8254x Family Of Gigabit Ethernet Controllers Software Developer’s Manual Pci X Gbe Dev

User Manual: Pdf

Open the PDF directly: View PDF .
Page Count: 410

Download
Open PDF In Browser	View PDF

PCI/PCI-X Family of Gigabit Ethernet
Controllers Software Developer’s
Manual
82540EP/EM, 82541xx, 82544GC/EI, 82545GM/EM, 82546GB/EB, and
82547xx

317453006EN.PDF
Revision 4.0

Legal Notice
INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL® PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY
ESTOPPEL OR OTHERWISE, TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN
INTEL'S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER, AND INTEL DISCLAIMS
ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES
RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER
INTELLECTUAL PROPERTY RIGHT.
Intel products are not intended for use in medical, life saving, life sustaining, critical control or safety systems, or in nuclear facility applications.
Intel may make changes to specifications and product descriptions at any time, without notice.
This document contains information on products in the design phase of development. The information here is subject to change without notice. Do not
finalize a design with this information.
Designers must not rely on the absence or characteristics of any features or instructions marked "reserved" or "undefined." Intel reserves these for
future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them.
This product has not been tested with every possible configuration/setting. Intel is not responsible for the product’s failure in any configuration/setting,
whether tested or untested.
The Intel product(s) discussed in this document may contain design defects or errors known as errata which may cause the product to deviate from
published specifications. Current characterized errata are available on request.
Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order.
Copies of documents which have an ordering number and are referenced in this document, or other Intel literature, may be obtained from:
Intel Corporation
P.O. Box 5937
Denver, CO 80217-9808
or call in North America 1-800-548-4725, Europe 44-0-1793-431-155, France 44-0-1793-421-777, Germany 44-0-1793-421-333, other Countries 708296-9333.
Intel® is a trademark or registered trademark of Intel Corporation or its subsidiaries in the United States and other countries.
*Other names and brands may be claimed as the property of others.
Copyright © Intel Corporation, 2009

Software Developer’s Manual

Revision History
Date

Version

Comments

Mar 2009

4.0

Updated Section 5 (added Table 5-15 “LED Configuration Defaults).

Oct 2008

3.9

Updated section 8.7.4 “Discard PAUSE Frames and Pass MAC Control
Frames”.

June 2008

3.8

Updated EEPROM Word 21h bit descriptions (section 5.6.18).

June 2008

3.7

Updated Sections 13.4.30 and 13.4.31 (added text stating to use the
Interrupt Throttling Register (ITR) instead of registers RDTR and RADV for
applications requiring an interrupt moderation mechanism).

Jan 2007

3.6

Added a note to sections 13.4.20 and 13.4.21 for the 82547Gi/EI.

Sept 2007

3.5

Updated section 13.4.16.

May 2007

3.4

Updated section 6.4.1. Changed acronym “WCR” to “WUC”.

Dec 2006

3.3

Updated Table 13-87. Changed bit 24 settings to:
0b = Cache line granularity.
1b = Descriptor granularity.

June 2006

3.2

Updated Table 13.47. Changed the default setting of reserved bit 3 from 0b
to 1b.

April 2006

3.1

Added bit definitions (bits 9:8) to PHY register PSCON (16d).
Updated Figure 3.2 (added Receive Queue artwork).
Changed 81541ER-C0 to 82541ER-CO in Table 5-1.

Nov 2005

3.0

Updated Device Control/Status, EEPROM Flash Control & Data, Extended
Device Control, and TCTL register bit assignments.
Updated PHY register 00d - 03d, 07d, 09d, 17d - 21d, and 23d bit assignments.

July 2005

2.5

Initial Public Release.

Software Developer’s Manual

iii

Note:

This page is intentionally left blank.

Software Developer’s Manual

Contents

Contents
1

Introduction .................................................................................................................. 1
1.1
1.2
1.3

1.4

1.5
1.6

Architectural Overview ............................................................................................ 7
2.1
2.2
2.3

