set to 1, the receiver sampling is 8 times higher than the baud rate clock. If OVER is
cleared, the sampling is performed at 16 times the baud rate clock.

The following formula performs the calculation of the Baud Rate.

696

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
SelectedClock
Baudrate = -------------------------------------------( 8 ( 2 – Over )CD )

This gives a maximum baud rate of MCK divided by 8, assuming that MCK is the highest possible clock and that OVER is programmed to 1.
Baud Rate Calculation Example
Table 35-4 shows calculations of CD to obtain a baud rate at 38400 bauds for different source
clock frequencies. This table also shows the actual resulting baud rate and the error.
Table 35-4.

Baud Rate Example (OVER = 0)

Source Clock

Expected Baud
Rate

MHz

Bit/s

3 686 400

38 400

6.00

38 400.00

0.00%

4 915 200

38 400

8.00

38 400.00

0.00%

5 000 000

38 400

8.14

39 062.50

1.70%

7 372 800

38 400

12.00

38 400.00

0.00%

8 000 000

38 400

13.02

38 461.54

0.16%

12 000 000

38 400

19.53

37 500.00

2.40%

12 288 000

38 400

20.00

38 400.00

0.00%

14 318 180

38 400

23.30

38 908.10

1.31%

14 745 600

38 400

24.00

38 400.00

0.00%

18 432 000

38 400

30.00

38 400.00

0.00%

24 000 000

38 400

39.06

38 461.54

0.16%

24 576 000

38 400

40.00

38 400.00

0.00%

25 000 000

38 400

40.69

38 109.76

0.76%

32 000 000

38 400

52.08

38 461.54

0.16%

32 768 000

38 400

53.33

38 641.51

0.63%

33 000 000

38 400

53.71

38 194.44

0.54%

40 000 000

38 400

65.10

38 461.54

0.16%

50 000 000

38 400

81.38

38 580.25

0.47%

Calculation Result

Actual Baud Rate

Error

Bit/s

The baud rate is calculated with the following formula:
BaudRate = MCK ⁄ CD × 16
The baud rate error is calculated with the following formula. It is not recommended to work with
an error higher than 5%.
ExpectedBaudRate
Error = 1 – ⎛⎝ ---------------------------------------------------⎞⎠
ActualBaudRate

697
6430F–ATARM–21-Feb-12

35.7.1.2

Fractional Baud Rate in Asynchronous Mode
The Baud Rate generator previously defined is subject to the following limitation: the output frequency changes by only integer multiples of the reference frequency. An approach to this
problem is to integrate a fractional N clock generator that has a high resolution. The generator
architecture is modified to obtain Baud Rate changes by a fraction of the reference source clock.
This fractional part is programmed with the FP field in the Baud Rate Generator Register
(US_BRGR). If FP is not 0, the fractional part is activated. The resolution is one eighth of the
clock divider. This feature is only available when using USART normal mode. The fractional
Baud Rate is calculated using the following formula:

SelectedClock
Baudrate = ---------------------------------------------------------------⎛ 8 ( 2 – Over ) ⎛ CD + FP
-------⎞ ⎞
⎝
⎝
8 ⎠⎠

The modified architecture is presented below:
Figure 35-4. Fractional Baud Rate Generator
FP

USCLKS

Modulus
Control
FP

MCK
MCK/DIV
SCK

Reserved

SCK

0
1
2

16-bit Counter

glitch-free
logic

1
0

FIDI

>1
0

SYNC

OVER
Sampling
Divider

0
Baud Rate
Clock

1
1
SYNC
USCLKS = 3

35.7.1.3

Sampling
Clock

Baud Rate in Synchronous Mode or SPI Mode
If the USART is programmed to operate in synchronous mode, the selected clock is simply
divided by the field CD in US_BRGR.
BaudRate = SelectedClock
-------------------------------------CD

In synchronous mode, if the external clock is selected (USCLKS = 3), the clock is provided
directly by the signal on the USART SCK pin. No division is active. The value written in
US_BRGR has no effect. The external clock frequency must be at least 3 times lower than the

698

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
system clock. In synchronous mode master (USCLKS = 0 or 1, CLK0 set to 1), the receive part
limits the SCK maximum frequency to MCK/3 in USART mode, or MCK/6 in SPI mode.
When either the external clock SCK or the internal clock divided (MCK/DIV) is selected, the
value programmed in CD must be even if the user has to ensure a 50:50 mark/space ratio on the
SCK pin. If the internal clock MCK is selected, the Baud Rate Generator ensures a 50:50 duty
cycle on the SCK pin, even if the value programmed in CD is odd.
35.7.1.4

Baud Rate in ISO 7816 Mode
The ISO7816 specification defines the bit rate with the following formula:
Di
B = ------ × f
Fi

where:
• B is the bit rate
• Di is the bit-rate adjustment factor
• Fi is the clock frequency division factor
• f is the ISO7816 clock frequency (Hz)
Di is a binary value encoded on a 4-bit field, named DI, as represented in Table 35-5.
Table 35-5.

Binary and Decimal Values for Di

DI field

0001

0010

0011

0100

0101

0110

1000

1001

Di (decimal)

Fi is a binary value encoded on a 4-bit field, named FI, as represented in Table 35-6.
Table 35-6.

Binary and Decimal Values for Fi

FI field

0000

0001

0010

0011

0100

0101

0110

1001

1010

1011

1100

1101

Fi (decimal)

372

558

744

1116

1488

1860

512

768

1024

1536

2048

Table 35-7 shows the resulting Fi/Di Ratio, which is the ratio between the ISO7816 clock and the
baud rate clock.
Table 35-7.

Possible Values for the Fi/Di Ratio

Fi/Di

372

558

774

1116

1488

1806

512

768

1024

1536

2048

372

558

744

1116

1488

1860

512

768

1024

1536

2048

186

279

372

558

744

930

256

384

512

768

1024

139.5

186

279

372

465

128

192

256

384

512

46.5

69.75

139.5

186

232.5

128

192

256

23.25

34.87

46.5

69.75

116.2

128

11.62

17.43

23.25

34.87

46.5

58.13

46.5

124

155

42.66

85.33

128

170.6

18.6

27.9

37.2

55.8

74.4

25.6

38.4

51.2

76.8

102.4

If the USART is configured in ISO7816 Mode, the clock selected by the USCLKS field in the
Mode Register (US_MR) is first divided by the value programmed in the field CD in the Baud

699
6430F–ATARM–21-Feb-12

Rate Generator Register (US_BRGR). The resulting clock can be provided to the SCK pin to
feed the smart card clock inputs. This means that the CLKO bit can be set in US_MR.
This clock is then divided by the value programmed in the FI_DI_RATIO field in the FI_DI_Ratio
register (US_FIDI). This is performed by the Sampling Divider, which performs a division by up
to 2047 in ISO7816 Mode. The non-integer values of the Fi/Di Ratio are not supported and the
user must program the FI_DI_RATIO field to a value as close as possible to the expected value.
The FI_DI_RATIO field resets to the value 0x174 (372 in decimal) and is the most common
divider between the ISO7816 clock and the bit rate (Fi = 372, Di = 1).
Figure 35-5 shows the relation between the Elementary Time Unit, corresponding to a bit time,
and the ISO 7816 clock.
Figure 35-5. Elementary Time Unit (ETU)
FI_DI_RATIO
ISO7816 Clock Cycles
ISO7816 Clock
on SCK
ISO7816 I/O Line
on TXD

1 ETU

35.7.2

Receiver and Transmitter Control
After reset, the receiver is disabled. The user must enable the receiver by setting the RXEN bit
in the Control Register (US_CR). However, the receiver registers can be programmed before the
receiver clock is enabled.
After reset, the transmitter is disabled. The user must enable it by setting the TXEN bit in the
Control Register (US_CR). However, the transmitter registers can be programmed before being
enabled.
The Receiver and the Transmitter can be enabled together or independently.
At any time, the software can perform a reset on the receiver or the transmitter of the USART by
setting the corresponding bit, RSTRX and RSTTX respectively, in the Control Register
(US_CR). The software resets clear the status flag and reset internal state machines but the
user interface configuration registers hold the value configured prior to software reset. Regardless of what the receiver or the transmitter is performing, the communication is immediately
stopped.
The user can also independently disable the receiver or the transmitter by setting RXDIS and
TXDIS respectively in US_CR. If the receiver is disabled during a character reception, the
USART waits until the end of reception of the current character, then the reception is stopped. If
the transmitter is disabled while it is operating, the USART waits the end of transmission of both
the current character and character being stored in the Transmit Holding Register (US_THR). If
a timeguard is programmed, it is handled normally.

700

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.7.3
35.7.3.1

Synchronous and Asynchronous Modes
Transmitter Operations
The transmitter performs the same in both synchronous and asynchronous operating modes
(SYNC = 0 or SYNC = 1). One start bit, up to 9 data bits, one optional parity bit and up to two
stop bits are successively shifted out on the TXD pin at each falling edge of the programmed
serial clock.
The number of data bits is selected by the CHRL field and the MODE 9 bit in the Mode Register
(US_MR). Nine bits are selected by setting the MODE 9 bit regardless of the CHRL field. The
parity bit is set according to the PAR field in US_MR. The even, odd, space, marked or none
parity bit can be configured. The MSBF field in US_MR configures which data bit is sent first. If
written to 1, the most significant bit is sent first. If written to 0, the less significant bit is sent first.
The number of stop bits is selected by the NBSTOP field in US_MR. The 1.5 stop bit is supported in asynchronous mode only.
Figure 35-6. Character Transmit
Example: 8-bit, Parity Enabled One Stop
Baud Rate
Clock
TXD

Start
Bit

Parity
Bit

Stop
Bit

The characters are sent by writing in the Transmit Holding Register (US_THR). The transmitter
reports two status bits in the Channel Status Register (US_CSR): TXRDY (Transmitter Ready),
which indicates that US_THR is empty and TXEMPTY, which indicates that all the characters
written in US_THR have been processed. When the current character processing is completed,
the last character written in US_THR is transferred into the Shift Register of the transmitter and
US_THR becomes empty, thus TXRDY rises.
Both TXRDY and TXEMPTY bits are low when the transmitter is disabled. Writing a character in
US_THR while TXRDY is low has no effect and the written character is lost.
Figure 35-7. Transmitter Status
Baud Rate
Clock
TXD
Start
D0
Bit

Parity Stop Start
D0
Bit Bit Bit

Parity Stop
Bit Bit

Write
US_THR
TXRDY

TXEMPTY

701
6430F–ATARM–21-Feb-12

35.7.3.2

Manchester Encoder
When the Manchester encoder is in use, characters transmitted through the USART are
encoded based on biphase Manchester II format. To enable this mode, set the MAN field in the
US_MR register to 1. Depending on polarity configuration, a logic level (zero or one), is transmitted as a coded signal one-to-zero or zero-to-one. Thus, a transition always occurs at the
midpoint of each bit time. It consumes more bandwidth than the original NRZ signal (2x) but the
receiver has more error control since the expected input must show a change at the center of a
bit cell. An example of Manchester encoded sequence is: the byte 0xB1 or 10110001 encodes
to 10 01 10 10 01 01 01 10, assuming the default polarity of the encoder. Figure 35-8 illustrates
this coding scheme.
Figure 35-8. NRZ to Manchester Encoding
NRZ
encoded
data
Manchester
encoded
data

Txd

The Manchester encoded character can also be encapsulated by adding both a configurable
preamble and a start frame delimiter pattern. Depending on the configuration, the preamble is a
training sequence, composed of a pre-defined pattern with a programmable length from 1 to 15
bit times. If the preamble length is set to 0, the preamble waveform is not generated prior to any
character. The preamble pattern is chosen among the following sequences: ALL_ONE,
ALL_ZERO, ONE_ZERO or ZERO_ONE, writing the field TX_PP in the US_MAN register, the
field TX_PL is used to configure the preamble length. Figure 35-9 illustrates and defines the
valid patterns. To improve flexibility, the encoding scheme can be configured using the
TX_MPOL field in the US_MAN register. If the TX_MPOL field is set to zero (default), a logic
zero is encoded with a zero-to-one transition and a logic one is encoded with a one-to-zero transition. If the TX_MPOL field is set to one, a logic one is encoded with a one-to-zero transition
and a logic zero is encoded with a zero-to-one transition.

702

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 35-9. Preamble Patterns, Default Polarity Assumed
Manchester
encoded
data

Txd

SFD

DATA

SFD

DATA

SFD

DATA

SFD

DATA

8 bit width "ALL_ONE" Preamble

Manchester
encoded
data

Txd

8 bit width "ALL_ZERO" Preamble
Manchester
encoded
data

Txd

8 bit width "ZERO_ONE" Preamble

Manchester
encoded
data

Txd

8 bit width "ONE_ZERO" Preamble

A start frame delimiter is to be configured using the ONEBIT field in the US_MR register. It consists of a user-defined pattern that indicates the beginning of a valid data. Figure 35-10
illustrates these patterns. If the start frame delimiter, also known as start bit, is one bit, (ONEBIT
to 1), a logic zero is Manchester encoded and indicates that a new character is being sent serially on the line. If the start frame delimiter is a synchronization pattern also referred to as sync
(ONEBIT to 0), a sequence of 3 bit times is sent serially on the line to indicate the start of a new
character. The sync waveform is in itself an invalid Manchester waveform as the transition
occurs at the middle of the second bit time. Two distinct sync patterns are used: the command
sync and the data sync. The command sync has a logic one level for one and a half bit times,
then a transition to logic zero for the second one and a half bit times. If the MODSYNC field in
the US_MR register is set to 1, the next character is a command. If it is set to 0, the next character is a data. When direct memory access is used, the MODSYNC field can be immediately
updated with a modified character located in memory. To enable this mode, VAR_SYNC field in
US_MR register must be set to 1. In this case, the MODSYNC field in US_MR is bypassed and
the sync configuration is held in the TXSYNH in the US_THR register. The USART character format is modified and includes sync information.

703
6430F–ATARM–21-Feb-12

Figure 35-10. Start Frame Delimiter
Preamble Length
is set to 0
SFD
Manchester
encoded
data

DATA

Txd

One bit start frame delimiter
SFD
Manchester
encoded
data

DATA

Txd

SFD
Manchester
encoded
data

Command Sync
start frame delimiter
DATA

Txd

Data Sync
start frame delimiter

Drift Compensation
Drift compensation is available only in 16X oversampling mode. An hardware recovery system
allows a larger clock drift. To enable the hardware system, the bit in the USART_MAN register
must be set. If the RXD edge is one 16X clock cycle from the expected edge, this is considered
as normal jitter and no corrective actions is taken. If the RXD event is between 4 and 2 clock
cycles before the expected edge, then the current period is shortened by one clock cycle. If the
RXD event is between 2 and 3 clock cycles after the expected edge, then the current period is
lengthened by one clock cycle. These intervals are considered to be drift and so corrective
actions are automatically taken.
Figure 35-11. Bit Resynchronization
Oversampling
16x Clock
RXD

Sampling
point
Expected edge
Synchro.
Error

35.7.3.3

704

Synchro.
Jump

Tolerance

Sync
Jump

Synchro.
Error

Asynchronous Receiver
If the USART is programmed in asynchronous operating mode (SYNC = 0), the receiver oversamples the RXD input line. The oversampling is either 16 or 8 times the Baud Rate clock,
depending on the OVER bit in the Mode Register (US_MR).

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The receiver samples the RXD line. If the line is sampled during one half of a bit time to 0, a start
bit is detected and data, parity and stop bits are successively sampled on the bit rate clock.
If the oversampling is 16, (OVER to 0), a start is detected at the eighth sample to 0. Then, data
bits, parity bit and stop bit are sampled on each 16 sampling clock cycle. If the oversampling is 8
(OVER to 1), a start bit is detected at the fourth sample to 0. Then, data bits, parity bit and stop
bit are sampled on each 8 sampling clock cycle.
The number of data bits, first bit sent and parity mode are selected by the same fields and bits
as the transmitter, i.e. respectively CHRL, MODE9, MSBF and PAR. For the synchronization
mechanism only, the number of stop bits has no effect on the receiver as it considers only one
stop bit, regardless of the field NBSTOP, so that resynchronization between the receiver and the
transmitter can occur. Moreover, as soon as the stop bit is sampled, the receiver starts looking
for a new start bit so that resynchronization can also be accomplished when the transmitter is
operating with one stop bit.
Figure 35-12 and Figure 35-13 illustrate start detection and character reception when USART
operates in asynchronous mode.
Figure 35-12. Asynchronous Start Detection
Baud Rate
Clock
Sampling
Clock (x16)
RXD
Sampling
1

9 10 11 12 13 14 15 16

D0
Sampling

Start
Detection
RXD
Sampling

0 1
Start
Rejection

Figure 35-13. Asynchronous Character Reception
Example: 8-bit, Parity Enabled

Baud Rate
Clock
RXD
Start
Detection

16
16
16
16
16
16
16
16
16
16
samples samples samples samples samples samples samples samples samples samples

Parity
Bit

Stop
Bit

705
6430F–ATARM–21-Feb-12

35.7.3.4

Manchester Decoder
When the MAN field in US_MR register is set to 1, the Manchester decoder is enabled. The
decoder performs both preamble and start frame delimiter detection. One input line is dedicated
to Manchester encoded input data.
An optional preamble sequence can be defined, its length is user-defined and totally independent of the emitter side. Use RX_PL in US_MAN register to configure the length of the preamble
sequence. If the length is set to 0, no preamble is detected and the function is disabled. In addition, the polarity of the input stream is programmable with RX_MPOL field in US_MAN register.
Depending on the desired application the preamble pattern matching is to be defined via the
RX_PP field in US_MAN. See Figure 35-9 for available preamble patterns.
Unlike preamble, the start frame delimiter is shared between Manchester Encoder and Decoder.
So, if ONEBIT field is set to 1, only a zero encoded Manchester can be detected as a valid start
frame delimiter. If ONEBIT is set to 0, only a sync pattern is detected as a valid start frame
delimiter. Decoder operates by detecting transition on incoming stream. If RXD is sampled during one quarter of a bit time to zero, a start bit is detected. See Figure 35-14. The sample pulse
rejection mechanism applies.
Figure 35-14. Asynchronous Start Bit Detection
Sampling
Clock
(16 x)
Manchester
encoded
data

Txd
Start
Detection
1

The receiver is activated and starts Preamble and Frame Delimiter detection, sampling the data
at one quarter and then three quarters. If a valid preamble pattern or start frame delimiter is
detected, the receiver continues decoding with the same synchronization. If the stream does not
match a valid pattern or a valid start frame delimiter, the receiver re-synchronizes on the next
valid edge.The minimum time threshold to estimate the bit value is three quarters of a bit time.
If a valid preamble (if used) followed with a valid start frame delimiter is detected, the incoming
stream is decoded into NRZ data and passed to USART for processing. Figure 35-15 illustrates
Manchester pattern mismatch. When incoming data stream is passed to the USART, the
receiver is also able to detect Manchester code violation. A code violation is a lack of transition
in the middle of a bit cell. In this case, MANE flag in US_CSR register is raised. It is cleared by
writing the Control Register (US_CR) with the RSTSTA bit to 1. See Figure 35-16 for an example of Manchester error detection during data phase.

706

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 35-15. Preamble Pattern Mismatch
Preamble Mismatch
Manchester coding error

Manchester
encoded
data

Preamble Mismatch
invalid pattern

SFD

Txd

DATA

Preamble Length is set to 8

Figure 35-16. Manchester Error Flag
Preamble Length
is set to 4
Elementary character bit time
SFD
Manchester
encoded
data

Txd
Entering USART character area

sampling points

Preamble subpacket
and Start Frame Delimiter
were successfully
decoded

Manchester
Coding Error
detected

When the start frame delimiter is a sync pattern (ONEBIT field to 0), both command and data
delimiter are supported. If a valid sync is detected, the received character is written as RXCHR
field in the US_RHR register and the RXSYNH is updated. RXCHR is set to 1 when the received
character is a command, and it is set to 0 if the received character is a data. This mechanism
alleviates and simplifies the direct memory access as the character contains its own sync field in
the same register.
As the decoder is setup to be used in unipolar mode, the first bit of the frame has to be a zero-toone transition.
35.7.3.5

Radio Interface: Manchester Encoded USART Application
This section describes low data rate RF transmission systems and their integration with a Manchester encoded USART. These systems are based on transmitter and receiver ICs that support
ASK and FSK modulation schemes.
The goal is to perform full duplex radio transmission of characters using two different frequency
carriers. See the configuration in Figure 35-17.

707
6430F–ATARM–21-Feb-12

Figure 35-17. Manchester Encoded Characters RF Transmission

Fup frequency Carrier
ASK/FSK
Upstream Receiver

Upstream
Emitter

LNA
VCO
RF filter
Demod

Serial
Configuration
Interface

control
Fdown frequency Carrier

bi-dir
line

Manchester
decoder

USART
Receiver

Manchester
encoder

USART
Emitter

ASK/FSK
downstream transmitter

Downstream
Receiver

PA
RF filter
Mod
VCO

control

The USART module is configured as a Manchester encoder/decoder. Looking at the downstream communication channel, Manchester encoded characters are serially sent to the RF
emitter. This may also include a user defined preamble and a start frame delimiter. Mostly, preamble is used in the RF receiver to distinguish between a valid data from a transmitter and
signals due to noise. The Manchester stream is then modulated. See Figure 35-18 for an example of ASK modulation scheme. When a logic one is sent to the ASK modulator, the power
amplifier, referred to as PA, is enabled and transmits an RF signal at downstream frequency.
When a logic zero is transmitted, the RF signal is turned off. If the FSK modulator is activated,
two different frequencies are used to transmit data. When a logic 1 is sent, the modulator outputs an RF signal at frequency F0 and switches to F1 if the data sent is a 0. See Figure 35-19.
From the receiver side, another carrier frequency is used. The RF receiver performs a bit check
operation examining demodulated data stream. If a valid pattern is detected, the receiver
switches to receiving mode. The demodulated stream is sent to the Manchester decoder.
Because of bit checking inside RF IC, the data transferred to the microcontroller is reduced by a
user-defined number of bits. The Manchester preamble length is to be defined in accordance
with the RF IC configuration.
Figure 35-18. ASK Modulator Output
1

NRZ stream
Manchester
encoded
data
default polarity
unipolar output

Txd

ASK Modulator
Output
Uptstream Frequency F0

708

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 35-19. FSK Modulator Output
1

NRZ stream
Manchester
encoded
data
default polarity
unipolar output

Txd

FSK Modulator
Output
Uptstream Frequencies
[F0, F0+offset]

35.7.3.6

Synchronous Receiver
In synchronous mode (SYNC = 1), the receiver samples the RXD signal on each rising edge of
the Baud Rate Clock. If a low level is detected, it is considered as a start. All data bits, the parity
bit and the stop bits are sampled and the receiver waits for the next start bit. Synchronous mode
operations provide a high speed transfer capability.
Configuration fields and bits are the same as in asynchronous mode.
Figure 35-20 illustrates a character reception in synchronous mode.
Figure 35-20. Synchronous Mode Character Reception
Example: 8-bit, Parity Enabled 1 Stop
Baud Rate
Clock

RXD
Sampling

Start

Stop Bit

D7
Parity Bit

35.7.3.7

Receiver Operations
When a character reception is completed, it is transferred to the Receive Holding Register
(US_RHR) and the RXRDY bit in the Status Register (US_CSR) rises. If a character is completed while the RXRDY is set, the OVRE (Overrun Error) bit is set. The last character is
transferred into US_RHR and overwrites the previous one. The OVRE bit is cleared by writing
the Control Register (US_CR) with the RSTSTA (Reset Status) bit to 1.

709
6430F–ATARM–21-Feb-12

Figure 35-21. Receiver Status
Baud Rate
Clock
RXD
Start
D0
Bit

Parity Stop Start
D0
Bit Bit Bit

Parity Stop
Bit Bit

RSTSTA = 1

Write
US_CR
Read
US_RHR

RXRDY
OVRE

710

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.7.3.8

Parity
The USART supports five parity modes selected by programming the PAR field in the Mode
Register (US_MR). The PAR field also enables the Multidrop mode, see “Multidrop Mode” on
page 712. Even and odd parity bit generation and error detection are supported.
If even parity is selected, the parity generator of the transmitter drives the parity bit to 0 if a number of 1s in the character data bit is even, and to 1 if the number of 1s is odd. Accordingly, the
receiver parity checker counts the number of received 1s and reports a parity error if the sampled parity bit does not correspond. If odd parity is selected, the parity generator of the
transmitter drives the parity bit to 1 if a number of 1s in the character data bit is even, and to 0 if
the number of 1s is odd. Accordingly, the receiver parity checker counts the number of received
1s and reports a parity error if the sampled parity bit does not correspond. If the mark parity is
used, the parity generator of the transmitter drives the parity bit to 1 for all characters. The
receiver parity checker reports an error if the parity bit is sampled to 0. If the space parity is
used, the parity generator of the transmitter drives the parity bit to 0 for all characters. The
receiver parity checker reports an error if the parity bit is sampled to 1. If parity is disabled, the
transmitter does not generate any parity bit and the receiver does not report any parity error.
Table 35-8 shows an example of the parity bit for the character 0x41 (character ASCII “A”)
depending on the configuration of the USART. Because there are two bits to 1, 1 bit is added
when a parity is odd, or 0 is added when a parity is even.
Table 35-8.

Parity Bit Examples

Character

Hexa

Binary

Parity Bit

Parity Mode

0x41

0100 0001

Odd

0x41

0100 0001

Even

0x41

0100 0001

Mark

0x41

0100 0001

Space

0x41

0100 0001

None

When the receiver detects a parity error, it sets the PARE (Parity Error) bit in the Channel Status
Register (US_CSR). The PARE bit can be cleared by writing the Control Register (US_CR) with
the RSTSTA bit to 1. Figure 35-22 illustrates the parity bit status setting and clearing.

711
6430F–ATARM–21-Feb-12

Figure 35-22. Parity Error

Baud Rate
Clock
RXD
Start
D0
Bit

Bad Stop
Parity Bit
Bit

RSTSTA = 1

Write
US_CR
PARE

RXRDY

35.7.3.9

Multidrop Mode
If the PAR field in the Mode Register (US_MR) is programmed to the value 0x6 or 0x07, the
USART runs in Multidrop Mode. This mode differentiates the data characters and the address
characters. Data is transmitted with the parity bit to 0 and addresses are transmitted with the
parity bit to 1.
If the USART is configured in multidrop mode, the receiver sets the PARE parity error bit when
the parity bit is high and the transmitter is able to send a character with the parity bit high when
the Control Register is written with the SENDA bit to 1.
To handle parity error, the PARE bit is cleared when the Control Register is written with the bit
RSTSTA to 1.
The transmitter sends an address byte (parity bit set) when SENDA is written to US_CR. In this
case, the next byte written to US_THR is transmitted as an address. Any character written in
US_THR without having written the command SENDA is transmitted normally with the parity to
0.

35.7.3.10

Transmitter Timeguard
The timeguard feature enables the USART interface with slow remote devices.
The timeguard function enables the transmitter to insert an idle state on the TXD line between
two characters. This idle state actually acts as a long stop bit.
The duration of the idle state is programmed in the TG field of the Transmitter Timeguard Register (US_TTGR). When this field is programmed to zero no timeguard is generated. Otherwise,
the transmitter holds a high level on TXD after each transmitted byte during the number of bit
periods programmed in TG in addition to the number of stop bits.
As illustrated in Figure 35-23, the behavior of TXRDY and TXEMPTY status bits is modified by
the programming of a timeguard. TXRDY rises only when the start bit of the next character is
sent, and thus remains to 0 during the timeguard transmission if a character has been written in
US_THR. TXEMPTY remains low until the timeguard transmission is completed as the timeguard is part of the current character being transmitted.

712

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 35-23. Timeguard Operations
TG = 4

TG = 4
Baud Rate
Clock
TXD
Start
D0
Bit

Parity Stop
Bit Bit

Start
D0
Bit

Parity Stop
Bit Bit

Write
US_THR
TXRDY

TXEMPTY

Table 35-9 indicates the maximum length of a timeguard period that the transmitter can handle
in relation to the function of the Baud Rate.
Table 35-9.

35.7.3.11

Maximum Timeguard Length Depending on Baud Rate

Baud Rate

Bit time

Timeguard

Bit/sec

μs

1 200

833

212.50

9 600

104

26.56

14400

69.4

17.71

19200

52.1

13.28

28800

34.7

8.85

33400

29.9

7.63

56000

17.9

4.55

57600

17.4

4.43

115200

8.7

2.21

Receiver Time-out
The Receiver Time-out provides support in handling variable-length frames. This feature detects
an idle condition on the RXD line. When a time-out is detected, the bit TIMEOUT in the Channel
Status Register (US_CSR) rises and can generate an interrupt, thus indicating to the driver an
end of frame.
The time-out delay period (during which the receiver waits for a new character) is programmed
in the TO field of the Receiver Time-out Register (US_RTOR). If the TO field is programmed to
0, the Receiver Time-out is disabled and no time-out is detected. The TIMEOUT bit in US_CSR
remains to 0. Otherwise, the receiver loads a 16-bit counter with the value programmed in TO.
This counter is decremented at each bit period and reloaded each time a new character is
received. If the counter reaches 0, the TIMEOUT bit in the Status Register rises. Then, the user
can either:
• Stop the counter clock until a new character is received. This is performed by writing the
Control Register (US_CR) with the STTTO (Start Time-out) bit to 1. In this case, the idle state

713
6430F–ATARM–21-Feb-12

on RXD before a new character is received will not provide a time-out. This prevents having
to handle an interrupt before a character is received and allows waiting for the next idle state
on RXD after a frame is received.
• Obtain an interrupt while no character is received. This is performed by writing US_CR with
the RETTO (Reload and Start Time-out) bit to 1. If RETTO is performed, the counter starts
counting down immediately from the value TO. This enables generation of a periodic interrupt
so that a user time-out can be handled, for example when no key is pressed on a keyboard.
If STTTO is performed, the counter clock is stopped until a first character is received. The idle
state on RXD before the start of the frame does not provide a time-out. This prevents having to
obtain a periodic interrupt and enables a wait of the end of frame when the idle state on RXD is
detected.
If RETTO is performed, the counter starts counting down immediately from the value TO. This
enables generation of a periodic interrupt so that a user time-out can be handled, for example
when no key is pressed on a keyboard.
Figure 35-24 shows the block diagram of the Receiver Time-out feature.
Figure 35-24. Receiver Time-out Block Diagram
TO

Baud Rate
Clock

Clock

16-bit Time-out
Counter

16-bit
Value
=

STTTO

Character
Received

Load

Clear

TIMEOUT

RETTO

Table 35-10 gives the maximum time-out period for some standard baud rates.
Table 35-10. Maximum Time-out Period

714

Baud Rate

Bit Time

Time-out

bit/sec

μs

600

1 667

109 225

1 200

833

54 613

2 400

417

27 306

4 800

208

13 653

9 600

104

6 827

14400

4 551

19200

3 413

28800

2 276

33400

1 962

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 35-10. Maximum Time-out Period (Continued)

35.7.3.12

Baud Rate

Bit Time

Time-out

56000

1 170

57600

1 138

200000

328

Framing Error
The receiver is capable of detecting framing errors. A framing error happens when the stop bit of
a received character is detected at level 0. This can occur if the receiver and the transmitter are
fully desynchronized.
A framing error is reported on the FRAME bit of the Channel Status Register (US_CSR). The
FRAME bit is asserted in the middle of the stop bit as soon as the framing error is detected. It is
cleared by writing the Control Register (US_CR) with the RSTSTA bit to 1.
Figure 35-25. Framing Error Status
Baud Rate
Clock
RXD
Start
D0
Bit

Parity Stop
Bit Bit

RSTSTA = 1

Write
US_CR
FRAME

RXRDY

35.7.3.13

Transmit Break
The user can request the transmitter to generate a break condition on the TXD line. A break condition drives the TXD line low during at least one complete character. It appears the same as a
0x00 character sent with the parity and the stop bits to 0. However, the transmitter holds the
TXD line at least during one character until the user requests the break condition to be removed.
A break is transmitted by writing the Control Register (US_CR) with the STTBRK bit to 1. This
can be performed at any time, either while the transmitter is empty (no character in either the
Shift Register or in US_THR) or when a character is being transmitted. If a break is requested
while a character is being shifted out, the character is first completed before the TXD line is held
low.
Once STTBRK command is requested further STTBRK commands are ignored until the end of
the break is completed.
The break condition is removed by writing US_CR with the STPBRK bit to 1. If the STPBRK is
requested before the end of the minimum break duration (one character, including start, data,
parity and stop bits), the transmitter ensures that the break condition completes.

715
6430F–ATARM–21-Feb-12

The transmitter considers the break as though it is a character, i.e. the STTBRK and STPBRK
commands are taken into account only if the TXRDY bit in US_CSR is to 1 and the start of the
break condition clears the TXRDY and TXEMPTY bits as if a character is processed.
Writing US_CR with both STTBRK and STPBRK bits to 1 can lead to an unpredictable result. All
STPBRK commands requested without a previous STTBRK command are ignored. A byte written into the Transmit Holding Register while a break is pending, but not started, is ignored.
After the break condition, the transmitter returns the TXD line to 1 for a minimum of 12 bit times.
Thus, the transmitter ensures that the remote receiver detects correctly the end of break and the
start of the next character. If the timeguard is programmed with a value higher than 12, the TXD
line is held high for the timeguard period.
After holding the TXD line for this period, the transmitter resumes normal operations.
Figure 35-26 illustrates the effect of both the Start Break (STTBRK) and Stop Break (STPBRK)
commands on the TXD line.
Figure 35-26. Break Transmission
Baud Rate
Clock
TXD
Start
D0
Bit

STTBRK = 1

Parity Stop
Bit Bit

Break Transmission

End of Break

STPBRK = 1

Write
US_CR
TXRDY

TXEMPTY

35.7.3.14

Receive Break
The receiver detects a break condition when all data, parity and stop bits are low. This corresponds to detecting a framing error with data to 0x00, but FRAME remains low.
When the low stop bit is detected, the receiver asserts the RXBRK bit in US_CSR. This bit may
be cleared by writing the Control Register (US_CR) with the bit RSTSTA to 1.
An end of receive break is detected by a high level for at least 2/16 of a bit period in asynchronous operating mode or one sample at high level in synchronous operating mode. The end of
break detection also asserts the RXBRK bit.

35.7.3.15

716

Hardware Handshaking
The USART features a hardware handshaking out-of-band flow control. The RTS and CTS pins
are used to connect with the remote device, as shown in Figure 35-27.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 35-27. Connection with a Remote Device for Hardware Handshaking
USART

Remote
Device

TXD

RXD

TXD

CTS

RTS

CTS

Setting the USART to operate with hardware handshaking is performed by writing the
USART_MODE field in the Mode Register (US_MR) to the value 0x2.
The USART behavior when hardware handshaking is enabled is the same as the behavior in
standard synchronous or asynchronous mode, except that the receiver drives the RTS pin as
described below and the level on the CTS pin modifies the behavior of the transmitter as
described below. Using this mode requires using the PDC channel for reception. The transmitter
can handle hardware handshaking in any case.
Figure 35-28 shows how the receiver operates if hardware handshaking is enabled. The RTS
pin is driven high if the receiver is disabled and if the status RXBUFF (Receive Buffer Full) coming from the PDC channel is high. Normally, the remote device does not start transmitting while
its CTS pin (driven by RTS) is high. As soon as the Receiver is enabled, the RTS falls, indicating
to the remote device that it can start transmitting. Defining a new buffer to the PDC clears the
status bit RXBUFF and, as a result, asserts the pin RTS low.
Figure 35-28. Receiver Behavior when Operating with Hardware Handshaking
RXD
RXEN = 1

RXDIS = 1

Write
US_CR
RTS
RXBUFF

Figure 35-29 shows how the transmitter operates if hardware handshaking is enabled. The CTS
pin disables the transmitter. If a character is being processing, the transmitter is disabled only
after the completion of the current character and transmission of the next character happens as
soon as the pin CTS falls.
Figure 35-29. Transmitter Behavior when Operating with Hardware Handshaking
CTS

TXD

717
6430F–ATARM–21-Feb-12

35.7.4

ISO7816 Mode
The USART features an ISO7816-compatible operating mode. This mode permits interfacing
with smart cards and Security Access Modules (SAM) communicating through an ISO7816 link.
Both T = 0 and T = 1 protocols defined by the ISO7816 specification are supported.
Setting the USART in ISO7816 mode is performed by writing the USART_MODE field in the
Mode Register (US_MR) to the value 0x4 for protocol T = 0 and to the value 0x5 for protocol T =
1.

35.7.4.1

ISO7816 Mode Overview
The ISO7816 is a half duplex communication on only one bidirectional line. The baud rate is
determined by a division of the clock provided to the remote device (see “Baud Rate Generator”
on page 696).
The USART connects to a smart card as shown in Figure 35-30. The TXD line becomes bidirectional and the Baud Rate Generator feeds the ISO7816 clock on the SCK pin. As the TXD pin
becomes bidirectional, its output remains driven by the output of the transmitter but only when
the transmitter is active while its input is directed to the input of the receiver. The USART is considered as the master of the communication as it generates the clock.
Figure 35-30. Connection of a Smart Card to the USART
USART
SCK
TXD

CLK
I/O

Smart
Card

When operating in ISO7816, either in T = 0 or T = 1 modes, the character format is fixed. The
configuration is 8 data bits, even parity and 1 or 2 stop bits, regardless of the values programmed in the CHRL, MODE9, PAR and CHMODE fields. MSBF can be used to transmit LSB
or MSB first. Parity Bit (PAR) can be used to transmit in normal or inverse mode. Refer to
“USART Mode Register” on page 737 and “PAR: Parity Type” on page 738.
The USART cannot operate concurrently in both receiver and transmitter modes as the communication is unidirectional at a time. It has to be configured according to the required mode by
enabling or disabling either the receiver or the transmitter as desired. Enabling both the receiver
and the transmitter at the same time in ISO7816 mode may lead to unpredictable results.
The ISO7816 specification defines an inverse transmission format. Data bits of the character
must be transmitted on the I/O line at their negative value. The USART does not support this format and the user has to perform an exclusive OR on the data before writing it in the Transmit
Holding Register (US_THR) or after reading it in the Receive Holding Register (US_RHR).
35.7.4.2

Protocol T = 0
In T = 0 protocol, a character is made up of one start bit, eight data bits, one parity bit and one
guard time, which lasts two bit times. The transmitter shifts out the bits and does not drive the
I/O line during the guard time.
If no parity error is detected, the I/O line remains to 1 during the guard time and the transmitter
can continue with the transmission of the next character, as shown in Figure 35-31.

718

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
If a parity error is detected by the receiver, it drives the I/O line to 0 during the guard time, as
shown in Figure 35-32. This error bit is also named NACK, for Non Acknowledge. In this case,
the character lasts 1 bit time more, as the guard time length is the same and is added to the
error bit time which lasts 1 bit time.
When the USART is the receiver and it detects an error, it does not load the erroneous character
in the Receive Holding Register (US_RHR). It appropriately sets the PARE bit in the Status Register (US_SR) so that the software can handle the error.
Figure 35-31. T = 0 Protocol without Parity Error
Baud Rate
Clock
RXD
Start
Bit

Parity Guard Guard Next
Bit Time 1 Time 2 Start
Bit

Figure 35-32. T = 0 Protocol with Parity Error
Baud Rate
Clock
Error

I/O
Start
Bit

Parity Guard
Bit Time 1

Guard Start
Time 2 Bit

Repetition

Receive Error Counter
The USART receiver also records the total number of errors. This can be read in the Number of
Error (US_NER) register. The NB_ERRORS field can record up to 255 errors. Reading US_NER
automatically clears the NB_ERRORS field.
Receive NACK Inhibit
The USART can also be configured to inhibit an error. This can be achieved by setting the
INACK bit in the Mode Register (US_MR). If INACK is to 1, no error signal is driven on the I/O
line even if a parity bit is detected.
Moreover, if INACK is set, the erroneous received character is stored in the Receive Holding
Register, as if no error occurred and the RXRDY bit does rise.
Transmit Character Repetition
When the USART is transmitting a character and gets a NACK, it can automatically repeat the
character before moving on to the next one. Repetition is enabled by writing the
MAX_ITERATION field in the Mode Register (US_MR) at a value higher than 0. Each character
can be transmitted up to eight times; the first transmission plus seven repetitions.
If MAX_ITERATION does not equal zero, the USART repeats the character as many times as
the value loaded in MAX_ITERATION.

719
6430F–ATARM–21-Feb-12

When the USART repetition number reaches MAX_ITERATION, the ITERATION bit is set in the
Channel Status Register (US_CSR). If the repetition of the character is acknowledged by the
receiver, the repetitions are stopped and the iteration counter is cleared.
The ITERATION bit in US_CSR can be cleared by writing the Control Register with the RSIT bit
to 1.
Disable Successive Receive NACK
The receiver can limit the number of successive NACKs sent back to the remote transmitter.
This is programmed by setting the bit DSNACK in the Mode Register (US_MR). The maximum
number of NACK transmitted is programmed in the MAX_ITERATION field. As soon as
MAX_ITERATION is reached, the character is considered as correct, an acknowledge is sent on
the line and the ITERATION bit in the Channel Status Register is set.
35.7.4.3

35.7.5

Protocol T = 1
When operating in ISO7816 protocol T = 1, the transmission is similar to an asynchronous format with only one stop bit. The parity is generated when transmitting and checked when
receiving. Parity error detection sets the PARE bit in the Channel Status Register (US_CSR).
IrDA Mode
The USART features an IrDA mode supplying half-duplex point-to-point wireless communication. It embeds the modulator and demodulator which allows a glueless connection to the
infrared transceivers, as shown in Figure 35-33. The modulator and demodulator are compliant
with the IrDA specification version 1.1 and support data transfer speeds ranging from 2.4 Kb/s to
115.2 Kb/s.
The USART IrDA mode is enabled by setting the USART_MODE field in the Mode Register
(US_MR) to the value 0x8. The IrDA Filter Register (US_IF) allows configuring the demodulator
filter. The USART transmitter and receiver operate in a normal asynchronous mode and all
parameters are accessible. Note that the modulator and the demodulator are activated.
Figure 35-33. Connection to IrDA Transceivers

USART

IrDA
Transceivers

Receiver

Demodulator

RXD

Transmitter

Modulator

TXD

RX
TX

The receiver and the transmitter must be enabled or disabled according to the direction of the
transmission to be managed.
To receive IrDA signals, the following needs to be done:
• Disable TX and Enable RX

720

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• Configure the TXD pin as PIO and set it as an output to 0 (to avoid LED emission). Disable
the internal pull-up (better for power consumption).
• Receive data
35.7.5.1

IrDA Modulation
For baud rates up to and including 115.2 Kbits/sec, the RZI modulation scheme is used. “0” is
represented by a light pulse of 3/16th of a bit time. Some examples of signal pulse duration are
shown in Table 35-11.
Table 35-11. IrDA Pulse Duration
Baud Rate

Pulse Duration (3/16)

2.4 Kb/s

78.13 μs

9.6 Kb/s

19.53 μs

19.2 Kb/s

9.77 μs

38.4 Kb/s

4.88 μs

57.6 Kb/s

3.26 μs

115.2 Kb/s

1.63 μs

Figure 35-34 shows an example of character transmission.
Figure 35-34. IrDA Modulation
Start
Bit
Transmitter
Output

Stop
Bit

Data Bits
1

TXD

3
16 Bit Period

Bit Period

35.7.5.2

IrDA Baud Rate
Table 35-12 gives some examples of CD values, baud rate error and pulse duration. Note that
the requirement on the maximum acceptable error of ±1.87% must be met.
Table 35-12. IrDA Baud Rate Error
Peripheral Clock

Baud Rate

Baud Rate Error

Pulse Time

3 686 400

115 200

0.00%

1.63

20 000 000

115 200

1.38%

1.63

32 768 000

115 200

1.25%

1.63

40 000 000

115 200

1.38%

1.63

3 686 400

57 600

0.00%

3.26

20 000 000

57 600

1.38%

3.26

32 768 000

57 600

1.25%

3.26

721
6430F–ATARM–21-Feb-12

Table 35-12. IrDA Baud Rate Error (Continued)
Peripheral Clock

35.7.5.3

Baud Rate

Baud Rate Error

Pulse Time

40 000 000

57 600

0.93%

3.26

3 686 400

38 400

0.00%

4.88

20 000 000

38 400

1.38%

4.88

32 768 000

38 400

0.63%

4.88

40 000 000

38 400

0.16%

4.88

3 686 400

19 200

0.00%

9.77

20 000 000

19 200

0.16%

9.77

32 768 000

19 200

107

0.31%

9.77

40 000 000

19 200

130

0.16%

9.77

3 686 400

9 600

0.00%

19.53

20 000 000

9 600

130

0.16%

19.53

32 768 000

9 600

213

0.16%

19.53

40 000 000

9 600

260

0.16%

19.53

3 686 400

2 400

0.00%

78.13

20 000 000

2 400

521

0.03%

78.13

32 768 000

2 400

853

0.04%

78.13

IrDA Demodulator
The demodulator is based on the IrDA Receive filter comprised of an 8-bit down counter which is
loaded with the value programmed in US_IF. When a falling edge is detected on the RXD pin,
the Filter Counter starts counting down at the Master Clock (MCK) speed. If a rising edge is
detected on the RXD pin, the counter stops and is reloaded with US_IF. If no rising edge is
detected when the counter reaches 0, the input of the receiver is driven low during one bit time.
Figure 35-35 illustrates the operations of the IrDA demodulator.

Figure 35-35. IrDA Demodulator Operations
MCK

RXD

Counter
Value

Receiver
Input

4 3
Pulse
Rejected

Pulse
Accepted

As the IrDA mode uses the same logic as the ISO7816, note that the FI_DI_RATIO field in
US_FIDI must be set to a value higher than 0 in order to assure IrDA communications operate
correctly.

722

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.7.6

RS485 Mode
The USART features the RS485 mode to enable line driver control. While operating in RS485
mode, the USART behaves as though in asynchronous or synchronous mode and configuration
of all the parameters is possible. The difference is that the RTS pin is driven high when the
transmitter is operating. The behavior of the RTS pin is controlled by the TXEMPTY bit. A typical
connection of the USART to a RS485 bus is shown in Figure 35-36.
Figure 35-36. Typical Connection to a RS485 Bus
USART

RXD

Differential
Bus

TXD
RTS

The USART is set in RS485 mode by programming the USART_MODE field in the Mode Register (US_MR) to the value 0x1.
The RTS pin is at a level inverse to the TXEMPTY bit. Significantly, the RTS pin remains high
when a timeguard is programmed so that the line can remain driven after the last character completion. Figure 35-37 gives an example of the RTS waveform during a character transmission
when the timeguard is enabled.
Figure 35-37. Example of RTS Drive with Timeguard
TG = 4
Baud Rate
Clock
TXD
Start
D0
Bit

Parity Stop
Bit Bit

Write
US_THR
TXRDY

TXEMPTY

RTS

723
6430F–ATARM–21-Feb-12

35.7.7

Modem Mode
The USART features modem mode, which enables control of the signals: DTR (Data Terminal
Ready), DSR (Data Set Ready), RTS (Request to Send), CTS (Clear to Send), DCD (Data Carrier Detect) and RI (Ring Indicator). While operating in modem mode, the USART behaves as a
DTE (Data Terminal Equipment) as it drives DTR and RTS and can detect level change on DSR,
DCD, CTS and RI.
Setting the USART in modem mode is performed by writing the USART_MODE field in the Mode
Register (US_MR) to the value 0x3. While operating in modem mode the USART behaves as
though in asynchronous mode and all the parameter configurations are available.
Table 35-13 gives the correspondence of the USART signals with modem connection standards.
Table 35-13. Circuit References
USART Pin

V24

CCITT

Direction

TXD

103

From terminal to modem

RTS

105

From terminal to modem

DTR

108.2

From terminal to modem

RXD

104

From modem to terminal

CTS

106

From terminal to modem

DSR

107

From terminal to modem

DCD

109

From terminal to modem

125

From terminal to modem

The control of the DTR output pin is performed by writing the Control Register (US_CR) with the
DTRDIS and DTREN bits respectively to 1. The disable command forces the corresponding pin
to its inactive level, i.e. high. The enable command forces the corresponding pin to its active
level, i.e. low. RTS output pin is automatically controlled in this mode
The level changes are detected on the RI, DSR, DCD and CTS pins. If an input change is
detected, the RIIC, DSRIC, DCDIC and CTSIC bits in the Channel Status Register (US_CSR)
are set respectively and can trigger an interrupt. The status is automatically cleared when
US_CSR is read. Furthermore, the CTS automatically disables the transmitter when it is
detected at its inactive state. If a character is being transmitted when the CTS rises, the character transmission is completed before the transmitter is actually disabled.

724

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.7.8

SPI Mode
The Serial Peripheral Interface (SPI) Mode is a synchronous serial data link that provides communication with external devices in Master or Slave Mode. It also enables communication
between processors if an external processor is connected to the system.
The Serial Peripheral Interface is essentially a shift register that serially transmits data bits to
other SPIs. During a data transfer, one SPI system acts as the “master” which controls the data
flow, while the other devices act as “slaves'' which have data shifted into and out by the master.
Different CPUs can take turns being masters and one master may simultaneously shift data into
multiple slaves. (Multiple Master Protocol is the opposite of Single Master Protocol, where one
CPU is always the master while all of the others are always slaves.) However, only one slave
may drive its output to write data back to the master at any given time.
A slave device is selected when its NSS signal is asserted by the master. The USART in SPI
Master mode can address only one SPI Slave because it can generate only one NSS signal.
The SPI system consists of two data lines and two control lines:
• Master Out Slave In (MOSI): This data line supplies the output data from the master shifted
into the input of the slave.
• Master In Slave Out (MISO): This data line supplies the output data from a slave to the input
of the master.
• Serial Clock (SCK): This control line is driven by the master and regulates the flow of the data
bits. The master may transmit data at a variety of baud rates. The SCK line cycles once for
each bit that is transmitted.
• Slave Select (NSS): This control line allows the master to select or deselect the slave.

35.7.8.1

Modes of Operation
The USART can operate in SPI Master Mode or in SPI Slave Mode.
Operation in SPI Master Mode is programmed by writing to 0xE the USART_MODE field in the
Mode Register. In this case the SPI lines must be connected as described below:
• the MOSI line is driven by the output pin TXD
• the MISO line drives the input pin RXD
• the SCK line is driven by the output pin SCK
• the NSS line is driven by the output pin RTS
Operation in SPI Slave Mode is programmed by writing to 0xF the USART_MODE field in the
Mode Register. In this case the SPI lines must be connected as described below:
• the MOSI line drives the input pin RXD
• the MISO line is driven by the output pin TXD
• the SCK line drives the input pin SCK
• the NSS line drives the input pin CTS
In order to avoid unpredicted behavior, any change of the SPI Mode must be followed by a software reset of the transmitter and of the receiver (except the initial configuration after a hardware
reset). (See Section 35.7.2 ”Receiver and Transmitter Control”).

725
6430F–ATARM–21-Feb-12

35.7.8.2

Baud Rate
In SPI Mode, the baudrate generator operates in the same way as in USART synchronous
mode: See “Baud Rate in Synchronous Mode or SPI Mode” on page 698. However, there are
some restrictions:
In SPI Master Mode:
• the external clock SCK must not be selected (USCLKS ≠ 0x3), and the bit CLKO must be set
to “1” in the Mode Register (US_MR), in order to generate correctly the serial clock on the
SCK pin.
• to obtain correct behavior of the receiver and the transmitter, the value programmed in CD
must be superior or equal to 6.
• if the internal clock divided (MCK/DIV) is selected, the value programmed in CD must be
even to ensure a 50:50 mark/space ratio on the SCK pin, this value can be odd if the internal
clock is selected (MCK).
In SPI Slave Mode:
• the external clock (SCK) selection is forced regardless of the value of the USCLKS field in the
Mode Register (US_MR). Likewise, the value written in US_BRGR has no effect, because
the clock is provided directly by the signal on the USART SCK pin.
• to obtain correct behavior of the receiver and the transmitter, the external clock (SCK)
frequency must be at least 6 times lower than the system clock.

726

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.7.8.3

Data Transfer
Up to 9 data bits are successively shifted out on the TXD pin at each rising or falling edge
(depending of CPOL and CPHA) of the programmed serial clock. There is no Start bit, no Parity
bit and no Stop bit.
The number of data bits is selected by the CHRL field and the MODE 9 bit in the Mode Register
(US_MR). The 9 bits are selected by setting the MODE 9 bit regardless of the CHRL field. The
MSB data bit is always sent first in SPI Mode (Master or Slave).
Four combinations of polarity and phase are available for data transfers. The clock polarity is
programmed with the CPOL bit in the Mode Register. The clock phase is programmed with the
CPHA bit. These two parameters determine the edges of the clock signal upon which data is
driven and sampled. Each of the two parameters has two possible states, resulting in four possible combinations that are incompatible with one another. Thus, a master/slave pair must use the
same parameter pair values to communicate. If multiple slaves are used and fixed in different
configurations, the master must reconfigure itself each time it needs to communicate with a different slave.
Table 35-14. SPI Bus Protocol Mode
SPI Bus Protocol Mode

CPOL

CPHA

727
6430F–ATARM–21-Feb-12

Figure 35-38. SPI Transfer Format (CPHA=1, 8 bits per transfer)
SCK cycle (for reference)

SCK
(CPOL = 0)

SCK
(CPOL = 1)

MOSI
SPI Master ->TXD
SPI Slave -> RXD

MISO
SPI Master ->RXD
SPI Slave -> TXD

MSB

LSB

NSS
SPI Master -> RTS
SPI Slave -> CTS

Figure 35-39. SPI Transfer Format (CPHA=0, 8 bits per transfer)
SCK cycle (for reference)

SCK
(CPOL = 0)

SCK
(CPOL = 1)

MOSI
SPI Master -> TXD
SPI Slave -> RXD

MSB

LSB

MISO
SPI Master -> RXD
SPI Slave -> TXD

MSB

LSB

NSS
SPI Master -> RTS
SPI Slave -> CTS

728

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.7.8.4

Receiver and Transmitter Control
See “Receiver and Transmitter Control” on page 700.

35.7.8.5

Character Transmission
The characters are sent by writing in the Transmit Holding Register (US_THR). An additional
condition for transmitting a character can be added when the USART is configured in SPI master mode. In the USART_MR register, the value configured on INACK field can prevent any
character transmission (even if US_THR has been written) while the receiver side is not ready
(character not read). When INACK equals 0, the character is transmitted whatever the receiver
status. If INACK is set to 1, the transmitter waits for the receiver holding register to be read
before transmitting the character (RXRDY flag cleared), thus preventing any overflow (character
loss) on the receiver side.
The transmitter reports two status bits in the Channel Status Register (US_CSR): TXRDY
(Transmitter Ready), which indicates that US_THR is empty and TXEMPTY, which indicates
that all the characters written in US_THR have been processed. When the current character processing is completed, the last character written in US_THR is transferred into the Shift Register
of the transmitter and US_THR becomes empty, thus TXRDY rises.
Both TXRDY and TXEMPTY bits are low when the transmitter is disabled. Writing a character in
US_THR while TXRDY is low has no effect and the written character is lost.
If the USART is in SPI Slave Mode and if a character must be sent while the Transmit Holding
Register (US_THR) is empty, the UNRE (Underrun Error) bit is set. The TXD transmission line
stays at high level during all this time. The UNRE bit is cleared by writing the Control Register
(US_CR) with the RSTSTA (Reset Status) bit to 1.
In SPI Master Mode, the slave select line (NSS) is asserted at low level 1 Tbit (Time bit) before
the transmission of the MSB bit and released at high level 1 Tbit after the transmission of the
LSB bit. So, the slave select line (NSS) is always released between each character transmission
and a minimum delay of 3 Tbits always inserted. However, in order to address slave devices
supporting the CSAAT mode (Chip Select Active After Transfer), the slave select line (NSS) can
be forced at low level by writing the Control Register (US_CR) with the RTSEN bit to 1. The
slave select line (NSS) can be released at high level only by writing the Control Register
(US_CR) with the RTSDIS bit to 1 (for example, when all data have been transferred to the slave
device).
In SPI Slave Mode, the transmitter does not require a falling edge of the slave select line (NSS)
to initiate a character transmission but only a low level. However, this low level must be present
on the slave select line (NSS) at least 1 Tbit before the first serial clock cycle corresponding to
the MSB bit.

35.7.8.6

Character Reception
When a character reception is completed, it is transferred to the Receive Holding Register
(US_RHR) and the RXRDY bit in the Status Register (US_CSR) rises. If a character is completed while RXRDY is set, the OVRE (Overrun Error) bit is set. The last character is transferred
into US_RHR and overwrites the previous one. The OVRE bit is cleared by writing the Control
Register (US_CR) with the RSTSTA (Reset Status) bit to 1.
To ensure correct behavior of the receiver in SPI Slave Mode, the master device sending the
frame must ensure a minimum delay of 1 Tbit between each character transmission. The
receiver does not require a falling edge of the slave select line (NSS) to initiate a character

729
6430F–ATARM–21-Feb-12

reception but only a low level. However, this low level must be present on the slave select line
(NSS) at least 1 Tbit before the first serial clock cycle corresponding to the MSB bit.
35.7.8.7

35.7.9

Receiver Timeout
Because the receiver baudrate clock is active only during data transfers in SPI Mode, a receiver
timeout is impossible in this mode, whatever the Time-out value is (field TO) in the Time-out
Register (US_RTOR).
Test Modes

WRITE BUFFER

READ BUFFER

DATA 0

NACT = SUBSCRIBE

APB bus
|
|
|
|

APB bus
USART3
LIN CONTROLLER

(Peripheral) DMA
Controller
TXRDY

DATA N

|
|
|
|

USART3
LIN CONTROLLER

(Peripheral) DMA
Controller
RXRDY

DATA N

The USART can be programmed to operate in three different test modes. The internal loopback
capability allows on-board diagnostics. In the loopback mode the USART interface pins are disconnected or not and reconfigured for loopback internally or externally.
35.7.9.1

Normal Mode
Normal mode connects the RXD pin on the receiver input and the transmitter output on the TXD
pin.

Figure 35-40. Normal Mode Configuration
RXD
Receiver

TXD
Transmitter

35.7.9.2

Automatic Echo Mode
Automatic echo mode allows bit-by-bit retransmission. When a bit is received on the RXD pin, it
is sent to the TXD pin, as shown in Figure 35-41. Programming the transmitter has no effect on
the TXD pin. The RXD pin is still connected to the receiver input, thus the receiver remains
active.

Figure 35-41. Automatic Echo Mode Configuration
RXD
Receiver

TXD
Transmitter

730

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.7.9.3

Local Loopback Mode
Local loopback mode connects the output of the transmitter directly to the input of the receiver,
as shown in Figure 35-42. The TXD and RXD pins are not used. The RXD pin has no effect on
the receiver and the TXD pin is continuously driven high, as in idle state.

Figure 35-42. Local Loopback Mode Configuration
RXD
Receiver

Transmitter

35.7.9.4

TXD

Remote Loopback Mode
Remote loopback mode directly connects the RXD pin to the TXD pin, as shown in Figure 35-43.
The transmitter and the receiver are disabled and have no effect. This mode allows bit-by-bit
retransmission.

Figure 35-43. Remote Loopback Mode Configuration
Receiver

RXD

TXD
Transmitter

731
6430F–ATARM–21-Feb-12

35.7.10

Write Protection Registers
To prevent any single software error that may corrupt USART behavior, certain address spaces
can be write-protected by setting the WPEN bit in the USART Write Protect Mode Register
(US_WPMR).
If a write access to the protected registers is detected, then the WPVS flag in the USART Write
Protect Status Register (US_WPSR) is set and the field WPVSRC indicates in which register the
write access has been attempted.
The WPVS flag is reset by writing the USART Write Protect Mode Register (US_WPMR) with
the appropriate access key, WPKEY.
The protected registers are:
• “USART Mode Register”
• “USART Baud Rate Generator Register”
• “USART Receiver Time-out Register”
• “USART Transmitter Timeguard Register”
• “USART FI DI RATIO Register”
• “USART IrDA FILTER Register”
• “USART Manchester Configuration Register”

732

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8

Universal Synchronous Asynchronous Receiver Transmitter (USART) User Interface

Table 35-15.

Offset

Name

Access

Reset

0x0000

Control Register

US_CR

Write-only

–

0x0004

Mode Register

US_MR

Read-write

–

0x0008

Interrupt Enable Register

US_IER

Write-only

–

0x000C

Interrupt Disable Register

US_IDR

Write-only

–

0x0010

Interrupt Mask Register

US_IMR

Read-only

0x0

0x0014

Channel Status Register

US_CSR

Read-only

–

0x0018

Receiver Holding Register

US_RHR

Read-only

0x0

0x001C

Transmitter Holding Register

US_THR

Write-only

–

0x0020

Baud Rate Generator Register

US_BRGR

Read-write

0x0

0x0024

Receiver Time-out Register

US_RTOR

Read-write

0x0

0x0028

Transmitter Timeguard Register

US_TTGR

Read-write

0x0

–

0x2C - 0x3C

Reserved

0x0040

FI DI Ratio Register

US_FIDI

Read-write

0x174

0x0044

Number of Errors Register

US_NER

Read-only

–

0x0048

Reserved

–

0x004C

IrDA Filter Register

US_IF

Read-write

0x0

0x0050

Manchester Encoder Decoder Register

US_MAN

Read-write

0x30011004

0xE4

Write Protect Mode Register

US_WPMR

Read-write

0x0

0xE8

Write Protect Status Register

US_WPSR

Read-only

0x0

Reserved

–

Reserved for PDC Registers

–

0x5C - 0xFC
0x100 - 0x128

733
6430F–ATARM–21-Feb-12

35.8.1
Name:

USART Control Register
US_CR

Address:

0x40090000 (0), 0x40094000 (1), 0x40098000 (2), 0x4009C000 (3)

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
RTSDIS/RCS

18
RTSEN/FCS

17
DTRDIS

16
DTREN

15
RETTO

14
RSTNACK

13
RSTIT

12
SENDA

11
STTTO

10
STPBRK

9
STTBRK

8
RSTSTA

7
TXDIS

6
TXEN

5
RXDIS

4
RXEN

3
RSTTX

2
RSTRX

1
–

0
–

• RSTRX: Reset Receiver
0: No effect.
1: Resets the receiver.
• RSTTX: Reset Transmitter
0: No effect.
1: Resets the transmitter.
• RXEN: Receiver Enable
0: No effect.
1: Enables the receiver, if RXDIS is 0.
• RXDIS: Receiver Disable
0: No effect.
1: Disables the receiver.
• TXEN: Transmitter Enable
0: No effect.
1: Enables the transmitter if TXDIS is 0.
• TXDIS: Transmitter Disable
0: No effect.
1: Disables the transmitter.

734

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• RSTSTA: Reset Status Bits
0: No effect.
1: Resets the status bits PARE, FRAME, OVRE, MANERR, UNRE and RXBRK in US_CSR.
• STTBRK: Start Break
0: No effect.
1: Starts transmission of a break after the characters present in US_THR and the Transmit Shift Register have been transmitted. No effect if a break is already being transmitted.
• STPBRK: Stop Break
0: No effect.
1: Stops transmission of the break after a minimum of one character length and transmits a high level during 12-bit periods.
No effect if no break is being transmitted.
• STTTO: Start Time-out
0: No effect.
1: Starts waiting for a character before clocking the time-out counter. Resets the status bit TIMEOUT in US_CSR.
• SENDA: Send Address
0: No effect.
1: In Multidrop Mode only, the next character written to the US_THR is sent with the address bit set.
• RSTIT: Reset Iterations
0: No effect.
1: Resets ITERATION in US_CSR. No effect if the ISO7816 is not enabled.
• RSTNACK: Reset Non Acknowledge
0: No effect
1: Resets NACK in US_CSR.
• RETTO: Rearm Time-out
0: No effect
1: Restart Time-out
• DTREN: Data Terminal Ready Enable
0: No effect.
1: Drives the pin DTR to 0.
• DTRDIS: Data Terminal Ready Disable
0: No effect.
1: Drives the pin DTR to 1.

735
6430F–ATARM–21-Feb-12

• RTSEN: Request to Send Enable
0: No effect.
1: Drives the pin RTS to 0.
• FCS: Force SPI Chip Select
– Applicable if USART operates in SPI Master Mode (USART_MODE = 0xE):
FCS = 0: No effect.
FCS = 1: Forces the Slave Select Line NSS (RTS pin) to 0, even if USART is no transmitting, in order to address SPI slave
devices supporting the CSAAT Mode (Chip Select Active After Transfer).
• RTSDIS: Request to Send Disable
0: No effect.
1: Drives the pin RTS to 1.
• RCS: Release SPI Chip Select
– Applicable if USART operates in SPI Master Mode (USART_MODE = 0xE):
RCS = 0: No effect.
RCS = 1: Releases the Slave Select Line NSS (RTS pin).

736

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8.2
Name:

USART Mode Register
US_MR

Address:

0x40090004 (0), 0x40094004 (1), 0x40098004 (2), 0x4009C004 (3)

Access:

Read-write

31
ONEBIT

30
MODSYNC

29
MAN

28
FILTER

27
–

25
MAX_ITERATION

23
INVDATA

22
VAR_SYNC

21
DSNACK

20
INACK

19
OVER

18
CLKO

17
MODE9

16
MSBF/CPOL

10
PAR

8
SYNC/CPHA

CHMODE
7

NBSTOP
6

CHRL

USCLKS

USART_MODE

This register can only be written if the WPEN bit is cleared in “USART Write Protect Mode Register” on page 760.
• USART_MODE
Value

Name

Description

0x0

NORMAL

Normal mode

0x1

RS485

0x2

HW_HANDSHAKING

0x3

MODEM

0x4

IS07816_T_0

IS07816 Protocol: T = 0

0x6

IS07816_T_1

IS07816 Protocol: T = 1

0x8

IRDA

0xE

SPI_MASTER

SPI Master

0xF

SPI_SLAVE

SPI Slave

RS485
Hardware Handshaking
Modem

IrDA

• USCLKS: Clock Selection
Value

Name

Description

MCK

Master Clock MCK is selected

DIV

Internal Clock Divided MCK/DIV (DIV=8) is selected

SCK

Serial Clock SLK is selected

737
6430F–ATARM–21-Feb-12

• CHRL: Character Length.
Value

Name

Description

5_BIT

Character length is 5 bits

6_BIT

Character length is 6 bits

7_BIT

Character length is 7 bits

8_BIT

Character length is 8 bits

• SYNC: Synchronous Mode Select
0: USART operates in Asynchronous Mode.
1: USART operates in Synchronous Mode.
• CPHA: SPI Clock Phase
– Applicable if USART operates in SPI Mode (USART_MODE = 0xE or 0xF):
CPHA = 0: Data is changed on the leading edge of SPCK and captured on the following edge of SPCK.
CPHA = 1: Data is captured on the leading edge of SPCK and changed on the following edge of SPCK.
CPHA determines which edge of SPCK causes data to change and which edge causes data to be captured. CPHA is used
with CPOL to produce the required clock/data relationship between master and slave devices.
• PAR: Parity Type
Value

Name

Description

EVEN

Even parity

ODD

Odd parity

SPACE

Parity forced to 0 (Space)

MARK

Parity forced to 1 (Mark)

MULTIDROP

No parity
Multidrop mode

• NBSTOP: Number of Stop Bits

738

Value

Name

Description

1_BIT

1 stop bit

1_5_BIT

2_BIT

1.5 stop bit (SYNC = 0) or reserved (SYNC = 1)
2 stop bits

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• CHMODE: Channel Mode
Value

Name

Description

NORMAL

Normal Mode

AUTOMATIC

LOCAL_LOOPBACK

REMOTE_LOOPBACK

Automatic Echo. Receiver input is connected to the TXD pin.
Local Loopback. Transmitter output is connected to the Receiver Input.
Remote Loopback. RXD pin is internally connected to the TXD pin.

• MSBF: Bit Order
0: Least Significant Bit is sent/received first.
1: Most Significant Bit is sent/received first.
• CPOL: SPI Clock Polarity
– Applicable if USART operates in SPI Mode (Slave or Master, USART_MODE = 0xE or 0xF):
CPOL = 0: The inactive state value of SPCK is logic level zero.
CPOL = 1: The inactive state value of SPCK is logic level one.
CPOL is used to determine the inactive state value of the serial clock (SPCK). It is used with CPHA to produce the required
clock/data relationship between master and slave devices.
• MODE9: 9-bit Character Length
0: CHRL defines character length.
1: 9-bit character length.
• CLKO: Clock Output Select
0: The USART does not drive the SCK pin.
1: The USART drives the SCK pin if USCLKS does not select the external clock SCK.
• OVER: Oversampling Mode
0: 16x Oversampling.
1: 8x Oversampling.
• INACK: Inhibit Non Acknowledge
0: The NACK is generated.
1: The NACK is not generated.
Note: In SPI master mode, if INACK = 0 the character transmission starts as soon as a character is written into US_THR
register (assuming TXRDY was set). When INACK is 1, an additional condition must be met. The character transmission
starts when a character is written and only if RXRDY flag is cleared (Receiver Holding Register has been read).
• DSNACK: Disable Successive NACK
0: NACK is sent on the ISO line as soon as a parity error occurs in the received character (unless INACK is set).
1: Successive parity errors are counted up to the value specified in the MAX_ITERATION field. These parity errors generate a NACK on the ISO line. As soon as this value is reached, no additional NACK is sent on the ISO line. The flag
ITERATION is asserted.

739
6430F–ATARM–21-Feb-12

• INVDATA: INverted Data
0: The data field transmitted on TXD line is the same as the one written in US_THR register or the content read in US_RHR
is the same as RXD line. Normal mode of operation.
1: The data field transmitted on TXD line is inverted (voltage polarity only) compared to the value written on US_THR register or the content read in US_RHR is inverted compared to what is received on RXD line (or ISO7816 IO line). Inverted
Mode of operation, useful for contactless card application. To be used with configuration bit MSBF.
• VAR_SYNC: Variable Synchronization of Command/Data Sync Start Frame Delimiter
0: User defined configuration of command or data sync field depending on MODSYNC value.
1: The sync field is updated when a character is written into US_THR register.
• MAX_ITERATION
Defines the maximum number of iterations in mode ISO7816, protocol T= 0.
• FILTER: Infrared Receive Line Filter
0: The USART does not filter the receive line.
1: The USART filters the receive line using a three-sample filter (1/16-bit clock) (2 over 3 majority).
• MAN: Manchester Encoder/Decoder Enable
0: Manchester Encoder/Decoder are disabled.
1: Manchester Encoder/Decoder are enabled.
• MODSYNC: Manchester Synchronization Mode
0:The Manchester Start bit is a 0 to 1 transition
1: The Manchester Start bit is a 1 to 0 transition.
• ONEBIT: Start Frame Delimiter Selector
0: Start Frame delimiter is COMMAND or DATA SYNC.
1: Start Frame delimiter is One Bit.

740

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8.3
Name:

USART Interrupt Enable Register
US_IER

Address:

0x40090008 (0), 0x40094008 (1), 0x40098008 (2), 0x4009C008 (3)

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
MANE

23
–

22
–

21
–

20
–

19
CTSIC

18
DCDIC

17
DSRIC

16
RIIC

15
–

14
–

13
NACK

12
RXBUFF

11
TXBUFE

10
ITER/UNRE

9
TXEMPTY

8
TIMEOUT

7
PARE

6
FRAME

5
OVRE

4
ENDTX

3
ENDRX

2
RXBRK

1
TXRDY

0
RXRDY

0: No effect
1: Enables the corresponding interrupt.
• RXRDY: RXRDY Interrupt Enable
• TXRDY: TXRDY Interrupt Enable
• RXBRK: Receiver Break Interrupt Enable
• ENDRX: End of Receive Transfer Interrupt Enable
• ENDTX: End of Transmit Interrupt Enable
• OVRE: Overrun Error Interrupt Enable
• FRAME: Framing Error Interrupt Enable
• PARE: Parity Error Interrupt Enable
• TIMEOUT: Time-out Interrupt Enable
• TXEMPTY: TXEMPTY Interrupt Enable
• ITER: Max number of Repetitions Reached
• UNRE: SPI Underrun Error
• TXBUFE: Buffer Empty Interrupt Enable
• RXBUFF: Buffer Full Interrupt Enable
• NACK: Non AcknowledgeInterrupt Enable
• RIIC: Ring Indicator Input Change Enable

741
6430F–ATARM–21-Feb-12

• DSRIC: Data Set Ready Input Change Enable
• DCDIC: Data Carrier Detect Input Change Interrupt Enable
• CTSIC: Clear to Send Input Change Interrupt Enable
• MANE: Manchester Error Interrupt Enable

742

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8.4
Name:

USART Interrupt Disable Register
US_IDR

Address:

0x4009000C (0), 0x4009400C (1), 0x4009800C (2), 0x4009C00C (3)

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
MANE

23
–

22
–

21
–

20
–

19
CTSIC

18
DCDIC

17
DSRIC

16
RIIC

15
–

14
–

13
NACK

12
RXBUFF

11
TXBUFE

10
ITER/UNRE

9
TXEMPTY

8
TIMEOUT

7
PARE

6
FRAME

5
OVRE

4
ENDTX

3
ENDRX

2
RXBRK

1
TXRDY

0
RXRDY

0: No effect
1: Disables the corresponding interrupt.
• RXRDY: RXRDY Interrupt Disable
• TXRDY: TXRDY Interrupt Disable
• RXBRK: Receiver Break Interrupt Disable
• ENDRX: End of Receive Transfer Interrupt Disable
• ENDTX: End of Transmit Interrupt Disable
• OVRE: Overrun Error Interrupt Disable
• FRAME: Framing Error Interrupt Disable
• PARE: Parity Error Interrupt Disable
• TIMEOUT: Time-out Interrupt Disable
• TXEMPTY: TXEMPTY Interrupt Disable
• ITER: Max number of Repetitions Reached Disable
• UNRE: SPI Underrun Error Disable
• TXBUFE: Buffer Empty Interrupt Disable
• RXBUFF: Buffer Full Interrupt Disable
• NACK: Non AcknowledgeInterrupt Disable
•

RIIC: Ring Indicator Input Change Disable

743
6430F–ATARM–21-Feb-12

• DSRIC: Data Set Ready Input Change Disable
• DCDIC: Data Carrier Detect Input Change Interrupt Disable
• CTSIC: Clear to Send Input Change Interrupt Disable
• MANE: Manchester Error Interrupt Disable

744

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8.5
Name:

USART Interrupt Mask Register
US_IMR

Address:

0x40090010 (0), 0x40094010 (1), 0x40098010 (2), 0x4009C010 (3)

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
MANE

23
–

22
–

21
–

20
–

19
CTSIC

18
DCDIC

17
DSRIC

16
RIIC

15
–

14
–

13
NACK

12
RXBUFF

11
TXBUFE

10
ITER/UNRE

9
TXEMPTY

8
TIMEOUT

7
PARE

6
FRAME

5
OVRE

4
ENDTX

3
ENDRX

2
RXBRK

1
TXRDY

0
RXRDY

0: The corresponding interrupt is not enabled.
1: The corresponding interrupt is enabled.
• RXRDY: RXRDY Interrupt Mask
• TXRDY: TXRDY Interrupt Mask
• RXBRK: Receiver Break Interrupt Mask
• ENDRX: End of Receive Transfer Interrupt Mask
• ENDTX: End of Transmit Interrupt Mask
• OVRE: Overrun Error Interrupt Mask
• FRAME: Framing Error Interrupt Mask
• PARE: Parity Error Interrupt Mask
• TIMEOUT: Time-out Interrupt Mask
• TXEMPTY: TXEMPTY Interrupt Mask
• ITER: Max number of Repetitions Reached Mask
• UNRE: SPI Underrun Error Mask
• TXBUFE: Buffer Empty Interrupt Mask
• RXBUFF: Buffer Full Interrupt Mask
• NACK: Non AcknowledgeInterrupt Mask
• RIIC: Ring Indicator Input Change Mask

745
6430F–ATARM–21-Feb-12

• DSRIC: Data Set Ready Input Change Mask
• DCDIC: Data Carrier Detect Input Change Interrupt Mask
• CTSIC: Clear to Send Input Change Interrupt Mask
• MANE: Manchester Error Interrupt Mask

746

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8.6
Name:

USART Channel Status Register
US_CSR

Address:

0x40090014 (0), 0x40094014 (1), 0x40098014 (2), 0x4009C014 (3)

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
MANERR

23
CTS

22
DCD

21
DSR

20
RI

19
CTSIC

18
DCDIC

17
DSRIC

16
RIIC

15
–

14
–

13
NACK

12
RXBUFF

11
TXBUFE

10
ITER/UNRE

9
TXEMPTY

8
TIMEOUT

7
PARE

6
FRAME

5
OVRE

4
ENDTX

3
ENDRX

2
RXBRK

1
TXRDY

0
RXRDY

• RXRDY: Receiver Ready
0: No complete character has been received since the last read of US_RHR or the receiver is disabled. If characters were
being received when the receiver was disabled, RXRDY changes to 1 when the receiver is enabled.
1: At least one complete character has been received and US_RHR has not yet been read.
• TXRDY: Transmitter Ready
0: A character is in the US_THR waiting to be transferred to the Transmit Shift Register, or an STTBRK command has been
requested, or the transmitter is disabled. As soon as the transmitter is enabled, TXRDY becomes 1.
1: There is no character in the US_THR.
• RXBRK: Break Received/End of Break
0: No Break received or End of Break detected since the last RSTSTA.
1: Break Received or End of Break detected since the last RSTSTA.
• ENDRX: End of Receiver Transfer
0: The End of Transfer signal from the Receive PDC channel is inactive.
1: The End of Transfer signal from the Receive PDC channel is active.
• ENDTX: End of Transmitter Transfer
0: The End of Transfer signal from the Transmit PDC channel is inactive.
1: The End of Transfer signal from the Transmit PDC channel is active.
• OVRE: Overrun Error
0: No overrun error has occurred since the last RSTSTA.
1: At least one overrun error has occurred since the last RSTSTA.
• FRAME: Framing Error
0: No stop bit has been detected low since the last RSTSTA.
1: At least one stop bit has been detected low since the last RSTSTA.

747
6430F–ATARM–21-Feb-12

• PARE: Parity Error
0: No parity error has been detected since the last RSTSTA.
1: At least one parity error has been detected since the last RSTSTA.
• TIMEOUT: Receiver Time-out
0: There has not been a time-out since the last Start Time-out command (STTTO in US_CR) or the Time-out Register is 0.
1: There has been a time-out since the last Start Time-out command (STTTO in US_CR).
• TXEMPTY: Transmitter Empty
0: There are characters in either US_THR or the Transmit Shift Register, or the transmitter is disabled.
1: There are no characters in US_THR, nor in the Transmit Shift Register.
• ITER: Max number of Repetitions Reached
0: Maximum number of repetitions has not been reached since the last RSTSTA.
1: Maximum number of repetitions has been reached since the last RSTSTA.
• UNRE: SPI Underrun Error
– Applicable if USART operates in SPI Slave Mode (USART_MODE = 0xF):
UNRE = 0: No SPI underrun error has occurred since the last RSTSTA.
UNRE = 1: At least one SPI underrun error has occurred since the last RSTSTA.
• TXBUFE: Transmission Buffer Empty
0: The signal Buffer Empty from the Transmit PDC channel is inactive.
1: The signal Buffer Empty from the Transmit PDC channel is active.
• RXBUFF: Reception Buffer Full
0: The signal Buffer Full from the Receive PDC channel is inactive.
1: The signal Buffer Full from the Receive PDC channel is active.
• NACK: Non AcknowledgeInterrupt
0: Non Acknowledge has not been detected since the last RSTNACK.
1: At least one Non Acknowledge has been detected since the last RSTNACK.
• RIIC: Ring Indicator Input Change Flag
0: No input change has been detected on the RI pin since the last read of US_CSR.
1: At least one input change has been detected on the RI pin since the last read of US_CSR.
• DSRIC: Data Set Ready Input Change Flag
0: No input change has been detected on the DSR pin since the last read of US_CSR.
1: At least one input change has been detected on the DSR pin since the last read of US_CSR.
• DCDIC: Data Carrier Detect Input Change Flag
0: No input change has been detected on the DCD pin since the last read of US_CSR.
1: At least one input change has been detected on the DCD pin since the last read of US_CSR.

748

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• CTSIC: Clear to Send Input Change Flag
0: No input change has been detected on the CTS pin since the last read of US_CSR.
1: At least one input change has been detected on the CTS pin since the last read of US_CSR.
• RI: Image of RI Input
0: RI is set to 0.
1: RI is set to 1.
• DSR: Image of DSR Input
0: DSR is set to 0
1: DSR is set to 1.
• DCD: Image of DCD Input
0: DCD is set to 0.
1: DCD is set to 1.
• CTS: Image of CTS Input
0: CTS is set to 0.
1: CTS is set to 1.
• MANERR: Manchester Error
0: No Manchester error has been detected since the last RSTSTA.
1: At least one Manchester error has been detected since the last RSTSTA.

749
6430F–ATARM–21-Feb-12

35.8.7
Name:

USART Receive Holding Register
US_RHR

Address:

0x40090018 (0), 0x40094018 (1), 0x40098018 (2), 0x4009C018 (3)

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
RXSYNH

14
–

13
–

12
–

11
–

10
–

9
–

8
RXCHR

RXCHR

• RXCHR: Received Character
Last character received if RXRDY is set.
• RXSYNH: Received Sync
0: Last Character received is a Data.
1: Last Character received is a Command.

750

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8.8
Name:

USART Transmit Holding Register
US_THR

Address:

0x4009001C (0), 0x4009401C (1), 0x4009801C (2), 0x4009C01C (3)

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
TXSYNH

14
–

13
–

12
–

11
–

10
–

9
–

8
TXCHR

TXCHR

• TXCHR: Character to be Transmitted
Next character to be transmitted after the current character if TXRDY is not set.
• TXSYNH: Sync Field to be transmitted
0: The next character sent is encoded as a data. Start Frame Delimiter is DATA SYNC.
1: The next character sent is encoded as a command. Start Frame Delimiter is COMMAND SYNC.