2.4
2.5
2.6
2.7

2.8

Scope .................................................................................................................... 1
Overview ............................................................................................................... 1
Ethernet Controller Features ................................................................................. 2
1.3.1
PCI Features ........................................................................................ 2
1.3.2
CSA Features (82547GI/EI Only) ......................................................... 2
1.3.3
Network Side Features ......................................................................... 2
1.3.4
Host Offloading Features ..................................................................... 3
1.3.5
Additional Performance Features ......................................................... 4
1.3.6
Manageability Features (Not Applicable to the 82544GC/EI or
82541ER)5
1.3.7
Additional Ethernet Controller Features ............................................... 5
1.3.8
Technology Features............................................................................ 5
Conventions .......................................................................................................... 6
1.4.1
Register and Bit References ................................................................ 6
1.4.2
Byte and Bit Designations .................................................................... 6
Related Documents............................................................................................... 6
Memory Alignment Terminology............................................................................ 6
Introduction............................................................................................................ 7
External Architecture ............................................................................................. 8
Microarchitecture................................................................................................. 10
2.3.1
PCI/PCI-X Core Interface ................................................................... 10
2.3.2
82547GI/EI CSA Interface .................................................................. 11
2.3.3
DMA Engine and Data FIFO .............................................................. 11
2.3.4
10/100/1000 Mb/s Receive and Transmit MAC Blocks ...................... 12
2.3.5
MII/GMII/TBI/Internal SerDes Interface Block .................................... 12
2.3.6
10/100/1000 Ethernet Transceiver (PHY) .......................................... 13
2.3.7
EEPROM Interface ............................................................................. 13
2.3.8
FLASH Memory Interface ................................................................... 14
DMA Addressing ................................................................................................. 14
Ethernet Addressing............................................................................................ 15
Interrupts ............................................................................................................. 16
Hardware Acceleration Capability ....................................................................... 17
2.7.1
Checksum Offloading ......................................................................... 17
2.7.2
TCP Segmentation ............................................................................. 17
Buffer and Descriptor Structure........................................................................... 17

Receive and Transmit Description .................................................................... 19
3.1
3.2

Introduction.......................................................................................................... 19
Packet Reception ................................................................................................ 19
3.2.1
Packet Address Filtering .................................................................... 19
3.2.2
Receive Data Storage ........................................................................ 20
3.2.3
Receive Descriptor Format.................................................................20
3.2.4
Receive Descriptor Fetching .............................................................. 25

Software Developer’s Manual

Contents

3.3

3.4

3.5

3.6

PCI Local Bus Interface ......................................................................................... 71
4.1

4.2
4.3

4.4
4.5
4.6
4.7

PCI Configuration ................................................................................................ 71
4.1.1
PCI-X Configuration Registers ........................................................... 79
4.1.2
Reserved and Undefined Addresses.................................................. 82
4.1.3
Message Signaled Interrupts.............................................................. 83
Commands .......................................................................................................... 85
PCI/PCI-X Command Usage............................................................................... 87
4.3.1
Memory Write Operations .................................................................. 87
4.3.2
Memory Read Operations .................................................................. 89
Cache Line Information ....................................................................................... 90
4.4.1
Target Transaction Termination ......................................................... 91
Interrupt Assignment (82547GI/EI Only) ............................................................. 91
LAN Disable ........................................................................................................ 91
CardBus Application (82541PI/GI/EI Only) ......................................................... 92

EEPROM Interface ................................................................................................... 93
5.1
5.2
5.3
5.4

3.2.5
Receive Descriptor Write-Back .......................................................... 26
3.2.6
Receive Descriptor Queue Structure.................................................. 26
3.2.7
Receive Interrupts .............................................................................. 28
3.2.8
82544GC/EI Receive Interrupts ......................................................... 31
3.2.9
Receive Packet Checksum Offloading ............................................... 31
Packet Transmission ........................................................................................... 34
3.3.1
Transmit Data Storage ....................................................................... 35
3.3.2
Transmit Descriptors .......................................................................... 35
3.3.3
Legacy Transmit Descriptor Format ................................................... 36
3.3.4
Transmit Descriptor Special Field Format .......................................... 40
3.3.5
TCP/IP Context Transmit Descriptor Format...................................... 41
3.3.6
TCP/IP Context Descriptor Layout ..................................................... 42
3.3.7
TCP/IP Data Descriptor Format ......................................................... 46
Transmit Descriptor Ring Structure ..................................................................... 51
3.4.1
Transmit Descriptor Fetching ............................................................. 53
3.4.2
Transmit Descriptor Write-back.......................................................... 53
3.4.3
Transmit Interrupts ............................................................................. 54
TCP Segmentation .............................................................................................. 55
3.5.1
Assumptions....................................................................................... 56
3.5.2
Transmission Process ........................................................................ 56
3.5.3
TCP Segmentation Performance ....................................................... 57
3.5.4
Packet Format .................................................................................... 57
3.5.5
TCP Segmentation Indication............................................................. 58
3.5.6
TCP Segmentation Use of Multiple Data Descriptors ........................ 59
3.5.7
IP and TCP/UDP Headers.................................................................. 60
3.5.8
Transmit Checksum Offloading with TCP Segmentation ................... 64
3.5.9
IP/TCP/UDP Header Updating ........................................................... 65
IP/TCP/UDP Transmit Checksum Offloading...................................................... 68

General Overview ............................................................................................... 93
Component Identification Via Programming Interface......................................... 94
EEPROM Device and Interface........................................................................... 95
5.3.1
Software Access................................................................................. 96
Signature and CRC Fields .................................................................................. 96

Software Developer’s Manual

Contents

5.5
5.6

5.7

EEUPDATE Utility ............................................................................................... 97
5.5.1
Command Line Parameters ............................................................... 97
EEPROM Address Map....................................................................................... 98
5.6.1
Ethernet Address (Words 00h-02h)..................................................103
5.6.2
Software Compatibility Word (Word 03h) .........................................103
5.6.3
SerDes Configuration (Word 04h) ....................................................104
5.6.4
EEPROM Image Version (Word 05h)...............................................104
5.6.5
Compatibility Fields (Word 05h - 07h) ..............................................104
5.6.6
PBA Number (Word 08h, 09h) .........................................................104
5.6.7
Initialization Control Word 1 (Word 0Ah) ..........................................105
5.6.8
Subsystem ID (Word 0Bh)................................................................106
5.6.9
Subsystem Vendor ID (Word 0Ch) ...................................................106
5.6.10
Device ID (Word 0Dh, 11h) ..............................................................107
5.6.11
Vendor ID (Word 0Eh) ......................................................................107
5.6.12
Initialization Control Word 2 (Word 0Fh) ..........................................107
5.6.13
PHY Register Address Data (Words 10h, 11h, and 13h - 1Eh) .......109
5.6.14
OEM Reserved Words (Words 10h, 11h, 13h - 1Fh) .......................109
5.6.15
EEPROM Size (Word 12h) ...............................................................109
5.6.16
Common Power (Word 12h).............................................................109
5.6.17
Software Defined Pins Control (Word 10h, 20h) ..............................109
5.6.18
CSA Port Configuration 2 (Word 21h) ..............................................111
5.6.19
Circuit Control (Word 21h)................................................................112
5.6.20
D0 Power (Word 22h high byte) .......................................................112
5.6.21
D3 Power (Word 22h low byte) ........................................................112
5.6.22
Reserved Words (23h - 2Eh)............................................................112
5.6.23
Reserved Words (23h - 2Fh) ............................................................112
5.6.24
Management Control (Word 13h, 23h) .............................................113
5.6.25
SMBus Slave Address (Word 14h low byte, 24h low byte) ..............114
5.6.26
Initialization Control 3 (Word 14h high byte, 24h high byte) .............115
5.6.27
IPv4 Address (Words 15h - 16h and 25h - 26h) ...............................116
5.6.28
IPv6 Address (words 17h - 1Eh1 and 27h - 2Eh) .............................116
5.6.29
LED Configuration Defaults (Word 2Fh)...........................................117
5.6.30
Boot Agent Main Setup Options (Word 30h) ....................................117
5.6.31
Boot Agent Configuration Customization Options (Word 31h) .........119
5.6.32
Boot Agent Configuration Customization Options (Word 32h) .........121
5.6.33
IBA Capabilities (Word 33h) .............................................................122
5.6.34
IBA Secondary Port Configuration (Words 34h-35h) .......................122
5.6.35
Checksum Word Calculation (Word 3Fh) .........................................123
5.6.36
82546GB/EB Dual-Channel Fiber Wake on LAN (WOL) Mode and
Functionality (Word 0Ah, 20h)123
5.6.37
EEPROM Images .............................................................................123
Parallel FLASH Memory....................................................................................124

FLASH Memory Interface ....................................................................................127
7.1
7.2

FLASH Interface Operation ...............................................................................127
FLASH Control and Accesses...........................................................................127
7.2.1
Read Accesses ................................................................................128
7.2.2
Write Accesses.................................................................................128

Software Developer’s Manual

vii

Contents

Power Management ............................................................................................... 131
6.1
6.2
6.3

6.4

Ethernet Interface .................................................................................................. 155
8.1
8.2

8.3
8.4

8.5
8.6

8.7

9.2

9.3

802.1q VLAN Packet Format ............................................................................ 179
9.1.1
802.1q Tagged Frames .................................................................... 179
Transmitting and Receiving 802.1q Packets ..................................................... 180
9.2.1
Adding 802.1q Tags on Transmits ................................................... 180
9.2.2
Stripping 802.1q Tags on Receives ................................................. 180
802.1q VLAN Packet Filtering ........................................................................... 180

Configurable LED Outputs ................................................................................. 183
10.1

viii

Introduction ....................................................................................................... 155
Link Interfaces Overview................................................................................... 155
8.2.1
Internal SerDes Interface/TBI Mode– 1Gb/s .................................... 156
8.2.2
GMII – 1 Gb/s ................................................................................... 157
8.2.3
MII – 10/100 Mb/s............................................................................. 158
Internal Interface ............................................................................................... 158
Duplex Operation .............................................................................................. 158
8.4.1
Full Duplex ....................................................................................... 159
8.4.2
Half Duplex....................................................................................... 159
Auto-Negotiation and Link Setup ...................................................................... 161
Auto-Negotiation and Link Setup ...................................................................... 161
8.6.1
Link Configuration in Internal Serdes/TBI Mode............................... 162
8.6.2
Internal GMII/MII Mode..................................................................... 165
8.6.3
Internal SerDes Mode Control Bit Resolution................................... 168
8.6.4
Internal PHY Mode Control Bit Resolution ....................................... 169
8.6.5
Loss of Signal/Link Status Indication................................................ 171
10/100 Mb/s Specific Performance Enhancements .......................................... 172
8.7.1
Adaptive IFS..................................................................................... 172
8.7.2
Flow Control ..................................................................................... 173
8.7.3
MAC Control Frames & Reception of Flow Control Packets ............ 173
8.7.4
Discard PAUSE Frames and Pass MAC Control Frames ................ 175
8.7.5
Transmission of PAUSE Frames...................................................... 176
8.7.6
Software Initiated PAUSE Frame Transmission............................... 176
8.7.7
External Control of Flow Control Operation...................................... 176