751
6430F–ATARM–21-Feb-12

35.8.9
Name:

USART Baud Rate Generator Register
US_BRGR

Address:

0x40090020 (0), 0x40094020 (1), 0x40098020 (2), 0x4009C020 (3)

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

17
FP

CD
7

4
CD

This register can only be written if the WPEN bit is cleared in “USART Write Protect Mode Register” on page 760.
• CD: Clock Divider
USART_MODE ≠ ISO7816
SYNC = 1
or
USART_MODE = SPI
(Master or Slave)

SYNC = 0
CD

OVER = 0

0
1 to 65535

OVER = 1

USART_MODE =
ISO7816

Baud Rate Clock Disabled
Baud Rate =
Selected Clock/(16*CD)

Baud Rate =
Selected Clock/(8*CD)

Baud Rate =
Selected Clock /CD

Baud Rate = Selected
Clock/(FI_DI_RATIO*CD)

• FP: Fractional Part
0: Fractional divider is disabled.
1 - 7: Baudrate resolution, defined by FP x 1/8.

752

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8.10
Name:

USART Receiver Time-out Register
US_RTOR

Address:

0x40090024 (0), 0x40094024 (1), 0x40098024 (2), 0x4009C024 (3)

Access:

Read-write

–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

TO
7

4
TO

This register can only be written if the WPEN bit is cleared in “USART Write Protect Mode Register” on page 760.
• TO: Time-out Value
0: The Receiver Time-out is disabled.
1 - 65535: The Receiver Time-out is enabled and the Time-out delay is TO x Bit Period.

753
6430F–ATARM–21-Feb-12

35.8.11
Name:

USART Transmitter Timeguard Register
US_TTGR

Address:

0x40090028 (0), 0x40094028 (1), 0x40098028 (2), 0x4009C028 (3)

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

This register can only be written if the WPEN bit is cleared in “USART Write Protect Mode Register” on page 760.
• TG: Timeguard Value
0: The Transmitter Timeguard is disabled.
1 - 255: The Transmitter timeguard is enabled and the timeguard delay is TG x Bit Period.

754

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8.12
Name:

USART FI DI RATIO Register
US_FIDI

Address:

0x40090040 (0), 0x40094040 (1), 0x40098040 (2), 0x4009C040 (3)

Access:

Read-write

Reset Value:

0x174

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

9
FI_DI_RATIO

FI_DI_RATIO

This register can only be written if the WPEN bit is cleared in “USART Write Protect Mode Register” on page 760.
• FI_DI_RATIO: FI Over DI Ratio Value
0: If ISO7816 mode is selected, the Baud Rate Generator generates no signal.
1 - 2047: If ISO7816 mode is selected, the Baud Rate is the clock provided on SCK divided by FI_DI_RATIO.

755
6430F–ATARM–21-Feb-12

35.8.13
Name:

USART Number of Errors Register
US_NER

Address:

0x40090044 (0), 0x40094044 (1), 0x40098044 (2), 0x4009C044 (3)

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

NB_ERRORS

• NB_ERRORS: Number of Errors
Total number of errors that occurred during an ISO7816 transfer. This register automatically clears when read.

756

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8.14
Name:

USART IrDA FILTER Register
US_IF

Address:

0x4009004C (0), 0x4009404C (1), 0x4009804C (2), 0x4009C04C (3)

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

IRDA_FILTER

This register can only be written if the WPEN bit is cleared in “USART Write Protect Mode Register” on page 760.
• IRDA_FILTER: IrDA Filter
Sets the filter of the IrDA demodulator.

757
6430F–ATARM–21-Feb-12

35.8.15
Name:

USART Manchester Configuration Register
US_MAN

Address:

0x40090050 (0), 0x40094050 (1), 0x40098050 (2), 0x4009C050 (3)

Access:

Read-write

31
–

30
DRIFT

29
1

28
RX_MPOL

27
–

26
–

23
–

22
–

21
–

20
–

15
–

14
–

13
–

12
TX_MPOL

11
–

10
–

7
–

6
–

5
–

4
–

24
RX_PP

RX_PL
8
TX_PP
0

TX_PL

This register can only be written if the WPEN bit is cleared in “USART Write Protect Mode Register” on page 760.
• TX_PL: Transmitter Preamble Length
0: The Transmitter Preamble pattern generation is disabled
1 - 15: The Preamble Length is TX_PL x Bit Period
• TX_PP: Transmitter Preamble Pattern
The following values assume that TX_MPOL field is not set:
Value

Name

Description

ALL_ONE

The preamble is composed of ‘1’s

ALL_ZERO

The preamble is composed of ‘0’s

ZERO_ONE

The preamble is composed of ‘01’s

ONE_ZERO

The preamble is composed of ‘10’s

• TX_MPOL: Transmitter Manchester Polarity
0: Logic Zero is coded as a zero-to-one transition, Logic One is coded as a one-to-zero transition.
1: Logic Zero is coded as a one-to-zero transition, Logic One is coded as a zero-to-one transition.
• RX_PL: Receiver Preamble Length
0: The receiver preamble pattern detection is disabled
1 - 15: The detected preamble length is RX_PL x Bit Period

758

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• RX_PP: Receiver Preamble Pattern detected
The following values assume that RX_MPOL field is not set:
Value

Name

Description

ALL_ONE

The preamble is composed of ‘1’s

ALL_ZERO

The preamble is composed of ‘0’s

ZERO_ONE

The preamble is composed of ‘01’s

ONE_ZERO

The preamble is composed of ‘10’s

• RX_MPOL: Receiver Manchester Polarity
0: Logic Zero is coded as a zero-to-one transition, Logic One is coded as a one-to-zero transition.
1: Logic Zero is coded as a one-to-zero transition, Logic One is coded as a zero-to-one transition.
• DRIFT: Drift compensation
0: The USART can not recover from an important clock drift
1: The USART can recover from clock drift. The 16X clock mode must be enabled.

759
6430F–ATARM–21-Feb-12

35.8.16
Name:

USART Write Protect Mode Register
US_WPMR

Address:

0x400900E4 (0), 0x400940E4 (1), 0x400980E4 (2), 0x4009C0E4 (3)

Access:

Read-write

Reset:

See Table 35-15

WPKEY
23

20
WPKEY

12
WPKEY

—

WPEN

• WPEN: Write Protect Enable
0 = Disables the Write Protect if WPKEY corresponds to 0x555341 (“USA” in ASCII).
1 = Enables the Write Protect if WPKEY corresponds to 0x555341 (“USA” in ASCII).
Protects the registers:
• “USART Mode Register” on page 737
• “USART Baud Rate Generator Register” on page 752
• “USART Receiver Time-out Register” on page 753
• “USART Transmitter Timeguard Register” on page 754
• “USART FI DI RATIO Register” on page 755
• “USART IrDA FILTER Register” on page 757
• “USART Manchester Configuration Register” on page 758
• WPKEY: Write Protect KEY
Should be written at value 0x555341 (“USA” in ASCII). Writing any other value in this field aborts the write operation of the
WPEN bit. Always reads as 0.

760

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
35.8.17
Name:

USART Write Protect Status Register
US_WPSR

Address:

0x400900E8 (0), 0x400940E8 (1), 0x400980E8 (2), 0x4009C0E8 (3)

Access:

Read-only

Reset:

See Table 35-15

—

WPVSRC
15

12
WPVSRC

—

WPVS

• WPVS: Write Protect Violation Status
0 = No Write Protect Violation has occurred since the last read of the US_WPSR register.
1 = A Write Protect Violation has occurred since the last read of the US_WPSR register. If this violation is an unauthorized
attempt to write a protected register, the associated violation is reported into field WPVSRC.
• WPVSRC: Write Protect Violation Source
When WPVS is active, this field indicates the write-protected register (through address offset or code) in which a write
access has been attempted.
Note:

Reading US_WPSR automatically clears all fields.

761
6430F–ATARM–21-Feb-12

762

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
36. Timer Counter (TC)
36.1

Description
The Timer Counter (TC) includes 9 identical 16-bit Timer Counter channels.
Each channel can be independently programmed to perform a wide range of functions including
frequency measurement, event counting, interval measurement, pulse generation, delay timing
and pulse width modulation.
Each channel has three external clock inputs, five internal clock inputs and two multi-purpose
input/output signals which can be configured by the user. Each channel drives an internal interrupt signal which can be programmed to generate processor interrupts.
The Timer Counter (TC) embeds a quadrature decoder logic connected in front of the timers and
driven by TIOA0, TIOB0 and TIOA1 inputs. When enabled, the quadrature decoder performs the
input lines filtering, decoding of quadrature signals and connects to the timers/counters in order
to read the position and speed of the motor through the user interface.
The Timer Counter block has two global registers which act upon all TC channels.
The Block Control Register allows the channels to be started simultaneously with the same
instruction.
The Block Mode Register defines the external clock inputs for each channel, allowing them to be
chained.
Table 36-1 gives the assignment of the device Timer Counter clock inputs common to Timer
Counter 0 to 2.
Table 36-1.

Timer Counter Clock Assignment

Name

Definition

TIMER_CLOCK1

MCK/2

TIMER_CLOCK2

MCK/8

TIMER_CLOCK3

MCK/32

TIMER_CLOCK4

MCK/128

TIMER_CLOCK5
Note:

36.2

(1)

SLCK

1. When Slow Clock is selected for Master Clock (CSS = 0 in PMC Master Clock Register),
TIMER_CLOCK5 input is equivalent to Master Clock.

Embedded Characteristics
• Provides 9 16-bit Timer Counter channels
• Wide range of functions including:
– Frequency measurement
– Event counting
– Interval measurement
– Pulse generation
– Delay timing
– Pulse Width Modulation
– Up/down capabilities

763
6430F–ATARM–21-Feb-12

– Quadrature decoder logic
• Each channel is user-configurable and contains:
– Three external clock inputs
– Five Internal clock inputs
– Two multi-purpose input/output signals
• Internal interrupt signal
• Two global registers that act on all TC channels

36.3

Block Diagram

Figure 36-1. Timer Counter Block Diagram
Parallel I/O
Controller

TIMER_CLOCK1

TCLK0
TIMER_CLOCK2

TIOA1
TIOA2

TIMER_CLOCK3

TCLK1
TIMER_CLOCK4

XC0
XC1

Timer/Counter
Channel 0

TIOA

TIOA0
TIOB0

TIOA0
TIOB

TCLK2

XC2

TIMER_CLOCK5

TC0XC0S

TIOB0
SYNC

TCLK0
TCLK1
TCLK2

INT0

TCLK0
TCLK1

XC0
TIOA0

XC1

TIOA2

XC2

Timer/Counter
Channel 1

TIOA

TIOA1
TIOB1

TIOA1
TIOB

TCLK2

TC1XC1S

TCLK0

XC0

TCLK1

XC1

TCLK2

XC2

TIOB1
SYNC

Timer/Counter
Channel 2

INT1

TIOA

TIOA2
TIOB2

TIOA2
TIOB

TIOA0
TIOA1

TC2XC2S

TIOB2
SYNC

INT2

Timer Counter
Interrupt
Controller

764

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series

Table 36-2.

Signal Name Description

Block/Channel

Signal Name
XC0, XC1, XC2

Channel Signal

Capture Mode: Timer Counter Input
Waveform Mode: Timer Counter Output

TIOB

Capture Mode: Timer Counter Input
Waveform Mode: Timer Counter Input/Output

SYNC

36.5.1

Interrupt Signal Output
Synchronization Input Signal

Pin Name List
Table 36-3.

36.5

External Clock Inputs

TIOA

INT

36.4

Description

TC pin list

Pin Name

Description

Type

TCLK0-TCLK2

External Clock Input

Input

TIOA0-TIOA2

I/O Line A

I/O

TIOB0-TIOB2

I/O Line B

I/O

36.5.2

I/O Lines

Instance

Signal

I/O Line

Peripheral

TC0

TCLK0

PA2

TC0

TCLK1

PB4

TC0

TCLK2

PA26

TC0

TIOA0

PA1

TC0

TIOA1

PB5

TC0

TIOA2

PA30

TC0

TIOB0

PA0

TC0

TIOB1

PB6

TC0

TIOB2

PA31

Power Management
The TC is clocked through the Power Management Controller (PMC), thus the programmer must
first configure the PMC to enable the Timer Counter clock.

765
6430F–ATARM–21-Feb-12

36.5.3

Interrupt
The TC has an interrupt line connected to the Interrupt Controller (IC). Handling the TC interrupt
requires programming the IC before configuring the TC.

36.6

Functional Description

36.6.1

TC Description
The 9 channels of the Timer Counter are independent and identical in operation except when
quadrature decoder is enabled. The registers for channel programming are listed in Table 36-5
on page 785.

36.6.2

16-bit Counter
Each channel is organized around a 16-bit counter. The value of the counter is incremented at
each positive edge of the selected clock. When the counter has reached the value 0xFFFF and
passes to 0x0000, an overflow occurs and the COVFS bit in TC_SR (Status Register) is set.
The current value of the counter is accessible in real time by reading the Counter Value Register, TC_CV. The counter can be reset by a trigger. In this case, the counter value passes to
0x0000 on the next valid edge of the selected clock.

36.6.3

Clock Selection
At block level, input clock signals of each channel can either be connected to the external inputs
TCLK0, TCLK1 or TCLK2, or be connected to the internal I/O signals TIOA0, TIOA1 or TIOA2
for chaining by programming the TC_BMR (Block Mode). See Figure 36-2 ”Clock Chaining
Selection”.
Each channel can independently select an internal or external clock source for its counter:
•

Internal clock signals: TIMER_CLOCK1, TIMER_CLOCK2, TIMER_CLOCK3,
TIMER_CLOCK4, TIMER_CLOCK5

•

External clock signals: XC0, XC1 or XC2

This selection is made by the TCCLKS bits in the TC Channel Mode Register.
The selected clock can be inverted with the CLKI bit in TC_CMR. This allows counting on the
opposite edges of the clock.
The burst function allows the clock to be validated when an external signal is high. The BURST
parameter in the Mode Register defines this signal (none, XC0, XC1, XC2). See Figure 36-3
”Clock Selection”
Note:

766

In all cases, if an external clock is used, the duration of each of its levels must be longer than the
master clock period. The external clock frequency must be at least 2.5 times lower than the master clock

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 36-2. Clock Chaining Selection
TC0XC0S
Timer/Counter
Channel 0

TCLK0
TIOA1

XC0

TIOA2

TIOA0

XC1 = TCLK1
XC2 = TCLK2

TIOB0

SYNC

TC1XC1S
Timer/Counter
Channel 1
TCLK1

XC0 = TCLK0
TIOA0

TIOA1

XC1

TIOA2

XC2 = TCLK2

TIOB1

SYNC

Timer/Counter
Channel 2

TC2XC2S

XC0 = TCLK0
TCLK2

TIOA2

XC1 = TCLK1
TIOA0

XC2

TIOB2

TIOA1
SYNC

Figure 36-3. Clock Selection
TCCLKS
CLKI

TIMER_CLOCK1

Synchronous
Edge Detection

TIMER_CLOCK2
TIMER_CLOCK3

Selected
Clock

TIMER_CLOCK4
TIMER_CLOCK5
XC0
XC1
XC2
MCK
BURST

767
6430F–ATARM–21-Feb-12

36.6.4

Clock Control
The clock of each counter can be controlled in two different ways: it can be enabled/disabled
and started/stopped. See Figure 36-4.
•

The clock can be enabled or disabled by the user with the CLKEN and the CLKDIS
commands in the Control Register. In Capture Mode it can be disabled by an RB load event
if LDBDIS is set to 1 in TC_CMR. In Waveform Mode, it can be disabled by an RC Compare
event if CPCDIS is set to 1 in TC_CMR. When disabled, the start or the stop actions have no
effect: only a CLKEN command in the Control Register can re-enable the clock. When the
clock is enabled, the CLKSTA bit is set in the Status Register.

•

The clock can also be started or stopped: a trigger (software, synchro, external or compare)
always starts the clock. The clock can be stopped by an RB load event in Capture Mode
(LDBSTOP = 1 in TC_CMR) or a RC compare event in Waveform Mode (CPCSTOP = 1 in
TC_CMR). The start and the stop commands have effect only if the clock is enabled.

Figure 36-4. Clock Control
Selected
Clock

Trigger

CLKSTA

CLKEN

Q
Q

CLKDIS

S
R

Counter
Clock

36.6.5

Stop
Event

Disable
Event

TC Operating Modes
Each channel can independently operate in two different modes:
•

Capture Mode provides measurement on signals.

•

Waveform Mode provides wave generation.

The TC Operating Mode is programmed with the WAVE bit in the TC Channel Mode Register.
In Capture Mode, TIOA and TIOB are configured as inputs.
In Waveform Mode, TIOA is always configured to be an output and TIOB is an output if it is not
selected to be the external trigger.
36.6.6

Trigger
A trigger resets the counter and starts the counter clock. Three types of triggers are common to
both modes, and a fourth external trigger is available to each mode.

768

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Regardless of the trigger used, it will be taken into account at the following active edge of the
selected clock. This means that the counter value can be read differently from zero just after a
trigger, especially when a low frequency signal is selected as the clock.
The following triggers are common to both modes:
•

Software Trigger: Each channel has a software trigger, available by setting SWTRG in
TC_CCR.

•

SYNC: Each channel has a synchronization signal SYNC. When asserted, this signal has
the same effect as a software trigger. The SYNC signals of all channels are asserted
simultaneously by writing TC_BCR (Block Control) with SYNC set.

•

Compare RC Trigger: RC is implemented in each channel and can provide a trigger when
the counter value matches the RC value if CPCTRG is set in TC_CMR.

The channel can also be configured to have an external trigger. In Capture Mode, the external
trigger signal can be selected between TIOA and TIOB. In Waveform Mode, an external event
can be programmed on one of the following signals: TIOB, XC0, XC1 or XC2. This external
event can then be programmed to perform a trigger by setting ENETRG in TC_CMR.
If an external trigger is used, the duration of the pulses must be longer than the master clock
period in order to be detected.
36.6.7

Capture Operating Mode
This mode is entered by clearing the WAVE parameter in TC_CMR (Channel Mode Register).
Capture Mode allows the TC channel to perform measurements such as pulse timing, frequency, period, duty cycle and phase on TIOA and TIOB signals which are considered as
inputs.
Figure 36-5 shows the configuration of the TC channel when programmed in Capture Mode.

36.6.8

Capture Registers A and B
Registers A and B (RA and RB) are used as capture registers. This means that they can be
loaded with the counter value when a programmable event occurs on the signal TIOA.
The LDRA parameter in TC_CMR defines the TIOA selected edge for the loading of register A,
and the LDRB parameter defines the TIOA selected edge for the loading of Register B.
RA is loaded only if it has not been loaded since the last trigger or if RB has been loaded since
the last loading of RA.
RB is loaded only if RA has been loaded since the last trigger or the last loading of RB.
Loading RA or RB before the read of the last value loaded sets the Overrun Error Flag (LOVRS)
in TC_SR (Status Register). In this case, the old value is overwritten.

36.6.9

Trigger Conditions
In addition to the SYNC signal, the software trigger and the RC compare trigger, an external trigger can be defined.
The ABETRG bit in TC_CMR selects TIOA or TIOB input signal as an external trigger. The
ETRGEDG parameter defines the edge (rising, falling or both) detected to generate an external
trigger. If ETRGEDG = 0 (none), the external trigger is disabled.

769
6430F–ATARM–21-Feb-12

770

MTIOA

MTIOB

ABETRG

CLKI

If RA is not loaded
or RB is Loaded

Edge
Detector

ETRGEDG

SWTRG

Timer/Counter Channel

BURST

MCK

Synchronous
Edge Detection

OVF

LDRB
Edge
Detector

Edge
Detector

Capture
Register A

LDBSTOP

CLKEN

LDRA

If RA is Loaded

CPCTRG

Counter

RESET
Trig

CLK

CLKSTA

LDBDIS

Capture
Register B

CLKDIS

TC1_SR

TIOA

TIOB

SYNC

XC2

XC1

XC0

TIMER_CLOCK5

TIMER_CLOCK4

TIMER_CLOCK3

TIMER_CLOCK2

TIMER_CLOCK1

TCCLKS

Compare RC =

COVFS
INT

Figure 36-5. Capture Mode

LOVRS

CPCS

LDRBS

LDRAS

ETRGS
TC1_IMR

SAM3U Series

6430F–ATARM–21-Feb-12

SAM3U Series
36.6.10

Waveform Operating Mode
Waveform operating mode is entered by setting the WAVE parameter in TC_CMR (Channel
Mode Register).
In Waveform Operating Mode the TC channel generates 1 or 2 PWM signals with the same frequency and independently programmable duty cycles, or generates different types of one-shot
or repetitive pulses.
In this mode, TIOA is configured as an output and TIOB is defined as an output if it is not used
as an external event (EEVT parameter in TC_CMR).
Figure 36-6 shows the configuration of the TC channel when programmed in Waveform Operating Mode.

36.6.11

Waveform Selection
Depending on the WAVSEL parameter in TC_CMR (Channel Mode Register), the behavior of
TC_CV varies.
With any selection, RA, RB and RC can all be used as compare registers.
RA Compare is used to control the TIOA output, RB Compare is used to control the TIOB output
(if correctly configured) and RC Compare is used to control TIOA and/or TIOB outputs.

771
6430F–ATARM–21-Feb-12

772

TIOB

SYNC

XC2

XC1

XC0

TIMER_CLOCK5

TIMER_CLOCK4

TIMER_CLOCK3

TIMER_CLOCK2

TIMER_CLOCK1

EEVT

BURST

ENETRG

CLKI

Timer/Counter Channel

Edge
Detector

EEVTEDG

SWTRG

MCK

Synchronous
Edge Detection

Trig

CLK

OVF

WAVSEL

RESET

Counter

WAVSEL

Compare RA =

CLKSTA

Compare RC =

Compare RB =

CPCSTOP

CPCDIS

CLKDIS

CLKEN

CPAS
INT

BSWTRG

BEEVT

BCPB

BCPC

ASWTRG

AEEVT

ACPA

ACPC

Output Controller
Output Controller

TCCLKS

TIOB

MTIOB

TIOA

MTIOA

Figure 36-6. Waveform Mode

CPCS
CPBS

COVFS

TC1_SR

ETRGS
TC1_IMR

SAM3U Series

6430F–ATARM–21-Feb-12

SAM3U Series
36.6.11.1

WAVSEL = 00
When WAVSEL = 00, the value of TC_CV is incremented from 0 to 0xFFFF. Once 0xFFFF has
been reached, the value of TC_CV is reset. Incrementation of TC_CV starts again and the cycle
continues. See Figure 36-7.
An external event trigger or a software trigger can reset the value of TC_CV. It is important to
note that the trigger may occur at any time. See Figure 36-8.
RC Compare cannot be programmed to generate a trigger in this configuration. At the same
time, RC Compare can stop the counter clock (CPCSTOP = 1 in TC_CMR) and/or disable the
counter clock (CPCDIS = 1 in TC_CMR).
Figure 36-7. WAVSEL= 00 without trigger
Counter Value

Counter cleared by compare match with 0xFFFF

0xFFFF

RC
RB

Waveform Examples

Time

TIOB

TIOA

773
6430F–ATARM–21-Feb-12

Figure 36-8. WAVSEL= 00 with trigger
Counter cleared by compare match with 0xFFFF

Counter Value
0xFFFF

Counter cleared by trigger

RC
RB

Time

Waveform Examples
TIOB

TIOA

36.6.11.2

WAVSEL = 10
When WAVSEL = 10, the value of TC_CV is incremented from 0 to the value of RC, then automatically reset on a RC Compare. Once the value of TC_CV has been reset, it is then
incremented and so on. See Figure 36-9.
It is important to note that TC_CV can be reset at any time by an external event or a software
trigger if both are programmed correctly. See Figure 36-10.
In addition, RC Compare can stop the counter clock (CPCSTOP = 1 in TC_CMR) and/or disable
the counter clock (CPCDIS = 1 in TC_CMR).
Figure 36-9. WAVSEL = 10 Without Trigger
Counter Value
0xFFFF
Counter cleared by compare match with RC
RC
RB

Waveform Examples

Time

TIOB

TIOA

774

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 36-10. WAVSEL = 10 With Trigger
Counter Value
0xFFFF
Counter cleared by compare match with RC

Counter cleared by trigger

RC
RB

Waveform Examples

Time

TIOB

TIOA

36.6.11.3

WAVSEL = 01
When WAVSEL = 01, the value of TC_CV is incremented from 0 to 0xFFFF. Once 0xFFFF is
reached, the value of TC_CV is decremented to 0, then re-incremented to 0xFFFF and so on.
See Figure 36-11.
A trigger such as an external event or a software trigger can modify TC_CV at any time. If a trigger occurs while TC_CV is incrementing, TC_CV then decrements. If a trigger is received while
TC_CV is decrementing, TC_CV then increments. See Figure 36-12.
RC Compare cannot be programmed to generate a trigger in this configuration.
At the same time, RC Compare can stop the counter clock (CPCSTOP = 1) and/or disable the
counter clock (CPCDIS = 1).

775
6430F–ATARM–21-Feb-12

Figure 36-11. WAVSEL = 01 Without Trigger
Counter decremented by compare match with 0xFFFF

Counter Value
0xFFFF

RC
RB

Time

Waveform Examples
TIOB

TIOA

Figure 36-12. WAVSEL = 01 With Trigger
Counter decremented by compare match with 0xFFFF

Counter Value
0xFFFF

Counter decremented
by trigger
RC
RB

Counter incremented
by trigger

Time

Waveform Examples
TIOB

TIOA

36.6.11.4

WAVSEL = 11
When WAVSEL = 11, the value of TC_CV is incremented from 0 to RC. Once RC is reached, the
value of TC_CV is decremented to 0, then re-incremented to RC and so on. See Figure 36-13.
A trigger such as an external event or a software trigger can modify TC_CV at any time. If a trigger occurs while TC_CV is incrementing, TC_CV then decrements. If a trigger is received while
TC_CV is decrementing, TC_CV then increments. See Figure 36-14.
RC Compare can stop the counter clock (CPCSTOP = 1) and/or disable the counter clock
(CPCDIS = 1).

776

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 36-13. WAVSEL = 11 Without Trigger
Counter Value
0xFFFF
Counter decremented by compare match with RC
RC
RB

Time

Waveform Examples
TIOB

TIOA

Figure 36-14. WAVSEL = 11 With Trigger
Counter Value
0xFFFF
Counter decremented by compare match with RC
RC
RB

Counter decremented
by trigger
Counter incremented
by trigger

Waveform Examples

Time

TIOB

TIOA

777
6430F–ATARM–21-Feb-12

36.6.12

External Event/Trigger Conditions
An external event can be programmed to be detected on one of the clock sources (XC0, XC1,
XC2) or TIOB. The external event selected can then be used as a trigger.
The EEVT parameter in TC_CMR selects the external trigger. The EEVTEDG parameter defines
the trigger edge for each of the possible external triggers (rising, falling or both). If EEVTEDG is
cleared (none), no external event is defined.
If TIOB is defined as an external event signal (EEVT = 0), TIOB is no longer used as an output
and the compare register B is not used to generate waveforms and subsequently no IRQs. In
this case the TC channel can only generate a waveform on TIOA.
When an external event is defined, it can be used as a trigger by setting bit ENETRG in
TC_CMR.
As in Capture Mode, the SYNC signal and the software trigger are also available as triggers. RC
Compare can also be used as a trigger depending on the parameter WAVSEL.

36.6.13

Output Controller
The output controller defines the output level changes on TIOA and TIOB following an event.
TIOB control is used only if TIOB is defined as output (not as an external event).
The following events control TIOA and TIOB: software trigger, external event and RC compare.
RA compare controls TIOA and RB compare controls TIOB. Each of these events can be programmed to set, clear or toggle the output as defined in the corresponding parameter in
TC_CMR.

36.6.14
36.6.14.1

Quadrature Decoder Logic
Description
The quadrature decoder logic is driven by TIOA0, TIOB0, TIOA1 input pins and drives the
timer/counter of channel 0 and 1. Channel 2 can be used as a time base in case of speed measurement requirements (refer to Figure 36.7 ”Timer Counter (TC) User Interface”).
When writing 0 in the QDEN field of the TC_BMR register, the quadrature decoder logic is totally
transparent.
TIOA0 and TIOB0 are to be driven by the 2 dedicated quadrature signals from a rotary sensor
mounted on the shaft of the off-chip motor.
A third signal from the rotary sensor can be processed through pin TIOA1 and is typically dedicated to be driven by an index signal if it is provided by the sensor. This signal is not required to
decode the quadrature signals PHA, PHB.
TCCLKS field of TC_CMR channels must be configured to select XC0 input (i.e. 0x101).
TC0XC0S field has no effect as soon as quadrature decoder is enabled.
Either speed or position/revolution can be measured. Position channel 0 accumulates the edges
of PHA, PHB input signals giving a high accuracy on motor position whereas channel 1 accumulates the index pulses of the sensor, therefore the number of rotations. Concatenation of both
values provides a high level of precision on motion system position.
In speed mode, position cannot be measured but revolution can be measured.
Inputs from the rotary sensor can be filtered prior to down-stream processing. Accommodation
of input polarity, phase definition and other factors are configurable.

778

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Interruptions can be generated on different events.
A compare function (using TC_RC register) is available on channel 0 (speed/position) or channel 1 (rotation) and can generate an interrupt by means of the CPCS flag in the TC_SR
registers.
Figure 36-15. Predefined Connection of the Quadrature Decoder with Timer Counters

Reset pulse
SPEEDEN

Quadrature
Decoder

1
1

(Filter + Edge
Detect + QD)

TIOA0
QDEN

PHEdges

TIOA0

TIOB0

TIOB1

1
TIOB0

TIOB 1

XC0

PHA

Speed/Position

QDEN

PHB

IDX

Timer/Counter
Channel 0

TIOA

Index

1
TIOB1

TIOB 1

XC0

Timer/Counter
Channel 1

XC0

Rotation
Direction
Timer/Counter
Channel 2

Speed Time Base

36.6.14.2

Input Pre-processing
Input pre-processing consists of capabilities to take into account rotary sensor factors such as
polarities and phase definition followed by configurable digital filtering.
Each input can be negated and swapping PHA, PHB is also configurable.

779
6430F–ATARM–21-Feb-12

By means of the MAXFILT field in TC_BMR, it is possible to configure a minimum duration for
which the pulse is stated as valid. When the filter is active, pulses with a duration lower than
MAXFILT+1 * tMCK ns are not passed to down-stream logic.
Filters can be disabled using the FILTER field in the TC_BMR register.
Figure 36-16. Input Stage

Input Pre-Processing

MAXFILT

SWAP
1

PHA

Filter

FILTER

PHedge

TIOA0
Direction
and
Edge
Detection

INVA
1

PHB

Filter

DIR

Filter

IDX

TIOB0
INVB
1
1

IDX

TIOB1
IDXPHB
INVIDX

Input filtering can efficiently remove spurious pulses that might be generated by the presence of
particulate contamination on the optical or magnetic disk of the rotary sensor.
Spurious pulses can also occur in environments with high levels of electro-magnetic interference. Or, simply if vibration occurs even when rotation is fully stopped and the shaft of the motor
is in such a position that the beginning of one of the reflective or magnetic bars on the rotary
sensor disk is aligned with the light or magnetic (Hall) receiver cell of the rotary sensor. Any
vibration can make the PHA, PHB signals toggle for a short duration.

780

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 36-17. Filtering Examples

MAXFILT=2

MCK

particulate contamination
PHA,B

Filter Out

Optical/Magnetic disk strips
PHA

PHB
motor shaft stopped in such a position that
rotary sensor cell is aligned with an edge of the disk
rotation
stop
PHA
PHB Edge area due to system vibration
PHB

Resulting PHA, PHB electrical waveforms
PHA

stop
mechanical shock on system

PHB

vibration
PHA, PHB electrical waveforms after filtering
PHA
PHB

781
6430F–ATARM–21-Feb-12

36.6.14.3

Direction Status and Change Detection
After filtering, the quadrature signals are analyzed to extract the rotation direction and edges of
the 2 quadrature signals detected in order to be counted by timer/counter logic downstream.
The direction status can be directly read at anytime on TC_QISR register. The polarity of the
direction flag status depends on the configuration written in TC_BMR register. INVA, INVB,
INVIDX, SWAP modify the polarity of DIR flag.
Any change in rotation direction is reported on TC_QISR register and can generate an interrupt.
The direction change condition is reported as soon as 2 consecutive edges on a phase signal
have sampled the same value on the other phase signal and there is an edge on the other signal. The 2 consecutive edges of 1 phase signal sampling the same value on other phase signal
is not sufficient to declare a direction change, for the reason that particulate contamination may
mask one or more reflective bar on the optical or magnetic disk of the sensor. (Refer to Figure
36-18 ”Rotation Change Detection” for waveforms.)

Figure 36-18. Rotation Change Detection

Direction Change under normal conditions

PHA

change condition
Report Time

PHB

DIR

DIRCHG

No direction change due to particulate contamination masking a reflective bar
missing pulse
PHA
same phase
PHB
DIR
spurious change condition (if detected in a simple way)
DIRCHG
The direction change detection is disabled when QDTRANS is set to 1 in TC_BMR. In this case
the DIR flag report must not be used.
782

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
A quadrature error is also reported by the quadrature decoder logic. Rather than reporting an
error only when 2 edges occur at the same time on PHA and PHB, which is unlikely to occur in
real life, there is a report if the time difference between 2 edges on PHA, PHB is lower than a
predefined value. This predefined value is configurable and corresponds to (MAXFILT+1) *
tMCK ns. After being filtered there is no reason to have 2 edges closer than (MAXFILT+1) *
tMCK ns under normal mode of operation. In the instance an anomaly occurs, a quadrature error
is reported on QERR flag on TC_QISR register.
Figure 36-19. Quadrature Error Detection

MAXFILT = 2

MCK

Abnormally formatted optical disk strips (theoretical view)
PHA

PHB
strip edge inaccurary due to disk etching/printing process
PHA

PHB
resulting PHA, PHB electrical waveforms
PHA
Even with an abnorrmaly formatted disk, there is no occurence of PHA, PHB switching at the same time.
PHB
duration < MAXFILT
QERR
MAXFILT must be tuned according to several factors such as the system clock frequency
(MCK), type of rotary sensor and rotation speed to be achieved.
36.6.14.4

Position and Rotation Measurement
When POSEN is set in TC_BMR register, position is processed on channel 0 (by means of the
PHA,PHB edge detections) and motor revolutions are accumulated in channel 1 timer/counter
and can be read through TC_CV0 and/or TC_CV1 register if the IDX signal is provided on
TIOA1 input.
Channel 0 and 1 must be configured in capture mode (WAVE = 0 in TC_CMR0).

783
6430F–ATARM–21-Feb-12

In parallel, the number of edges are accumulated on timer/counter channel 0 and can be read on
the TC_CV0 register.
Therefore, the accurate position can be read on both TC_CV registers and concatenated to form
a 32-bit word.
The timer/counter channel 0 is cleared for each increment of IDX count value.
Depending on the quadrature signals, the direction is decoded and allows to count up or down in
timer/counter channels 0 and 1. The direction status is reported on TC_QISR register.
36.6.14.5

Speed Measurement
When SPEEDEN is set in TC_BMR register, the speed measure is enabled on channel 0.
A time base must be defined on channel 2 by writing the TC_RC2 period register. Channel 2
must be configured in waveform mode (WAVE bit field set) in TC_CMR2 register. WAVSEL bit
field must be defined with 0x10 to clear the counter by comparison and matching with TC_RC
value. ACPC field must be defined at 0x11 to toggle TIOA output.
This time base is automatically fed back to TIOA of channel 0 when QDEN and SPEEDEN are
set.
Channel 0 must be configured in capture mode (WAVE = 0 in TC_CMR0). ABETRG bit field of
TC_CMR0 must be configured at 1 to get TIOA as a trigger for this channel.
EDGTRG can be set to 0x01, to clear the counter on a rising edge of the TIOA signal and LDRA
field must be set accordingly to 0x01, to load TC_RA0 at the same time as the counter is cleared
(LDRB must be set to 0x01). As a consequence, at the end of each time base period the differentiation required for the speed calculation is performed.
The process must be started by configuring the TC_CR register with CLKEN and SWTRG.
The speed can be read on TC_RA0 register in TC_CMR0.
Channel 1 can still be used to count the number of revolutions of the motor.

784

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
36.7

Timer Counter (TC) User Interface

Table 36-5.

Name

Access

Reset

0x00 + channel * 0x40 + 0x00

Channel Control Register

TC_CCR

Write-only

–

0x00 + channel * 0x40 + 0x04

Channel Mode Register

TC_CMR

Read-write

0x00 + channel * 0x40 + 0x08

Reserved

0x00 + channel * 0x40 + 0x0C

Reserved

0x00 + channel * 0x40 + 0x10

Counter Value

TC_CV

Read-only

0x00 + channel * 0x40 + 0x14

TC_RA

Read-write(2)

0x00 + channel * 0x40 + 0x18

TC_RB

Read-write(2)

0x00 + channel * 0x40 + 0x1C

TC_RC

Read-write

0x00 + channel * 0x40 + 0x20

Status Register

TC_SR

Read-only

0x00 + channel * 0x40 + 0x24

Interrupt Enable Register

TC_IER

Write-only

–

0x00 + channel * 0x40 + 0x28

Interrupt Disable Register

TC_IDR

Write-only

–

0x00 + channel * 0x40 + 0x2C

Interrupt Mask Register

TC_IMR

Read-only

0xC0

Block Control Register

TC_BCR

Write-only

–

0xC4

Block Mode Register

TC_BMR

Read-write

0xC8

QDEC Interrupt Enable Register

TC_QIER

Write-only

–

0xCC

QDEC Interrupt Disable Register

TC_QIDR

Write-only

–

0xD0

QDEC Interrupt Mask Register

TC_QIMR

Read-only

0xD4

QDEC Interrupt Status Register

TC_QISR

Read-only

0xD8

Reserved

0xE4

Reserved

0xFC

Reserved

–

Notes:

1. Channel index ranges from 0 to 2.
2. Read-only if WAVE = 0

785
6430F–ATARM–21-Feb-12

36.7.1
Name:

TC Block Control Register
TC_BCR

Address:

0x400800C0

Access:

Write-only

–

SYNC

• SYNC: Synchro Command
0 = no effect.
1 = asserts the SYNC signal which generates a software trigger simultaneously for each of the channels.

786

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
36.7.2
Name:

TC Block Mode Register
TC_BMR

Address:

0x400800C4

Access:

Read-write

–

FILTER

–

IDXPHB

SWAP

MAXFILT

24
MAXFILT

INVIDX

INVB

INVA

EDGPHA

QDTRANS

SPEEDEN

POSEN

QDEN

–

TC2XC2S

TC1XC1S

0
TC0XC0S

• TC0XC0S: External Clock Signal 0 Selection
Value

Name

Description

TCLK0

Signal connected to XC0: TCLK0

–

Reserved

TIOA1

Signal connected to XC0: TIOA1

TIOA2

Signal connected to XC0: TIOA2

• TC1XC1S: External Clock Signal 1 Selection
Value

Name

Description

TCLK1

Signal connected to XC1: TCLK1

–

Reserved

TIOA0

Signal connected to XC1: TIOA0

TIOA2

Signal connected to XC1: TIOA2

• TC2XC2S: External Clock Signal 2 Selection
Value

Name

Description

TCLK2

Signal connected to XC2: TCLK2

–

Reserved

TIOA1

Signal connected to XC2: TIOA1

TIOA2

Signal connected to XC2: TIOA2

• QDEN: Quadrature Decoder ENabled
0 = disabled.
1 = enables the quadrature decoder logic (filter, edge detection and quadrature decoding).
quadrature decoding (direction change) can be disabled using QDTRANS bit.
One of the POSEN or SPEEDEN bits must be also enabled.

787
6430F–ATARM–21-Feb-12

• POSEN: POSition ENabled
0 = disable position.
1 = enables the position measure on channel 0 and 1
• SPEEDEN: SPEED ENabled
0 = disabled.
1 = enables the speed measure on channel 0, the time base being provided by channel 2.
• QDTRANS: Quadrature Decoding TRANSparent
0 = full quadrature decoding logic is active (direction change detected).
1 = quadrature decoding logic is inactive (direction change inactive) but input filtering and edge detection are performed.
• EDGPHA: EDGe on PHA count mode
0 = edges are detected on both PHA and PHB.
1 = edges are detected on PHA only.
• INVA: INVerted phA
0 = PHA (TIOA0) is directly driving quadrature decoder logic.
1 = PHA is inverted before driving quadrature decoder logic.
• INVB: INVerted phB
0 = PHB (TIOB0) is directly driving quadrature decoder logic.
1 = PHB is inverted before driving quadrature decoder logic.
• SWAP: SWAP PHA and PHB
0 = no swap between PHA and PHB.
1 = swap PHA and PHB internally, prior to driving quadrature decoder logic.
• INVIDX: INVerted InDeX
0 = IDX (TIOA1) is directly driving quadrature logic.
1 = IDX is inverted before driving quadrature logic.
• IDXPHB: InDeX pin is PHB pin
0 = IDX pin of the rotary sensor must drive TIOA1.
1 = IDX pin of the rotary sensor must drive TIOB0.
• FILTER:
0 = IDX,PHA, PHB input pins are not filtered.
1 = IDX,PHA, PHB input pins are filtered using MAXFILT value.
• MAXFILT: MAXimum FILTer
1.. 63: defines the filtering capabilities
Pulses with a period shorter than MAXFILT+1 MCK clock cycles are discarded.

788

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
36.7.3
Name:

TC Channel Control Register
TC_CCRx [x=0..2]

Address:

0x40080000 (0)[0], 0x40080040 (0)[1], 0x40080080 (0)[2]

Access:

Write-only

–

SWTRG

CLKDIS

CLKEN

• CLKEN: Counter Clock Enable Command
0 = no effect.
1 = enables the clock if CLKDIS is not 1.
• CLKDIS: Counter Clock Disable Command
0 = no effect.
1 = disables the clock.
• SWTRG: Software Trigger Command
0 = no effect.
1 = a software trigger is performed: the counter is reset and the clock is started.

789
6430F–ATARM–21-Feb-12

36.7.4
Name:

TC QDEC Interrupt Enable Register
TC_QIER

Address:

0x400800C8

Access:

Write-only

–

QERR

DIRCHG

IDX

• IDX: InDeX
0 = no effect.
1 = enables the interrupt when a rising edge occurs on IDX input.
• DIRCHG: DIRection CHanGe
0 = no effect.
1 = enables the interrupt when a change on rotation direction is detected.
• QERR: Quadrature ERRor
0 = no effect.
1 = enables the interrupt when a quadrature error occurs on PHA,PHB.

790

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
36.7.5
Name:

TC QDEC Interrupt Disable Register
TC_QIDR

Address:

0x400800CC

Access:

Write-only

–

QERR

DIRCHG

IDX

• IDX: InDeX
0 = no effect.
1 = disables the interrupt when a rising edge occurs on IDX input.
• DIRCHG: DIRection CHanGe
0 = no effect.
1 = disables the interrupt when a change on rotation direction is detected.
• QERR: Quadrature ERRor
0 = no effect.
1 = disables the interrupt when a quadrature error occurs on PHA, PHB.

791
6430F–ATARM–21-Feb-12

36.7.6
Name:

TC QDEC Interrupt Mask Register
TC_QIMR

Address:

0x400800D0

Access:

Read-only

–

QERR

DIRCHG

IDX

• IDX: InDeX
0 = the interrupt on IDX input is disabled.
1 = the interrupt on IDX input is enabled.
• DIRCHG: DIRection CHanGe
0 = the interrupt on rotation direction change is disabled.
1 = the interrupt on rotation direction change is enabled.
• QERR: Quadrature ERRor
0 = the interrupt on quadrature error is disabled.
1 = the interrupt on quadrature error is enabled.

792

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
36.7.7
Name:

TC QDEC Interrupt Status Register
TC_QISR

Address:

0x400800D4

Access:

Read-only

–

DIR

–

QERR

DIRCHG

IDX

• IDX: InDeX
0 = no Index input change since the last read of TC_QISR.
1 = the IDX input has change since the last read of TC_QISR.
• DIRCHG: DIRection CHanGe
0 = no change on rotation direction since the last read of TC_QISR.
1 = the rotation direction changed since the last read of TC_QISR.
• QERR: Quadrature ERRor
0 = no quadrature error since the last read of TC_QISR.
1 = A quadrature error occurred since the last read of TC_QISR.
• DIR: Direction
Returns an image of the actual rotation direction.

793
6430F–ATARM–21-Feb-12

36.7.8
Name:

TC Channel Mode Register: Capture Mode
TC_CMRx [x=0..2] (WAVE = 0)

Address:

0x40080004 (0)[0], 0x40080044 (0)[1], 0x40080084 (0)[2]

Access:

Read-write

–

WAVE

CPCTRG

–

ABETRG

LDBDIS

LDBSTOP

4
BURST

LDRB

CLKI

LDRA
9

8
ETRGEDG

TCCLKS

• TCCLKS: Clock Selection
Value

Name

Description

TIMER_CLOCK1

Clock selected: TCLK1

TIMER_CLOCK2

Clock selected: TCLK2

TIMER_CLOCK3

Clock selected: TCLK3

TIMER_CLOCK4

Clock selected: TCLK4

TIMER_CLOCK5

Clock selected: TCLK5

XC0

Clock selected: XC0

XC1

Clock selected: XC1

XC2

Clock selected: XC2

• CLKI: Clock Invert
0 = counter is incremented on rising edge of the clock.
1 = counter is incremented on falling edge of the clock.
• BURST: Burst Signal Selection
Value

Name

Description

NONE

The clock is not gated by an external signal.

XC0

XC0 is ANDed with the selected clock.

XC1

XC1 is ANDed with the selected clock.

XC2

XC2 is ANDed with the selected clock.

• LDBSTOP: Counter Clock Stopped with RB Loading
0 = counter clock is not stopped when RB loading occurs.
1 = counter clock is stopped when RB loading occurs.

794

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• LDBDIS: Counter Clock Disable with RB Loading
0 = counter clock is not disabled when RB loading occurs.
1 = counter clock is disabled when RB loading occurs.
• ETRGEDG: External Trigger Edge Selection
Value

Name

Description

NONE

The clock is not gated by an external signal.

RISING

Rising edge

FALLING

Falling edge

EDGE

Each edge

• ABETRG: TIOA or TIOB External Trigger Selection
0 = TIOB is used as an external trigger.
1 = TIOA is used as an external trigger.
• CPCTRG: RC Compare Trigger Enable
0 = RC Compare has no effect on the counter and its clock.
1 = RC Compare resets the counter and starts the counter clock.
• WAVE: Waveform Mode
0 = Capture Mode is enabled.
1 = Capture Mode is disabled (Waveform Mode is enabled).
• LDRA: RA Loading Edge Selection
Value

Name

Description

NONE

None

RISING

Rising edge of TIOA

FALLING

Falling edge of TIOA

EDGE

Each edge of TIOA

• LDRB: RB Loading Edge Selection
Value

Name

Description

NONE

None

RISING

Rising edge of TIOA

FALLING

Falling edge of TIOA

EDGE

Each edge of TIOA

795
6430F–ATARM–21-Feb-12

36.7.9
Name:

TC Channel Mode Register: Waveform Mode
TC_CMRx [x=0..2] (WAVE = 1)

Address:

0x40080004 (0)[0], 0x40080044 (0)[1], 0x40080084 (0)[2]

Access:

Read-write

BSWTRG
23

ASWTRG
15

BEEVT
20

AEEVT
14

WAVE

CPCDIS

CPCSTOP

24
BCPB

ENETRG
5

ACPC
12

WAVSEL

26
BCPC

4
BURST

ACPA
10

EEVT
3
CLKI

8
EEVTEDG

TCCLKS

• TCCLKS: Clock Selection
Value

Name

Description

TIMER_CLOCK1

Clock selected: TCLK1

TIMER_CLOCK2

Clock selected: TCLK2

TIMER_CLOCK3

Clock selected: TCLK3

TIMER_CLOCK4

Clock selected: TCLK4

TIMER_CLOCK5

Clock selected: TCLK5

XC0

Clock selected: XC0

XC1

Clock selected: XC1

XC2

Clock selected: XC2

• CLKI: Clock Invert
0 = counter is incremented on rising edge of the clock.
1 = counter is incremented on falling edge of the clock.
• BURST: Burst Signal Selection
Value

Name

Description

NONE

The clock is not gated by an external signal.

XC0

XC0 is ANDed with the selected clock.

XC1

XC1 is ANDed with the selected clock.

XC2

XC2 is ANDed with the selected clock.

• CPCSTOP: Counter Clock Stopped with RC Compare
0 = counter clock is not stopped when counter reaches RC.
1 = counter clock is stopped when counter reaches RC.

796

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• CPCDIS: Counter Clock Disable with RC Compare
0 = counter clock is not disabled when counter reaches RC.
1 = counter clock is disabled when counter reaches RC.
• EEVTEDG: External Event Edge Selection
Value

Name

Description

NONE

None

RISING

Rising edge

FALLING

Falling edge

EDGE

Each edge

• EEVT: External Event Selection
Signal selected as external event.
Value

Name

Description

TIOB Direction

TIOB

TIOB(1)

input

XC0

output

XC1

output

XC2

output

Note:

1. If TIOB is chosen as the external event signal, it is configured as an input and no longer generates waveforms and subsequently no IRQs.

• ENETRG: External Event Trigger Enable
0 = the external event has no effect on the counter and its clock. In this case, the selected external event only controls the
TIOA output.
1 = the external event resets the counter and starts the counter clock.
• WAVSEL: Waveform Selection
Value

Name

Description

UP mode without automatic trigger on RC Compare

UPDOWN

UPDOWN mode without automatic trigger on RC Compare

UP_RC

UP mode with automatic trigger on RC Compare

UPDOWN_RC

UPDOWN mode with automatic trigger on RC Compare

• WAVE: Waveform Mode
0 = Waveform Mode is disabled (Capture Mode is enabled).
1 = Waveform Mode is enabled.

797
6430F–ATARM–21-Feb-12

• ACPA: RA Compare Effect on TIOA
Value

Name

Description

NONE

None

SET

Set

CLEAR

Clear

TOGGLE

Toggle

• ACPC: RC Compare Effect on TIOA
Value

Name

Description

NONE

None

SET

Set

CLEAR

Clear

TOGGLE

Toggle

• AEEVT: External Event Effect on TIOA
Value

Name

Description

NONE

None

SET

Set

CLEAR

Clear

TOGGLE

Toggle

• ASWTRG: Software Trigger Effect on TIOA
Value

Name

Description

NONE

None

SET

Set

CLEAR

Clear

TOGGLE

Toggle

• BCPB: RB Compare Effect on TIOB

798

Value

Name

Description

NONE

None

SET

Set

CLEAR

Clear

TOGGLE

Toggle

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• BCPC: RC Compare Effect on TIOB
Value

Name

Description

NONE

None

SET

Set

CLEAR

Clear

TOGGLE

Toggle

• BEEVT: External Event Effect on TIOB
Value

Name

Description

NONE

None

SET

Set

CLEAR

Clear

TOGGLE

Toggle

• BSWTRG: Software Trigger Effect on TIOB
Value

Name

Description

NONE

None

SET

Set

CLEAR

Clear

TOGGLE

Toggle

799
6430F–ATARM–21-Feb-12

36.7.10
Name:

TC Counter Value Register
TC_CVx [x=0..2]

Address:

0x40080010 (0)[0], 0x40080050 (0)[1], 0x40080090 (0)[2]

Access:

Read-only

CV
23

20
CV

12
CV

4
CV

• CV: Counter Value
CV contains the counter value in real time.

800

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
36.7.11
Name:

TC Register A
TC_RAx [x=0..2]

Address:

0x40080014 (0)[0], 0x40080054 (0)[1], 0x40080094 (0)[2]

Access:

Read-only if WAVE = 0, Read-write if WAVE = 1

RA
23

20
RA

12
RA

4
RA

• RA: Register A
RA contains the Register A value in real time.

801
6430F–ATARM–21-Feb-12

36.7.12
Name:

TC Register B
TC_RBx [x=0..2]

Address:

0x40080018 (0)[0], 0x40080058 (0)[1], 0x40080098 (0)[2]

Access:

Read-only if WAVE = 0, Read-write if WAVE = 1

RB
23

20
RB

12
RB

4
RB

• RB: Register B
RB contains the Register B value in real time.

802

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
36.7.13
Name:

TC Register C
TC_RCx [x=0..2]

Address:

0x4008001C (0)[0], 0x4008005C (0)[1], 0x4008009C (0)[2]

Access:

Read-write

RC
23

20
RC

12
RC

4
RC

• RC: Register C
RC contains the Register C value in real time.

803
6430F–ATARM–21-Feb-12

36.7.14
Name:

TC Status Register
TC_SRx [x=0..2]

Address:

0x40080020 (0)[0], 0x40080060 (0)[1], 0x400800A0 (0)[2]

Access:

Read-only

–

MTIOB

MTIOA

CLKSTA

–

ETRGS

LDRBS

LDRAS

CPCS

CPBS

CPAS

LOVRS

COVFS

• COVFS: Counter Overflow Status
0 = no counter overflow has occurred since the last read of the Status Register.
1 = a counter overflow has occurred since the last read of the Status Register.
• LOVRS: Load Overrun Status
0 = Load overrun has not occurred since the last read of the Status Register or WAVE = 1.
1 = RA or RB have been loaded at least twice without any read of the corresponding register since the last read of the Status Register, if WAVE = 0.
• CPAS: RA Compare Status
0 = RA Compare has not occurred since the last read of the Status Register or WAVE = 0.
1 = RA Compare has occurred since the last read of the Status Register, if WAVE = 1.
• CPBS: RB Compare Status
0 = RB Compare has not occurred since the last read of the Status Register or WAVE = 0.
1 = RB Compare has occurred since the last read of the Status Register, if WAVE = 1.
• CPCS: RC Compare Status
0 = RC Compare has not occurred since the last read of the Status Register.
1 = RC Compare has occurred since the last read of the Status Register.
• LDRAS: RA Loading Status
0 = RA Load has not occurred since the last read of the Status Register or WAVE = 1.
1 = RA Load has occurred since the last read of the Status Register, if WAVE = 0.
• LDRBS: RB Loading Status
0 = RB Load has not occurred since the last read of the Status Register or WAVE = 1.
1 = RB Load has occurred since the last read of the Status Register, if WAVE = 0.

804

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• ETRGS: External Trigger Status
0 = external trigger has not occurred since the last read of the Status Register.
1 = external trigger has occurred since the last read of the Status Register.
• CLKSTA: Clock Enabling Status
0 = clock is disabled.
1 = clock is enabled.
• MTIOA: TIOA Mirror
0 = TIOA is low. If WAVE = 0, this means that TIOA pin is low. If WAVE = 1, this means that TIOA is driven low.
1 = TIOA is high. If WAVE = 0, this means that TIOA pin is high. If WAVE = 1, this means that TIOA is driven high.
• MTIOB: TIOB Mirror
0 = TIOB is low. If WAVE = 0, this means that TIOB pin is low. If WAVE = 1, this means that TIOB is driven low.
1 = TIOB is high. If WAVE = 0, this means that TIOB pin is high. If WAVE = 1, this means that TIOB is driven high.

805
6430F–ATARM–21-Feb-12

36.7.15
Name:

TC Interrupt Enable Register
TC_IERx [x=0..2]

Address:

0x40080024 (0)[0], 0x40080064 (0)[1], 0x400800A4 (0)[2]

Access:

Write-only

–

ETRGS

LDRBS

LDRAS

CPCS

CPBS

CPAS

LOVRS

COVFS

• COVFS: Counter Overflow
0 = no effect.
1 = enables the Counter Overflow Interrupt.
• LOVRS: Load Overrun
0 = no effect.
1 = enables the Load Overrun Interrupt.
• CPAS: RA Compare
0 = no effect.
1 = enables the RA Compare Interrupt.
• CPBS: RB Compare
0 = no effect.
1 = enables the RB Compare Interrupt.
• CPCS: RC Compare
0 = no effect.
1 = enables the RC Compare Interrupt.
• LDRAS: RA Loading
0 = no effect.
1 = enables the RA Load Interrupt.
• LDRBS: RB Loading
0 = no effect.
1 = enables the RB Load Interrupt.
• ETRGS: External Trigger
0 = no effect.
1 = enables the External Trigger Interrupt.
806

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
36.7.16
Name:

TC Interrupt Disable Register
TC_IDRx [x=0..2]

Address:

0x40080028 (0)[0], 0x40080068 (0)[1], 0x400800A8 (0)[2]

Access:

Write-only

–

ETRGS

LDRBS

LDRAS

CPCS

CPBS

CPAS

LOVRS

COVFS

• COVFS: Counter Overflow
0 = no effect.
1 = disables the Counter Overflow Interrupt.
• LOVRS: Load Overrun
0 = no effect.
1 = disables the Load Overrun Interrupt (if WAVE = 0).
• CPAS: RA Compare
0 = no effect.
1 = disables the RA Compare Interrupt (if WAVE = 1).
• CPBS: RB Compare
0 = no effect.
1 = disables the RB Compare Interrupt (if WAVE = 1).
• CPCS: RC Compare
0 = no effect.
1 = disables the RC Compare Interrupt.
• LDRAS: RA Loading
0 = no effect.
1 = disables the RA Load Interrupt (if WAVE = 0).
• LDRBS: RB Loading
0 = no effect.
1 = disables the RB Load Interrupt (if WAVE = 0).
• ETRGS: External Trigger
0 = no effect.
1 = disables the External Trigger Interrupt.

807
6430F–ATARM–21-Feb-12

36.7.17
Name:

TC Interrupt Mask Register
TC_IMRx [x=0..2]

Address:

0x4008002C (0)[0], 0x4008006C (0)[1], 0x400800AC (0)[2]

Access:

Read-only

–

ETRGS

LDRBS

LDRAS

CPCS

CPBS

CPAS

LOVRS

COVFS

• COVFS: Counter Overflow
0 = the Counter Overflow Interrupt is disabled.
1 = the Counter Overflow Interrupt is enabled.
• LOVRS: Load Overrun
0 = the Load Overrun Interrupt is disabled.
1 = the Load Overrun Interrupt is enabled.
• CPAS: RA Compare
0 = the RA Compare Interrupt is disabled.
1 = the RA Compare Interrupt is enabled.
• CPBS: RB Compare
0 = the RB Compare Interrupt is disabled.
1 = the RB Compare Interrupt is enabled.
• CPCS: RC Compare
0 = the RC Compare Interrupt is disabled.
1 = the RC Compare Interrupt is enabled.
• LDRAS: RA Loading
0 = the Load RA Interrupt is disabled.
1 = the Load RA Interrupt is enabled.
• LDRBS: RB Loading
0 = the Load RB Interrupt is disabled.
1 = the Load RB Interrupt is enabled.
• ETRGS: External Trigger
0 = the External Trigger Interrupt is disabled.
1 = the External Trigger Interrupt is enabled.
808

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37. High Speed Multimedia Card Interface (HSMCI)
37.1

Description
The High Speed Multimedia Card Interface (HSMCI) supports the MultiMedia Card (MMC)
Specification V4.3, the SD Memory Card Specification V2.0, the SDIO V2.0 specification and
CE-ATA V1.1.
The HSMCI includes a command register, response registers, data registers, timeout counters
and error detection logic that automatically handle the transmission of commands and, when
required, the reception of the associated responses and data with a limited processor overhead.
The HSMCI supports stream, block and multi block data read and write, and is compatible with
the DMA Controller (DMAC), minimizing processor intervention for large buffer transfers.
The HSMCI operates at a rate of up to Master Clock divided by 2 and supports the interfacing of
1 slot(s). Each slot may be used to interface with a High Speed MultiMediaCard bus (up to 30
Cards) or with an SD Memory Card. Only one slot can be selected at a time (slots are multiplexed). A bit field in the SD Card Register performs this selection.
The SD Memory Card communication is based on a 9-pin interface (clock, command, four data
and three power lines) and the High Speed MultiMedia Card on a 7-pin interface (clock, command, one data, three power lines and one reserved for future use).
The SD Memory Card interface also supports High Speed MultiMedia Card operations. The
main differences between SD and High Speed MultiMedia Cards are the initialization process
and the bus topology.
HSMCI fully supports CE-ATA Revision 1.1, built on the MMC System Specification v4.0. The
module includes dedicated hardware to issue the command completion signal and capture the
host command completion signal disable.

37.2

Embedded Characteristics
•
•
•
•
•
•
•
•
•

•
•
•
•
•

Compatible with MultiMedia Card Specification Version 4.3
Compatible with SD Memory Card Specification Version 2.0
Compatible with SDIO Specification Version 2.0
Compatible with CE-ATA Specification 1.1
Cards Clock Rate Up to Master Clock Divided by 2
Boot Operation Mode Support
High Speed Mode Support
Embedded Power Management to Slow Down Clock Rate When Not Used
Supports 1 Multiplexed Slot(s)
– Each Slot for either a High Speed MultiMediaCard Bus (Up to 30 Cards) or an SD
Memory Card
Support for Stream, Block and Multi-block Data Read and Write
Supports Connection to DMA Controller (DMAC)
– Minimizes Processor Intervention for Large Buffer Transfers
Built in FIFO (from 16 to 256 bytes) with Large Memory Aperture Supporting Incremental
Access
Support for CE-ATA Completion Signal Disable Command
Protection Against Unexpected Modification On-the-Fly of the Configuration Registers

809
6430F–ATARM–21-Feb-12

37.3

Block Diagram

Figure 37-1. Block Diagram
APB Bridge

DMAC
APB
MCCK

HSMCI Interface
PMC

MCK

PIO

(1)

MCCDA

(1)

MCDA0

(1)

MCDA1 (1)
MCDA2

(1)

MCDA3

(1)

MCDA4

(1)

MCDA5

(1)

MCDA6 (1)
Interrupt Control

MCDA7

(1)

HSMCI Interrupt

Note:

810

1. When several HSMCI (x HSMCI) are embedded in a product, MCCK refers to HSMCIx_CK, MCCDA to HSMCIx_CDA,
MCDAy to HSMCIx_DAy.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.4

Application Block Diagram
Figure 37-2. Application Block Diagram
Application Layer
ex: File System, Audio, Security, etc.

Physical Layer
HSMCI Interface

1 2 3 4 5 6 7

1 2 3 4 5 6 78
9
9 1011

1213 8

SDCard

MMC

37.5

Pin Name List

Table 37-1.

I/O Lines Description for 8-bit Configuration
Pin Description

Type(1)

Comments

MCCDA

Command/response

I/O/PP/OD

CMD of an MMC or SDCard/SDIO

MCCK

Clock

I/O

CLK of an MMC or SD Card/SDIO

MCDA0 - MCDA7

Data 0..7 of Slot A

I/O/PP

DAT[0..7] of an MMC
DAT[0..3] of an SD Card/SDIO

MCDB0 - MCDB7

Data 0..7 of Slot B

I/O/PP

DAT[0..7] of an MMC
DAT[0..3] of an SD Card/SDIO

MCDC0 - MCDC7

Data 0..7 of Slot C

I/O/PP

DAT[0..7] of an MMC
DAT[0..3] of an SD Card/SDIO

MCDD0 - MCDD7

Data 0..7 of Slot D

I/O/PP

DAT[0..7] of an MMC
DAT[0..3] of an SD Card/SDIO

Pin Name

Notes:

(2)

1. I: Input, O: Output, PP: Push/Pull, OD: Open Drain.
2. When several HSMCI (x HSMCI) are embedded in a product, MCCK refers to HSMCIx_CK, MCCDA to HSMCIx_CDA,
MCCDB to HSMCIx_CDB, MCCDC to HSMCIx_CDC, MCCDD to HSMCIx_CDD, MCDAy to HSMCIx_DAy, MCDBy to
HSMCIx_DBy, MCDCy to HSMCIx_DCy, MCDDy to HSMCIx_DDy.

811
6430F–ATARM–21-Feb-12

37.6
37.6.1

Product Dependencies
I/O Lines
The pins used for interfacing the High Speed MultiMedia Cards or SD Cards are multiplexed with
PIO lines. The programmer must first program the PIO controllers to assign the peripheral functions to HSMCI pins.
Table 37-2.

I/O Lines

Instance

Signal

I/O Line

Peripheral

HSMCI

MCCDA

PA4

HSMCI

MCCK

PA3

HSMCI

MCDA0

PA5

HSMCI

MCDA1

PA6

HSMCI

MCDA2

PA7

HSMCI

MCDA3

PA8

HSMCI

MCDA4

PC28

HSMCI

MCDA5

PC29

HSMCI

MCDA6

PC30

HSMCI

MCDA7

PC31

37.6.2

Power Management
The HSMCI is clocked through the Power Management Controller (PMC), so the programmer
must first configure the PMC to enable the HSMCI clock.

37.6.3

Interrupt
The HSMCI interface has an interrupt line connected to the Nested Vector Interrupt Controller
(NVIC).
Handling the HSMCI interrupt requires programming the NVIC before configuring the HSMCI.

37.7

Bus Topology
Figure 37-3. High Speed MultiMedia Memory Card Bus Topology
1 2 3 4 5 6 7

9 1011

1213 8

MMC

812

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The High Speed MultiMedia Card communication is based on a 13-pin serial bus interface. It has
three communication lines and four supply lines.
Table 37-3.

Bus Topology

Pin
Number

Name

Type(1)

Description

HSMCI Pin Name(2)
(Slot z)

DAT[3]

I/O/PP

Data

MCDz3

CMD

I/O/PP/OD

Command/response

MCCDz

VSS1

Supply voltage ground

VSS

VDD

Supply voltage

VDD

CLK

I/O

Clock

MCCK

VSS2

Supply voltage ground

VSS

DAT[0]

I/O/PP

Data 0

MCDz0

DAT[1]

I/O/PP

Data 1

MCDz1

DAT[2]

I/O/PP

Data 2

MCDz2

DAT[4]

I/O/PP

Data 4

MCDz4

DAT[5]

I/O/PP

Data 5

MCDz5

DAT[6]

I/O/PP

Data 6

MCDz6

DAT[7]

I/O/PP

Data 7

MCDz7

Notes:

1. I: Input, O: Output, PP: Push/Pull, OD: Open Drain.
2. When several HSMCI (x HSMCI) are embedded in a product, MCCK refers to HSMCIx_CK,
MCCDA to HSMCIx_CDA, MCDAy to HSMCIx_DAy.

Figure 37-4. MMC Bus Connections (One Slot)

HSMCI
MCDA0

MCCDA

MCCK

1 2 3 4 5 6 7

9 1011

1213 8

MMC1

Note:

1213 8

MMC2

1213 8

MMC3

When several HSMCI (x HSMCI) are embedded in a product, MCCK refers to HSMCIx_CK,
MCCDA to HSMCIx_CDA MCDAy to HSMCIx_DAy.

813
6430F–ATARM–21-Feb-12

Figure 37-5. SD Memory Card Bus Topology
1 2 3 4 5 6 78
9

SD CARD

The SD Memory Card bus includes the signals listed in Table 37-4.
Table 37-4.

SD Memory Card Bus Signals

Pin
Number

Name

Type(1)

Description

HSMCI Pin Name(2)
(Slot z)

CD/DAT[3]

I/O/PP

Card detect/ Data line Bit 3

MCDz3

CMD

Command/response

MCCDz

VSS1

Supply voltage ground

VSS

VDD

Supply voltage

VDD

CLK

I/O

Clock

MCCK

VSS2

Supply voltage ground

VSS

DAT[0]

I/O/PP

Data line Bit 0

MCDz0

DAT[1]

I/O/PP

Data line Bit 1 or Interrupt

MCDz1

DAT[2]

I/O/PP

Data line Bit 2

MCDz2

Notes:

1. I: input, O: output, PP: Push Pull, OD: Open Drain.
2. When several HSMCI (x HSMCI) are embedded in a product, MCCK refers to HSMCIx_CK,
MCCDA to HSMCIx_CDA, MCDAy to HSMCIx_DAy.

MCDA0 - MCDA3
MCCK

SD CARD

MCCDA

1 2 3 4 5 6 78

Figure 37-6. SD Card Bus Connections with One Slot

Note:

When several HSMCI (x HSMCI) are embedded in a product, MCCK refers to HSMCIx_CK,
MCCDA to HSMCIx_CDA MCDAy to HSMCIx_DAy.

When the HSMCI is configured to operate with SD memory cards, the width of the data bus can
be selected in the HSMCI_SDCR register. Clearing the SDCBUS bit in this register means that
the width is one bit; setting it means that the width is four bits. In the case of High Speed MultiMedia cards, only the data line 0 is used. The other data lines can be used as independent
PIOs.

814

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.8

High Speed MultiMediaCard Operations
After a power-on reset, the cards are initialized by a special message-based High Speed MultiMediaCard bus protocol. Each message is represented by one of the following tokens:
• Command: A command is a token that starts an operation. A command is sent from the host
either to a single card (addressed command) or to all connected cards (broadcast
command). A command is transferred serially on the CMD line.
• Response: A response is a token which is sent from an addressed card or (synchronously)
from all connected cards to the host as an answer to a previously received command. A
response is transferred serially on the CMD line.
• Data: Data can be transferred from the card to the host or vice versa. Data is transferred via
the data line.
Card addressing is implemented using a session address assigned during the initialization
phase by the bus controller to all currently connected cards. Their unique CID number identifies
individual cards.
The structure of commands, responses and data blocks is described in the High Speed MultiMedia-Card System Specification. See also Table 37-5 on page 816.
High Speed MultiMediaCard bus data transfers are composed of these tokens.
There are different types of operations. Addressed operations always contain a command and a
response token. In addition, some operations have a data token; the others transfer their information directly within the command or response structure. In this case, no data token is present
in an operation. The bits on the DAT and the CMD lines are transferred synchronous to the clock
HSMCI Clock.
Two types of data transfer commands are defined:
• Sequential commands: These commands initiate a continuous data stream. They are
terminated only when a stop command follows on the CMD line. This mode reduces the
command overhead to an absolute minimum.
• Block-oriented commands: These commands send a data block succeeded by CRC bits.
Both read and write operations allow either single or multiple block transmission. A multiple
block transmission is terminated when a stop command follows on the CMD line similarly to the
sequential read or when a multiple block transmission has a pre-defined block count (See “Data
Transfer Operation” on page 818.).
The HSMCI provides a set of registers to perform the entire range of High Speed MultiMedia
Card operations.

37.8.1

Command - Response Operation
After reset, the HSMCI is disabled and becomes valid after setting the MCIEN bit in the
HSMCI_CR Control Register.
The PWSEN bit saves power by dividing the HSMCI clock by 2PWSDIV + 1 when the bus is
inactive.
The two bits, RDPROOF and WRPROOF in the HSMCI Mode Register (HSMCI_MR) allow
stopping the HSMCI Clock during read or write access if the internal FIFO is full. This will guarantee data integrity, not bandwidth.
All the timings for High Speed MultiMedia Card are defined in the High Speed MultiMediaCard
System Specification.
815

6430F–ATARM–21-Feb-12

The two bus modes (open drain and push/pull) needed to process all the operations are defined
in the HSMCI command register. The HSMCI_CMDR allows a command to be carried out.
For example, to perform an ALL_SEND_CID command:
Host Command
CMD

Content

CRC

NID Cycles
E

******

CID
Z

Content

The command ALL_SEND_CID and the fields and values for the HSMCI_CMDR Control Register are described in Table 37-5 and Table 37-6.
Table 37-5.

ALL_SEND_CID Command Description

CMD Index

Type

Argument

Resp

Abbreviation

CMD2

bcr(1)

[31:0] stuff bits

ALL_SEND_CID

Note:

Command
Description
Asks all cards to send
their CID numbers on
the CMD line

1. bcr means broadcast command with response.

Table 37-6.

Fields and Values for HSMCI_CMDR Command Register

Field

Value

CMDNB (command number)

2 (CMD2)

RSPTYP (response type)

2 (R2: 136 bits response)

SPCMD (special command)

0 (not a special command)

OPCMD (open drain command)

MAXLAT (max latency for command to response)

0 (NID cycles ==> 5 cycles)

TRCMD (transfer command)

0 (No transfer)

TRDIR (transfer direction)

X (available only in transfer command)

TRTYP (transfer type)

X (available only in transfer command)

IOSPCMD (SDIO special command)

0 (not a special command)

The HSMCI_ARGR contains the argument field of the command.
To send a command, the user must perform the following steps:
• Fill the argument register (HSMCI_ARGR) with the command argument.
• Set the command register (HSMCI_CMDR) (see Table 37-6).
The command is sent immediately after writing the command register.
While the card maintains a busy indication (at the end of a STOP_TRANSMISSION command
CMD12, for example), a new command shall not be sent. The NOTBUSY flag in the status register (HSMCI_SR) is asserted when the card releases the busy indication.
If the command requires a response, it can be read in the HSMCI response register
(HSMCI_RSPR). The response size can be from 48 bits up to 136 bits depending on the command. The HSMCI embeds an error detection to prevent any corrupted data during the transfer.

816

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The following flowchart shows how to send a command to the card and read the response if
needed. In this example, the status register bits are polled but setting the appropriate bits in the
interrupt enable register (HSMCI_IER) allows using an interrupt method.
Figure 37-7. Command/Response Functional Flow Diagram
Set the command argument
HSMCI_ARGR = Argument(1)

Set the command
HSMCI_CMDR = Command

Read HSMCI_SR

Wait for command
ready status flag

0
CMDRDY

Check error bits in the
status register (1)

Yes
Status error flags?
(1)
RETURN ERROR

Read response if required

Does the command involve
a busy indication?

RETURN OK

Read HSMCI_SR

0
NOTBUSY

1
RETURN OK

817
6430F–ATARM–21-Feb-12

Note:

37.8.2

1. If the command is SEND_OP_COND, the CRC error flag is always present (refer to R3 response in the High Speed MultiMedia Card specification).

Data Transfer Operation
The High Speed MultiMedia Card allows several read/write operations (single block, multiple
blocks, stream, etc.). These kinds of transfer can be selected setting the Transfer Type (TRTYP)
field in the HSMCI Command Register (HSMCI_CMDR).
These operations can be done using the features of the DMA Controller.

In all cases, the block length (BLKLEN field) must be defined either in the mode register
HSMCI_MR, or in the Block Register HSMCI_BLKR. This field determines the size of the data
block.
Consequent to MMC Specification 3.1, two types of multiple block read (or write) transactions
are defined (the host can use either one at any time):
• Open-ended/Infinite Multiple block read (or write):
The number of blocks for the read (or write) multiple block operation is not defined. The card
will continuously transfer (or program) data blocks until a stop transmission command is
received.
• Multiple block read (or write) with pre-defined block count (since version 3.1 and higher):
The card will transfer (or program) the requested number of data blocks and terminate the
transaction. The stop command is not required at the end of this type of multiple block read
(or write), unless terminated with an error. In order to start a multiple block read (or write)
with pre-defined block count, the host must correctly program the HSMCI Block Register
(HSMCI_BLKR). Otherwise the card will start an open-ended multiple block read. The BCNT
field of the Block Register defines the number of blocks to transfer (from 1 to 65535 blocks).
Programming the value 0 in the BCNT field corresponds to an infinite block transfer.
37.8.3

818

Read Operation
The following flowchart (Figure 37-8) shows how to read a single block with or without use of
DMAC facilities. In this example, a polling method is used to wait for the end of read. Similarly,
the user can configure the interrupt enable register (HSMCI_IER) to trigger an interrupt at the
end of read.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 37-8. Read Functional Flow Diagram
Send SELECT/DESELECT_CARD
command(1) to select the card

Send SET_BLOCKLEN command(1)

Yes
Read with DMAC

Reset the DMAEN bit
HSMCI_DMA &= ~DMAEN
Set the block length (in bytes)
HSMCI_MR l= (BlockLength<<16) (2)
Set the block count (if neccessary)
HSMCI_BLKR l= (BlockCount<<0)

Set the DMAEN bit
HSMCI_DMA |= DMAEN
Set the block length (in bytes)
HSMCI_BLKR |= (BlockLength << 16)(2)

Configure the DMA channel X
DMAC_SADDRx = Data Address
DMAC_BTSIZE = BlockLength/4
DMACHEN[X] = TRUE

Send READ_SINGLE_BLOCK
command(1)

Number of words to read = BlockLength/4
Send READ_SINGLE_BLOCK
command(1)

Yes
Number of words to read = 0 ?
Read status register HSMCI_SR
No
Read status register HSMCI_SR
Poll the bit
XFRDONE = 0?

Poll the bit
RXRDY = 0?

Yes

Yes
No

RETURN

Read data = HSMCI_RDR
Number of words to read =
Number of words to read -1

RETURN

Notes:

1. It is assumed that this command has been correctly sent (see Figure 37-7).
2. This field is also accessible in the HSMCI Block Register (HSMCI_BLKR).

819
6430F–ATARM–21-Feb-12

37.8.4

Write Operation
In write operation, the HSMCI Mode Register (HSMCI_MR) is used to define the padding value
when writing non-multiple block size. If the bit PADV is 0, then 0x00 value is used when padding
data, otherwise 0xFF is used.
If set, the bit DMAEN in the HSMCI_DMA register enables DMA transfer.
The following flowchart (Figure 37-9) shows how to write a single block with or without use of
DMA facilities. Polling or interrupt method can be used to wait for the end of write according to
the contents of the Interrupt Mask Register (HSMCI_IMR).

820

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 37-9. Write Functional Flow Diagram
Send SELECT/DESELECT_CARD
command(1) to select the card

Send SET_BLOCKLEN command(1)

Yes

No
Write using DMAC

Reset theDMAEN bit
HSMCI_DMA &= ~DMAEN
Set the block length (in bytes)
HSMCI_MR |= (BlockLength) <<16)(2)
Set the block count (if necessary)
HSMCI_BLKR |= (BlockCount << 0)

Set the DMAEN bit
HSMCI_DMA |= DMAEN
Set the block length (in bytes)
(2)
HSMCI_BLKR |= (BlockLength << 16)

Send WRITE_SINGLE_BLOCK
command(1)

Send WRITE_SINGLE_BLOCK
(1)
command

Configure the DMA channel X
DMAC_DADDRx = Data Address to write
DMAC_BTSIZE = BlockLength/4

Number of words to write = BlockLength/4

DMAC_CHEN[X] = TRUE
Yes
Number of words to write = 0 ?
Read status register HSMCI_SR
No
Read status register HSMCI_SR
Poll the bit
XFRDONE = 0?
Poll the bit
TXRDY = 0?

Yes

Yes
No

No
RETURN

HSMCI_TDR = Data to write
Number of words to write =
Number of words to write -1

RETURN

Note:

1. It is assumed that this command has been correctly sent (see Figure 37-7).
2. This field is also accessible in the HSMCI Block Register (HSMCI_BLKR).

821
6430F–ATARM–21-Feb-12

The following flowchart (Figure 37-10) shows how to manage read multiple block and write multiple block transfers with the DMA Controller. Polling or interrupt method can be used to wait for
the end of write according to the contents of the Interrupt Mask Register (HSMCI_IMR).
Figure 37-10. Read Multiple Block and Write Multiple Block
Send SELECT/DESELECT_CARD
command(1) to select the card

Send SET_BLOCKLEN command(1)

Set the block length
HSMCI_MR |= (BlockLength << 16)
Set the DMAEN bit
HSMCI_DMA |= DMAEN

Send WRITE_MULTIPLE_BLOCK or
READ_MULTIPLE_BLOCK command(1)

Configure the HDMA channel X
DMAC_SADDRx and DMAC_DADDRx
DMAC_BTSIZE = BlockLength/4

DMAC_CHEN[X] = TRUE

Read status register DMAC_EBCISR
and Poll Bit CBTC[X]

New Buffer ?(2)

Yes

No
Read status register HSMCI_SR
and Poll Bit FIFOEMPTY

Send STOP_TRANSMISSION
command(1)

Poll the bit
XFRDONE = 1

Yes
RETURN

Notes:

1. It is assumed that this command has been correctly sent (see Figure 37-7).
2. Handle errors reported in HSMCI_SR.

822

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.8.5

WRITE_SINGLE_BLOCK Operation using DMA Controller
1. Wait until the current command execution has successfully terminated.
c.

Check that CMDRDY and NOTBUSY fields are asserted in HSMCI_SR

2. Program the block length in the card. This value defines the value block_length.
3. Program the block length in the HSMCI configuration register with block_length value.
4. Program HSMCI_DMA register with the following fields:
– OFFSET field with dma_offset.
– CHKSIZE is user defined and set according to DMAC_DCSIZE.
– DMAEN is set to true to enable DMA hardware handshaking in the HSMCI. This bit
was previously set to false.
5. Issue a WRITE_SINGLE_BLOCK command writing HSMCI_ARG then HSMCI_CMDR.
6. Program the DMA Controller.
a. Read the channel Register to choose an available (disabled) channel.
b.

Clear any pending interrupts on the channel from the previous DMAC transfer by
reading the DMAC_EBCISR register.

Program the channel registers.

d. The DMAC_SADDRx register for channel x must be set to the location of the
source data. When the first data location is not word aligned, the two LSB bits
define the temporary value called dma_offset. The two LSB bits of
DMAC_SADDRx must be set to 0.
e. The DMAC_DADDRx register for channel x must be set with the starting address of
the HSMCI_FIFO address.
f.

Program DMAC_CTRLAx register of channel x with the following field’s values:
–DST_WIDTH is set to WORD.
–SRC_WIDTH is set to WORD.
–DCSIZE must be set according to the value of HSMCI_DMA, CHKSIZE field.
–BTSIZE is programmed with CEILING((block_length + dma_offset) / 4), where
the ceiling function is the function that returns the smallest integer not less than
x.

g. Program DMAC_CTRLBx register for channel x with the following field’s values:
–DST_INCR is set to INCR, the block_length value must not be larger than the
HSMCI_FIFO aperture.
–SRC_INCR is set to INCR.
–FC field is programmed with memory to peripheral flow control mode.
–both DST_DSCR and SRC_DSCR are set to 1 (descriptor fetch is disabled).
–DIF and SIF are set with their respective layer ID. If SIF is different from DIF, the
DMA controller is able to prefetch data and write HSMCI simultaneously.
h. Program DMAC_CFGx register for channel x with the following field’s values:
–FIFOCFG defines the watermark of the DMAC channel FIFO.
–DST_H2SEL is set to true to enable hardware handshaking on the destination.
–DST_PER is programmed with the hardware handshaking ID of the targeted
HSMCI Host Controller.

823
6430F–ATARM–21-Feb-12

Enable Channel x, writing one to DMAC_CHER[x]. The DMAC is ready and waiting
for request.

7. Wait for XFRDONE in HSMCI_SR register.
37.8.6
37.8.6.1

READ_SINGLE_BLOCK Operation using DMA Controller
Block Length is Multiple of 4
1. Wait until the current command execution has successfully completed.
a. Check that CMDRDY and NOTBUSY are asserted in HSMCI_SR.
2. Program the block length in the card. This value defines the value block_length.
3. Program the block length in the HSMCI configuration register with block_length value.
4. Set RDPROOF bit in HSMCI_MR to avoid overflow.
5. Program HSMCI_DMA register with the following fields:
– ROPT field is set to 0.
– OFFSET field is set to 0.
– CHKSIZE is user defined.
– DMAEN is set to true to enable DMAC hardware handshaking in the HSMCI. This bit
was previously set to false.
6. Issue a READ_SINGLE_BLOCK command.
7. Program the DMA controller.
a. Read the channel Register to choose an available (disabled) channel.
b.

Clear any pending interrupts on the channel from the previous DMA transfer by
reading the DMAC_EBCISR register.

Program the channel registers.

d. The DMAC_SADDRx register for channel x must be set with the starting address of
the HSMCI_FIFO address.
e. The DMAC_DADDRx register for channel x must be word aligned.
f.

Program DMAC_CTRLAx register of channel x with the following field’s values:
–DST_WIDTH is set to WORD.
–SRC_WIDTH is set to WORD.
–SCSIZE must be set according to the value of HSMCI_DMA, CHKSIZE field.
–BTSIZE is programmed with block_length/4.

g. Program DMAC_CTRLBx register for channel x with the following field’s values:
–DST_INCR is set to INCR.
–SRC_INCR is set to INCR.
–FC field is programmed with peripheral to memory flow control mode.
–both DST_DSCR and SRC_DSCR are set to 1 (descriptor fetch is disabled).
–DIF and SIF are set with their respective layer ID. If SIF is different from DIF, the
DMA controller is able to prefetch data and write HSMCI simultaneously.
h. Program DMAC_CFGx register for channel x with the following field’s values:
–FIFOCFG defines the watermark of the DMA channel FIFO.
–SRC_H2SEL is set to true to enable hardware handshaking on the destination.
–SRC_PER is programmed with the hardware handshaking ID of the targeted
HSMCI Host Controller.

824

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
–Enable Channel x, writing one to DMAC_CHER[x]. The DMAC is ready and
waiting for request.
8. Wait for XFRDONE in HSMCI_SR register.
37.8.6.2

Block Length is Not Multiple of 4 and Padding Not Used (ROPT field in HSMCI_DMA register set to 0)
In the previous DMA transfer flow (block length multiple of 4), the DMA controller is configured to
use only WORD AHB access. When the block length is no longer a multiple of 4 this is no longer
true. The DMA controller is programmed to copy exactly the block length number of bytes using
2 transfer descriptors.
1. Use the previous step until READ_SINGLE_BLOCK then
2. Program the DMA controller to use a two descriptors linked list.
a. Read the channel Register to choose an available (disabled) channel.
b.

Clear any pending interrupts on the channel from the previous DMA transfer by
reading the DMAC_EBCISR register.

Program the channel registers in the Memory for the first descriptor. This descriptor
will be word oriented. This descriptor is referred to as LLI_W, standing for LLI word
oriented transfer.

d. The LLI_W.DMAC_SADDRx field in memory must be set with the starting address
of the HSMCI_FIFO address.
e. The LLI_W.DMAC_DADDRx field in the memory must be word aligned.
f.

Program LLI_W.DMAC_CTRLAx with the following field’s values:
–DST_WIDTH is set to WORD.
–SRC_WIDTH is set to WORD.
–SCSIZE must be set according to the value of HSMCI_DMA, CHKSIZE field.
–BTSIZE is programmed with block_length/4. If BTSIZE is zero, this descriptor is
skipped later.

g. Program LLI_W.DMAC_CTRLBx with the following field’s values:
–DST_INCR is set to INCR
–SRC_INCR is set to INCR
–FC field is programmed with peripheral to memory flow control mode.
–SRC_DSCR is set to zero. (descriptor fetch is enabled for the SRC)
–DST_DSCR is set to one. (descriptor fetch is disabled for the DST)
–DIF and SIF are set with their respective layer ID. If SIF is different from DIF, DMA
controller is able to prefetch data and write HSMCI simultaneously.
h. Program LLI_W.DMAC_CFGx register for channel x with the following field’s
values:
–FIFOCFG defines the watermark of the DMA channel FIFO.
–DST_REP is set to zero meaning that address are contiguous.
–SRC_H2SEL is set to true to enable hardware handshaking on the destination.
–SRC_PER is programmed with the hardware handshaking ID of the targeted
HSMCI Host Controller.
i.

Program LLI_W.DMAC_DSCRx with the address of LLI_B descriptor. And set
DSCRx_IF to the AHB Layer ID. This operation actually links the Word oriented

825
6430F–ATARM–21-Feb-12

descriptor on the second byte oriented descriptor. When block_length[1:0] is equal
to 0 (multiple of 4) LLI_W.DMAC_DSCRx points to 0, only LLI_W is relevant.
j.

Program the channel registers in the Memory for the second descriptor. This
descriptor will be byte oriented. This descriptor is referred to as LLI_B, standing for
LLI Byte oriented.

The LLI_B.DMAC_SADDRx field in memory must be set with the starting address
of the HSMCI_FIFO address.

The LLI_B.DMAC_DADDRx is not relevant if previous word aligned descriptor was
enabled. If 1, 2 or 3 bytes are transferred that address is user defined and not word
aligned.

m. Program LLI_B.DMAC_CTRLAx with the following field’s values:
–DST_WIDTH is set to BYTE.
–SRC_WIDTH is set to BYTE.
–SCSIZE must be set according to the value of HSMCI_DMA, CHKSIZE field.
–BTSIZE is programmed with block_length[1:0]. (last 1, 2, or 3 bytes of the buffer).
n. Program LLI_B.DMAC_CTRLBx with the following field’s values:
–DST_INCR is set to INCR
–SRC_INCR is set to INCR
–FC field is programmed with peripheral to memory flow control mode.
–Both SRC_DSCR and DST_DSCR are set to 1 (descriptor fetch is disabled) or
Next descriptor location points to 0.
–DIF and SIF are set with their respective layer ID. If SIF is different from DIF, DMA
Controller is able to prefetch data and write HSMCI simultaneously.
o.

Program LLI_B.DMAC_CFGx memory location for channel x with the following
field’s values:
–FIFOCFG defines the watermark of the DMA channel FIFO.
–SRC_H2SEL is set to true to enable hardware handshaking on the destination.
–SRC_PER is programmed with the hardware handshaking ID of the targeted
HSMCI Host Controller.

Program LLI_B.DMAC_DSCR with 0.

q. Program DMAC_CTRLBx register for channel x with 0. its content is updated with
the LLI fetch operation.
r.

Program DMAC_DSCRx with the address of LLI_W if block_length greater than 4
else with address of LLI_B.

Enable Channel x writing one to DMAC_CHER[x]. The DMAC is ready and waiting
for request.

3. Wait for XFRDONE in HSMCI_SR register.
37.8.6.3

Block Length is Not Multiple of 4, with Padding Value (ROPT field in HSMCI_DMA register set to 1)
When the ROPT field is set to one, The DMA Controller performs only WORD access on the bus
to transfer a non-multiple of 4 block length. Unlike previous flow, in which the transfer size is
rounded to the nearest multiple of 4.
1. Program the HSMCI Interface, see previous flow.
– ROPT field is set to 1.
2. Program the DMA Controller

826

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
a. Read the channel Register to choose an available (disabled) channel.
b.

Clear any pending interrupts on the channel from the previous DMA transfer by
reading the DMAC_EBCISR register.

Program the channel registers.

d. The DMAC_SADDRx register for channel x must be set with the starting address of
the HSMCI_FIFO address.
e. The DMAC_DADDRx register for channel x must be word aligned.
f.

Program DMAC_CTRLAx register of channel x with the following field’s values:
–DST_WIDTH is set to WORD
–SRC_WIDTH is set to WORD
–SCSIZE must be set according to the value of HSMCI_DMA.CHKSIZE Field.
–BTSIZE is programmed with CEILING(block_length/4).

g. Program DMAC_CTRLBx register for channel x with the following field’s values:
–DST_INCR is set to INCR
–SRC_INCR is set to INCR
–FC field is programmed with peripheral to memory flow control mode.
–both DST_DSCR and SRC_DSCR are set to 1. (descriptor fetch is disabled)
–DIF and SIF are set with their respective layer ID. If SIF is different from DIF, the
DMA Controller is able to prefetch data and write HSMCI simultaneously.
h. Program DMAC_CFGx register for channel x with the following field’s values:
–FIFOCFG defines the watermark of the DMA channel FIFO.
–SRC_H2SEL is set to true to enable hardware handshaking on the destination.
–SRC_PER is programmed with the hardware handshaking ID of the targeted
HSMCI Host Controller.
–Enable Channel x writing one to DMAC_CHER[x]. The DMAC is ready and
waiting for request.
3. Wait for XFRDONE in HSMCI_SR register.
37.8.7
37.8.7.1

WRITE_MULTIPLE_BLOCK
One Block per Descriptor
1. Wait until the current command execution has successfully terminated.
a. Check that CMDRDY and NOTBUSY are asserted in HSMCI_SR.
2. Program the block length in the card. This value defines the value block_length.
3. Program the block length in the HSMCI configuration register with block_length value.
4. Program HSMCI_DMA register with the following fields:
– OFFSET field with dma_offset.
– CHKSIZE is user defined.
– DMAEN is set to true to enable DMAC hardware handshaking in the HSMCI. This bit
was previously set to false.
5. Issue a WRITE_MULTIPLE_BLOCK command.
6. Program the DMA Controller to use a list of descriptors. Each descriptor transfers one
block of data. Block n of data is transferred with descriptor LLI(n).

827
6430F–ATARM–21-Feb-12

a. Read the channel Register to choose an available (disabled) channel.
b.

Clear any pending interrupts on the channel from the previous DMAC transfer by
reading the DMAC_EBCISR register.

Program a List of descriptors.

d. The LLI(n).DMAC_SADDRx memory location for channel x must be set to the location of the source data. When the first data location is not word aligned, the two
LSB bits define the temporary value called dma_offset. The two LSB bits of
LLI(n).DMAC_SADDRx must be set to 0.
e. The LLI(n).DMAC_DADDRx register for channel x must be set with the starting
address of the HSMCI_FIFO address.
f.

Program LLI(n).DMAC_CTRLAx register of channel x with the following field’s
values:
–DST_WIDTH is set to WORD.
–SRC_WIDTH is set to WORD.
–DCSIZE must be set according to the value of HSMCI_DMA, CHKSIZE field.
–BTSIZE is programmed with CEILING((block_length + dma_offset)/4).

g. Program LLI(n).DMAC_CTRLBx register for channel x with the following field’s
values:
–DST_INCR is set to INCR.
–SRC_INCR is set to INCR.
–DST_DSCR is set to 0 (fetch operation is enabled for the destination).
–SRC_DSCR is set to 1 (source address is contiguous).
–FC field is programmed with memory to peripheral flow control mode.
–Both DST_DSCR and SRC_DSCR are set to 1 (descriptor fetch is disabled).
–DIF and SIF are set with their respective layer ID. If SIF is different from DIF, DMA
Controller is able to prefetch data and write HSMCI simultaneously.
h. Program LLI(n).DMAC_CFGx register for channel x with the following field’s values:
–FIFOCFG defines the watermark of the DMA channel FIFO.
–DST_H2SEL is set to true to enable hardware handshaking on the destination.
–SRC_REP is set to 0. (contiguous memory access at block boundary)
–DST_PER is programmed with the hardware handshaking ID of the targeted
HSMCI Host Controller.
i.

If LLI(n) is the last descriptor, then LLI(n).DSCR points to 0 else LLI(n) points to the
start address of LLI(n+1).

Program DMAC_CTRLBx for channel register x with 0. Its content is updated with
the LLI fetch operation.

Program DMAC_DSCRx for channel register x with the address of the first descriptor LLI(0).

Enable Channel x writing one to DMAC_CHER[x]. The DMA is ready and waiting
for request.

7. Poll CBTC[x] bit in the DMAC_EBCISR Register.
8. If a new list of buffers shall be transferred, repeat step 6. Check and handle HSMCI
errors.
9. Poll FIFOEMPTY field in the HSMCI_SR.

828

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
10. Send The STOP_TRANSMISSION command writing HSMCI_ARG then
HSMCI_CMDR.
11. Wait for XFRDONE in HSMCI_SR register.
37.8.8
37.8.8.1

READ_MULTIPLE_BLOCK
Block Length is a Multiple of 4
1. Wait until the current command execution has successfully terminated.
a. Check that CMDRDY and NOTBUSY are asserted in HSMCI_SR.
2. Program the block length in the card. This value defines the value block_length.
3. Program the block length in the HSMCI configuration register with block_length value.
4. Set RDPROOF bit in HSMCI_MR to avoid overflow.
5. Program HSMCI_DMA register with the following fields:
– ROPT field is set to 0.
– OFFSET field is set to 0.
– CHKSIZE is user defined.
– DMAEN is set to true to enable DMAC hardware handshaking in the HSMCI. This bit
was previously set to false.
6. Issue a READ_MULTIPLE_BLOCK command.
7. Program the DMA Controller to use a list of descriptors:
a. Read the channel Register to choose an available (disabled) channel.
b.

Clear any pending interrupts on the channel from the previous DMA transfer by
reading the DMAC_EBCISR register.

Program the channel registers in the Memory with the first descriptor. This descriptor will be word oriented. This descriptor is referred to as LLI_W(n), standing for LLI
word oriented transfer for block n.

d. The LLI_W(n).DMAC_SADDRx field in memory must be set with the starting
address of the HSMCI_FIFO address.
e. The LLI_W(n).DMAC_DADDRx field in the memory must be word aligned.
f.

Program LLI_W(n).DMAC_CTRLAx with the following field’s values:
–DST_WIDTH is set to WORD
–SRC_WIDTH is set to WORD
–SCSIZE must be set according to the value of HSMCI_DMA, CHKSIZE field.
–BTSIZE is programmed with block_length/4.

g. Program LLI_W(n).DMAC_CTRLBx with the following field’s values:
–DST_INCR is set to INCR.
–SRC_INCR is set to INCR.
–FC field is programmed with peripheral to memory flow control mode.
–SRC_DSCR is set to 0 (descriptor fetch is enabled for the SRC).
–DST_DSCR is set to TRUE (descriptor fetch is disabled for the DST).
–DIF and SIF are set with their respective layer ID. If SIF is different from DIF, the
DMA Controller is able to prefetch data and write HSMCI simultaneously.

829
6430F–ATARM–21-Feb-12

h. Program LLI_W(n).DMAC_CFGx register for channel x with the following field’s
values:
–FIFOCFG defines the watermark of the DMA channel FIFO.
–DST_REP is set to zero. Addresses are contiguous.
–SRC_H2SEL is set to true to enable hardware handshaking on the destination.
–SRC_PER is programmed with the hardware handshaking ID of the targeted
HSMCI Host Controller.
i.

Program LLI_W(n).DMAC_DSCRx with the address of LLI_W(n+1) descriptor. And
set the DSCRx_IF to the AHB Layer ID. This operation actually links descriptors
together. If LLI_W(n) is the last descriptor then LLI_W(n).DMAC_DSCRx points to
0.

Program DMAC_CTRLBx register for channel x with 0. its content is updated with
the LLI Fetch operation.

Program DMAC_DSCRx register for channel x with the address of LLI_W(0).

Enable Channel x writing one to DMAC_CHER[x]. The DMA is ready and waiting
for request.

8. Poll CBTC[x] bit in the DMAC_EBCISR Register.
9. If a new list of buffer shall be transferred repeat step 6. Check and handle HSMCI
errors.
10. Poll FIFOEMPTY field in the HSMCI_SR.
11. Send The STOP_TRANSMISSION command writing the HSMCI_ARG then the
HSMCI_CMDR.
12. Wait for XFRDONE in HSMCI_SR register.
37.8.8.2

Block Length is Not Multiple of 4. (ROPT field in HSMCI_DMA register set to 0)
Two DMA Transfer descriptors are used to perform the HSMCI block transfer.
1. Use the previous step to configure the HSMCI to perform a READ_MULTIPLE_BLOCK
command.
2. Issue a READ_MULTIPLE_BLOCK command.
3. Program the DMA Controller to use a list of descriptors.
a. Read the channel register to choose an available (disabled) channel.
b.

Clear any pending interrupts on the channel from the previous DMAC transfer by
reading the DMAC_EBCISR register.

For every block of data repeat the following procedure:

d. Program the channel registers in the Memory for the first descriptor. This descriptor
will be word oriented. This descriptor is referred to as LLI_W(n) standing for LLI
word oriented transfer for block n.
e. The LLI_W(n).DMAC_SADDRx field in memory must be set with the starting
address of the HSMCI_FIFO address.
f.

The LLI_W(n).DMAC_DADDRx field in the memory must be word aligned.

g. Program LLI_W(n).DMAC_CTRLAx with the following field’s values:
–DST_WIDTH is set to WORD.
–SRC_WIDTH is set to WORD.
–SCSIZE must be set according to the value of HSMCI_DMA, CHKSIZE field.
–BTSIZE is programmed with block_length/4. If BTSIZE is zero, this descriptor is
skipped later.
830

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
h. Program LLI_W(n).DMAC_CTRLBx with the following field’s values:
–DST_INCR is set to INCR.
–SRC_INCR is set to INCR.
–FC field is programmed with peripheral to memory flow control mode.
–SRC_DSCR is set to 0 (descriptor fetch is enabled for the SRC).
–DST_DSCR is set to TRUE (descriptor fetch is disabled for the DST).
–DIF and SIF are set with their respective layer ID. If SIF is different from DIF, the
DMA Controller is able to prefetch data and write HSMCI simultaneously.
i.

Program LLI_W(n).DMAC_CFGx register for channel x with the following field’s
values:
–FIFOCFG defines the watermark of the DMA channel FIFO.
–DST_REP is set to zero. Address are contiguous.
–SRC_H2SEL is set to true to enable hardware handshaking on the destination.
–SRC_PER is programmed with the hardware handshaking ID of the targeted
HSMCI Host Controller.

Program LLI_W(n).DMAC_DSCRx with the address of LLI_B(n) descriptor. And set
the DSCRx_IF to the AHB Layer ID. This operation actually links the Word oriented
descriptor on the second byte oriented descriptor. When block_length[1:0] is equal
to 0 (multiple of 4) LLI_W(n).DMAC_DSCRx points to 0, only LLI_W(n) is relevant.

Program the channel registers in the Memory for the second descriptor. This
descriptor will be byte oriented. This descriptor is referred to as LLI_B(n), standing
for LLI Byte oriented.

The LLI_B(n).DMAC_SADDRx field in memory must be set with the starting
address of the HSMCI_FIFO address.

m. The LLI_B(n).DMAC_DADDRx is not relevant if previous word aligned descriptor
was enabled. If 1, 2 or 3 bytes are transferred, that address is user defined and not
word aligned.
n. Program LLI_B(n).DMAC_CTRLAx with the following field’s values:
–DST_WIDTH is set to BYTE.
–SRC_WIDTH is set to BYTE.
–SCSIZE must be set according to the value of HSMCI_DMA, CHKSIZE field.
–BTSIZE is programmed with block_length[1:0]. (last 1, 2, or 3 bytes of the buffer).
o.

Program LLI_B(n).DMAC_CTRLBx with the following field’s values:
–DST_INCR is set to INCR.
–SRC_INCR is set to INCR.
–FC field is programmed with peripheral to memory flow control mode.
–Both SRC_DSCR and DST_DSCR are set to 1 (descriptor fetch is disabled) or
Next descriptor location points to 0.
–DIF and SIF are set with their respective layer ID. If SIF is different from DIF, the
DMA Controller is able to prefetch data and write HSMCI simultaneously.

Program LLI_B(n).DMAC_CFGx memory location for channel x with the following
field’s values:
–FIFOCFG defines the watermark of the DMAC channel FIFO.
–SRC_H2SEL is set to true to enable hardware handshaking on the destination.

831
6430F–ATARM–21-Feb-12

–SRC_PER is programmed with the hardware handshaking ID of the targeted
HSMCI Host Controller
q. Program LLI_B(n).DMAC_DSCR with address of descriptor LLI_W(n+1). If
LLI_B(n) is the last descriptor, then program LLI_B(n).DMAC_DSCR with 0.
r.

Program DMAC_CTRLBx register for channel x with 0, its content is updated with
the LLI Fetch operation.

Program DMAC_DSCRx with the address of LLI_W(0) if block_length is greater
than 4 else with address of LLI_B(0).

Enable Channel x writing one to DMAC_CHER[x]. The DMAC is ready and waiting
for request.

4. Enable DMADONE interrupt in the HSMCI_IER register.
5. Poll CBTC[x] bit in the DMAC_EBCISR Register.
6. If a new list of buffers shall be transferred, repeat step 7. Check and handle HSMCI
errors.
7. Poll FIFOEMPTY field in the HSMCI_SR.
8. Send The STOP_TRANSMISSION command writing HSMCI_ARG then
HSMCI_CMDR.
9. Wait for XFRDONE in HSMCI_SR register.
37.8.8.3

Block Length is Not a Multiple of 4. (ROPT field in HSMCI_DMA register set to 1)
One DMA Transfer descriptor is used to perform the HSMCI block transfer, the DMA writes a
rounded up value to the nearest multiple of 4.
1. Use the previous step to configure the HSMCI to perform a READ_MULTIPLE_BLOCK.
2. Set the ROPT field to 1 in the HSMCI_DMA register.
3. Issue a READ_MULTIPLE_BLOCK command.
4. Program the DMA controller to use a list of descriptors:
a. Read the channel Register to choose an available (disabled) channel.
b.

Clear any pending interrupts on the channel from the previous DMAC transfer by
reading the DMAC_EBCISR register.

d. The LLI_W(n).DMAC_SADDRx field in memory must be set with the starting
address of the HSMCI_FIFO address.
e. The LLI_W(n).DMAC_DADDRx field in the memory must be word aligned.
f.

Program LLI_W(n).DMAC_CTRLAx with the following field’s values:
–DST_WIDTH is set to WORD.
–SRC_WIDTH is set to WORD.
–SCSIZE must be set according to the value of HSMCI_DMA, CHKSIZE field.
–BTSIZE is programmed with Ceiling(block_length/4).

g. Program LLI_W(n).DMAC_CTRLBx with the following field’s values:
–DST_INCR is set to INCR
–SRC_INCR is set to INCR
–FC field is programmed with peripheral to memory flow control mode.
–SRC_DSCR is set to 0. (descriptor fetch is enabled for the SRC)

832

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
–DST_DSCR is set to TRUE. (descriptor fetch is disabled for the DST)
–DIF and SIF are set with their respective layer ID. If SIF is different from DIF, the
DMA Controller is able to prefetch data and write HSMCI simultaneously.
h. Program LLI_W(n).DMAC_CFGx register for channel x with the following field’s
values:
–FIFOCFG defines the watermark of the DMA channel FIFO.
–DST_REP is set to zero. Address are contiguous.
–SRC_H2SEL is set to true to enable hardware handshaking on the destination.
–SRC_PER is programmed with the hardware handshaking ID of the targeted
HSMCI Host Controller.
i.

Program DMAC_CTRLBx register for channel x with 0. its content is updated with
the LLI Fetch operation.

Program DMAC_DSCRx register for channel x with the address of LLI_W(0).

Enable Channel x writing one to DMAC_CHER[x]. The DMAC is ready and waiting
for request.

5. Poll CBTC[x] bit in the DMAC_EBCISR Register.
6. If a new list of buffers shall be transferred repeat step 7. Check and handle HSMCI
errors.
7. Poll FIFOEMPTY field in the HSMCI_SR.
8. Send The STOP_TRANSMISSION command writing the HSMCI_ARG then the
HSMCI_CMDR.
9. Wait for XFRDONE in HSMCI_SR register.

37.9

SD/SDIO Card Operation
The High Speed MultiMedia Card Interface allows processing of SD Memory (Secure Digital
Memory Card) and SDIO (SD Input Output) Card commands.
SD/SDIO cards are based on the Multi Media Card (MMC) format, but are physically slightly
thicker and feature higher data transfer rates, a lock switch on the side to prevent accidental
overwriting and security features. The physical form factor, pin assignment and data transfer
protocol are forward-compatible with the High Speed MultiMedia Card with some additions. SD
slots can actually be used for more than flash memory cards. Devices that support SDIO can
use small devices designed for the SD form factor, such as GPS receivers, Wi-Fi or Bluetooth
adapters, modems, barcode readers, IrDA adapters, FM radio tuners, RFID readers, digital cameras and more.
SD/SDIO is covered by numerous patents and trademarks, and licensing is only available
through the Secure Digital Card Association.
The SD/SDIO Card communication is based on a 9-pin interface (Clock, Command, 4 x Data
and 3 x Power lines). The communication protocol is defined as a part of this specification. The
main difference between the SD/SDIO Card and the High Speed MultiMedia Card is the initialization process.

833
6430F–ATARM–21-Feb-12

The SD/SDIO Card Register (HSMCI_SDCR) allows selection of the Card Slot and the data bus
width.
The SD/SDIO Card bus allows dynamic configuration of the number of data lines. After power
up, by default, the SD/SDIO Card uses only DAT0 for data transfer. After initialization, the host
can change the bus width (number of active data lines).
37.9.1

SDIO Data Transfer Type
SDIO cards may transfer data in either a multi-byte (1 to 512 bytes) or an optional block format
(1 to 511 blocks), while the SD memory cards are fixed in the block transfer mode. The TRTYP
field in the HSMCI Command Register (HSMCI_CMDR) allows to choose between SDIO Byte or
SDIO Block transfer.
The number of bytes/blocks to transfer is set through the BCNT field in the HSMCI Block Register (HSMCI_BLKR). In SDIO Block mode, the field BLKLEN must be set to the data block size
while this field is not used in SDIO Byte mode.
An SDIO Card can have multiple I/O or combined I/O and memory (called Combo Card). Within
a multi-function SDIO or a Combo card, there are multiple devices (I/O and memory) that share
access to the SD bus. In order to allow the sharing of access to the host among multiple devices,
SDIO and combo cards can implement the optional concept of suspend/resume (Refer to the
SDIO Specification for more details). To send a suspend or a resume command, the host must
set the SDIO Special Command field (IOSPCMD) in the HSMCI Command Register.

37.9.2

SDIO Interrupts
Each function within an SDIO or Combo card may implement interrupts (Refer to the SDIO
Specification for more details). In order to allow the SDIO card to interrupt the host, an interrupt
function is added to a pin on the DAT[1] line to signal the card’s interrupt to the host. An SDIO
interrupt on each slot can be enabled through the HSMCI Interrupt Enable Register. The SDIO
interrupt is sampled regardless of the currently selected slot.

37.10 CE-ATA Operation
CE-ATA maps the streamlined ATA command set onto the MMC interface. The ATA task file is
mapped onto MMC register space.
CE-ATA utilizes five MMC commands:
• GO_IDLE_STATE (CMD0): used for hard reset.
• STOP_TRANSMISSION (CMD12): causes the ATA command currently executing to be
aborted.
• FAST_IO (CMD39): Used for single register access to the ATA taskfile registers, 8 bit access
only.
• RW_MULTIPLE_REGISTERS (CMD60): used to issue an ATA command or to access the
control/status registers.
• RW_MULTIPLE_BLOCK (CMD61): used to transfer data for an ATA command.
CE-ATA utilizes the same MMC command sequences for initialization as traditional MMC
devices.
37.10.1

Executing an ATA Polling Command
1. Issue READ_DMA_EXT with RW_MULTIPLE_REGISTER (CMD60) for 8kB of DATA.
2. Read the ATA status register until DRQ is set.

834

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
3. Issue RW_MULTIPLE_BLOCK (CMD61) to transfer DATA.
4. Read the ATA status register until DRQ && BSY are set to 0.
37.10.2

Executing an ATA Interrupt Command
1. Issue READ_DMA_EXT with RW_MULTIPLE_REGISTER (CMD60) for 8kB of DATA
with nIEN field set to zero to enable the command completion signal in the device.
2. Issue RW_MULTIPLE_BLOCK (CMD61) to transfer DATA.
3. Wait for Completion Signal Received Interrupt.

37.10.3

Aborting an ATA Command
If the host needs to abort an ATA command prior to the completion signal it must send a special
command to avoid potential collision on the command line. The SPCMD field of the
HSMCI_CMDR must be set to 3 to issue the CE-ATA completion Signal Disable Command.

37.10.4

CE-ATA Error Recovery
Several methods of ATA command failure may occur, including:
• No response to an MMC command, such as RW_MULTIPLE_REGISTER (CMD60).
• CRC is invalid for an MMC command or response.
• CRC16 is invalid for an MMC data packet.
• ATA Status register reflects an error by setting the ERR bit to one.
• The command completion signal does not arrive within a host specified time out period.
Error conditions are expected to happen infrequently. Thus, a robust error recovery mechanism
may be used for each error event. The recommended error recovery procedure after a timeout
is:
• Issue the command completion signal disable if nIEN was cleared to zero and the
RW_MULTIPLE_BLOCK (CMD61) response has been received.
• Issue STOP_TRANSMISSION (CMD12) and successfully receive the R1 response.
• Issue a software reset to the CE-ATA device using FAST_IO (CMD39).
If STOP_TRANMISSION (CMD12) is successful, then the device is again ready for ATA commands. However, if the error recovery procedure does not work as expected or there is another
timeout, the next step is to issue GO_IDLE_STATE (CMD0) to the device. GO_IDLE_STATE
(CMD0) is a hard reset to the device and completely resets all device states.
Note that after issuing GO_IDLE_STATE (CMD0), all device initialization needs to be completed
again. If the CE-ATA device completes all MMC commands correctly but fails the ATA command
with the ERR bit set in the ATA Status register, no error recovery action is required. The ATA
command itself failed implying that the device could not complete the action requested, however, there was no communication or protocol failure. After the device signals an error by setting
the ERR bit to one in the ATA Status register, the host may attempt to retry the command.

37.11 HSMCI Boot Operation Mode
In boot operation mode, the processor can read boot data from the slave (MMC device) by keeping the CMD line low after power-on before issuing CMD1. The data can be read from either the
boot area or user area, depending on register setting. As it is not possible to boot directly on SDCARD, a preliminary boot code must be stored in internal Flash.

835
6430F–ATARM–21-Feb-12

37.11.1

Boot Procedure, Processor Mode
1. Configure the HSMCI data bus width programming SDCBUS Field in the
HSMCI_SDCR register. The BOOT_BUS_WIDTH field located in the device Extended
CSD register must be set accordingly.
2. Set the byte count to 512 bytes and the block count to the desired number of blocks,
writing BLKLEN and BCNT fields of the HSMCI_BLKR Register.
3. Issue the Boot Operation Request command by writing to the HSMCI_CMDR register
with SPCMD field set to BOOTREQ, TRDIR set to READ and TRCMD set to “start data
transfer”.
4. The BOOT_ACK field located in the HSMCI_CMDR register must be set to one, if the
BOOT_ACK field of the MMC device located in the Extended CSD register is set to one.
5. Host processor can copy boot data sequentially as soon as the RXRDY flag is
asserted.
6. When Data transfer is completed, host processor shall terminate the boot stream by
writing the HSMCI_CMDR register with SPCMD field set to BOOTEND.

37.11.2

Boot Procedure DMA Mode
1. Configure the HSMCI data bus width by programming SDCBUS Field in the
HSMCI_SDCR register. The BOOT_BUS_WIDTH field in the device Extended CSD
register must be set accordingly.
2. Set the byte count to 512 bytes and the block count to the desired number of blocks by
writing BLKLEN and BCNT fields of the HSMCI_BLKR Register.
3. Enable DMA transfer in the HSMCI_DMA register.
4. Configure DMA controller, program the total amount of data to be transferred and
enable the relevant channel.
5. Issue the Boot Operation Request command by writing to the HSMCI_CMDR register
with SPCND set to BOOTREQ, TRDIR set to READ and TRCMD set to “start data
transfer”.
6. DMA controller copies the boot partition to the memory.
7. When DMA transfer is completed, host processor shall terminate the boot stream by
writing the HSMCI_CMDR register with SPCMD field set to BOOTEND.

37.12 HSMCI Transfer Done Timings
37.12.1

Definition
The XFRDONE flag in the HSMCI_SR indicates exactly when the read or write sequence is
finished.

37.12.2

Read Access
During a read access, the XFRDONE flag behaves as shown in Figure 37-11.

836

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 37-11. XFRDONE During a Read Access
CMD line
HSMCI read CMD

Card response

The CMDRDY flag is released 8 tbit after the end of the card response.

CMDRDY flag

Data
Last Block

1st Block
Not busy flag

XFRDONE flag

37.12.3

Write Access
During a write access, the XFRDONE flag behaves as shown in Figure 37-12.
Figure 37-12. XFRDONE During a Write Access
CMD line
HSMCI write CMD

CMDRDY flag

Card response

The CMDRDY flag is released 8 tbit after the end of the card response.

D0 is tied by the card
D0 is released

D0
1st Block

Last Block

1st Block

Last Block

Data bus - D0

Not busy flag

XFRDONE flag

837
6430F–ATARM–21-Feb-12

37.13 Write Protection Registers
To prevent any single software error that may corrupt HSMCI behavior, the entire HSMCI
address space from address offset 0x000 to 0x00FC can be write-protected by setting the
WPEN bit in the “HSMCI Write Protect Mode Register” (HSMCI_WPMR).
If a write access to anywhere in the HSMCI address space from address offset 0x000 to 0x00FC
is detected, then the WPVS flag in the HSMCI Write Protect Status Register (HSMCI_WPSR) is
set and the field WPVSRC indicates in which register the write access has been attempted.
The WPVS flag is reset by writing the HSMCI Write Protect Mode Register (HSMCI_WPMR)
with the appropriate access key, WPKEY.
The protected registers are:
• “HSMCI Mode Register” on page 841
• “HSMCI Data Timeout Register” on page 843
• “HSMCI SDCard/SDIO Register” on page 844
• “HSMCI Completion Signal Timeout Register” on page 850
• “HSMCI DMA Configuration Register” on page 864
• “HSMCI Configuration Register” on page 865

838

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.14 High Speed Multimedia Card Interface (HSMCI) User Interface
Table 37-7.

Offset

Name

Access

Reset

0x00

Control Register

HSMCI_CR

Write

–

0x04

Mode Register

HSMCI_MR

Read-write

0x0

0x08

Data Timeout Register

HSMCI_DTOR

Read-write

0x0

0x0C

SD/SDIO Card Register

HSMCI_SDCR

Read-write

0x0

0x10

Argument Register

HSMCI_ARGR

Read-write

0x0

0x14

Command Register

HSMCI_CMDR

Write

–

0x18

Block Register

HSMCI_BLKR

Read-write

0x0

0x1C

Completion Signal Timeout Register

0x20
0x24
0x28

HSMCI_CSTOR

Read-write

0x0

(1)

HSMCI_RSPR

Read

0x0

(1)

HSMCI_RSPR

Read

0x0

(1)

HSMCI_RSPR

Read

0x0

(1)

HSMCI_RSPR

Read

0x0

Response Register

Response Register
Response Register

0x2C

Response Register

0x30

Receive Data Register

HSMCI_RDR

Read

0x0

0x34

Transmit Data Register

HSMCI_TDR

Write

–

0x38 - 0x3C

Reserved

0x40

Status Register

HSMCI_SR

Read

0xC0E5

0x44

Interrupt Enable Register

HSMCI_IER

Write

–

0x48

Interrupt Disable Register

HSMCI_IDR

Write

–

0x4C

Interrupt Mask Register

HSMCI_IMR

Read

0x0

0x50

DMA Configuration Register

HSMCI_DMA

Read-write

0x00

0x54

Configuration Register

HSMCI_CFG

Read-write

0x00

–

0x58-0xE0

Reserved

0xE4

Write Protection Mode Register

HSMCI_WPMR

Read-write

–

0xE8

Write Protection Status Register

HSMCI_WPSR

Read-only

–

0xEC - 0xFC

Reserved

–

0x100-0x1FC

Reserved

–

HSMCI_FIFO0

Read-write

0x0

...

HSMCI_FIFO255

Read-write

0x0

0x200
...
0x5FC
Note:

FIFO Memory Aperture0
...
FIFO Memory Aperture255

1. The response register can be read by N accesses at the same HSMCI_RSPR or at consecutive addresses (0x20 to 0x2C).
N depends on the size of the response.

839
6430F–ATARM–21-Feb-12

37.14.1
Name:

HSMCI Control Register
HSMCI_CR

Address:

0x40000000

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
SWRST

6
–

5
–

4
–

3
PWSDIS

2
PWSEN

1
MCIDIS

0
MCIEN

• MCIEN: Multi-Media Interface Enable
0 = No effect.
1 = Enables the Multi-Media Interface if MCDIS is 0.
• MCIDIS: Multi-Media Interface Disable
0 = No effect.
1 = Disables the Multi-Media Interface.
• PWSEN: Power Save Mode Enable
0 = No effect.
1 = Enables the Power Saving Mode if PWSDIS is 0.
Warning: Before enabling this mode, the user must set a value different from 0 in the PWSDIV field (Mode Register,
HSMCI_MR).
• PWSDIS: Power Save Mode Disable
0 = No effect.
1 = Disables the Power Saving Mode.
• SWRST: Software Reset
0 = No effect.
1 = Resets the HSMCI. A software triggered hardware reset of the HSMCI interface is performed.

840

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.14.2
Name:

HSMCI Mode Register
HSMCI_MR

Address:

0x40000004

Access:

Read-write

BLKLEN
23

20
BLKLEN

15
–

14
PADV

13
FBYTE

12
WRPROOF

11
RDPROOF

9
PWSDIV

CLKDIV

This register can only be written if the WPEN bit is cleared in “HSMCI Write Protect Mode Register” on page 866.
• CLKDIV: Clock Divider
High Speed MultiMedia Card Interface clock (MCCK or HSMCI_CK) is Master Clock (MCK) divided by (2*(CLKDIV+1)).
• PWSDIV: Power Saving Divider
High Speed MultiMedia Card Interface clock is divided by 2(PWSDIV) + 1 when entering Power Saving Mode.
Warning: This value must be different from 0 before enabling the Power Save Mode in the HSMCI_CR (HSMCI_PWSEN
bit).
• RDPROOF Read Proof Enable
Enabling Read Proof allows to stop the HSMCI Clock during read access if the internal FIFO is full. This will guarantee data
integrity, not bandwidth.
0 = Disables Read Proof.
1 = Enables Read Proof.
• WRPROOF Write Proof Enable
Enabling Write Proof allows to stop the HSMCI Clock during write access if the internal FIFO is full. This will guarantee data
integrity, not bandwidth.
0 = Disables Write Proof.
1 = Enables Write Proof.
• FBYTE: Force Byte Transfer
Enabling Force Byte Transfer allow byte transfers, so that transfer of blocks with a size different from modulo 4 can be
supported.
Warning: BLKLEN value depends on FBYTE.
0 = Disables Force Byte Transfer.
1 = Enables Force Byte Transfer.

841
6430F–ATARM–21-Feb-12

• PADV: Padding Value
0 = 0x00 value is used when padding data in write transfer.
1 = 0xFF value is used when padding data in write transfer.
PADV may be only in manual transfer.
• BLKLEN: Data Block Length
This field determines the size of the data block.
This field is also accessible in the HSMCI Block Register (HSMCI_BLKR).
Bits 16 and 17 must be set to 0 if FBYTE is disabled.
Note:

842

In SDIO Byte mode, BLKLEN field is not used.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.14.3
Name:

HSMCI Data Timeout Register
HSMCI_DTOR

Address:

0x40000008

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

5
DTOMUL

DTOCYC

This register can only be written if the WPEN bit is cleared in “HSMCI Write Protect Mode Register” on page 866.
• DTOCYC: Data Timeout Cycle Number
These fields determine the maximum number of Master Clock cycles that the HSMCI waits between two data block transfers. It equals (DTOCYC x Multiplier).
• DTOMUL: Data Timeout Multiplier
Multiplier is defined by DTOMUL as shown in the following table:
Value

Name

Description

DTOCYC

DTOCYC x 16

128

DTOCYC x 128

256

DTOCYC x 256

1024

DTOCYC x 1024

4096

DTOCYC x 4096

65536

DTOCYC x 65536

1048576

DTOCYC x 1048576

If the data time-out set by DTOCYC and DTOMUL has been exceeded, the Data Time-out Error flag (DTOE) in the HSMCI
Status Register (HSMCI_SR) rises.

843
6430F–ATARM–21-Feb-12

37.14.4
Name:

HSMCI SDCard/SDIO Register
HSMCI_SDCR

Address:

0x4000000C

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

5
–

4
–

3
–

2
–

7
SDCBUS

0
SDCSEL

This register can only be written if the WPEN bit is cleared in “HSMCI Write Protect Mode Register” on page 866.
• SDCSEL: SDCard/SDIO Slot
Value

Name

Description

SLOTA

Slot A is selected.

SLOTB

–

SLOTC

–

SLOTD

–

• SDCBUS: SDCard/SDIO Bus Width

844

Value

Name

Description

1 bit

–

Reserved

4 bit

8 bit

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.14.5
Name:

HSMCI Argument Register
HSMCI_ARGR

Address:

0x40000010

Access:

Read-write

ARG
23

20
ARG

12
ARG

4
ARG

• ARG: Command Argument

845
6430F–ATARM–21-Feb-12

37.14.6
Name:

HSMCI Command Register
HSMCI_CMDR

Address:

0x40000014

Access:

Write-only

31
–

30
–

29
–

28
–

27
BOOT_ACK

26
ATACS

23
–

22
–

20
TRTYP

18
TRDIR

15
–

14
–

13
–

12
MAXLAT

11
OPDCMD

9
SPCMD

7
RSPTYP

24
IOSPCMD
16
TRCMD

CMDNB

This register is write-protected while CMDRDY is 0 in HSMCI_SR. If an Interrupt command is sent, this register is only writable by an interrupt response (field SPCMD). This means that the current command execution cannot be interrupted or
modified.
• CMDNB: Command Number
This is the command index.
• RSPTYP: Response Type
Value

Name

Description

NORESP

No response.

48_BIT

48-bit response.

136_BIT

136-bit response.

R1B

R1b response type

• SPCMD: Special Command

846

Value

Name

Description

STD

Not a special CMD.

INIT

Initialization CMD:
74 clock cycles for initialization sequence.

SYNC

Synchronized CMD:
Wait for the end of the current data block transfer before sending the pending command.

CE_ATA

CE-ATA Completion Signal disable Command.
The host cancels the ability for the device to return a command completion signal on the
command line.

IT_CMD

Interrupt command:
Corresponds to the Interrupt Mode (CMD40).

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Value

Name

Description

IT_RESP

BOR

Boot Operation Request.
Start a boot operation mode, the host processor can read boot data from the MMC device directly.

EBO

End Boot Operation.
This command allows the host processor to terminate the boot operation mode.

Interrupt response:
Corresponds to the Interrupt Mode (CMD40).

• OPDCMD: Open Drain Command
0 (PUSHPULL) = Push pull command.
1 (OPENDRAIN) = Open drain command.
• MAXLAT: Max Latency for Command to Response
0 (5) = 5-cycle max latency.
1 (64) = 64-cycle max latency.
• TRCMD: Transfer Command
Value

Name

Description

NO_DATA

START_DATA

No data transfer
Start data transfer

STOP_DATA

Stop data transfer

–

Reserved

• TRDIR: Transfer Direction
0 (WRITE) = Write.
1 (READ) = Read.
• TRTYP: Transfer Type
Value

Name

SINGLE

MULTIPLE

STREAM

Description
MMC/SDCard Single Block
MMC/SDCard Multiple Block
MMC Stream

BYTE

SDIO Byte

BLOCK

SDIO Block

• IOSPCMD: SDIO Special Command
Value

Name

Description

STD

SUSPEND

SDIO Suspend Command

RESUME

SDIO Resume Command

Not an SDIO Special Command

847
6430F–ATARM–21-Feb-12

• ATACS: ATA with Command Completion Signal
0 (NORMAL) = Normal operation mode.
1 (COMPLETION) = This bit indicates that a completion signal is expected within a programmed amount of time
(HSMCI_CSTOR).
• BOOT_ACK: Boot Operation Acknowledge.
The master can choose to receive the boot acknowledge from the slave when a Boot Request command is issued. When
set to one this field indicates that a Boot acknowledge is expected within a programmable amount of time defined with
DTOMUL and DTOCYC fields located in the HSMCI_DTOR register. If the acknowledge pattern is not received then an
acknowledge timeout error is raised. If the acknowledge pattern is corrupted then an acknowledge pattern error is set.

848

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.14.7
Name:

HSMCI Block Register
HSMCI_BLKR

Address:

0x40000018

Access:

Read-write

BLKLEN
23

20
BLKLEN

12
BCNT

4
BCNT

• BCNT: MMC/SDIO Block Count - SDIO Byte Count
This field determines the number of data byte(s) or block(s) to transfer.
The transfer data type and the authorized values for BCNT field are determined by the TRTYP field in the HSMCI Command Register (HSMCI_CMDR):
Value

Name

Description

MULTIPLE

MMC/SDCARD Multiple Block
From 1 to 65635: Value 0 corresponds to an infinite block transfer.

BYTE

SDIO Byte
From 1 to 512 bytes: Value 0 corresponds to a 512-byte transfer.
Values from 0x200 to 0xFFFF are forbidden.

BLOCK

SDIO Block
From 1 to 511 blocks: Value 0 corresponds to an infinite block transfer.
Values from 0x200 to 0xFFFF are forbidden.

Warning: In SDIO Byte and Block modes, writing to the 7 last bits of BCNT field is forbidden and may lead to unpredictable
results.
• BLKLEN: Data Block Length
This field determines the size of the data block.
This field is also accessible in the HSMCI Mode Register (HSMCI_MR).
Bits 16 and 17 must be set to 0 if FBYTE is disabled.
Note:

In SDIO Byte mode, BLKLEN field is not used.

849
6430F–ATARM–21-Feb-12

37.14.8
Name:

HSMCI Completion Signal Timeout Register
HSMCI_CSTOR

Address:

0x4000001C

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

5
CSTOMUL

CSTOCYC

This register can only be written if the WPEN bit is cleared in “HSMCI Write Protect Mode Register” on page 866.
• CSTOCYC: Completion Signal Timeout Cycle Number
These fields determine the maximum number of Master Clock cycles that the HSMCI waits between two data block transfers. Its value is calculated by (CSTOCYC x Multiplier).
• CSTOMUL: Completion Signal Timeout Multiplier
These fields determine the maximum number of Master Clock cycles that the HSMCI waits between two data block transfers. Its value is calculated by (CSTOCYC x Multiplier).
These fields determine the maximum number of Master Clock cycles that the HSMCI waits between the end of the data
transfer and the assertion of the completion signal. The data transfer comprises data phase and the optional busy phase. If
a non-DATA ATA command is issued, the HSMCI starts waiting immediately after the end of the response until the completion signal.
Multiplier is defined by CSTOMUL as shown in the following table:
Value

Name

Description

CSTOCYC x 1

CSTOCYC x 16

128

CSTOCYC x 128

256

CSTOCYC x 256

1024

CSTOCYC x 1024

4096

CSTOCYC x 4096

65536

CSTOCYC x 65536

1048576

CSTOCYC x 1048576

If the data time-out set by CSTOCYC and CSTOMUL has been exceeded, the Completion Signal Time-out Error flag
(CSTOE) in the HSMCI Status Register (HSMCI_SR) rises.

850

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.14.9
Name:

HSMCI Response Register
HSMCI_RSPR

Address:

0x40000020

Access:

Read-only

RSP
23

20
RSP

12
RSP

4
RSP

• RSP: Response
Note:

1. The response register can be read by N accesses at the same HSMCI_RSPR or at consecutive addresses (0x20 to 0x2C).
N depends on the size of the response.

851
6430F–ATARM–21-Feb-12

37.14.10 HSMCI Receive Data Register
Name:
HSMCI_RDR
Address:

0x40000030

Access:

Read-only

DATA
23

20
DATA

12
DATA

4
DATA

• DATA: Data to Read

852

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.14.11 HSMCI Transmit Data Register
Name:
HSMCI_TDR
Address:

0x40000034

Access:

Write-only

DATA
23

20
DATA

12
DATA

4
DATA

• DATA: Data to Write

853
6430F–ATARM–21-Feb-12

37.14.12 HSMCI Status Register
Name:
HSMCI_SR
Address:

0x40000040

Access:

Read-only

31
UNRE

30
OVRE

29
ACKRCVE

28
ACKRCV

27
XFRDONE

26
FIFOEMPTY

25
DMADONE

24
BLKOVRE

23
CSTOE

22
DTOE

21
DCRCE

20
RTOE

19
RENDE

18
RCRCE

17
RDIRE

16
RINDE

–

CSRCV

SDIOWAIT

–

8
MCI_SDIOIR
QA

7
–

6
–

5
NOTBUSY

4
DTIP

3
BLKE

2
TXRDY

1
RXRDY

0
CMDRDY

• CMDRDY: Command Ready
0 = A command is in progress.
1 = The last command has been sent. Cleared when writing in the HSMCI_CMDR.
• RXRDY: Receiver Ready
0 = Data has not yet been received since the last read of HSMCI_RDR.
1 = Data has been received since the last read of HSMCI_RDR.
• TXRDY: Transmit Ready
0= The last data written in HSMCI_TDR has not yet been transferred in the Shift Register.
1= The last data written in HSMCI_TDR has been transferred in the Shift Register.
• BLKE: Data Block Ended
This flag must be used only for Write Operations.
0 = A data block transfer is not yet finished. Cleared when reading the HSMCI_SR.
1 = A data block transfer has ended, including the CRC16 Status transmission.
the flag is set for each transmitted CRC Status.
Refer to the MMC or SD Specification for more details concerning the CRC Status.
• DTIP: Data Transfer in Progress
0 = No data transfer in progress.
1 = The current data transfer is still in progress, including CRC16 calculation. Cleared at the end of the CRC16 calculation.
• NOTBUSY: HSMCI Not Busy
This flag must be used only for Write Operations.
A block write operation uses a simple busy signalling of the write operation duration on the data (DAT0) line: during a data
transfer block, if the card does not have a free data receive buffer, the card indicates this condition by pulling down the data
line (DAT0) to LOW. The card stops pulling down the data line as soon as at least one receive buffer for the defined data
transfer block length becomes free.

854

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The NOTBUSY flag allows to deal with these different states.
0 = The HSMCI is not ready for new data transfer. Cleared at the end of the card response.
1 = The HSMCI is ready for new data transfer. Set when the busy state on the data line has ended. This corresponds to a
free internal data receive buffer of the card.
Refer to the MMC or SD Specification for more details concerning the busy behavior.
For all the read operations, the NOTBUSY flag is cleared at the end of the host command.
For the Infinite Read Multiple Blocks, the NOTBUSY flag is set at the end of the STOP_TRANSMISSION host command
(CMD12).
For the Single Block Reads, the NOTBUSY flag is set at the end of the data read block.
For the Multiple Block Reads with pre-defined block count, the NOTBUSY flag is set at the end of the last received data
block.
• SDIOIRQA: SDIO Interrupt for Slot A
0 = No interrupt detected on SDIO Slot A.
1 = An SDIO Interrupt on Slot A occurred. Cleared when reading the HSMCI_SR.
• SDIOWAIT: SDIO Read Wait Operation Status
0 = Normal Bus operation.
1 = The data bus has entered IO wait state.
• CSRCV: CE-ATA Completion Signal Received
0 = No completion signal received since last status read operation.
1 = The device has issued a command completion signal on the command line. Cleared by reading in the HSMCI_SR
register.
• RINDE: Response Index Error
0 = No error.
1 = A mismatch is detected between the command index sent and the response index received. Cleared when writing in
the HSMCI_CMDR.
• RDIRE: Response Direction Error
0 = No error.
1 = The direction bit from card to host in the response has not been detected.
• RCRCE: Response CRC Error
0 = No error.
1 = A CRC7 error has been detected in the response. Cleared when writing in the HSMCI_CMDR.
• RENDE: Response End Bit Error
0 = No error.
1 = The end bit of the response has not been detected. Cleared when writing in the HSMCI_CMDR.

855
6430F–ATARM–21-Feb-12

• RTOE: Response Time-out Error
0 = No error.
1 = The response time-out set by MAXLAT in the HSMCI_CMDR has been exceeded. Cleared when writing in the
HSMCI_CMDR.
• DCRCE: Data CRC Error
0 = No error.
1 = A CRC16 error has been detected in the last data block. Cleared by reading in the HSMCI_SR register.
• DTOE: Data Time-out Error
0 = No error.
1 = The data time-out set by DTOCYC and DTOMUL in HSMCI_DTOR has been exceeded. Cleared by reading in the
HSMCI_SR register.
• CSTOE: Completion Signal Time-out Error
0 = No error.
1 = The completion signal time-out set by CSTOCYC and CSTOMUL in HSMCI_CSTOR has been exceeded. Cleared by
reading in the HSMCI_SR register. Cleared by reading in the HSMCI_SR register.
• BLKOVRE: DMA Block Overrun Error
0 = No error.
1 = A new block of data is received and the DMA controller has not started to move the current pending block, a block overrun is raised. Cleared by reading in the HSMCI_SR register.
• DMADONE: DMA Transfer done
0 = DMA buffer transfer has not completed since the last read of HSMCI_SR register.
1 = DMA buffer transfer has completed.
• FIFOEMPTY: FIFO empty flag
0 = FIFO contains at least one byte.
1 = FIFO is empty.
• XFRDONE: Transfer Done flag
0 = A transfer is in progress.
1 = Command register is ready to operate and the data bus is in the idle state.
• ACKRCV: Boot Operation Acknowledge Received
0 = No Boot acknowledge received since the last read of the status register.
1 = A Boot acknowledge signal has been received. Cleared by reading the HSMCI_SR register.
• ACKRCVE: Boot Operation Acknowledge Error
0 = No error
1 = Corrupted Boot Acknowledge signal received.

856

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• OVRE: Overrun
0 = No error.
1 = At least one 8-bit received data has been lost (not read). Cleared when sending a new data transfer command.
When FERRCTRL in HSMCI_CFG is set to 1, OVRE becomes reset after read.
• UNRE: Underrun
0 = No error.
1 = At least one 8-bit data has been sent without valid information (not written). Cleared when sending a new data transfer
command or when setting FERRCTRL in HSMCI_CFG to 1.
When FERRCTRL in HSMCI_CFG is set to 1, UNRE becomes reset after read.

857
6430F–ATARM–21-Feb-12

37.14.13 HSMCI Interrupt Enable Register
Name:
HSMCI_IER
Address:

0x40000044

Access:

Write-only

31
UNRE

30
OVRE

29
ACKRCVE

28
ACKRCV

27
XFRDONE

26
FIFOEMPTY

25
DMADONE

24
BLKOVRE

23
CSTOE

22
DTOE

21
DCRCE

20
RTOE

19
RENDE

18
RCRCE

17
RDIRE

16
RINDE

–

CSRCV

SDIOWAIT

–

8
MCI_SDIOIR
QA

7
–

6
–

5
NOTBUSY

4
DTIP

3
BLKE

2
TXRDY

1
RXRDY

0
CMDRDY

• CMDRDY: Command Ready Interrupt Enable
• RXRDY: Receiver Ready Interrupt Enable
• TXRDY: Transmit Ready Interrupt Enable
• BLKE: Data Block Ended Interrupt Enable
• DTIP: Data Transfer in Progress Interrupt Enable
• NOTBUSY: Data Not Busy Interrupt Enable
• SDIOIRQA: SDIO Interrupt for Slot A Interrupt Enable
• SDIOIRQD: SDIO Interrupt for Slot D Interrupt Enable
• SDIOWAIT: SDIO Read Wait Operation Status Interrupt Enable
• CSRCV: Completion Signal Received Interrupt Enable
• RINDE: Response Index Error Interrupt Enable
• RDIRE: Response Direction Error Interrupt Enable
• RCRCE: Response CRC Error Interrupt Enable
• RENDE: Response End Bit Error Interrupt Enable
• RTOE: Response Time-out Error Interrupt Enable
• DCRCE: Data CRC Error Interrupt Enable
• DTOE: Data Time-out Error Interrupt Enable
• CSTOE: Completion Signal Timeout Error Interrupt Enable

858

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• BLKOVRE: DMA Block Overrun Error Interrupt Enable
• DMADONE: DMA Transfer completed Interrupt Enable
• FIFOEMPTY: FIFO empty Interrupt enable
• XFRDONE: Transfer Done Interrupt enable
• ACKRCV: Boot Acknowledge Interrupt Enable
• ACKRCVE: Boot Acknowledge Error Interrupt Enable
• OVRE: Overrun Interrupt Enable
• UNRE: Underrun Interrupt Enable
0 = No effect.
1 = Enables the corresponding interrupt.

859
6430F–ATARM–21-Feb-12

37.14.14 HSMCI Interrupt Disable Register
Name:
HSMCI_IDR
Address:

0x40000048

Access:

Write-only

31
UNRE

30
OVRE

29
ACKRCVE

28
ACKRCV

27
XFRDONE

26
FIFOEMPTY

25
DMADONE

24
BLKOVRE

23
CSTOE

22
DTOE

21
DCRCE

20
RTOE

19
RENDE

18
RCRCE

17
RDIRE

16
RINDE

–

CSRCV

SDIOWAIT

–

8
MCI_SDIOIR
QA

7
–

6
–

5
NOTBUSY

4
DTIP

3
BLKE

2
TXRDY

1
RXRDY

0
CMDRDY

• CMDRDY: Command Ready Interrupt Disable
• RXRDY: Receiver Ready Interrupt Disable
• TXRDY: Transmit Ready Interrupt Disable
• BLKE: Data Block Ended Interrupt Disable
• DTIP: Data Transfer in Progress Interrupt Disable
• NOTBUSY: Data Not Busy Interrupt Disable
• SDIOIRQA: SDIO Interrupt for Slot A Interrupt Disable
• SDIOWAIT: SDIO Read Wait Operation Status Interrupt Disable
• CSRCV: Completion Signal received interrupt Disable
• RINDE: Response Index Error Interrupt Disable
• RDIRE: Response Direction Error Interrupt Disable
• RCRCE: Response CRC Error Interrupt Disable
• RENDE: Response End Bit Error Interrupt Disable
• RTOE: Response Time-out Error Interrupt Disable
• DCRCE: Data CRC Error Interrupt Disable
• DTOE: Data Time-out Error Interrupt Disable
• CSTOE: Completion Signal Time out Error Interrupt Disable
• BLKOVRE: DMA Block Overrun Error Interrupt Disable

860

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• DMADONE: DMA Transfer completed Interrupt Disable
• FIFOEMPTY: FIFO empty Interrupt Disable
• XFRDONE: Transfer Done Interrupt Disable
• ACKRCV: Boot Acknowledge Interrupt Disable
• ACKRCVE: Boot Acknowledge Error Interrupt Disable
• OVRE: Overrun Interrupt Disable
• UNRE: Underrun Interrupt Disable
0 = No effect.
1 = Disables the corresponding interrupt.

861
6430F–ATARM–21-Feb-12

37.14.15 HSMCI Interrupt Mask Register
Name:
HSMCI_IMR
Address:

0x4000004C

Access:

Read-only

31
UNRE

30
OVRE

29
ACKRCVE

28
ACKRCV

27
XFRDONE

26
FIFOEMPTY

25
DMADONE

24
BLKOVRE

23
CSTOE

22
DTOE

21
DCRCE

20
RTOE

19
RENDE

18
RCRCE

17
RDIRE

16
RINDE

–

CSRCV

SDIOWAIT

–

8
MCI_SDIOIR
QA

7
–

6
–

5
NOTBUSY

4
DTIP

3
BLKE

2
TXRDY

1
RXRDY

0
CMDRDY

• CMDRDY: Command Ready Interrupt Mask
• RXRDY: Receiver Ready Interrupt Mask
• TXRDY: Transmit Ready Interrupt Mask
• BLKE: Data Block Ended Interrupt Mask
• DTIP: Data Transfer in Progress Interrupt Mask
• NOTBUSY: Data Not Busy Interrupt Mask
• SDIOIRQA: SDIO Interrupt for Slot A Interrupt Mask
• SDIOWAIT: SDIO Read Wait Operation Status Interrupt Mask
• CSRCV: Completion Signal Received Interrupt Mask
• RINDE: Response Index Error Interrupt Mask
• RDIRE: Response Direction Error Interrupt Mask
• RCRCE: Response CRC Error Interrupt Mask
• RENDE: Response End Bit Error Interrupt Mask
• RTOE: Response Time-out Error Interrupt Mask
• DCRCE: Data CRC Error Interrupt Mask
• DTOE: Data Time-out Error Interrupt Mask
• CSTOE: Completion Signal Time-out Error Interrupt Mask
• BLKOVRE: DMA Block Overrun Error Interrupt Mask

862

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• DMADONE: DMA Transfer Completed Interrupt Mask
• FIFOEMPTY: FIFO Empty Interrupt Mask
• XFRDONE: Transfer Done Interrupt Mask
• ACKRCV: Boot Operation Acknowledge Received Interrupt Mask
• ACKRCVE: Boot Operation Acknowledge Error Interrupt Mask
• OVRE: Overrun Interrupt Mask
• UNRE: Underrun Interrupt Mask
0 = The corresponding interrupt is not enabled.
1 = The corresponding interrupt is enabled.

863
6430F–ATARM–21-Feb-12

37.14.16 HSMCI DMA Configuration Register
Name:
HSMCI_DMA
Address:

0x40000050

Access:

Read-write

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
ROPT

11
–

10
–

9
–

8
DMAEN

7
–

6
–

5
–

4
CHKSIZE

3
–

2
–

0
OFFSET

This register can only be written if the WPEN bit is cleared in “HSMCI Write Protect Mode Register” on page 866.
• OFFSET: DMA Write Buffer Offset
This field indicates the number of discarded bytes when the DMA writes the first word of the transfer.
• CHKSIZE: DMA Channel Read and Write Chunk Size
The CHKSIZE field indicates the number of data available when the DMA chunk transfer request is asserted.
Value

Name

Description

1 data available

4 data available

• DMAEN: DMA Hardware Handshaking Enable
0 = DMA interface is disabled.
1 = DMA Interface is enabled.
Note:

To avoid unpredictable behavior, DMA hardware handshaking must be disabled when CPU transfers are performed.

• ROPT: Read Optimization with padding
0: BLKLEN bytes are moved from the Memory Card to the system memory, two DMA descriptors are used when the transfer size is not a multiple of 4.
1: Ceiling(BLKLEN/4) * 4 bytes are moved from the Memory Card to the system memory, only one DMA descriptor is used.

864

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.14.17 HSMCI Configuration Register
Name:
HSMCI_CFG
Address:

0x40000054

Access:

Read-write

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
LSYNC

11
–

10
–

9
–

8
HSMODE

7
–

6
–

5
–

4
FERRCTRL

3
–

2
–

1
–

0
FIFOMODE

This register can only be written if the WPEN bit is cleared in “HSMCI Write Protect Mode Register” on page 866.
• FIFOMODE: HSMCI Internal FIFO control mode
0 = A write transfer starts when a sufficient amount of data is written into the FIFO.
When the block length is greater than or equal to 3/4 of the HSMCI internal FIFO size, then the write transfer starts as soon
as half the FIFO is filled. When the block length is greater than or equal to half the internal FIFO size, then the write transfer
starts as soon as one quarter of the FIFO is filled. In other cases, the transfer starts as soon as the total amount of data is
written in the internal FIFO.
1 = A write transfer starts as soon as one data is written into the FIFO.
• FERRCTRL: Flow Error flag reset control mode
0= When an underflow/overflow condition flag is set, a new Write/Read command is needed to reset the flag.
1= When an underflow/overflow condition flag is set, a read status resets the flag.
• HSMODE: High Speed Mode
0= Default bus timing mode.
1= If set to one, the host controller outputs command line and data lines on the rising edge of the card clock. The Host
driver shall check the high speed support in the card registers.
• LSYNC: Synchronize on the last block
0= The pending command is sent at the end of the current data block.
1= The pending command is sent at the end of the block transfer when the transfer length is not infinite. (block count shall
be different from zero)

865
6430F–ATARM–21-Feb-12

37.14.18 HSMCI Write Protect Mode Register
Name:
HSMCI_WPMR
Address:

0x400000E4

Access:

Read-write

28
27
WP_KEY (0x4D => “M”)

20
19
WP_KEY (0x43 => C”)

12
11
WP_KEY (0x49 => “I”)

0
WP_EN

• WP_EN: Write Protection Enable
0 = Disables the Write Protection if WP_KEY corresponds to 0x4D4349 (“MCI’ in ASCII).
1 = Enables the Write Protection if WP_KEY corresponds to 0x4D4349 (“MCI’ in ASCII).
• WP_KEY: Write Protection Key password
Should be written at value 0x4D4349 (ASCII code for “MCI”). Writing any other value in this field has no effect.
Protects the registers:
• “HSMCI Mode Register” on page 841
• “HSMCI Data Timeout Register” on page 843
• “HSMCI SDCard/SDIO Register” on page 844
• “HSMCI Completion Signal Timeout Register” on page 850
• “HSMCI DMA Configuration Register” on page 864
• “HSMCI Configuration Register” on page 865

866

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.14.19 HSMCI Write Protect Status Register
Name:
HSMCI_WPSR
Address:

0x400000E8

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

WP_VSRC
15

12
WP_VSRC

7
–

6
–

5
–

4
–

WP_VS

• WP_VS: Write Protection Violation Status
Value

Name

Description

NONE

No Write Protection Violation occurred since the last read of this
register (WP_SR)

WRITE

Write Protection detected unauthorized attempt to write a control
register had occurred (since the last read.)

RESET

Software reset had been performed while Write Protection was
enabled (since the last read).

BOTH

Both Write Protection violation and software reset with Write
Protection enabled have occurred since the last read.

• WP_VSRC: Write Protection Violation SouRCe
When WPVS is active, this field indicates the write-protected register (through address offset or code) in which a write
access has been attempted.

867
6430F–ATARM–21-Feb-12

37.14.20 HSMCI FIFOx Memory Aperture
Name:
HSMCI_FIFOx[x=0..255]
Address:

0x40000200

Access:

Read-write

DATA
23

20
DATA

12
DATA

4
DATA

• DATA: Data to Read or Data to Write

868

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38. Pulse Width Modulation (PWM)
38.1

Description
The PWM macrocell controls 4 channels independently. Each channel controls two complementary square output waveforms. Characteristics of the output waveforms such as period, dutycycle, polarity and dead-times (also called dead-bands or non-overlapping times) are configured
through the user interface. Each channel selects and uses one of the clocks provided by the
clock generator. The clock generator provides several clocks resulting from the division of the
PWM master clock (MCK).
All PWM macrocell accesses are made through registers mapped on the peripheral bus. All
channels integrate a double buffering system in order to prevent an unexpected output waveform while modifying the period, the duty-cycle or the dead-times.
Channels can be linked together as synchronous channels to be able to update their duty-cycle
or dead-times at the same time.
The update of duty-cycles of synchronous channels can be performed by the Peripheral DMA
Controller Channel (PDC) which offers buffer transfer without processor Intervention.
The PWM macrocell provides 8 independent comparison units capable of comparing a programmed value to the counter of the synchronous channels (counter of channel 0). These
comparisons are intended to generate software interrupts, to trigger pulses on the 2 independent event lines (in order to synchronize ADC conversions with a lot of flexibility independently of
the PWM outputs), and to trigger PDC transfer requests.
The PWM outputs can be overridden synchronously or asynchronously to their channel counter.
The PWM block provides a fault protection mechanism with 4 fault inputs, capable of detecting a
fault condition and to override the PWM outputs asynchronously.
For safety usage, some control registers are write-protected.

38.2

Embedded Characteristics
• 4 Channels
• Common Clock Generator Providing Thirteen Different Clocks
– A Modulo n Counter Providing Eleven Clocks
– Two Independent Linear Dividers Working on Modulo n Counter Outputs
• Independent Channels
– Independent 16-bit Counter for Each Channel
– Independent Complementary Outputs with 12-bit Dead-Time Generator (Also Called
Dead-Band or Non-Overlapping Time) for Each Channel
– Independent Enable Disable Command for Each Channel
– Independent Clock Selection for Each Channel
– Independent Period, Duty-Cycle and Dead-Time for Each Channel
– Independent Double Buffering of Period, Duty-Cycle and Dead-Times for Each
Channel
– Independent Programmable Selection of The Output Waveform Polarity for Each
Channel

869
6430F–ATARM–21-Feb-12

– Independent Programmable Center or Left Aligned Output Waveform for Each
Channel
– Independent Output Override for Each Channel
• Synchronous Channel Mode
– Synchronous Channels Share the Same Counter
– Mode to Update the Synchronous Channels Registers after a Programmable
Number of Periods
– Synchronous Channels Supports Connection of one Peripheral DMA Controller
Channel (PDC) Which Offers Buffer Transfer Without Processor Intervention To
Update Duty-Cycle Registers
• 2 Independent Events Lines Intended to Synchronize ADC Conversions
• 8 Comparison Units Intended to Generate Interrupts, Pulses on Event Lines and PDC
Transfer Requests
• 4 Programmable Fault Inputs Providing an Asynchronous Protection of PWM Outputs
– User Driven through PIO inputs
– PMC Driven when Crystal Oscillator Clock Fails
– ADC Controller Driven through Configurable Comparison Function
• Write-Protect Registers

870

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.3

Block Diagram

Figure 38-1. Pulse Width Modulation Controller Block Diagram
PWM Controller
Channel x
Update

Period

Duty-Cycle
MUX
Counter
Channel x

Clock
Selector

OCx

DTOHx
Dead-Time
Generator DTOLx

OOOHx
Output
Override

OOOLx

PWMHx
Fault
Protection PWMLx

PWMHx
PWMLx

SYNCx

Comparator

PIO

Channel 0
Update

Period
Comparator

OC0

DTOH0
Dead-Time
Generator DTOL0

OOOH0
Output
Override

OOOL0

PWMH0
Fault
Protection PWML0

PWMH0
PWML0

Duty-Cycle

Counter
Channel 0

Clock
Selector
PWMFIx
PIO
PWMFI0

event line 0
event line 1
Comparison
Units

Events
Generator

ADC
event line x

PMC

MCK

CLOCK
Generator

APB
Interface
Interrupt Generator

Interrupt
Controller

PDC

APB

38.4

I/O Lines Description
Each channel outputs two complementary external I/O lines.
Table 38-1.

I/O Line Description

Name

Description

Type

PWMHx

PWM Waveform Output High for channel x

Output

PWMLx

PWM Waveform Output Low for channel x

Output

PWMFIx

PWM Fault Input x

Input

871
6430F–ATARM–21-Feb-12

38.5
38.5.1

Product Dependencies
I/O Lines
The pins used for interfacing the PWM are multiplexed with PIO lines. The programmer must first
program the PIO controller to assign the desired PWM pins to their peripheral function. If I/O
lines of the PWM are not used by the application, they can be used for other purposes by the PIO
controller.
All of the PWM outputs may or may not be enabled. If an application requires only four channels,
then only four PIO lines will be assigned to PWM outputs.
Table 38-2.

872

I/O Lines

Instance

Signal

I/O Line

Peripheral

PWM

PWMFI0

PA11

PWM

PWMFI1

PA12

PWM

PWMFI2

PA18

PWM

PWMH0

PA4

PWM

PWMH0

PA28

PWM

PWMH0

PB0

PWM

PWMH0

PB13

PWM

PWMH0

PC24

PWM

PWMH1

PA5

PWM

PWMH1

PA29

PWM

PWMH1

PB1

PWM

PWMH1

PB14

PWM

PWMH1

PC25

PWM

PWMH2

PA6

PWM

PWMH2

PA15

PWM

PWMH2

PB2

PWM

PWMH2

PB15

PWM

PWMH2

PC26

PWM

PWMH3

PA20

PWM

PWMH3

PB3

PWM

PWMH3

PB16

PWM

PWMH3

PC27

PWM

PWML0

PA7

PWM

PWML0

PB17

PWM

PWML0

PB25

PWM

PWML0

PC6

PWM

PWML0

PC29

PWM

PWML1

PA8

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 38-2.

38.5.2

I/O Lines

PWM

PWML1

PB18

PWM

PWML1

PB26

PWM

PWML1

PC7

PWM

PWML1

PC30

PWM

PWML2

PA9

PWM

PWML2

PB19

PWM

PWML2

PB27

PWM

PWML2

PC8

PWM

PWML2

PC31

PWM

PWML3

PA10

PWM

PWML3

PB20

PWM

PWML3

PB28

PWM

PWML3

PC9

PWM

PWML3

PC16

Power Management
The PWM is not continuously clocked. The programmer must first enable the PWM clock in the
Power Management Controller (PMC) before using the PWM. However, if the application does
not require PWM operations, the PWM clock can be stopped when not needed and be restarted
later. In this case, the PWM will resume its operations where it left off.
In the PWM description, Master Clock (MCK) is the clock of the peripheral bus to which the PWM
is connected.

38.5.3

Interrupt Sources
The PWM interrupt line is connected on one of the internal sources of the Interrupt Controller.
Using the PWM interrupt requires the Interrupt Controller to be programmed first. Note that it is
not recommended to use the PWM interrupt line in edge sensitive mode.

38.5.4

Fault Inputs
The PWM has the FAULT inputs connected to the different modules. Please refer to the implementation of these module within the product for detailed information about the fault generation
procedure. The PWM receives faults from PIO inputs, PMC, ADC controller,

Table 38-3.

Fault Inputs

External PWM Fault Input Number

Polarity Level(1)

Fault Input ID

Main OSC

–

PA18

PWMFI2

User Defined

PA11

PWMFI0

User Defined

PWMFI1

User Defined

PA12
Note:

1. FPOL bit in PWMC_FMR.

873
6430F–ATARM–21-Feb-12

38.6

Functional Description
The PWM macrocell is primarily composed of a clock generator module and 4 channels.
• Clocked by the master clock (MCK), the clock generator module provides 13 clocks.
• Each channel can independently choose one of the clock generator outputs.
• Each channel generates an output waveform with attributes that can be defined
independently for each channel through the user interface registers.

38.6.1

PWM Clock Generator
Figure 38-2. Functional View of the Clock Generator Block Diagram
MCK

modulo n counter
MCK
MCK/2
MCK/4
MCK/8
MCK/16
MCK/32
MCK/64
MCK/128
MCK/256
MCK/512
MCK/1024

Divider A

PREA

clkA

DIVA

PWM_MR

Divider B

PREB

clkB

DIVB

PWM_MR

The PWM master clock (MCK) is divided in the clock generator module to provide different
clocks available for all channels. Each channel can independently select one of the divided
clocks.
The clock generator is divided in three blocks:
– a modulo n counter which provides 11 clocks: FMCK, FMCK/2, FMCK/4, FMCK/8,
FMCK/16, FMCK/32, FMCK/64, FMCK/128, FMCK/256, FMCK/512, FMCK/1024
– two linear dividers (1, 1/2, 1/3, ... 1/255) that provide two separate clocks: clkA and
clkB

874

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Each linear divider can independently divide one of the clocks of the modulo n counter. The
selection of the clock to be divided is made according to the PREA (PREB) field of the PWM
Clock register (PWM_CLK). The resulting clock clkA (clkB) is the clock selected divided by DIVA
(DIVB) field value.
After a reset of the PWM controller, DIVA (DIVB) and PREA (PREB) are set to 0. This implies
that after reset clkA (clkB) are turned off.
At reset, all clocks provided by the modulo n counter are turned off except clock ”MCK”. This situation is also true when the PWM master clock is turned off through the Power Management
Controller.

CAUTION:
• Before using the PWM macrocell, the programmer must first enable the PWM clock in the
Power Management Controller (PMC).

875
6430F–ATARM–21-Feb-12

38.6.2

PWM Channel

38.6.2.1

Channel Block Diagram

Figure 38-3. Functional View of the Channel Block Diagram
Channel x
Update

Period
Comparator

OCx

DTOHx
Dead-Time
Generator DTOLx

OOOHx
Output
Override

OOOLx

PWMHx
Fault
Protection PWMLx

Duty-Cycle
MUX
from
Clock
Generator
from APB
Peripheral Bus

Clock
Selector

SYNCx

Counter
Channel x

Counter
Channel 0

Each of the 4 channels is composed of six blocks:
• A clock selector which selects one of the clocks provided by the clock generator (described in
Section 38.6.1 on page 874).
• A counter clocked by the output of the clock selector. This counter is incremented or
decremented according to the channel configuration and comparators matches. The size of
the counter is 16 bits.
• A comparator used to compute the OCx output waveform according to the counter value and
the configuration. The counter value can be the one of the channel counter or the one of the
channel 0 counter according to SYNCx bit in the “PWM Sync Channels Mode Register” on
page 913 (PWM_SCM).
• A dead-time generator providing two complementary outputs (DTOHx/DTOLx) which allows
to drive external power control switches safely.
• An output override block that can force the two complementary outputs to a programmed
value (OOOHx/OOOLx).
• An asynchronous fault protection mechanism that has the highest priority to override the two
complementary outputs in case of fault detection (PWMHx/PWMLx).

876

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.6.2.2

Comparator
The comparator continuously compares its counter value with the channel period defined by
CPRD in the “PWM Channel Period Register” on page 945 (PWM_CPRDx) and the duty-cycle
defined by CDTY in the “PWM Channel Duty Cycle Register” on page 943 (PWM_CDTYx) to
generate an output signal OCx accordingly.
The different properties of the waveform of the output OCx are:
• the clock selection. The channel counter is clocked by one of the clocks provided by the
clock generator described in the previous section. This channel parameter is defined in the
CPRE field of the “PWM Channel Mode Register” on page 941 (PWM_CMRx). This field is
reset at 0.
• the waveform period. This channel parameter is defined in the CPRD field of the
PWM_CPRDx register.
If the waveform is left aligned, then the output waveform period depends on the counter
source clock and can be calculated:
By using the PWM master clock (MCK) divided by an X given prescaler value (with X being 1,
2, 4, 8, 16, 32, 64, 128, 256, 512, or 1024), the resulting period formula will be:
(------------------------------X × CPRD )MCK

By using the PWM master clock (MCK) divided by one of both DIVA or DIVB divider, the
formula becomes, respectively:
(-----------------------------------------CRPD × DIVA )( CRPD × DIVB )
or ------------------------------------------MCK
MCK

If the waveform is center aligned then the output waveform period depends on the counter
source clock and can be calculated:
By using the PWM master clock (MCK) divided by an X given prescaler value
(with X being 1, 2, 4, 8, 16, 32, 64, 128, 256, 512, or 1024). The resulting period formula will
be:
(-----------------------------------------2 × X × CPRD )MCK

By using the PWM master clock (MCK) divided by one of both DIVA or DIVB divider, the
formula becomes, respectively:
(----------------------------------------------------2 × CPRD × DIVA )( 2 × CPRD × DIVB )
or -----------------------------------------------------MCK
MCK

• the waveform duty-cycle. This channel parameter is defined in the CDTY field of the
PWM_CDTYx register.
If the waveform is left aligned then:
period – 1 ⁄ fchannel_x_clock × CDTY )duty cycle = (----------------------------------------------------------------------------------------------------------period
If the waveform is center aligned, then:
( ( period ⁄ 2 ) – 1 ⁄ fchannel_x_clock × CDTY ) )
duty cycle = -----------------------------------------------------------------------------------------------------------------------------( period ⁄ 2 )

877
6430F–ATARM–21-Feb-12

• the waveform polarity. At the beginning of the period, the signal can be at high or low level.
This property is defined in the CPOL field of the PWM_CMRx register. By default the signal
starts by a low level.
• the waveform alignment. The output waveform can be left or center aligned. Center aligned
waveforms can be used to generate non overlapped waveforms. This property is defined in
the CALG field of the PWM_CMRx register. The default mode is left aligned.
Figure 38-4. Non Overlapped Center Aligned Waveforms
No overlap

OC0

OC1

Period

Note:

1. See Figure 38-5 on page 879 for a detailed description of center aligned waveforms.

When center aligned, the channel counter increases up to CPRD and decreases down to 0. This
ends the period.
When left aligned, the channel counter increases up to CPRD and is reset. This ends the period.
Thus, for the same CPRD value, the period for a center aligned channel is twice the period for a
left aligned channel.
Waveforms are fixed at 0 when:
• CDTY = CPRD and CPOL = 0
• CDTY = 0 and CPOL = 1

Waveforms are fixed at 1 (once the channel is enabled) when:
• CDTY = 0 and CPOL = 0
• CDTY = CPRD and CPOL = 1

The waveform polarity must be set before enabling the channel. This immediately affects the
channel output level. Changes on channel polarity are not taken into account while the channel
is enabled.
Besides generating output signals OCx, the comparator generates interrupts in function of the
counter value. When the output waveform is left aligned, the interrupt occurs at the end of the
counter period. When the output waveform is center aligned, the bit CES of the PWM_CMRx
register defines when the channel counter interrupt occurs. If CES is set to 0, the interrupt
occurs at the end of the counter period. If CES is set to 1, the interrupt occurs at the end of the
counter period and at half of the counter period.
Figure 38-5 “Waveform Properties” illustrates the counter interrupts in function of the
configuration.