802.1q VLAN Support ........................................................................................... 179
9.1

Introduction to Power Management .................................................................. 131
Assumptions...................................................................................................... 131
D3cold support .................................................................................................. 132
6.3.1
Power States .................................................................................... 132
6.3.2
Timing............................................................................................... 134
6.3.3
PCI Power Management Registers .................................................. 139
Wakeup ............................................................................................................. 143
6.4.1
Advanced Power Management Wakeup .......................................... 143
6.4.2
ACPI Power Management Wakeup.................................................. 144
6.4.3
Wakeup Packets .............................................................................. 145

Configurable LED Outputs ................................................................................ 183
10.1.1
Selecting an LED Output Source ..................................................... 183
10.1.2
Polarity Inversion.............................................................................. 184

Software Developer’s Manual

Contents

10.1.3

Blink Control .....................................................................................184

PHY Functionality and Features ......................................................................187
11.1

11.2

11.3
11.4

11.5
11.6

11.7

11.8

11.9
11.10

11.11

11.12
11.13

11.14

Auto-Negotiation................................................................................................187
11.1.1
Overview ..........................................................................................187
11.1.2
Next Page Exchanges......................................................................188
11.1.3
Register Update ...............................................................................188
11.1.4
Status ...............................................................................................189
MDI/MDI-X Crossover (copper only) .................................................................189
11.2.1
Polarity Correction (copper only) ......................................................190
11.2.2
10/100 Downshift (82540EP/EM Only).............................................190
Cable Length Detection (copper only)...............................................................191
PHY Power Management (copper only)............................................................191
11.4.1
Link Down – Energy Detect (copper only) ........................................191
11.4.2
D3 State, No Link Required (copper only)........................................192
11.4.3
D3 Link-Up, Speed-Management Enabled (copper only).................192
11.4.4
D3 Link-Up, Speed-Management Disabled (copper only) ................192
Initialization........................................................................................................193
11.5.1
MDIO Control Mode .........................................................................193
Determining Link State ......................................................................................194
11.6.1
False Link .........................................................................................195
11.6.2
Forced Operation .............................................................................195
11.6.3
Auto Negotiation ...............................................................................196
11.6.4
Parallel Detection .............................................................................196
Link Criteria .......................................................................................................196
11.7.1
1000BASE-T ....................................................................................196
11.7.2
100BASE-TX ....................................................................................196
11.7.3
10BASE-T ........................................................................................197
Link Enhancements...........................................................................................197
11.8.1
SmartSpeed .....................................................................................197
11.8.2
Flow Control .....................................................................................197
Management Data Interface..............................................................................198
Low Power Operation........................................................................................198
11.10.1 Powerdown via the PHY Register ....................................................199
11.10.2 Smart Power-Down ..........................................................................199
1000 Mbps Operation........................................................................................199
11.11.1 Introduction.......................................................................................199
11.11.2 Transmit Functions ...........................................................................201
11.11.3 Transmit FIFO ..................................................................................201
11.11.4 Receive Functions............................................................................203
100 Mbps Operation..........................................................................................204
10 Mbps Operation............................................................................................204
11.13.1 Link Test ...........................................................................................205
11.13.2 10Base-T Link Failure Criteria and Override ....................................205
11.13.3 Jabber ..............................................................................................205
11.13.4 Polarity Correction ............................................................................205
11.13.5 Dribble Bits .......................................................................................205
PHY Line Length Indication...............................................................................205

Software Developer’s Manual

Contents

Dual Port Characteristics .................................................................................... 207
12.1
12.2

12.3

12.4
12.5

Register Descriptions........................................................................................... 215
13.1
13.2