878

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 38-5. Waveform Properties
Channel x
slected clock
CHIDx(PWM_SR)

CHIDx(PWM_ENA)
CHIDx(PWM_DIS)
Center Aligned
CALG(PWM_CMRx) = 1

PWM_CCNTx
CPRD(PWM_CPRDx)
CDTY(PWM_CDTYx)

Period
Output Waveform OCx
CPOL(PWM_CMRx) = 0

Output Waveform OCx
CPOL(PWM_CMRx) = 1
Counter Event
CHIDx(PWM_ISR)
CES(PWM_CMRx) = 0
Counter Event
CHIDx(PWM_ISR)
CES(PWM_CMRx) = 1
Left Aligned
CALG(PWM_CMRx) = 0

PWM_CCNTx
CPRD(PWM_CPRDx)
CDTY(PWM_CDTYx)

Period
Output Waveform OCx
CPOL(PWM_CMRx) = 0

Output Waveform OCx
CPOL(PWM_CMRx) = 1

Counter Event
CHIDx(PWM_ISR)

879
6430F–ATARM–21-Feb-12

38.6.2.3

Dead-Time Generator
The dead-time generator uses the comparator output OCx to provide the two complementary
outputs DTOHx and DTOLx, which allows the PWM macrocell to drive external power control
switches safely. When the dead-time generator is enabled by setting the bit DTE to 1 or 0 in the
“PWM Channel Mode Register” (PWM_CMRx), dead-times (also called dead-bands or nonoverlapping times) are inserted between the edges of the two complementary outputs DTOHx
and DTOLx. Note that enabling or disabling the dead-time generator is allowed only if the channel is disabled.
The dead-time is adjustable by the “PWM Channel Dead Time Register” (PWM_DTx). Both outputs of the dead-time generator can be adjusted separately by DTH and DTL. The dead-time
values can be updated synchronously to the PWM period by using the “PWM Channel Dead
Time Update Register” (PWM_DTUPDx).
The dead-time is based on a specific counter which uses the same selected clock that feeds the
channel counter of the comparator. Depending on the edge and the configuration of the deadtime, DTOHx and DTOLx are delayed until the counter has reached the value defined by DTH or
DTL. An inverted configuration bit (DTHI and DTLI bit in the PWM_CMRx register) is provided
for each output to invert the dead-time outputs. The following figure shows the waveform of the
dead-time generator.
Figure 38-6. Complementary Output Waveforms
output waveform OCx
CPOLx = 0
output waveform DTOHx
DTHIx = 0
output waveform DTOLx
DTLIx = 0
output waveform DTOHx
DTHIx = 1
output waveform DTOLx
DTLIx = 1
DTHx

DTLx

DTHx

DTLx

output waveform OCx
CPOLx = 1
output waveform DTOHx
DTHIx = 0
output waveform DTOLx
DTLIx = 0
output waveform DTOHx
DTHIx = 1
output waveform DTOLx
DTLIx = 1

880

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.6.2.4

Output Override
The two complementary outputs DTOHx and DTOLx of the dead-time generator can be forced
to a value defined by the software.
Figure 38-7. Override Output Selection
DTOHx

0
OOOHx

OOVHx

OSHx
DTOLx

0
OOOLx

OOVLx

OSLx

The fields OSHx and OSLx in the “PWM Output Selection Register” (PWM_OS) allow the outputs of the dead-time generator DTOHx and DTOLx to be overridden by the value defined in the
fields OOVHx and OOVLx in the“PWM Output Override Value Register” (PWM_OOV).
The set registers “PWM Output Selection Set Register” and “PWM Output Selection Set Update
Register” (PWM_OSS and PWM_OSSUPD) enable the override of the outputs of a channel
regardless of other channels. In the same way, the clear registers “PWM Output Selection Clear
Register” and “PWM Output Selection Clear Update Register” (PWM_OSC and
PWM_OSCUPD) disable the override of the outputs of a channel regardless of other channels.
By using buffer registers PWM_OSSUPD and PWM_OSCUPD, the output selection of PWM
outputs is done synchronously to the channel counter, at the beginning of the next PWM period.
By using registers PWM_OSS and PWM_OSC, the output selection of PWM outputs is done
asynchronously to the channel counter, as soon as the register is written.
The value of the current output selection can be read in PWM_OS.
While overriding PWM outputs, the channel counters continue to run, only the PWM outputs are
forced to user defined values.

881
6430F–ATARM–21-Feb-12

38.6.2.5

Fault Protection
4 inputs provide fault protection which can force any of the PWM output pair to a programmable
value. This mechanism has priority over output overriding.

Figure 38-8. Fault Protection
0
fault input 0

Glitch
Filter

FIV0

0
FMOD0

SET

OUT

Fault 0 Status
FS0
FPEx[0]

CLR

FPE0[0]
FFIL0

0
fault input 1

Glitch
Filter

FIV1

Write FCLR0 at 1

FPOL0

FMOD0

0
FMOD1

SET

OUT

Fault 1 Status
FS1
FPEx[1]
FPE0[1]

FPOL1

Write FCLR1 at 1

0
1

SYNCx
from fault 1

CLR

FFIL1

from fault 0

FMOD1

From Output
Override
OOHx

FPVHx

PWMHx

0
Fault protection
on PWM
channel x

1
from fault y
SYNCx

fault input y
FPVLx

OOLx
From Output
Override

PWMLx

The polarity level of the fault inputs is configured by the FPOL field in the “PWM Fault Mode
Register” (PWM_FMR). For fault inputs coming from internal peripherals such as ADC, Timer
Counter, to name but a few, the polarity level must be FPOL = 1. For fault inputs coming from
external GPIO pins the polarity level depends on the user's implementation.
The configuration of the Fault Activation Mode (FMOD bit in PWMC_FMR) depends on the
peripheral generating the fault. If the corresponding peripheral does not have “Fault Clear” management, then the FMOD configuration to use must be FMOD = 1, to avoid spurious fault
detection. Check the corresponding peripheral documentation for details on handling fault
generation.
The fault inputs can be glitch filtered or not in function of the FFIL field in the PWM_FMR register. When the filter is activated, glitches on fault inputs with a width inferior to the PWM master
clock (MCK) period are rejected.
A fault becomes active as soon as its corresponding fault input has a transition to the programmed polarity level. If the corresponding bit FMOD is set to 0 in the PWM_FMR register, the
fault remains active as long as the fault input is at this polarity level. If the corresponding FMOD
bit is set to 1, the fault remains active until the fault input is not at this polarity level anymore and
until it is cleared by writing the corresponding bit FCLR in the “PWM Fault Clear Register”
(PWM_FSCR). By reading the “PWM Fault Status Register” (PWM_FSR), the user can read the
current level of the fault inputs by means of the field FIV, and can know which fault is currently
active thanks to the FS field.
Each fault can be taken into account or not by the fault protection mechanism in each channel.
To be taken into account in the channel x, the fault y must be enabled by the bit FPEx[y] in the
“PWM Fault Protection Enable Registers” (PWM_FPE1). However the synchronous channels
(see Section 38.6.2.6 “Synchronous Channels”) do not use their own fault enable bits, but those
of the channel 0 (bits FPE0[y]).
The fault protection on a channel is triggered when this channel is enabled and when any one of
the faults that are enabled for this channel is active. It can be triggered even if the PWM master
clock (MCK) is not running but only by a fault input that is not glitch filtered.

882

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
When the fault protection is triggered on a channel, the fault protection mechanism forces the
channel outputs to the values defined by the fields FPVHx and FPVLx in the “PWM Fault Protection Value Register” (PWM_FPV) and leads to a reset of the counter of this channel. The output
forcing is made asynchronously to the channel counter.

CAUTION:
• To prevent an unexpected activation of the status flag FSy in the PWM_FSR register, the
FMODy bit can be set to “1” only if the FPOLy bit has been previously configured to its final
value.
• To prevent an unexpected activation of the Fault Protection on the channel x, the bit FPEx[y]
can be set to “1” only if the FPOLy bit has been previously configured to its final value.

If a comparison unit is enabled (see Section 38.6.3 “PWM Comparison Units”) and if a fault is
triggered in the channel 0, in this case the comparison cannot match.
As soon as the fault protection is triggered on a channel, an interrupt (different from the interrupt
generated at the end of the PWM period) can be generated but only if it is enabled and not
masked. The interrupt is reset by reading the interrupt status register, even if the fault which has
caused the trigger of the fault protection is kept active.

883
6430F–ATARM–21-Feb-12

38.6.2.6

Synchronous Channels
Some channels can be linked together as synchronous channels. They have the same source
clock, the same period, the same alignment and are started together. In this way, their counters
are synchronized together.
The synchronous channels are defined by the SYNCx bits in the “PWM Sync Channels Mode
Register” (PWM_SCM). Only one group of synchronous channels is allowed.
When a channel is defined as a synchronous channel, the channel 0 is automatically defined as
a synchronous channel too, because the channel 0 counter configuration is used by all the synchronous channels.
If a channel x is defined as a synchronous channel, it uses the following configuration fields of
the channel 0 instead of its own:
• CPRE0 field in PWM_CMR0 register instead of CPREx field in PWM_CMRx register (same
source clock)
• CPRD0 field in PWM_CMR0 register instead of CPRDx field in PWM_CMRx register (same
period)
• CALG0 field in PWM_CMR0 register instead of CALGx field in PWM_CMRx register (same
alignment)
Thus writing these fields of a synchronous channel has no effect on the output waveform of this
channel (except channel 0 of course).
Because counters of synchronous channels must start at the same time, they are all enabled
together by enabling the channel 0 (by the CHID0 bit in PWM_ENA register). In the same way,
they are all disabled together by disabling channel 0 (by the CHID0 bit in PWM_DIS register).
However, a synchronous channel x different from channel 0 can be enabled or disabled independently from others (by the CHIDx bit in PWM_ENA and PWM_DIS registers).
Defining a channel as a synchronous channel while it is an asynchronous channel (by writing the
bit SYNCx to 1 while it was at 0) is allowed only if the channel is disabled at this time (CHIDx = 0
in PWM_SR register). In the same way, defining a channel as an asynchronous channel while it is
a synchronous channel (by writing the SYNCx bit to 0 while it was 1) is allowed only if the channel is disabled at this time.
The field UPDM (Update Mode) in the PWM_SCM register allow to select one of the three methods to update the registers of the synchronous channels:
• Method 1 (UPDM = 0): the period value, the duty-cycle values and the dead-time values must
be written by the CPU in their respective update registers (respectively PWM_CPRDUPDx,
PWM_CDTYUPDx and PWM_DTUPDx).The update is triggered at the next PWM period as
soon as the bit UPDULOCK in the “PWM Sync Channels Update Control Register”
(PWM_SCUC) is set to 1 (see “Method 1: Manual write of duty-cycle values and manual
trigger of the update” on page 886).
• Method 2 (UPDM = 1): the period value, the duty-cycle values, the dead-time values and the
update period value must be written by the CPU in their respective update registers
(respectively PWM_CPRDUPDx, PWM_CDTYUPDx and PWM_DTUPD). The update of the
period value and of the dead-time values is triggered at the next PWM period as soon as the
bit UPDULOCK in the “PWM Sync Channels Update Control Register” (PWM_SCUC) is set
to 1. The update of the duty-cycle values and the update period value is triggered
automatically after an update period defined by the field UPR in the “PWM Sync Channels

884

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Update Period Register” (PWM_SCUP) (see “Method 2: Manual write of duty-cycle values
and automatic trigger of the update” on page 887).
• Method 3 (UPDM = 2): same as Method 2 apart from the fact that the duty-cycle values of
ALL synchronous channels are written by the Peripheral DMA Controller (PDC) (see “Method
3: Automatic write of duty-cycle values and automatic trigger of the update” on page 889).
The user can choose to synchronize the PDC transfer request with a comparison match (see
Section 38.6.3 “PWM Comparison Units”), by the fields PTRM and PTRCS in the PWM_SCM
register.

Table 38-4.

Summary of the Update of Registers of Synchronous Channels
UPDM=0

UPDM=1

UPDM=2

Write by the CPU
Period Value
(PWM_CPRDUPDx)

Update is triggered at the
next PWM period as soon as
the bit UPDULOCK is set to 1
Write by the CPU

Dead-Time Values
(PWM_DTUPDx)

Update is triggered at the
next PWM period as soon as
the bit UPDULOCK is set to 1
Write by the CPU

Duty-Cycle Values
(PWM_CDTYUPDx)

Update Period Value
(PWM_SCUPUPD)

Write by the CPU

Write by the PDC

Update is triggered at the next
PWM period as soon as the bit
UPDULOCK is set to 1

Update is triggered at the next
PWM period as soon as the update period
counter has reached the value UPR

Not applicable

Write by the CPU

Not applicable

Update is triggered at the next
PWM period as soon as the update period
counter has reached the value UPR

885
6430F–ATARM–21-Feb-12

Method 1: Manual write of duty-cycle values and manual trigger of the update
In this mode, the update of the period value, the duty-cycle values and the dead-time values
must be done by writing in their respective update registers with the CPU (respectively
PWM_CPRDUPDx, PWM_CDTYUPDx and PWM_DTUPDx).
To trigger the update, the user must use the bit UPDULOCK of the “PWM Sync Channels
Update Control Register” (PWM_SCUC) which allows to update synchronously (at the same
PWM period) the synchronous channels:
• If the bit UPDULOCK is set to 1, the update is done at the next PWM period of the
synchronous channels.
• If the UPDULOCK bit is not set to 1, the update is locked and cannot be performed.
After writing the UPDULOCK bit to 1, it is held at this value until the update occurs, then it is read
0.

Sequence for Method 1:
1. Select the manual write of duty-cycle values and the manual update by setting the
UPDM field to 0 in the PWM_SCM register
2. Define the synchronous channels by the SYNCx bits in the PWM_SCM register.
3. Enable the synchronous channels by writing CHID0 in the PWM_ENA register.
4. If an update of the period value and/or the duty-cycle values and/or the dead-time values is required, write registers that need to be updated (PWM_CPRDUPDx,
PWM_CDTYUPDx and PWM_DTUPDx).
5. Set UPDULOCK to 1 in PWM_SCUC.
6. The update of the registers will occur at the beginning of the next PWM period. At this
moment the UPDULOCK bit is reset, go to Step 4.) for new values.
Figure 38-9. Method 1 (UPDM = 0)
CCNT0

CDTYUPD

0x20

0x40

0x20

0x40

0x60

UPDULOCK

CDTY

886

0x60

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Method 2: Manual write of duty-cycle values and automatic trigger of the update
In this mode, the update of the period value, the duty-cycle values, the dead-time values and the
update period value must be done by writing in their respective update registers with the CPU
(respectively PWM_CPRDUPDx, PWM_CDTYUPDx, PWM_DTUPDx and PWM_SCUPUPD).
To trigger the update of the period value and the dead-time values, the user must use the bit
UPDULOCK of the “PWM Sync Channels Update Control Register” (PWM_SCUC) which
allows to update synchronously (at the same PWM period) the synchronous channels:
• If the bit UPDULOCK is set to 1, the update is done at the next PWM period of the
synchronous channels.
• If the UPDULOCK bit is not set to 1, the update is locked and cannot be performed.
After writing the UPDULOCK bit to 1, it is held at this value until the update occurs, then it is read
0.
The update of the duty-cycle values and the update period is triggered automatically after an
update period.
To configure the automatic update, the user must define a value for the Update Period by the
UPR field in the “PWM Sync Channels Update Period Register” (PWM_SCUP). The PWM controller waits UPR+1 period of synchronous channels before updating automatically the duty
values and the update period value.
The status of the duty-cycle value write is reported in the “PWM Interrupt Status Register 2”
(PWM_ISR2) by the following flags:
• WRDY: this flag is set to 1 when the PWM Controller is ready to receive new duty-cycle
values and a new update period value. It is reset to 0 when the PWM_ISR2 register is read.
Depending on the interrupt mask in the PWM_IMR2 register, an interrupt can be generated by
these flags.
Sequence for Method 2:
1. Select the manual write of duty-cycle values and the automatic update by setting the
field UPDM to 1 in the PWM_SCM register
2. Define the synchronous channels by the bits SYNCx in the PWM_SCM register.
3. Define the update period by the field UPR in the PWM_SCUP register.
4. Enable the synchronous channels by writing CHID0 in the PWM_ENA register.
5. If an update of the period value and/or of the dead-time values is required, write registers that need to be updated (PWM_CPRDUPDx, PWM_DTUPDx), else go to Step 8.
6. Set UPDULOCK to 1 in PWM_SCUC.
7. The update of these registers will occur at the beginning of the next PWM period. At
this moment the bit UPDULOCK is reset, go to Step 5. for new values.
8. If an update of the duty-cycle values and/or the update period is required, check first
that write of new update values is possible by polling the flag WRDY (or by waiting for
the corresponding interrupt) in the PWM_ISR2 register.
9. Write registers that need to be updated (PWM_CDTYUPDx, PWM_SCUPUPD).
10. The update of these registers will occur at the next PWM period of the synchronous
channels when the Update Period is elapsed. Go to Step 8. for new values.

887
6430F–ATARM–21-Feb-12

Figure 38-10. Method 2 (UPDM=1)
CCNT0

CDTYUPD

UPRUPD

0x1

UPR

0x1

UPRCNT

0x0

CDTY

0x60

0x40

0x20

0x3

0x1

0x0

0x20

0x1

0x0

0x40

0x1

0x0

0x1

0x2

0x3

0x0

0x1

0x2

0x60

WRDY

888

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Method 3: Automatic write of duty-cycle values and automatic trigger of the update
In this mode, the update of the duty cycle values is made automatically by the Peripheral DMA
Controller (PDC). The update of the period value, the dead-time values and the update period
value must be done by writing in their respective update registers with the CPU (respectively
PWM_CPRDUPDx, PWM_DTUPDx and PWM_SCUPUPD).
To trigger the update of the period value and the dead-time values, the user must use the bit
UPDULOCK which allows to update synchronously (at the same PWM period) the synchronous
channels:
• If the bit UPDULOCK is set to 1, the update is done at the next PWM period of the
synchronous channels.
• If the UPDULOCK bit is not set to 1, the update is locked and cannot be performed.
After writing the UPDULOCK bit to 1, it is held at this value until the update occurs, then it is read
0.
The update of the duty-cycle values and the update period value is triggered automatically after
an update period.
To configure the automatic update, the user must define a value for the Update Period by the
field UPR in the “PWM Sync Channels Update Period Register” (PWM_SCUP). The PWM controller waits UPR+1 periods of synchronous channels before updating automatically the duty
values and the update period value.
Using the PDC removes processor overhead by reducing its intervention during the transfer.
This significantly reduces the number of clock cycles required for a data transfer, which
improves microcontroller performance.
The PDC must write the duty-cycle values in the synchronous channels index order. For example if the channels 0, 1 and 3 are synchronous channels, the PDC must write the duty-cycle of
the channel 0 first, then the duty-cycle of the channel 1, and finally the duty-cycle of the channel
3.
The status of the PDC transfer is reported in the “PWM Interrupt Status Register 2”
(PWM_ISR2) by the following flags:
• WRDY: this flag is set to 1 when the PWM Controller is ready to receive new duty-cycle
values and a new update period value. It is reset to 0 when the PWM_ISR2 register is read.
The user can choose to synchronize the WRDY flag and the PDC transfer request with a
comparison match (see Section 38.6.3 “PWM Comparison Units”), by the fields PTRM and
PTRCS in the PWM_SCM register.
• ENDTX: this flag is set to 1 when a PDC transfer is completed
• TXBUFE: this flag is set to 1 when the PDC buffer is empty (no pending PDC transfers)
• UNRE: this flag is set to 1 when the update period defined by the UPR field has elapsed
while the whole data has not been written by the PDC. It is reset to 0 when the PWM_ISR2
register is read.
Depending on the interrupt mask in the PWM_IMR2 register, an interrupt can be generated by
these flags.

889
6430F–ATARM–21-Feb-12

Sequence for Method 3:
1. Select the automatic write of duty-cycle values and automatic update by setting the field
UPDM to 2 in the PWM_SCM register.
2. Define the synchronous channels by the bits SYNCx in the PWM_SCM register.
3. Define the update period by the field UPR in the PWM_SCUP register.
4. Define when the WRDY flag and the corresponding PDC transfer request must be set in
the update period by the PTRM bit and the PTRCS field in the PWM_SCM register (at
the end of the update period or when a comparison matches).
5. Define the PDC transfer settings for the duty-cycle values and enable it in the PDC
registers
6. Enable the synchronous channels by writing CHID0 in the PWM_ENA register.
7. If an update of the period value and/or of the dead-time values is required, write registers that need to be updated (PWM_CPRDUPDx, PWM_DTUPDx), else go to Step 10.
8. Set UPDULOCK to 1 in PWM_SCUC.
9. The update of these registers will occur at the beginning of the next PWM period. At
this moment the bit UPDULOCK is reset, go to Step 7. for new values.
10. If an update of the update period value is required, check first that write of a new update
value is possible by polling the flag WRDY (or by waiting for the corresponding interrupt) in the PWM_ISR2 register, else go to Step 13.
11. Write the register that needs to be updated (PWM_SCUPUPD).
12. The update of this register will occur at the next PWM period of the synchronous channels when the Update Period is elapsed. Go to Step 10. for new values.
13. Check the end of the PDC transfer by the flag ENDTX. If the transfer has ended, define
a new PDC transfer in the PDC registers for new duty-cycle values. Go to Step 5.

890

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 38-11. Method 3 (UPDM=2 and PTRM=0)
CCNT0

CDTYUPD

UPRUPD

0x1

UPR

0x1

UPRCNT

0x0

CDTY

0x60

0x40

0x20

0x80

0xB0

0xA0

0x3

0x1

0x0

0x1

0x0

0x1

0x2

0x3

0x0

0x1

0x80

0x60

0x40

0x20

0x0

0x2

0xA0

PDC transfer request
WRDY

Figure 38-12. Method 3 (UPDM=2 and PTRM=1 and PTRCS=0)
CCNT0

CDTYUPD

0x20

UPRUPD

0x1

UPR

0x1

UPRCNT

0x0

CDTY

0x20

0x60

0x40

0x80

0xB0

0xA0

0x3

0x1

0x0

0x40

0x1

0x0

0x60

0x1

0x0

0x80

0x1

0x2

0x3

0x0

0x1

0x2

0xA0

CMP0 match
PDC transfer request
WRDY

891
6430F–ATARM–21-Feb-12

38.6.3

PWM Comparison Units
The PWM provides 8 independent comparison units able to compare a programmed value with
the current value of the channel 0 counter (which is the channel counter of all synchronous
channels, Section 38.6.2.6 “Synchronous Channels”). These comparisons are intended to generate pulses on the event lines (used to synchronize ADC, see Section 38.6.4 “PWM Event
Lines”), to generate software interrupts and to trigger PDC transfer requests for the synchronous
channels (see “Method 3: Automatic write of duty-cycle values and automatic trigger of the
update” on page 889).
Figure 38-13. Comparison Unit Block Diagram
CEN [PWM_CMPM]x
fault on channel 0

CV [PWM_CMPVx]

CNT [PWM_CCNT0]

Comparison x

=
1

CNT [PWM_CCNT0] is decrementing

0
1

CVM [PWM_CMPVx]
CALG [PWM_CMR0]

CPRCNT [PWM_CMPMx]

CTR [PWM_CMPMx]

The comparison x matches when it is enabled by the bit CEN in the “PWM Comparison x Mode
Register” (PWM_CMPMx for the comparison x) and when the counter of the channel 0 reaches
the comparison value defined by the field CV in “PWM Comparison x Value Register”
(PWM_CMPVx for the comparison x). If the counter of the channel 0 is center aligned (CALG =
1 in “PWM Channel Mode Register” ), the bit CVM (in PWM_CMPVx) defines if the comparison
is made when the counter is counting up or counting down (in left alignment mode CALG=0, this
bit is useless).
If a fault is active on the channel 0, the comparison is disabled and cannot match (see Section
38.6.2.5 “Fault Protection”).
The user can define the periodicity of the comparison x by the fields CTR and CPR (in
PWM_CMPVx). The comparison is performed periodically once every CPR+1 periods of the
counter of the channel 0, when the value of the comparison period counter CPRCNT (in
PWM_CMPMx) reaches the value defined by CTR. CPR is the maximum value of the comparison period counter CPRCNT. If CPR=CTR=0, the comparison is performed at each period of the
counter of the channel 0.
The comparison x configuration can be modified while the channel 0 is enabled by using the
“PWM Comparison x Mode Update Register” (PWM_CMPMUPDx registers for the comparison
x). In the same way, the comparison x value can be modified while the channel 0 is enabled by
using the “PWM Comparison x Value Update Register” (PWM_CMPVUPDx registers for the
comparison x).
892

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
The update of the comparison x configuration and the comparison x value is triggered periodically after the comparison x update period. It is defined by the field CUPR in the PWM_CMPMx.
The comparison unit has an update period counter independent from the period counter to trigger this update. When the value of the comparison update period counter CUPRCNT (in
PWM_CMPMx) reaches the value defined by CUPR, the update is triggered. The comparison x
update period CUPR itself can be updated while the channel 0 is enabled by using the
PWM_CMPMUPDx register.
CAUTION: to be taken into account, the write of the register PWM_CMPVUPDx must be followed by a write of the register PWM_CMPMUPDx.
The comparison match and the comparison update can be source of an interrupt, but only if it is
enabled and not masked. These interrupts can be enabled by the “PWM Interrupt Enable Register 2” and disabled by the “PWM Interrupt Disable Register 2” . The comparison match interrupt
and the comparison update interrupt are reset by reading the “PWM Interrupt Status Register 2”
.

Figure 38-14. Comparison Waveform
CCNT0

CVUPD

0x6

0x2

0x6

CVMVUPD
CTRUPD

0x1

0x2

CPRUPD

0x1

0x3

CUPRUPD

0x3

0x2

0x6

0x2

CTR

0x1

0x2

CPR

0x1

0x3

CUPR

0x3

0x2

CUPRCNT

0x0

0x1

0x2

0x3

0x0

0x1

0x2

0x0

0x1

0x2

0x0

0x1

CPRCNT

0x0

0x1

0x0

0x1

0x0

0x1

0x2

0x3

0x0

0x1

0x2

0x3

0x6

CVM

Comparison Update
CMPU
Comparison Match
CMPM

893
6430F–ATARM–21-Feb-12

38.6.4

PWM Event Lines
The PWM provides 2 independent event lines intended to trigger actions in other peripherals (in
particular for ADC (Analog-to-Digital Converter)).
A pulse (one cycle of the master clock (MCK)) is generated on an event line, when at least one
of the selected comparisons is matching. The comparisons can be selected or unselected independently by the CSEL bits in the “PWM Event Line x Register” (PWM_ELMRx for the Event
Line x).
Figure 38-15. Event Line Block Diagram
CMPS0 (PWM_ISR2)
CSEL0 (PWM_ELMRx)
CMPS1 (PWM_ISR2)
CSEL1 (PWM_ELMRx)
CMPS2 (PWM_ISR2)
CSEL2 (PWM_ELMRx)
PULSE
GENERATOR

Event Line x

CMPS7 (PWM_ISR2)
CSEL7 (PWM_ELMRx)

894

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.6.5
38.6.5.1

PWM Controller Operations
Initialization
Before enabling the channels, they must have been configured by the software application:
• Unlock User Interface by writing the WPCMD field in the PWM_WPCR Register.
• Configuration of the clock generator (DIVA, PREA, DIVB, PREB in the PWM_CLK register if
required).
• Selection of the clock for each channel (CPRE field in the PWM_CMRx register)
• Configuration of the waveform alignment for each channel (CALG field in the PWM_CMRx
register)
• Selection of the counter event selection (if CALG = 1) for each channel (CES field in the
PWM_CMRx register)
• Configuration of the output waveform polarity for each channel (CPOL in the PWM_CMRx
register)
• Configuration of the period for each channel (CPRD in the PWM_CPRDx register). Writing in
PWM_CPRDx register is possible while the channel is disabled. After validation of the
channel, the user must use PWM_CPRDUPDx register to update PWM_CPRDx as
explained below.
• Configuration of the duty-cycle for each channel (CDTY in the PWM_CDTYx register).
Writing in PWM_CDTYx register is possible while the channel is disabled. After validation of
the channel, the user must use PWM_CDTYUPDx register to update PWM_CDTYx as
explained below.
• Configuration of the dead-time generator for each channel (DTH and DTL in PWM_DTx) if
enabled (DTE bit in the PWM_CMRx register). Writing in the PWM_DTx register is possible
while the channel is disabled. After validation of the channel, the user must use
PWM_DTUPDx register to update PWM_DTx
• Selection of the synchronous channels (SYNCx in the PWM_SCM register)
• Selection of the moment when the WRDY flag and the corresponding PDC transfer request
are set (PTRM and PTRCS in the PWM_SCM register)
• Configuration of the update mode (UPDM in the PWM_SCM register)
• Configuration of the update period (UPR in the PWM_SCUP register) if needed.
• Configuration of the comparisons (PWM_CMPVx and PWM_CMPMx).
• Configuration of the event lines (PWM_ELMRx).
• Configuration of the fault inputs polarity (FPOL in PWM_FMR)
• Configuration of the fault protection (FMOD and FFIL in PWM_FMR, PWM_FPV and
PWM_FPE1)
• Enable of the Interrupts (writing CHIDx and FCHIDx in PWM_IER1 register, and writing
WRDYE, ENDTXE, TXBUFE, UNRE, CMPMx and CMPUx in PWM_IER2 register)
• Enable of the PWM channels (writing CHIDx in the PWM_ENA register)

895
6430F–ATARM–21-Feb-12

38.6.5.2

Source Clock Selection Criteria
The large number of source clocks can make selection difficult. The relationship between the
value in the “PWM Channel Period Register” (PWM_CPRDx) and the “PWM Channel Duty
Cycle Register” (PWM_CDTYx) can help the user in choosing. The event number written in the
Period Register gives the PWM accuracy. The Duty-Cycle quantum cannot be lower than
1/CPRDx value. The higher the value of PWM_CPRDx, the greater the PWM accuracy.
For example, if the user sets 15 (in decimal) in PWM_CPRDx, the user is able to set a value
from between 1 up to 14 in PWM_CDTYx Register. The resulting duty-cycle quantum cannot be
lower than 1/15 of the PWM period.

38.6.5.3

Changing the Duty-Cycle, the Period and the Dead-Times
It is possible to modulate the output waveform duty-cycle, period and dead-times.
To prevent unexpected output waveform, the user must use the “PWM Channel Duty Cycle
Update Register” , the “PWM Channel Period Update Register” and the “PWM Channel Dead
Time Update Register” (PWM_CDTYUPDx, PWM_CPRDUPDx and PWM_DTUPDx) to change
waveform parameters while the channel is still enabled.
• If the channel is an asynchronous channel (SYNCx = 0 in “PWM Sync Channels Mode
Register” (PWM_SCM)), these registers hold the new period, duty-cycle and dead-times
values until the end of the current PWM period and update the values for the next period.
• If the channel is a synchronous channel and update method 0 is selected (SYNCx = 1 and
UPDM = 0 in PWM_SCM register), these registers hold the new period, duty-cycle and deadtimes values until the bit UPDULOCK is written at “1” (in “PWM Sync Channels Update
Control Register” (PWM_SCUC)) and the end of the current PWM period, then update the
values for the next period.
• If the channel is a synchronous channel and update method 1 or 2 is selected (SYNCx=1 and
UPDM=1 or 2 in PWM_SCM register):
– registers PWM_CPRDUPDx and PWM_DTUPDx hold the new period and deadtimes values until the bit UPDULOCK is written at “1” (in PWM_SCUC register) and
the end of the current PWM period, then update the values for the next period.
– register PWM_CDTYUPDx holds the new duty-cycle value until the end of the
update period of synchronous channels (when UPRCNT is equal to UPR in “PWM
Sync Channels Update Period Register” (PWM_SCUP)) and the end of the current
PWM period, then updates the value for the next period.

Note:

896

If the update registers PWM_CDTYUPDx, PWM_CPRDUPDx and PWM_DTUPDx are written
several times between two updates, only the last written value is taken into account.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 38-16. Synchronized Period, Duty-Cycle and Dead-Times Update
User's Writing

User's Writing

PWM_DTUPDx Value

PWM_CPRDUPDx Value

PWM_CDTYUPDx Value

PWM_CPRDx

PWM_DTx

PWM_CDTYx

- If Asynchronous Channel
-> End of PWM period
- If Synchronous Channel
-> End of PWM period and UPDULOCK = 1
- If Asynchronous Channel
-> End of PWM period
- If Synchronous Channel
- If UPDM = 0
-> End of PWM period and UPDULOCK = 1
- If UPDM = 1 or 2
-> End of PWM period and end of Update Period

897
6430F–ATARM–21-Feb-12

38.6.5.4

Changing the Synchronous Channels Update Period
It is possible to change the update period of synchronous channels while they are enabled. (See
“Method 2: Manual write of duty-cycle values and automatic trigger of the update” on page 887
and “Method 3: Automatic write of duty-cycle values and automatic trigger of the update” on
page 889.)
To prevent an unexpected update of the synchronous channels registers, the user must use the
“PWM Sync Channels Update Period Update Register” (PWM_SCUPUPD) to change the
update period of synchronous channels while they are still enabled. This register holds the new
value until the end of the update period of synchronous channels (when UPRCNT is equal to
UPR in “PWM Sync Channels Update Period Register” (PWM_SCUP)) and the end of the current PWM period, then updates the value for the next period.
Note:

If the update register PWM_SCUPUPD is written several times between two updates, only the last
written value is taken into account.

Note:

Changing the update period does make sense only if there is one or more synchronous channels
and if the update method 1 or 2 is selected (UPDM = 1 or 2 in “PWM Sync Channels Mode Register” ).

Figure 38-17. Synchronized Update of Update Period Value of Synchronous Channels
User's Writing

PWM_SCUPUPD Value

PWM_SCUP

End of PWM period and
end of Update Period
of Synchronous Channels

898

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.6.5.5

Changing the Comparison Value and the Comparison Configuration
It is possible to change the comparison values and the comparison configurations while the
channel 0 is enabled (see Section 38.6.3 “PWM Comparison Units”).
To prevent unexpected comparison match, the user must use the “PWM Comparison x Value
Update Register” and the “PWM Comparison x Mode Update Register” (PWM_CMPVUPDx
and PWM_CMPMUPDx) to change respectively the comparison values and the comparison
configurations while the channel 0 is still enabled. These registers hold the new values until the
end of the comparison update period (when CUPRCNT is equal to CUPR in “PWM Comparison
x Mode Register” (PWM_CMPMx) and the end of the current PWM period, then update the values for the next period.
CAUTION: to be taken into account, the write of the register PWM_CMPVUPDx must be followed by a write of the register PWM_CMPMUPDx.
Note:

If the update registers PWM_CMPVUPDx and PWM_CMPMUPDx are written several times
between two updates, only the last written value are taken into account.

Figure 38-18. Synchronized Update of Comparison Values and Configurations
User's Writing

User's Writing

PWM_CMPVUPDx Value
Comparison Value
for comparison x

PWM_CMPMUPDx Value
Comparison configuration
for comparison x

PWM_CMPVx

PWM_CMPMx

End of channel0 PWM period and
end of Comparison Update Period and
and PWM_CMPMx written
End of channel0 PWM period and
end of Comparison Update Period

899
6430F–ATARM–21-Feb-12

38.6.5.6

Interrupts
Depending on the interrupt mask in the PWM_IMR1 and PWM_IMR2 registers, an interrupt can
be generated at the end of the corresponding channel period (CHIDx in the PWM_ISR1 register), after a fault event (FCHIDx in the PWM_ISR1 register), after a comparison match (CMPMx
in the PWM_ISR2 register), after a comparison update (CMPUx in the PWM_ISR2 register) or
according to the transfer mode of the synchronous channels (WRDY, ENDTX, TXBUFE and
UNRE in the PWM_ISR2 register).
If the interrupt is generated by the flags CHIDx or FCHIDx, the interrupt remains active until a
read operation in the PWM_ISR1 register occurs.
If the interrupt is generated by the flags WRDY or UNRE or CMPMx or CMPUx, the interrupt
remains active until a read operation in the PWM_ISR2 register occurs.
A channel interrupt is enabled by setting the corresponding bit in the PWM_IER1 and
PWM_IER2 registers. A channel interrupt is disabled by setting the corresponding bit in the
PWM_IDR1 and PWM_IDR2 registers.

38.6.5.7

Write Protect Registers
To prevent any single software error that may corrupt PWM behavior, the registers listed below
can be write-protected by writing the field WPCMD in the “PWM Write Protect Control Register”
on page 934 (PWM_WPCR). They are divided into 6 groups:
• Register group 0:
– “PWM Clock Register” on page 905
• Register group 1:
– “PWM Disable Register” on page 907
• Register group 2:
– “PWM Sync Channels Mode Register” on page 913
– “PWM Channel Mode Register” on page 941
–
• Register group 3:
– “PWM Channel Period Register” on page 945
– “PWM Channel Period Update Register” on page 946
• Register group 4:
– “PWM Channel Dead Time Register” on page 948
– “PWM Channel Dead Time Update Register” on page 949
• Register group 5:
– “PWM Fault Mode Register” on page 928
– “PWM Fault Protection Value Register” on page 931
–
There are two types of Write Protect:
• Write Protect SW, which can be enabled or disabled.
• Write Protect HW, which can just be enabled, only a hardware reset of the PWM controller
can disable it.

900

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Both types of Write Protect can be applied independently to a particular register group by means
of the WPCMD and WPRG fields in PWM_WPCR register. If at least one Write Protect is active,
the register group is write-protected. The field WPCMD allows to perform the following actions
depending on its value:
• 0 = Disabling the Write Protect SW of the register groups of which the bit WPRG is at 1.
• 1 = Enabling the Write Protect SW of the register groups of which the bit WPRG is at 1.
• 2 = Enabling the Write Protect HW of the register groups of which the bit WPRG is at 1.

At any time, the user can determine which Write Protect is active in which register group by the
fields WPSWS and WPHWS in the “PWM Write Protect Status Register” on page 936
(PWM_WPSR).
If a write access in a write-protected register is detected, then the WPVS flag in the
PWM_WPSR register is set and the field WPVSRC indicates in which register the write access
has been attempted, through its address offset without the two LSBs.
The WPVS and PWM_WPSR fields are automatically reset after reading the PWM_WPSR
register.

901
6430F–ATARM–21-Feb-12

38.7

Pulse Width Modulation (PWM) User Interface

Table 38-5.

Offset

Name

Access

Reset

0x00

PWM Clock Register

PWM_CLK

Read-write

0x0

0x04

PWM Enable Register

PWM_ENA

Write-only

–

0x08

PWM Disable Register

PWM_DIS

Write-only

–

0x0C

PWM Status Register

PWM_SR

Read-only

0x0

0x10

PWM Interrupt Enable Register 1

PWM_IER1

Write-only

–

0x14

PWM Interrupt Disable Register 1

PWM_IDR1

Write-only

–

0x18

PWM Interrupt Mask Register 1

PWM_IMR1

Read-only

0x0

0x1C

PWM Interrupt Status Register 1

PWM_ISR1

Read-only

0x0

0x20

PWM Sync Channels Mode Register

PWM_SCM

Read-write

0x0

0x24

Reserved

–

0x28

PWM Sync Channels Update Control Register

PWM_SCUC

Read-write

0x0

0x2C

PWM Sync Channels Update Period Register

PWM_SCUP

Read-write

0x0

0x30

PWM Sync Channels Update Period Update Register

PWM_SCUPUPD

Write-only

0x0

0x34

PWM Interrupt Enable Register 2

PWM_IER2

Write-only

–

0x38

PWM Interrupt Disable Register 2

PWM_IDR2

Write-only

–

0x3C

PWM Interrupt Mask Register 2

PWM_IMR2

Read-only

0x0

0x40

PWM Interrupt Status Register 2

PWM_ISR2

Read-only

0x0

0x44

PWM Output Override Value Register

PWM_OOV

Read-write

0x0

0x48

PWM Output Selection Register

PWM_OS

Read-write

0x0

0x4C

PWM Output Selection Set Register

PWM_OSS

Write-only

–

0x50

PWM Output Selection Clear Register

PWM_OSC

Write-only

–

0x54

PWM Output Selection Set Update Register

PWM_OSSUPD

Write-only

–

0x58

PWM Output Selection Clear Update Register

PWM_OSCUPD

Write-only

–

0x5C

PWM Fault Mode Register

PWM_FMR

Read-write

0x0

0x60

PWM Fault Status Register

PWM_FSR

Read-only

0x0

0x64

PWM Fault Clear Register

PWM_FCR

Write-only

–

0x68

PWM Fault Protection Value Register

PWM_FPV

Read-write

0x0

0x6C

PWM Fault Protection Enable Register

PWM_FPE

Read-write

0x0

0x70-0x78

Reserved

–

0x7C

PWM Event Line 0 Mode Register

PWM_ELMR0

Read-write

0x0

0x80

PWM Event Line 1 Mode Register

PWM_ELMR1

Read-write

0x0

0x84-AC

Reserved

–

0xB4-E0

Reserved

–

0xE4

PWM Write Protect Control Register

Write-only

–

902

–

PWM_WPCR

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 38-5.

Offset

Name

Access

Reset

0xE8

PWM Write Protect Status Register

PWM_WPSR

Read-only

0x0

0xEC - 0xFC

Reserved

–

0x100 - 0x128

Reserved for PDC registers

–

0x12C

Reserved

–

0x130

PWM Comparison 0 Value Register

PWM_CMPV0

Read-write

0x0

0x134

PWM Comparison 0 Value Update Register

PWM_CMPVUPD0

Write-only

–

0x138

PWM Comparison 0 Mode Register

PWM_CMPM0

Read-write

0x0

0x13C

PWM Comparison 0 Mode Update Register

PWM_CMPMUPD0

Write-only

–

0x140

PWM Comparison 1 Value Register

PWM_CMPV1

Read-write

0x0

0x144

PWM Comparison 1 Value Update Register

PWM_CMPVUPD1

Write-only

–

0x148

PWM Comparison 1 Mode Register

PWM_CMPM1

Read-write

0x0

0x14C

PWM Comparison 1 Mode Update Register

PWM_CMPMUPD1

Write-only

–

0x150

PWM Comparison 2 Value Register

PWM_CMPV2

Read-write

0x0

0x154

PWM Comparison 2 Value Update Register

PWM_CMPVUPD2

Write-only

–

0x158

PWM Comparison 2 Mode Register

PWM_CMPM2

Read-write

0x0

0x15C

PWM Comparison 2 Mode Update Register

PWM_CMPMUPD2

Write-only

–

0x160

PWM Comparison 3 Value Register

PWM_CMPV3

Read-write

0x0

0x164

PWM Comparison 3 Value Update Register

PWM_CMPVUPD3

Write-only

–

0x168

PWM Comparison 3 Mode Register

PWM_CMPM3

Read-write

0x0

0x16C

PWM Comparison 3 Mode Update Register

PWM_CMPMUPD3

Write-only

–

0x170

PWM Comparison 4 Value Register

PWM_CMPV4

Read-write

0x0

0x174

PWM Comparison 4 Value Update Register

PWM_CMPVUPD4

Write-only

–

0x178

PWM Comparison 4 Mode Register

PWM_CMPM4

Read-write

0x0

0x17C

PWM Comparison 4 Mode Update Register

PWM_CMPMUPD4

Write-only

–

0x180

PWM Comparison 5 Value Register

PWM_CMPV5

Read-write

0x0

0x184

PWM Comparison 5 Value Update Register

PWM_CMPVUPD5

Write-only

–

0x188

PWM Comparison 5 Mode Register

PWM_CMPM5

Read-write

0x0

0x18C

PWM Comparison 5 Mode Update Register

PWM_CMPMUPD5

Write-only

–

0x190

PWM Comparison 6 Value Register

PWM_CMPV6

Read-write

0x0

0x194

PWM Comparison 6 Value Update Register

PWM_CMPVUPD6

Write-only

–

0x198

PWM Comparison 6 Mode Register

PWM_CMPM6

Read-write

0x0

0x19C

PWM Comparison 6 Mode Update Register

PWM_CMPMUPD6

Write-only

–

0x1A0

PWM Comparison 7 Value Register

PWM_CMPV7

Read-write

0x0

0x1A4

PWM Comparison 7 Value Update Register

PWM_CMPVUPD7

Write-only

–

0x1A8

PWM Comparison 7 Mode Register

PWM_CMPM7

Read-write

0x0

0x1AC

PWM Comparison 7 Mode Update Register

PWM_CMPMUPD7

Write-only

–

903
6430F–ATARM–21-Feb-12

Table 38-5.

Offset

0x1B0 - 0x1FC

Reserved

0x200 + ch_num *
0x20 + 0x00

PWM Channel Mode Register(1)

0x200 + ch_num *
0x20 + 0x04

Access

Reset

–

PWM_CMR

Read-write

0x0

PWM Channel Duty Cycle Register(1)

PWM_CDTY

Read-write

0x0

0x200 + ch_num *
0x20 + 0x08

PWM Channel Duty Cycle Update Register(1)

PWM_CDTYUPD

Write-only

–

0x200 + ch_num *
0x20 + 0x0C

PWM Channel Period Register(1)

PWM_CPRD

Read-write

0x0

0x200 + ch_num *
0x20 + 0x10

PWM Channel Period Update Register(1)

PWM_CPRDUPD

Write-only

–

0x200 + ch_num *
0x20 + 0x14

PWM Channel Counter Register(1)

PWM_CCNT

Read-only

0x0

0x200 + ch_num *
0x20 + 0x18

PWM Channel Dead Time Register(1)

PWM_DT

Read-write

0x0

0x200 + ch_num *
0x20 + 0x1C

PWM Channel Dead Time Update Register(1)

PWM_DTUPD

Write-only

–

Notes:

904

Name
–

1. Some registers are indexed with “ch_num” index ranging from 0 to 3.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.1
Name:

PWM Clock Register
PWM_CLK

Address:

0x4008C000

Access:

Read-write

31
–

30
–

29
–

28
–

PREB
19

DIVB
15
–

14
–

13
–

12
–

PREA
3

DIVA

This register can only be written if the bits WPSWS0 and WPHWS0 are cleared in “PWM Write Protect Status Register” on
page 936.
• DIVA, DIVB: CLKA, CLKB Divide Factor
DIVA, DIVB

CLKA, CLKB

CLKA, CLKB clock is turned off

CLKA, CLKB clock is clock selected by PREA, PREB

2-255

CLKA, CLKB clock is clock selected by PREA, PREB divided by DIVA, DIVB factor.

• PREA, PREB: CLKA, CLKB Source Clock Selection
PREA, PREB

Divider Input Clock

MCK

MCK/2

MCK/4

MCK/8

MCK/16

MCK/32

MCK/64

MCK/128

MCK/256

MCK/512

MCK/1024

Other

Reserved

905
6430F–ATARM–21-Feb-12

38.7.2
Name:

PWM Enable Register
PWM_ENA

Address:

0x4008C004

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
CHID3

2
CHID2

1
CHID1

0
CHID0

• CHIDx: Channel ID
0 = No effect.
1 = Enable PWM output for channel x.

906

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.3
Name:

PWM Disable Register
PWM_DIS

Address:

0x4008C008

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
CHID3

2
CHID2

1
CHID1

0
CHID0

This register can only be written if the bits WPSWS1 and WPHWS1 are cleared in “PWM Write Protect Status Register” on
page 936.
• CHIDx: Channel ID
0 = No effect.
1 = Disable PWM output for channel x.

907
6430F–ATARM–21-Feb-12

38.7.4
Name:

PWM Status Register
PWM_SR

Address:

0x4008C00C

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
CHID3

2
CHID2

1
CHID1

0
CHID0

• CHIDx: Channel ID
0 = PWM output for channel x is disabled.
1 = PWM output for channel x is enabled.

908

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.5
Name:

PWM Interrupt Enable Register 1
PWM_IER1

Address:

0x4008C010

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
FCHID3

18
FCHID2

17
FCHID1

16
FCHID0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
CHID3

2
CHID2

1
CHID1

0
CHID0

• CHIDx: Counter Event on Channel x Interrupt Enable
• FCHIDx: Fault Protection Trigger on Channel x Interrupt Enable

909
6430F–ATARM–21-Feb-12

38.7.6
Name:

PWM Interrupt Disable Register 1
PWM_IDR1

Address:

0x4008C014

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
FCHID3

18
FCHID2

17
FCHID1

16
FCHID0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
CHID3

2
CHID2

1
CHID1

0
CHID0

• CHIDx: Counter Event on Channel x Interrupt Disable
• FCHIDx: Fault Protection Trigger on Channel x Interrupt Disable

910

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.7
Name:

PWM Interrupt Mask Register 1
PWM_IMR1

Address:

0x4008C018

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
FCHID3

18
FCHID2

17
FCHID1

16
FCHID0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
CHID3

2
CHID2

1
CHID1

0
CHID0

• CHIDx: Counter Event on Channel x Interrupt Mask
• FCHIDx: Fault Protection Trigger on Channel x Interrupt Mask

911
6430F–ATARM–21-Feb-12

38.7.8
Name:

PWM Interrupt Status Register 1
PWM_ISR1

Address:

0x4008C01C

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
FCHID3

18
FCHID2

17
FCHID1

16
FCHID0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
CHID3

2
CHID2

1
CHID1

0
CHID0

• CHIDx: Counter Event on Channel x
0 = No new counter event has occurred since the last read of the PWM_ISR1 register.
1 = At least one counter event has occurred since the last read of the PWM_ISR1 register.
• FCHIDx: Fault Protection Trigger on Channel x
0 = No new trigger of the fault protection since the last read of the PWM_ISR1 register.
1 = At least one trigger of the fault protection since the last read of the PWM_ISR1 register.

Note:

912

Reading PWM_ISR1 automatically clears CHIDx and FCHIDx flags.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.9
Name:

PWM Sync Channels Mode Register
PWM_SCM

Address:

0x4008C020

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

22
PTRCS

20
PTRM

19
–

18
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
SYNC3

2
SYNC2

1
SYNC1

0
SYNC0

16
UPDM

This register can only be written if the bits WPSWS2 and WPHWS2 are cleared in “PWM Write Protect Status Register” on
page 936.
• SYNCx: Synchronous Channel x
0 = Channel x is not a synchronous channel.
1 = Channel x is a synchronous channel.
• UPDM: Synchronous Channels Update Mode
Value

Name

MODE0

Manual write of double buffer registers and manual update of synchronous channels(1)

MODE1

Manual write of double buffer registers and automatic update of synchronous channels(2)

MODE2

Automatic write of duty-cycle update registers by the PDC and automatic update of synchronous
channels(2)

–

Notes:

Description

Reserved

1. The update occurs at the beginning of the next PWM period, when the UPDULOCK bit in “PWM Sync Channels Update
Control Register” is set.
2. The update occurs when the Update Period is elapsed.

• PTRM: PDC Transfer Request Mode
UPDM

PTRM

WRDY Flag and PDC Transfer Request

The WRDY flag in “PWM Interrupt Status Register 2” on page 921 and the PDC transfer request
are never set to 1.

The WRDY flag in “PWM Interrupt Status Register 2” on page 921 is set to 1 as soon as the
update period is elapsed, the PDC transfer request is never set to 1.

The WRDY flag in “PWM Interrupt Status Register 2” on page 921 and the PDC transfer request
are set to 1 as soon as the update period is elapsed.

The WRDY flag in “PWM Interrupt Status Register 2” on page 921 and the PDC transfer request
are set to 1 as soon as the selected comparison matches.

913
6430F–ATARM–21-Feb-12

• PTRCS: PDC Transfer Request Comparison Selection
Selection of the comparison used to set the flag WRDY and the corresponding PDC transfer request.

914

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.10
Name:

PWM Sync Channels Update Control Register
PWM_SCUC

Address:

0x4008C028

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
–

0
UPDULOCK

• UPDULOCK: Synchronous Channels Update Unlock
0 = No effect
1 = If the UPDM field is set to “0” in “PWM Sync Channels Mode Register” on page 913, writing the UPDULOCK bit to “1”
triggers the update of the period value, the duty-cycle and the dead-time values of synchronous channels at the beginning
of the next PWM period. If the field UPDM is set to “1” or “2”, writing the UPDULOCK bit to “1” triggers only the update of
the period value and of the dead-time values of synchronous channels.
This bit is automatically reset when the update is done.

915
6430F–ATARM–21-Feb-12

38.7.11
Name:

PWM Sync Channels Update Period Register
PWM_SCUP

Address:

0x4008C02C

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

UPRCNT

UPR

• UPR: Update Period
Defines the time between each update of the synchronous channels if automatic trigger of the update is activated
(UPDM = 1 or UPDM = 2 in “PWM Sync Channels Mode Register” on page 913). This time is equal to UPR+1 periods of
the synchronous channels.
• UPRCNT: Update Period Counter
Reports the value of the Update Period Counter.

916

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.12
Name:

PWM Sync Channels Update Period Update Register
PWM_SCUPUPD

Address:

0x4008C030

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

UPRUPD

This register acts as a double buffer for the UPR value. This prevents an unexpected automatic trigger of the update of synchronous channels.
• UPRUPD: Update Period Update
Defines the wanted time between each update of the synchronous channels if automatic trigger of the update is activated
(UPDM = 1 or UPDM = 2 in “PWM Sync Channels Mode Register” on page 913). This time is equal to UPR+1 periods of
the synchronous channels.

917
6430F–ATARM–21-Feb-12

38.7.13
Name:

PWM Interrupt Enable Register 2
PWM_IER2

Address:

0x4008C034

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
CMPU7

22
CMPU6

21
CMPU5

20
CMPU4

19
CMPU3

18
CMPU2

17
CMPU1

16
CMPU0

15
CMPM7

14
CMPM6

13
CMPM5

12
CMPM4

11
CMPM3

10
CMPM2

9
CMPM1

8
CMPM0

7
–

6
–

5
–

4
–

3
UNRE

2
TXBUFE

1
ENDTX

0
WRDY

• WRDY: Write Ready for Synchronous Channels Update Interrupt Enable
• ENDTX: PDC End of TX Buffer Interrupt Enable
• TXBUFE: PDC TX Buffer Empty Interrupt Enable
• UNRE: Synchronous Channels Update Underrun Error Interrupt Enable
• CMPMx: Comparison x Match Interrupt Enable
• CMPUx: Comparison x Update Interrupt Enable

918

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.14
Name:

PWM Interrupt Disable Register 2
PWM_IDR2

Address:

0x4008C038

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
CMPU7

22
CMPU6

21
CMPU5

20
CMPU4

19
CMPU3

18
CMPU2

17
CMPU1

16
CMPU0

15
CMPM7

14
CMPM6

13
CMPM5

12
CMPM4

11
CMPM3

10
CMPM2

9
CMPM1

8
CMPM0

7
–

6
–

5
–

4
–

3
UNRE

2
TXBUFE

1
ENDTX

0
WRDY

• WRDY: Write Ready for Synchronous Channels Update Interrupt Disable
• ENDTX: PDC End of TX Buffer Interrupt Disable
• TXBUFE: PDC TX Buffer Empty Interrupt Disable
• UNRE: Synchronous Channels Update Underrun Error Interrupt Disable
• CMPMx: Comparison x Match Interrupt Disable
• CMPUx: Comparison x Update Interrupt Disable

919
6430F–ATARM–21-Feb-12

38.7.15
Name:

PWM Interrupt Mask Register 2
PWM_IMR2

Address:

0x4008C03C

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
CMPU7

22
CMPU6

21
CMPU5

20
CMPU4

19
CMPU3

18
CMPU2

17
CMPU1

16
CMPU0

15
CMPM7

14
CMPM6

13
CMPM5

12
CMPM4

11
CMPM3

10
CMPM2

9
CMPM1

8
CMPM0

7
–

6
–

5
–

4
–

3
UNRE

2
TXBUFE

1
ENDTX

0
WRDY

• WRDY: Write Ready for Synchronous Channels Update Interrupt Mask
• ENDTX: PDC End of TX Buffer Interrupt Mask
• TXBUFE: PDC TX Buffer Empty Interrupt Mask
• UNRE: Synchronous Channels Update Underrun Error Interrupt Mask
• CMPMx: Comparison x Match Interrupt Mask
• CMPUx: Comparison x Update Interrupt Mask

920

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.16
Name:

PWM Interrupt Status Register 2
PWM_ISR2

Address:

0x4008C040

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
CMPU7

22
CMPU6

21
CMPU5

20
CMPU4

19
CMPU3

18
CMPU2

17
CMPU1

16
CMPU0

15
CMPM7

14
CMPM6

13
CMPM5

12
CMPM4

11
CMPM3

10
CMPM2

9
CMPM1

8
CMPM0

7
–

6
–

5
–

4
–

3
UNRE

2
TXBUFE

1
ENDTX

0
WRDY

• WRDY: Write Ready for Synchronous Channels Update
0 = New duty-cycle and dead-time values for the synchronous channels cannot be written.
1 = New duty-cycle and dead-time values for the synchronous channels can be written.
• ENDTX: PDC End of TX Buffer
0 = The Transmit Counter register has not reached 0 since the last write of the PDC.
1 = The Transmit Counter register has reached 0 since the last write of the PDC.
• TXBUFE: PDC TX Buffer Empty
0 = PWM_TCR or PWM_TCNR has a value other than 0.
1 = Both PWM_TCR and PWM_TCNR have a value other than 0.
• UNRE: Synchronous Channels Update Underrun Error
0 = No Synchronous Channels Update Underrun has occurred since the last read of the PWM_ISR2 register.
1 = At least one Synchronous Channels Update Underrun has occurred since the last read of the PWM_ISR2 register.
• CMPMx: Comparison x Match
0 = The comparison x has not matched since the last read of the PWM_ISR2 register.
1 = The comparison x has matched at least one time since the last read of the PWM_ISR2 register.
• CMPUx: Comparison x Update
0 = The comparison x has not been updated since the last read of the PWM_ISR2 register.
1 = The comparison x has been updated at least one time since the last read of the PWM_ISR2 register.

Note:

Reading PWM_ISR2 automatically clears flags WRDY, UNRE and CMPSx.

921
6430F–ATARM–21-Feb-12

38.7.17
Name:

PWM Output Override Value Register
PWM_OOV

Address:

0x4008C044

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
OOVL3

18
OOVL2

17
OOVL1

16
OOVL0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
OOVH3

2
OOVH2

1
OOVH1

0
OOVH0

• OOVHx: Output Override Value for PWMH output of the channel x
0 = Override value is 0 for PWMH output of channel x.
1 = Override value is 1 for PWMH output of channel x.
• OOVLx: Output Override Value for PWML output of the channel x
0 = Override value is 0 for PWML output of channel x.
1 = Override value is 1 for PWML output of channel x.

922

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.18
Name:

PWM Output Selection Register
PWM_OS

Address:

0x4008C048

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
OSL3

18
OSL2

17
OSL1

16
OSL0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
OSH3

2
OSH2

1
OSH1

0
OSH0

• OSHx: Output Selection for PWMH output of the channel x
0 = Dead-time generator output DTOHx selected as PWMH output of channel x.
1 = Output override value OOVHx selected as PWMH output of channel x.
• OSLx: Output Selection for PWML output of the channel x
0 = Dead-time generator output DTOLx selected as PWML output of channel x.
1 = Output override value OOVLx selected as PWML output of channel x.

923
6430F–ATARM–21-Feb-12

38.7.19
Name:

PWM Output Selection Set Register
PWM_OSS

Address:

0x4008C04C

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
OSSL3

18
OSSL2

17
OSSL1

16
OSSL0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
OSSH3

2
OSSH2

1
OSSH1

0
OSSH0

• OSSHx: Output Selection Set for PWMH output of the channel x
0 = No effect.
1 = Output override value OOVHx selected as PWMH output of channel x.
• OSSLx: Output Selection Set for PWML output of the channel x
0 = No effect.
1 = Output override value OOVLx selected as PWML output of channel x.

924

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.20
Name:

PWM Output Selection Clear Register
PWM_OSC

Address:

0x4008C050

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
OSCL3

18
OSCL2

17
OSCL1

16
OSCL0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
OSCH3

2
OSCH2

1
OSCH1

0
OSCH0

• OSCHx: Output Selection Clear for PWMH output of the channel x
0 = No effect.
1 = Dead-time generator output DTOHx selected as PWMH output of channel x.
• OSCLx: Output Selection Clear for PWML output of the channel x
0 = No effect.
1 = Dead-time generator output DTOLx selected as PWML output of channel x.

925
6430F–ATARM–21-Feb-12

38.7.21
Name:

PWM Output Selection Set Update Register
PWM_OSSUPD

Address:

0x4008C054

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
OSSUPL3

18
OSSUPL2

17
OSSUPL1

16
OSSUPL0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
OSSUPH3

2
OSSUPH2

1
OSSUPH1

0
OSSUPH0

• OSSUPHx: Output Selection Set for PWMH output of the channel x
0 = No effect.
1 = Output override value OOVHx selected as PWMH output of channel x at the beginning of the next channel x PWM
period.
• OSSUPLx: Output Selection Set for PWML output of the channel x
0 = No effect.
1 = Output override value OOVLx selected as PWML output of channel x at the beginning of the next channel x PWM
period.

926

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.22
Name:

PWM Output Selection Clear Update Register
PWM_OSCUPD

Address:

0x4008C058

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
OSCUPL3

18
OSCUPL2

17
OSCUPL1

16
OSCUPL0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
OSCUPH3

2
OSCUPH2

1
OSCUPH1

0
OSCUPH0

• OSCUPHx: Output Selection Clear for PWMH output of the channel x
0 = No effect.
1 = Dead-time generator output DTOHx selected as PWMH output of channel x at the beginning of the next channel x
PWM period.
• OSCUPLx: Output Selection Clear for PWML output of the channel x
0 = No effect.
1 = Dead-time generator output DTOLx selected as PWML output of channel x at the beginning of the next channel x PWM
period.

927
6430F–ATARM–21-Feb-12

38.7.23
Name:

PWM Fault Mode Register
PWM_FMR

Address:

0x4008C05C

Access:

Read-write

–
23

20
FFIL

12
FMOD

4
FPOL

This register can only be written if the bits WPSWS5 and WPHWS5 are cleared in “PWM Write Protect Status Register” on
page 936.
• FPOL: Fault Polarity (fault input bit varies from 0 to 3)
For each field bit y (fault input number):
0 = The fault y becomes active when the fault input y is at 0.
1 = The fault y becomes active when the fault input y is at 1.
• FMOD: Fault Activation Mode (fault input bit varies from 0 to 3)
For each field bit y (fault input number):
0 = The fault y is active until the Fault condition is removed at the peripheral(1) level.
1 = The fault y stays active until the Fault condition is removed at the peripheral(1) level AND until it is cleared in the
“PWM Fault Clear Register” .
Note:

1. The Peripheral generating the fault.

• FFIL: Fault Filtering (fault input bit varies from 0 to 3)
For each field bit y (fault input number):
0 = The fault input y is not filtered.
1 = The fault input y is filtered.

CAUTION: To prevent an unexpected activation of the status flag FSy in the “PWM Fault Status Register” on page 929, the
bit FMODy can be set to “1” only if the FPOLy bit has been previously configured to its final value.

928

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.24
Name:

PWM Fault Status Register
PWM_FSR

Address:

0x4008C060

Access:

Read-only

–
23

20
–

12
FS

4
FIV

• FIV: Fault Input Value (fault input bit varies from 0 to 3)
For each field bit y (fault input number):
0 = The current sampled value of the fault input y is 0 (after filtering if enabled).
1 = The current sampled value of the fault input y is 1 (after filtering if enabled).
• FS: Fault Status (fault input bit varies from 0 to 3)
For each field bit y (fault input number):
0 = The fault y is not currently active.
1 = The fault y is currently active.

929
6430F–ATARM–21-Feb-12

38.7.25
Name:

PWM Fault Clear Register
PWM_FCR

Address:

0x4008C064

Access:

Write-only

–
23

20
–

12
–

4
FCLR

• FCLR: Fault Clear (fault input bit varies from 0 to 3)
For each field bit y (fault input number):
0 = No effect.
1 = If bit y of FMOD field is set to 1 and if the fault input y is not at the level defined by the bit y of FPOL field, the fault y
is cleared and becomes inactive (FMOD and FPOL fields belong to “PWM Fault Mode Register” on page 928), else
writing this bit to 1 has no effect.

930

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.26
Name:

PWM Fault Protection Value Register
PWM_FPV

Address:

0x4008C068

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
FPVL3

18
FPVL2

17
FPVL1

16
FPVL0

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
FPVH3

2
FPVH2

1
FPVH1

0
FPVH0

This register can only be written if the bits WPSWS5 and WPHWS5 are cleared in “PWM Write Protect Status Register” on
page 936.
• FPVHx: Fault Protection Value for PWMH output on channel x
0 = PWMH output of channel x is forced to 0 when fault occurs.
1 = PWMH output of channel x is forced to 1 when fault occurs.
• FPVLx: Fault Protection Value for PWML output on channel x
0 = PWML output of channel x is forced to 0 when fault occurs.
1 = PWML output of channel x is forced to 1 when fault occurs.

931
6430F–ATARM–21-Feb-12

38.7.27
Name:

PWM Fault Protection Enable Register
PWM_FPE

Address:

0x4008C06C

Access:

Read-write

FPE3
23

20
FPE2

12
FPE1

4
FPE0

This register can only be written if the bits WPSWS5 and WPHWS5 are cleared in “PWM Write Protect Status Register” on
page 936.
Only the first 4 bits (number of fault input pins) of fields FPE0, FPE1, FPE2 and FPE3 are significant.
• FPEx: Fault Protection Enable for channel x (fault input bit varies from 0 to 3)
For each field bit y (fault input number):
0 = Fault y is not used for the Fault Protection of channel x.
1 = Fault y is used for the Fault Protection of channel x.

CAUTION: To prevent an unexpected activation of the Fault Protection, the bit y of FPEx field can be set to “1” only if the
corresponding FPOL bit has been previously configured to its final value in “PWM Fault Mode Register” on page 928.

932

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.28
Name:

PWM Event Line x Register
PWM_ELMRx

Address:

0x4008C07C

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
CSEL7

6
CSEL6

5
CSEL5

4
CSEL4

3
CSEL3

2
CSEL2

1
CSEL1

0
CSEL0

• CSELy: Comparison y Selection
0 = A pulse is not generated on the event line x when the comparison y matches.
1 = A pulse is generated on the event line x when the comparison y match.

933
6430F–ATARM–21-Feb-12

38.7.29
Name:

PWM Write Protect Control Register
PWM_WPCR

Address:

0x4008C0E4

Access:

Write-only

3
WPRG1

2
WPRG0

WPKEY
23

20
WPKEY

12
WPKEY

7
WPRG5

6
WPRG4

5
WPRG3

4
WPRG2

0
WPCMD

• WPCMD: Write Protect Command
This command is performed only if the WPKEY value is correct.
0 = Disable the Write Protect SW of the register groups of which the bit WPRGx is at 1.
1 = Enable the Write Protect SW of the register groups of which the bit WPRGx is at 1.
2 = Enable the Write Protect HW of the register groups of which the bit WPRGx is at 1.
Moreover, to meet security requirements, in this mode of operation, the PIO lines associated with PWM can not be configured through the PIO interface, not even by the PIO controller.
3 = No effect.
Note:

Only a hardware reset of the PWM controller can disable the Write Protect HW.

• WPRGx: Write Protect Register Group x
0 = The WPCMD command has no effect on the register group x.
1 = The WPCMD command is applied to the register group x.
• WPKEY: Write Protect Key
Should be written at value 0x50574D (“PWM” in ASCII). Writing any other value in this field aborts the write operation of the
WPCMD field. Always reads as 0.
List of register groups:
• Register group 0:
– “PWM Clock Register” on page 905
• Register group 1:
– “PWM Disable Register” on page 907
• Register group 2:
– “PWM Sync Channels Mode Register” on page 913
– “PWM Channel Mode Register” on page 941
–

934

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• Register group 3:
– “PWM Channel Period Register” on page 945
– “PWM Channel Period Update Register” on page 946
• Register group 4:
– “PWM Channel Dead Time Register” on page 948
– “PWM Channel Dead Time Update Register” on page 949
• Register group 5:
– “PWM Fault Mode Register” on page 928
– “PWM Fault Protection Value Register” on page 931

935
6430F–ATARM–21-Feb-12

38.7.30
Name:

PWM Write Protect Status Register
PWM_WPSR

Address:

0x4008C0E8

Access:

Read-only

WPVSRC
23

20
WPVSRC

15
–

14
–

13
WPHWS5

12
WPHWS4

11
WPHWS3

10
WPHWS2

9
WPHWS1

8
WPHWS0

7
WPVS

6
–

5
WPSWS5

4
WPSWS4

3
WPSWS3

2
WPSWS2

1
WPSWS1

0
WPSWS0

• WPSWSx: Write Protect SW Status
0 = The Write Protect SW x of the register group x is disabled.
1 = The Write Protect SW x of the register group x is enabled.
• WPHWSx: Write Protect HW Status
0 = The Write Protect HW x of the register group x is disabled.
1 = The Write Protect HW x of the register group x is enabled.
• WPVS: Write Protect Violation Status
0 = No Write Protect violation has occurred since the last read of the PWM_WPSR register.
1 = At least one Write Protect violation has occurred since the last read of the PWM_WPSR register. If this violation is an
unauthorized attempt to write a protected register, the associated violation is reported into field WPVSRC.
• WPVSRC: Write Protect Violation Source
When WPVS is active, this field indicates the write-protected register (through address offset) in which a write access has
been attempted.
Note:

The two LSBs of the address offset of the write-protected register are not reported

Note:

Reading PWM_WPSR automatically clears WPVS and WPVSRC fields.

936

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.31
Name:

PWM Comparison x Value Register
PWM_CMPVx

Address:

0x4008C130 [0], 0x4008C140 [1], 0x4008C150 [2], 0x4008C160 [3], 0x4008C170 [4], 0x4008C180 [5],
0x4008C190 [6], 0x4008C1A0 [7]

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
CVM

CV
15

12
CV

4
CV

Only the first 16 bits (channel counter size) of field CV are significant.
• CV: Comparison x Value
Define the comparison x value to be compared with the counter of the channel 0.
• CVM: Comparison x Value Mode
0 = The comparison x between the counter of the channel 0 and the comparison x value is performed when this counter is
incrementing.
1 = The comparison x between the counter of the channel 0 and the comparison x value is performed when this counter is
decrementing.
Note:

This bit is useless if the counter of the channel 0 is left aligned (CALG = 0 in “PWM Channel Mode Register” on page 941)

937
6430F–ATARM–21-Feb-12

38.7.32
Name:

PWM Comparison x Value Update Register
PWM_CMPVUPDx

Address:

0x4008C134 [0], 0x4008C144 [1], 0x4008C154 [2], 0x4008C164 [3], 0x4008C174 [4], 0x4008C184 [5],
0x4008C194 [6], 0x4008C1A4 [7]

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
CVMUPD

CVUPD
15

12
CVUPD

4
CVUPD

This register acts as a double buffer for the CV and CVM values. This prevents an unexpected comparison x match.
Only the first 16 bits (channel counter size) of field CVUPD are significant.
• CVUPD: Comparison x Value Update
Define the comparison x value to be compared with the counter of the channel 0.
• CVMUPD: Comparison x Value Mode Update
0 = The comparison x between the counter of the channel 0 and the comparison x value is performed when this counter is
incrementing.
1 = The comparison x between the counter of the channel 0 and the comparison x value is performed when this counter is
decrementing.
Note:

This bit is useless if the counter of the channel 0 is left aligned (CALG = 0 in “PWM Channel Mode Register” on page 941)

CAUTION: to be taken into account, the write of the register PWM_CMPVUPDx must be followed by a write of the register
PWM_CMPMUPDx.

938

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.33
Name:

PWM Comparison x Mode Register
PWM_CMPMx

Address:

0x4008C138 [0], 0x4008C148 [1], 0x4008C158 [2], 0x4008C168 [3], 0x4008C178 [4], 0x4008C188 [5],
0x4008C198 [6], 0x4008C1A8 [7]

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

1
–

0
CEN

CUPRCNT
15

CUPR
13

CPRCNT
7

CPR
5

CTR

3
–

2
–

• CEN: Comparison x Enable
0 = The comparison x is disabled and can not match.
1 = The comparison x is enabled and can match.
• CTR: Comparison x Trigger
The comparison x is performed when the value of the comparison x period counter (CPRCNT) reaches the value defined
by CTR.
• CPR: Comparison x Period
CPR defines the maximum value of the comparison x period counter (CPRCNT). The comparison x value is performed
periodically once every CPR+1 periods of the channel 0 counter.
• CPRCNT: Comparison x Period Counter
Reports the value of the comparison x period counter.
Note:

The field CPRCNT is read-only

• CUPR: Comparison x Update Period
Defines the time between each update of the comparison x mode and the comparison x value. This time is equal to
CUPR+1 periods of the channel 0 counter.
• CUPRCNT: Comparison x Update Period Counter
Reports the value of the comparison x update period counter.
Note:

The field CUPRCNT is read-only

939
6430F–ATARM–21-Feb-12

38.7.34
Name:

PWM Comparison x Mode Update Register
PWM_CMPMUPDx

Address:

0x4008C13C [0], 0x4008C14C [1], 0x4008C15C [2], 0x4008C16C [3], 0x4008C17C [4], 0x4008C18C [5],
0x4008C19C [6], 0x4008C1AC [7]

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

15
–

14
–

13
–

12
–

3
–

1
–

0
CENUPD

CTRUPD

CUPRUPD
10
CPRUPD
2
–

This register acts as a double buffer for the CEN, CTR, CPR and CUPR values. This prevents an unexpected comparison
x match.
• CENUPD: Comparison x Enable Update
0 = The comparison x is disabled and can not match.
1 = The comparison x is enabled and can match.
• CTRUPD: Comparison x Trigger Update
The comparison x is performed when the value of the comparison x period counter (CPRCNT) reaches the value defined
by CTR.
• CPRUPD: Comparison x Period Update
CPR defines the maximum value of the comparison x period counter (CPRCNT). The comparison x value is performed
periodically once every CPR+1 periods of the channel 0 counter.
• CUPRUPD: Comparison x Update Period Update
Defines the time between each update of the comparison x mode and the comparison x value. This time is equal to
CUPR+1 periods of the channel 0 counter.

940

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.35
Name:

PWM Channel Mode Register
PWM_CMRx [x=0..3]

Address:

0x4008C200 [0], 0x4008C220 [1], 0x4008C240 [2], 0x4008C260 [3]

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
DTLI

17
DTHI

16
DTE

15
–

14
–

13
–

12
–

11
–

10
CES

9
CPOL

8
CALG

7
–

6
–

5
–

4
–

CPRE

This register can only be written if the bits WPSWS2 and WPHWS2 are cleared in “PWM Write Protect Status Register” on
page 936.
• CPRE: Channel Pre-scaler
Value

Name

Description

0b0000

MCK

Master clock

0b0001

MCK_DIV_2

Master clock/2

0b0010

MCK_DIV_4

Master clock/4

0b0011

MCK_DIV_8

Master clock/8

0b0100

MCK_DIV_16

Master clock/16

0b0101

MCK_DIV_32

Master clock/32

0b0110

MCK_DIV_64

Master clock/64

0b0111

MCK_DIV_128

Master clock/128

0b1000

MCK_DIV_256

Master clock/256

0b1001

MCK_DIV_512

Master clock/512

0b1010

MCK_DIV_1024

Master clock/1024

0b1011

CLKA

Clock A

0b1100

CLKB

Clock B

• CALG: Channel Alignment
0 = The period is left aligned.
1 = The period is center aligned.
• CPOL: Channel Polarity
0 = The OCx output waveform (output from the comparator) starts at a low level.
1 = The OCx output waveform (output from the comparator) starts at a high level.

941
6430F–ATARM–21-Feb-12

• CES: Counter Event Selection
The bit CES defines when the channel counter event occurs when the period is center aligned (flag CHIDx in the “PWM
Interrupt Status Register 1” on page 912).
CALG = 0 (Left Alignment):
0/1 = The channel counter event occurs at the end of the PWM period.
CALG = 1 (Center Alignment):
0 = The channel counter event occurs at the end of the PWM period.
1 = The channel counter event occurs at the end of the PWM period and at half the PWM period.
• DTE: Dead-Time Generator Enable
0 = The dead-time generator is disabled.
1 = The dead-time generator is enabled.
• DTHI: Dead-Time PWMHx Output Inverted
0 = The dead-time PWMHx output is not inverted.
1 = The dead-time PWMHx output is inverted.
• DTLI: Dead-Time PWMLx Output Inverted
0 = The dead-time PWMLx output is not inverted.
1 = The dead-time PWMLx output is inverted.

942

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.36
Name:

PWM Channel Duty Cycle Register
PWM_CDTYx [x=0..3]

Address:

0x4008C204 [0], 0x4008C224 [1], 0x4008C244 [2], 0x4008C264 [3]

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

CDTY
15

12
CDTY

4
CDTY

Only the first 16 bits (channel counter size) are significant.
• CDTY: Channel Duty-Cycle
Defines the waveform duty-cycle. This value must be defined between 0 and CPRD (PWM_CPRx).

943
6430F–ATARM–21-Feb-12

38.7.37
Name:

PWM Channel Duty Cycle Update Register
PWM_CDTYUPDx [x=0..3]

Address:

0x4008C208 [0], 0x4008C228 [1], 0x4008C248 [2], 0x4008C268 [3]

Access:

Write-only.

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

CDTYUPD
15

12
CDTYUPD

4
CDTYUPD

This register acts as a double buffer for the CDTY value. This prevents an unexpected waveform when modifying the waveform duty-cycle.
Only the first 16 bits (channel counter size) are significant.
• CDTYUPD: Channel Duty-Cycle Update
Defines the waveform duty-cycle. This value must be defined between 0 and CPRD (PWM_CPRx).

944

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.38
Name:

PWM Channel Period Register
PWM_CPRDx [x=0..3]

Address:

0x4008C20C [0], 0x4008C22C [1], 0x4008C24C [2], 0x4008C26C [3]

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

CPRD
15

12
CPRD

4
CPRD

This register can only be written if the bits WPSWS3 and WPHWS3 are cleared in “PWM Write Protect Status Register” on
page 936.
Only the first 16 bits (channel counter size) are significant.
• CPRD: Channel Period
If the waveform is left-aligned, then the output waveform period depends on the channel counter source clock and can be
calculated:
– By using the PWM master clock (MCK) divided by an X given prescaler value (with X being 1, 2, 4, 8, 16, 32,
64, 128, 256, 512, or 1024). The resulting period formula will be:
(------------------------------X × CPRD )MCK

– By using the PWM master clock (MCK) divided by one of both DIVA or DIVB divider, the formula becomes,
respectively:
(-----------------------------------------CRPD × DIVA )( CRPD × DIVB )
or ------------------------------------------MCK
MCK

If the waveform is center-aligned, then the output waveform period depends on the channel counter source clock and can
be calculated:
– By using the PWM master clock (MCK) divided by an X given prescaler value (with X being 1, 2, 4, 8, 16, 32,
64, 128, 256, 512, or 1024). The resulting period formula will be:
(-----------------------------------------2 × X × CPRD )MCK

– By using the PWM master clock (MCK) divided by one of both DIVA or DIVB divider, the formula becomes,
respectively:
(----------------------------------------------------2 × CPRD × DIVA )( 2 × CPRD × DIVB )
or -----------------------------------------------------MCK
MCK

945
6430F–ATARM–21-Feb-12

38.7.39
Name:

PWM Channel Period Update Register
PWM_CPRDUPDx [x=0..3]

Address:

0x4008C210 [0], 0x4008C230 [1], 0x4008C250 [2], 0x4008C270 [3]

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

CPRDUPD
15

12
CPRDUPD

4
CPRDUPD

This register can only be written if the bits WPSWS3 and WPHWS3 are cleared in “PWM Write Protect Status Register” on
page 936.
This register acts as a double buffer for the CPRD value. This prevents an unexpected waveform when modifying the
waveform period.
Only the first 16 bits (channel counter size) are significant.
• CPRDUPD: Channel Period Update
If the waveform is left-aligned, then the output waveform period depends on the channel counter source clock and can be
calculated:
– By using the PWM master clock (MCK) divided by an X given prescaler value (with X being 1, 2, 4, 8, 16, 32,
64, 128, 256, 512, or 1024). The resulting period formula will be:
(--------------------------------------------X × CPRDUPD )
MCK

– By using the PWM master clock (MCK) divided by one of both DIVA or DIVB divider, the formula becomes,
respectively:
(-------------------------------------------------------CRPDUPD × DIVA )( CRPDUPD × DIVB )
or --------------------------------------------------------MCK
MCK

– By using the PWM master clock (MCK) divided by one of both DIVA or DIVB divider, the formula becomes,
respectively:
(-----------------------------------------------------------------2 × CPRDUPD × DIVA )( 2 × CPRDUPD × DIVB )
or ------------------------------------------------------------------MCK
MCK

946

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.40
Name:

PWM Channel Counter Register
PWM_CCNTx [x=0..3]

Address:

0x4008C214 [0], 0x4008C234 [1], 0x4008C254 [2], 0x4008C274 [3]

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

CNT
15

12
CNT

4
CNT

Only the first 16 bits (channel counter size) are significant.
• CNT: Channel Counter Register
Channel counter value. This register is reset when:
• the channel is enabled (writing CHIDx in the PWM_ENA register).
• the channel counter reaches CPRD value defined in the PWM_CPRDx register if the waveform is left aligned.

947
6430F–ATARM–21-Feb-12

38.7.41
Name:

PWM Channel Dead Time Register
PWM_DTx [x=0..3]

Address:

0x4008C218 [0], 0x4008C238 [1], 0x4008C258 [2], 0x4008C278 [3]

Access:

Read-write

DTL
23

20
DTL

12
DTH

4
DTH

This register can only be written if the bits WPSWS4 and WPHWS4 are cleared in “PWM Write Protect Status Register” on
page 936.
Only the first 12 bits (dead-time counter size) of fields DTH and DTL are significant.
• DTH: Dead-Time Value for PWMHx Output
Defines the dead-time value for PWMHx output. This value must be defined between 0 and CPRD-CDTY (PWM_CPRx
and PWM_CDTYx).
• DTL: Dead-Time Value for PWMLx Output
Defines the dead-time value for PWMLx output. This value must be defined between 0 and CDTY (PWM_CDTYx).

948

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
38.7.42
Name:

PWM Channel Dead Time Update Register
PWM_DTUPDx [x=0..3]

Address:

0x4008C21C [0], 0x4008C23C [1], 0x4008C25C [2], 0x4008C27C [3]

Access:

Write-only

DTLUPD
23

20
DTLUPD

12
DTHUPD

4
DTHUPD

This register can only be written if the bits WPSWS4 and WPHWS4 are cleared in “PWM Write Protect Status Register” on
page 936.
This register acts as a double buffer for the DTH and DTL values. This prevents an unexpected waveform when modifying
the dead-time values.
Only the first 12 bits (dead-time counter size) of fields DTHUPD and DTLUPD are significant.
• DTHUPD: Dead-Time Value Update for PWMHx Output
Defines the dead-time value for PWMHx output. This value must be defined between 0 and CPRD-CDTY (PWM_CPRx
and PWM_CDTYx). This value is applied only at the beginning of the next channel x PWM period.
• DTLUPD: Dead-Time Value Update for PWMLx Output
Defines the dead-time value for PWMLx output. This value must be defined between 0 and CDTY (PWM_CDTYx). This
value is applied only at the beginning of the next channel x PWM period.