13.3
13.4

Introduction ....................................................................................................... 207
Features of Each MAC...................................................................................... 207
12.2.1
PCI/PCI-X interface .......................................................................... 207
12.2.2
MAC Configuration Register Space ................................................. 209
12.2.3
SDP, LED, INT# output .................................................................... 209
Shared EEPROM .............................................................................................. 210
12.3.1
EEPROM Map.................................................................................. 210
12.3.2
EEPROM Arbitration ........................................................................ 210
Shared FLASH .................................................................................................. 211
12.4.1
FLASH Access Contention............................................................... 211
LAN Disable ...................................................................................................... 212
12.5.1
Overview .......................................................................................... 212
12.5.2
Values Sampled on Reset................................................................ 212
12.5.3
Multi-Function Advertisement........................................................... 213
12.5.4
Interrupt Use..................................................................................... 213
12.5.5
Power Reporting............................................................................... 213
12.5.6
Summary .......................................................................................... 214
Introduction ....................................................................................................... 215
Register Conventions ........................................................................................ 215
13.2.1
Memory and I/O Address Decoding ................................................. 216
13.2.2
I/O-Mapped Internal Register, Internal Memory, and Flash ............. 217
PCI-X Register Access Split.............................................................................. 223
Main Register Descriptions ............................................................................... 224
13.4.1
Device Control Register ................................................................... 224
13.4.2
Device Status Register..................................................................... 229
13.4.3
EEPROM/Flash Control & Data Register ......................................... 232
13.4.4
EEPROM Read Register.................................................................. 234
13.4.5
Flash Access .................................................................................... 236
13.4.6
Extended Device Control Register ................................................... 237
13.4.7
MDI Control Register........................................................................ 242
13.4.8
Flow Control Address Low ............................................................... 283
13.4.9
Flow Control Address High............................................................... 283
13.4.10 Flow Control Type ............................................................................ 284
13.4.11 VLAN Ether Type ............................................................................. 284
13.4.12 Flow Control Transmit Timer Value.................................................. 285
13.4.13 Transmit Configuration Word Register ............................................. 286
13.4.14 Receive Configuration Word Register .............................................. 287
13.4.15 LED Control...................................................................................... 289
13.4.16 Packet Buffer Allocation ................................................................... 292
13.4.17 Interrupt Cause Read Register......................................................... 293
13.4.18 Interrupt Throttling Register.............................................................. 295
13.4.19 Interrupt Cause Set Register............................................................ 296
13.4.20 Interrupt Mask Set/Read Register .................................................... 297
13.4.21 Interrupt Mask Clear Register .......................................................... 298
13.4.22 Receive Control Register ................................................................. 300
13.4.23 Flow Control Receive Threshold Low............................................... 304
13.4.24 Flow Control Receive Threshold High.............................................. 305

Software Developer’s Manual

Contents

13.5

13.6

13.7

13.4.25 Receive Descriptor Base Address Low ............................................306
13.4.26 Receive Descriptor Base Address High ...........................................306
13.4.27 Receive Descriptor Length ...............................................................307
13.4.28 Receive Descriptor Head .................................................................307
13.4.29 Receive Descriptor Tail ....................................................................308
13.4.30 Receive Delay Timer Register..........................................................308
13.4.31 Receive Interrupt Absolute Delay Timer...........................................309
13.4.32 Receive Small Packet Detect Interrupt.............................................310
13.4.33 Transmit Control Register ................................................................310
13.4.34 Transmit IPG Register ......................................................................312
13.4.35 Adaptive IFS Throttle - AIT ...............................................................314
13.4.36 Transmit Descriptor Base Address Low ...........................................315
13.4.37 Transmit Descriptor Base Address High ..........................................316
13.4.38 Transmit Descriptor Length ..............................................................316
13.4.39 Transmit Descriptor Head ................................................................317
13.4.40 Transmit Descriptor Tail ...................................................................318
13.4.41 Transmit Interrupt Delay Value.........................................................318
13.4.42 TX DMA Control (82544GC/EI only) ................................................319
13.4.43 Transmit Descriptor Control .............................................................319
13.4.44 Transmit Absolute Interrupt Delay Value..........................................321
13.4.45 TCP Segmentation Pad And Minimum Threshold............................322
13.4.46 Receive Descriptor Control ..............................................................324
13.4.47 Receive Checksum Control ..............................................................325
Filter Registers ..................................................................................................327
13.5.1
Multicast Table Array........................................................................327
13.5.2
Receive Address Low.......................................................................329
13.5.3
Receive Address High ......................................................................329
13.5.4
VLAN Filter Table Array ...................................................................330
Wakeup Registers .............................................................................................331
13.6.1
Wakeup Control Register .................................................................331
13.6.2
Wakeup Filter Control Register ........................................................332
13.6.3
Wakeup Status Register...................................................................333
13.6.4
IP Address Valid ...............................................................................335
13.6.5
IPv4 Address Table ..........................................................................336
13.6.6
IPv6 Address Table ..........................................................................337
13.6.7
Wakeup Packet Length ....................................................................338
13.6.8
Wakeup Packet Memory (128 Bytes) ...............................................338
13.6.9
Flexible Filter Length Table ..............................................................338
13.6.10 Flexible Filter Mask Table ................................................................339
13.6.11 Flexible Filter Value Table ................................................................340
Statistics Registers............................................................................................340
13.7.1
CRC Error Count ..............................................................................341
13.7.2
Alignment Error Count ......................................................................341
13.7.3
Symbol Error Count ..........................................................................342
13.7.4
RX Error Count .................................................................................342
13.7.5
Missed Packets Count......................................................................343
13.7.6
Single Collision Count ......................................................................343
13.7.7
Excessive Collisions Count ..............................................................344
13.7.8
Multiple Collision Count ....................................................................344
13.7.9
Late Collisions Count .......................................................................345