949
6430F–ATARM–21-Feb-12

950

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39. USB High Speed Device Port (UDPHS)
39.1

Description
The USB High Speed Device Port (UDPHS) is compliant with the Universal Serial Bus (USB),
rev 2.0 High Speed device specification.
Each endpoint can be configured in one of several USB transfer types. It can be associated with
one, two or three banks of a Dual-port RAM used to store the current data payload. If two or
three banks are used, one DPR bank is read or written by the processor, while the other is read
or written by the USB device peripheral. This feature is mandatory for isochronous endpoints.

39.2

Embedded Characteristics
• 1 Device High Speed
• USB v2.0 High Speed Compliant, 480 Mbits Per Second
• UTMI Compliant
• 7 Endpoints
• Embedded Dual-port RAM for Endpoints
• Suspend/Resume Logic (Command of UTMI)
• Up to Three Memory Banks for Endpoints (Not for Control Endpoint)
• 4 KBytes of DPRAM

951
6430F–ATARM–21-Feb-12

Table 39-1.

UDPHS Endpoint Description
Mnemonic

Nb Bank

DMA

High Band
Width

Max. Endpoint Size

Endpoint Type

EPT_0

Control

EPT_1

512

Ctrl/Bulk/Iso(1)/Interrupt

EPT_2

512

Ctrl/Bulk/Iso(1)/Interrupt

EPT_3

Ctrl/Bulk/Iso(1)/Interrupt

EPT_4

Ctrl/Bulk/Iso(1)/Interrupt

EPT_5

1024

Ctrl/Bulk/Iso(1)/Interrupt

EPT_6

1024

Ctrl/Bulk/Iso(1)/Interrupt

Endpoint #

Note:

1. In Isochronous Mode (Iso), it is preferable that High Band Width capability is available.

The size of internal DPRAM is 4 KB.
Suspend and resume are automatically detected by the UDPHS device, which notifies the processor by raising an interrupt.

952

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.3

Block Diagram

Figure 39-1. Block Diagram

APB
Interface

APB bus

ctrl
status

DHSDP
DHSDM

AHB1
AHB bus

Rd/Wr/Ready
DMA

AHB0

APB bus

UTMI

USB2.0
CORE

DFSDP

DFSDM

Master
AHB
Multiplexeur
Slave
Local
AHB
Slave
interface

EPT
Alloc

32 bits

DPRAM

System Clock
Domain

16/8 bits

USB Clock
Domain

PMC

953
6430F–ATARM–21-Feb-12

39.4

Typical Connection
Figure 39-2. Board Schematic

PIO (VBUS DETECT)

15k Ω

(1)

"B" Receptacle
1 = VBUS
2 = D3 = D+
4 = GND
1

DHSDM
39 ± 1% Ω
DFSDM
Shell = Shield

(1)

22k Ω

CRPB

DHSDP
39 ± 1% Ω

CRPB:1µF to 10µF

DFSDP

6K8 ± 1% Ω
VBG
10 pF
GNDUTMI

Note:

39.5
39.5.1

The values shown on the 22 kΩ and 15 kΩ resistors are only valid with 3V3 supplied PIOs.
Both 39 Ω resistors need to be placed as close to the device pins as possible.

Product Dependencies
Power Management
The UDPHS is not continuously clocked.
For using the UDPHS, the programmer must first enable the UDPHS Clock in the Power Management Controller (PMC_PCER register). Then enable the PLL (PMC_UCKR register).
However, if the application does not require UDPHS operations, the UDPHS clock can be
stopped when not needed and restarted later.

39.5.2

Interrupt
The UDPHS interrupt line is connected on one of the internal sources of the Interrupt Controller.
Using the UDPHS interrupt requires the Interrupt Controller to be programmed first.
Table 39-2.

39.6
39.6.1

954

Peripheral IDs

Instance

UDPHS

Functional Description
USB V2.0 High Speed Device Port Introduction
The USB V2.0 High Speed Device Port provides communication services between host and
attached USB devices. Each device is offered with a collection of communication flows (pipes)
associated with each endpoint. Software on the host communicates with a USB Device through
a set of communication flows.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.6.2

USB V2.0 High Speed Transfer Types
A communication flow is carried over one of four transfer types defined by the USB device.
A device provides several logical communication pipes with the host. To each logical pipe is
associated an endpoint. Transfer through a pipe belongs to one of the four transfer types:
• Control Transfers: Used to configure a device at attach time and can be used for other devicespecific purposes, including control of other pipes on the device.
• Bulk Data Transfers: Generated or consumed in relatively large burst quantities and have
wide dynamic latitude in transmission constraints.
• Interrupt Data Transfers: Used for timely but reliable delivery of data, for example, characters
or coordinates with human-perceptible echo or feedback response characteristics.
• Isochronous Data Transfers: Occupy a prenegotiated amount of USB bandwidth with a
prenegotiated delivery latency. (Also called streaming real time transfers.)
As indicated below, transfers are sequential events carried out on the USB bus.
Endpoints must be configured according to the transfer type they handle.

Table 39-3.

USB Communication Flow

Transfer

Direction

Bandwidth

Endpoint Size

Error Detection

Retrying

Bidirectional

Not guaranteed

8, 16, 32, 64

Yes

Automatic

Isochronous

Unidirectional

Guaranteed

8-1024

Yes

Interrupt

Unidirectional

Not guaranteed

8-1024

Yes

Bulk

Unidirectional

Not guaranteed

8-512

Yes

Control

39.6.3

USB Transfer Event Definitions
A transfer is composed of one or several transactions;

Table 39-4.

USB Transfer Events

CONTROL
(bidirectional)

IN
(device toward host)

Control Transfers (1)

• Setup transaction →Data IN transactions ∅Status OUT transaction
• Setup transaction →Data OUT transactions ∅Status IN transaction
• Setup transaction →Status IN transaction

Bulk IN Transfer

• Data IN transaction →Data IN transaction
• Data IN transaction →Data IN transaction

Interrupt IN Transfer
Isochronous IN Transfer

• Data IN transaction →Data IN transaction

(2)

• Data OUT transaction →Data OUT transaction

Bulk OUT Transfer
OUT
(host toward device)

Isochronous OUT Transfer
Notes:

• Data OUT transaction →Data OUT transaction

Interrupt OUT Transfer
(2)

• Data OUT transaction→Data OUT transaction

1. Control transfer must use endpoints with one bank and can be aborted using a stall handshake.
2. Isochronous transfers must use endpoints configured with two or three banks.

An endpoint handles all transactions related to the type of transfer for which it has been
configured.

955
6430F–ATARM–21-Feb-12

39.6.4

USB V2.0 High Speed BUS Transactions
Each transfer results in one or more transactions over the USB bus.
There are five kinds of transactions flowing across the bus in packets:
1. Setup Transaction
2. Data IN Transaction
3. Data OUT Transaction
4. Status IN Transaction
5. Status OUT Transaction
Figure 39-3. Control Read and Write Sequences
Setup Stage

Control Write

Setup TX

Setup Stage

Control Read

No Data
Control

Setup TX

Status Stage

Data Stage

Data OUT TX

Data Stage

Data IN TX

Setup Stage

Status Stage

Setup TX

Status IN TX

Data IN TX

Status IN TX

Status Stage

Status OUT TX

A status IN or OUT transaction is identical to a data IN or OUT transaction.
39.6.5

Endpoint Configuration
The endpoint 0 is always a control endpoint, it must be programmed and active in order to be
enabled when the End Of Reset interrupt occurs.
To configure the endpoints:
• Fill the configuration register (UDPHS_EPTCFG) with the endpoint size, direction (IN or
OUT), type (CTRL, Bulk, IT, ISO) and the number of banks.
• Fill the number of transactions (NB_TRANS) for isochronous endpoints.
Note: For control endpoints the direction has no effect.
• Verify that the EPT_MAPD flag is set. This flag is set if the endpoint size and the number of
banks are correct compared to the FIFO maximum capacity and the maximum number of
allowed banks.
• Configure control flags of the endpoint and enable it in UDPHS_EPTCTLENBx according to
“UDPHS Endpoint Control Register” on page 1002.
Control endpoints can generate interrupts and use only 1 bank.

956

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
All endpoints (except endpoint 0) can be configured either as Bulk, Interrupt or Isochronous. See
Table 39-1. UDPHS Endpoint Description.
The maximum packet size they can accept corresponds to the maximum endpoint size.
Note: The endpoint size of 1024 is reserved for isochronous endpoints.
The size of the DPRAM is 4 KB. The DPR is shared by all active endpoints. The memory size
required by the active endpoints must not exceed the size of the DPRAM.

SIZE_DPRAM = SIZE _EPT0
+ NB_BANK_EPT1 x SIZE_EPT1
+ NB_BANK_EPT2 x SIZE_EPT2
+ NB_BANK_EPT3 x SIZE_EPT3
+ NB_BANK_EPT4 x SIZE_EPT4
+ NB_BANK_EPT5 x SIZE_EPT5
+ NB_BANK_EPT6 x SIZE_EPT6
+... (refer to 39.7.11 UDPHS Endpoint Configuration Register)
If a user tries to configure endpoints with a size the sum of which is greater than the DPRAM,
then the EPT_MAPD is not set.
The application has access to the physical block of DPR reserved for the endpoint through a 64
KB logical address space.
The physical block of DPR allocated for the endpoint is remapped all along the 64 KB logical
address space. The application can write a 64 KB buffer linearly.
Figure 39-4. Logical Address Space for DPR Access
DPR
8 to 64 B

1 bank

Logical address
8 to 64 B
64 KB
EP0
8 to1024 B
64 KB 8 to1024 B
8 to1024 B
EP1
...
64 KB
EP2

8 to1024 B

x banks

y banks

z banks
8 to1024 B

64 KB
EP3
...

957
6430F–ATARM–21-Feb-12

Configuration examples of UDPHS_EPTCTLx (UDPHS Endpoint Control Register) for Bulk IN
endpoint type follow below.
• With DMA
– AUTO_VALID: Automatically validate the packet and switch to the next bank.
– EPT_ENABL: Enable endpoint.
• Without DMA:
– TX_BK_RDY: An interrupt is generated after each transmission.
– EPT_ENABL: Enable endpoint.
Configuration examples of Bulk OUT endpoint type follow below.
• With DMA
– AUTO_VALID: Automatically validate the packet and switch to the next bank.
– EPT_ENABL: Enable endpoint.
• Without DMA
– RX_BK_RDY: An interrupt is sent after a new packet has been stored in the endpoint
FIFO.
– EPT_ENABL: Enable endpoint.

958

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.6.6

DPRAM Management
Endpoints can only be allocated in ascending order, from the endpoint 0 to the last endpoint to
be allocated. The user shall therefore configure them in the same order.
The allocation of an endpoint x starts when the Number of Banks field in the UDPHS Endpoint
Configuration Register (UDPHS_EPTCFGx.BK_NUMBER) is different from zero. Then, the
hardware allocates a memory area in the DPRAM and inserts it between the x-1 and x+1 endpoints. The x+1 endpoint memory window slides up and its data is lost. Note that the following
endpoint memory windows (from x+2) do not slide.
Disabling an endpoint, by writing a one to the Endpoint Disable bit in the UDPHS Endpoint Control Disable Register (UDPHS_EPTCTLDISx.EPT_DISABL), does not reset its configuration:
• the Endpoint Banks (UDPHS_EPTCFGx.BK_NUMBER),
• the Endpoint Size (UDPHS_EPTCFGx.EPT_SIZE),
• the Endpoint Direction (UDPHS_EPTCFGx.EPT_DIR),
• and the Endpoint Type (UDPHS_EPTCFGx.EPT_TYPE).
To free its memory, the user shall write a zero to the UDPHS_EPTCFGx.BK_NUMBER field.
The x+1 endpoint memory window then slides down and its data is lost. Note that the following
endpoint memory windows (from x+2) do not slide.
Figure 39-5 on page 959 illustrates the allocation and reorganization of the DPRAM in a typical
example.

Figure 39-5. Allocation and Reorganization of the DPRAM
Free Memory

Free Memory

EPT5

EPT4

EPT3

EPT3
(always allocated)

EPT4

EPT2

EPT1

EPT0

EPT4

Device:

Endpoint 3
Disabled

EPT3 (larger size)

Device:

UDPHS_EPTCTLENBx.EPT_ENABL = 1 UDPHS_EPTCTLDIS3.EPT_DISABL = 1 UDPHS_EPTCFG3.BK_NUMBER = 0
UDPHS_EPTCFGx.BK_NUMBER <> 0

Endpoints 0..5
Activated

Conflict

EPT4 Lost Memory

Endpoint 3
Memory Freed

UDPHS_EPTCTLENB3.EPT_ENABL = 1
UDPHS_EPTCFG3.BK_NUMBER <> 0

Endpoint 3
Activated

1. The endpoints 0 to 5 are enabled, configured and allocated in ascending order. Each
endpoint then owns a memory area in the DPRAM.
2. The endpoint 3 is disabled, but its memory is kept allocated by the controller.
3. In order to free its memory, its UDPHS_EPTCFGx.BK_NUMBER field is written to zero.
The endpoint 4 memory window slides down, but the endpoint 5 does not move.
4. If the user chooses to reconfigure the endpoint 3 with a larger size, the controller allocates a memory area after the endpoint 2 memory area and automatically slides up the
959
6430F–ATARM–21-Feb-12

endpoint 4 memory window. The endpoint 5 does not move and a memory conflict
appears as the memory windows of the endpoints 4 and 5 overlap. The data of these
endpoints is potentially lost.
Notes:

1. There is no way the data of the endpoint 0 can be lost (except if it is de-allocated) as the memory allocation and de-allocation may affect only higher endpoints.
2. Deactivating then reactivating the same endpoint with the same configuration only modifies
temporarily the controller DPRAM pointer and size for this endpoint. Nothing changes in the
DPRAM, higher endpoints seem not to have been moved and their data is preserved as far as
nothing has been written or received into them while changing the allocation state of the first
endpoint.
3. When the user writes a value different from zero to the UDPHS_EPTCFGx.BK_NUMBER
field, the Endpoint Mapped bit (UDPHS_EPTCFGx.EPT_MAPD) is set only if the configured
size and number of banks are correct as compared to the endpoint maximal allowed values
and to the maximal FIFO size (i.e. the DPRAM size). The UDPHS_EPTCFGx.EPT_MAPD
value does not consider memory allocation conflicts.

39.6.7

Transfer With DMA
USB packets of any length may be transferred when required by the UDPHS Device. These
transfers always feature sequential addressing.
Packet data AHB bursts may be locked on a DMA buffer basis for drastic overall AHB bus bandwidth performance boost with paged memories. These clock-cycle consuming memory row (or
bank) changes will then likely not occur, or occur only once instead of dozens times, during a
single big USB packet DMA transfer in case another AHB master addresses the memory. This
means up to 128-word single-cycle unbroken AHB bursts for Bulk endpoints and 256-word single-cycle unbroken bursts for isochronous endpoints. This maximum burst length is then
contro lled by the lowest programm ed USB endpoint size (EPT_SIZE field in the
UDPHS_EPTCFGx register) and DMA Size (BUFF_LENGTH field in the
UDPHS_DMACONTROLx register).
The USB 2.0 device average throughput may be up to nearly 60 MBytes. Its internal slave average access latency decreases as burst length increases due to the 0 wait-state side effect of
unchanged endpoints. If at least 0 wait-state word burst capability is also provided by the external DMA AHB bus slaves, each of both DMA AHB busses need less than 50% bandwidth
allocation for full USB 2.0 bandwidth usage at 30 MHz, and less than 25% at 60 MHz.
The UDPHS DMA Channel Transfer Descriptor is described in “UDPHS DMA Channel Transfer
Descriptor” on page 1013.
Note:

960

In case of debug, be careful to address the DMA to an SRAM address even if a remap is done.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 39-6. Example of DMA Chained List
Transfer Descriptor
UDPHS Registers
(Current Transfer Descriptor)

Next Descriptor Address
DMA Channel Address

Transfer Descriptor

UDPHS Next Descriptor

DMA Channel Control

Next Descriptor Address

DMA Channel Address

Transfer Descriptor

DMA Channel Control

Next Descriptor Address

DMA Channel Control
DMA Channel Address
DMA Channel Control
Null
Memory Area
Data Buff 1

Data Buff 2
Data Buff 3

39.6.8

Transfer Without DMA
Important. If the DMA is not to be used, it is necessary that it be disabled because otherwise it
can be enabled by previous versions of software without warning. If this should occur, the DMA
can process data before an interrupt without knowledge of the user.
The recommended means to disable DMA is as follows:
// Reset IP UDPHS
AT91C_BASE_UDPHS->UDPHS_CTRL &= ~AT91C_UDPHS_EN_UDPHS;
AT91C_BASE_UDPHS->UDPHS_CTRL |= AT91C_UDPHS_EN_UDPHS;
// With OR without DMA !!!
for( i=1; i<=((AT91C_BASE_UDPHS->UDPHS_IPFEATURES &
AT91C_UDPHS_DMA_CHANNEL_NBR)>>4); i++ ) {
// RESET endpoint canal DMA:
// DMA stop channel command
AT91C_BASE_UDPHS->UDPHS_DMA[i].UDPHS_DMACONTROL = 0;

// STOP

command
// Disable endpoint
AT91C_BASE_UDPHS->UDPHS_EPT[i].UDPHS_EPTCTLDIS |= 0XFFFFFFFF;
// Reset endpoint config
AT91C_BASE_UDPHS->UDPHS_EPT[i].UDPHS_EPTCTLCFG = 0;
// Reset DMA channel (Buff count and Control field)
AT91C_BASE_UDPHS->UDPHS_DMA[i].UDPHS_DMACONTROL = 0x02;

// NON

STOP command
// Reset DMA channel 0 (STOP)

961
6430F–ATARM–21-Feb-12

AT91C_BASE_UDPHS->UDPHS_DMA[i].UDPHS_DMACONTROL = 0;

// STOP

command
// Clear DMA channel status (read the register for clear it)
AT91C_BASE_UDPHS->UDPHS_DMA[i].UDPHS_DMASTATUS =
AT91C_BASE_UDPHS->UDPHS_DMA[i].UDPHS_DMASTATUS;
}

39.6.9
39.6.9.1

Handling Transactions with USB V2.0 Device Peripheral
Setup Transaction
The setup packet is valid in the DPR while RX_SETUP is set. Once RX_SETUP is cleared by
the application, the UDPHS accepts the next packets sent over the device endpoint.
When a valid setup packet is accepted by the UDPHS:
• the UDPHS device automatically acknowledges the setup packet (sends an ACK response)
• payload data is written in the endpoint
• sets the RX_SETUP interrupt
• the BYTE_COUNT field in the UDPHS_EPTSTAx register is updated
An endpoint interrupt is generated while RX_SETUP in the UDPHS_EPTSTAx register is not
cleared. This interrupt is carried out to the microcontroller if interrupts are enabled for this
endpoint.
Thus, firmware must detect RX_SETUP polling UDPHS_EPTSTAx or catching an interrupt, read
the setup packet in the FIFO, then clear the RX_SETUP bit in the UDPHS_EPTCLRSTA register
to acknowledge the setup stage.
If STALL_SNT was set to 1, then this bit is automatically reset when a setup token is detected by
the device. Then, the device still accepts the setup stage. (See Section 39.6.9.15 “STALL” on
page 973).

39.6.9.2

NYET
NYET is a High Speed only handshake. It is returned by a High Speed endpoint as part of the
PING protocol.
High Speed devices must support an improved NAK mechanism for Bulk OUT and control endpoints (except setup stage). This mechanism allows the device to tell the host whether it has
sufficient endpoint space for the next OUT transfer (see USB 2.0 spec 8.5.1 NAK Limiting via
Ping Flow Control).
The NYET/ACK response to a High Speed Bulk OUT transfer and the PING response are automatically handled by hardware in the UDPHS_EPTCTLx register (except when the user wants to
force a NAK response by using the NYET_DIS bit).
If the endpoint responds instead to the OUT/DATA transaction with an NYET handshake, this
means that the endpoint accepted the data but does not have room for another data payload.
The host controller must return to using a PING token until the endpoint indicates it has space
available.

962

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 39-7. NYET Example with Two Endpoint Banks

data 0 ACK

t=0

data 1 NYET

t = 125 μs

Bank 1 E
Bank 0 F

PING

t = 250 μs

ACK

data 0 NYET

t = 375 μs

Bank 1 F Bank 1 F
Bank 0 E' Bank 0 E

Bank 1 F
Bank 0 E

PING

t = 500 μs

Bank 1 F
Bank 0 F

NACK

PING

t = 625 μs

Bank 1 E'
Bank 0 F

ACK

E: empty
E': begin to empty
F: full

Bank 1 E
Bank 0 F

39.6.9.3

Data IN

39.6.9.4

Bulk IN or Interrupt IN
Data IN packets are sent by the device during the data or the status stage of a control transfer or
during an (interrupt/bulk/isochronous) IN transfer. Data buffers are sent packet by packet under
the control of the application or under the control of the DMA channel.
There are three ways for an application to transfer a buffer in several packets over the USB:
• packet by packet (see 39.6.9.5 below)
• 64 KB (see 39.6.9.5 below)
• DMA (see 39.6.9.6 below)

39.6.9.5

Bulk IN or Interrupt IN: Sending a Packet Under Application Control (Device to Host)
The application can write one or several banks.
A simple algorithm can be used by the application to send packets regardless of the number of
banks associated to the endpoint.
Algorithm Description for Each Packet:
• The application waits for TX_PK_RDY flag to be cleared in the UDPHS_EPTSTAx register
before it can perform a write access to the DPR.
• The application writes one USB packet of data in the DPR through the 64 KB endpoint logical
memory window.
• The application sets TX_PK_RDY flag in the UDPHS_EPTSETSTAx register.
The application is notified that it is possible to write a new packet to the DPR by the
TX_PK_RDY interrupt. This interrupt can be enabled or masked by setting the TX_PK_RDY bit
in the UDPHS_EPTCTLENB/UDPHS_EPTCTLDIS register.
Algorithm Description to Fill Several Packets:
Using the previous algorithm, the application is interrupted for each packet. It is possible to
reduce the application overhead by writing linearly several banks at the same time. The
AUTO_VALID bit in the UDPHS_EPTCTLx must be set by writing the AUTO_VALID bit in the
UDPHS_EPTCTLENBx register.
The auto-valid-bank mechanism allows the transfer of data (IN and OUT) without the intervention of the CPU. This means that bank validation (set TX_PK_RDY or clear the RX_BK_RDY bit)
is done by hardware.

963
6430F–ATARM–21-Feb-12

• The application checks the BUSY_BANK_STA field in the UDPHS_EPTSTAx register. The
application must wait that at least one bank is free.
• The application writes a number of bytes inferior to the number of free DPR banks for the
endpoint. Each time the application writes the last byte of a bank, the TX_PK_RDY signal is
automatically set by the UDPHS.
• If the last packet is incomplete (i.e., the last byte of the bank has not been written) the
application must set the TX_PK_RDY bit in the UDPHS_EPTSETSTAx register.
The application is notified that all banks are free, so that it is possible to write another burst of
packets by the BUSY_BANK interrupt. This interrupt can be enabled or masked by setting the
BUSY_BANK flag in the UDPHS_EPTCTLENB and UDPHS_EPTCTLDIS registers.
This algorithm must not be used for isochronous transfer. In this case, the ping-pong mechanism
does not operate.
A Zero Length Packet can be sent by setting just the TX_PKTRDY flag in the
UDPHS_EPTSETSTAx register.
39.6.9.6

Bulk IN or Interrupt IN: Sending a Buffer Using DMA (Device to Host)
The UDPHS integrates a DMA host controller. This DMA controller can be used to transfer a buffer from the memory to the DPR or from the DPR to the processor memory under the UDPHS
control. The DMA can be used for all transfer types except control transfer.
Example DMA configuration:
1. Program UDPHS_DMAADDRESS x with the address of the buffer that should be
transferred.
2. Enable the interrupt of the DMA in UDPHS_IEN
3. Program UDPHS_ DMACONTROLx:
– Size of buffer to send: size of the buffer to be sent to the host.
– END_B_EN: The endpoint can validate the packet (according to the values
programmed in the AUTO_VALID and SHRT_PCKT fields of UDPHS_EPTCTLx.)
(See “UDPHS Endpoint Control Register” on page 1002 and Figure 39-12. Autovalid
with DMA)
– END_BUFFIT: generate an interrupt when the BUFF_COUNT in
UDPHS_DMASTATUSx reaches 0.
– CHANN_ENB: Run and stop at end of buffer
The auto-valid-bank mechanism allows the transfer of data (IN & OUT) without the intervention
of the CPU. This means that bank validation (set TX_PK_RDY or clear the RX_BK_RDY bit) is
done by hardware.
A transfer descriptor can be used. Instead of programming the register directly, a descriptor
should be programmed and the address of this descriptor is then given to
UDPHS_DMANXTDSC to be processed after setting the LDNXT_DSC field (Load Next Descriptor Now) in UDPHS_DMACONTROLx register.
The structure that defines this transfer descriptor must be aligned.
Each buffer to be transferred must be described by a DMA Transfer descriptor (see “UDPHS
DMA Channel Transfer Descriptor” on page 1013). Transfer descriptors are chained. Before
executing transfer of the buffer, the UDPHS may fetch a new transfer descriptor from the mem-

964

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
ory address pointed by the UDPHS_DMANXTDSCx register. Once the transfer is complete, the
transfer status is updated in the UDPHS_DMASTATUSx register.
To chain a new transfer descriptor with the current DMA transfer, the DMA channel must be
stopped. To do so, INTDIS_DMA and TX_BK_RDY may be set in the UDPHS_EPTCTLENBx
register. It is also possible for the application to wait for the completion of all transfers. In this
case the LDNXT_DSC field in the last transfer descriptor UDPHS_DMACONTROLx register
must be set to 0 and CHANN_ENB set to 1.
Then the application can chain a new transfer descriptor.
The INTDIS_DMA can be used to stop the current DMA transfer if an enabled interrupt is triggered. This can be used to stop DMA transfers in case of errors.
The application can be notified at the end of any buffer transfer (ENB_BUFFIT bit in the
UDPHS_DMACONTROLx register).
Figure 39-8. Data IN Transfer for Endpoint with One Bank
Prevous Data IN TX

USB Bus
Packets

Token IN

Microcontroller Loads Data in FIFO

Data IN 1

TX_PK_RDY
Flag
(UDPHS_EPTSTAx) Set by firmware

ACK

Token IN

NAK

Cleared by hardware

Data is Sent on USB Bus

Token IN

Data IN 2

Set by the firmware

Cleared by hardware

Interrupt Pending
TX_COMPLT Flag
(UDPHS_EPTSTAx)

Interrupt Pending
Payload in FIFO

Set by hardware
DPR access by firmware

FIFO
Content

ACK

Data IN 1

Load in progress

Cleared by firmware

DPR access by hardware
Data IN 2

965
6430F–ATARM–21-Feb-12

Figure 39-9. Data IN Transfer for Endpoint with Two Banks
Microcontroller
Load Data IN Bank 0
USB Bus
Packets

Microcontroller Load Data IN Bank 1
UDPHS Device Send Bank 0

Data IN

Token IN

ACK

Microcontroller Load Data IN Bank 0
UDPHS Device Send Bank 1

Data IN

Token IN

ACK

Set by Firmware,
Cleared by Hardware
Data Payload Written switch to next bank
in FIFO Bank 0
Virtual TX_PK_RDY
bank 0
(UDPHS_EPTSTAx)

Cleared by Hardware
Data Payload Fully Transmitted

Virtual TX_PK_RDY
bank 1
(UDPHS_EPTSTAx)

Set by Firmware,
Data Payload Written in FIFO Bank 1
Interrupt Pending

TX_COMPLT
Flag
(UDPHS_EPTSTAx)
FIFO
(DPR)
Bank 0

Set by Hardware

Set by Hardware
Interrupt Cleared by Firmware

Written by
Microcontroller

Read by USB Device

FIFO
(DPR)
Bank1

Written by
Microcontroller

Read by UDPHS Device

Figure 39-10. Data IN Followed By Status OUT Transfer at the End of a Control Transfer
Device Sends the Last
Data Payload to Host
USB Bus
Packets

Token IN

Data IN

Device Sends a
Status OUT to Host

ACK

Token OUT

Data OUT (ZLP)

ACK

Token OUT

Data OUT (ZLP)

ACK

Interrupt
Pending
RX_BK_RDY
(UDPHS_EPTSTAx)

Set by Hardware
Cleared by Firmware

TX_COMPLT
(UDPHS_EPTSTAx)
Set by Hardware

Note:

966

Cleared by Firmware

A NAK handshake is always generated at the first status stage token.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 39-11. Data OUT Followed by Status IN Transfer
Host Sends the Last
Data Payload to the Device

USB Bus
Packets

Token OUT

Data OUT

Device Sends a Status IN
to the Host

ACK

Token IN

Data IN

ACK

Interrupt Pending
RX_BK_RDY
(UDPHS_EPTSTAx)

Cleared by Firmware
Set by Hardware

TX_PK_RDY
(UDPHS_EPTSTAx)
Set by Firmware

Note:

Clear by Hardware

Before proceeding to the status stage, the software should determine that there is no risk of extra data from the host (data
stage). If not certain (non-predictable data stage length), then the software should wait for a NAK-IN interrupt before proceeding
to the status stage. This precaution should be taken to avoid collision in the FIFO.

Figure 39-12. Autovalid with DMA
Bank (system)

Write

Bank 0

Bank 1

write bank 0

write bank 1

bank 0 is full

Bank 1

Bank 0

Bank 1

write bank 0

bank 1 is full

bank 0 is full

Bank 0
IN data 0

Bank (usb)

Bank 0

IN data 1

Bank 1

IN data 0

Bank 0

Bank 1

Virtual TX_PK_RDY Bank 0

Virtual TX_PK_RDY Bank 1

TX_PK_RDY
(Virtual 0 & Virtual 1)

Note:

In the illustration above Autovalid validates a bank as full, although this might not be the case, in order to continue processing
data and to send to DMA.

967
6430F–ATARM–21-Feb-12

39.6.9.7

Isochronous IN
Isochronous-IN is used to transmit a stream of data whose timing is implied by the delivery rate.
Isochronous transfer provides periodic, continuous communication between host and device.
It guarantees bandwidth and low latencies appropriate for telephony, audio, video, etc.
If the endpoint is not available (TX_PK_RDY = 0), then the device does not answer to the host.
An ERR_FL_ISO interrupt is generated in the UDPHS_EPTSTAx register and once enabled,
then sent to the CPU.
The STALL_SNT command bit is not used for an ISO-IN endpoint.

39.6.9.8

High Bandwidth Isochronous Endpoint Handling: IN Example
For high bandwidth isochronous endpoints, the DMA can be programmed with the number of
transactions (BUFF_LENGTH field in UDPHS_DMACONTROLx) and the system should provide
the required number of packets per microframe, otherwise, the host will notice a sequencing
problem.
A response should be made to the first token IN recognized inside a microframe under the following conditions:
• If at least one bank has been validated, the correct DATAx corresponding to the programmed
Number Of Transactions per Microframe (NB_TRANS) should be answered. In case of a
subsequent missed or corrupted token IN inside the microframe, the USB 2.0 Core available
data bank(s) that should normally have been transmitted during that microframe shall be
flushed at its end. If this flush occurs, an error condition is flagged (ERR_FLUSH is set in
UDPHS_EPTSTAx).
• If no bank is validated yet, the default DATA0 ZLP is answered and underflow is flagged
(ERR_FL_ISO is set in UDPHS_EPTSTAx). Then, no data bank is flushed at microframe
end.
• If no data bank has been validated at the time when a response should be made for the
second transaction of NB_TRANS = 3 transactions microframe, a DATA1 ZLP is answered
and underflow is flagged (ERR_FL_ISO is set in UDPHS_EPTSTAx). If and only if remaining
untransmitted banks for that microframe are available at its end, they are flushed and an error
condition is flagged (ERR_FLUSH is set in UDPHS_EPTSTAx).
• If no data bank has been validated at the time when a response should be made for the last
programmed transaction of a microframe, a DATA0 ZLP is answered and underflow is flagged
(ERR_FL_ISO is set in UDPHS_EPTSTAx). If and only if the remaining untransmitted data
bank for that microframe is available at its end, it is flushed and an error condition is flagged
(ERR_FLUSH is set in UDPHS_EPTSTAx).
• If at the end of a microframe no valid token IN has been recognized, no data bank is flushed
and no error condition is reported.
At the end of a microframe in which at least one data bank has been transmitted, if less than
NB_TRANS banks have been validated for that microframe, an error condition is flagged
(ERR_TRANS is set in UDPHS_EPTSTAx).
Cases of Error (in UDPHS_EPTSTAx)
• ERR_FL_ISO: There was no data to transmit inside a microframe, so a ZLP is answered by
default.

968

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• ERR_FLUSH: At least one packet has been sent inside the microframe, but the number of
token IN received is lesser than the number of transactions actually validated (TX_BK_RDY)
and likewise with the NB_TRANS programmed.
• ERR_TRANS: At least one packet has been sent inside the microframe, but the number of
token IN received is lesser than the number of programmed NB_TRANS transactions and the
packets not requested were not validated.
• ERR_FL_ISO + ERR_FLUSH: At least one packet has been sent inside the microframe, but
the data has not been validated in time to answer one of the following token IN.
• ERR_FL_ISO + ERR_TRANS: At least one packet has been sent inside the microframe, but
the data has not been validated in time to answer one of the following token IN and the data
can be discarded at the microframe end.
• ERR_FLUSH + ERR_TRANS: The first token IN has been answered and it was the only one
received, a second bank has been validated but not the third, whereas NB_TRANS was
waiting for three transactions.
• ERR_FL_ISO + ERR_FLUSH + ERR_TRANS: The first token IN has been treated, the data
for the second Token IN was not available in time, but the second bank has been validated
before the end of the microframe. The third bank has not been validated, but three
transactions have been set in NB_TRANS.
39.6.9.9

Data OUT

39.6.9.10

Bulk OUT or Interrupt OUT
Like data IN, data OUT packets are sent by the host during the data or the status stage of control transfer or during an interrupt/bulk/isochronous OUT transfer. Data buffers are sent packet
by packet under the control of the application or under the control of the DMA channel.

39.6.9.11

Bulk OUT or Interrupt OUT: Receiving a Packet Under Application Control (Host to Device)
Algorithm Description for Each Packet:
• The application enables an interrupt on RX_BK_RDY.
• When an interrupt on RX_BK_RDY is received, the application knows that UDPHS_EPTSTAx
register BYTE_COUNT bytes have been received.
• The application reads the BYTE_COUNT bytes from the endpoint.
• The application clears RX_BK_RDY.
Note:

If the application does not know the size of the transfer, it may not be a good option to use
AUTO_VALID. Because if a zero-length-packet is received, the RX_BK_RDY is automatically
cleared by the AUTO_VALID hardware and if the endpoint interrupt is triggered, the software will
not find its originating flag when reading the UDPHS_EPTSTAx register.

Algorithm to Fill Several Packets:
• The application enables the interrupts of BUSY_BANK and AUTO_VALID.
• When a BUSY_BANK interrupt is received, the application knows that all banks available for
the endpoint have been filled. Thus, the application can read all banks available.
If the application doesn’t know the size of the receive buffer, instead of using the BUSY_BANK
interrupt, the application must use RX_BK_RDY.
39.6.9.12

Bulk OUT or Interrupt OUT: Sending a Buffer Using DMA (Host To Device)
To use the DMA setting, the AUTO_VALID field is mandatory.

969
6430F–ATARM–21-Feb-12

See 39.6.9.6 Bulk IN or Interrupt IN: Sending a Buffer Using DMA (Device to Host) for more
information.
DMA Configuration Example:
1. First program UDPHS_DMAADDRESSx with the address of the buffer that should be
transferred.
2. Enable the interrupt of the DMA in UDPHS_IEN
3. Program the DMA Channelx Control Register:
– Size of buffer to be sent.
– END_B_EN: Can be used for OUT packet truncation (discarding of unbuffered
packet data) at the end of DMA buffer.
– END_BUFFIT: Generate an interrupt when BUFF_COUNT in the
UDPHS_DMASTATUSx register reaches 0.
– END_TR_EN: End of transfer enable, the UDPHS device can put an end to the
current DMA transfer, in case of a short packet.
– END_TR_IT: End of transfer interrupt enable, an interrupt is sent after the last USB
packet has been transferred by the DMA, if the USB transfer ended with a short
packet. (Beneficial when the receive size is unknown.)
– CHANN_ENB: Run and stop at end of buffer.
For OUT transfer, the bank will be automatically cleared by hardware when the application has
read all the bytes in the bank (the bank is empty).
Notes:

1. When a zero-length-packet is received, RX_BK_RDY bit in UDPHS_EPTSTAx is cleared automatically by AUTO_VALID, and the application knows of the end of buffer by the presence of
the END_TR_IT.
2. If the host sends a zero-length packet, and the endpoint is free, then the device sends an ACK.
No data is written in the endpoint, the RX_BY_RDY interrupt is generated, and the
BYTE_COUNT field in UDPHS_EPTSTAx is null.

Figure 39-13. Data OUT Transfer for Endpoint with One Bank
Host Sends Data Payload

USB Bus
Packets

Token OUT

Data OUT 1

Token OUT

Set by Hardware

Data OUT 1
Written by UDPHS Device

970

ACK

Data OUT 2

Host Resends the Next Data Payload

NAK

Token OUT

Data OUT 2

ACK

Interrupt Pending

RX_BK_RDY
(UDPHS_EPTSTAx)

FIFO (DPR)
Content

Microcontroller Transfers Data
Host Sends the Next Data Payload

Data OUT 1
Microcontroller Read

Cleared by Firmware,
Data Payload Written in FIFO
Data OUT 2
Written by UDPHS Device

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 39-14. Data OUT Transfer for an Endpoint with Two Banks
Microcontroller reads Data 1 in bank 0,
Host sends second data payload

Host sends first data payload

USB Bus
Packets

Token OUT

Virtual RX_BK_RDY
Bank 0

Data OUT 1

ACK

Token OUT

Data OUT 2

ACK

Token OUT

Set by Hardware
Data Payload written
in FIFO endpoint bank 1

Virtual RX_BK_RDY
Bank 1

Data OUT 3

Cleared by Firmware

Interrupt pending
Set by Hardware,
Data payload written
in FIFO endpoint bank 0

Microcontroller reads Data 2 in bank 1,
Host sends third data payload

Cleared by Firmware
Interrupt pending

RX_BK_RDY = (virtual bank 0 | virtual bank 1)
(UDPHS_EPTSTAx)
FIFO (DPR)
Bank 0

Data OUT 1

Data OUT 3

Read by Microcontroller

Write in progress

Data OUT 1
Write by UDPHS Device

FIFO (DPR)
Bank 1

Data OUT 2

Write by Hardware

39.6.9.13

Read by Microcontroller

High Bandwidth Isochronous Endpoint OUT

Figure 39-15. Bank Management, Example of Three Transactions per Microframe
USB bus
Transactions

MDATA0

MDATA1

DATA2

t=0
RX_BK_RDY
Microcontroller FIFO
(DPR) Access

Read Bank 1

MDATA0

t = 52.5 μs
(40% of 125 μs)

Read Bank 2

MDATA1

DATA2

USB line

t = 125 μs
RX_BK_RDY

Read Bank 3

Read Bank 1

USB 2.0 supports individual High Speed isochronous endpoints that require data rates up to 192
Mb/s (24 MB/s): 3x1024 data bytes per microframe.
To support such a rate, two or three banks may be used to buffer the three consecutive data
packets. The microcontroller (or the DMA) should be able to empty the banks very rapidly (at
least 24 MB/s on average).
NB_TRANS field in UDPHS_EPTCFGx register = Number Of Transactions per Microframe.
If NB_TRANS > 1 then it is High Bandwidth.

Example:

971
6430F–ATARM–21-Feb-12

• If NB_TRANS = 3, the sequence should be either
– MData0
– MData0/Data1
– MData0/Data1/Data2
• If NB_TRANS = 2, the sequence should be either
– MData0
– MData0/Data1
• If NB_TRANS = 1, the sequence should be
– Data0
39.6.9.14

Isochronous Endpoint Handling: OUT Example
The user can ascertain the bank status (free or busy), and the toggle sequencing of the data
packet for each bank with the UDPHS_EPTSTAx register in the three bit fields as follows:
• TOGGLESQ_STA: PID of the data stored in the current bank
• CURRENT_BANK: Number of the bank currently being accessed by the microcontroller.
• BUSY_BANK_STA: Number of busy bank
This is particularly useful in case of a missing data packet.
If the inter-packet delay between the OUT token and the Data is greater than the USB standard,
then the ISO-OUT transaction is ignored. (Payload data is not written, no interrupt is generated
to the CPU.)
If there is a data CRC (Cyclic Redundancy Check) error, the payload is, none the less, written in
the endpoint. The ERR_CRISO flag is set in UDPHS_EPTSTAx register.
If the endpoint is already full, the packet is not written in the DPRAM. The ERR_FL_ISO flag is
set in UDPHS_EPTSTAx.
If the payload data is greater than the maximum size of the endpoint, then the ERR_OVFLW flag
is set. It is the task of the CPU to manage this error. The data packet is written in the endpoint
(except the extra data).
If the host sends a Zero Length Packet, and the endpoint is free, no data is written in the endpoint, the RX_BK_RDY flag is set, and the BYTE_COUNT field in UDPHS_EPTSTAx register is
null.
The FRCESTALL command bit is unused for an isochonous endpoint.
Otherwise, payload data is written in the endpoint, the RX_BK_RDY interrupt is generated and
the BYTE_COUNT in UDPHS_EPTSTAx register is updated.

972

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.6.9.15

STALL
STALL is returned by a function in response to an IN token or after the data phase of an OUT or
in response to a PING transaction. STALL indicates that a function is unable to transmit or
receive data, or that a control pipe request is not supported.
• OUT
To stall an endpoint, set the FRCESTALL bit in UDPHS_EPTSETSTAx register and after the
STALL_SNT flag has been set, set the TOGGLE_SEG bit in the UDPHS_EPTCLRSTAx
register.
• IN
Set the FRCESTALL bit in UDPHS_EPTSETSTAx register.
Figure 39-16. Stall Handshake Data OUT Transfer
USB Bus
Packets

Token OUT

Data OUT

Stall PID

FRCESTALL
Set by Firmware

Cleared by Firmware
Interrupt Pending

STALL_SNT
Set by Hardware

Cleared by Firmware

Figure 39-17. Stall Handshake Data IN Transfer
USB Bus
Packets

Token IN

Stall PID

FRCESTALL
Cleared by Firmware

Set by Firmware

Interrupt Pending
STALL_SNT
Set by Hardware

Cleared by Firmware

973
6430F–ATARM–21-Feb-12

39.6.10

Speed Identification
The high speed reset is managed by the hardware.
At the connection, the host makes a reset which could be a classic reset (full speed) or a high
speed reset.
At the end of the reset process (full or high), the ENDRESET interrupt is generated.
Then the CPU should read the SPEED bit in UDPHS_INTSTAx to ascertain the speed mode of
the device.

39.6.11

USB V2.0 High Speed Global Interrupt
Interrupts are defined in Section 39.7.3 ”UDPHS Interrupt Enable Register” (UDPHS_IEN) and
in Section 39.7.4 ”UDPHS Interrupt Status Register” (UDPHS_INTSTA).

39.6.12

Endpoint Interrupts
Interrupts are enabled in UDPHS_IEN (see Section 39.7.3 ”UDPHS Interrupt Enable Register”)
and individually masked in UDPHS_EPTCTLENBx (see Section 39.7.12 ”UDPHS Endpoint
Control Enable Register”).

Table 39-5.

974

Endpoint Interrupt Source Masks

SHRT_PCKT

Short Packet Interrupt

BUSY_BANK

Busy Bank Interrupt

NAK_OUT

NAKOUT Interrupt

NAK_IN/ERR_FLUSH

NAKIN/Error Flush Interrupt

STALL_SNT/ERR_CRISO/ERR_NB_TRA

Stall Sent/CRC error/Number of Transaction
Error Interrupt

RX_SETUP/ERR_FL_ISO

Received SETUP/Error Flow Interrupt

TX_PK_RD /ERR_TRANS

TX Packet Read/Transaction Error Interrupt

TX_COMPLT

Transmitted IN Data Complete Interrupt

RX_BK_RDY

Received OUT Data Interrupt

ERR_OVFLW

Overflow Error Interrupt

MDATA_RX

MDATA Interrupt

DATAX_RX

DATAx Interrupt

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 39-18. UDPHS Interrupt Control Interface
(UDPHS_IEN)

Global IT mask
Global IT sources

DET_SUSPD
MICRO_SOF
USB Global
IT Sources

INT_SOF
ENDRESET
WAKE_UP
ENDOFRSM
UPSTR_RES

(UDPHS_EPTCTLENBx)
SHRT_PCKT

EP mask

BUSY_BANK
EP sources

NAK_OUT

(UDPHS_IEN)
EPT_0
husb2dev
interrupt

NAK_IN/ERR_FLUSH
STALL_SNT/ERR_CRISO/ERR_NBTRA
EPT0 IT
Sources

RX_SETUP/ERR_FL_ISO
TX_BK_RDY/ERR_TRANS
TX_COMPLT
RX_BK_RDY
ERR_OVFLW
MDATA_RX
DATAX_RX
(UDPHS_IEN)
EPT_x

EP mask
EP sources

(UDPHS_EPTCTLx)
INTDIS_DMA

EPT1-6 IT
Sources

disable DMA
channelx request

(UDPHS_DMACONTROLx)

mask
(UDPHS_IEN)
DMA_x

EN_BUFFIT
mask
DMA CH x

END_TR_IT
mask
DESC_LD_IT

975
6430F–ATARM–21-Feb-12

39.6.13
39.6.13.1

Power Modes
Controlling Device States
A USB device has several possible states. Refer to Chapter 9 (USB Device Framework) of the
Universal Serial Bus Specification, Rev 2.0.
Figure 39-19. UDPHS Device State Diagram

Attached

Hub Reset
Hub
or
Configured
Deconfigured
Bus Inactive
Powered

Suspended
Bus Activity

Power
Interruption

Reset
Bus Inactive
Suspended

Default
Bus Activity

Reset
Address
Assigned

Bus Inactive
Address

Suspended
Bus Activity

Device
Deconfigured

Device
Configured
Bus Inactive

Configured

Suspended
Bus Activity

Movement from one state to another depends on the USB bus state or on standard requests
sent through control transactions via the default endpoint (endpoint 0).
After a period of bus inactivity, the USB device enters Suspend Mode. Accepting Suspend/Resume requests from the USB host is mandatory. Constraints in Suspend Mode are very
strict for bus-powered applications; devices may not consume more than 500 μA on the USB
bus.
While in Suspend Mode, the host may wake up a device by sending a resume signal (bus activity) or a USB device may send a wake-up request to the host, e.g., waking up a PC by moving a
USB mouse.
The wake-up feature is not mandatory for all devices and must be negotiated with the host.

976

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.6.13.2

Not Powered State
Self powered devices can detect 5V VBUS using a PIO. When the device is not connected to a
host, device power consumption can be reduced by the DETACH bit in UDPHS_CTRL. Disabling the transceiver is automatically done. HSDM, HSDP, FSDP and FSDP lines are tied to
GND pull-downs integrated in the hub downstream ports.

39.6.13.3

Entering Attached State
When no device is connected, the USB FSDP and FSDM signals are tied to GND by 15 KΩ pulldowns integrated in the hub downstream ports. When a device is attached to an hub downstream port, the device connects a 1.5 KΩ pull-up on FSDP. The USB bus line goes into IDLE
state, FSDP is pulled-up by the device 1.5 KΩ resistor to 3.3V and FSDM is pulled-down by the
15 KΩ resistor to GND of the host.
After pull-up connection, the device enters the powered state. The transceiver remains disabled
until bus activity is detected.
In case of low power consumption need, the device can be stopped. When the device detects
the VBUS, the software must enable the USB transceiver by enabling the EN_UDPHS bit in
UDPHS_CTRL register.
The software can detach the pull-up by setting DETACH bit in UDPHS_CTRL register.

39.6.13.4

From Powered State to Default State (Reset)
After its connection to a USB host, the USB device waits for an end-of-bus reset. The unmasked
flag ENDRESET is set in the UDPHS_IEN register and an interrupt is triggered.
Once the ENDRESET interrupt has been triggered, the device enters Default State. In this state,
the UDPHS software must:
• Enable the default endpoint, setting the EPT_ENABL flag in the UDPHS_EPTCTLENB[0]
register and, optionally, enabling the interrupt for endpoint 0 by writing 1 in EPT_0 of the
UDPHS_IEN register. The enumeration then begins by a control transfer.
• Configure the Interrupt Mask Register which has been reset by the USB reset detection
• Enable the transceiver.
In this state, the EN_UDPHS bit in UDPHS_CTRL register must be enabled.

39.6.13.5

From Default State to Address State (Address Assigned)
After a Set Address standard device request, the USB host peripheral enters the address state.
Warning: before the device enters address state, it must achieve the Status IN transaction of
the control transfer, i.e., the UDPHS device sets its new address once the TX_COMPLT flag in
the UDPHS_EPTCTL[0] register has been received and cleared.
To move to address state, the driver software sets the DEV_ADDR field and the FADDR_EN
flag in the UDPHS_CTRL register.

39.6.13.6

From Address State to Configured State (Device Configured)
Once a valid Set Configuration standard request has been received and acknowledged, the
device enables endpoints corresponding to the current configuration. This is done by setting the
BK_NUMBER, EPT_TYPE, EPT_DIR and EPT_SIZE fields in the UDPHS_EPTCFGx registers
and enabling them by setting the EPT_ENABL flag in the UDPHS_EPTCTLENBx registers, and,
optionally, enabling corresponding interrupts in the UDPHS_IEN register.

977
6430F–ATARM–21-Feb-12

39.6.13.7

Entering Suspend State (Bus Activity)
When a Suspend (no bus activity on the USB bus) is detected, the DET_SUSPD signal in the
UDPHS_STA register is set. This triggers an interrupt if the corresponding bit is set in the
UDPHS_IEN register. This flag is cleared by writing to the UDPHS_CLRINT register. Then the
device enters Suspend Mode.
In this state bus powered devices must drain less than 500 μA from the 5V VBUS. As an example, the microcontroller switches to slow clock, disables the PLL and main oscillator, and goes
into Idle Mode. It may also switch off other devices on the board.
The UDPHS device peripheral clocks can be switched off. Resume event is asynchronously
detected.

39.6.13.8

Receiving a Host Resume
In Suspend mode, a resume event on the USB bus line is detected asynchronously, transceiver
and clocks disabled (however the pull-up should not be removed).
Once the resume is detected on the bus, the signal WAKE_UP in the UDPHS_INTSTA is set. It
may generate an interrupt if the corresponding bit in the UDPHS_IEN register is set. This interrupt may be used to wake-up the core, enable PLL and main oscillators and configure clocks.

39.6.13.9

Sending an External Resume
In Suspend State it is possible to wake-up the host by sending an external resume.
The device waits at least 5 ms after being entered in Suspend State before sending an external
resume.
The device must force a K state from 1 to 15 ms to resume the host.

978

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.6.14

Test Mode
A device must support the TEST_MODE feature when in the Default, Address or Configured
High Speed device states.
TEST_MODE can be:
• Test_J
• Test_K
• Test_Packet
• Test_SEO_NAK
(See Section 39.7.7 “UDPHS Test Register” on page 990 for definitions of each test mode.)
const char test_packet_buffer[] = {
0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,0x00,

// JKJKJKJK *

0xAA,0xAA,0xAA,0xAA,0xAA,0xAA,0xAA,0xAA,

// JJKKJJKK *

0xEE,0xEE,0xEE,0xEE,0xEE,0xEE,0xEE,0xEE,

// JJKKJJKK *

9
8
8
0xFE,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF,0xFF, //
JJJJJJJKKKKKKK * 8
0x7F,0xBF,0xDF,0xEF,0xF7,0xFB,0xFD,
0xFC,0x7E,0xBF,0xDF,0xEF,0xF7,0xFB,0xFD,0x7E
* 10}, JK

// JJJJJJJK * 8
// {JKKKKKKK

};

979
6430F–ATARM–21-Feb-12

39.7

USB High Speed Device Port (UDPHS) User Interface

Table 39-6.

Offset

0x00

UDPHS Control Register

Access

Reset

UDPHS_CTRL

Name

Read-write

0x0000_0200

Read-only

0x0000_0000

–

0x04

UDPHS Frame Number Register

UDPHS_FNUM

0x08 - 0x0C

Reserved

–

0x10

UDPHS Interrupt Enable Register

UDPHS_IEN

Read-write

0x0000_0010

0x14

UDPHS Interrupt Status Register

UDPHS_INTSTA

Read-only

0x0000_0000

0x18

UDPHS Clear Interrupt Register

UDPHS_CLRINT

Write-only

–

0x1C

UDPHS Endpoints Reset Register

UDPHS_EPTRST

Write-only

–

0x20 - 0xCC

Reserved

–

0xE0

UDPHS Test Register

UDPHS_TST

0xE4 - 0xE8

Reserved

–

0xF0

UDPHS Name1 Register

UDPHS_IPNAME1

Read-only

0x4855_5342

0xF4

UDPHS Name2 Register

UDPHS_IPNAME2

Read-only

0x3244_4556

0xF8

UDPHS Features Register

UDPHS_IPFEATURES

Read-only

0x100 + endpoint * 0x20 + 0x00

UDPHS Endpoint Configuration Register

UDPHS_EPTCFG

Read-write

0x0000_0000

0x100 + endpoint * 0x20 + 0x04

UDPHS Endpoint Control Enable Register

UDPHS_EPTCTLENB

Write-only

–

0x100 + endpoint * 0x20 + 0x08

UDPHS Endpoint Control Disable Register

UDPHS_EPTCTLDIS

Write-only

–

0x100 + endpoint * 0x20 + 0x0C

UDPHS Endpoint Control Register

UDPHS_EPTCTL

Read-only

0x0000_0000(1)

0x100 + endpoint * 0x20 + 0x10

Reserved (for endpoint)

–

0x100 + endpoint * 0x20 + 0x14

UDPHS Endpoint Set Status Register

UDPHS_EPTSETSTA

0x100 + endpoint * 0x20 + 0x18

UDPHS Endpoint Clear Status Register

UDPHS_EPTCLRSTA

Write-only

–

0x100 + endpoint * 0x20 + 0x1C

UDPHS Endpoint Status Register

UDPHS_EPTSTA

Read-only

0x0000_0040

0x120 - 0x1DC

UDPHS Endpoint1 to 6 (2) Registers

0x300 + channel * 0x10 + 0x00

UDPHS DMA Next Descriptor Address Register

UDPHS_DMANXTDSC

Read-write

0x0000_0000

0x300 + channel * 0x10 + 0x04

UDPHS DMA Channel Address Register

UDPHS_DMAADDRESS

Read-write

0x0000_0000

0x300 + channel * 0x10 + 0x08

UDPHS DMA Channel Control Register

UDPHS_DMACONTROL

Read-write

0x0000_0000

0x300 + channel * 0x10 + 0x0C

UDPHS DMA Channel Status Register

UDPHS_DMASTATUS

Read-write

0x0000_0000

0x310 - 0x370

DMA Channel1 to 5 (3) Registers

Notes:

–

Read-write

0x0000_0000

–

Write-only

–

1. The reset value for UDPHS_EPTCTL0 is 0x0000_0001.
2. The addresses for the UDPHS Endpoint registers shown here are for UDPHS Endpoint0. The structure of this group of registers is repeated successively for each endpoint according to the consecution of endpoint registers located between 0x120
and 0x1DC.
3. The DMA channel index refers to the corresponding EP number. When no DMA channel is assigned to one EP, the associated registers are reserved. This is the case for EP0, so DMA Channel 0 registers are reserved.

980

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.7.1
Name:

UDPHS Control Register
UDPHS_CTRL

Address:

0x400A4000

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
PULLD_DIS

10
REWAKEUP

9
DETACH

8
EN_UDPHS

7
FADDR_EN

3
DEV_ADDR

• DEV_ADDR: UDPHS Address
This field contains the default address (0) after power-up or UDPHS bus reset (read), or it is written with the value set by a
SET_ADDRESS request received by the device firmware (write).
• FADDR_EN: Function Address Enable
0 = Device is not in address state (read), or only the default function address is used (write).
1 = Device is in address state (read), or this bit is set by the device firmware after a successful status phase of a
SET_ADDRESS transaction (write). When set, the only address accepted by the UDPHS controller is the one stored in the
UDPHS Address field. It will not be cleared afterwards by the device firmware. It is cleared by hardware on hardware reset,
or when UDPHS bus reset is received.
• EN_UDPHS: UDPHS Enable
0 = UDPHS is disabled (read), or this bit disables and resets the UDPHS controller (write). Disable the UTMI transceiver.
The UTMI may disable the pull-up.
1 = UDPHS is enabled (read), or this bit enables the UDPHS controller (write).
• DETACH: Detach Command
0 = UDPHS is attached (read), or this bit pulls up the DP line (attach command) (write).
1 = UDPHS is detached, UTMI transceiver is suspended (read), or this bit simulates a detach on the UDPHS line and
forces the UTMI transceiver into suspend state (Suspend M = 0) (write).
See PULLD_DIS description below.
• REWAKEUP: Send Remote Wake Up
0 = Remote Wake Up is disabled (read), or this bit has no effect (write).
1 = Remote Wake Up is enabled (read), or this bit forces an external interrupt on the UDPHS controller for Remote Wake
UP purposes.
An Upstream Resume is sent only after the UDPHS bus has been in SUSPEND state for at least 5 ms.
This bit is automatically cleared by hardware at the end of the Upstream Resume.

981
6430F–ATARM–21-Feb-12

• PULLD_DIS: Pull-Down Disable
When set, there is no pull-down on DP & DM. (DM Pull-Down = DP Pull-Down = 0).
Note:

If the DETACH bit is also set, device DP & DM are left in high impedance state.

(See DETACH description above.)
DETACH

PULLD_DIS

Condition

Pull up

Pull down

not recommended

Pull up

High impedance
state

VBUS present

Pull down

No VBUS

High impedance
state

VBUS present &
software disconnect

982

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.7.2
Name:

UDPHS Frame Number Register
UDPHS_FNUM

Address:

0x400A4004

Access:

Read-only

31
FNUM_ERR

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

11
10
FRAME_NUMBER

5
FRAME_NUMBER

1
MICRO_FRAME_NUM

• MICRO_FRAME_NUM: Microframe Number
Number of the received microframe (0 to 7) in one frame.This field is reset at the beginning of each new frame (1 ms).
One microframe is received each 125 microseconds (1 ms/8).
• FRAME_NUMBER: Frame Number as defined in the Packet Field Formats
This field is provided in the last received SOF packet (see INT_SOF in the UDPHS Interrupt Status Register).
• FNUM_ERR: Frame Number CRC Error
This bit is set by hardware when a corrupted Frame Number in Start of Frame packet (or Micro SOF) is received.
This bit and the INT_SOF (or MICRO_SOF) interrupt are updated at the same time.

983
6430F–ATARM–21-Feb-12

39.7.3
Name:

UDPHS Interrupt Enable Register
UDPHS_IEN

Address:

0x400A4010

Access:

Read-write

31
–

30
DMA_6

29
DMA_5

28
DMA_4

27
DMA_3

26
DMA_2

25
DMA_1

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
EPT_6

13
EPT_5

12
EPT_4

11
EPT_3

10
EPT_2

9
EPT_1

8
EPT_0

7
UPSTR_RES

6
ENDOFRSM

5
WAKE_UP

4
ENDRESET

3
INT_SOF

2
MICRO_SOF

1
DET_SUSPD

0
–

• DET_SUSPD: Suspend Interrupt Enable
0 = disable Suspend Interrupt.
1 = enable Suspend Interrupt.
• MICRO_SOF: Micro-SOF Interrupt Enable
0 = disable Micro-SOF Interrupt.
1 = enable Micro-SOF Interrupt.
• INT_SOF: SOF Interrupt Enable
0 = disable SOF Interrupt.
1 = enable SOF Interrupt.
• ENDRESET: End Of Reset Interrupt Enable
0 = disable End Of Reset Interrupt.
1 = enable End Of Reset Interrupt. Automatically enabled after USB reset.
• WAKE_UP: Wake Up CPU Interrupt Enable
0 = disable Wake Up CPU Interrupt.
1 = enable Wake Up CPU Interrupt.
• ENDOFRSM: End Of Resume Interrupt Enable
0 = disable Resume Interrupt.
1 = enable Resume Interrupt.
• UPSTR_RES: Upstream Resume Interrupt Enable
0 = disable Upstream Resume Interrupt.
1 = enable Upstream Resume Interrupt.

984

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• EPT_x: Endpoint x Interrupt Enable
0 = disable the interrupts for this endpoint.
1 = enable the interrupts for this endpoint.
• DMA_x: DMA Channel x Interrupt Enable
0 = disable the interrupts for this channel.
1 = enable the interrupts for this channel.

985
6430F–ATARM–21-Feb-12

39.7.4
Name:

UDPHS Interrupt Status Register
UDPHS_INTSTA

Address:

0x400A4014

Access:

Read-only

31
–

30
DMA_6

29
DMA_5

28
DMA_4

27
DMA_3

26
DMA_2

25
DMA_1

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
EPT_6

13
EPT_5

12
EPT_4

11
EPT_3

10
EPT_2

9
EPT_1

8
EPT_0

7
UPSTR_RES

6
ENDOFRSM

5
WAKE_UP

4
ENDRESET

3
INT_SOF

2
MICRO_SOF

1
DET_SUSPD

0
SPEED

• SPEED: Speed Status
0 = reset by hardware when the hardware is in Full Speed mode.
1 = set by hardware when the hardware is in High Speed mode
• DET_SUSPD: Suspend Interrupt
0 = cleared by setting the DET_SUSPD bit in UDPHS_CLRINT register
1 = set by hardware when a UDPHS Suspend (Idle bus for three frame periods, a J state for 3 ms) is detected. This triggers
a UDPHS interrupt when the DET_SUSPD bit is set in UDPHS_IEN register.
• MICRO_SOF: Micro Start Of Frame Interrupt
0 = cleared by setting the MICRO_SOF bit in UDPHS_CLRINT register.
1 = set by hardware when an UDPHS micro start of frame PID (SOF) has been detected (every 125 us) or synthesized by
the macro. This triggers a UDPHS interrupt when the MICRO_SOF bit is set in UDPHS_IEN. In case of detected SOF, the
MICRO_FRAME_NUM field in UDPHS_FNUM register is incremented and the FRAME_NUMBER field doesn’t change.
Note:

The Micro Start Of Frame Interrupt (MICRO_SOF), and the Start Of Frame Interrupt (INT_SOF) are not generated at the same
time.

• INT_SOF: Start Of Frame Interrupt
0 = cleared by setting the INT_SOF bit in UDPHS_CLRINT.
1 = set by hardware when an UDPHS Start Of Frame PID (SOF) has been detected (every 1 ms) or synthesized by the
macro. This triggers a UDPHS interrupt when the INT_SOF bit is set in UDPHS_IEN register. In case of detected SOF, in
High Speed mode, the MICRO_FRAME_NUMBER field is cleared in UDPHS_FNUM register and the FRAME_NUMBER
field is updated.
• ENDRESET: End Of Reset Interrupt
0 = cleared by setting the ENDRESET bit in UDPHS_CLRINT.
1 = set by hardware when an End Of Reset has been detected by the UDPHS controller. This triggers a UDPHS interrupt
when the ENDRESET bit is set in UDPHS_IEN.

986

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• WAKE_UP: Wake Up CPU Interrupt
0 = cleared by setting the WAKE_UP bit in UDPHS_CLRINT.
1 = set by hardware when the UDPHS controller is in SUSPEND state and is re-activated by a filtered non-idle signal from
the UDPHS line (not by an upstream resume). This triggers a UDPHS interrupt when the WAKE_UP bit is set in
UDPHS_IEN register. When receiving this interrupt, the user has to enable the device controller clock prior to operation.
Note:

this interrupt is generated even if the device controller clock is disabled.

• ENDOFRSM: End Of Resume Interrupt
0 = cleared by setting the ENDOFRSM bit in UDPHS_CLRINT.
1 = set by hardware when the UDPHS controller detects a good end of resume signal initiated by the host. This triggers a
UDPHS interrupt when the ENDOFRSM bit is set in UDPHS_IEN.
• UPSTR_RES: Upstream Resume Interrupt
0 = cleared by setting the UPSTR_RES bit in UDPHS_CLRINT.
1 = set by hardware when the UDPHS controller is sending a resume signal called “upstream resume”. This triggers a
UDPHS interrupt when the UPSTR_RES bit is set in UDPHS_IEN.
• EPT_x: Endpoint x Interrupt
0 = reset when the UDPHS_EPTSTAx interrupt source is cleared.
1 = set by hardware when an interrupt is triggered by the UDPHS_EPTSTAx register and this endpoint interrupt is enabled
by the EPT_x bit in UDPHS_IEN.
• DMA_x: DMA Channel x Interrupt
0 = reset when the UDPHS_DMASTATUSx interrupt source is cleared.
1 = set by hardware when an interrupt is triggered by the DMA Channelx and this endpoint interrupt is enabled by the
DMA_x bit in UDPHS_IEN.

987
6430F–ATARM–21-Feb-12

39.7.5
Name:

UDPHS Clear Interrupt Register
UDPHS_CLRINT

Address:

0x400A4018

Access:

Write only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
UPSTR_RES

6
ENDOFRSM

5
WAKE_UP

4
ENDRESET

3
INT_SOF

2
MICRO_SOF

1
DET_SUSPD

0
–

• DET_SUSPD: Suspend Interrupt Clear
0 = no effect.
1 = clear the DET_SUSPD bit in UDPHS_INTSTA.
• MICRO_SOF: Micro Start Of Frame Interrupt Clear
0 = no effect.
1 = clear the MICRO_SOF bit in UDPHS_INTSTA.
• INT_SOF: Start Of Frame Interrupt Clear
0 = no effect.
1 = clear the INT_SOF bit in UDPHS_INTSTA.
• ENDRESET: End Of Reset Interrupt Clear
0 = no effect.
1 = clear the ENDRESET bit in UDPHS_INTSTA.
• WAKE_UP: Wake Up CPU Interrupt Clear
0 = no effect.
1 = clear the WAKE_UP bit in UDPHS_INTSTA.
• ENDOFRSM: End Of Resume Interrupt Clear
0 = no effect.
1 = clear the ENDOFRSM bit in UDPHS_INTSTA.
• UPSTR_RES: Upstream Resume Interrupt Clear
0 = no effect.
1 = clear the UPSTR_RES bit in UDPHS_INTSTA.

988

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.7.6
Name:

UDPHS Endpoints Reset Register
UDPHS_EPTRST

Address:

0x400A401C

Access:

Write only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
EPT_6

5
EPT_5

4
EPT_4

3
EPT_3

2
EPT_2

1
EPT_1

0
EPT_0

• EPT_x: Endpoint x Reset
0 = no effect.
1 = reset the Endpointx state.
Setting this bit clears the Endpoint status UDPHS_EPTSTAx register, except for the TOGGLESQ_STA field.

989
6430F–ATARM–21-Feb-12

39.7.7
Name:

UDPHS Test Register
UDPHS_TST

Address:

0x400A40E0

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
OPMODE2

4
TST_PKT

3
TST_K

2
TST_J

0
SPEED_CFG

• SPEED_CFG: Speed Configuration
Speed Configuration:
Value
0

Name

Description2

NORMAL

Normal Mode: The macro is in Full Speed mode, ready to make a High Speed identification, if the host
supports it and then to automatically switch to High Speed mode

Reserved

HIGH_SPEED

Force High Speed: Set this value to force the hardware to work in High Speed mode. Only for debug
or test purpose.

FULL_SPEED

Force Full Speed: Set this value to force the hardware to work only in Full Speed mode. In this
configuration, the macro will not respond to a High Speed reset handshake.

• TST_J: Test J Mode
0 = no effect.
1 = set to send the J state on the UDPHS line. This enables the testing of the high output drive level on the D+ line.
• TST_K: Test K Mode
0 = no effect.
1 = set to send the K state on the UDPHS line. This enables the testing of the high output drive level on the D- line.
• TST_PKT: Test Packet Mode
0 = no effect.
1 = set to repetitively transmit the packet stored in the current bank. This enables the testing of rise and fall times, eye patterns, jitter, and any other dynamic waveform specifications.
• OPMODE2: OpMode2
0 = no effect.
1 = set to force the OpMode signal (UTMI interface) to “10”, to disable the bit-stuffing and the NRZI encoding.

990

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Note:

For the Test mode, Test_SE0_NAK (see Universal Serial Bus Specification, Revision 2.0: 7.1.20, Test Mode Support). Force the
device in High Speed mode, and configure a bulk-type endpoint. Do not fill this endpoint for sending NAK to the host.
Upon command, a port’s transceiver must enter the High Speed receive mode and remain in that mode until the exit action is
taken. This enables the testing of output impedance, low level output voltage and loading characteristics. In addition, while in
this mode, upstream facing ports (and only upstream facing ports) must respond to any IN token packet with a NAK handshake
(only if the packet CRC is determined to be correct) within the normal allowed device response time. This enables testing of the
device squelch level circuitry and, additionally, provides a general purpose stimulus/response test for basic functional testing.

991
6430F–ATARM–21-Feb-12

39.7.8
Name:

UDPHS Name1 Register
UDPHS_IPNAME1

Address:

0x400A40F0

Access:

Read-only

IP_NAME1
23

20
IP_NAME1

12
IP_NAME1

4
IP_NAME1

• IP_NAME1
ASCII string “HUSB”

992

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.7.9
Name:

UDPHS Name2 Register
UDPHS_IPNAME2

Address:

0x400A40F4

Access:

Read-only

IP_NAME2
23

20
IP_NAME2

12
IP_NAME2

4
IP_NAME2

• IP_NAME2
ASCII string “2DEV”

993
6430F–ATARM–21-Feb-12

39.7.10
Name:

UDPHS Features Register
UDPHS_IPFEATURES

Address:

0x400A40F8

Access:

Read-only

31
ISO_EPT_15

30
ISO_EPT_14

29
ISO_EPT_13

28
ISO_EPT_12

27
ISO_EPT_11

26
ISO_EPT_10

25
ISO_EPT_9

24
ISO_EPT_8

23
ISO_EPT_7

22
ISO_EPT_6

21
ISO_EPT_5

20
ISO_EPT_4

19
ISO_EPT_3

18
ISO_EPT_2

17
ISO_EPT_1

16
DATAB16_8

15
BW_DPRAM

13
FIFO_MAX_SIZE

10
9
DMA_FIFO_WORD_DEPTH

7
DMA_B_SIZ

5
DMA_CHANNEL_NBR

2
1
EPT_NBR_MAX

• EPT_NBR_MAX: Max Number of Endpoints
Give the max number of endpoints.
0 = if 16 endpoints are hardware implemented.
1 = if 1 endpoint is hardware implemented.
2 = if 2 endpoints are hardware implemented.
...
15 = if 15 endpoints are hardware implemented.
• DMA_CHANNEL_NBR: Number of DMA Channels
Give the number of DMA channels.
1 = if 1 DMA channel is hardware implemented.
2 = if 2 DMA channels are hardware implemented.
...
7 = if 7 DMA channels are hardware implemented.
• DMA_B_SIZ: DMA Buffer Size
0 = if the DMA Buffer size is 16 bits.
1 = if the DMA Buffer size is 24 bits.
• DMA_FIFO_WORD_DEPTH: DMA FIFO Depth in Words
0 = if FIFO is 16 words deep.
1 = if FIFO is 1 word deep.
2 = if FIFO is 2 words deep.
...
15 = if FIFO is 15 words deep.

994

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• FIFO_MAX_SIZE: DPRAM Size
0 = if DPRAM is 128 bytes deep.
1 = if DPRAM is 256 bytes deep.
2 = if DPRAM is 512 bytes deep.
3 = if DPRAM is 1024 bytes deep.
4 = if DPRAM is 2048 bytes deep.
5 = if DPRAM is 4096 bytes deep.
6 = if DPRAM is 8192 bytes deep.
7 = if DPRAM is 16384 bytes deep.
• BW_DPRAM: DPRAM Byte Write Capability
0 = if DPRAM Write Data Shadow logic is implemented.
1 = if DPRAM is byte write capable.
• DATAB16_8: UTMI DataBus16_8
0 = if the UTMI uses an 8-bit parallel data interface (60 MHz, unidirectional).
1 = if the UTMI uses a 16-bit parallel data interface (30 MHz, bidirectional).
• ISO_EPT_x: Endpointx High Bandwidth Isochronous Capability
0 = if the endpoint does not have isochronous High Bandwidth Capability.
1 = if the endpoint has isochronous High Bandwidth Capability.

995
6430F–ATARM–21-Feb-12

39.7.11
Name:

UDPHS Endpoint Configuration Register
UDPHS_EPTCFGx [x=0..6]

Address:

0x400A4100 [0], 0x400A4120 [1], 0x400A4140 [2], 0x400A4160 [3], 0x400A4180 [4], 0x400A41A0 [5],
0x400A41C0 [6]

Access:

Read-write

31
EPT_MAPD

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

3
EPT_DIR

1
EPT_SIZE

7
BK_NUMBER

EPT_TYPE

NB_TRANS
0

• EPT_SIZE: Endpoint Size
Set this field according to the endpoint size in bytes (see Section 39.6.5 ”Endpoint Configuration”).
Endpoint Size (1)
Value

Note:

Name

Description

8 bytes

16 bytes

32 bytes

64 bytes

128

128 bytes

256

256 bytes

512

512 bytes

1024

1024 bytes

1. 1024 bytes is only for isochronous endpoint.

• EPT_DIR: Endpoint Direction
0 = Clear this bit to configure OUT direction for Bulk, Interrupt and Isochronous endpoints.
1 = set this bit to configure IN direction for Bulk, Interrupt and Isochronous endpoints.
For Control endpoints this bit has no effect and should be left at zero.
• EPT_TYPE: Endpoint Type
Set this field according to the endpoint type (see Section 39.6.5 ”Endpoint Configuration”).
(Endpoint 0 should always be configured as control)

996

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Endpoint Type
Value

Name

Description

CTRL8

Control endpoint

ISO

Isochronous endpoint

BULK

Bulk endpoint

INT

Interrupt endpoint

• BK_NUMBER: Number of Banks
Set this field according to the endpoint’s number of banks (see Section 39.6.5 ”Endpoint Configuration”).
Number of Banks
Value

Name

Description

Zero bank, the endpoint is not mapped in memory

One bank (bank 0)

Double bank (Ping-Pong: bank0/bank1)

Triple bank (bank0/bank1/bank2)

• NB_TRANS: Number Of Transaction per Microframe
The Number of transactions per microframe is set by software.
Note:

Meaningful for high bandwidth isochronous endpoint only.

• EPT_MAPD: Endpoint Mapped
0 = the user should reprogram the register with correct values.
1 = set by hardware when the endpoint size (EPT_SIZE) and the number of banks (BK_NUMBER) are correct regarding:
– the fifo max capacity (FIFO_MAX_SIZE in UDPHS_IPFEATURES register)
– the number of endpoints/banks already allocated
– the number of allowed banks for this endpoint

997
6430F–ATARM–21-Feb-12

39.7.12
Name:

UDPHS Endpoint Control Enable Register
UDPHS_EPTCTLENBx [x=0..6]

Address:

0x400A4104 [0], 0x400A4124 [1], 0x400A4144 [2], 0x400A4164 [3], 0x400A4184 [4], 0x400A41A4 [5],
0x400A41C4 [6]

Access:

Write-only

31
SHRT_PCKT

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
BUSY_BANK

17
–

16
–

NAK_OUT

NAK_IN/
ERR_FLUSH

13
STALL_SNT/
ERR_CRISO/
ERR_NBTRA

RX_SETUP/
ERR_FL_ISO

TX_PK_RDY/
ERR_TRANS

TX_COMPLT

RX_BK_RDY

ERR_OVFLW

7
MDATA_RX

6
DATAX_RX

4
NYET_DIS

3
INTDIS_DMA

2
–

1
AUTO_VALID

0
EPT_ENABL

5
–

For additional Information, see “UDPHS Endpoint Control Register” on page 1002.
• EPT_ENABL: Endpoint Enable
0 = no effect.
1 = enable endpoint according to the device configuration.
• AUTO_VALID: Packet Auto-Valid Enable
0 = no effect.
1 = enable this bit to automatically validate the current packet and switch to the next bank for both IN and OUT transfers.
• INTDIS_DMA: Interrupts Disable DMA
0 = no effect.
1 = If set, when an enabled endpoint-originated interrupt is triggered, the DMA request is disabled.
• NYET_DIS: NYET Disable (Only for High Speed Bulk OUT endpoints)
0 = no effect.
1 = forces an ACK response to the next High Speed Bulk OUT transfer instead of a NYET response.
• DATAX_RX: DATAx Interrupt Enable (Only for high bandwidth Isochronous OUT endpoints)
0 = no effect.
1 = enable DATAx Interrupt.
• MDATA_RX: MDATA Interrupt Enable (Only for high bandwidth Isochronous OUT endpoints)
0 = no effect.
1 = enable MDATA Interrupt.

998

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• ERR_OVFLW: Overflow Error Interrupt Enable
0 = no effect.
1 = enable Overflow Error Interrupt.
• RX_BK_RDY: Received OUT Data Interrupt Enable
0 = no effect.
1 = enable Received OUT Data Interrupt.
• TX_COMPLT: Transmitted IN Data Complete Interrupt Enable
0 = no effect.
1 = enable Transmitted IN Data Complete Interrupt.
• TX_PK_RDY/ERR_TRANS: TX Packet Ready/Transaction Error Interrupt Enable
0 = no effect.
1 = enable TX Packet Ready/Transaction Error Interrupt.
• RX_SETUP/ERR_FL_ISO: Received SETUP/Error Flow Interrupt Enable
0 = no effect.
1 = enable RX_SETUP/Error Flow ISO Interrupt.
• STALL_SNT/ERR_CRISO/ERR_NBTRA: Stall Sent /ISO CRC Error/Number of Transaction Error Interrupt Enable
0 = no effect.
1 = enable Stall Sent/Error CRC ISO/Error Number of Transaction Interrupt.
• NAK_IN/ERR_FLUSH: NAKIN/Bank Flush Error Interrupt Enable
0 = no effect.
1 = enable NAKIN/Bank Flush Error Interrupt.
• NAK_OUT: NAKOUT Interrupt Enable
0 = no effect.
1 = enable NAKOUT Interrupt.
• BUSY_BANK: Busy Bank Interrupt Enable
0 = no effect.
1 = enable Busy Bank Interrupt.
• SHRT_PCKT: Short Packet Send/Short Packet Interrupt Enable
For OUT endpoints:
0 = no effect.
1 = enable Short Packet Interrupt.
For IN endpoints:
Guarantees short packet at end of DMA Transfer if the UDPHS_DMACONTROLx register END_B_EN and
UDPHS_EPTCTLx register AUTOVALID bits are also set.

999
6430F–ATARM–21-Feb-12

39.7.13
Name:

UDPHS Endpoint Control Disable Register
UDPHS_EPTCTLDISx [x=0..6]

Address:

0x400A4108 [0], 0x400A4128 [1], 0x400A4148 [2], 0x400A4168 [3], 0x400A4188 [4], 0x400A41A8 [5],
0x400A41C8 [6]

Access:

Write-only

31
SHRT_PCKT

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
BUSY_BANK

17
–

16
–

NAK_OUT

NAK_IN/
ERR_FLUSH

13
STALL_SNT/
ERR_CRISO/
ERR_NBTRA

RX_SETUP/
ERR_FL_ISO

TX_PK_RDY/
ERR_TRANS

TX_COMPLT

RX_BK_RDY

ERR_OVFLW

7
MDATA_RX

6
DATAX_RX

4
NYET_DIS

3
INTDIS_DMA

2
–

1
AUTO_VALID

0
EPT_DISABL

5
–

For additional Information, see “UDPHS Endpoint Control Register” on page 1002.
• EPT_DISABL: Endpoint Disable
0 = no effect.
1 = disable endpoint.
• AUTO_VALID: Packet Auto-Valid Disable
0 = no effect.
1 = disable this bit to not automatically validate the current packet.
• INTDIS_DMA: Interrupts Disable DMA
0 = no effect.
1 = disable the “Interrupts Disable DMA”.
• NYET_DIS: NYET Enable (Only for High Speed Bulk OUT endpoints)
0 = no effect.
1 = let the hardware handle the handshake response for the High Speed Bulk OUT transfer.
• DATAX_RX: DATAx Interrupt Disable (Only for High Bandwidth Isochronous OUT endpoints)
0 = no effect.
1 = disable DATAx Interrupt.
• MDATA_RX: MDATA Interrupt Disable (Only for High Bandwidth Isochronous OUT endpoints)
0 = no effect.
1 = disable MDATA Interrupt.

1000

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• ERR_OVFLW: Overflow Error Interrupt Disable
0 = no effect.
1 = disable Overflow Error Interrupt.
• RX_BK_RDY: Received OUT Data Interrupt Disable
0 = no effect.
1 = disable Received OUT Data Interrupt.
• TX_COMPLT: Transmitted IN Data Complete Interrupt Disable
0 = no effect.
1 = disable Transmitted IN Data Complete Interrupt.
• TX_PK_RDY/ERR_TRANS: TX Packet Ready/Transaction Error Interrupt Disable
0 = no effect.
1 = disable TX Packet Ready/Transaction Error Interrupt.
• RX_SETUP/ERR_FL_ISO: Received SETUP/Error Flow Interrupt Disable
0 = no effect.
1 = disable RX_SETUP/Error Flow ISO Interrupt.
• STALL_SNT/ERR_CRISO/ERR_NBTRA: Stall Sent/ISO CRC Error/Number of Transaction Error Interrupt Disable
0 = no effect.
1 = disable Stall Sent/Error CRC ISO/Error Number of Transaction Interrupt.
• NAK_IN/ERR_FLUSH: NAKIN/bank flush error Interrupt Disable
0 = no effect.
1 = disable NAKIN/ Bank Flush Error Interrupt.
• NAK_OUT: NAKOUT Interrupt Disable
0 = no effect.
1 = disable NAKOUT Interrupt.
• BUSY_BANK: Busy Bank Interrupt Disable
0 = no effect.
1 = disable Busy Bank Interrupt.
• SHRT_PCKT: Short Packet Interrupt Disable
For OUT endpoints:
0 = no effect.
1 = disable Short Packet Interrupt.
For IN endpoints:
Never automatically add a zero length packet at end of DMA transfer.