Software Developer’s Manual

Contents

13.8

xii

13.7.10 Collision Count ................................................................................. 345
13.7.11 Defer Count ...................................................................................... 346
13.7.12 Transmit with No CRS...................................................................... 346
13.7.13 Sequence Error Count...................................................................... 347
13.7.14 Carrier Extension Error Count .......................................................... 347
13.7.15 Receive Length Error Count............................................................. 348
13.7.16 XON Received Count ....................................................................... 348
13.7.17 XON Transmitted Count ................................................................... 349
13.7.18 XOFF Received Count ..................................................................... 349
13.7.19 XOFF Transmitted Count ................................................................. 349
13.7.20 FC Received Unsupported Count .................................................... 350
13.7.21 Packets Received (64 Bytes) Count................................................. 350
13.7.22 Packets Received (65-127 Bytes) Count ......................................... 351
13.7.23 Packets Received (128-255 Bytes) Count ....................................... 351
13.7.24 Packets Received (256-511 Bytes) Count ....................................... 352
13.7.25 Packets Received (512-1023 Bytes) Count ..................................... 352
13.7.26 Packets Received (1024 to Max Bytes) Count................................. 353
13.7.27 Good Packets Received Count ........................................................ 353
13.7.28 Broadcast Packets Received Count................................................. 354
13.7.29 Multicast Packets Received Count................................................... 354
13.7.30 Good Packets Transmitted Count .................................................... 355
13.7.31 Good Octets Received Count........................................................... 355
13.7.32 Good Octets Transmitted Count....................................................... 356
13.7.33 Receive No Buffers Count................................................................ 356
13.7.34 Receive Undersize Count................................................................. 357
13.7.35 Receive Fragment Count ................................................................. 357
13.7.36 Receive Oversize Count................................................................... 358
13.7.37 Receive Jabber Count...................................................................... 358
13.7.38 Management Packets Received Count ............................................ 359
13.7.39 Management Packets Dropped Count1 ........................................................360
13.7.40 Management Pkts Transmitted Count.............................................. 360
13.7.41 Total Octets Received ...................................................................... 360
13.7.42 Total Octets Transmitted .................................................................. 361
13.7.43 Total Packets Received.................................................................... 362
13.7.44 Total Packets Transmitted................................................................ 362
13.7.45 Packets Transmitted (64 Bytes) Count............................................. 363
13.7.46 Packets Transmitted (65-127 Bytes) Count ..................................... 363
13.7.47 Packets Transmitted (128-255 Bytes) Count ................................... 364
13.7.48 Packets Transmitted (256-511 Bytes) Count ................................... 364
13.7.49 Packets Transmitted (512-1023 Bytes) Count ................................. 365
13.7.50 Packets Transmitted (1024 Bytes or Greater) Count ....................... 365
13.7.51 Multicast Packets Transmitted Count............................................... 366
13.7.52 Broadcast Packets Transmitted Count............................................. 366
13.7.53 TCP Segmentation Context Transmitted Count ............................... 367
13.7.54 TCP Segmentation Context Transmit Fail Count ............................. 367
Diagnostics Registers ....................................................................................... 368
13.8.1
Receive Data FIFO Head Register................................................... 368
13.8.2
Receive Data FIFO Tail Register ..................................................... 368
13.8.3
Receive Data FIFO Head Saved Register ....................................... 369
13.8.4
Receive Data FIFO Tail Saved Register .......................................... 369

Software Developer’s Manual

Contents

13.8.5
13.8.6
13.8.7
13.8.8
13.8.9
13.8.10
13.8.11

General Initialization and Reset Operation ..................................................375
14.1
14.2
14.3
14.4
14.5

14.6
14.7
14.8

Receive Data FIFO Packet Count ....................................................370
Transmit Data FIFO Head Register..................................................370
Transmit Data FIFO Tail Register ....................................................371
Transmit Data FIFO Head Saved Register ......................................371
Transmit Data FIFO Tail Saved Register .........................................372
Transmit Data FIFO Packet Count ...................................................372
Packet Buffer Memory ......................................................................373

Introduction........................................................................................................375
Power Up State .................................................................................................375
General Configuration .......................................................................................375
Receive Initialization..........................................................................................376
Transmit Initialization.........................................................................................377
14.5.1
Signal Interface ................................................................................380
14.5.2
GMII/MII Features not Supported .....................................................381
14.5.3
Avoiding GMII Test Mode(s).............................................................382
14.5.4
MAC Configuration ...........................................................................382
14.5.5
Link Setup ........................................................................................383
PHY Initialization (10/100/1000 Mb/s Copper Media) .......................................384
Reset Operation ................................................................................................385
Initialization of Statistics ....................................................................................388

Diagnostics and Testability ...............................................................................389
15.1

15.2

Diagnostics........................................................................................................389
15.1.1
FIFO State ........................................................................................389
15.1.2
FIFO Data.........................................................................................389
15.1.3
Loopback..........................................................................................389
Testability ..........................................................................................................390
15.2.1
EXTEST Instruction ..........................................................................391
15.2.2
SAMPLE/PRELOAD Instruction .......................................................391
15.2.3
IDCODE Instruction ..........................................................................391
15.2.4
BYPASS Instruction .........................................................................391

Appendix (Changes From 82544EI/82544GC) ............................................393

Appendix (82540EP/EM and 82545GM/EM Differences) .........................395

Software Developer’s Manual

xiii

Contents

Note:

xiv

This page intentionally left blank.

Software Developer’s Manual

Introduction

Introduction
1.1

Scope
This document serves as a software developer’s manual for 82546GB/EB, 82545GM/EM,
82544GC/EI, 82541(PI/GI/EI), 82541ER, 82547GI/EI, and 82540EP/EM Gigabit Ethernet
Controllers. Throughout this manual references are made to the PCI/PCI-X Family of Gigabit
Ethernet Controllers or Ethernet controllers. Unless specifically noted, these references apply to all
the Ethernet controllers listed above.

1.2

Overview
The PCI/PCI-X Family of Gigabit Ethernet Controllers are highly integrated, high-performance
Ethernet LAN devices for 1000 Mb/s, 100 Mb/s and 10 Mb/s data rates. They are optimized for
LAN on Motherboard (LOM) designs, enterprise networking, and Internet appliances that use the
Peripheral Component Interconnect (PCI) and PCI-X bus.
Note:

The 82541xx and 82540EP/EM do not support the PCI-X bus.
The 82547GI(EI) connects to the motherboard chipset through a Communications Streaming
Architecture (CSA) port. CSA is designed for low memory latency and higher performance than a
comparable PCI interface.
The remaining Ethernet controllers provide a 32-/64-bit, 33/66 MHz direct interface to the PCI
Local Bus Specification (revision 2.2 or 2.3), as well as the emerging PCI-X extension to the PCI
Local Bus (revision 1.0a).
The Ethernet controllers provide an interface to the host processor by using on-chip command and
status registers and a shared host memory area, set up mainly during initialization. The controllers
provide a highly optimized architecture to deliver high performance and PCI/CSA/PCI-X bus
efficiency. By implementing hardware acceleration capabilities, the controllers enable offloading
various tasks such as TCP/UDP/IP checksum calculations from the host processor. They also
minimize I/O accesses and interrupts required to manage the Ethernet controllers and provide a
highly configurable design that can be used effectively in various environments.
The PCI/PCI-X Family of Gigabit Ethernet Controllers handle all IEEE 802.3 receive and transmit
MAC functions. They contain fully integrated physical-layer circuitry for 1000 Base-T, 100 BaseTX, and 10 Base-T applications (IEEE 802.3, 802.3u, and 802.3ab) as well as on-chip Serializer/
Deserializer (SerDes)1 functionality that fully complies with IEEE 802.3z PCS.

1. The 82541xx, 82547GI/EI, and 82540EP/EM do not support any SerDes functionality.

Software Developer’s Manual

Introduction

For the 82544GC/EI, when connected to an appropriate SerDes, it can alternatively provide an
Ethernet interface for 1000 Base-SX or LX applications (IEEE 802.3z).
Note:

The 82546EB/82545EM is SerDes PICMG 2.16 compliant. The 82546GB/82545GM is SerDes
PICMG 3.1 compliant.
82546GB/EB Ethernet controllers also provide features in an integrated dual-port solution
comprised of two distinct MAC/PHY instances. As a result, they appear as multi-function PCI
devices containing two identically-functioning Ethernet controllers. See Section 12 for details.

1.3

Ethernet Controller Features
This section describes the features of the PCI/PCI-X Family of Gigabit Ethernet Controllers.

1.3.1

PCI Features
• 32/64-bit 33/66 MHz, PCI Rev 2.3 and PCI-X 1.0a compliant Host interface (82546GB/
82545GM)

• 32/64-bit 33/66 MHz, PCI Rev 2.2 and PCI-X 1.0a compliant Host interface (82546EB,
82545EM, and 82544GC/EI)

•
•
•
•
•
1.3.2

32/64-bit 33/66 MHz, PCI Rev 2.2 compliant Host interface (82540EP/EM)
64-bit addressing for systems with more than 4 GB of physical memory
Efficient PCI bus master operation
Command usage optimization for advanced PCI commands

CSA Features (82547GI/EI Only)
•
•
•
•

1.3.3

32/64-bit 33/66 MHz, PCI Rev 2.3 compliant Host interface (82541xx)