1001
6430F–ATARM–21-Feb-12

39.7.14
Name:

UDPHS Endpoint Control Register
UDPHS_EPTCTLx [x=0..6]

Address:

0x400A410C [0], 0x400A412C [1], 0x400A414C [2], 0x400A416C [3], 0x400A418C [4], 0x400A41AC [5],
0x400A41CC [6]

Access:

Read-only

31
SHRT_PCKT

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
BUSY_BANK

17
–

16
–

NAK_OUT

NAK_IN/
ERR_FLUSH

13
STALL_SNT/
ERR_CRISO/
ERR_NBTRA

RX_SETUP/
ERR_FL_ISO

TX_PK_RDY/
ERR_TRANS

TX_COMPLT

RX_BK_RDY

ERR_OVFLW

7
MDATA_RX

6
DATAX_RX

4
NYET_DIS

3
INTDIS_DMA

2
–

1
AUTO_VALID

0
EPT_ENABL

5
–

• EPT_ENABL: Endpoint Enable
0 = If cleared, the endpoint is disabled according to the device configuration. Endpoint 0 should always be enabled after a
hardware or UDPHS bus reset and participate in the device configuration.
1 = If set, the endpoint is enabled according to the device configuration.
• AUTO_VALID: Packet Auto-Valid Enabled (Not for CONTROL Endpoints)
Set this bit to automatically validate the current packet and switch to the next bank for both IN and OUT endpoints.
For IN Transfer:
If this bit is set, then the UDPHS_EPTSTAx register TX_PK_RDY bit is set automatically when the current bank is full
and at the end of DMA buffer if the UDPHS_DMACONTROLx register END_B_EN bit is set.
The user may still set the UDPHS_EPTSTAx register TX_PK_RDY bit if the current bank is not full, unless the user
wants to send a Zero Length Packet by software.
For OUT Transfer:
If this bit is set, then the UDPHS_EPTSTAx register RX_BK_RDY bit is automatically reset for the current bank when
the last packet byte has been read from the bank FIFO or at the end of DMA buffer if the UDPHS_DMACONTROLx
register END_B_EN bit is set. For example, to truncate a padded data packet when the actual data transfer size is
reached.
The user may still clear the UDPHS_EPTSTAx register RX_BK_RDY bit, for example, after completing a DMA buffer
by software if UDPHS_DMACONTROLx register END_B_EN bit was disabled or in order to cancel the read of the
remaining data bank(s).
• INTDIS_DMA: Interrupt Disables DMA
If set, when an enabled endpoint-originated interrupt is triggered, the DMA request is disabled regardless of the
UDPHS_IEN register EPT_x bit for this endpoint. Then, the firmware will have to clear or disable the interrupt source or
clear this bit if transfer completion is needed.

1002

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
If the exception raised is associated with the new system bank packet, then the previous DMA packet transfer is normally
completed, but the new DMA packet transfer is not started (not requested).
If the exception raised is not associated to a new system bank packet (NAK_IN, NAK_OUT, ERR_FL_ISO...), then the
request cancellation may happen at any time and may immediately stop the current DMA transfer.
This may be used, for example, to identify or prevent an erroneous packet to be transferred into a buffer or to complete a
DMA buffer by software after reception of a short packet, or to perform buffer truncation on ERR_FL_ISO interrupt for
adaptive rate.
• NYET_DIS: NYET Disable (Only for High Speed Bulk OUT endpoints)
0 = If clear, this bit lets the hardware handle the handshake response for the High Speed Bulk OUT transfer.
1 = If set, this bit forces an ACK response to the next High Speed Bulk OUT transfer instead of a NYET response.
Note:

According to the Universal Serial Bus Specification, Rev 2.0 (8.5.1.1 NAK Responses to OUT/DATA During PING Protocol), a
NAK response to an HS Bulk OUT transfer is expected to be an unusual occurrence.

• DATAX_RX: DATAx Interrupt Enabled (Only for High Bandwidth Isochronous OUT endpoints)
0 = no effect.
1 = send an interrupt when a DATA2, DATA1 or DATA0 packet has been received meaning the whole microframe data
payload has been received.
• MDATA_RX: MDATA Interrupt Enabled (Only for High Bandwidth Isochronous OUT endpoints)
0 = no effect.
1 = send an interrupt when an MDATA packet has been received and so at least one packet of the microframe data payload has been received.
• ERR_OVFLW: Overflow Error Interrupt Enabled
0 = Overflow Error Interrupt is masked.
1 = Overflow Error Interrupt is enabled.
• RX_BK_RDY: Received OUT Data Interrupt Enabled
0 = Received OUT Data Interrupt is masked.
1 = Received OUT Data Interrupt is enabled.
• TX_COMPLT: Transmitted IN Data Complete Interrupt Enabled
0 = Transmitted IN Data Complete Interrupt is masked.
1 = Transmitted IN Data Complete Interrupt is enabled.
• TX_PK_RDY/ERR_TRANS: TX Packet Ready/Transaction Error Interrupt Enabled
0 = TX Packet Ready/Transaction Error Interrupt is masked.
1 = TX Packet Ready/Transaction Error Interrupt is enabled.
Caution: Interrupt source is active as long as the corresponding UDPHS_EPTSTAx register TX_PK_RDY flag remains
low. If there are no more banks available for transmitting after the software has set UDPHS_EPTSTAx/TX_PK_RDY
for the last transmit packet, then the interrupt source remains inactive until the first bank becomes free again to transmit
at UDPHS_EPTSTAx/TX_PK_RDY hardware clear.

1003
6430F–ATARM–21-Feb-12

• RX_SETUP/ERR_FL_ISO: Received SETUP/Error Flow Interrupt Enabled
0 = Received SETUP/Error Flow Interrupt is masked.
1 = Received SETUP/Error Flow Interrupt is enabled.
• STALL_SNT/ERR_CRISO/ERR_NBTRA: Stall Sent/ISO CRC Error/Number of Transaction Error Interrupt Enabled
0 = Stall Sent/ISO CRC error/number of Transaction Error Interrupt is masked.
1 = Stall Sent /ISO CRC error/number of Transaction Error Interrupt is enabled.
• NAK_IN/ERR_FLUSH: NAKIN/Bank Flush Error Interrupt Enabled
0 = NAKIN Interrupt is masked.
1 = NAKIN/Bank Flush Error Interrupt is enabled.
• NAK_OUT: NAKOUT Interrupt Enabled
0 = NAKOUT Interrupt is masked.
1 = NAKOUT Interrupt is enabled.
• BUSY_BANK: Busy Bank Interrupt Enabled
0 = BUSY_BANK Interrupt is masked.
1 = BUSY_BANK Interrupt is enabled.
For OUT endpoints: an interrupt is sent when all banks are busy.
For IN endpoints: an interrupt is sent when all banks are free.
• SHRT_PCKT: Short Packet Interrupt Enabled
For OUT endpoints: send an Interrupt when a Short Packet has been received.
0 = Short Packet Interrupt is masked.
1 = Short Packet Interrupt is enabled.
For IN endpoints: a Short Packet transmission is guaranteed upon end of the DMA Transfer, thus signaling a BULK or
INTERRUPT end of transfer or an end of isochronous (micro-)frame data, but only if the UDPHS_DMACONTROLx
register END_B_EN and UDPHS_EPTCTLx register AUTO_VALID bits are also set.

1004

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.7.15
Name:

UDPHS Endpoint Set Status Register
UDPHS_EPTSETSTAx [x=0..6]

Address:

0x400A4114 [0], 0x400A4134 [1], 0x400A4154 [2], 0x400A4174 [3], 0x400A4194 [4], 0x400A41B4 [5],
0x400A41D4 [6]

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
TX_PK_RDY

10
–

9
KILL_BANK

8
–

7
–

6
–

5
FRCESTALL

4
–

3
–

2
–

1
–

0
–

• FRCESTALL: Stall Handshake Request Set
0 = no effect.
1 = set this bit to request a STALL answer to the host for the next handshake
Refer to chapters 8.4.5 (Handshake Packets) and 9.4.5 (Get Status) of the Universal Serial Bus Specification, Rev 2.0 for
more information on the STALL handshake.
• KILL_BANK: KILL Bank Set (for IN Endpoint)
0 = no effect.
1 = kill the last written bank.
• TX_PK_RDY: TX Packet Ready Set
0 = no effect.
1 = set this bit after a packet has been written into the endpoint FIFO for IN data transfers
– This flag is used to generate a Data IN transaction (device to host).
– Device firmware checks that it can write a data payload in the FIFO, checking that TX_PK_RDY is cleared.
– Transfer to the FIFO is done by writing in the “Buffer Address” register.
– Once the data payload has been transferred to the FIFO, the firmware notifies the UDPHS device setting
TX_PK_RDY to one.
– UDPHS bus transactions can start.
– TXCOMP is set once the data payload has been received by the host.
– Data should be written into the endpoint FIFO only after this bit has been cleared.
– Set this bit without writing data to the endpoint FIFO to send a Zero Length Packet.

1005
6430F–ATARM–21-Feb-12

39.7.16
Name:

UDPHS Endpoint Clear Status Register
UDPHS_EPTCLRSTAx [x=0..6]

Address:

0x400A4118 [0], 0x400A4138 [1], 0x400A4158 [2], 0x400A4178 [3], 0x400A4198 [4], 0x400A41B8 [5],
0x400A41D8 [6]

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

14
NAK_IN/
ERR_FLUSH

13
STALL_SNT/
ERR_NBTRA

12
RX_SETUP/
ERR_FL_ISO

–

TX_COMPLT

RX_BK_RDY

–

6
TOGGLESQ

5
FRCESTALL

4
–

3
–

2
–

1
–

0
–

NAK_OUT

7
–

• FRCESTALL: Stall Handshake Request Clear
0 = no effect.
1 = clear the STALL request. The next packets from host will not be STALLed.
• TOGGLESQ: Data Toggle Clear
0 = no effect.
1 = clear the PID data of the current bank
For OUT endpoints, the next received packet should be a DATA0.
For IN endpoints, the next packet will be sent with a DATA0 PID.
• RX_BK_RDY: Received OUT Data Clear
0 = no effect.
1 = clear the RX_BK_RDY flag of UDPHS_EPTSTAx.
• TX_COMPLT: Transmitted IN Data Complete Clear
0 = no effect.
1 = clear the TX_COMPLT flag of UDPHS_EPTSTAx.
• RX_SETUP/ERR_FL_ISO: Received SETUP/Error Flow Clear
0 = no effect.
1 = clear the RX_SETUP/ERR_FL_ISO flags of UDPHS_EPTSTAx.
• STALL_SNT/ERR_NBTRA: Stall Sent/Number of Transaction Error Clear
0 = no effect.
1 = clear the STALL_SNT/ERR_NBTRA flags of UDPHS_EPTSTAx.

1006

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• NAK_IN/ERR_FLUSH: NAKIN/Bank Flush Error Clear
0 = no effect.
1 = clear the NAK_IN/ERR_FLUSH flags of UDPHS_EPTSTAx.
• NAK_OUT: NAKOUT Clear
0 = no effect.
1 = clear the NAK_OUT flag of UDPHS_EPTSTAx.

1007
6430F–ATARM–21-Feb-12

39.7.17
Name:

UDPHS Endpoint Status Register
UDPHS_EPTSTAx [x=0..6]

Address:

0x400A411C [0], 0x400A413C [1], 0x400A415C [2], 0x400A417C [3], 0x400A419C [4], 0x400A41BC [5],
0x400A41DC [6]

Access:

Read-only

31
SHRT_PCKT

27
BYTE_COUNT

17
16
CURRENT_BANK/
CONTROL_DIR

BYTE_COUNT

NAK_OUT

NAK_IN/
ERR_FLUSH

7
6
TOGGLESQ_STA

13
STALL_SNT/
ERR_CRISO/
ERR_NBTRA

BUSY_BANK_STA

RX_SETUP/
ERR_FL_ISO

TX_PK_RDY/
ERR_TRANS

TX_COMPLT

RX_BK_RDY/
KILL_BANK

ERR_OVFLW

4
–

3
–

2
–

1
–

0
–

5
FRCESTALL

• FRCESTALL: Stall Handshake Request
0 = no effect.
1= If set a STALL answer will be done to the host for the next handshake.
This bit is reset by hardware upon received SETUP.
• TOGGLESQ_STA: Toggle Sequencing
Toggle Sequencing:
– IN endpoint: it indicates the PID Data Toggle that will be used for the next packet sent. This is not relative to
the current bank.
– CONTROL and OUT endpoint:
These bits are set by hardware to indicate the PID data of the current bank:
Value

Name

Description

DATA0

DATA1

DATA2

Data2 (only for High Bandwidth Isochronous Endpoint)

MDATA

MData (only for High Bandwidth Isochronous Endpoint)

Notes:

1. In OUT transfer, the Toggle information is meaningful only when the current bank is busy (Received OUT Data = 1).
2. These bits are updated for OUT transfer:
- a new data has been written into the current bank.
- the user has just cleared the Received OUT Data bit to switch to the next bank.
3. For High Bandwidth Isochronous Out endpoint, it is recommended to check the UDPHS_EPTSTAx/ERR_TRANS bit to know
if the toggle sequencing is correct or not.
4. This field is reset to DATA1 by the UDPHS_EPTCLRSTAx register TOGGLESQ bit, and by UDPHS_EPTCTLDISx (disable
endpoint).

1008

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• ERR_OVFLW: Overflow Error
This bit is set by hardware when a new too-long packet is received.
Example: If the user programs an endpoint 64 bytes wide and the host sends 128 bytes in an OUT transfer, then the Overflow Error bit is set.
This bit is updated at the same time as the BYTE_COUNT field.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint) and by UDPHS_EPTCTLDISx (disable endpoint).
• RX_BK_RDY/KILL_BANK: Received OUT Data/KILL Bank
– Received OUT Data: (For OUT endpoint or Control endpoint)
This bit is set by hardware after a new packet has been stored in the endpoint FIFO.
This bit is cleared by the device firmware after reading the OUT data from the endpoint.
For multi-bank endpoints, this bit may remain active even when cleared by the device firmware, this if an other packet has
been received meanwhile.
Hardware assertion of this bit may generate an interrupt if enabled by the UDPHS_EPTCTLx register RX_BK_RDY bit.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint) and by UDPHS_EPTCTLDISx (disable endpoint).
– KILL Bank: (For IN endpoint)
– the bank is really cleared or the bank is sent, BUSY_BANK_STA is decremented.
– the bank is not cleared but sent on the IN transfer, TX_COMPLT
– the bank is not cleared because it was empty. The user should wait that this bit is cleared before trying to clear
another packet.
Note:

“Kill a packet” may be refused if at the same time, an IN token is coming and the current packet is sent on the UDPHS line. In
this case, the TX_COMPLT bit is set. Take notice however, that if at least two banks are ready to be sent, there is no problem to
kill a packet even if an IN token is coming. In fact, in that case, the current bank is sent (IN transfer) and the last bank is killed.

• TX_COMPLT: Transmitted IN Data Complete
This bit is set by hardware after an IN packet has been transmitted for isochronous endpoints and after it has been
accepted (ACK’ed) by the host for Control, Bulk and Interrupt endpoints.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint), and by UDPHS_EPTCTLDISx (disable endpoint).
• TX_PK_RDY/ERR_TRANS: TX Packet Ready/Transaction Error
– TX Packet Ready:
This bit is cleared by hardware, as soon as the packet has been sent for isochronous endpoints, or after the host has
acknowledged the packet for Control, Bulk and Interrupt endpoints.
For Multi-bank endpoints, this bit may remain clear even after software is set if another bank is available to transmit.
Hardware clear of this bit may generate an interrupt if enabled by the UDPHS_EPTCTLx register TX_PK_RDY bit.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint), and by UDPHS_EPTCTLDISx (disable endpoint).
– Transaction Error: (For high bandwidth isochronous OUT endpoints) (Read-Only)
This bit is set by hardware when a transaction error occurs inside one microframe.
If one toggle sequencing problem occurs among the n-transactions (n = 1, 2 or 3) inside a microframe, then this bit is still
set as long as the current bank contains one “bad” n-transaction. (see “CURRENT_BANK/CONTROL_DIR: Current
Bank/Control Direction” on page 1011) As soon as the current bank is relative to a new “good” n-transactions, then this bit
is reset.
1009
6430F–ATARM–21-Feb-12

Notes:

1. A transaction error occurs when the toggle sequencing does not respect the Universal Serial Bus Specification, Rev 2.0
(5.9.2 High Bandwidth Isochronous endpoints) (Bad PID, missing data....)
2. When a transaction error occurs, the user may empty all the “bad” transactions by clearing the Received OUT Data flag
(RX_BK_RDY).

If this bit is reset, then the user should consider that a new n-transaction is coming.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint), and by UDPHS_EPTCTLDISx (disable endpoint).
• RX_SETUP/ERR_FL_ISO: Received SETUP/Error Flow
– Received SETUP: (for Control endpoint only)
This bit is set by hardware when a valid SETUP packet has been received from the host.
It is cleared by the device firmware after reading the SETUP data from the endpoint FIFO.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint), and by UDPHS_EPTCTLDISx (disable endpoint).
– Error Flow: (for isochronous endpoint only)
This bit is set by hardware when a transaction error occurs.
– Isochronous IN transaction is missed, the micro has no time to fill the endpoint (underflow).
– Isochronous OUT data is dropped because the bank is busy (overflow).
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint) and by UDPHS_EPTCTLDISx (disable endpoint).
• STALL_SNT/ERR_CRISO/ERR_NBTRA: Stall Sent/CRC ISO Error/Number of Transaction Error
– STALL_SNT: (for Control, Bulk and Interrupt endpoints)
This bit is set by hardware after a STALL handshake has been sent as requested by the UDPHS_EPTSTAx register
FRCESTALL bit.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint) and by UDPHS_EPTCTLDISx (disable endpoint).
– ERR_CRISO: (for Isochronous OUT endpoints) (Read-only)
This bit is set by hardware if the last received data is corrupted (CRC error on data).
This bit is updated by hardware when new data is received (Received OUT Data bit).
– ERR_NBTRA: (for High Bandwidth Isochronous IN endpoints)
This bit is set at the end of a microframe in which at least one data bank has been transmitted, if less than the number of
transactions per micro-frame banks (UDPHS_EPTCFGx register NB_TRANS) have been validated for transmission inside
this microframe.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint) and by UDPHS_EPTCTLDISx (disable endpoint).
• NAK_IN/ERR_FLUSH: NAK IN/Bank Flush Error
– NAK_IN:
This bit is set by hardware when a NAK handshake has been sent in response to an IN request from the Host.
This bit is cleared by software.
– ERR_FLUSH: (for High Bandwidth Isochronous IN endpoints)
This bit is set when flushing unsent banks at the end of a microframe.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint) and by EPT_CTL_DISx (disable endpoint).

1010

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• NAK_OUT: NAK OUT
This bit is set by hardware when a NAK handshake has been sent in response to an OUT or PING request from the Host.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint) and by EPT_CTL_DISx (disable endpoint).
• CURRENT_BANK/CONTROL_DIR: Current Bank/Control Direction
– Current Bank: (all endpoints except Control endpoint)
These bits are set by hardware to indicate the number of the current bank.
Value

Name

Description

BANK0

Bank 0 (or single bank)

BANK1

Bank 1

BANK2

Bank 2

Note:

The current bank is updated each time the user:
- Sets the TX Packet Ready bit to prepare the next IN transfer and to switch to the next bank.
- Clears the received OUT data bit to access the next bank.

This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint) and by UDPHS_EPTCTLDISx (disable endpoint).
– Control Direction: (for Control endpoint only)
0 = a Control Write is requested by the Host.
1 = a Control Read is requested by the Host.
Notes:

1. This bit corresponds with the 7th bit of the bmRequestType (Byte 0 of the Setup Data).
2. This bit is updated after receiving new setup data.

• BUSY_BANK_STA: Busy Bank Number
These bits are set by hardware to indicate the number of busy banks.
IN endpoint: it indicates the number of busy banks filled by the user, ready for IN transfer.
OUT endpoint: it indicates the number of busy banks filled by OUT transaction from the Host.
Value

Name

Description

1BUSYBANK

1 busy bank

2BUSYBANKS

2 busy banks

3BUSYBANKS

3 busy banks

• BYTE_COUNT: UDPHS Byte Count
Byte count of a received data packet.
This field is incremented after each write into the endpoint (to prepare an IN transfer).
This field is decremented after each reading into the endpoint (OUT transfer).
This field is also updated at RX_BK_RDY flag clear with the next bank.
This field is also updated at TX_PK_RDY flag set with the next bank.
This field is reset by EPT_x of UDPHS_EPTRST register.

1011
6430F–ATARM–21-Feb-12

• SHRT_PCKT: Short Packet
An OUT Short Packet is detected when the receive byte count is less than the configured UDPHS_EPTCFGx register
EPT_Size.
This bit is updated at the same time as the BYTE_COUNT field.
This bit is reset by UDPHS_EPTRST register EPT_x (reset endpoint) and by UDPHS_EPTCTLDISx (disable endpoint).

1012

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.7.18

UDPHS DMA Channel Transfer Descriptor
The DMA channel transfer descriptor is loaded from the memory.
Be careful with the alignment of this buffer.
The structure of the DMA channel transfer descriptor is defined by three parameters as
described below:
Offset 0:
The address must be aligned: 0xXXXX0
Next Descriptor Address Register: UDPHS_DMANXTDSCx
Offset 4:
The address must be aligned: 0xXXXX4
DMA Channelx Address Register: UDPHS_DMAADDRESSx
Offset 8:
The address must be aligned: 0xXXXX8
DMA Channelx Control Register: UDPHS_DMACONTROLx
To use the DMA channel transfer descriptor, fill the structures with the correct value (as
described in the following pages).
Then write directly in UDPHS_DMANXTDSCx the address of the descriptor to be used first.
Then write 1 in the LDNXT_DSC bit of UDPHS_DMACONTROLx (load next channel transfer
descriptor). The descriptor is automatically loaded upon Endpointx request for packet transfer.

1013
6430F–ATARM–21-Feb-12

39.7.19
Name:

UDPHS DMA Next Descriptor Address Register
UDPHS_DMANXTDSCx [x = 0..5]

Address:

0x400A4300 [0], 0x400A4310 [1], 0x400A4320 [2], 0x400A4330 [3], 0x400A4340 [4], 0x400A4350 [5]

Access:

Read-write

Note:

28
27
NXT_DSC_ADD

20
19
NXT_DSC_ADD

12
11
NXT_DSC_ADD

4
3
NXT_DSC_ADD

Channel 0 is not used.

• NXT_DSC_ADD
This field points to the next channel descriptor to be processed. This channel descriptor must be aligned, so bits 0 to 3 of
the address must be equal to zero.

1014

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.7.20
Name:

UDPHS DMA Channel Address Register
UDPHS_DMAADDRESSx [x = 0..5]

Address:

0x400A4304 [0], 0x400A4314 [1], 0x400A4324 [2], 0x400A4334 [3], 0x400A4344 [4], 0x400A4354 [5]

Access:

Read-write

BUFF_ADD
23

20
BUFF_ADD

12
BUFF_ADD

4
BUFF_ADD

Note:

Channel 0 is not used.

• BUFF_ADD
This field determines the AHB bus starting address of a DMA channel transfer.
Channel start and end addresses may be aligned on any byte boundary.
The firmware may write this field only when the UDPHS_DMASTATUS register CHANN_ENB bit is clear.
This field is updated at the end of the address phase of the current access to the AHB bus. It is incrementing of the access
byte width. The access width is 4 bytes (or less) at packet start or end, if the start or end address is not aligned on a word
boundary.
The packet start address is either the channel start address or the next channel address to be accessed in the channel
buffer.
The packet end address is either the channel end address or the latest channel address accessed in the channel buffer.
The channel start address is written by software or loaded from the descriptor, whereas the channel end address is either
determined by the end of buffer or the UDPHS device, USB end of transfer if the UDPHS_DMACONTROLx register
END_TR_EN bit is set.

1015
6430F–ATARM–21-Feb-12

39.7.21
Name:

UDPHS DMA Channel Control Register
UDPHS_DMACONTROLx [x = 0..5]

Address:

0x400A4308 [0], 0x400A4318 [1], 0x400A4328 [2], 0x400A4338 [3], 0x400A4348 [4], 0x400A4358 [5]

Access:

Read-write

28
27
BUFF_LENGTH

20
19
BUFF_LENGTH

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
BURST_LCK

6
DESC_LD_IT

5
END_BUFFIT

4
END_TR_IT

3
END_B_EN

2
END_TR_EN

1
LDNXT_DSC

0
CHANN_ENB

Note:

Channel 0 is not used.

• CHANN_ENB (Channel Enable Command)
0 = DMA channel is disabled at and no transfer will occur upon request. This bit is also cleared by hardware when the channel source bus is disabled at end of buffer.
If the UDPHS_DMACONTROL register LDNXT_DSC bit has been cleared by descriptor loading, the firmware will have to
set the corresponding CHANN_ENB bit to start the described transfer, if needed.
If the UDPHS_DMACONTROL register LDNXT_DSC bit is cleared, the channel is frozen and the channel registers may
then be read and/or written reliably as soon as both UDPHS_DMASTATUS register CHANN_ENB and CHANN_ACT flags
read as 0.
If a channel request is currently serviced when this bit is cleared, the DMA FIFO buffer is drained until it is empty, then the
UDPHS_DMASTATUS register CHANN_ENB bit is cleared.
If the LDNXT_DSC bit is set at or after this bit clearing, then the currently loaded descriptor is skipped (no data transfer
occurs) and the next descriptor is immediately loaded.
1 = UDPHS_DMASTATUS register CHANN_ENB bit will be set, thus enabling DMA channel data transfer. Then any pending request will start the transfer. This may be used to start or resume any requested transfer.
• LDNXT_DSC: Load Next Channel Transfer Descriptor Enable (Command)
0 = no channel register is loaded after the end of the channel transfer.
1 = the channel controller loads the next descriptor after the end of the current transfer, i.e. when the
UDPHS_DMASTATUS/CHANN_ENB bit is reset.
If the UDPHS_DMA CONTROL/CHANN_ENB bit is cleared, the next descriptor is immediately loaded upon transfer
request.

1016

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
DMA Channel Control Command Summary
LDNXT_DSC

CHANN_ENB

Description

Stop now

Run and stop at end of buffer

Load next descriptor now

Run and link at end of buffer

• END_TR_EN: End of Transfer Enable (Control)
Used for OUT transfers only.
0 = USB end of transfer is ignored.
1 = UDPHS device can put an end to the current buffer transfer.
When set, a BULK or INTERRUPT short packet or the last packet of an ISOCHRONOUS (micro) frame (DATAX) will close
the current buffer and the UDPHS_DMASTATUSx register END_TR_ST flag will be raised.
This is intended for UDPHS non-prenegotiated end of transfer (BULK or INTERRUPT) or ISOCHRONOUS microframe
data buffer closure.
• END_B_EN: End of Buffer Enable (Control)
0 = DMA Buffer End has no impact on USB packet transfer.
1 = endpoint can validate the packet (according to the values programmed in the UDPHS_EPTCTLx register AUTO_VALID
and SHRT_PCKT fields) at DMA Buffer End, i.e. when the UDPHS_DMASTATUS register BUFF_COUNT reaches 0.
This is mainly for short packet IN validation initiated by the DMA reaching end of buffer, but could be used for OUT packet
truncation (discarding of unwanted packet data) at the end of DMA buffer.
• END_TR_IT: End of Transfer Interrupt Enable
0 = UDPHS device initiated buffer transfer completion will not trigger any interrupt at UDPHS_STATUSx/END_TR_ST
rising.
1 = an interrupt is sent after the buffer transfer is complete, if the UDPHS device has ended the buffer transfer.
Use when the receive size is unknown.
• END_BUFFIT: End of Buffer Interrupt Enable
0 = UDPHS_DMA_STATUSx/END_BF_ST rising will not trigger any interrupt.
1 = an interrupt is generated when the UDPHS_DMASTATUSx register BUFF_COUNT reaches zero.
• DESC_LD_IT: Descriptor Loaded Interrupt Enable
0 = UDPHS_DMASTATUSx/DESC_LDST rising will not trigger any interrupt.
1 = an interrupt is generated when a descriptor has been loaded from the bus.
• BURST_LCK: Burst Lock Enable
0 = the DMA never locks bus access.
1 = USB packets AHB data bursts are locked for maximum optimization of the bus bandwidth usage and maximization of
fly-by AHB burst duration.

1017
6430F–ATARM–21-Feb-12

• BUFF_LENGTH: Buffer Byte Length (Write-only)
This field determines the number of bytes to be transferred until end of buffer. The maximum channel transfer size (64
KBytes) is reached when this field is 0 (default value). If the transfer size is unknown, this field should be set to 0, but the
transfer end may occur earlier under UDPHS device control.
When this field is written, The UDPHS_DMASTATUSx register BUFF_COUNT field is updated with the write value.
Notes:

1. Bits [31:2] are only writable when issuing a channel Control Command other than “Stop Now”.
2. For reliability it is highly recommended to wait for both UDPHS_DMASTATUSx register CHAN_ACT and CHAN_ENB flags
are at 0, thus ensuring the channel has been stopped before issuing a command other than “Stop Now”.

1018

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
39.7.22
Name:

UDPHS DMA Channel Status Register
UDPHS_DMASTATUSx [x = 0..5]

Address:

0x400A430C [0], 0x400A431C [1], 0x400A432C [2], 0x400A433C [3], 0x400A434C [4], 0x400A435C [5]

Access:

Read-write

Note:

28
27
BUFF_COUNT

20
19
BUFF_COUNT

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
DESC_LDST

5
END_BF_ST

4
END_TR_ST

3
–

2
–

1
CHANN_ACT

0
CHANN_ENB

Channel 0 is not used.

• CHANN_ENB: Channel Enable Status
0 = if cleared, the DMA channel no longer transfers data, and may load the next descriptor if the UDPHS_DMACONTROLx
register LDNXT_DSC bit is set.
When any transfer is ended either due to an elapsed byte count or a UDPHS device initiated transfer end, this bit is automatically reset.
1 = if set, the DMA channel is currently enabled and transfers data upon request.
This bit is normally set or cleared by writing into the UDPHS_DMACONTROLx register CHANN_ENB bit field either by software or descriptor loading.
If a channel request is currently serviced when the UDPHS_DMACONTROLx register CHANN_ENB bit is cleared, the
DMA FIFO buffer is drained until it is empty, then this status bit is cleared.
• CHANN_ACT: Channel Active Status
0 = the DMA channel is no longer trying to source the packet data.
When a packet transfer is ended this bit is automatically reset.
1 = the DMA channel is currently trying to source packet data, i.e. selected as the highest-priority requesting channel.
When a packet transfer cannot be completed due to an END_BF_ST, this flag stays set during the next channel descriptor
load (if any) and potentially until UDPHS packet transfer completion, if allowed by the new descriptor.
• END_TR_ST: End of Channel Transfer Status
0 = cleared automatically when read by software.
1 = set by hardware when the last packet transfer is complete, if the UDPHS device has ended the transfer.
Valid until the CHANN_ENB flag is cleared at the end of the next buffer transfer.

1019
6430F–ATARM–21-Feb-12

• END_BF_ST: End of Channel Buffer Status
0 = cleared automatically when read by software.
1 = set by hardware when the BUFF_COUNT downcount reach zero.
Valid until the CHANN_ENB flag is cleared at the end of the next buffer transfer.
• DESC_LDST: Descriptor Loaded Status
0 = cleared automatically when read by software.
1 = set by hardware when a descriptor has been loaded from the system bus.
Valid until the CHANN_ENB flag is cleared at the end of the next buffer transfer.
• BUFF_COUNT: Buffer Byte Count
This field determines the current number of bytes still to be transferred for this buffer.
This field is decremented from the AHB source bus access byte width at the end of this bus address phase.
The access byte width is 4 by default, or less, at DMA start or end, if the start or end address is not aligned on a word
boundary.
At the end of buffer, the DMA accesses the UDPHS device only for the number of bytes needed to complete it.
This field value is reliable (stable) only if the channel has been stopped or frozen (UDPHS_EPTCTLx register
NT_DIS_DMA bit is used to disable the channel request) and the channel is no longer active CHANN_ACT flag is 0.

Note:

1020

For OUT endpoints, if the receive buffer byte length (BUFF_LENGTH) has been defaulted to zero because the USB transfer
length is unknown, the actual buffer byte length received will be 0x10000-BUFF_COUNT.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
40. DMA Controller (DMAC)
40.1

Description
The DMA Controller (DMAC) is an AHB-central DMA controller core that transfers data from a
source peripheral to a destination peripheral over one or more AMBA buses. One channel is
required for each source/destination pair. In the most basic configuration, the DMAC has one
master interface and one channel. The master interface reads the data from a source and writes
it to a destination. Two AMBA transfers are required for each DMAC data transfer. This is also
known as a dual-access transfer.
The DMAC is programmed via the APB interface.
The DMAC embeds four channels:
DMAC Channel Number

FIFO Size

8 Bytes

32 Bytes

For hardware interface numbers, see Table 40-2, “Register Mapping,” on page 1040.

1021
6430F–ATARM–21-Feb-12

40.2

Block Diagram

Figure 40-1. DMA Controller (DMAC) Block Diagram

DMAC Channel n
DMA Destination

APB Interface

DMAC Channel 2

Status
Registers

DMAC Channel 1
DMAC Channel 0
DMAC Channel 0
Write data path
to destination

DMAC
APB
Interface

Configuration
Registers

DMAC Destination
Control State Machine
Destination Pointer
Management

DMAC Interrupt
Controller

DMAC Interrupt

DMA FIFO Controller

DMA FIFO

Trigger Manager
External
Triggers

Soft
Triggers

DMAC Channel 0
Read data path
from source

DMAC Read
Datapath Bundles

DMAC
REQ/ACK
Interface

DMAC
Hardware
Handshaking
Interface

DMAC Source
Control State Machine
Source Pointer
Management

DMA Source
Requests Pool

DMAC Write
Datapath Bundles
DMA Global Control
and Data Mux

DMA Global
Request Arbiter

DMAC AHB Lite Master Interface

AMBA AHB

40.3
40.3.1

Functional Description
Basic Definitions
Source peripheral: Device on an AMBA layer from where the DMAC reads data, which is then
stored in the channel FIFO. The source peripheral teams up with a destination peripheral to form
a channel.
Destination peripheral: Device to which the DMAC writes the stored data from the FIFO (previously read from the source peripheral).
Memory: Source or destination that is always “ready” for a DMAC transfer and does not require
a handshaking interface to interact with the DMAC.
Channel: Read/write datapath between a source peripheral on one configured AMBA layer and
a destination peripheral on the same or different AMBA layer that occurs through the channel

1022

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
FIFO. If the source peripheral is not memory, then a source handshaking interface is assigned to
the channel. If the destination peripheral is not memory, then a destination handshaking interface is assigned to the channel. Source and destination handshaking interfaces can be assigned
dynamically by programming the channel registers.
Master interface: DMAC is a master on the AHB bus reading data from the source and writing it
to the destination over the AHB bus.
Slave interface: The APB interface over which the DMAC is programmed. The slave interface
in practice could be on the same layer as any of the master interfaces or on a separate layer.
Handshaking interface: A set of signal registers that conform to a protocol and handshake
between the DMAC and source or destination peripheral to control the transfer of a single or
chunk transfer between them. This interface is used to request, acknowledge, and control a
DMAC transaction. A channel can receive a request through one of two types of handshaking
interface: hardware or software.
Hardware handshaking interface: Uses hardware signals to control the transfer of a single or
chunk transfer between the DMAC and the source or destination peripheral.
Software handshaking interface: Uses software registers to contr5ol the transfer of a single or
chunk transfer between the DMAC and the source or destination peripheral. No special DMAC
handshaking signals are needed on the I/O of the peripheral. This mode is useful for interfacing
an existing peripheral to the DMAC without modifying it.
Flow controller: The device (either the DMAC or source/destination peripheral) that determines
the length of and terminates a DMAC buffer transfer. If the length of a buffer is known before
enabling the channel, then the DMAC should be programmed as the flow controller.
Transfer hierarchy: Figure 40-2 on page 1023 illustrates the hierarchy between DMAC transfers, buffer transfers, chunk or single, and AMBA transfers (single or burst) for non-memory
peripherals. Figure 40-3 on page 1023 shows the transfer hierarchy for memory.
Figure 40-2. DMAC Transfer Hierarchy for Non-Memory Peripheral
HDMA Transfer

Buffer

Chunk
Transfer

AMBA
Burst
Transfer

DMA Transfer
Level

Buffer

Chunk
Transfer

AMBA
Burst
Transfer

Buffer Transfer
Level

Buffer

Chunk
Transfer

AMBA
Burst
Transfer

AMBA
Single
Transfer

Single
Transfer

DMA Transaction
Level

AMBA
Single
Transfer

AMBA Transfer
Level

Figure 40-3. DMAC Transfer Hierarchy for Memory

1023
6430F–ATARM–21-Feb-12

HDMA Transfer

Buffer

AMBA
Burst
Transfer

Buffer

AMBA
Burst
Transfer

DMA Transfer
Level

Buffer

AMBA
Burst
Transfer

AMBA
Single
Transfer

Buffer Transfer
Level

AMBA Transfer
Level

Buffer: A buffer of DMAC data. The amount of data (length) is determined by the flow controller.
For transfers between the DMAC and memory, a buffer is broken directly into a sequence of
AMBA bursts and AMBA single transfers.
For transfers between the DMAC and a non-memory peripheral, a buffer is broken into a
sequence of DMAC transactions (single and chunks). These are in turn broken into a sequence
of AMBA transfers.
Transaction: A basic unit of a DMAC transfer as determined by either the hardware or software
handshaking interface. A transaction is only relevant for transfers between the DMAC and a
source or destination peripheral if the source or destination peripheral is a non-memory device.
There are two types of transactions: single transfer and chunk transfer.
– Single transfer: The length of a single transaction is always 1 and is converted to a
single AMBA access.
– Chunk transfer: The length of a chunk is programmed into the DMAC. The chunk is
then converted into a sequence of AHB access.DMAC executes each AMBA burst
transfer by performing incremental bursts that are no longer than 16 beats.
DMAC transfer: Software controls the number of buffers in a DMAC transfer. Once the DMAC
transfer has completed, then hardware within the DMAC disables the channel and can generate
an interrupt to signal the completion of the DMAC transfer. You can then re-program the channel
for a new DMAC transfer.
Single-buffer DMAC transfer: Consists of a single buffer.
Multi-buffer DMAC transfer: A DMAC transfer may consist of multiple DMAC buffers. Multi-buffer DMAC transfers are supported through buffer chaining (linked list pointers), auto-reloading of
channel registers, and contiguous buffers. The source and destination can independently select
which method to use.
– Linked lists (buffer chaining) – A descriptor pointer (DSCR) points to the location
in system memory where the next linked list item (LLI) exists. The LLI is a set of
registers that describe the next buffer (buffer descriptor) and a descriptor pointer
register. The DMAC fetches the LLI at the beginning of every buffer when buffer
chaining is enabled.
– Contiguous buffers – Where the address of the next buffer is selected to be a
continuation from the end of the previous buffer.
Channel locking: Software can program a channel to keep the AHB master interface by locking
the arbitration for the master bus interface for the duration of a DMAC transfer, buffer, or chunk.

1024

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Bus locking: Software can program a channel to maintain control of the AMBA bus by asserting
hmastlock for the duration of a DMAC transfer, buffer, or transaction (single or chunk). Channel
locking is asserted for the duration of bus locking at a minimum.
40.3.2

Memory Peripherals
Figure 40-3 on page 1023 shows the DMAC transfer hierarchy of the DMAC for a memory
peripheral. There is no handshaking interface with the DMAC, and therefore the memory peripheral can never be a flow controller. Once the channel is enabled, the transfer proceeds
immediately without waiting for a transaction request. The alternative to not having a transaction-level handshaking interface is to allow the DMAC to attempt AMBA transfers to the
peripheral once the channel is enabled. If the peripheral slave cannot accept these AMBA transfers, it inserts wait states onto the bus until it is ready; it is not recommended that more than 16
wait states be inserted onto the bus. By using the handshaking interface, the peripheral can signal to the DMAC that it is ready to transmit/receive data, and then the DMAC can access the
peripheral without the peripheral inserting wait states onto the bus.

40.3.3

Handshaking Interface
Handshaking interfaces are used at the transaction level to control the flow of single or chunk
transfers. The operation of the handshaking interface is different and depends on whether the
peripheral or the DMAC is the flow controller.
The peripheral uses the handshaking interface to indicate to the DMAC that it is ready to transfer/accept data over the AMBA bus. A non-memory peripheral can request a DMAC transfer
through the DMAC using one of two handshaking interfaces:
• Hardware handshaking
• Software handshaking
Software selects between the hardware or software handshaking interface on a per-channel
basis. Software handshaking is accomplished through memory-mapped registers, while hardware handshaking is accomplished using a dedicated handshaking interface.

40.3.3.1

Software Handshaking
When the slave peripheral requires the DMAC to perform a DMAC transaction, it communicates
this request by sending an interrupt to the CPU or interrupt controller.
The interrupt service routine then uses the software registers to initiate and control a DMAC
transaction. These software registers are used to implement the software handshaking
interface.
The SRC_H2SEL/DST_H2SEL bit in the DMAC_CFGx channel configuration register must be
set to zero to enable software handshaking.
When the peripheral is not the flow controller, then the last transaction register DMAC_LAST is
not used, and the values in these registers are ignored.

40.3.3.2

Chunk Transactions
Writing a 1 to the DMAC_CREQ[2x] register starts a source chunk transaction request, where x
is the channel number. Writing a 1 to the DMAC_CREQ[2x+1] register starts a destination chunk
transfer request, where x is the channel number.
Upon completion of the chunk transaction, the hardware clears the DMAC_CREQ[2x] or
DMAC_CREQ[2x+1].

1025
6430F–ATARM–21-Feb-12

40.3.3.3

Single Transactions
Writing a 1 to the DMAC_SREQ[2x] register starts a source single transaction request, where x
is the channel number. Writing a 1 to the DMAC_SREQ[2x+1] register starts a destination single
transfer request, where x is the channel number.
Upon completion of the chunk transaction, the hardware clears the DMAC_SREQ[x] or
DMAC_SREQ[2x+1].
Software can poll the relevant channel bit in the DMAC_CREQ[2x]/DMAC_CREQ[2x+1] and
DMAC_SREQ[x]/DMAC_SREQ[2x+1] registers. When both are 0, then either the requested
chunk or single transaction has completed.

40.3.4

DMAC Transfer Types
A DMAC transfer may consist of single or multi-buffers transfers. On successive buffers of a
multi-buffer transfer, the DMAC_SADDRx/DMAC_DADDRx registers in the DMAC are reprogrammed using either of the following methods:
• Buffer chaining using linked lists
• Contiguous address between buffers
On successive buffers of a multi-buffer transfer, the DMAC_CTRLAx and DMAC_CTRLBx registers in the DMAC are re-programmed using either of the following methods:
• Buffer chaining using linked lists
When buffer chaining, using linked lists is the multi-buffer method of choice, and on successive
buffers, the DMAC_DSCRx register in the DMAC is re-programmed using the following method:
• Buffer chaining using linked lists
A buffer descriptor (LLI) consists of following registers, DMAC_SADDRx, DMAC_DADDRx,
DMAC_DSCRx, DMAC_CTRLAx, DMAC_CTRLBx.These registers, along with the
DMAC_CFGx register, are used by the DMAC to set up and describe the buffer transfer.

40.3.4.1

Multi-buffer Transfers

40.3.4.2

Buffer Chaining Using Linked Lists
In this case, the DMAC re-programs the channel registers prior to the start of each buffer by
fetching the buffer descriptor for that buffer from system memory. This is known as an LLI
update.
DMAC buffer chaining is supported by using a Descriptor Pointer register (DMAC_DSCRx) that
stores the address in memory of the next buffer descriptor. Each buffer descriptor contains the
corresponding buffer descriptor (DMAC_SADDRx, DMAC_DADDRx, DMAC_DSCRx,
DMAC_CTRLAx DMAC_CTRLBx).
To set up buffer chaining, a sequence of linked lists must be programmed in memory.
The DMAC_SADDRx, DMAC_DADDRx, DMAC_DSCRx, DMAC_CTRLAx and DMAC_CTRLBx
registers are fetched from system memory on an LLI update. The updated content of the
DMAC_CTRLAx register is written back to memory on buffer completion. Figure 40-4 on page
1027 shows how to use chained linked lists in memory to define multi-buffer transfers using buffer chaining.
The Linked List multi-buffer transfer is initiated by programming DMAC_DSCRx with DSCRx(0)
(LLI(0) base address) and DMAC_CTRLBx register with both SRC_DSCR and DST_DSCR set

1026

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
to 0. Other fields and registers are ignored and overwritten when the descriptor is retrieved from
memory.
The last transfer descriptor must be written to memory with its next descriptor address set to 0.
Figure 40-4. Multi Buffer Transfer Using Linked List
System Memory
LLI(0)

DSCRx(0)

LLI(1)

DSCRx(1)= DSCRx(0) + 0x10

DSCRx(2)= DSCRx(1) + 0x10

CTRLBx= DSCRx(0) + 0xC

CTRLBx= DSCRx(1) + 0xC

CTRLAx= DSCRx(0) + 0x8

CTRLBx= DSCRx(1) + 0x8

DADDRx= DSCRx(0) + 0x4

DADDRx= DSCRx(1) + 0x4
SADDRx= DSCRx(1) + 0x0

SADDRx= DSCRx(0) + 0x0
DSCRx(1)

DSCRx(2)
(points to 0 if
LLI(1) is the last
transfer descriptor

1027
6430F–ATARM–21-Feb-12

40.3.4.3

Programming DMAC for Multiple Buffer Transfers

Table 40-1.

Multiple Buffers Transfer Management Table

Transfer Type

AUTO

SRC_REP

DST_REP

SRC_DSCR

DST_DSCR

BTSIZE

SADDR

DADDR

Other
Fields

1) Single Buffer or Last
buffer of a multiple buffer
transfer

–

USR

2) Multi Buffer transfer
with contiguous DADDR

–

LLI

CONT

LLI

3) Multi Buffer transfer
with contiguous SADDR

–

LLI

CONT

LLI

4) Multi Buffer transfer
with LLI support

–

LLI

Notes:

1. USR means that the register field is manually programmed by the user.
2. CONT means that address are contiguous.
3. LLI means that the register field is updated with the content of the linked list item.

40.3.4.4

Contiguous Address Between Buffers
In this case, the address between successive buffers is selected to be a continuation from the
end of the previous buffer. Enabling the source or destination address to be contiguous between
buffers is a function of DMAC_CTRLAx.SRC_DSCR and DMAC_CTRLAx.DST_DSCR
registers.

40.3.4.5

Suspension of Transfers Between buffers
At the end of every buffer transfer, an end of buffer interrupt is asserted if:
• the channel buffer interrupt is unmasked, DMAC_EBCIMR.BTC[n] = ‘1’, where n is the
channel number.
Note:

The buffer complete interrupt is generated at the completion of the buffer transfer to the
destination.

At the end of a chain of multiple buffers, an end of linked list interrupt is asserted if:
• the channel end of chained buffer interrupt is unmasked, DMAC_EBCIMR.CBTC[n] = ‘1’,
when n is the channel number.
40.3.4.6

Ending Multi-buffer Transfers
All multi-buffer transfers must end as shown in Row 1 of Table 40-1 on page 1028. At the end of
every buffer transfer, the DMAC samples the row number, and if the DMAC is in Row 1 state,
then the previous buffer transferred was the last buffer and the DMAC transfer is terminated.
For rows 2, 3, and 4 the user must setup the last buffer descriptor in memory such that both
LLI.DMAC_CTRLBx.SRC_DSCR and LLI.DMAC_CTRLBx.DST_DSCR are one and
LLI.DMAC_DSCRx is set to 0.

1028

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
40.3.5

Programming a Channel
Four registers, the DMAC_DSCRx, the DMAC_CTRLAx, the DMAC_CTRLBx and
DMAC_CFGx, need to be programmed to set up whether single or multi-buffer transfers take
place, and which type of multi-buffer transfer is used. The different transfer types are shown in
Table 40-1 on page 1028.
The “BTSIZE, SADDR and DADDR” columns indicate where the values of DMAC_SARx,
DMAC_DARx, DMAC_CTLx, and DMAC_LLPx are obtained for the next buffer transfer when
multi-buffer DMAC transfers are enabled.

40.3.5.1

Programming Examples

40.3.5.2

Single-buffer Transfer (Row 1)
1. Read the Channel Handler Status Register DMAC_CHSR.ENABLE Field to choose a
free (disabled) channel.
2. Clear any pending interrupts on the channel from the previous DMAC transfer by reading the interrupt status register, DMAC_EBCISR.
3. Program the following channel registers:
a. Write the starting source address in the DMAC_SADDRx register for channel x.
b.

Write the starting destination address in the DMAC_DADDRx register for channel
x.

Program DMAC_CTRLAx, DMAC_CTRLBx and DMAC_CFGx according to Row 1
as shown in Table 40-1 on page 1028. Program the DMAC_CTRLBx register with
both DST_DSCR and SRC_DSCR fields set to one.

d. Write the control information for the DMAC transfer in the DMAC_CTRLAx and
DMAC_CTRLBx registers for channel x. For example, in the register, you can program the following:
– i. Set up the transfer type (memory or non-memory peripheral for source and
destination) and flow control device by programming the FC of the DMAC_CTRLBx
register.
– ii. Set up the transfer characteristics, such as:
– Transfer width for the source in the SRC_WIDTH field.
– Transfer width for the destination in the DST_WIDTH field.
– Incrementing/decrementing or fixed address for source in SRC_INC field.
– Incrementing/decrementing or fixed address for destination in DST_INC field.
e. Write the channel configuration information into the DMAC_CFGx register for channel x.
– i. Designate the handshaking interface type (hardware or software) for the source
and destination peripherals. This is not required for memory. This step requires
programming the SRC_H2SEL/DST_H2SEL bits, respectively. Writing a ‘1’ activates
the hardware handshaking interface to handle source/destination requests. Writing a
‘0’ activates the software handshaking interface to handle source/destination
requests.
– ii. If the hardware handshaking interface is activated for the source or destination
peripheral, assign a handshaking interface to the source and destination peripheral.
This requires programming the SRC_PER and DST_PER bits, respectively.

1029
6430F–ATARM–21-Feb-12

4. After the DMAC selected channel has been programmed, enable the channel by writing
a ‘1’ to the DMAC_CHER.ENABLE[n] bit, where n is the channel number. Make sure
that bit 0 of DMAC_EN.ENABLE register is enabled.
5. Source and destination request single and chunk DMAC transactions to transfer the
buffer of data (assuming non-memory peripherals). The DMAC acknowledges at the
completion of every transaction (chunk and single) in the buffer and carry out the buffer
transfer.
6. Once the transfer completes, hardware sets the interrupts and disables the channel. At
this time you can either respond to the buffer Complete or Transfer Complete interrupts,
or poll for the Channel Handler Status Register (DMAC_CHSR.ENABLE[n]) bit until it is
cleared by hardware, to detect when the transfer is complete.
40.3.5.3

Multi-buffer Transfer with Linked List for Source and Linked List for Destination (Row 4)
1. Read the Channel Enable register to choose a free (disabled) channel.
2. Set up the chain of Linked List Items (otherwise known as buffer descriptors) in memory. Write the control information in the LLI.DMAC_CTRLAx and LLI.DMAC_CTRLBx
registers location of the buffer descriptor for each LLI in memory (see Figure 40-5 on
page 1032) for channel x. For example, in the register, you can program the following:
a. Set up the transfer type (memory or non-memory peripheral for source and destination) and flow control device by programming the FC of the DMAC_CTRLBx
register.
b.

Set up the transfer characteristics, such as:

– i. Transfer width for the source in the SRC_WIDTH field.
– ii. Transfer width for the destination in the DST_WIDTH field.
– v. Incrementing/decrementing or fixed address for source in SRC_INCR field.
– vi. Incrementing/decrementing or fixed address for destination DST_INCR field.
3. Write the channel configuration information into the DMAC_CFGx register for channel
x.
a. Designate the handshaking interface type (hardware or software) for the source
and destination peripherals. This is not required for memory. This step requires programming the SRC_H2SEL/DST_H2SEL bits, respectively. Writing a ‘1’ activates
the hardware handshaking interface to handle source/destination requests for the
specific channel. Writing a ‘0’ activates the software handshaking interface to handle source/destination requests.
b.

If the hardware handshaking interface is activated for the source or destination
peripheral, assign the handshaking interface to the source and destination peripheral. This requires programming the SRC_PER and DST_PER bits, respectively.

4. Make sure that the LLI.DMAC_CTRLBx register locations of all LLI entries in memory
(except the last) are set as shown in Row 4 of Table 40-1 on page 1028. The
LLI.DMAC_CTRLBx register of the last Linked List Item must be set as described in
Row 1 of Table 40-1. Figure 40-4 on page 1027 shows a Linked List example with two
list items.
5. Make sure that the LLI.DMAC_DSCRx register locations of all LLI entries in memory
(except the last) are non-zero and point to the base address of the next Linked List
Item.
6. Make sure that the LLI.DMAC_SADDRx/LLI.DMAC_DADDRx register locations of all
LLI entries in memory point to the start source/destination buffer address preceding
that LLI fetch.

1030

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
7. Make sure that the LLI.DMAC_CTRLAx.DONE field of the LLI.DMAC_CTRLAx register
locations of all LLI entries in memory are cleared.
8. If source picture-picture mode is enabled (DMAC_CTRLBx.SRC_PIP is enabled), program the DMAC_SPIPx register for channel x.
9. If destination picture-in-picture is enabled (DMAC_CTRLBx.DST_PIP is enabled), program the DMAC_DPIPx register for channel x.
10. Clear any pending interrupts on the channel from the previous DMAC transfer by reading the status register: DMAC_EBCISR.
11. Program the DMAC_CTRLBx, DMAC_CFGx registers according to Row 4 as shown in
Table 40-1 on page 1028.
12. Program the DMAC_DSCRx register with DMAC_DSCRx(0), the pointer to the first
Linked List item.
13. Finally, enable the channel by writing a ‘1’ to the DMAC_CHER.ENABLE[n] bit, where n
is the channel number. The transfer is performed.
14. The DMAC fetches the first LLI from the location pointed to by DMAC_DSCRx(0).
Note:

The LLI.DMAC_SADDRx, LLI. DMAC_DADDRx, LLI.DMAC_DSCRx, LLI.DMAC_CTRLAx and
LLI.DMAC_CTRLBx registers are fetched. The DMAC automatically reprograms the
DMAC_SADDRx, DMAC_DADDRx, DMAC_DSCRx, DMAC_CTRLBx and DMAC_CTRLAx channel registers from the DMAC_DSCRx(0).

15. Source and destination request single and chunk DMAC transactions to transfer the
buffer of data (assuming non-memory peripheral). The DMAC acknowledges at the
completion of every transaction (chunk and single) in the buffer and carry out the buffer
transfer.
16. Once the buffer of data is transferred, the DMAC_CTRLAx register is written out to system memory at the same location and on the same layer where it was originally
fetched, that is, the location of the DMAC_CTRLAx register of the linked list item
fetched prior to the start of the buffer transfer. Only DMAC_CTRLAx register is written
out because only the DMAC_CTRLAx.BTSIZE and DMAC_CTRLAX.DONE bits have
been updated by DMAC hardware. Additionally, the DMAC_CTRLAx.DONE bit is
asserted when the buffer transfer has completed.
Note:

Do not poll the DMAC_CTRLAx.DONE bit in the DMAC memory map. Instead, poll the
LLI.DMAC_CTRLAx.DONE bit in the LLI for that buffer. If the poll LLI.DMAC_CTRLAx.DONE bit is
asserted, then this buffer transfer has completed. This LLI.DMAC_CTRLAx.DONE bit was cleared
at the start of the transfer.

17. The DMAC does not wait for the buffer interrupt to be cleared, but continues fetching
the next LLI from the memory location pointed to by current DMAC_DSCRx register
and automatically reprograms the DMAC_SADDRx, DMAC_DADDRx, DMAC_DSCRx,
DMAC_CTRLAx and DMAC_CTRLBx channel registers. The DMAC transfer continues
until the DMAC determines that the DMAC_CTRLBx and DMAC_DSCRx registers at
the end of a buffer transfer match described in Row 1 of Table 40-1 on page 1028. The
DMAC then knows that the previous buffer transferred was the last buffer in the DMAC
transfer. The DMAC transfer might look like that shown in Figure 40-5 on page 1032.

1031
6430F–ATARM–21-Feb-12

Figure 40-5. Multi-buffer with Linked List Address for Source and Destination
Address of
Destination Layer

Address of
Source Layer

Buffer 2
SADDR(2)

Buffer 2
DADDR(2)

Buffer 1
SADDR(1)

Buffer 1
DADDR(1)

Buffer 0

Buffer 0
DADDR(0)

SADDR(0)
Source Buffers

Destination Buffers

If the user needs to execute a DMAC transfer where the source and destination address are
contiguous but the amount of data to be transferred is greater than the maximum buffer size
DMAC_CTRLAx.BTSIZE, then this can be achieved using the type of multi-buffer transfer as
shown in Figure 40-6 on page 1033.

1032

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 40-6. Multi-buffer with Linked Address for Source and Destination Buffers are Contiguous
Address of
Source Layer

Address of
Destination Layer

Buffer 2
DADDR(3)
Buffer 2

Buffer 2

SADDR(3)

DADDR(2)
Buffer 2

Buffer 1

SADDR(2)

DADDR(1)
Buffer 1

SADDR(1)

Buffer 0
DADDR(0)

Buffer 0
SADDR(0)
Source Buffers

Destination Buffers

The DMAC transfer flow is shown in Figure 40-7 on page 1034.

1033
6430F–ATARM–21-Feb-12

Figure 40-7. DMAC Transfer Flow for Source and Destination Linked List Address
Channel enabled by
software

LLI Fetch

Hardware reprograms
SADDRx, DADDRx, CTRLA/Bx, DSCRx

DMAC buffer transfer

Writeback of HDMA_CTRLAx
register in system memory
Buffer Complete interrupt
generated here

Is HDMA in
Row1 of
HDMA State Machine Table?

HDMA Transfer Complete
interrupt generated here

yes

Channel Disabled by
hardware

40.3.5.4

Multi-buffer DMAC Transfer with Linked List for Source and Contiguous Destination Address (Row 2)
1. Read the Channel Enable register to choose a free (disabled) channel.
2. Set up the linked list in memory. Write the control information in the
LLI.DMAC_CTRLAx and LLI.DMAC_CTRLBx register location of the buffer descriptor
for each LLI in memory for channel x. For example, in the register, you can program the
following:

1034

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
a. Set up the transfer type (memory or non-memory peripheral for source and destination) and flow control device by programming the FC of the DMAC_CTRLBx
register.
b.

Set up the transfer characteristics, such as:

The values in the LLI.DMAC_DADDRx register location of each Linked List Item (LLI) in memory,
although fetched during an LLI fetch, are not used.

4. Write the channel configuration information into the DMAC_CFGx register for channel
x.
a. Designate the handshaking interface type (hardware or software) for the source
and destination peripherals. This is not required for memory. This step requires programming the SRC_H2SEL/DST_H2SEL bits, respectively. Writing a ‘1’ activates
the hardware handshaking interface to handle source/destination requests for the
specific channel. Writing a ‘0’ activates the software handshaking interface to handle source/destination requests.
b.

If the hardware handshaking interface is activated for the source or destination
peripheral, assign handshaking interface to the source and destination peripherals.
This requires programming the SRC_PER and DST_PER bits, respectively.

5. Make sure that all LLI.DMAC_CTRLBx register locations of the LLI (except the last) are
set as shown in Row 2 of Table 40-1 on page 1028, while the LLI.DMAC_CTRLBx register of the last Linked List item must be set as described in Row 1 of Table 40-1. Figure
40-4 on page 1027 shows a Linked List example with two list items.
6. Make sure that the LLI.DMAC_DSCRx register locations of all LLIs in memory (except
the last) are non-zero and point to the next Linked List Item.
7. Make sure that the LLI.DMAC_SADDRx register location of all LLIs in memory point to
the start source buffer address proceeding that LLI fetch.
8. Make sure that the LLI.DMAC_CTRLAx.DONE field of the LLI.DMAC_CTRLAx register
locations of all LLIs in memory is cleared.
9. Clear any pending interrupts on the channel from the previous DMAC transfer by reading the interrupt status register.
10. Program the DMAC_CTRLAx, DMAC_CTRLBx and DMAC_CFGx registers according
to Row 2 as shown in Table 40-1 on page 1028
11. Program the DMAC_DSCRx register with DMAC_DSCRx(0), the pointer to the first
Linked List item.
12. Finally, enable the channel by writing a ‘1’ to the DMAC_CHER.ENABLE[n] bit. The
transfer is performed. Make sure that bit 0 of the DMAC_EN register is enabled.
13. The DMAC fetches the first LLI from the location pointed to by DMAC_DSCRx(0).
Note:

The LLI.DMAC_SADDRx, LLI.DMAC_DADDRx, LLI.DMAC_DSCRx and LLI.DMAC_CTRLA/Bx
registers are fetched. The LLI.DMAC_DADDRx register location of the LLI although fetched is not
used. The DMAC_DADDRx register in the DMAC remains unchanged.

14. Source and destination requests single and chunk DMAC transactions to transfer the
buffer of data (assuming non-memory peripherals). The DMAC acknowledges at the
completion of every transaction (chunk and single) in the buffer and carry out the buffer
transfer

1035
6430F–ATARM–21-Feb-12

15. Once the buffer of data is transferred, the DMAC_CTRLAx register is written out to system memory at the same location and on the same layer (DMAC_DSCRx.DSCR_IF)
where it was originally fetched, that is, the location of the DMAC_CTRLAx register of
the linked list item fetched prior to the start of the buffer transfer. Only DMAC_CTRLAx
register is written out because only the DMAC_CTRLAx.BTSIZE and
DMAC_CTRLAX.DONE fields have been updated by DMAC hardware. Additionally, the
DMAC_CTRLAx.DONE bit is asserted when the buffer transfer has completed.
Note:

16. The DMAC does not wait for the buffer interrupt to be cleared, but continues and
fetches the next LLI from the memory location pointed to by current DMAC_DSCRx
register and automatically reprograms the DMAC_SADDRx, DMAC_CTRLAx,
DMAC_CTRLBx and DMAC_DSCRx channel registers. The DMAC_DADDRx register
is left unchanged. The DMAC transfer continues until the DMAC samples the
DMAC_CTRLAx, DMAC_CTRLBx and DMAC_DSCRx registers at the end of a buffer
transfer match that described in Row 1 of Table 40-1 on page 1028. The DMAC then
knows that the previous buffer transferred was the last buffer in the DMAC transfer.
The DMAC transfer might look like that shown in Figure 40-8 on page 1036 Note that the destination address is decrementing.
Figure 40-8. DMAC Transfer with Linked List Source Address and Contiguous Destination Address
Address of
Source Layer

Address of
Destination Layer

Buffer 2
SADDR(2)
Buffer 2
DADDR(2)

Buffer 1

SADDR(1)

DADDR(1)

Buffer 0
DADDR(0)

Buffer 0
SADDR(0)

Source Buffers

Destination Buffers

The DMAC transfer flow is shown in Figure 40-9 on page 1037.

1036

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 40-9. DMAC Transfer Flow for Linked List Source Address and Contiguous Destination Address
Channel Enabled by
software

LLI Fetch

Hardware reprograms
SADDRx, CTRLAx,CTRLBx, DSCRx

HDMA buffer transfer

Writeback of control
information of LLI

Buffer Complete interrupt
generated here

Is HDMA in
Row 1 ?

HDMA Transfer Complete
interrupt generated here

yes

Channel Disabled by
hardware

40.3.6

Disabling a Channel Prior to Transfer Completion
Under normal operation, software enables a channel by writing a ‘1’ to the Channel Handler
Enable Register, DMAC_CHER.ENABLE[n], and hardware disables a channel on transfer completion by clearing the DMAC_CHSR.ENABLE[n] register bit.
The recommended way for software to disable a channel without losing data is to use the SUSPEND[n] bit in conjunction with the EMPTY[n] bit in the Channel Handler Status Register.

1037
6430F–ATARM–21-Feb-12

1. If software wishes to disable a channel n prior to the DMAC transfer completion, then it
can set the DMAC_CHER.SUSPEND[n] bit to tell the DMAC to halt all transfers from
the source peripheral. Therefore, the channel FIFO receives no new data.
2. Software can now poll the DMAC_CHSR.EMPTY[n] bit until it indicates that the channel
n FIFO is empty, where n is the channel number.
3. The DMAC_CHER.ENABLE[n] bit can then be cleared by software once the channel n
FIFO is empty, where n is the channel number.
When DMAC_CTRLAx.SRC_WIDTH is less than DMAC_CTRLAx.DST_WIDTH and the
DMAC_CHSRx.SUSPEND[n] bit is high, the DMAC_CHSRx.EMPTY[n] is asserted once the
contents of the FIFO do not permit a single word of DMAC_CTRLAx.DST_WIDTH to be formed.
However, there may still be data in the channel FIFO but not enough to form a single transfer of
DMAC_CTLx.DST_WIDTH width. In this configuration, once the channel is disabled, the remaining data in the channel FIFO are not transferred to the destination peripheral. It is permitted to
remove the channel from the suspension state by writing a ‘1’ to the DMAC_CHER.RESUME[n]
field register. The DMAC transfer completes in the normal manner. n defines the channel
number.
Note:

40.3.6.1

If a channel is disabled by software, an active single or chunk transaction is not guaranteed to
receive an acknowledgement.

Abnormal Transfer Termination
A DMAC transfer may be terminated abruptly by software by clearing the channel enable bit,
DMAC_CHDR.ENABLE[n] where n is the channel number. This does not mean that the channel
is disabled immediately after the DMAC_CHSR.ENABLE[n] bit is cleared over the APB interface. Consider this as a request to disable the channel. The DMAC_CHSR.ENABLE[n] must be
polled and then it must be confirmed that the channel is disabled by reading back 0.
Software may terminate all channels abruptly by clearing the global enable bit in the DMAC Configuration Register (DMAC_EN.ENABLE bit). Again, this does not mean that all channels are
disabled immediately after the DMAC_EN.ENABLE is cleared over the APB slave interface.
Consider this as a request to disable all channels. The DMAC_CHSR.ENABLE must be polled
and then it must be confirmed that all channels are disabled by reading back ‘0’.

40.4

Note:

If the channel enable bit is cleared while there is data in the channel FIFO, this data is not sent to
the destination peripheral and is not present when the channel is re-enabled. For read sensitive
source peripherals, such as a source FIFO, this data is therefore lost. When the source is not a
read sensitive device (i.e., memory), disabling a channel without waiting for the channel FIFO to
empty may be acceptable as the data is available from the source peripheral upon request and is
not lost.

Note:

If a channel is disabled by software, an active single or chunk transaction is not guaranteed to
receive an acknowledgement.

DMAC Software Requirements
• There must not be any write operation to Channel registers in an active channel after the
channel enable is made HIGH. If any channel parameters must be reprogrammed, this can
only be done after disabling the DMAC channel.
• You must program the DMAC_SADDRx and DMAC_DADDRx channel registers with a byte,
half-word and word aligned address depending on the source width and destination width.
• After the software disables a channel by writing into the channel disable register, it must reenable the channel only after it has polled a 0 in the corresponding channel enable status
register. This is because the current AHB Burst must terminate properly.

1038

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• If you program the BTSIZE field in the DMAC_CTRLA, as zero, and the DMAC is defined as
the flow controller, then the channel is automatically disabled.
• When hardware handshaking interface protocol is fully implemented, a peripheral is expected
to deassert any sreq or breq signals on receiving the ack signal irrespective of the request
the ack was asserted in response to.
• Multiple Transfers involving the same peripheral must not be programmed and enabled on
different channel, unless this peripheral integrates several hardware handshaking interface.
• When a Peripheral is flow controller, the targeted DMAC Channel must be enabled before the
Peripheral. If you do not ensure this the DMAC Channel might miss a Last Transfer Flag, if
the First DMAC request is also the last transfer.

1039
6430F–ATARM–21-Feb-12

40.5

DMA Controller (DMAC) User Interface

Table 40-2.

Offset

Name

Access

Reset

0x000

DMAC Global Configuration Register

DMAC_GCFG

Read-write

0x10

0x004

DMAC Enable Register

DMAC_EN

Read-write

0x0

0x008

DMAC Software Single Request Register

DMAC_SREQ

Read-write

0x0

0x00C

DMAC Software Chunk Transfer Request Register

DMAC_CREQ

Read-write

0x0

0x010

DMAC Software Last Transfer Flag Register

DMAC_LAST

Read-write

0x0

0x014

Reserved

–

0x018

DMAC Error, Chained Buffer transfer completed and Buffer
transfer completed Interrupt Enable register.

DMAC_EBCIER

Write-only

–

0x01C

DMAC Error, Chained Buffer transfer completed and Buffer
transfer completed Interrupt Disable register.

DMAC_EBCIDR

Write-only

–

0x020

DMAC Error, Chained Buffer transfer completed and Buffer
transfer completed Mask Register.

DMAC_EBCIMR

Read-only

0x0

0x024

DMAC Error, Chained Buffer transfer completed and Buffer
transfer completed Status Register.

DMAC_EBCISR

Read-only

0x0

0x028

DMAC Channel Handler Enable Register

DMAC_CHER

Write-only

–

0x02C

DMAC Channel Handler Disable Register

DMAC_CHDR

Write-only

–

0x030

DMAC Channel Handler Status Register

DMAC_CHSR

Read-only

0x00FF0000

0x034

Reserved

–

0x038

Reserved

–

0x03C+ch_num*(0x28)+(0x0)

DMAC Channel Source Address Register

DMAC_SADDR

Read-write

0x0

0x03C+ch_num*(0x28)+(0x4)

DMAC Channel Destination Address Register

DMAC_DADDR

Read-write

0x0

0x03C+ch_num*(0x28)+(0x8)

DMAC Channel Descriptor Address Register

DMAC_DSCR

Read-write

0x0

0x03C+ch_num*(0x28)+(0xC)

DMAC Channel Control A Register

DMAC_CTRLA

Read-write

0x0

0x03C+ch_num*(0x28)+(0x10)

DMAC Channel Control B Register

DMAC_CTRLB

Read-write

0x0

0x03C+ch_num*(0x28)+(0x14)

DMAC Channel Configuration Register

DMAC_CFG

Read-write

0x01000000

0x03C+ch_num*(0x28)+(0x18)

Reserved

–

0x03C+ch_num*(0x28)+(0x1C)

Reserved

–

0x03C+ch_num*(0x28)+(0x20)

Reserved

–

0x03C+ch_num*(0x28)+(0x24)

Reserved

–

0x064 - 0xC8

DMAC Channel 1 to 3 Register(1)

Read-write

0x0

0x017C- 0x1FC

Reserved

–

Note:

1040

–

1. The addresses for the DMAC registers shown here are for DMA Channel 0. This sequence of registers is repeated successively for each DMA channel located between 0x064 and 0xC8.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
40.5.1
Name:

DMAC Global Configuration Register
DMAC_GCFG

Address:

0x400B0000

Access:

Read-write

Reset:

0x00000010

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
ARB_CFG

3
–

2
–

1
–

0
–

• ARB_CFG
0: Fixed priority arbiter.
1: Modified round robin arbiter.

1041
6430F–ATARM–21-Feb-12

40.5.2
Name:

DMAC Enable Register
DMAC_EN

Address:

0x400B0004

Access:

Read-write

Reset: 0x00000000
31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
–

0
ENABLE

• ENABLE
0: DMA Controller is disabled.
1: DMA Controller is enabled.

1042

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
40.5.3
Name:

DMAC Software Single Request Register
DMAC_SREQ

Address:

0x400B0008

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
DSREQ3

6
SSREQ3

5
DSREQ2–

4
SSREQ2–

3
DSREQ1

2
SSREQ1

1
DSREQ0

0
SSREQ0

• DSREQx
Request a destination single transfer on channel i.
• SSREQx
Request a source single transfer on channel i.

1043
6430F–ATARM–21-Feb-12

40.5.4
Name:

DMAC Software Chunk Transfer Request Register
DMAC_CREQ

Address:

0x400B000C

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
DCREQ3

6
SCREQ3

5
DCREQ2–

4
SCREQ2–

3
DCREQ1

2
SCREQ1

1
DCREQ0

0
SCREQ0

• DCREQx
Request a destination chunk transfer on channel i.
• SCREQx
Request a source chunk transfer on channel i.

1044

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
40.5.5
Name:

DMAC Software Last Transfer Flag Register
DMAC_LAST

Address:

0x400B0010

Access:

Read-write

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
DLAST3

6
SLAST3

5
DLAST2

4
SLAST2

3
DLAST1

2
SLAST1

1
DLAST0

0
SLAST0

• DLASTx
Writing one to DLASTx prior to writing one to DSREQx or DCREQx indicates that this destination request is the last transfer
of the buffer.
• SLASTx
Writing one to SLASTx prior to writing one to SSREQx or SCREQx indicates that this source request is the last transfer of
the buffer.

1045
6430F–ATARM–21-Feb-12

40.5.6
Name:

DMAC Error, Buffer Transfer and Chained Buffer Transfer Interrupt Enable Register
DMAC_EBCIER

Address:

0x400B0018

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
ERR3

18
ERR2

17
ERR1

16
ERR0

15
–

14
–

13
–

12
–

11
CBTC3

10
CBTC2

9
CBTC1

8
CBTC0

7
–

6
–

5
–

4
–

3
BTC3

2
BTC2

1
BTC1

0
BTC0

• BTC[3:0]
Buffer Transfer Completed Interrupt Enable Register. Set the relevant bit in the BTC field to enable the interrupt for
channel i.
• CBTC[3:0]
Chained Buffer Transfer Completed Interrupt Enable Register. Set the relevant bit in the CBTC field to enable the interrupt
for channel i.
• ERR[3:0]
Access Error Interrupt Enable Register. Set the relevant bit in the ERR field to enable the interrupt for channel i.

1046

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
40.5.7
Name:

DMAC Error, Buffer Transfer and Chained Buffer Transfer Interrupt Disable Register
DMAC_EBCIDR

Address:

0x400B001C

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
ERR3

18
ERR2

17
ERR1

16
ERR0

15
–

14
–

13
–

12
–

11
CBTC3

10
CBTC2

9
CBTC1

8
CBTC0

7
–

6
–

5
–

4
–

3
BTC3

2
BTC2

1
BTC1

0
BTC0

• BTC[3:0]
Buffer transfer completed Disable Interrupt Register. When set, a bit of the BTC field disables the interrupt from the relevant DMAC channel.
• CBTC[3:0]
Chained Buffer transfer completed Disable Register. When set, a bit of the CBTC field disables the interrupt from the relevant DMAC channel.
• ERR[3:0]
Access Error Interrupt Disable Register. When set, a bit of the ERR field disables the interrupt from the relevant DMAC
channel.

1047
6430F–ATARM–21-Feb-12

40.5.8
Name:

DMAC Error, Buffer Transfer and Chained Buffer Transfer Interrupt Mask Register
DMAC_EBCIMR

Address:

0x400B0020

Access:

Read-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
ERR3

18
ERR2

17
ERR1

16
ERR0

15
–

14
–

13
–

12
–

11
CBTC3

10
CBTC2

9
CBTC1

8
CBTC0

7
–

6
–

5
–

4
–

3
BTC3

2
BTC2

1
BTC1

0
BTC0

• BTC[3:0]
0: Buffer Transfer completed interrupt is disabled for channel i.
1: Buffer Transfer completed interrupt is enabled for channel i.
• CBTC[3:0]
0: Chained Buffer Transfer interrupt is disabled for channel i.
1: Chained Buffer Transfer interrupt is enabled for channel i.
• ERR[3:0]
0: Transfer Error Interrupt is disabled for channel i.
1: Transfer Error Interrupt is enabled for channel i.

1048

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
40.5.9
Name:

DMAC Error, Buffer Transfer and Chained Buffer Transfer Status Register
DMAC_EBCISR

Address:

0x400B0024

Access:

Read-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
ERR3

18
ERR2

17
ERR1

16
ERR0

15
–

14
–

13
–

12
–

11
CBTC3

10
CBTC2

9
CBTC1

8
CBTC0

7
–

6
–

5
–

4
–

3
BTC3

2
BTC2

1
BTC1

0
BTC0

• BTC[3:0]
When BTC[i] is set, Channel i buffer transfer has terminated.
• CBTC[3:0]
When CBTC[i] is set, Channel i Chained buffer has terminated. LLI Fetch operation is disabled.
• ERR[3:0]
When ERR[i] is set, Channel i has detected an AHB Read or Write Error Access.

1049
6430F–ATARM–21-Feb-12

40.5.10
Name:

DMAC Channel Handler Enable Register
DMAC_CHER

Address:

0x400B0028

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
KEEP3

26
KEEP2

25
KEEP1

24
KEEP0

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
SUSP3

10
SUSP2

9
SUSP1

8
SUSP0

7
–

6
–

5
–

4
–

3
ENA3

2
ENA2

1
ENA1

0
ENA0

• ENA[3:0]
When set, a bit of the ENA field enables the relevant channel.
• SUSP[3:0]
When set, a bit of the SUSP field freezes the relevant channel and its current context.
• KEEP[3:0]
When set, a bit of the KEEP field resumes the current channel from an automatic stall state.

1050

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
40.5.11
Name:

DMAC Channel Handler Disable Register
DMAC_CHDR

Address:

0x400B002C

Access:

Write-only

Reset:

0x00000000

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
RES3

10
RES2

9
RES1

8
RES0

7
–

6
–

5
–

4
–

3
DIS3

2
DIS2

1
DIS1

0
DIS0

• DIS[3:0]
Write one to this field to disable the relevant DMAC Channel. The content of the FIFO is lost and the current AHB access is
terminated. Software must poll DIS[3:0] field in the DMAC_CHSR register to be sure that the channel is disabled.
• RES[3:0]
Write one to this field to resume the channel transfer restoring its context.

1051
6430F–ATARM–21-Feb-12

40.5.12
Name:

DMAC Channel Handler Status Register
DMAC_CHSR

Address:

0x400B0030

Access:

Read-only

Reset:

0x00FF0000

31
–

30
–

29
–

28
–

27
STAL3

26
STAL2

25
STAL1

24
STAL0

23
–

22
–

21
–

20
–

19
EMPT3

18
EMPT2

17
EMPT1

16
EMPT0

15
–

14
–

13
–

12
–

11
SUSP3

10
SUSP2

9
SUSP1

8
SUSP0

7
–

6
–

5
–

4
–

3
ENA3

2
ENA2

1
ENA1

0
ENA0

• ENA[3:0]
A one in any position of this field indicates that the relevant channel is enabled.
• SUSP[3:0]
A one in any position of this field indicates that the channel transfer is suspended.
• EMPT[3:0]
A one in any position of this field indicates that the relevant channel is empty.
• STAL[3:0]
A one in any position of this field indicates that the relevant channel is stalling.

1052

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
40.5.13
Name:

DMAC Channel x [x = 0..3] Source Address Register
DMAC_SADDRx [x = 0..3]

Addresses: 0x400B003C [0], 0x400B0064 [1], 0x400B008C [2], 0x400B00B4 [3]
Access:

Read-write

Reset:

0x00000000

SADDRx
23

20
SADDRx

12
SADDRx

4
SADDRx

• SADDRx
Channel x source address. This register must be aligned with the source transfer width.

1053
6430F–ATARM–21-Feb-12

40.5.14
Name:

DMAC Channel x [x = 0..3] Destination Address Register
DMAC_DADDRx [x = 0..3]

Addresses: 0x400B0040 [0], 0x400B0068 [1], 0x400B0090 [2], 0x400B00B8 [3]
Access:

Read-write

Reset:

0x00000000

DADDRx
23

20
DADDRx

12
DADDRx

4
DADDRx

• DADDRx
Channel x destination address. This register must be aligned with the destination transfer width.

1054

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
40.5.15
Name:

DMAC Channel x [x = 0..3] Descriptor Address Register
DMAC_DSCRx [x = 0..3]

Addresses: 0x400B0044 [0], 0x400B006C [1], 0x400B0094 [2], 0x400B00BC [3]
Access:

Read-write

Reset:

0x00000000

DSCRx
23

20
DSCRx

12
DSCRx

4
DSCRx

–

• DSCRx
Buffer Transfer descriptor address. This address is word aligned.

1055
6430F–ATARM–21-Feb-12

40.5.16
Name:

DMAC Channel x [x = 0..3] Control A Register
DMAC_CTRLAx [x = 0..3]

Addresses: 0x400B0048 [0], 0x400B0070 [1], 0x400B0098 [2], 0x400B00C0 [3]
Access:

Read-write

Reset:

0x00000000

31
DONE

30
–

23
–

22
–

21
–

27
–

26
–

20
DCSIZE

19
–

18
–

17
–

16
SCSIZE

DST_WIDTH

24
SRC_WIDTH

BTSIZE
7

BTSIZE

• BTSIZE
Buffer Transfer Size. The transfer size relates to the number of transfers to be performed, that is, for writes it refers to the
number of source width transfers to perform when DMAC is flow controller. For Reads, BTSIZE refers to the number of
transfers completed on the Source Interface. When this field is set to 0, the DMAC module is automatically disabled when
the relevant channel is enabled.
• SCSIZE
Source Chunk Transfer Size.
SCSIZE value

Number of data transferred

• DCSIZE
Destination Chunk Transfer size.
DCSIZE

Number of data transferred

• SRC_WIDTH

1056

SRC_WIDTH

Single Transfer Size

BYTE

HALF-WORD

WORD

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• DST_WIDTH
DST_WIDTH

Single Transfer Size

BYTE

HALF-WORD

WORD

• DONE
0: The transfer is performed.
1: If SOD field of DMAC_CFG register is set to true, then the DMAC is automatically disabled when an LLI updates the content of this register.
The DONE field is written back to memory at the end of the transfer.

1057
6430F–ATARM–21-Feb-12

40.5.17
Name:

DMAC Channel x [x = 0..3] Control B Register
DMAC_CTRLBx [x = 0..3]

Addresses: 0x400B004C [0], 0x400B0074 [1], 0x400B009C [2], 0x400B00C4 [3]
Access:

Read-write

Reset:

0x00000000

31
–

30
IEN

23
–

15
–
7
–

28
DST_INCR

27
–

26
–

24
SRC_INCR

20
DST_DSCR

19
–

18
–

17
–

16
SRC_DSCR

14
–

13
–

12
–

11
–

10
–

9
–

8
–

6
–

5
–

4
–

3
–

2
–

1
–

0
–

• SRC_DSCR
0: Source address is updated when the descriptor is fetched from the memory.
1: Buffer Descriptor Fetch operation is disabled for the source.
• DST_DSCR
0: Destination address is updated when the descriptor is fetched from the memory.
1: Buffer Descriptor Fetch operation is disabled for the destination.
• FC
This field defines which device controls the size of the buffer transfer, also referred as to the Flow Controller.
FC

Type of transfer

Flow Controller

000

Memory-to-Memory

DMA Controller

001

Memory-to-Peripheral

DMA Controller

010

Peripheral-to-Memory

DMA Controller

011

Peripheral-to-Peripheral

DMA Controller

• SRC_INCR

1058

SRC_INCR

Type of addressing mode

INCREMENTING

FIXED

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• DST_INCR
DST_INCR

Type of addressing scheme

INCREMENTING

FIXED

• IEN
If this bit is cleared, when the buffer transfer is completed, the BTC[x] flag is set in the EBCISR status register. This bit is
active low.

1059
6430F–ATARM–21-Feb-12

40.5.18
Name:

DMAC Channel x [x = 0..3] Configuration Register
DMAC_CFGx [x = 0..3]

Addresses: 0x400B0050 [0], 0x400B0078 [1], 0x400B00A0 [2], 0x400B00C8 [3]
Access:

Read-write

Reset:

0x0100000000

31
–

30
–

27
–

25
AHB_PROT

23
–

22
LOCK_IF_L

21
LOCK_B

20
LOCK_IF

19
–

18
–

17
–

16
SOD

15
–

14
–

13
DST_H2SEL

12
–

11
–

10
–

9
SRC_H2SEL

8
–

FIFOCFG

DST_PER

SRC_PER

• SRC_PER
Channel x Source Request is associated with peripheral identifier coded SRC_PER handshaking interface.
• DST_PER
Channel x Destination Request is associated with peripheral identifier coded DST_PER handshaking interface.
• SRC_H2SEL
0: Software handshaking interface is used to trigger a transfer request.
1: Hardware handshaking interface is used to trigger a transfer request.
• DST_H2SEL
0: Software handshaking interface is used to trigger a transfer request.
1: Hardware handshaking interface is used to trigger a transfer request.
• SOD
0: STOP ON DONE disabled, the descriptor fetch operation ignores DONE Field of CTRLA register.
1: STOP ON DONE activated, the DMAC module is automatically disabled if DONE FIELD is set to 1.
• LOCK_IF
0: Interface Lock capability is disabled
1: Interface Lock capability is enabled
• LOCK_B
0: AHB Bus Locking capability is disabled.
1: AHB Bus Locking capability is enabled.
• LOCK_IF_L
0: The Master Interface Arbiter is locked by the channel x for a chunk transfer.

1060

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
1: The Master Interface Arbiter is locked by the channel x for a buffer transfer.
• AHB_PROT
AHB_PROT field provides additional information about a bus access and is primarily used to implement some level of
protection.
HPROT[3]

HPROT[2]

HPROT[1]

HPROT[0]

Description

Data access
0: User Access
1: Privileged Access

AHB_PROT[0]

0: Not Bufferable
1: Bufferable

AHB_PROT[1]

0: Not cacheable
1: Cacheable

AHB_PROT[2]

• FIFOCFG
FIFOCFG

FIFO request

The largest defined length AHB burst is performed on the destination AHB interface.

When half FIFO size is available/filled, a source/destination request is serviced.

When there is enough space/data available to perform a single AHB access, then the request is serviced.

1061
6430F–ATARM–21-Feb-12

1062

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
41. 12-bit Analog-to-Digital Converter (ADC12B)
41.1

Description
The ADC12B is based on a Cyclic Pipeline 12-bit Analog-to-Digital Converter (ADC12B).
It also integrates an 8-to-1 analog multiplexer, making possible the analog-to-digital conversions
of 8 analog lines. The conversions extend from 0V to AD12BVREF.
The ADC12B supports a 10-bit or 12-bit resolution mode, and conversion results are reported in
a common register for all channels, as well as in a channel-dedicated register. Software trigger,
external trigger on rising edge of the AD12BTRG pin, internal triggers from Timer Counter output(s) or PWM Event lines are configurable.
The ADC12B also integrates a Sleep Mode and a conversion sequencer and connects with a
PDC channel. These features reduce both power consumption and processor intervention.
This ADC12B has a selectable single-ended or fully differential input and benefits from a 2-bit
programmable gain. A whole set of reference voltage is generated internally from a single external reference voltage node that may be equal to the analog supply voltage. An external
decoupling capacitance is required for noise filtering.
A digital error correction circuit based on the multi-bit redundant signed digit (RSD) algorithm is
employed in order to reduce INL and DNL errors.
Finally, the user can configure ADC12B timings, such as Startup Time and Sample & Hold Time.

41.2

Block Diagram

Figure 41-1. Analog-to-Digital Converter Block Diagram
Timer
Counter
Channels

ADC12B

Trigger
Selection

AD12BTRG

Control
Logic

ADC12B Interrupt

NVIC
AHB

VDDANA

PDC
AD12BVREF

AD12B0

Analog
Inputs

Peripheral Bridge

PIO
AD12B1

AD12Bn

IN+
IN-

OFFSET

S/H

PGA

Cyclic Pipeline
12-bit Analog-to-Digital
Converter

User
Interface
APB

GND

1063
6430F–ATARM–21-Feb-12

41.3

Signal Description

Table 41-1.

ADC12B Pin Description

Pin Name

Description

AD12B0 - AD12B7

Analog input channels

AD12BTRG

External trigger

41.4
41.4.1

Product Dependencies
Power Management
The ADC12B Controller is not continuously clocked. The programmer must first enable the
ADC12B Controller clock in the Power Management Controller (PMC) before using the ADC12B
Controller. However, if the application does not require ADC12B operations, the ADC12B Controller clock can be stopped when not needed and restarted when necessary.
Configuring the ADC12B Controller does not require the ADC12B Controller clock to be
enabled.

41.4.2

Interrupt Sources
The ADC12B interrupt line is connected on one of the sources of the Nested Vectored Interrupt
Controller (NVIC). Using the ADC12B interrupt requires the NVIC to be programmed first.
Table 41-2.

41.4.3

Peripheral IDs

Instance

ADC12B

Analog Inputs
The analog input pins are multiplexed with PIO lines. The assignment of the ADC12B input is
automatically done as soon as the corresponding channel is enabled by writing the register
ADC12B_CHER. By default, after reset, the PIO line is configured as an input with its pull-up
enabled and the ADC12B input is connected to the GND.

41.4.4

I/O Lines
The AD12BTRG pin is shared with other peripheral functions through the PIO Controller. In this
case, the PIO Controller needs to be set accordingly to assign the AD12BTRG pin to the
ADC12B function.
Table 41-3.

41.4.5

I/O Lines

Instance

Signal

I/O Line

Peripheral

ADC12B

AD12BTRG

PA2

Timer Triggers
Timer Counters may or may not be used as hardware triggers depending on user requirements.
Thus, some or all of the timer counters may be non-connected.

41.4.6

1064

PWM Event Lines
PWM Event Lines may or may not be used as hardware triggers depending on user
requirements.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
41.4.7

41.5
41.5.1

Conversion Performances
For performance and electrical characteristics of the ADC12B, see the DC Characteristics section of the product datasheet.

Functional Description
Analog-to-digital Conversion
The ADC12B uses the ADC12B Clock to perform conversions. Converting a single analog value
to 12-bit digital data requires Sample and Hold Clock cycles as defined in the SHTIM field of the
“ADC12B Mode Register” on page 1074 and 10 ADC12B Clock cycles. The ADC12B Clock frequency is selected in the PRESCAL field of the Mode Register (ADC12B_MR).
The ADC12B clock range is between MCK/2, if PRESCAL is 0, and MCK/128, if PRESCAL is
set to 63 (0x3F). PRESCAL must be programmed in order to provide an ADC12B clock frequency according to the parameters given in the Electrical Characteristics section of the product
datasheet.

41.5.2

Conversion Reference
The conversion is performed on a full range between 0V and the reference voltage pin
AD12BVREF Analog inputs between these voltages convert to values based on a linear
conversion.

41.5.3

Conversion Resolution
The ADC12B supports 10-bit or 12-bit resolution. The 10-bit selection is performed by setting the
LOWRES bit in the ADC12B Mode Register (ADC12B_MR). By default, after a reset, the resolution is the highest and the DATA field in the data registers is fully used. By setting the LOWRES
bit, the ADC12B switches in the lowest resolution and the conversion results can be read in the
eight lowest significant bits of the data registers. The two highest bits of the DATA field in the
corresponding ADC12B_CDR register and of the LDATA field in the ADC12B_LCDR register
read 0.
Moreover, when a PDC channel is connected to the ADC12B, 12-bit or 10-bit resolution sets the
transfer request size to 16 bits.

41.5.4

Differential Inputs
The ADC12B can be used either as a single ended ADC12B (DIFF bit equal to 0) or as a fully
differential ADC12B (DIFF bit equal to 1) as shown in Figure 41-2. By default, after a reset, the
ADC12B is in single ended mode.
The same inputs are used in single ended or differential mode.
In single ended mode, inputs are managed by an 8:1 channels analog multiplexer. In the fully
differential mode, inputs are managed by a 4:1 channels analog multiplexer. See Table 41-4 and
Table 41-5.

1065
6430F–ATARM–21-Feb-12

Table 41-4.

Input Pins and Channel Number in Single Ended Mode

Input Pins

Channel Number

AD12B0

CH0

AD12B1

CH1

AD12B2

CH2

AD12B3

CH3

AD12B4

CH4

AD12B5

CH5

AD12B6

CH6

AD12B7

CH7

Table 41-5.

41.5.5

Input Pins and Channel Number In Differential Mode

Input Pins

Channel Number

AD12B0-AD12B1

CH0

AD12B2-AD12B3

CH2

AD12B4-AD125B

CH4

AD12B6-AD12B7

CH6

Input Gain and Offset
The ADC12B has a built in Programmable Gain Amplifier (PGA) and Programmable Offset.
The Programmable Gain Amplifier can be set to gains of 1/2, 1, 2 and 4. The Programmable
Gain Amplifier can be used either for single ended applications or for fully differential
applications.
The gain is configurable through the GAIN bit as shown in Table 41-6.
Table 41-6.

Gain of the Sample and Hold Unit: GAIN Bits and DIFF Bit.

GAIN<0:1>

GAIN (DIFF = 0)

GAIN (DIFF = 1)

0.5

To allow full range, analog offset of the ADC12B can be configured by the OFFSET bit. The Offset can only be changed in single ended mode. In fully differential mode the offset is always set
to Vrefin/2.
Table 41-7.

Offset of the Sample and Hold Unit: OFFSET DIFF and Gain (G)

OFFSET Bit

OFFSET (DIFF = 0)

Vrefin/2G

Vrefin/2

OFFSET (DIFF = 1)
Vrefin/2

1066

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 41-2. Analog Full Scale Ranges in Single Ended/Differential Applications Versus Gain and Offset

single ended
se0fd1=0

fully differential
se0fd1=1
vrefin
VIN+

VIN+

same as
gain=1

gain=0.5

(½)vrefin

(00)
VIN-

0
vrefin
(¾)vrefin

VIN+

gain=1

(½)vrefin

VIN-

(01)

(¼)vrefin

0
vrefin

offset=1

offset=0
(¾)vrefin
(5/8)vrefin

gain=2

(½)vrefin

VIN+

(3/8)vrefin

(10)
VIN+

VIN+
VIN-

(¼)vrefin

0
vrefin

offset=1

gain=4

same as
gain=2

offset=0
(5/8)vrefin

VIN+

(½)vrefin
(3/8)vrefin

(11)
VIN+

VIN+
VIN-

(¼)vrefin
(1/8)vrefin

1067
6430F–ATARM–21-Feb-12

41.5.6

Power Consumption Adjustment
The power consumption of the ADC12B can be adjusted through a 2-bit bias control (IBCTL bit
in ADC12B_ACR register) providing possibilities for smart optimization of power and effective
resolution relative to the application speed request.
Please refer to the Electrical Characteristics of the product datasheet for further details.

41.5.7

Conversion Results
When a conversion is completed, the resulting 12-bit digital value is stored in the Channel Data
Register (ADC12B_CDR) of the current channel and in the ADC12B Last Converted Data Register (ADC12B_LCDR).
The channel EOC bit in the Status Register (ADC12B_SR) is set and the DRDY bit is set. In the
case of a connected PDC channel, DRDY rising triggers a data transfer request. In any case,
either EOC and DRDY can trigger an interrupt.
Reading one of the ADC12B_CDR registers clears the corresponding EOC bit. Reading
ADC12B_LCDR clears the DRDY bit and the EOC bit corresponding to the last converted
channel.

Figure 41-3. EOCx and DRDY Flag Behavior
Write the ADC_CR
with START = 1

Read the ADC_CDRx

Write the ADC_CR
with START = 1

Read the ADC_LCDR

CHx
(ADC_CHSR)
EOCx
(ADC_SR)
Conversion Time

Conversion Time

DRDY
(ADC_SR)

1068

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
If the ADC12B_CDR is not read before further incoming data is converted, the corresponding
Overrun Error (OVRE) flag is set in the Status Register (ADC12B_SR).
Likewise, new data converted when DRDY is high sets the GOVRE bit (General Overrun Error)
in ADC12B_SR.
The OVRE and GOVRE flags are automatically cleared when ADC12B_SR is read.
Figure 41-4. GOVRE and OVREx Flag Behavior
Read ADC12B_SR

Trigger
CH0
(ADC12B_CHSR)
CH1
(ADC12B_CHSR)
ADC12B_LCDR

Undefined Data

ADC12B_CDR0

Undefined Data

ADC12B_CDR1

EOC0
(ADC12B_SR)

EOC1
(ADC12B_SR)

Data CH0

Data CH1

Data CH0

Undefined Data

Data CH1

Conversion

Read ADC12B_CDR0

Read ADC12B_CDR1

GOVRE
(ADC12B_SR)
DRDY
(ADC12B_SR)
OVRE0
(ADC12B_SR)

Warning: If the corresponding channel is disabled during a conversion or if it is disabled and
then reenabled during a conversion, its associated data and its corresponding EOC and OVRE
flags in ADC12B_SR are unpredictable.
The ADC12B can be triggered externally by software or internally from the Timer Counter or
PWM.

1069
6430F–ATARM–21-Feb-12

41.5.8

Conversion Triggers
Conversions of the active analog channels are started with a software or a hardware trigger. The
software trigger is provided by writing the Control Register (ADC12B_CR) with the START bit at
1.
The hardware trigger can be one of the TIOA outputs of the Timer Counter channels, PWM
Event lines or the external trigger input of the ADC12B (AD12BTRG). The hardware trigger is
selected with the field TRGSEL in the Mode Register (ADC12B_MR). The selected hardware
trigger is enabled with the TRGEN bit in the Mode Register (ADC12B_MR).
If a hardware trigger is selected, the start of a conversion is triggered after a delay starting at
each rising edge of the selected signal. Due to asynchronous handling, the delay may vary in a
range of 2 MCK clock periods to 1 ADC12B clock period.
trigger
start
delay

If one of the TIOA outputs is selected, the corresponding Timer Counter channel must be programmed in Waveform Mode.
Only one start command is necessary to initiate a conversion sequence on all the channels. The
ADC12B hardware logic automatically performs the conversions on the active channels, then
waits for a new request. The Channel Enable (ADC12B_CHER) and Channel Disable
(ADC12B_CHDR) Registers enable the analog channels to be enabled or disabled
independently.
If the ADC12B is used with a PDC, only the transfers of converted data from enabled channels
are performed and the resulting data buffers should be interpreted accordingly.
Warning: Enabling hardware triggers does not disable the software trigger functionality. Thus, if
a hardware trigger is selected, the start of a conversion can be initiated either by the hardware or
the software trigger.
41.5.9

Sleep Mode and Conversion Sequencer
The ADC12B Sleep Mode maximizes power saving by automatically deactivating the ADC12B
when it is not being used for conversions. Sleep Mode is selected by setting the SLEEP bit in the
Mode Register ADC12B_MR.
Two sleep Mode are selectable (OFFMODES): STANDBY Mode and OFF Mode. In Standby
Mode, the ADC12B is powered off except voltage reference to allow fast startup. In OFF Mode
the ADC12B is totally powered off.
Table 41-8.

Low Power Modes According SLEEP Bit and OFFMODES Bit.

SLEEP Bit

OFFMODES Bit

Low Power Mode

Normal Mode

Standby Mode

Off Mode

The SLEEP mode is automatically managed by a conversion sequencer, which can automatically process the conversions of all channels at lowest power consumption.

1070

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
When a start conversion request occurs, the ADC12B is automatically activated. As the analog
cell requires a start-up time, the logic waits during this time and starts the conversion on the
enabled channels. When all conversions are complete, the ADC12B is deactivated until the next
trigger. Triggers occurring during the sequence are not taken into account.
The conversion sequencer allows automatic processing with minimum processor intervention
and optimized power consumption. Conversion sequences can be performed periodically using
a Timer/Counter output or a PWM Event line. The periodic acquisition of several samples can be
processed automatically without any intervention of the processor thanks to the PDC.
The conversion sequencer can only be used if all ADC12B inputs have the same input configuration, e.g. same PGA gain, same input type (differential or single ended) and same input offset.
If input have different configurations, sequencer can’t be used because PGA gain, input type
and input offset can’t be changed.
Note:

41.5.10

The reference voltage pins always remain connected in normal mode as in sleep mode.

ADC12B Timings
Each ADC12B has its own minimal Startup Time that is programmed through the field
STARTUP in the Mode Register (ADC12B_MR).
In the same way, a minimal Sample and Hold Time is necessary for the ADC12B to guarantee
the best converted final value between the two channels selection. This time has to be programmed through the SHTIM bitfield in the Mode Register (ADC12B_MR).
Warning: No input buffer amplifier to isolate the source is included in the ADC12B. This must be
taken into consideration to program a precise value in the SHTIM field. See the section, ADC12B
Characteristics in the product datasheet.

1071
6430F–ATARM–21-Feb-12

41.6

12-bit Analog-to-Digital Converter (ADC12B) User Interface

Table 41-9.
Offset

Name

Access

Reset

0x00

Control Register

ADC12B_CR

Write-only

–

0x04

Mode Register

ADC12B_MR

Read-write

0x00000000

0x08

Reserved

–

0x0C

Reserved

–

0x10

Channel Enable Register

ADC12B_CHER

Write-only

–

0x14

Channel Disable Register

ADC12B_CHDR

Write-only

–

0x18

Channel Status Register

ADC12B_CHSR

Read-only

0x00000000

0x1C

Status Register

ADC12B_SR

Read-only

0x000C0000

0x20

Last Converted Data Register

ADC12B_LCDR

Read-only

0x00000000

0x24

Interrupt Enable Register

ADC12B_IER

Write-only

–

0x28

Interrupt Disable Register

ADC12B_IDR

Write-only

–

0x2C

Interrupt Mask Register

ADC12B_IMR

Read-only

0x00000000

0x30

Channel Data Register 0

ADC12B_CDR0

Read-only

0x00000000

0x34

Channel Data Register 1

ADC12B_CDR1

Read-only

0x00000000

...

0x4C

Channel Data Register 7

ADC12B_CDR7

Read-only

0x00000000

0x64

Analog Control Register

ADC12B_ACR

Read-write

0x00000000

0x68

Extended Mode Register

ADC12B_EMR

Read-write

0x00000000

–

0x50 - 0xFC

1072

Reserved

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
41.6.1
Name:

ADC12B Control Register
ADC12B_CR

Address:

0x400A8000

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
START

0
SWRST

• SWRST: Software Reset
0 = No effect.
1 = Resets the ADC12B simulating a hardware reset.
• START: Start Conversion
0 = No effect.
1 = Begins analog-to-digital conversion.

1073
6430F–ATARM–21-Feb-12

41.6.2
Name:

ADC12B Mode Register
ADC12B_MR

Address:

0x400A8004

Access:

Read-write

31
–

30
–

29
–

28
–

SHTIM
19

2
TRGSEL

0
TRGEN

STARTUP
15

12
PRESCAL

7
–

6
–

5
SLEEP

4
LOWRES

• TRGEN: Trigger Enable
TRGEN

Selected TRGEN

Hardware triggers are disabled. Starting a conversion is only possible by software.

Hardware trigger selected by TRGSEL field is enabled.

• TRGSEL: Trigger Selection
TRGSEL

Selected TRGSEL

External trigger

TIO Output of the Timer Counter Channel 0

TIO Output of the Timer Counter Channel 1

TIO Output of the Timer Counter Channel 2

PWM Event Line 0

PWM Event Line 1

Reserved

• LOWRES: Resolution
LOWRES

Selected Resolution

12-bit resolution

10-bit resolution

• SLEEP: Sleep Mode
SLEEP

1074

Selected Mode

Normal Mode

Sleep Modes (see OFFMODES register)

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• PRESCAL: Prescaler Rate Selection
ADC12BClock = MCK/( (PRESCAL+1) * 2 )
• STARTUP: Start Up Time
Startup Time = (STARTUP+1) * 8/ADC12BClock
• SHTIM: Sample & Hold Time
Sample and Hold Time = SHTIM/ADC12BClock

1075
6430F–ATARM–21-Feb-12

41.6.3
Name:

ADC12B Channel Enable Register
ADC12B_CHER

Address:

0x400A8010

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
CH7

6
CH6

5
CH5

4
CH4

3
CH3

2
CH2

1
CH1

0
CH0

• CHx: Channel x Enable
0 = No effect.
1 = Enables the corresponding channel.

1076

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
41.6.4
Name:

ADC12B Channel Disable Register
ADC12B_CHDR

Address:

0x400A8014

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
CH7

6
CH6

5
CH5

4
CH4

3
CH3

2
CH2

1
CH1

0
CH0

• x: Channel x Disable
0 = No effect.
1 = Disables the corresponding channel.
Warning: If the corresponding channel is disabled during a conversion or if it is disabled then reenabled during a conversion, its associated data and its corresponding EOC and OVRE flags in ADC12B_SR are unpredictable.

1077
6430F–ATARM–21-Feb-12

41.6.5
Name:

ADC12B Channel Status Register
ADC12B_CHSR

Address:

0x400A8018

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
CH7

6
CH6

5
CH5

4
CH4

3
CH3

2
CH2

1
CH1

0
CH0

• CHx: Channel x Status
0 = Corresponding channel is disabled.
1 = Corresponding channel is enabled.

1078

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
41.6.6
Name:

ADC12B Analog Control Register
ADC12B_ACR

Address:

0x400A8064

Access:

Read-write

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
OFFSET

16
DIFF

15
–

14
–

13
–

12
–

11
–

10
–

7
–

6
–

5
–

4
–

3
–

2
–

IBCTL
0
GAIN

• GAIN: Input Gain
Gain of the sample and hold unit according to GAIN bits and DIFF bit
GAIN<0:1>

GAIN (DIFFx = 0)

GAIN (DIFF = 1)

0.5

• IBCTL: Bias Current Control
Bias Current Control
IBCTL<0:1>

Current

typ - 20%

typ

typ + 20%

typ + 40%

• DIFF: Differential Mode
0 = Single Ended Mode
1 = Fully Differential Mode
• OFFSET: Input OFFSET
Offset of the sample and hold unit according to OFFSET bit, DIFF bit and Gain (G).
OFFSET

OFFSET (DIFF = 0)

Vrefin/2G

Vrefin/2

OFFSET (DIFF = 1)
Vrefin/2

1079
6430F–ATARM–21-Feb-12

41.6.7
Name:

ADC12B Extended Mode Register
ADC12B_EMR

Address:

0x400A8068

Access:

Read-write

31
–

30
–

29
–

28
–

15
–

14
–

13
–

12
–

7
–

6
–

5
–

4
–

27
–

26
–

25
–

24
–

11
–

10
–

9
–

8
–

3
–

2
–

1
–

0
OFFMODES

20
19
OFF_MODE_STARTUP_TIME

• OFFMODES: Off Mode if Sleep Bit (ADC12B_MR) = 1
0 = Standby Mode
1 = Off Mode
• OFF_MODE_STARTUP_TIME: Startup Time
Off Mode Startup Time = (OFF_MODE_STARTUP_TIME+1) * 8/ADC12BClock

1080

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
41.6.8
Name:

ADC12B Status Register
ADC12B_SR

Address:

0x400A801C

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
RXBUFF

18
ENDRX

17
GOVRE

16
DRDY

15
OVRE7

14
OVRE6

13
OVRE5

12
OVRE4

11
OVRE3

10
OVRE2

9
OVRE1

8
OVRE0

7
EOC7

6
EOC6

5
EOC5

4
EOC4

3
EOC3

2
EOC2

1
EOC1

0
EOC0

• EOCx: End of Conversion x
0 = Corresponding analog channel is disabled, or the conversion is not finished.
1 = Corresponding analog channel is enabled and conversion is complete.
• OVREx: Overrun Error x
0 = No overrun error on the corresponding channel since the last read of ADC12B_SR.
1 = There has been an overrun error on the corresponding channel since the last read of ADC12B_SR.
• DRDY: Data Ready
0 = No data has been converted since the last read of ADC12B_LCDR.
1 = At least one data has been converted and is available in ADC12B_LCDR.
• GOVRE: General Overrun Error
0 = No General Overrun Error occurred since the last read of ADC12B_SR.
1 = At least one General Overrun Error has occurred since the last read of ADC12B_SR.
• ENDRX: End of RX Buffer
0 = The Receive Counter Register has not reached 0 since the last write in ADC12B_RCR or ADC12B_RNCR.
1 = The Receive Counter Register has reached 0 since the last write in ADC12B_RCR or ADC12B_RNCR.
• RXBUFF: RX Buffer Full
0 = ADC12B_RCR or ADC12B_RNCR have a value other than 0.
1 = Both ADC12B_RCR and ADC12B_RNCR have a value of 0.

1081
6430F–ATARM–21-Feb-12

41.6.9
Name:

ADC12B Last Converted Data Register
ADC12B_LCDR

Address:

0x400A8020

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

LDATA
3

LDATA

• LDATA: Last Data Converted
The analog-to-digital conversion data is placed into this register at the end of a conversion and remains until a new conversion is completed.

1082

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
41.6.10
Name:

ADC12B Interrupt Enable Register
ADC12B_IER

Address:

0x400A8024

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
RXBUFF

18
ENDRX

17
GOVRE

16
DRDY

15
OVRE7

14
OVRE6

13
OVRE5

12
OVRE4

11
OVRE3

10
OVRE2

9
OVRE1

8
OVRE0

7
EOC7

6
EOC6

5
EOC5

4
EOC4

3
EOC3

2
EOC2

1
EOC1

0
EOC0

• EOCx: End of Conversion Interrupt Enable x
• OVREx: Overrun Error Interrupt Enable x
• DRDY: Data Ready Interrupt Enable
• GOVRE: General Overrun Error Interrupt Enable
• ENDRX: End of Receive Buffer Interrupt Enable
• RXBUFF: Receive Buffer Full Interrupt Enable
0 = No effect.
1 = Enables the corresponding interrupt.

1083
6430F–ATARM–21-Feb-12

41.6.11
Name:

ADC12B Interrupt Disable Register
ADC12B_IDR

Address:

0x400A8028

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
RXBUFF

18
ENDRX

17
GOVRE

16
DRDY

15
OVRE7

14
OVRE6

13
OVRE5

12
OVRE4

11
OVRE3

10
OVRE2

9
OVRE1

8
OVRE0

7
EOC7

6
EOC6

5
EOC5

4
EOC4

3
EOC3

2
EOC2

1
EOC1

0
EOC0

• EOCx: End of Conversion Interrupt Disable x
• OVREx: Overrun Error Interrupt Disable x
• DRDY: Data Ready Interrupt Disable
• GOVRE: General Overrun Error Interrupt Disable
• ENDRX: End of Receive Buffer Interrupt Disable
• RXBUFF: Receive Buffer Full Interrupt Disable
0 = No effect.
1 = Disables the corresponding interrupt.

1084

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
41.6.12
Name:

ADC12B Interrupt Mask Register
ADC12B_IMR

Address:

0x400A802C

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
RXBUFF

18
ENDRX

17
GOVRE

16
DRDY

15
OVRE7

14
OVRE6

13
OVRE5

12
OVRE4

11
OVRE3

10
OVRE2

9
OVRE1

8
OVRE0

7
EOC7

6
EOC6

5
EOC5

4
EOC4

3
EOC3

2
EOC2

1
EOC1

0
EOC0

• EOCx: End of Conversion Interrupt Mask x
• OVREx: Overrun Error Interrupt Mask x
• DRDY: Data Ready Interrupt Mask
• GOVRE: General Overrun Error Interrupt Mask
• ENDRX: End of Receive Buffer Interrupt Mask
• RXBUFF: Receive Buffer Full Interrupt Mask
0 = The corresponding interrupt is disabled.
1 = The corresponding interrupt is enabled.

1085
6430F–ATARM–21-Feb-12

41.6.13
Name:

ADC12B Channel Data Register
ADC12B_CDRx

Address:

0x400A8030

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

DATA
3

DATA

• DATA: Converted Data
The analog-to-digital conversion data is placed into this register at the end of a conversion and remains until a new conversion is completed. The Convert Data Register (CDR) is only loaded if the corresponding analog channel is enabled.

1086

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
42. Analog-to-Digital Converter (ADC)
42.1

Description
The ADC is based on a Successive Approximation Register (SAR) 10-bit Analog-to-Digital Converter (ADC). It also integrates an 8-to-1 analog multiplexer, making possible the analog-todigital conversions of 8 analog lines. The conversions extend from 0V to ADVREF. The ADC
supports an 8-bit or 10-bit resolution mode, and conversion results are reported in a common
register for all channels, as well as in a channel-dedicated register. Software trigger, external
trigger on rising edge of the ADTRG pin or internal triggers from Timer Counter output(s) or
PWM Event lines are configurable.
The ADC also integrates a Sleep Mode and a conversion sequencer and connects with a PDC
channel. These features reduce both power consumption and processor intervention.
Finally, the user can configure ADC timings, such as Startup Time and Sample & Hold Time.

42.2

Block Diagram

Figure 42-1. Analog-to-Digital Converter Block Diagram
Timer
Counter
Channels

PMC
MCK

ADC Controller
Trigger
Selection

ADTRG

Control
Logic

ADC Interrupt

NVIC

VDDANA
ADC cell
ADVREF
System Bus
AD0

Analog Inputs
Multiplexed
with I/O lines

PDC

PIO

AD1
User
Interface
ADn

Peripheral Bridge

Successive
Approximation
Register
Analog-to-Digital
Converter
APB

GND

42.3

Signal Description

Table 42-1.

ADC Pin Description

Pin Name

Description

AD0 - AD7

Analog input channels

ADTRG

External trigger

1087
6430F–ATARM–21-Feb-12

42.4

Product Dependencies

42.4.1

Power Management
The MCK of the ADC Controller is not continuously clocked. The programmer must first enable
the ADC Controller MCK in the Power Management Controller (PMC) before using the ADC
Controller. However, if the application does not require ADC operations, the ADC Controller
clock can be stopped when not needed and restarted when necessary. Configuring the ADC
Controller does not require the ADC Controller clock to be enabled.

42.4.2

Interrupt Sources
The ADC interrupt line is connected on one of the internal sources of the Advanced Interrupt
Controller. Using the ADC interrupt requires the NVIC to be programmed first.
Table 42-2.

42.4.3

Peripheral IDs

Instance

ADC

Analog Inputs
The analog input pins can be multiplexed with PIO lines. In this case, the assignment of the ADC
input is automatically done as soon as the corresponding channel is enabled by writing the register ADC_CHER. By default, after reset, the PIO line is configured as input with its pull-up
enabled and the ADC input is connected to the GND.

42.4.4

I/O Lines
The pin ADTRG may be shared with other peripheral functions through the PIO Controller. In
this case, the PIO Controller should be set accordingly to assign the pin ADTRG to the ADC
function.

42.4.5

Timer Triggers
Timer Counters may or may not be used as hardware triggers depending on user requirements.
Thus, some or all of the timer counters may be non-connected.

42.4.6

PWM Event Lines
PWM Event Lines may or may not be used as hardware triggers depending on user
requirements.

42.4.7

Conversion Performances
For performance and electrical characteristics of the ADC, see the DC Characteristics section.

1088

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
42.5
42.5.1

Functional Description
Analog-to-digital Conversion
The ADC uses the ADC Clock to perform conversions. Converting a single analog value to a 10bit digital data requires Sample and Hold Clock cycles as defined in the field SHTIM of the “ADC
Mode Register” on page 1096 and 10 ADC Clock cycles. The ADC Clock frequency is selected
in the PRESCAL field of the Mode Register (ADC_MR).
The ADC clock range is between MCK/2, if PRESCAL is 0, and MCK/128, if PRESCAL is set to
63 (0x3F). PRESCAL must be programmed in order to provide an ADC clock frequency according to the parameters given in the Product definition section.

42.5.2

Conversion Reference
The conversion is performed on a full range between 0V and the reference voltage pin ADVREF
Analog inputs between these voltages convert to values based on a linear conversion.

42.5.3

Conversion Resolution
The ADC supports 8-bit or 10-bit resolutions. The 8-bit selection is performed by setting the bit
LOWRES in the ADC Mode Register (ADC_MR). By default, after a reset, the resolution is the
highest and the DATA field in the data registers is fully used. By setting the bit LOWRES, the
ADC switches in the lowest resolution and the conversion results can be read in the eight lowest
significant bits of the data registers. The two highest bits of the DATA field in the corresponding
ADC_CDR register and of the LDATA field in the ADC_LCDR register read 0.
Moreover, when a PDC channel is connected to the ADC, 10-bit resolution sets the transfer
request sizes to 16-bit. Setting the bit LOWRES automatically switches to 8-bit data transfers. In
this case, the destination buffers are optimized.

1089
6430F–ATARM–21-Feb-12

42.5.4

Conversion Results
When a conversion is completed, the resulting 10-bit digital value is stored in the Channel Data
Register (ADC_CDR) of the current channel and in the ADC Last Converted Data Register
(ADC_LCDR).
The channel EOC bit in the Status Register (ADC_SR) is set and the DRDY is set. In the case of
a connected PDC channel, DRDY rising triggers a data transfer request. In any case, either
EOC and DRDY can trigger an interrupt.
Reading one of the ADC_CDR registers clears the corresponding EOC bit. Reading ADC_LCDR
clears the DRDY bit and the EOC bit corresponding to the last converted channel.

Figure 42-2. EOCx and DRDY Flag Behavior
Write the ADC_CR
with START = 1

Read the ADC_CDRx

Write the ADC_CR
with START = 1

Read the ADC_LCDR

CHx
(ADC_CHSR)
EOCx
(ADC_SR)
Conversion Time

Conversion Time

DRDY
(ADC_SR)

1090

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
If the ADC_CDR is not read before further incoming data is converted, the corresponding Overrun Error (OVRE) flag is set in the Status Register (ADC_SR).
In the same way, new data converted when DRDY is high sets the bit GOVRE (General Overrun
Error) in ADC_SR.
The OVRE and GOVRE flags are automatically cleared when ADC_SR is read.
Figure 42-3. GOVRE and OVREx Flag Behavior
Read ADC_SR

ADTRG
CH0
(ADC_CHSR)
CH1
(ADC_CHSR)
ADC_LCDR

Undefined Data

ADC_CDR0

Undefined Data

ADC_CDR1

EOC0
(ADC_SR)

EOC1
(ADC_SR)

Data B

Data A

Data C

Data A

Data C

Undefined Data

Data B

Conversion

Read ADC_CDR0

Read ADC_CDR1

GOVRE
(ADC_SR)
DRDY
(ADC_SR)
OVRE0
(ADC_SR)

1091
6430F–ATARM–21-Feb-12

42.5.5

Conversion Triggers
Conversions of the active analog channels are started with a software or a hardware trigger. The
software trigger is provided by writing the Control Register (ADC_CR) with the bit START at 1.
The hardware trigger can be one of the TIOA outputs of the Timer Counter channels, PWM
Event lines or the external trigger input of the ADC (ADTRG). The hardware trigger is selected
with the field TRGSEL in the Mode Register (ADC_MR). The selected hardware trigger is
enabled with the bit TRGEN in the Mode Register (ADC_MR).
If a hardware trigger is selected, the start of a conversion is triggered after a delay starting at
each rising edge of the selected signal.Due to asynchronism handling, the delay may vary in a
range of 2 MCK clock periods to 1 ADC clock period.
trigger
start
delay

If one of the TIOA outputs is selected, the corresponding Timer Counter channel must be programmed in Waveform Mode.
Only one start command is necessary to initiate a conversion sequence on all the channels. The
ADC hardware logic automatically performs the conversions on the active channels, then waits
for a new request. The Channel Enable (ADC_CHER) and Channel Disable (ADC_CHDR) Registers enable the analog channels to be enabled or disabled independently.
If the ADC is used with a PDC, only the transfers of converted data from enabled channels are
performed and the resulting data buffers should be interpreted accordingly.
Warning: Enabling hardware triggers does not disable the software trigger functionality. Thus, if
a hardware trigger is selected, the start of a conversion can be initiated either by the hardware or
the software trigger.
42.5.6

Sleep Mode and Conversion Sequencer
The ADC Sleep Mode maximizes power saving by automatically deactivating the ADC when it is
not being used for conversions. Sleep Mode is selected by setting the bit SLEEP in the Mode
Register ADC_MR.
The SLEEP mode is automatically managed by a conversion sequencer, which can automatically process the conversions of all channels at lowest power consumption.
When a start conversion request occurs, the ADC is automatically activated. As the analog cell
requires a start-up time, the logic waits during this time and starts the conversion on the enabled
channels. When all conversions are complete, the ADC is deactivated until the next trigger. Triggers occurring during the sequence are not taken into account.
The conversion sequencer allows automatic processing with minimum processor intervention
and optimized power consumption. Conversion sequences can be performed periodically using
a Timer/Counter output or a PWM Event line. The periodic acquisition of several samples can be
processed automatically without any intervention of the processor thanks to the PDC.
Note:

42.5.7

The reference voltage pins always remain connected in normal mode as in sleep mode.

ADC Timings
Each ADC has its own minimal Startup Time that is programmed through the field STARTUP in
the Mode Register ADC_MR.

1092

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
In the same way, a minimal Sample and Hold Time is necessary for the ADC to guarantee the
best converted final value between two channels selection. This time has to be programmed
through the bitfield SHTIM in the Mode Register ADC_MR.
Warning: No input buffer amplifier to isolate the source is included in the ADC. This must be
taken into consideration to program a precise value in the SHTIM field. See the section, ADC
Characteristics in the product datasheet.

1093
6430F–ATARM–21-Feb-12

42.6

Analog-to-Digital Converter (ADC) User Interface

Table 42-3.
Offset

Name

Access

Reset

0x00

Control Register

ADC_CR

Write-only

–

0x04

Mode Register

ADC_MR

Read-write

0x00000000

0x08

Reserved

–

0x0C

Reserved

–

0x10

Channel Enable Register

ADC_CHER

Write-only

–

0x14

Channel Disable Register

ADC_CHDR

Write-only

–

0x18

Channel Status Register

ADC_CHSR

Read-only

0x00000000

0x1C

Status Register

ADC_SR

Read-only

0x000C0000

0x20

Last Converted Data Register

ADC_LCDR

Read-only

0x00000000

0x24

Interrupt Enable Register

ADC_IER

Write-only

–

0x28

Interrupt Disable Register

ADC_IDR

Write-only

–

0x2C

Interrupt Mask Register

ADC_IMR

Read-only

0x00000000

0x30

Channel Data Register 0

ADC_CDR0

Read-only

0x00000000

0x34

Channel Data Register 1

ADC_CDR1

Read-only

0x00000000

...

ADC_CDR7

Read-only

0x00000000

–

...
0x4C
0x50 - 0xFC

1094

...
Channel Data Register 7
Reserved

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
42.6.1
Name:

ADC Control Register
ADC_CR

Address:

0x400AC000

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
–

6
–

5
–

4
–

3
–

2
–

1
START

0
SWRST

• SWRST: Software Reset
0 = No effect.
1 = Resets the ADC simulating a hardware reset.
• START: Start Conversion
0 = No effect.
1 = Begins analog-to-digital conversion.

1095
6430F–ATARM–21-Feb-12

42.6.2
Name:

ADC Mode Register
ADC_MR

Address:

0x400AC004

Access:

Read-write

31
–

30
–

29
–

28
–

23
–

19
STARTUP

2
TRGSEL

0
TRGEN

SHTIM

PRESCAL
7
–

6
–

5
SLEEP

4
LOWRES

• TRGEN: Trigger Enable
TRGEN

Selected TRGEN

Hardware triggers are disabled. Starting a conversion is only possible by software.

Hardware trigger selected by TRGSEL field is enabled.

• TRGSEL: Trigger Selection
TRGSEL

Selected TRGSEL

TIO Output of the Timer Counter Channel 0

TIO Output of the Timer Counter Channel 1

TIO Output of the Timer Counter Channel 2

PWM Event Line 0

PWM Event Line 1

Reserved

External trigger

Reserved

• LOWRES: Resolution
LOWRES

Selected Resolution

10-bit resolution

8-bit resolution

• SLEEP: Sleep Mode
SLEEP

1096

Selected Mode

Normal Mode

Sleep Mode

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
• PRESCAL: Prescaler Rate Selection
ADCClock = MCK / ( (PRESCAL+1) * 2 )
• STARTUP: Start Up Time
Startup Time = (STARTUP+1) * 8 / ADCClock
• SHTIM: Sample & Hold Time
Sample & Hold Time = SHTIM/ADCClock

1097
6430F–ATARM–21-Feb-12

42.6.3
Name:

ADC Channel Enable Register
ADC_CHER

Address:

0x400AC010

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
CH7

6
CH6

5
CH5

4
CH4

3
CH3

2
CH2

1
CH1

0
CH0

• CHx: Channel x Enable
0 = No effect.
1 = Enables the corresponding channel.

1098

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
42.6.4
Name:

ADC Channel Disable Register
ADC_CHDR

Address:

0x400AC014

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
CH7

6
CH6

5
CH5

4
CH4

3
CH3

2
CH2

1
CH1

0
CH0

• CHx: Channel x Disable
0 = No effect.
1 = Disables the corresponding channel.
Warning: If the corresponding channel is disabled during a conversion or if it is disabled then reenabled during a conversion, its associated data and its corresponding EOC and OVRE flags in ADC_SR are unpredictable.

1099
6430F–ATARM–21-Feb-12

42.6.5
Name:

ADC Channel Status Register
ADC_CHSR

Address:

0x400AC018

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

9
–

8
–

7
CH7

6
CH6

5
CH5

4
CH4

3
CH3

2
CH2

1
CH1

0
CH0

• CHx: Channel x Status
0 = Corresponding channel is disabled.
1 = Corresponding channel is enabled.

1100

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
42.6.6
Name:

ADC Status Register
ADC_SR

Address:

0x400AC01C

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
RXBUFF

18
ENDRX

17
GOVRE

16
DRDY

15
OVRE7

14
OVRE6

13
OVRE5

12
OVRE4

11
OVRE3

10
OVRE2

9
OVRE1

8
OVRE0

7
EOC7

6
EOC6

5
EOC5

4
EOC4

3
EOC3

2
EOC2

1
EOC1

0
EOC0

• EOCx: End of Conversion x
0 = Corresponding analog channel is disabled, or the conversion is not finished.
1 = Corresponding analog channel is enabled and conversion is complete.
• OVREx: Overrun Error x
0 = No overrun error on the corresponding channel since the last read of ADC_SR.
1 = There has been an overrun error on the corresponding channel since the last read of ADC_SR.
• DRDY: Data Ready
0 = No data has been converted since the last read of ADC_LCDR.
1 = At least one data has been converted and is available in ADC_LCDR.
• GOVRE: General Overrun Error
0 = No General Overrun Error occurred since the last read of ADC_SR.
1 = At least one General Overrun Error has occurred since the last read of ADC_SR.
• ENDRX: End of RX Buffer
0 = The Receive Counter Register has not reached 0 since the last write in ADC_RCR or ADC_RNCR.
1 = The Receive Counter Register has reached 0 since the last write in ADC_RCR or ADC_RNCR.
• RXBUFF: RX Buffer Full
0 = ADC_RCR or ADC_RNCR have a value other than 0.
1 = Both ADC_RCR and ADC_RNCR have a value of 0.

1101
6430F–ATARM–21-Feb-12

42.6.7
Name:

ADC Last Converted Data Register
ADC_LCDR

Address:

0x400AC020

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

LDATA
0

LDATA

• LDATA: Last Data Converted
The analog-to-digital conversion data is placed into this register at the end of a conversion and remains until a new conversion is completed.

1102

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
42.6.8
Name:

ADC Interrupt Enable Register
ADC_IER

Address:

0x400AC024

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
RXBUFF

18
ENDRX

17
GOVRE

16
DRDY

15
OVRE7

14
OVRE6

13
OVRE5

12
OVRE4

11
OVRE3

10
OVRE2

9
OVRE1

8
OVRE0

7
EOC7

6
EOC6

5
EOC5

4
EOC4

3
EOC3

2
EOC2

1
EOC1

0
EOC0

1103
6430F–ATARM–21-Feb-12

42.6.9
Name:

ADC Interrupt Disable Register
ADC_IDR

Address:

0x400AC028

Access:

Write-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
RXBUFF

18
ENDRX

17
GOVRE

16
DRDY

15
OVRE7

14
OVRE6

13
OVRE5

12
OVRE4

11
OVRE3

10
OVRE2

9
OVRE1

8
OVRE0

7
EOC7

6
EOC6

5
EOC5

4
EOC4

3
EOC3

2
EOC2

1
EOC1

0
EOC0

1104

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
42.6.10
Name:

ADC Interrupt Mask Register
ADC_IMR

Address:

0x400AC02C

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
RXBUFF

18
ENDRX

17
GOVRE

16
DRDY

15
OVRE7

14
OVRE6

13
OVRE5

12
OVRE4

11
OVRE3

10
OVRE2

9
OVRE1

8
OVRE0

7
EOC7

6
EOC6

5
EOC5

4
EOC4

3
EOC3

2
EOC2

1
EOC1

0
EOC0

1105
6430F–ATARM–21-Feb-12

42.6.11
Name:

ADC Channel Data Register
ADC_CDRx

Address:

0x400AC030

Access:

Read-only

31
–

30
–

29
–

28
–

27
–

26
–

25
–

24
–

23
–

22
–

21
–

20
–

19
–

18
–

17
–

16
–

15
–

14
–

13
–

12
–

11
–

10
–

DATA
0

DATA

1106

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
43. SAM3U4/2/1 Electrical Characteristics
43.1

Absolute Maximum Ratings

Table 43-1.

Absolute Maximum Ratings*

Operating Temperature (Industrial) ................-40° C to + 85° C
Storage Temperature.....................................-60°C to + 150°C
Voltage on Input Pins
with Respect to Ground...... ..............................-0.3V to + 4.0V
Maximum Operating Voltage
(VDDCORE) ..... ................................................................2.0V

*NOTICE:

Stresses beyond those listed under “Absolute Maximum
Ratings” may cause permanent damage to the device.
This is a stress rating only and functional operation of the
device at these or other conditions beyond those indicated in the operational sections of this specification is
not implied. Exposure to absolute maximum rating
conditions for extended periods may affect device
reliability.

Maximum Operating Voltage
(VDDIO) ..............................................................................4.0V
Total DC Output Current on all I/O lines
100-lead LQFP ..............................................................100 mA
144-lead LQFP ..............................................................130 mA
100-ball TFBGA .............................................................100 mA
144-ball LFBGA .............................................................130 mA

1107
6430F–ATARM–21-Feb-12

43.2

DC Characteristics
The following characteristics are applicable to the operating temperature range: TA = -40°C
to 85°C, unless otherwise specified.

Table 43-2.

DC Characteristics

Symbol

Parameter

VDDCORE

DC Supply Core

VVDDIO

DC Supply I/Os

VVDDBU

Backup I/O Lines Power
Supply

VVDDUTMII

Typ

Max

1.62

1.8

1.95

1.62

3.3

3.6

1.62

3.6

USB UTMI+ Interface
Power Supply

3.0

3.6

VVDDPLL

PLL A, UPLL and Main
Oscillator Supply

1.62

1.95

VVDDANA

ADC Analog Power Supply

(1)

VIL

Input Low-level Voltage

PIOA/B/C[0-31]

-0.3

0.3 x
VVDDIO

VIH

Input High-level Voltage

PIOA/B/C[0-31]

0.7 x VVDDIO

VVDDIO
+0.3V

VOH

Output High-level Voltage

PIOA/B/C[0-31]
IOH ~ 0
IOH > 0 (See IOH details below)

VVDDIO -0.2V
VVDDIO -0.4V

VOL

Output Low-level Voltage

PIOA/B/C[0-31]
IOH ~ 0
IOH > 0 (See IOL details below)

VHys

1108

Hysteresis Voltage

Conditions

Min

Units

0.2
0.4

PIOA/B/C[0-31]
except PIOA[14], PB[9-16], PB[25-PB31]
and PC[20-27]

150

500

ERASE, TST, FWUP, JTAGSEL

230

700

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 43-2.
Symbol

DC Characteristics (Continued)
Parameter

IOH (or ISOURCE)

Conditions

Min

Typ

1.62V < VDDIO < 1.95V; VOH = VVDDIO - 0.4
- PA3 (SPCK), PA15(MCCK) pins
- Other pins(2)

-8
-3

3.0V < VDDIO < 3.6V; VOH = VVDDIO - 0.4
- PA3 (SPCK), PA15(MCCK) pins
- Other pins(2)

-15
-3

1.62V < VDDIO < 3.6V; VOH = VVDDIO - 0.4
- NRST, TDO

IOL (or ISINK)

IIL

Input High
Leakage Current

IIH

1.62V < VDDIO < 1.95V; VOL = 0.4V
- PA3 (SPCK), PA15(MCCK) pins
- Other pins(2)

8
4

3.0V < VDDIO < 3.6V; VOL = 0.4V
- PA3 (SPCK), PA15(MCCK) pins
- Other pins(2)

9
6

1.62V < VDDIO < 3.6V; VOL = 0.4V
- NRST, TDO

14
9

VVDDIO powered pins(3)
No pull-up or pull-down; VIN=GND; VDDIO Max.
(Typ: TA = 25°C, Max: TA = 85°C)

VDDBU powered pins(4)
No pull-up or pull-down; VIN=GND; VDDBU Max.
(Typ: TA = 25°C, Max: TA = 85°C)
VVDDIO powered pins(3)
No pull-up or pull-down; VIN=VDD; VDDIO Max.
(Typ: TA = 25°C, Max: TA = 85°C)

VDDBU powered pins(4)
No pull-up or pull-down; VIN=VDD; VDDBU Max.
(Typ: TA = 25°C, Max: TA = 85°C)

μA

150

kΩ

NRSTB

kΩ

Pull-down Resistor

TST, ERASE, JTAGSEL

kΩ

On-die Series Termination
Resistor

PA0-PA31, PB0-PB31, PC0-PC31

Input Capacitance

Digital Inputs

RPULLDOWN

RODT

100

Pull-up Resistor

CIN

PA0-PA31, PB0-PB31, PC0-PC31

RPULLUP

Notes:

-24
-9

Relaxed Mode:
3.0V < VDDIO < 3.6V; VOL = 0.6V
- PA3 (SPCK), PA15(MCCK) pins
- Other pins(2)

Input Low
Leakage Current

Units

-2

Relaxed Mode:
3.0V < VDDIO < 3.6V; VOH = 2.2V
- PA3 (SPCK), PA15(MCCK) pins
- Other pins(2)

Max

36
TBD

1. Refer to Section 43.7 “12-Bit ADC Characteristics” and Section 43.8 “AC Characteristics”
2. PA[0-2], PA[4-14], PA[16-31]; PB[0-31]; PC[0-31]
3. PA[0-31], PB[0-31], PC[0-31]
4. FWUP, JTAGSEL, NRSTB, ERASE, TST

1109
6430F–ATARM–21-Feb-12

Table 43-3.

1.8V Voltage Regulator Characteristics

Symbol

Parameter

VVDDIN

DC Input Voltage Range

VVDDOUT

DC Output Voltage

Normal Mode
Standby Mode

VACCURACY

Output Voltage Accuracy

ILoad = 0.5mA to 150 mA

ILOAD

Maximum DC Output
Current
Maximum Peak Current
during startup(3)

DDROPOUT
VLINE

ILOAD-START

Conditions

Min

Typ

Max

Units

1.8

3.3

3.6

1.8
0
-3

V
3

VVDDIN > 2.2V
VVDDIN ≤2.2V

150
60

See Note (3).

300

Dropout Voltage

VVDDIN = 1.8V, ILoad = 60 mA

150

Line Regulation

VVDDIN from 2.7V to 3.6V;

100

ILoad MAX
VLINE-TR

Transient Line regulation

VVDDIN from 2.7V to 3.6V;
tr = tf = 5μs; ILoad Max
CDOUT = 4.7μF

VLOAD

Load Regulation

VVDDIN ≥ 2.2V;
ILoad = 10% to 90% MAX

VLOAD-TR

Transient Load Regulation

VVDDIN ≥ 2.2V;

ILoad = 10% to 90% MAX

100

7
700

10
1200
1

tr = tf = 5 μs
CDOUT = 4.7 μF
Normal Mode;
@ ILoad = 0 mA
@ ILoad = 150 mA
Standby Mode;

Quiescent Current

CDIN

Input Decoupling Capacitor Cf. External Capacitor Requirements (1)

μF

Cf. External Capacitor Requirements (2)

4.7

μF

CDOUT

Output Decoupling
Capacitor

ESR

0.5

μA

Ohm

TON

Turn on Time

CDOUT= 4.7μF, VVDDOUT reaches VTH+
(core power brownout detector supply rising
threshold)

120

250

μs

TON

Turn on Time

CDOUT= 4.7μF, VVDDOUT reaches 1.8V (+/- 3%)

200

400

μs

Notes:

1. A 10 μF or higher ceramic capacitor must be connected between VDDIN and the closest GND pin of the device.
This large decoupling capacitor is mandatory to reduce startup current, improving transient response and noise rejection.
2. To ensure stability, an external 4.7μF output capacitor, CDOUT must be connected between the VDDOUT and the closest
GND pin of the device. The ESR (Equivalent Series Resistance) of the capacitor must be in the range 0.5 to 10 Ohms.
Solid tantalum, and multilayer ceramic capacitors are all suitable as output capacitor.
A 100 nF bypass capacitor between VDDOUT and the closest GND pin of the device decreases output noise and improves
the load transient response.
3. Defined as the current needed to charge external bypass/decoupling capacitor network.

1110

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series

Table 43-4.
Symbol

Core Power Supply Brownout Detector Characteristics
Parameter

Conditions
(1)

Min

Typ

Max

Units

1.52

1.55

1.58

1.50

1.62

250

VTH-

Supply Falling Threshold

VHYST-

Hysteresis VTH-

VTH+

Supply Rising Threshold

1.35

VHYST+

Hysteresis VTH+

100

IDDON

Current Consumption on
VDDCORE

IDDOFF
Td-

VTH- detection propagation
time

Td+

VTH+ detection propagation
time

TSTART

Startup Time

Note:

Brownout Detector enabled

μA

Brownout Detector disabled

200

VDDCORE = VTH+ to (VTH- - 100mV)

200

350

μs

100

200

μs

Max

Units

100
From disabled state to enabled state

1. The product is guaranteed to be functional at VTH-

Figure 43-1. Core Brownout Output Waveform
VDDCORE

Vth+
Vtht
BOD OUTPUT

td-

td+

Table 43-5.

VDDUTMI Supply Monitor

Symbol

Parameter

Conditions

Min

VTH

Supply Monitor Threshold

16 selectable steps of 100mV

1.9

3.4

TACCURACY

Threshold Level Accuracy

-1.5

+1.5

VHYST

Hysteresis

IDDON
IDDOFF
TSTART

Current Consumption on
VDDCORE
Startup Time

enabled

Typ

μA
disabled
From disabled state to enabled state

1
140

μs

1111
6430F–ATARM–21-Feb-12

Figure 43-2. VDDUTMI Supply Monitor
VDDUTMI

Vth+
Vhyst
Vth

Reset

Table 43-6.

Backup Power Supply Zero-Power-on Reset Characteristics

Symbol

Parameter

Conditions

Min

Typ

Max

Units

Vth+

Threshold voltage rising

At Startup

1.50

1.55

1.60

Vth-

Threshold voltage falling

1.40

1.45

1.50

Tres

Reset Time-out Period

150

μs

Figure 43-3. Zero-Power-on Reset Characteristics
VDDBU

Vth+
Vth-

Reset

1112

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series

Table 43-7.
Symbol

ISB

Parameter

Conditions

Typ

Max

Units

Standby current

@25°C onto VDDCORE = 1.8V
@85°C onto VDDCORE = 1.8V
@25°C onto VDDCORE = 1.95V
@85°C onto VDDCORE = 1.95V

<1
14
<1
15

1.5
40
1.8
50

μA
μA
μA
μA

128-Bit Mode Read Access:
Maximum Read Frequency onto VDDCORE = 1.8V @ 25 °C
Maximum Read Frequency onto VDDCORE = 1.95V @ 25 °C

15
20

20
25

mA
mA

64-Bit Mode Read Access:
Maximum Read Frequency onto VDDCORE = 1.8V @ 25 °C
Maximum Read Frequency onto VDDCORE = 1.95V @ 25 °C

7.5
10

10
12.5

mA
mA

Write onto VDDCORE = 1.8V @ 25 °C
Write onto VDDCORE = 1.95V @ 25 °C

3.6
5.0

4.5
6.0

mA
mA

Active current

ICC

43.3

DC Flash Characteristics

Power Consumption
• Power consumption of the device according to the different Low Power Mode Capabilities
(Backup, Wait, Sleep) and Active Mode.
• Power consumption on power supply in different modes: Backup, Wait, Sleep and Active.
• Power consumption by peripheral: calculated as the difference in current measurement after
having enabled then disabled the corresponding clock.

43.3.1

43.3.1.1

Backup Mode Current Consumption
The Backup Mode configuration and measurements are defined as follow.
Configuration A
• All Power supplies OFF, except VDDBU
• Supply Monitor on VDDUTMI is disabled
• RTT and RTC not used
• Embedded RC Oscillator used
• Wake-Up pin FWUP = VDDBU
• Current measurement on AMP1

1113
6430F–ATARM–21-Feb-12

Figure 43-4. Measurement Setup
VDDANA
VDDIO
VDDUTMI
AMP1
VDDBU
2.5, 3V, 3.3V
VDDCORE,
VDDPLL
XIN32

XOUT32

Table 43-8.

43.3.1.2

Power Consumption for Backup Mode Configuration A
Conditions

VDDBU Consumption (AMP1)

Unit

VDDBU = 3.3V @25°C
VDDBU = 3.0V @25°C
VDDBU = 2.5V @25°C
VDDBU = 1.8V @25°C

3.0
2.7
2.2
1.6

μA

Configuration B
• All Power supplies OFF, except VDDBU and VDDIO
• Supply Monitor on VDDUTMI is disabled
• RTC ON, RTT ON
• 32 KHz Crystal Oscillator used
• FWUP pin = VDDBU
• Wake-up pins WKUP0 to 15 = VDDIO
• Current measurement on AMP1 and on AMP2

1114

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 43-5. Measurement Setup
AMP1

VDDBU
VDDANA
VDDIO

AMP2

3.3V

VDDUTMI
VDDIN
Voltage
Regulator

VDDOUT

VDDCORE
VDDPLL

Table 43-9.

43.3.2

43.3.2.1

Power Consumption for Backup Mode Configuration B
Conditions

VDDBU Consumption
(AMP1)

Total Consumption
(AMP2)

VDDBU = 3.3V @25°C
VDDBU = 3.0V @25°C
VDDBU = 2.5V @25°C
VDDBU = 1.8V @25°C

3.0
2.7
2.2
1.6

3.06
2.75
2.25
1.64

Unit

μA

Wait and Sleep Mode Current Consumption
The Wait Mode and Sleep Mode configuration and measurements are defined below.
Sleep Mode
• All power supplies are powered
• Core Clock OFF
• Master Clock (MCK) running at various frequencies with PLLA or the fast RC oscillator.
• Fast start-up through WKUP0-15 pins
• Current measurement on AMP1 (VDDOUT=VDDCORE + VDDPLL)
• All peripheral clocks deactivated

1115
6430F–ATARM–21-Feb-12

Figure 43-6. Measurement Setup for Sleep Mode

VDDBU
VDDANA
VDDIO

AMP2

3.3V

VDDUTMI
VDDIN
AMP1

Voltage
Regulator

VDDOUT

VDDCORE
VDDPLL

Table 43-10 gives current consumption in typical conditions.

Table 43-10. Typical Current Consumption for Sleep Mode

Conditions
Figure 43-6 on page 1116 @25°C
MCK = 48 MHz
There is no activity on the I/Os of the device.

1116

VDDOUT
Consumption
(AMP1)

Total
Consumption
(AMP2)

Unit

11.8

11.9

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 43-7. Current Consumption in Sleep Mode (AMP1) versus Master Clock ranges (Condition from Table 43-10)

24
VDOUT (IDDCORE + IDDPLL) in mA

22
20
18
16
14
12
10
8
6
4
2
0
0

8 12 16 20 24 28 32 36 40 44 48 52 56 60 64 68 72 76 80 84 88 92 96
Processor and Peripheral Clocks in MHz

Table 43-11. Sleep mode Current consumption versus Master Clock (MCK) variation
Core Clock/MCK (MHz)

AMP1 (VDDOUT) Consumption

Unit

21.9

17.3

14.5

11.72

9.6

6.56

4.8

2.835

1.937

1.013

0.567

0.343

0.5

0.23

0.25

0.174

0.125

0.146

1117
6430F–ATARM–21-Feb-12

43.3.2.2

Wait Mode
• All power supplies are powered
• Core Clock and Master Clock Stopped
• Current measurement on AMP1, AMP2 and AMP3
• All Peripheral clocks deactivated
Figure 43-8.

Measurement Setup for Wait Mode
VDDBU
VDDANA
VDDIO

AMP2

3.3V

VDDUTMI
AMP3

VDDIN
AMP1

Voltage
Regulator

VDDOUT

VDDCORE
VDDPLL

Table 43-12 gives current consumption in typical conditions.

Table 43-12. Typical Current Consumption in Wait Mode

Conditions
See Figure 43-8 on page 1118 @25°C
There is no activity on the I/Os of the device.

43.3.3

VDDOUT
Consumption
(AMP1)

Total
Consumption
(AMP2)

Regulator and
Core
Consumption
(AMP3)

Unit

15.4

27.3

23.4

μA

Active Mode Power Consumption
The Active Mode configuration and measurements are defined as follows:
• VDDIO = 3.3V
• VDDCORE = 1.8V (Internal Voltage regulator used)
• TA = 25° C
• Application Running from Flash Memory with 128-bit access Mode
• All Peripheral clocks are deactivated.
• Master Clock (MCK) running at various frequencies with PLLA or the fast RC oscillator.
• Current measurement on AMP1 (VDDOUT = VDDCORE + VDDPLL) and total current on
AMP2

1118

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 43-9. Active Mode Measurement Setup

VDDBU
VDDANA
VDDIO

AMP2

3.3V

VDDUTMI
VDDIN
AMP1

Voltage
Regulator

VDDOUT

VDDCORE
VDDPLL

Tables below give Active Mode Current Consumption in typical conditions.
• VDDCORE at 1.8V
• Temperature = 25°C

Table 43-13. Master Clock (MCK) variation with PLLA
AMP1 (VDDOUT)
Consumption

Core Clock/MCK (MHz)

AMP2 (total)
Consumption

48.3

44.3

39.4

39.7

35.2

35.4

30.2

30.5

25.8

26.1

20.4

20.6

18.3

18.5

Unit

1119
6430F–ATARM–21-Feb-12

Table 43-14. Master Clock (MCK) Variation with Fast RC Oscillator
AMP1 (VDDOUT)
Consumption

Core Clock/MCK (MHz)

43.3.4

Unit

AMP2 (total)
Consumption

10.08

7.5

7.54

5.03

2.68

2.70

1.4

1.42

0.5

0.77

0.78

0.25

0.44

0.45

0.125

0.28

0.29

0.032

0.046

0.061

Peripheral Power Consumption in Active Mode
Table 43-15. Power Consumption on VDDCORE(1)
Peripheral

Consumption (Typ)

PIO Controller

Unit

USART

31.3

PWM

53.8

TWI

SPI

2.55

Timer Counter Channels

9.3
μA/MHz

ADC12B

17.45

ADC

15.7

HSMCI

SMC

78.8

SSC

16.3

UDPHS
Note:

1120

1. Note: VDDIO = 3.3V, VDDCORE = 1.80V, TA = 25° C

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
43.4

Crystal Oscillators Characteristics

43.4.1

32 kHz RC Oscillator Characteristics

Table 43-16. 32 kHz RC Oscillator Characteristics
Symbol

Parameter

Min

Typ

Max

Unit

RC Oscillator Frequency

kHz

Frequency Supply Dependency

-3

%/V

-11

Frequency Temperature Dependency
Duty

Duty Cycle

TON

Startup Time

IDDON

Current Consumption

43.4.2

Conditions

Over temperature range (-40°C/
+85°C) versus 25°C

After Startup Time
Temp. Range = -40°C to +85°C
Typical Consumption at 2.2V
Supply and Temp = 25°C

100

μs

540

870

Typ

Max

Unit

MHz

5.4
8.4
12.6

MHz

TBD

μs

120
160
210

μA

0.2

μA

4/8/12 MHz RC Oscillators Characteristics

Table 43-17. 4/8/12 MHz RC Oscillators Characteristics
Symbol

Parameter

FRange

RC Oscillator Frequency Range

Conditions

Min
4

1.62V= Oscillator Drive Level. Having a crystal drive level number
lower than the oscillator specification may damage the crystal.
• Equivalent Series Resistor (ESR)

– Crystal ESR <= Oscillator ESR Max. Having a crystal with ESR value higher than
the oscillator may cause the oscillator to not start.
• Shunt Capacitance
– Max. crystal Shunt capacitance <= Oscillator Shunt Capacitance (CSHUNT). Having a
crystal with ESR value higher than the oscillator may cause the oscillator to not start.
43.4.9.2

1126

Printed Circuit Board (PCB)
SAM3U Oscillators are low power oscillators requiring particular attention when designing PCB
systems. A board design example is given on Atmel’s website in the MCU Technical Center
Section.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
43.5

UPLL, PLLA Characteristics

Table 43-24. Supply Voltage Phase Lock Loop Characteristics
Symbol

Parameter

Conditions

Supply Voltage Range
VDDPLL
Allowable Voltage Ripple

Min

Typ

Max

Unit

1.6

1.8

30
10

Max

Unit

RMS Value 10 kHz to 10 MHz
RMS Value > 10 MHz

Table 43-25. PLLA Characteristics
Symbol

Parameter

FIN

Input Frequency

MHz

FOUT

Output Frequency

192

MHz

1.3
2
3

μA

200

μS

Max

Unit

IPLL

Current Consumption

Conditions

Min

Active mode @ 96MHz @1.8V
Active mode @ 160MHz @1.8V
Active mode @ 192MHz @1.8V

Typ

1
1.6
2.4

Standby mode
PLLA Settling Time

TSTART

Table 43-26. UPLL Characteristics for USB High Speed Device Port
Symbol

Parameter

FIN

Input Frequency

Delta FIN

Input Frequency Accuracy

FOUT

Output Frequency

IPLL

Current Consumption

Conditions

Min

12
See note (1)

-0.05

MHz
+0.05

480
Active mode @ 480MHz @1.8V

Note:

Typ

Standby mode

2.5

%
MHz

TBD

μA

1. Required for 12MHz Clock Signal injection on XIN pin (oscillator in bypass mode).

1127
6430F–ATARM–21-Feb-12

43.6

USB High Speed Port

43.6.1

Typical Connection
For typical connection please refer to the USB Device Section.

43.6.2

Electrical Characteristics

Table 43-27. Electrical Parameters
Symbol

Parameter

Conditions

RPUI

Bus Pull-up Resistor on Upstream Port (idle
bus)

in FS or HS Mode

1.5

kOhm

RPUA

Bus Pull-up Resistor on Upstream Port
(upstream port receiving)

in FS or HS Mode

kOhm

43.6.2.1

43.6.3

Min

Typ

Max

Unit

USB Transceiver
USB 2.0 Compliant in full-speed and high-speed modes. Refer to Chapter 7 of the USB 2.0,
Revision 2.0 April 27, 2000.
Static Power Consumption

Table 43-28. Static Power Consumption
Symbol

Parameter

IBIAS

Bias current consumption on VBG

Conditions

Min

Typ

Max

Unit

μA

Typ

Max

Unit

0.7

0.8

HS Transceiver & I/O current consumption
IVDDUTMII
Note:

FS / HS Transceiver & I/O current consumption

no connection

(1)

1. If cable is connected add 200 μA (Typical) due to Pull-up/Pull-down current consumption.

43.6.4

Dynamic Power Consumption

Table 43-29. Dynamic Power Consumption
Symbol

Parameter

IBIAS

Bias current consumption on VBG

IVDDUTMII

Note:

43.6.5

1128

Conditions

Min

HS Transceiver current consumption

HS transmission

HS Transceiver current consumption

HS reception

FS/HS Transceiver current consumption

FS transmission 0m
cable(1)

FS/HS Transceiver current consumption

FS transmission 5m
cable(1)

FS/HS Transceiver current consumption

FS reception(1)

4.5

1. Including 1 mA due to Pull-up/Pull-down current consumption.

USB High Speed Design Guidelines
In order to facilitate hardware design around the SAM3U USB High Speed Port, Atmel provides
an application note on www.atmel.com.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
43.7

12-Bit ADC Characteristics

Table 43-30. Analog Power Supply Characteristics
Symbol
VVDDIN

IVDDIN

Note:

Parameter

Conditions

Min

Typ

Max

Units

ADC Analog Supply

12-bit or 10 bit resolution

2.4

3.0

3.6

ADC Analog Supply

10 bit resolution

2.0

3.6

Max. Voltage Ripple

rms value, 10 kHz to 20 MHz

Current Consumption

Sleep Mode
Fast Wake Up Mode
Normal Mode (IBCTL= 00)
Normal Mode (IBCTL= 01)

1
2.6
7.1
9

μA
mA
mA
mA

Max

Units

0.1
1.8
4.7
6

Use IBCTL = 00 for Sampling Frequency below 500 kHz and IBCTL = 01 between 500 kHz and 1MHz.

Table 43-31. Channel Conversion Time and ADC Clock
Symbol

Parameter

Conditions

Min

fADC

ADC Clock Frequency

MHz

tCP_ADC

ADC Clock Period

1000

Sampling Frequency

0.05

MHz

From OFF Mode to Normal Mode:
- Voltage Reference OFF
- Analog Circuitry OFF
tSTART-UP

Typ

ADC Startup time

tTRACKTIM

Track and Hold Time

tCONV

Conversion Time

tSETTLE

Settling Time

μs
From Standby Mode to Normal Mode:
- Voltage Reference ON
- Analog Circuitry OFF

See Section “Track and Hold Time
versus Source Output Impedance” for
more details

160

ns
20

Settling time to change offset and gain

TCP_ADC

200

Table 43-32. External Voltage Reference Input
Parameter

Conditions

Min

Typ

Max

Units

ADVREF Input Voltage Range, 12-bit

2.4V < VVDDIN < 3.6V

2.4

VDDIN

ADVREF Input Voltage Range, 10-bit

2.0V < VVDDIN < 3.6V

2.0

VDDIN

250

μA

ADVREF Current
ADVREF Input DC impedance

kΩ

1129
6430F–ATARM–21-Feb-12

43.7.0.1

Static performance characteristics
Minimal code=0
Maximal code=4095
ADC resolution = 12 bits (4096)
In the following tables, the LSB is relative to analog scale:
• Single Ended (ex: ADVREF=3.0V),
– Gain = 1, LSB = (3.0V / 4096) = 732uV
– Gain = 2, LSB = (1.5V / 4096) = 366uV
– Gain = 4, LSB = (750mV / 4096) = 183uV
• Differential (ex: ADVREF=3.0V),
– Gain = 0.5, LSB = (6.0V / 4096) = 1465uV
– Gain = 1, LSB = (3.0V / 4096) = 732uV
– Gain = 2, LSB = (750mV / 4096) = 366uV

Table 43-33. INL, DNL, 12-bit mode, VDDIN supply voltage conditions, Temperature range [-40°, +100°], FADC=2 MHz,
IBCTL=01.
Parameter

Conditions

Min

Typ

Max

Units

Integral Non-linearity (INL)

VDDIN 2.4V to <3.0V
Differential, DIFF=1, OFF=x, GAIN=xx

-2.2

±1

+2.2

LSB

Integral Non-linearity (INL)

VDDIN 2.4V to <3.0V
Single ended, DIFF=0, OFF=x, GAIN=xx

-6

±1

LSB

Integral Non-linearity (INL)

VDDIN 3.0V to <3.6V
Differential, DIFF=1, OFF=x, GAIN=xx

-1.5

±1

+1.5

LSB

Integral Non-linearity (INL)

VDDIN 3.0V to <3.6V
Single ended, DIFF=0, OFF=x, GAIN=xx

-5

±1

LSB

Differential Non-linearity (DNL)

VDDIN 2.4V to <3.0V
Differential, DIFF=1, OFF=x, GAIN=xx

-1.5

±0.5

+1.5

LSB

Differential Non-linearity (DNL)

VDDIN 2.4V to <3.0V
Single ended, DIFF=0, OFF=x, GAIN=xx

-2.2

±0.5

+2.2

LSB

Differential Non-linearity (DNL)

VDDIN 3.0V to <3.6V
Differential, DIFF=1, OFF=x, GAIN=xx

-1.5

±0.5

+1.5

LSB

Differential Non-linearity (DNL)

VDDIN 3.0V to <3.6V
Single ended, DIFF=0, OFF=x, GAIN=xx

-2.2

±0.5

+2.2

LSB

x stand for digital 0 or 1

1130

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series

Table 43-34. Ge, Oe, 12-bit mode, VDDIN supply voltage conditions, Temperature range [-40°, +100°], FADC=2 MHz,
IBCTL=01.(1)
Parameter

Conditions

Min

Typ

Max

Units

Gain error (Ge)

VDDIN 2.4V to <3.6V
Differential, DIFF=1, OFF=x, GAIN=xx

-1.56
-64

-0.56
-23

+0.29
+12

%
LSB

Gain error (Ge)

VDDIN 2.4V to <3.6V
Single ended, DIFF=0, OFF=x, GAIN=xx

-1.56
-64

-0.56
-23

+0.78
+32

%
LSB

Offset error (Oe)

VDDIN 2.4V to <3.6V
Differential, DIFF=1, OFF=x, GAIN=xx

-30

+64

LSB

Offset error (Oe)

VDDIN 2.4V to <3.6V
Single ended, DIFF=0, OFF=x, GAIN=xx

-60

+80

LSB

x stand for digital 0 or 1
(1) Offset and Gain errors are given without calibration. A software calibration can be done to reduce Gain and offset
errors.

Figure 43-10. Offset and Gain Definitions

Ya=actual adc codes
Ymax=4095
Ya=Ga×Xi+Oe
Ga: actual gain

YaH

Ga=(YaH-YaL)/XiH
Ga=1+Ge(%)
YiM=2047

Ge(%): gain error

YaL

Ge(lsb): Ge(%)×Xmax

Oe=actual offset
0
Single Ended case:
Offset and Gain
definitions

VL
XiL

ADVref/2
XiM=2047

VH ADVref
XiH Xmax=4095

Vin
Xi=ideal adc codes

1131
6430F–ATARM–21-Feb-12

Table 43-35. Static Performance Characteristics -10 bits mode (1)
Parameter

Conditions

Min

Resolution

Typ

Max

Integral Non-linearity (INL)

Units
Bit

-1

±0.5

LSB

Differential Non-linearity (DNL)

no missing code

-1

±0.5

LSB

Offset Error

all gain, Differential or Single ended, no
calibration

-8

+20

LSB

Gain Error without calibration

all gain, Differential or Single ended, no
calibration

-16

-6

LSB

Min

Typ

Max

Units

Note:

1. Single ended or differential mode, any gain values.

LSB of 10 bit ADC, LSB=3.0V/1024

Table 43-36. Dynamic Performance Characteristics in Single ended and 12 bits mode (1)
Parameter

Conditions

Signal to Noise Ratio - SNR

Single ended, DIFF=0, OFF=x, GAIN=xx

Total Harmonic Distortion - THD

Single ended, DIFF=0, OFF=x, GAIN=xx

-66

-72

Signal to Noise and Distortion - SINAD

Single ended, DIFF=0, OFF=x, GAIN=xx

ENOB

Single ended, DIFF=0, OFF=x, GAIN=xx

9.0

10.0

Bits

Note:

1. ADC Clock (FADC) = 20MHz, Fs=1MHz, Fin = 127 kHz, IBCTL = 01, FFT using 1024 points or more, Frequency band =
[1kHz, 500kHz] – Nyquist conditions fulfilled.

Table 43-37. Dynamic Performance Characteristics in Differential and 12 bits mode(1)
Parameter

Conditions

Min

Typ

Signal to Noise Ratio - SNR

Differential, DIFF=1, OFF=x, GAIN=xx

Total Harmonic Distortion - THD

Differential, DIFF=1, OFF=x, GAIN=xx

-69

-80

Signal to Noise and Distortion - SINAD

Differential, DIFF=1, OFF=x, GAIN=xx

ENOB

Differential, DIFF=1, OFF=x, GAIN=xx

9.5

10.3

Bits

Note:

Max

Units

ADC Clock (FADC)= 20MHz, Fs=1MHz, Fin=127kHz, IBCTL = 01, FFT using 1024 points or more, Frequency band = [1kHz,
500kHz] – Nyquist conditions fulfilled.

Track and Hold Time versus Source Output Impedance
The following figure gives a simplified acquisition path.

1132

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 43-11. Simplified Acquisition Path
ADC
Input
Zsource

Mux.

Sample & Hold
12-bit
ADC
Core

Ron

Csample

During the tracking phase the ADC needs to track the input signal during the tracking time
shown below:
• 10-bit mode: tTRACK = 0.042 x Zsource + 160
• 12-bit mode: tTRACK = 0.054 x Zsource + 205
With tTRACK expressed in ns and ZSOURCE expressed in Ohms.
Two cases must be considered:
1. The calculated tracking time (tTRACK) is lower than 15 tCP_ADC.
Set TRANSFER = 1 and TRACTIM = 0.
In this case, the allowed Zsource can be computed versus the ADC frequency with
the hypothesis of tTRACK = 15 × tCP_ADC:
Where tCP_ADC = 1/fADC . See Table 43-38 on page 1133.
Table 43-38. Source impedance values
fADC = ADC clock (MHz)

ZSOURCE (kOhms) for 12 bits

ZSOURCE (kOhms) for 10 bits

20.00

16.00

10.67

8.00

6.40

5.33

4.57

4.00

3.56

3.20

108

2.91

119

2.67

100

130

2.46

109

141

2.29

118

152

2.13

126

164

2.00

135

175

1.00

274

353

1133
6430F–ATARM–21-Feb-12

2. The calculated tracking time (tTRACK) is higher than 15 tCP_ADC.
Set TRANSFER = 1 and TRACTIM = 0.
In this case, a timer will trigger the ADC in order to set the correct sampling rate according to the
Track time.
The maximum possible sampling frequency will be defined by tTRACK in nano seconds, computed
by the previous formula but with minus 15 × tCP_ADC and plus TRANSFER time.
• 10 bit mode: 1/fS = tTRACK - 15 × tCP_ADC + 5 tCP_ADC
• 12 bit mode: 1/fS = tTRACK - 15 × tCP_ADC + 5 tCP_ADC
Note:

Csample and Ron are taken into account in the formulas

Table 43-39. Analog Inputs
Parameter
Input Voltage Range

Min
0

Input Leakage Current
Input Capacitance
Note:

Typ

Max

Units

VADVREF
±0.5

μA

1. Input Voltage range can be up to VDDIN without destruction or over-consumption.
If VDDIO < VADVREF max input voltage is VDDIO.

ADC Application Information
For more information on data converter terminology, please refer to the application note:
Data Converter Terminology, Atmel lit° 6022.
http://www.atmel.com/dyn/resources/prod_documents/doc6022.pdf

1134

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
43.8

AC Characteristics

43.8.1

Master Clock Characteristics

Table 43-40. Master Clock Waveform Parameters
Symbol

Parameter

1/(tCPMCK)

Master Clock Frequency

43.8.2

Conditions

Min

Max

VDDCORE @ 1.62V

VDDCORE @ 1.8V

Units
MHz

I/O Characteristics
Criteria used to define the maximum frequency of the I/Os:
– output duty cycle (40%-60%)
– minimum output swing: 100 mV to VDDIO - 100 mV
– minimum output swing: 100 mV to VDDIO - 100 mV
– Addition of rising and falling time inferior to 75% of the period

Table 43-41. I/O Characteristics
Symbol

FreqMax1

PulseminH1

PulseminL1

Parameter

Conditions

Min

30 pF

VDDIO = 1.62V
VDDIO = 3.0V

45
65

45 pF

VDDIO = 1.62V
VDDIO = 3.0V

34
45

30 pF

VDDIO = 1.62V
VDDIO = 3.0V

11
7.7

45 pF

VDDIO = 1.62V
VDDIO = 3.0V

14.7
11

30 pF

VDDIO = 1.62V
VDDIO = 3.0V

11
7.7

45 pF

VDDIO = 1.62V
VDDIO = 3.0V

14.7
11

Pin Group 1 (1) Maximum output frequency

Pin Group 1 (1) High Level Pulse Width

Pin Group 1 (1) Low Level Pulse Width

FreqMax2

Pin Group 2 (2) Maximum output frequency

Load: 25 pF
1.62V < VDDIO < 3.6V

PulseminH2

Pin Group 2 (2) High Level Pulse Width

Load: 25pF
1.62V < VDDIO < 3.6V

14.5

PulseminL2

Pin Group 2 (2) Low Level Pulse Width

Load: 25pF
1.62V < VDDIO < 3.6V

14.5

Notes:

Max

Units

MHz

MHz
ns
ns

1. Pin Group 1 = PA3, PA15
2. Pin Group 2 = PA[0-2], PA[4-14], PA[16-31], PB[0-31], PC[0-31]

Table 43-42. NRSTB
Symbol

Parameter

Filtered Pulse Width

Tuf

Unfiltered Pulse Width

Conditions

Min

100

Typ

Max

Units

μs
μs

1135
6430F–ATARM–21-Feb-12

43.8.3

SPI Characteristics
Figure 43-12. SPI Master Mode with (CPOL= NCPHA = 0) or (CPOL= NCPHA= 1)

SPCK

SPI1

SPI0
MISO

SPI2
MOSI

Figure 43-13. SPI Master Mode with (CPOL = 0 and NCPHA=1) or (CPOL=1 and NCPHA= 0)

SPCK

SPI4

SPI3
MISO

SPI5
MOSI

Figure 43-14. SPI Slave Mode with (CPOL=0 and NCPHA=1) or (CPOL=1 and NCPHA=0)
NPCS0

SPI13

SPI12
SCK

SPI6
MISO

SPI7

SPI8

MOSI

1136

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 43-15. SPI Slave Mode with (CPOL = NCPHA = 0) or (CPOL= NCPHA= 1)
NPCS0

SPI14
SPI15

SCK

SPI9
MISO

SPI10

SPI11

MOSI

43.8.3.1

Maximum SPI Frequency
The following formulas give maximum SPI frequency in Master read and write modes and in
Slave read and write modes.

Master Write Mode
The SPI is only sending data to a slave device such as an LCD, for example. The limit is
given by SPI2 (or SPI5) timing. Since it gives a maximum frequency above the maximum pad
speed (see Section 43.8.2 “I/O Characteristics”), the max SPI frequency is the one from the
pad.
Master Read Mode
1
f SPCK Max = -------------------------------------------------------SPI 0 ( orSPI 3 ) + T valid
Tvalid is the slave time response to output data after detecting an SPCK edge. For Atmel SPI
DataFlash (AT45DB642D), Tvalid (orTv ) is 12 ns Max.
In the formula above, FSPCKMax = 38.5 MHz @ VDDIO = 3.3V.
Slave Read Mode
In slave mode, SPCK is the input clock for the SPI. The max SPCK frequency is given by
setup and hold timings SPI7/SPI8(or SPI10/SPI11). Since this gives a frequency well above
the pad limit, the limit in slave read mode is given by SPCK pad.
Slave Write Mode
1
f SPCK Max = -------------------------------------------------------SPI 6 ( orSPI 9 ) + T setup
For 3.3V I/O domain and SPI6, FSPCKMax = 33 MHz. Tsetup is the setup time from the master
before sampling data.
43.8.3.2

SPI Timings

Table 43-43. SPI Timings
Symbol
SPI0

Parameter
MISO Setup time before SPCK rises (master)

Conditions

Min

3.3V domain(1)

(2)

1.8V domain

Max

Units

1137
6430F–ATARM–21-Feb-12

Table 43-43. SPI Timings (Continued)
Symbol
SPI1

Parameter
MISO Hold time after SPCK rises (master)

Conditions

Min

3.3V domain(1)

1.8V domain(2)

(1)

SPI2

SPCK rising to MOSI Delay (master)

SPI3

MISO Setup time before SPCK falls (master)

SPI4
SPI5
SPI6

MISO Hold time after SPCK falls (master)
SPCK falling to MOSI Delay (master)
SPCK falling to MISO Delay (slave)

3.3V domain

1.8V domain(2)

SPI8
SPI9

MOSI Setup time before SPCK rises (slave)
MOSI Hold time after SPCK rises (slave)
SPCK rising to MISO Delay (slave)

SPI10

MOSI Setup time before SPCK falls (slave)

SPI11

MOSI Hold time after SPCK falls (slave)

SPI12
SPI13

NPCS Setup time to SPCK (slave)
NPCS Hold time after SPCK (slave)

SPI15
Notes:

NPCS Setup time to SPCK (slave)
NPCS Hold time after SPCK (slave)

3.5

(2)

1.8V domain

3.3V domain(1)

(2)

1.8V domain

3.3V domain(1)

ns
3

(2)

3.5

(1)

3.3V domain

1.8V domain(2)

1.8V domain

3.3V domain

0.5

(2)

1.8V domain

3.3V domain(1)

0.5

1.8V domain(2)

(1)

3.3V domain

1.8V domain(2)

(1)

3.3V domain

1.8V domain(2)

3.3V domain(1)

1.5

(2)

1.8V domain

1.5

3.3V domain(1)

5.2

(2)

1.8V domain

3.3V domain(1)

1.8V domain(2)

(1)

SPI14

Units

3.3V domain(1)

(1)

SPI7

Max

3.3V domain

4.2

1.8V domain(2)

(1)

(2)

3.3V domain

1.8V domain

1. 3.3V domain: VVDDIO from 3.0V to 3.6V, maximum external capacitor = 30 pF.
2. 1.8V domain: VVDDIO from 1.65V to 1.95V, maximum external capacitor = 30 pF.

Note that in SPI master mode the SAM3U does not sample the data (MISO) on the opposite edge where data clocks out
(MOSI) but the same edge is used as shown in Figure 43-12 and Figure 43-13.
43.8.4

MCI Timings
The High Speed MultiMedia Card Interface (HSMCI) supports the MultiMedia Card (MMC)
Specification V4.3, the SD Memory Card Specification V2.0, the SDIO V2.0 specification and
CE-ATA V1.1.

43.8.5

SSC Timings
Timings are given assuming the following VDDIO supply and load.
VDDIO = 1.62V @25pF

1138

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
VDDIO = 3V @25pF
Figure 43-16. SSC Transmitter, TK and TF as output

TK (CKI =0)

TK (CKI =1)

SSC0
TF/TD

Figure 43-17. SSC Transmitter, TK as input and TF as output

TK (CKI =0)

TK (CKI =1)

SSC1
TF/TD

Figure 43-18. SSC Transmitter, TK as output and TF as input
TK (CKI=0)

TK (CKI=1)

SSC2

SSC3

SSC4
TD

1139
6430F–ATARM–21-Feb-12

Figure 43-19. SSC Transmitter, TK and TF as input
TK (CKI=1)

TK (CKI=0)

SSC5

SSC6

TF
SSC7
TD

Figure 43-20. SSC Receiver RK and RF as input
RK (CKI=0)

RK (CKI=1)

SSC8

SSC9

RF/RD

Figure 43-21. SSC Receiver, RK as input and RF as output
RK (CKI=1)

RK (CKI=0)

SSC8

SSC9

RD
SSC10
RF

1140

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 43-22. SSC Receiver, RK and RF as output
RK (CKI=1)

RK (CKI=0)

SSC11

SSC12

RD
SSC13
RF

Figure 43-23. SSC Receiver, RK as output and RF as input
RK (CKI=0)

RK (CKI=1)

SSC11

SSC12

RF/RD

1141
6430F–ATARM–21-Feb-12

43.8.5.1

SSC Timings

Table 43-44. SSC Timings
Symbol

Parameter

Condition

Min

Max

Units

Transmitter
SSC0

TK edge to TF/TD (TK output, TF output)

1.8v domain(3)
3.3v domain(4)

0(2)
0(2)

2 (2)
1.5(2)

SSC1

TK edge to TF/TD (TK input, TF output)

1.8v domain(3)
3.3v domain(4)

5(2)
5(2)

17.5 (2)
14.5(2)

SSC2

TF setup time before TK edge (TK output)

1.8v domain(3)
3.3v domain(4)

17 - tCPMCK
15 - tCPMCK

SSC3

TF hold time after TK edge (TK output)

1.8v domain(3)
3.3v domain(4)

tCPMCK - 5
tCPMCK - 5

SSC4(1)

TK edge to TF/TD (TK output, TF input)

1.8v domain(3)
3.3v domain(4)

0 (+2*tCPMCK)(1)(2)
0 (+2*tCPMCK)(1)(2)

SSC5

TF setup time before TK edge (TK input)

1.8v domain(3)
3.3v domain(4)

0
0

SSC6

TF hold time after TK edge (TK input)

1.8v domain(3)
3.3v domain(4)

tCPMCK
tCPMCK

SSC7(1)

TK edge to TF/TD (TK input, TF input)

1.8v domain(3)
3.3v domain(4)

5 (+3*tCPMCK)(1)(2)
5 (+3*tCPMCK)(1)(2)

2 (+2*tCPMCK)(1)(2)
2 (+2*tCPMCK)(1)(2)

18 (+3*tCPMCK)(1)(2)
15 (+3*tCPMCK)(1)(2)

Receiver
SSC8

RF/RD setup time before RK edge (RK input)

1.8v domain(3)
3.3v domain(4)

0
0

SSC9

RF/RD hold time after RK edge (RK input)

1.8v domain(3)
3.3v domain(4)

tCPMCK
tCPMCK

SSC10

RK edge to RF (RK input)

1.8v domain(3)
3.3v domain(4)

5(2)
5(2)

SSC11

RF/RD setup time before RK edge (RK output)

1.8v domain(3)
3.3v domain(4)

17 - tCPMCK
15 - tCPMCK

SSC12

RF/RD hold time after RK edge (RK output)

1.8v domain(3)
3.3v domain(4)

tCPMCK - 5
tCPMCK - 5

SSC13

RK edge to RF (RK output)

1.8v domain(3)
3.3v domain(4)

0(2)
0(2)

Notes:

18(2)
15(2)

3(2)
2(2)

1. Timings SSC4 and SSC7 depend on the start condition. When STTDLY = 0 (Receive start delay) and START = 4, or 5 or
7(Receive Start Selection), two Periods of the MCK must be added to timings.
2. For output signals (TF, TD, RF), Min and Max access times are defined. The Min access time is the time between the TK (or
RK) edge and the signal change. The Max access timing is the time between the TK edge and the signal stabilization. Figure
43-24 illustrates Min and Max accesses for SSC0. The same applies for SSC1, SSC4, and SSC7, SSC10 and SSC13.
3. 1.8V domain: VVDDIO from 1.65V to 1.95V, maximum external capacitor = 25 pF.
4. 3.3V domain: VVDDIO from 3.0V to 3.6V, maximum external capacitor = 25 pF.

1142

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 43-24. Min and Max Access Time of Output Signals

TK (CKI =1)

TK (CKI =0)
SSC0min
SSC0max
TF/TD

43.8.6

SMC Timings
SMC Timings are given with the following conditions.
VDDIO = 1.62V @ 30 pF
VDDIO = 3V @ 50 pF
Timings are given assuming a capacitance load on data, control and address pads:
In the following tables tCPMCK is MCK period. Timing extraction

43.8.6.1

Read Timings

Table 43-45. SMC Read Signals - NRD Controlled (READ_MODE = 1)
Symbol

Parameter

Min

VDDIO Supply

1.8V

(2)

Max
(3)

3.3V

1.8V

(2)

Units
(3)

3.3V

NO HOLD SETTINGS (nrd hold = 0)
SMC1

Data Setup before NRD High

SMC2

Data Hold after NRD High

17.5

HOLD SETTINGS (nrd hold ≠ 0)
SMC3

Data Setup before NRD High

SMC4

Data Hold after NRD High

HOLD or NO HOLD SETTINGS (nrd hold ≠ 0, nrd hold = 0)
SMC5

NBS0/A0, NBS1, NBS2/A1, NBS3,
A2 - A25 Valid before NRD High

SMC6

NCS low before NRD High

SMC7

NRD Pulse Width

(nrd setup +
nrd pulse)*
tCPMCK + 7

(nrd setup +
nrd pulse)*
tCPMCK + 6.5

(nrd setup +
nrd pulse - ncs
rd setup) *
tCPMCK + 8

(nrd setup +
nrd pulse - ncs
rd setup) *
tCPMCK + 7

nrd pulse *
tCPMCK - 5

1143
6430F–ATARM–21-Feb-12

Table 43-46. SMC Read Signals - NCS Controlled (READ_MODE= 0)
Symbol

Parameter

Min

Max

(2)

VDDIO supply

3.3V

1.8V

(3)

1.8V

(2)

Units
(3)

3.3V

NO HOLD SETTINGS (ncs rd hold = 0)
SMC8

Data Setup before NCS High

SMC9

Data Hold after NCS High

HOLD SETTINGS (ncs rd hold ≠ 0)
SMC10

Data Setup before NCS High

SMC11

Data Hold after NCS High

HOLD or NO HOLD SETTINGS (ncs rd hold ≠ 0, ncs rd hold = 0)
NBS0/A0, NBS1, NBS2/A1, NBS3,
A2 - A25 valid before NCS High

(ncs rd setup +
ncs rd pulse)*
tCPMCK - 3.5

(ncs rd setup
+ ncs rd
pulse)* tCPMCK
-3

SMC13

NRD low before NCS High

(ncs rd setup +
ncs rd pulse nrd setup)*
tCPMCK - 2

(ncs rd setup
+ ncs rd pulse
- nrd setup)*
tCPMCK - 2

SMC14

NCS Pulse Width

ncs rd pulse
length * tCPMCK
-5

SMC12

43.8.6.2

Write Timings

Table 43-47. SMC Write Signals - NWE Controlled (WRITE_MODE = 1)
Min
Symbol

(2)

Parameter

1.8V

Max
3.3V

(3)

1.8V

(2)

3.3V(3)

Units

HOLD or NO HOLD SETTINGS (nwe hold ≠ 0, nwe hold = 0)
SMC15

Data Out Valid before NWE High

nwe pulse *
tCPMCK - 4

nwe pulse *
tCPMCK - 3.5

SMC16

NWE Pulse Width

nwe pulse *
tCPMCK - 5

SMC17

NBS0/A0 NBS1, NBS2/A1, NBS3,
A2 - A25 valid before NWE low

nwe setup *
tCPMCK + 8

nwe setup *
tCPMCK + 7

(nwe setup ncs rd setup +
nwe pulse) *
tCPMCK + 2

(nwe setup ncs rd setup +
nwe pulse) *
tCPMCK + 3

SMC18

1144

NCS low before NWE high

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Table 43-47. SMC Write Signals - NWE Controlled (WRITE_MODE = 1) (Continued)
Min
Symbol

1.8V(2)

Parameter

Max
3.3V(3)

1.8V(2)

3.3V(3)

Units

HOLD SETTINGS (nwe hold ≠ 0)
SMC19

NWE High to Data OUT, NBS0/A0
NBS1, NBS2/A1, NBS3, A2 - A25
change

SMC20

NWE High to NCS Inactive (1)

nwe hold *
tCPMCK - 5.5

(nwe hold - ncs
wr hold)* tCPMCK
-3

(nwe hold - ncs
wr hold)*
tCPMCK - 3

NO HOLD SETTINGS (nwe hold = 0)
NWE High to Data OUT, NBS0/A0
NBS1, NBS2/A1, NBS3, A2 - A25,
NCS change(1)

SMC21

Table 43-48. SMC Write NCS Controlled (WRITE_MODE = 0)
Min

Max

1.8V(2)

3.3V(3)

Data Out Valid before NCS High

ncs wr pulse *
tCPMCK - 3

ncs wr pulse *
tCPMCK - 2

SMC23

NCS Pulse Width

ncs wr pulse *
tCPMCK - 5

SMC24

NBS0/A0 NBS1, NBS2/A1, NBS3,
A2 - A25 valid before NCS low

ncs wr setup *
tCPMCK - 3

ncs wr setup *
tCPMCK - 2.5

SMC25

NWE low before NCS high

(ncs wr setup nwe setup +
ncs pulse)*
tCPMCK - 2.5

(ncs wr setup nwe setup +
ncs pulse)*
tCPMCK - 2

SMC26

NCS High to Data Out, NBS0/A0,
NBS1, NBS2/A1, NBS3, A2 - A25,
change

ncs wr hold *
tCPMCK - 6.5

ncs wr hold *
tCPMCK - 5.5

SMC27

NCS High to NWE Inactive

(ncs wr hold nwe hold)*
tCPMCK - 5

(ncs wr hold nwe hold)*
tCPMCK - 4.5

Symbol

Parameter

SMC22

Notes:

1.8V(2)

3.3V(3)

Units

1. hold length = total cycle duration - setup duration - pulse duration. “hold length” is for “ncs wr hold length” or “NWE hold
length”.
2. 1.8V domain: VDDIO from 1.65V to 1.95V, maximum external capacitor = 25 pF
3. 3.3V domain: VDDIO from 3.0V to 3.6V, maximum external capacitor = 25 pF.

1145
6430F–ATARM–21-Feb-12

Figure 43-25. SMC Timings - NCS Controlled Read and Write

SMC12

SMC26

SMC24

A0/A1/NBS[3:0]/A2-A25
SMC13

SMC13

NRD

SMC14

NCS

SMC14

SMC8

SMC9

SMC23

SMC10

SMC11

SMC22

SMC26

D0 - D15
SMC27

SMC25

NWE

NCS Controlled READ
with NO HOLD

NCS Controlled READ
with HOLD

NCS Controlled WRITE

Figure 43-26. SMC Timings - NRD Controlled Read and NWE Controlled Write
SMC21

SMC17

SMC5

SMC5
SMC17

SMC19

A0/A1/NBS[3:0]/
A2-A25
SMC6

SMC21 SMC6

SMC18

SMC20

NCS

NRD

SMC7

SMC1

SMC2

SMC15

SMC21

SMC3

SMC4

SMC15

SMC19

D0 - D31

NWE

SMC16

NRD Controlled READ
with NO HOLD

1146

NWE Controlled WRITE
with NO HOLD

SMC16

NRD Controlled READ
with HOLD

NWE Controlled WRITE
with HOLD

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Timings are given with the following conditions.
VDDIO = 1.62V and 3V
SCK/MISO/MOSI Load = 30 pF
Figure 43-27. USART SPI Master Mode
NPCSx
SPI5

SPI3
CPOL=1
SPI0
SCK
CPOL=0
SPI4
MISO

SPI4

SPI1

SPI2
LSB

MSB

MOSI

• the MOSI line is driven by the output pin TXD
• the MISO line drives the input pin RXD
• the SCK line is driven by the output pin SCK
• the NSS line is driven by the output pin RTS
Figure 43-28. USART SPI Slave mode (Mode 1 or 2)

NPCS0

SPI13

SPI12
SCK

SPI6
MISO

SPI7

SPI8

MOSI

• the MOSI line drives the input pin RXD
• the MISO line is driven by the output pin TXD
• the SCK line drives the input pin SCK
• the NSS line drives the input pin CTS

1147
6430F–ATARM–21-Feb-12

Figure 43-29. USART SPI Slave mode (Mode 0 or 3)

NPCS0

SPI14
SPI15

SCK

SPI9
MISO

SPI10

SPI11

MOSI

Table 43-49. USART SPI Timings
Symbol

Parameter

Conditions

Min

Max

Units

Master Mode
SPI0

SCK Period

1.8v domain
3.3v domain

TBD

SPI1

Input Data Setup Time

1.8v domain
3.3v domain

TBD

SPI2

Input Data Hold Time

1.8v domain
3.3v domain

TBD
TBD

SPI3

Chip Select Active to Serial Clock

1.8v domain
3.3v domain

TBD
TBD

SPI4

Output Data Setup Time

1.8v domain
3.3v domain

TBD
TBD

SPI5

Serial Clock to Chip Select Inactive

1.8v domain
3.3v domain

TBD
TBD

Slave Mode
SPI6

SCK falling to MISO

1.8V domain
3.3V domain

TBD
TBD

SPI7

MOSI Setup time before SCK rises

1.8V domain
3.3V domain

TBD
TBD

SPI8

MOSI Hold time after SCK rises

1.8v domain
3.3v domain

TBD
TBD

SPI9

SCK rising to MISO

1.8v domain
3.3v domain

TBD
TBD

1148

SAM3U Series

TBD

6430F–ATARM–21-Feb-12

SAM3U Series
Table 43-49. USART SPI Timings
Symbol

Parameter

Conditions

Min

SPI10

MOSI Setup time before SCK falls

1.8v domain
3.3v domain

TBD
TBD

SPI11

MOSI Hold time after SCK falls

1.8v domain
3.3v domain

TBD
TBD

SPI12

NPCS0 setup to SCK rising

1.8v domain
3.3v domain

TBD
TBD

SPI13

NPCS0 hold after SCK falling

1.8v domain
3.3v domain

TBD
TBD

SPI14

NPCS0 setup to SCK falling

1.8v domain
3.3v domain

TBD
TBD

SPI15

NPCS0 hold after SCK rising

1.8v domain
3.3v domain

TBD
TBD

Notes:

Max

Units

1. 1.8V domain: VDDIO from 1.65V to 1.95V, maximum external capacitor = 25 pF
2. 3.3V domain: VDDIO from 3.0V to 3.6V, maximum external capacitor = 25 pF.

1149
6430F–ATARM–21-Feb-12

43.8.7
Two-wire Serial Interface Characteristics
Table 30 describes the requirements for devices connected to the Two-wire Serial Bus. For timing symbols refer to Figure
43-30.
Table 43-50. Two-wire Serial Bus Requirements
Symbol

Parameter

VIL
VIH
Vhys

Hysteresis of Schmitt Trigger Inputs

VOL

Output Low-voltage

Rise Time for both TWD and TWCK

tof

Output Fall Time from VIHmin to VILmax

Ci(1)

Capacitance for each I/O Pin

fTWCK

TWCK Clock Frequency

Condition

Min

Max

Input Low-voltage

-0.3

0.3 VVDDIO

Input High-voltage

0.7xVVDDIO

VCC + 0.3

0.150

–

3 mA sink current

10 pF < Cb < 400 pF
Figure 43-30

0.4

20 + 0.1Cb(1)(2)

300

20 + 0.1Cb(1)(2)

250

–

400

kHz

1000ns
------------------Cb

fTWCK ≤100 kHz
Rp

tLOW

Value of Pull-up resistor

Low Period of the TWCK clock

tHIGH

High period of the TWCK clock

tHD;STA

Hold Time (repeated) START Condition

tSU;STA

Set-up time for a repeated START condition

tHD;DAT

Data hold time

tSU;DAT

Setup time for STOP condition

tHD;STA

Hold Time (repeated) START Condition

Note:

V VDDIO – 0,4V
-------------------------------------3mA

fTWCK > 100 kHz

V VDDIO – 0,4V
-------------------------------------3mA

300ns
---------------Cb

fTWCK ≤100 kHz

(3)

–

μs

fTWCK > 100 kHz

(3)

–

μs

fTWCK ≤100 kHz

(4)

–

μs

fTWCK > 100 kHz

(4)

–

μs

fTWCK ≤100 kHz

tHIGH

–

μs

fTWCK > 100 kHz

tHIGH

–

μs

fTWCK ≤100 kHz

tHIGH

–

μs

fTWCK > 100 kHz

tHIGH

–

μs

fTWCK ≤100 kHz

3 x TCP_MCK

μs

fTWCK > 100 kHz

3 x TCP_MCK

μs

fTWCK ≤100 kHz

tLOW - 3 x
TCP_MCK

–

fTWCK > 100 kHz

tLOW - 3 x
TCP_MCK

–

fTWCK ≤100 kHz

tHIGH

–

μs

fTWCK > 100 kHz

tHIGH

–

μs

fTWCK ≤100 kHz

tHIGH

–

μs

fTWCK > 100 kHz

tHIGH

–

μs

Data setup time

tSU;STO

Units

1. Required only for fTWCK > 100 kHz.
2. Cb = capacitance of one bus line in pF. Per I2C Standard compatibility, Cb Max = 400pF

3. The TWCK low Period is defined as follow: T low = ( ( CLDIV × 2

CKDIV

4. The TWCK high period is defined as follows: T high = ( ( CHDIV × 2

) + 4 ) × T MCK

CKDIV

) + 4 ) × T MCK

5. TCP_MCK = MCK Bus Period.

1150

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Figure 43-30. Two-wire Serial Bus Timing
tHIGH

tof
tLOW

tr
tLOW

TWCK
tSU;STA

tHD;STA

tHD;DAT

tSU;DAT

tSU;STO

TWD

tBUF

43.8.8
Embedded Flash Characteristics
The maximum operating frequency is given in tables 43-51 and 43-52 below but is limited by the Embedded Flash access
time when the processor is fetching code out of it. The tables 43-51 and 43-52 below give the device maximum operating
frequency depending on the field FWS of the MC_FMR register. This field defines the number of wait states required to
access the Embedded Flash Memory.
Note: The embedded flash is fully tested during production test, the flash contents is not set to a known state prior to shipment. Therefore, the flash contents should be erased prior to programming an application.
Table 43-51. Embedded Flash Wait State VDDCORE set at 1.62V
FWS

Read Operations

Maximum Operating Frequency (MHz)

1 cycle

2 cycles

3 cycles

4 cycles

Table 43-52. Embedded Flash Wait State VDDCORE set at 1.80V
FWS

Read Operations

Maximum Operating Frequency (MHz)

1 cycle

2 cycles

3 cycles

4 cycles

Table 43-53. AC Flash Characteristics
Parameter

Conditions

Min

Typ

Max

Units

per page including auto-erase

4.6

per page without auto-erase

2.3

11.5

Program Cycle Time
Full Chip Erase
Data Retention

10
Not Powered or Powered

Years
30K

Write/Erase cycles @ 25°C
Endurance

cycles
Write/Erase cycles @ 85°C

10K

1151
6430F–ATARM–21-Feb-12

44. SAM3U4/2/1 Mechanical Characteristics
44.1

100-lead LQFP Package

Figure 44-1. 100-lead LQFP Package Drawing

Table 44-1.

Device and LQFP Package Maximum Weight

SAM3UE4/2/1

Table 44-2.

800

Package Reference

JEDEC Drawing Reference

MS-026

JESD97 Classification

Table 44-3.

LQFP Package Characteristics

Moisture Sensitivity Level

This package respects the recommendations of the NEMI User Group.

1152

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
44.2

100-ball TFBGA Package
Figure 44-2. 100-ball TFBGA Package Drawing

Table 44-4.

Soldering Information (Substrate Level)

Ball Land

TBD

Soldering Mask Opening

TBD

Table 44-5.

Device Maximum Weight

TBD

Table 44-6.

100-ball Package Characteristics

Moisture Sensitivity Level

Table 44-7.

Package Reference

JEDEC Drawing Reference

TBD

JESD97 Classification

1153
6430F–ATARM–21-Feb-12

44.3

144-lead LQFP Package

Figure 44-3. 144-lead LQFP Package Drawing

Notes:

1. This drawing is for general information only; refer to JEDEe Drawing MS-026 for additional information.
2. The top package body size may be smaller than the bottom package size by as much as 0.15 mm.
3. Dimensions D1 and E1 do not include mold protrusions. Allowable protrusion is 0.25 mm per side. D1 and E1 are maximum
plastic body size dimensions including mold mismatch.
4. b dimension by more than 0.08 mm. Dambar cannot be located on the lower radius or the foot. Minimum space between protrusion and an adjacent lead is 0.07 mm for 0.4 and 0.5 mm pitch packages.
5. These dimensions apply to the flat section of the lead between 0.10 mm and 0.25 mm from the lead tip.
6. A1 is defined as the distance from the seating place to the lowest point on the package body.

Table 44-8.

Device Maximum Weight

TBD

Table 44-9.

144-lead Package Characteristics

Moisture Sensitivity Level

Table 44-10. Package Reference
JEDEC Drawing Reference

MS-026

JESD97 Classification

1154

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
44.4

144-ball LFBGA Package

Figure 44-4. 144-ball LFBGA Package Drawing

All dimensions are in mm.

Table 44-11. Soldering Information (Substrate Level)
Ball Land

0.380 mm

Soldering Mask Opening

0.280 mm

Table 44-12. Device and 144-ball BGA Package Maximum Weight
300

Table 44-13. 144-ball BGA Package Characteristics
Moisture Sensitivity Level

Table 44-14.
JEDEC Drawing Reference

none

JESD97 Classification

This package respects the recommendations of the NEMI User Group.
1155
6430F–ATARM–21-Feb-12

44.5

Soldering Profile
Table 44-15 gives the recommended soldering profile from J-STD-020C.
Table 44-15. Soldering Profile
Profile Feature

Green Package

Average Ramp-up Rate (217°C to Peak)

3° C/sec. max.

Preheat Temperature 175°C ±25°C

180 sec. max.

Temperature Maintained Above 217°C

60 sec. to 150 sec.

Time within 5° C of Actual Peak Temperature

20 sec. to 40 sec.

Peak Temperature Range

260° C

Ramp-down Rate

6° C/sec. max.

Time 25° C to Peak Temperature

8 min. max.

Note:

The package is certified to be backward compatible with Pb/Sn soldering profile.

A maximum of three reflow passes is allowed per component.

44.6

Packaging Resources
Land Pattern Definition.
Refer to the following IPC Standards:
• IPC-7351A and IPC-782 (Generic Requirements for Surface Mount Design and Land Pattern
Standards) http://landpatterns.ipc.org/default.asp
• Atmel Green and RoHS Policy and Package Material Declaration Data Sheet
http://www.atmel.com/green/

1156

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
45. Ordering Information

Table 45-1.

ATSAM3U4/2/1 Ordering Information

Ordering Code

MRL

Flash
(Kbytes)

Package

Package Type

ATSAM3U4EA-AU

256

LQFP144

Green

Industrial
-40°C to 85°C

ATSAM3U4EA-CU

256

LFBGA 144

Green

Industrial
-40°C to 85°C

ATSAM3U4CA-AU

256

LQFP 100

Green

Industrial
-40°C to 85°C

ATSAM3U4CA-CU

256

TFBGA100

Green

Industrial
-40°C to 85°C

ATSAM3U2EA-AU

128

LQFP144

Green

Industrial
-40°C to 85°C

ATSAM3U2EA-CU

128

LFBGA144

Green

Industrial
-40°C to 85°C

ATSAM3U2CA-AU

128

LQFP100

Green

Industrial
-40°C to 85°C

ATSAM3U2CA-CU

128

TFBGA100

Green

Industrial
-40°C to 85°C

ATSAM3U1EA-AU

LQFP144

Green

Industrial
-40°C to 85°C

ATSAM3U1EA-CU

LFBGA144

Green

Industrial
-40°C to 85°C

ATSAM3U1CA-AU

LQFP100

Green

Industrial
-40°C to 85°C

ATSAM3U1CA-CU

TFBGA100

Green

Industrial
-40°C to 85°C

ATSAM3U1EB-AU

LQFP144

Green

Industrial
-40°C to 85°C

ATSAM3U1EB-CU

LFBGA144

Green

Industrial
-40°C to 85°C

ATSAM3U1CB-AU

LQFP100

Green

Industrial
-40°C to 85°C

ATSAM3U1CB-CU

TFBGA100

Green

Industrial
-40°C to 85°C

Temperature
Operating Range

1157
6430F–ATARM–21-Feb-12

1158

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
46. SAM3U Series Errata
46.1

Marking
All devices are marked with the Atmel logo and the ordering code.
Additional marking is as follows:

YYWW
V
XXXXXXXXX

ARM

where
• “YY”: manufactory year
• “WW”: manufactory week
• “V”: revision
• “XXXXXXXXX”: lot number

At the time of publication, in addition to the following “SAM3U Errata - Rev. A Parts”, there is an
errata sheet on engineering samples of the SAM3U devices. This document is available on
Atmel’s web site:
http://www.atmel.com/dyn/resources/prod_documents/doc6483.pdf

1159
6430F–ATARM–21-Feb-12

46.2

SAM3U Errata - Rev. A Parts
Revision A parts Chip IDs are as follows:
SAM3U4C (Rev A)

0x28000961

SAM3U2C (Rev A)

0x280A0761

SAM3U1C (Rev A)

0x28090561

SAM3U4E (Rev A)

0x28100961

SAM3U2E (Rev A)

0x281A0761

SAM3U1E (Rev A)

0x28190561

Refer also to Section 46.1 “Marking” on page 1159
46.2.1
46.2.1.1

Flash
Flash: Issue Running at Frequency Lower than 2.5 MHz
When the system clock (MCK) is lower than 2.5 MHz with 2 Wait States (WS) programmed in
the EEFC Flash Mode Register (EEFC_FMR), the Cortex fetches erroneous instructions.
Problem Fix/Workaround
Do not use 2 WS when running at a frequency lower than 2.5 MHz.

46.2.1.2

Flash: Read Flash in 64-bit Mode
Higher than expected power consumption can be seen when reading Flash in 64-bit mode.
Problem Fix/Workaround
Use 128-bit mode instead.

46.2.1.3

Flash: Flash Programming
When writing data into the Flash memory plane (either through the EEFC, using the IAP function, or FFPI), the data may not be correctly written (i.e the data written is not the one expected).
Problem Fix/Workaround
Set the number of Wait States (WS) to 6 (FWS=6) during the programming.

46.2.1.4

Flash: Fetching Error after Reading the Unique Identifier
After reading the Unique Identifier (or using the STUI/SPUI command), the processor may fetch
wrong instructions. It depends on the code and on the region of the code.
Problem Fix/Workaround
In order to avoid this problem, follow the steps below:
1. Set bit 16 of EEFC Flash Mode Register to 1
2. Send the Start Read Unique Identifier command (STUI) by writing the Flash Command
Register with the STUI command.
3. Wait for the FRDY bit to fall
4. Read the Unique ID (and next bits if required)
5. Send the Stop Read Unique Identifier command (SPUI) by writing the Flash Command
Register with the SPUI command.
6. Wait for the FRDY bit to rise
7. Clear bit 16 of EEFC Flash Mode Register
Note:

1160

During the sequence, the software cannot run out of Flash (so needs to run out of SRAM).

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
46.2.2
46.2.2.1

12-bit Analog-to-Digital Converter (ADC12B)
ADC12B: Current Consumption in Backup Mode on VDDANA
In Backup mode, the current consumption on VDDANA is around 1.0 mA instead of 0.1 μA
Problem Fix/Workaround
Four workarounds are possible:
1. Do not supply VDDANA and VDDIO in Backup mode using an external switch managed
by SHDN pin.
2. Do not supply VDDANA in Backup mode using an external switch managed by the
SHDN and set all PIOs with ADC inputs (PA22, PA30, PB3-PB8, PC15-PC18, PC28C21) at low level (either externally or by software).
3. Do not supply VDDANA in Backup mode using an external switch managed by any PIO
and set all PIOs with ADC inputs (PA22, PA30, PB3-PB8, PC15-PC18, PC28-C21) at
low level (either externally or by software). Since the PIO state is preserved when in
backup mode, any free PIO line can be used to switch off the external switch by driving
the PIO line at low level (PIO is input, pull-up enabled after backup reset).
4. Use wait mode instead of Backup mode.

46.2.2.2

ADC: Trigger Launches only One Conversion
A start command initiates a conversion sequence of one channel but not all activated channels
as expected.
Problem Fix/Workaround
Send as many start commands as the number of activated channels, or use free run mode.

46.2.2.3

ADC12B: Saturation
When the ADC12B works in saturation (measurements below 0V or above ADREF) the results
may be erratic, the value deviation can be around 30 LSB to the expected data.
Problem Fix/Workaround
None.

1161
6430F–ATARM–21-Feb-12

46.2.2.4

ADC: Wrong first Conversions
The first conversions done by the ADC may be erroneous if the maximum gain (x4 in single
ended or x2 in differential mode) is not used. The issue appears after the power-up or if a conversion has not occured for 1 minute.
Problem Fix/Workaround
Three workarounds are possible :
1) Perform 16 dummy conversions on one channel (whatever conditions used in term of setup of
gain, single/differential, offset, and channel selected). The next conversions will be correct for
any channels and any settings. Note that these dummy conversions need to be performed if no
conversion has occured for 1 minute or for a new chip start-up.
2) Perform a dummy conversion on a single ended channel on which an external voltage of
ADVREF/2 (+/-10%) is applied. Use the following conditions for this conversion: gain at 4, offset
set at 1. The next conversions will be correct for any channels and any settings. Note that this
dummy conversion needs to be performed if no conversion has occured for 1 minute or for a
new chip start-up.
3) Perform a dummy conversion on a differential channel on which the two inputs are connected
together and connected to any voltage (from 0 to ADVREF). Use the following conditions for this
conversion: gain at 4, offset set at 1. The next conversions will be correct for any channels and
any settings. Note that this dummy conversion needs to be performed if no conversion has
occured for 1 minute or for a new chip start-up.

46.2.3
46.2.3.1

Power Management Controller (PMC)
PMC: Main Oscillator Frequency selection if the Main On-chip RC Oscillator is OFF
When the 4/8/12 MHz RC Oscillator is off, the frequency selection (MOSCRCF bitfield in
CKGR_MOR) can not be changed. The register can be written but the modification to
MOSCRCF will not be taken into account.
Problem Fix/Workaround
Modify MOSCRCF while 4/8/12 MHz RC Oscillator is on (MOSCREN = 1).

1162

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
46.3

SAM3U Errata - Rev. B Parts
Revision B parts Chip IDs are as follows:
SAM3U1C (Rev B) 0x28090562
SAM3U1E (Rev B) 0x28190562
Refer also to Section 46.1 “Marking” on page 1159

46.3.1
46.3.1.1

46.3.1.2

46.3.1.3

During the sequence, the software cannot run out of Flash (so needs to run out of SRAM).

1163
6430F–ATARM–21-Feb-12

46.3.2
46.3.2.1

46.3.2.2

1164

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
46.3.2.3

46.3.3
46.3.3.1

1165
6430F–ATARM–21-Feb-12

1166

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
47. Revision History
In the tables that follow, the most recent version of the document appears first.
Note:

Doc.
Rev.
6430F

“rfo” indicates changes requested during document review and approval loop.

Comments

Change
Request
Ref.

PMC:
Section 28.9 ”Fast Startup”, SUPC_FSMR --> PMC_FSMR and SUPC_FSPR --> PMC_FSPR
Section 28.3 ”Master Clock Controller”, removed bogus sentence about Master Clock divider functionality
Section 28.1 ”Description”, changed sentence “Processor Clock (HCLK), must be switched off...”

8010
rfo
8217

Memories:
Section 9.1.3.1 ”Flash Overview”, Flash size should be 64KBytes instead of 256KBytes

8029

Electrical Characteristics:
Section 43.7, 43.7.1 Gain and Offset Calibration removed
8031
Section 43.4 ”Crystal Oscillators Characteristics”:
8174
Table 43-16, “32 kHz RC Oscillator Characteristics”, changed parameter ‘Frequency Temperature Dependency’
Table 43-4, “Core Power Supply Brownout Detector Characteristics”, changed MAX value of VTH+
Section 43.8.8 ”Embedded Flash Characteristics”, added note regarding erasing Flash contents
8223
Errata:
Section 46.3 ”SAM3U Errata - Rev. B Parts”, added errata section for rev. B
Section 46.2 ”SAM3U Errata - Rev. A Parts”: Section 46.2.2.2 ”ADC: Trigger Launches only One Conversion”,
added errata
Section 46.3 ”SAM3U Errata - Rev. B Parts”: Section 46.3.2.2 ”ADC: Trigger Launches only One Conversion”,
added errata
Section 46.2 ”SAM3U Errata - Rev. A Parts”: Section 46.2.2.4 ”ADC: Wrong first Conversions”, added errata
Section 46.3 ”SAM3U Errata - Rev. B Parts”: Section 46.3.2.3 ”ADC: Wrong first Conversions”, added errata

8131
rfo
rfo
8164

Overview & Mechanical Characteristics:
Replaced all occurrences of '100-ball LFBGA' into '100-ball TFBGA'

8044

Ordering Information:
Table 45-1, “ATSAM3U4/2/1 Ordering Information”, updated with MRL B devices

8130

SPI:
Section 32.8.11 ”SPI Write Protection Status Register”, description of register simplified
Section 32.8.10 ”SPI Write Protection Mode Register”, removed ‘SPI’ from register description names
Section 32. ”Serial Peripheral Interface (SPI) Programmer Datasheet”, SPI version updated to version ‘R’

8136
rfo
rfo

TC:
Section 36.1 ”Description”: Table 36-1, “Timer Counter Clock Assignment”, footnote for table updated

8159

1167
6430F–ATARM–21-Feb-12

Doc.
Rev.
6430E

Change
Request
Ref.

Comments
Overview:
Comment in front of rows PA24 and PA25 removed, and put as a footnote(3) for TWD1 and TWCK1.
Table 11-2, “Multiplexing on PIO Controller A (PIOA)”, “Peripheral B” column, PA2 and PA17 texts exchanged.
Figure 5-5 ”Fast Start-Up Sources”, ‘Falling/Rising Edge Detector’ changed to ‘High/Low Level Detector’ in 3
blocks.
Wait mode consumption updated in Table 5-1, “Low Power Mode Configuration Summary”
Clock Generator:
Last sentence removed from Section 27.5.1 ”Divider and Phase Lock Loop Programming”, as PLLADIV2 does
not exist.
Third bullet edited in Section 27.1 ”Description”.

7751
7908

rfo

SUPC:
Section 19.3.2 ”Slow Clock Generator”, variable VDD_SUPPLY_MONITOR changed to VDD_Backup, in order
to get VDDBU instead of VDDUTMI.

7743

USART:
Section 35. ”Universal Synchronous Asynchronous Receiver Transmitter (USART)”, PDC condition shown
instead of DMA.

1168

rfo

Electrical Characteristics:
3 values updated in Table 43-12, “Typical Current Consumption in Wait Mode”.
Section 43.7.1 ”Sample and Hold Time versus Source Output Impedance” created.

PMC:
Section 28.8 ”Programmable Clock Output Controller”, UPLLCK --> UPLLCK/2.
Figure 28-1, “General Clock Block Diagram” edited (UPLLCK --> UPLLCK/2, and UPLLDIV /1/2 --> Divider /2)

Doc.
Rev.
6430D

7724
7954
7922

7898
7912
7804

Errata:
Section 46.2.1.4 ”Flash: Fetching Error after Reading the Unique Identifier” added.

7978

Comments

Change
Request
Ref.

Overview:
Section 5.5.2 ”Wait Mode”, sentence starting with ‘By configuring...’ --> ‘This is done by configuring...’

7492

Section 10.13 ”Chip Identification”, (Rev A) was removed from Table 10-1.

7642

TWD1 and TWCK1 removed from Figure 2-2, “100-pin SAM3U4/2/1C Block Diagram”.

7624

Table 3-1, “Signal Description List”, Note (4) added to TDO Output.

7635

A typo fixed in Section 9.1.1 ”Internal SRAM”: 4224 Kbytes --> 4224 bytes.

7305

Debug and Test Features:
Table 14-1, “Debug and Test Signal List”, Note (1) added to TDO Output.
Section 14.4.8 ”ID Code Register”, Chip ID and JTAG ID Code edited.
Section 14.4.7 ”IEEE 1149.1 JTAG Boundary Scan”, second paragraph updated.

7635
7543
7485

Chip ID:
CHIPID_CIDR column in Table 29-1, “ATSAM3U Chip IDs Register” updated as in Table 10-2

7642

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Doc.
Rev.
6430D

Comments (Continued)

Change
Request
Ref.

Clock Generator:
Section 27.6 ”UTMI Phase Lock Loop Programming”, first sentence edited, together with Table 27-5, “UTMI PLL
7484
Block Diagram”.
PMC:
Section 28.9 ”Fast Startup”, a sentence starting with ‘Important’ added as a second paragraph.
In the first paragraph, ‘LPM bit is at 0’ replaced by ‘LPM bit is at 1’.

7539

Section 28.14.15 ”PMC Status Register”, MOSCSELS bit descriptions reversed.

7389

Electrical Characteristics:
In Table 43-18, “32.768 kHz Crystal Oscillator Characteristics”, CLEXT Maximum value is 22 instead of 20.

7589

Section 43.8.7 ”USART in SPI Mode Timings” updated: UART --> USART, SPCK --> SCK, Figures and titles
updated.
Errata:
Section 46.2.2.1 ”ADC12B: Current Consumption in Backup Mode on VDDANA” edited.
Section 46.2.1.3 ”Flash: Flash Programming” added.
Backpage:
A typo fixed: ‘tehincal’ --> ‘technical’

7320

7420
7205
7536

1169
6430F–ATARM–21-Feb-12

Doc.
Rev.
6430C

Change
Request
Ref.

Comments
Overview:
Section 2. ”SAM3U Block Diagram”, changed orientation of block diagrams.
Section 5. ”Power Considerations”, fixed grammar in Voltage ranges.

rfo

Section 3. ”Signal Description”, USART signal DCD0 is an Input

6681

Figure 5-1, “Single Supply”, Main supply range is 1.8V-3.6V.
Figure 5-1, Figure 5-2, Figure 5-3, updated “Note” below figures, “With Main Supply <2.0V USB and ADC are
not usable.

6698

Section 5.5 ”Low Power Modes”, stray references to WUPx pins, renamed WKUPx

6711

Table 5-1, “Low Power Mode Configuration Summary”, updated footnote “5”.

6964

Table 11-2, “Multiplexing on PIO Controller A (PIOA)”, TWD1 and TWCK1 only available on 144-pin version.

6686

Section 10.13 ”Chip Identification”
Table 10-2, “SAM3U Chip IDs Register - Revision A Parts”, added to datasheet.
Section 12.4 ”Universal Synchronous Asynchronous Receiver Transmitter (USART)”...”SCK up to MCK/6”

6951
rfo→7097

CHIP ID:
”ARCH: Architecture Identifier” bit field updated with SAM3S, SAM3N identifiers. Bitfield tables reorganized.

6967

CORTEX:
Section 13.19 ”Nested Vectored Interrupt Controller”, text on 3rd bullet updated: “Level detection of interrupt
signals.”

6823

EFC:
Section 21.3.3.7 ”Unique Identifier”, Unique Identifier is located ... 0x80000-0x8000F.
Figure 21-6, “Example of Partial Page Programming” text added below figure,..”works only with 128-bit (or
higher) boundaries...”

rfo
6827

HSMCI:
Section 37.10 ”HSMCI Boot Operation Mode”, added precision on boot from internal Flash.
”SDCBUS: SDCard/SDIO Bus Width”, bitfield table updated.
Values, Names, Descriptions updated in bitfield description tables. (global)
Section 37.13.18 ”HSMCI Write Protect Mode Register”, ASCI code is for MCI.

6745
7125
rfo

MATRIX:
Section 24.6.6 ”Write Protect Status Register”, WPVS identified in bitfield description.

6718

PMC:
“PMC Fast Startup Polarity Register”, FSTPx: bitfield typo fixed.
“PMC Clock Generator PLLA Register”, removed STMODE bitfield.
Section 28.11 ”Programming Sequence” Step 5 “Selection of Programmable Clocks”, updated: “Four clock
options are available: main clock, slow clock, PLLACK and UPLLCK. By default, the clock source selected is
main clock.”
Section 28.10 ”Clock Failure Detector”, updated with new information.
PWM:
Section 38.6.23 ”PWM Fault Mode Register”, Section 38.6.24 ”PWM Fault Status Register”, Section 38.6.25
”PWM Fault Clear Register”, Section 38.6.27 ”PWM Fault Protection Enable Register”, bitfield descriptions
updated.
Table 38-2, “I/O Lines”, new to datasheet.

1170

6688
6706

7128
6824

rfo

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Doc.
Rev.
6430C

Comments (Continued)

Change
Request
Ref.

RTC:
Section 18.3.2 ”Interrupt”, updated.
Section 18.5 ”Real Time Clock (RTC) User Interface”, the reset for RTC_CALR is 0x01210720.
TIMEVSEL, CALEVSEL bitfield descriptions reorganized.

7071
7046/7087
6796

SSC:
Redundant letter C removed from title.
SUPC:
Section 19.4 ”Supply Controller (SUPC) User Interface”, offset updated for GPBR: 0x90-0xDC.
FWUPDBC, WUPDBC bitfield descriptions reorganized.
Backup supply is VDDBU
TC:
Figure 36-2 ”Clock Chaining Selection”, channel 1 updated.
updated bitfields: TC0XC0S, TC1XC1S, TC2XC2S, TCCLKS, BURST, ETRGEDG, LDRA, LDRB, TCCLKS,
BURST, EEVTEDG, EEVT, WAVSEL, ACPA, ACPC, AEEVT, ASWTRG, BCPB, BCPC, BEEVT, BSWTRG
UDHP:
Figure 39-4 ”Logical Address Space for DPR Access”, EP0 has but 1 bank
Figure 39-1 ”Block Diagram”, 1 PMC to UTMI signal line. Notes removed.
Figure 39.4 ”Product Dependencies”, added to datasheet.
Figure 39-6 ”Register Mapping”, DMA offset updated to 0x300 + channel *...
USART:
Section 35.6.7 ”Modem Mode”, is available.
Section 35.6 ”Functional Description”, ...SCK up to MCK/6
Section 35.6.8.2 ”Baud Rate”
SPI Master Mode: ...”the value programmed in CD must be superior or equal to 6.”
SPI Slave Mode:...”the external clock (SCK) frequency must be at least 6 times lower than the system clock.”
Section 35.6.1 ”Baud Rate Generator”,”...signal provided on SCK must be at least 3 times lower than MCK in
USART mode, or 6 in SPI mode.”
Section 35.6.1.3 ”Baud Rate in Synchronous Mode or SPI Mode”, ...”limits the SCK maximum frequency to
MCK/3 in USART mode, or MCK/6 in SPI mode.”

6949

6950
6796
6714
6687
6796

6750
6792
rfo
6822
6791
rfo→7097

1171
6430F–ATARM–21-Feb-12

Doc.
Rev.
6430C

Change
Request
Ref.

Comments (Continued)
Electrical Characteristics:
Table 43-18, “32.768 kHz Crystal Oscillator Characteristics”, last row of the table changed from CL to CPARA
“Internal Parasitic Capacitance”. Schematic and Crystal formula updated
Table 43-21, “3 to 20 MHz Crystal Oscillator Characteristics”, CLEXT row, Parameter cell updated. Added 1K
resistor to Schematic and Crystal formula updated.

6701

Table 43-16, “32 kHz RC Oscillator Characteristics”, table updated.
6951
Table 43-17, “4/8/12 MHz RC Oscillators Characteristics”, updated “Frequency Temperature Dependency”
Table 43-31, “Channel Conversion Time and ADC Clock”, TSTART-UP “From OFF mode...” startup times updated.
Table 43-32, “External Voltage Reference Input”, ADVREF changed to AD12BVREF.
Table 43-37, “Analog Inputs”, VADVREF changed to VAD12BVREF
Errata:
Section 46.2 ”SAM3U Errata - Rev. A Parts”, added to Errata.

1172

7098

6951

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
.

Doc Rev
6430B

Comments

Change
Request
Ref.

Introduction:
Section 1. ”SAM3U Description”, Updated: 52 Kbytes of SRAM. 4x USARTs (SAM3U1C/2C/4C have 3), up to 6400
2x TWIs (SAM3U1C/2C/4C have 1), up to 5x SPIs SAM3U1C/2C/4C have 4),
Table 1-1, “Configuration Summary” EBI column updated, 8 bits for SAM3U1C/2C/4C
SAM3U4/3/2C rows FWUP replaces NO in FWUP,SHDN pins column.
6642
Figure 2-1 ”144-pin SAM3U4/2/1E Block Diagram” and Figure 2-2 ”100-pin SAM3U4/2/1C Block Diagram”
updated, SM cell removed; UART moved to peripheral area, added Flash Unique block, removed 12B from
ADC block, added SysTick counter and Fmax 96MHz to M3 block. FWUP replaces WKUP in fig 2-1, FWUP
added to fig 2-2.
Figure 2-2 ”100-pin SAM3U4/2/1C Block Diagram”, NWR1/NBS1, NXRP0, A0 removed from block diagram,

6482/6642

Table 3-1, “Signal Description List”, Schmit Trigger added ”PIO Controller - PIOA - PIOB - PIOC”. exception
details given in footnote.
VDDIN, VDDOUT added to table.
”Serial Wire/JTAG Debug Port (SWJ-DP)” replaced ICE and JTAG. This section of the table updated
status of pulldowns and pullups specified.

6480

Section 4. ”Package and Pinout”; reorganized according to product.
Section 4.1 ”SAM3U4/2/1E Package and Pinout”, Section 4.2 ”SAM3U4/2/1C Package and Pinout”, pinouts
finalized in datasheet.

6471/rfo
6607

Section 5.5.1 ”Backup Mode”, BOD replaced by Supply Monitor/SM.
Figure 5-4 ”Wake-up Source”, BODEN replaced by SMEN. FWUP →Falling Edge Detector.
Table 5-1, “Low Power Mode Configuration Summary”, PIO state in Low Power Mode, backup mode is;
“Previous state saved.”
Figure 5-3 ”Backup Batteries Used”, FWUP replaces FWKUP.

rfo

Section 6.6 ”NRSTB Pin”, VDDIO changed to VDDBU
Section 6. ”Input/Output Lines” replaces Section 5.8 “Programmable I/O Lines”.
Section 6.1 ”General Purpose I/O Lines (GPIO)” and Section 6.2 ”System I/O Lines” replace Section 6. “I/O
Line Considerations”.
Figure 6-1 ”On-Die Termination schematic”, added.
Section 6.8 “PIO Controllers”, removed.
Section 8. ”Product Mapping”, title changed from “Memories”.
Section 9. ”Memories”; now comprises Section 9.1 ”Embedded Memories” and Section 9.2 ”External
Memories”
Section 9.1.3.5 ”Security Bit Feature”, updated

6646
6481/rfo

Table 7-3, “SAM3U Master to Slave Access”, Slave 9, High Speed Peripheral Bridge line added.
Section 7.2 ”APB/AHB Bridges”, reference to ADC updated “10-bit ADC, 12-bit ADC (ADC12B)”.
Table 11-3, “Multiplexing on PIO Controller B (PIOB)” ADC12B2, ADC12B3 properly listed.
Section 12.10.1 ”12-bit High Speed ADC”, Section 12.10.2 ”10-bit Low Power ADC”, titles changed.
“Quadrature Decoder Logic” on page 51 properly stated in list of TC functions.

6663
6397

Section 12.10.1 ”12-bit High Speed ADC”, 2nd item on list updated.
Section 12.10.2 ”10-bit Low Power ADC”, Ksample values updated on 2nd item of list.

rfo

ADC12B:
Section 41.6.6 ”ADC12B Analog Control Register”,IBCTL reasigned to fields 8 and 9

rfo/

rfo

6645
6642

6649

1173
6430F–ATARM–21-Feb-12

Doc Rev
6430B

Change
Request
Ref.

Comments (Continued)
CORTEX-M3:
Table 13-31, “Priority grouping”updated.
Section 13.19.7.1 ”IP27”, title changed to IP27, value in bitfields 0 to 7 changed to IP28.

6394:

Section 13.20.6 ”Application Interrupt and Reset Control Register”read/write values to VECTKEY changed

6436

Section 13.5.2.2 ”Non Maskable Interrupt (NMI)” added to datasheet.
Section 13.6.2 ”Fault escalation and hard faults”, last sentence updated with NMI function.
Section 13.3.5 ”Data types” Condition tags sorted out.
Table 13-11, “Faults”updated footnote 1
Table 13-29, “CMSIS functions for NVIC control”Description in 4th row updated.

6483

on page 53, “...copyright ARM Ltd., 2008 - 2009.” precise years given.
Table 13-4, “Memory access behavior”last row assinged to Reserved in Memory Map.
Big Endien not used in this product.

rfo

Debug and Test:
Section 14.1 ”Overview”, SWJ-DP...also embeds a serial trace.
Table 14-1, “Debug and Test Signal List”, reorganized.

rfo

FFPI:
Section 22.2.5.4 ”Flash Lock Commands”, last sentence removed from 2nd paragraph (ref to EA command).

6677

HSMCI:
Section 37.12 ”Write Protection Registers”, Section 37.13.18 ”HSMCI Write Protect Mode Register” and
Section 37.13.19 ”HSMCI Write Protect Status Register”,added.

6432

MATRIX:
Table 24-1, “Register Mapping”, added offsets for Write Protection Registers

6431

PDC:
Section 26. ”Peripheral DMA Controller (PDC)” Section replaced.

6468

PIO:
Section 30.5.1 ”Write Protection Registers”, Section 30.6.42 ”PIO Write Protect Mode Register” and Section
30.6.43 ”PIO Write Protect Status Register” added with links to protected registers.
PMC:
Section 28.14.16 ”PMC Interrupt Mask Register”, Section 28.14.13 ”PMC Interrupt Enable Register”, Section
28.14.14 ”PMC Interrupt Disable Register”, 0 and 1read or write values described.
Section 28.13 ”Write Protection Registers”, Section 28.14.20 ”PMC Write Protect Mode Register” and
Section 28.14.20 ”PMC Write Protect Mode Register” added with links to protected registers.
Section 28.10 ”Clock Failure Detector”, added sentence on Fast RC
Section 28.11 ”Programming Sequence”, added step 1 and reordered subsequent numbering sequence.
removed code example, bitfield names updated.
PWM:
Table 38-4, “Fault Inputs”, “PWM Fault Input Number” column, typos fixed
RSTC:
Section 16.2 ”Block Diagram”, “backup_nreset” replaces “core_backup_reset”.
Section 16.3.1 ”Reset Controller Overview”, RSTC_MR...”is powered with VDDBU...”
Section 16.3.4.2 ”Backup Reset”, “The vddcore_nreset signal is asserted by the SUPC...” and
“backup_nreset” replaces “core_backup_reset”.
SMC:
Table 25-4, “External Memory Mapping”, restored rows for chip selects 0, 1, 2 to the table.

1174

6430

6311/rfo
6432
6469
6591
rfo
6397

rfo

6397

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Doc Rev
6430B

Comments (Continued)

Change
Request
Ref.

SPI:
Section 32.7.10 ”SPI Write Protection Control Register” and Section 32.7.11 ”SPI Write Protection Status
Register”, added.

6432

SSC:
Section 31.7.1 ”Write Protection Registers”, Section 31.8.17 ”SSC Write Protect Mode Register” and Section
31.8.18 ”SSC Write Protect Status Register”

6429

SUPC:
Figure 19-5 ”Raising the VDDUTMI Power Supply”, updated, i.e., shutdown polarity changed, Fast RC
Oscillator, NRST, periph_nreset, proc_nreset signals added, detailed startup time added up to instruction
fetch.
Figure 19-6 ”NRSTB Reset”, updated SHDN / vr_standby waveform.
Section 19.4.8 ”Supply Controller Status Register”, added note concerning “status register flag reset”.
Figure 19-2 ”Separated Backup Supply Powering Scheme”
USART:
Section 35.7.2 ”USART Mode Register”, VAR_SYNC bitfield description updated.
Section 35.7.1 ”USART Control Register” RSTSTA description updated with UNRE bit.
Section ”Receive NACK Inhibit”, updated. Section 35.7.1 ”USART Control Register”, NACK description,
grammar fixed.
Section 35.6.10 ”Write Protection Registers”, Section 35.7.16 ”USART Write Protect Mode Register” and
Section 35.7.17 ”USART Write Protect Status Register”, added.
Electrical Characteristics:
Section 43. ”SAM3U4/2/1 Electrical Characteristics”, updated
Section 43.8.3.1 ”Maximum SPI Frequency”, added
Figure 43-5 ”Measurement Setup”, updated.
Table 43-41, “I/O Characteristics”, in the Conditions column: VDDIO = 1.62V
Mechanical Characteristics:
Section 44.5 ”Soldering Profile”, added.

6604
rfo
6653
6642

6282
6283
6422
6432

rfo
6663

rfo

1175
6430F–ATARM–21-Feb-12

Doc.
Rev.

Date

Comments

6430A

24-Mar-09

First Issue - advance information

16-May-09

Review

29-May-09

Approved

1176

Change
Request
Ref.

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
Features ..................................................................................................... 1
1

ATSAM3U4/2/1 Description ..................................................................... 2
1.1 Configuration Summary ...........................................................................................2

ATSAM3U4/2/1 Block Diagram ................................................................ 3

Signal Description .................................................................................... 5
3.1 Design Considerations ............................................................................................8

Package and Pinout ................................................................................. 9
4.1 SAM3U4/2/1E Package and Pinout .........................................................................9
4.2 SAM3U4/2/1C Package and Pinout ......................................................................12

Power Considerations ........................................................................... 15
5.1 Power Supplies ......................................................................................................15
5.2 Voltage Regulator ..................................................................................................15
5.3 Typical Powering Schematics ................................................................................15
5.4 Active Mode ...........................................................................................................19
5.5 Low Power Modes .................................................................................................19
5.6 Wake-up Sources ..................................................................................................22
5.7 Fast Start-Up .........................................................................................................23

Input/Output Lines ................................................................................. 24
6.1 General Purpose I/O Lines (GPIO) ........................................................................24
6.2 System I/O Lines ...................................................................................................24
6.3 Serial Wire JTAG Debug Port (SWJ-DP) ..............................................................24
6.4 Test Pin .................................................................................................................25
6.5 NRST Pin ...............................................................................................................25
6.6 NRSTB Pin ............................................................................................................25
6.7 ERASE Pin ............................................................................................................25

Processor and Architecture .................................................................. 26
7.1 ARM Cortex-M3 Processor ....................................................................................26
7.2 APB/AHB Bridges ..................................................................................................26
7.3 Matrix Masters .......................................................................................................26
7.4 Matrix Slaves .........................................................................................................27
7.5 Master to Slave Access .........................................................................................27
7.6 DMA Controller ......................................................................................................28
7.7 Peripheral DMA Controller .....................................................................................28
7.8 Debug and Test Features ......................................................................................29
i

6430F–ATARM–21-Feb-12

Product Mapping .................................................................................... 30

Memories ................................................................................................ 31
9.1 Embedded Memories ............................................................................................31
9.2 External Memories .................................................................................................34

10 System Controller .................................................................................. 36
10.1 System Controller and Peripheral Mapping .........................................................38
10.2 Power-on-Reset, Brownout and Supply Monitor .................................................38
10.3 Reset Controller ...................................................................................................38
10.4 Supply Controller .................................................................................................38
10.5 Clock Generator ..................................................................................................39
10.6 Power Management Controller ............................................................................40
10.7 Watchdog Timer ..................................................................................................41
10.8 SysTick Timer ......................................................................................................41
10.9 Real-time Timer ...................................................................................................41
10.10 Real-time Clock .................................................................................................41
10.11 General-Purpose Back-up Registers .................................................................41
10.12 Nested Vectored Interrupt Controller .................................................................41
10.13 Chip Identification ..............................................................................................42
10.14 PIO Controllers ..................................................................................................42

11 Peripherals .............................................................................................. 43
11.1 Peripheral Identifiers ............................................................................................43
11.2 Peripheral Signal Multiplexing on I/O Lines .........................................................44

12 Embedded Peripherals Overview ......................................................... 48
12.1 Serial Peripheral Interface (SPI) ..........................................................................48
12.2 Two Wire Interface (TWI) ....................................................................................48
12.3 Universal Asynchronous Receiver Transceiver (UART) ......................................48
12.4 Universal Synchronous Asynchronous Receiver Transmitter (USART) ..............49
12.5 Serial Synchronous Controller (SSC) ..................................................................49
12.6 Timer Counter (TC) .............................................................................................49
12.7 Pulse Width Modulation Controller (PWM) ..........................................................50
12.8 High Speed Multimedia Card Interface (HSMCI) .................................................51
12.9 USB High Speed Device Port (UDPHS) ..............................................................51
12.10 Analog-to-Digital Converter (ADC) ....................................................................51

13 ARM Cortex® M3 Processor .................................................................. 53

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
13.1 About this section ................................................................................................53
13.2 About the Cortex-M3 processor and core peripherals .........................................53
13.3 Programmers model ............................................................................................55
13.4 Memory model .....................................................................................................68
13.5 Exception model ..................................................................................................77
13.6 Fault handling ......................................................................................................84
13.7 Power management ............................................................................................87
13.8 Instruction set summary ......................................................................................89
13.9 Intrinsic functions .................................................................................................92
13.10 About the instruction descriptions ......................................................................93
13.11 Memory access instructions ............................................................................101
13.12 General data processing instructions ..............................................................117
13.13 Multiply and divide instructions ........................................................................133
13.14 Saturating instructions .....................................................................................137
13.15 Bitfield instructions ...........................................................................................139
13.16 Branch and control instructions .......................................................................143
13.17 Miscellaneous instructions ...............................................................................151
13.18 About the Cortex-M3 peripherals .....................................................................164
13.19 Nested Vectored Interrupt Controller ...............................................................165
13.20 System control block .......................................................................................178
13.21 System timer, SysTick .....................................................................................205
13.22 Memory protection unit ....................................................................................210
13.23 Glossary ..........................................................................................................224

14 Debug and Test Features .................................................................... 229
14.1 Overview ............................................................................................................229
14.2 Application Examples ........................................................................................230
14.3 Debug and Test Pin Description ........................................................................231
14.4 Functional Description .......................................................................................231

15 Watchdog Timer (WDT) ....................................................................... 237
15.1 Description .........................................................................................................237
15.2 Block Diagram ...................................................................................................237
15.3 Functional Description .......................................................................................238
15.4 Watchdog Timer (WDT) User Interface .............................................................240

16 Reset Controller (RSTC) ...................................................................... 245
16.1 Overview ............................................................................................................245

iii
6430F–ATARM–21-Feb-12

16.2 Block Diagram ...................................................................................................245
16.3 Functional Description .......................................................................................245
16.4 Reset Controller (RSTC) User Interface ............................................................253

17 Real-time Timer (RTT) .......................................................................... 257
17.1 Description .........................................................................................................257
17.2 Embedded Characteristics ................................................................................257
17.3 Block Diagram ...................................................................................................257
17.4 Functional Description .......................................................................................258
17.5 Real-time Timer (RTT) User Interface ...............................................................260

18 Real Time Clock (RTC) ......................................................................... 265
18.1 Description .........................................................................................................265
18.2 Block Diagram ...................................................................................................265
18.3 Product Dependencies ......................................................................................265
18.4 Functional Description .......................................................................................266
18.5 Real Time Clock (RTC) User Interface .............................................................269

19 Supply Controller (SUPC) .................................................................... 283
19.1 Description .........................................................................................................283
19.2 Embedded Characteristics ................................................................................283
19.3 Block Diagram ...................................................................................................284
19.4 Supply Controller Functional Description ..........................................................285
19.5 Supply Controller (SUPC) User Interface ..........................................................294

20 General Purpose Backup Registers (GPBR) ..................................... 305
20.1 Embedded Characteristics ................................................................................305
20.2 Description .........................................................................................................305

21 Enhanced Embedded Flash Controller (EEFC) ................................. 307
21.1 Description .......................................................................................................307
21.2 Embedded Characteristics .............................................................................307
21.3 Product Dependencies ......................................................................................307
21.4 Functional Description .......................................................................................307
21.5 Enhanced Embedded Flash Controller (EEFC) User Interface .........................317

22 Fast Flash Programming Interface (FFPI) .......................................... 323
22.1 Overview ............................................................................................................323
22.2 Parallel Fast Flash Programming ......................................................................323

23 SAM3U4/2/1 Boot Program .................................................................. 333
iv

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
23.1 Description .........................................................................................................333
23.2 Flow Diagram ....................................................................................................333
23.3 Device Initialization ............................................................................................333
23.4 SAM-BA Monitor ................................................................................................334
23.5 Hardware and Software Constraints ..................................................................338

24 Bus Matrix (MATRIX) ............................................................................ 339
24.1 Description .........................................................................................................339
24.2 Memory Mapping ...............................................................................................339
24.3 Special Bus Granting Techniques .....................................................................339
24.4 Arbitration ..........................................................................................................340
24.5 Write Protect Registers ......................................................................................342
24.6 Bus Matrix (MATRIX) User Interface .................................................................343

25 Static Memory Controller (SMC) ......................................................... 353
25.1 Description .........................................................................................................353
25.2 Embedded Characteristics ................................................................................353
25.3 Block Diagram ...................................................................................................354
25.4 I/O Lines Description .........................................................................................355
25.5 Multiplexed Signals ............................................................................................355
25.6 Application Example ..........................................................................................356
25.7 Product Dependencies ......................................................................................356
25.8 External Memory Mapping .................................................................................357
25.9 Connection to External Devices ........................................................................358
25.10 Standard Read and Write Protocols ................................................................360
25.11 Scrambling/Unscrambling Function .................................................................366
25.12 Automatic Wait States .....................................................................................367
25.13 Data Float Wait States .....................................................................................370
25.14 External Wait ...................................................................................................375
25.15 Slow Clock Mode .............................................................................................381
25.16 NAND Flash Controller Operations .................................................................384
25.17 SMC Error Correcting Code Functional Description ........................................396
25.18 Power Management Controller (PMC) User Interface .....................................400

26 Peripheral DMA Controller (PDC) ....................................................... 441
26.1 Description .........................................................................................................441
26.2 Embedded Characteristics ................................................................................441
26.3 Block Diagram ...................................................................................................442

v
6430F–ATARM–21-Feb-12

26.4 Functional Description .......................................................................................442
26.5 Peripheral DMA Controller (PDC) User Interface ..............................................445

27 Clock Generator ................................................................................... 457
27.1 Description .........................................................................................................457
27.2 Block Diagram ...................................................................................................458
27.3 Slow Clock .........................................................................................................459
27.4 Main Clock .........................................................................................................460
27.5 Divider and PLLA Block .....................................................................................463
27.6 UTMI Phase Lock Loop Programming ..............................................................464

28 Power Management Controller (PMC) ................................................ 465
28.1 Description .........................................................................................................465
28.2 Block Diagram ...................................................................................................466
28.3 Master Clock Controller .....................................................................................466
28.4 Processor Clock Controller ................................................................................467
28.5 SysTick Clock ....................................................................................................467
28.6 Peripheral Clock Controller ................................................................................467
28.7 Free Running Processor Clock ..........................................................................467
28.8 Programmable Clock Output Controller .............................................................468
28.9 Fast Startup .......................................................................................................468
28.10 Clock Failure Detector .....................................................................................470
28.11 Programming Sequence ..................................................................................471
28.12 Clock Switching Details ...................................................................................474
28.13 Write Protection Registers ...............................................................................477
28.14 Power Management Controller (PMC) User Interface ....................................478

29 Chip Identifier (CHIPID) ....................................................................... 501
29.1 Description .........................................................................................................501
29.2 Embedded Characteristics ................................................................................501
29.3 Chip Identifier (CHIPID) User Interface ............................................................502

30 Parallel Input/Output Controller (PIO) ................................................ 509
30.1 Description .........................................................................................................509
30.2 Embedded Characteristics ................................................................................509
30.3 Block Diagram ...................................................................................................510
30.4 Product Dependencies ......................................................................................511
30.5 Functional Description .......................................................................................512
30.6 I/O Lines Programming Example .......................................................................520
vi

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
30.7 Parallel Input/Output Controller (PIO) User Interface ........................................522

31 Synchronous Serial Controller (SSC) ................................................ 547
31.1 Description .........................................................................................................547
31.2 Embedded Characteristics .............................................................................547
31.3 Block Diagram ...................................................................................................548
31.4 Application Block Diagram .................................................................................548
31.5 Pin Name List ....................................................................................................549
31.6 Product Dependencies ......................................................................................549
31.7 Functional Description .......................................................................................551
31.8 SSC Application Examples ................................................................................562
31.9 Synchronous Serial Controller (SSC) User Interface ........................................565

32 Serial Peripheral Interface (SPI) Programmer Datasheet ................. 593
32.1 Description .........................................................................................................593
32.2 Embedded Characteristics ................................................................................593
32.3 Block Diagram ...................................................................................................594
32.4 Application Block Diagram .................................................................................595
32.5 Signal Description .............................................................................................595
32.6 Product Dependencies ......................................................................................596
32.7 Functional Description .......................................................................................597
32.8 Serial Peripheral Interface (SPI) User Interface ................................................611

33 Two-wire Interface (TWI) ...................................................................... 627
33.1 Description .........................................................................................................627
33.2 Embedded Characteristics ................................................................................627
33.3 List of Abbreviations ..........................................................................................628
33.4 Block Diagram ...................................................................................................628
33.5 Application Block Diagram .................................................................................629
33.6 Product Dependencies ......................................................................................629
33.7 Functional Description .......................................................................................630
33.8 Master Mode ......................................................................................................631
33.9 Multi-master Mode .............................................................................................643
33.10 Slave Mode ......................................................................................................646
33.11 Two-wire Interface (TWI) User Interface .........................................................654

34 Universal Asynchronous Receiver Transceiver (UART) .................. 669
34.1 Description .........................................................................................................669
34.2 Embedded Characteristics ................................................................................669
vii
6430F–ATARM–21-Feb-12

34.3 Block Diagram ...................................................................................................670
34.4 Product Dependencies ......................................................................................670
34.5 UART Operations ..............................................................................................671
34.6 Universal Asynchronous Receiver Transceiver (UART) User Interface ...........677

35 Universal Synchronous Asynchronous Receiver Transmitter (USART) 689
35.1 Description .........................................................................................................689
35.2 Embedded Characteristics ................................................................................689
35.3 Block Diagram ...................................................................................................690
35.4 Application Block Diagram .................................................................................691
35.5 I/O Lines Description ........................................................................................692
35.6 Product Dependencies ......................................................................................693
35.7 Functional Description .......................................................................................695
35.8 Universal Synchronous Asynchronous Receiver Transmitter (USART) User Interface 733

36 Timer Counter (TC) .............................................................................. 763
36.1 Description .........................................................................................................763
36.2 Embedded Characteristics ................................................................................763
36.3 Block Diagram ...................................................................................................764
36.4 Pin Name List ....................................................................................................765
36.5 Product Dependencies ......................................................................................765
36.6 Functional Description .......................................................................................766
36.7 Timer Counter (TC) User Interface ....................................................................785

37 High Speed Multimedia Card Interface (HSMCI) ............................... 809
37.1 Description .........................................................................................................809
37.2 Embedded Characteristics ................................................................................809
37.3 Block Diagram ...................................................................................................810
37.4 Application Block Diagram .................................................................................811
37.5 Pin Name List ....................................................................................................811
37.6 Product Dependencies ......................................................................................812
37.7 Bus Topology .....................................................................................................812
37.8 High Speed MultiMediaCard Operations ...........................................................815
37.9 SD/SDIO Card Operation ..................................................................................833
37.10 CE-ATA Operation ...........................................................................................834
37.11 HSMCI Boot Operation Mode ..........................................................................835
37.12 HSMCI Transfer Done Timings .......................................................................836
37.13 Write Protection Registers ...............................................................................838

viii

SAM3U Series
6430F–ATARM–21-Feb-12

SAM3U Series
37.14 High Speed Multimedia Card Interface (HSMCI) User Interface .....................839

38 Pulse Width Modulation (PWM) .......................................................... 869
38.1 Description .........................................................................................................869
38.2 Embedded Characteristics ................................................................................869
38.3 Block Diagram ...................................................................................................871
38.4 I/O Lines Description .........................................................................................871
38.5 Product Dependencies ......................................................................................872
38.6 Functional Description .......................................................................................874
38.7 Pulse Width Modulation (PWM) User Interface .................................................902

39 USB High Speed Device Port (UDPHS) .............................................. 951
39.1 Description .........................................................................................................951
39.2 Embedded Characteristics ................................................................................951
39.3 Block Diagram ...................................................................................................953
39.4 Typical Connection ............................................................................................954
39.5 Product Dependencies ......................................................................................954
39.6 Functional Description .......................................................................................954
39.7 USB High Speed Device Port (UDPHS) User Interface ....................................980

40 DMA Controller (DMAC) ..................................................................... 1021
40.1 Description .......................................................................................................1021
40.2 Block Diagram .................................................................................................1022
40.3 Functional Description .....................................................................................1022
40.4 DMAC Software Requirements .......................................................................1038
40.5 DMA Controller (DMAC) User Interface ..........................................................1040

41 12-bit Analog-to-Digital Converter (ADC12B) .................................. 1063
41.1 Description .......................................................................................................1063
41.2 Block Diagram .................................................................................................1063
41.3 Signal Description ............................................................................................1064
41.4 Product Dependencies ....................................................................................1064
41.5 Functional Description .....................................................................................1065
41.6 12-bit Analog-to-Digital Converter (ADC12B) User Interface ..........................1072

42 Analog-to-Digital Converter (ADC) ................................................... 1087
42.1 Description .......................................................................................................1087
42.2 Block Diagram .................................................................................................1087
42.3 Signal Description ............................................................................................1087

ix
6430F–ATARM–21-Feb-12

42.4 Product Dependencies ....................................................................................1088
42.5 Functional Description .....................................................................................1089
42.6 Analog-to-Digital Converter (ADC) User Interface ...........................................1094

43 SAM3U4/2/1 Electrical Characteristics ............................................. 1107
43.1 Absolute Maximum Ratings .............................................................................1107
43.2 DC Characteristics ...........................................................................................1108
43.3 Power Consumption ........................................................................................1113
43.4 Crystal Oscillators Characteristics ...................................................................1121
43.5 UPLL, PLLA Characteristics ............................................................................1127
43.6 USB High Speed Port ......................................................................................1128
43.7 12-Bit ADC Characteristics ..............................................................................1129
43.8 AC Characteristics ...........................................................................................1135

44 SAM3U4/2/1 Mechanical Characteristics ......................................... 1152
44.1 100-lead LQFP Package .................................................................................1152
44.2 100-ball TFBGA Package ................................................................................1153
44.3 144-lead LQFP Package .................................................................................1154
44.4 144-ball LFBGA Package ................................................................................1155
44.5 Soldering Profile ..............................................................................................1156
44.6 Packaging Resources ......................................................................................1156

45 Ordering Information ......................................................................... 1157
46 SAM3U Series Errata ......................................................................... 1159
46.1 Marking ............................................................................................................1159
46.2 SAM3U Errata - Rev. A Parts ..........................................................................1160
46.3 SAM3U Errata - Rev. B Parts ..........................................................................1163

47 Revision History ................................................................................. 1167

SAM3U Series
6430F–ATARM–21-Feb-12

Headquarters

International

Atmel Corporation
2325 Orchard Parkway
San Jose, CA 95131
USA
Tel: (+1) (408) 441-0311
Fax: (+1) (408) 487-2600

Atmel Asia Limited
Unit 01-5 & 16, 19F
BEA Tower, Millennium City 5
418 Kwun Tong Road
Kwun Tong, Kowloon
HONG KONG
Tel: (+852) 2245-6100
Fax: (+852) 2722-1369

Atmel Munich GmbH
Business Campus
Parkring 4
D-85748 Garching b. Munich
GERMANY
Tel: (+49) 89-31970-0
Fax: (+49) 89-3194621

Atmel Japan
9F, Tonetsu Shinkawa Bldg.
1-24-8 Shinkawa
Chuo-ku, Tokyo 104-0033
JAPAN
Tel: (81) 3-3523-3551
Fax: (81) 3-3523-7581

Technical Support
AT91SAM Support
Atmel technical support

Sales Contacts
www.atmel.com/contacts/

Product Contact
Web Site
www.atmel.com
www.atmel.com/AT91SAM
Literature Requests
www.atmel.com/literature

Disclaimer: The information in this document is provided in connection with Atmel products. No license, express or implied, by estoppel or otherwise, to any
intellectual property right is granted by this document or in connection with the sale of Atmel products. EXCEPT AS SET FORTH IN ATMEL’S TERMS AND CONDITIONS OF SALE LOCATED ON ATMEL’S WEB SITE, ATMEL ASSUMES NO LIABILITY WHATSOEVER AND DISCLAIMS ANY EXPRESS, IMPLIED OR STATUTORY
WARRANTY RELATING TO ITS PRODUCTS INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR
PURPOSE, OR NON-INFRINGEMENT. IN NO EVENT SHALL ATMEL BE LIABLE FOR ANY DIRECT, INDIRECT, CONSEQUENTIAL, PUNITIVE, SPECIAL OR INCIDENTAL DAMAGES (INCLUDING, WITHOUT LIMITATION, DAMAGES FOR LOSS OF PROFITS, BUSINESS INTERRUPTION, OR LOSS OF INFORMATION) ARISING OUT
OF THE USE OR INABILITY TO USE THIS DOCUMENT, EVEN IF ATMEL HAS BEEN ADVISED OF THE POSSIBILITY OF SUCH DAMAGES. Atmel makes no
representations or warranties with respect to the accuracy or completeness of the contents of this document and reserves the right to make changes to specifications and product descriptions at any time without notice. Atmel does not make any commitment to update the information contained herein. Unless specifically provided otherwise, Atmel products are not suitable for, and shall not be used in, automotive applications. Atmel’s products are not intended, authorized, or warranted
for use as components in applications intended to support or sustain life.

© 2011 Atmel Corporation. All rights reserved. Atmel ®, logo and combinations thereof DataFlash®, SAM-BA® and others are registered trademarks or trademarks of Atmel Corporation or its subsidiaries. ARM ®, Thumb ® and the ARMPowered logo ® and others are registered trademarks
or trademarks ARM Ltd. Windows® and others are registered trademarks or trademarks of Microsoft Corporation in the US and/or other countries. Other terms and product names may be trademarks of others.

6430F–ATARM–21-Feb-12

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.6
Linearized                      : Yes
Encryption                      : Standard V2.3 (128-bit)
User Access                     : Print, Copy, Annotate, Fill forms, Extract, Print high-res
Page Mode                       : UseOutlines
XMP Toolkit                     : Adobe XMP Core 4.0-c316 44.253921, Sun Oct 01 2006 17:14:39
Format                          : application/pdf
Creator                         : anders.rose
Title                           : 6430F.book
Create Date                     : 2012:02:21 15:47:29Z
Creator Tool                    : FrameMaker 8.0
Modify Date                     : 2012:02:21 16:30:04+01:00
Metadata Date                   : 2012:02:21 16:30:04+01:00
Producer                        : Acrobat Distiller 8.1.0 (Windows)
Document ID                     : uuid:3e9ef178-1798-4adc-ac73-dfceb7fa853b
Instance ID                     : uuid:b6b3a8c7-1450-4c67-9fa6-da4deeebcfd1
Page Count                      : 1187
Author                          : anders.rose

EXIF Metadata provided by EXIF.tools

6430F SAM3U User Guide

Navigation menu

Versions of this User Manual:

Views

Navigation