Uses dedicated port for client LAN controller directly on an MCH device
High-speed interface with twice the peak bandwidth of a 32-bit 33 MHz PCI bus
PCI power management registers recognized by the MCH
Interface only uses 13 signals

Network Side Features
• Auto-Negotiation and Link Setup
— Automatic link configuration including speed, duplex and flow control under IEEE
802.3ab for copper media
— For GMII/MII mode, the driver complies with the IEEE 802.3ab standard requirements
for speed, duplex, and flow control Auto-Negotiation capabilities

• Supports half and full duplex operation at 10 Mb/s and 100 Mb/s speeds while working with
the internal PHY

Software Developer’s Manual

Introduction

• IEEE 802.3x compliant flow control support
— Enables control of the transmission of Pause packets through software or hardware
triggering
— Provides indications of receive FIFO status

• State-of-the-art internal transceiver (PHY) with DSP architecture implementation
— Digital adaptive equalization and crosstalk
— Echo and crosstalk cancellation
— Automatic MDI/MDI-X crossover at all speeds and compensation for cable length
— Media Independent Interfaces (MII) IEEE 802.3e for supporting 10/10BASE-T
transceivers

• Integrated dual-port solution comprised of two distinct MAC/PHY instances (82546GB/EB)
• Provides on-chip IEEE 802.3z PCS SerDes functionality (82546GB/EB and 82545GM/EM)
1.3.4

Host Offloading Features
•
•
•
•

Receive and transmit IP and TCP/UDP checksum offloading capabilities
Transmit TCP Segmentation (operating system support required)
Packet filtering based on checksum errors
Support for various address filtering modes:
— 16 exact matches (unicast, or multicast)
— 4096-bit hash filter for multicast frames
— Promiscuous, unicast and promiscuous multicast transfer modes

• IEEE 802.1q VLAN support1
— Ability to add and strip IEEE 802.1q VLAN tags
— Packet filtering based on VLAN tagging, supporting 4096 tags

• SNMP and RMON statistic counters
• Support for IPv6 including (not applicable to the 82544GC/EI):
— IP/TCP and IP/UDP receive checksum offload
— Wake up filters
— TCP segmentation

1. Not applicable to the 82541ER.

Software Developer’s Manual

Introduction

1.3.5

Additional Performance Features
• Provides adaptive Inter Frame Spacing (IFS) capability, enabling collision reduction in half
duplex networks (82544GC/EI)

• Programmable host memory receive buffers (256 B to 16 KB)
• Programmable cache line size from 16 B to 128 B for efficient usage of PCI bandwidth
• Implements a total of 64 KB (40 KB for the 82547GI/EI) of configurable receive and transmit
data FIFOs. Default allocation is 48 KB for the receive data FIFO and 16 KB for the transmit
data FIFO

• Descriptor ring management hardware for transmit and receive. Optimized descriptor fetching
and write-back mechanisms for efficient system memory and PCI bandwidth usage

• Provides interrupt coalescing to reduce the number of interrupts generated by receive and
transmit operations (82544GC/EI)

• Supports reception and transmission of packets with length up to 16 KB
• New intelligent interrupt generation features to enhance driver performance (not applicable to
the 82544GC/EI):
— Packet interrupt coalescing timers (packet timers) and absolute-delay interrupt timers for
both transmit and receive operation
— Short packet detection interrupt for improved response time to TCP acknowledges
— Transmit Descriptor Ring “Low” signaling
— Interrupt throttling control to limit maximum interrupt rate and improve CPU utilization

Software Developer’s Manual

Introduction

1.3.6

Manageability Features (Not Applicable to the 82544GC/EI or
82541ER)
• Manageability support for ASF 1.0 and AoL 2.0 by way of SMBus 2.0 interface and either:
— TCO mode SMBus-based management packet transmit / receive support
— Internal ASF-compliant TCO controller

1.3.7

Additional Ethernet Controller Features
• Implements ACPI1 register set and power down functionality supporting D0 and D3 states
• Supports Wake on LAN (WoL)1
• Provides four wire serial EEPROM interface for loading product configuration information
— Allows use of either 3.3 V dc or 5 V dc powered EEPROM

• Provides external parallel interface for up to 512 KB of FLASH memory for support of PreBoot Execution Environment (PXE)

•
•
•
•

Provides seven general purpose user mode pins
Provides Activity and Link LED indications
Supports little-endian byte ordering for 32- and 64-bit systems
Provides loopback capabilities under TBI (82544GC/EI)2 (internal SerDes for the 82546GB/
EB and 82545GM/EM) and GMII/MII modes of operation

• Provides IEEE JTAG boundary scan support
• Four programmable LED outputs (Not applicable to the 82544GC/EI).
— For the 82546GB/EB, four programmable LED outputs for each port

• Detection and improved power-management with LAN cable unconnected (82546GB/EB)
1.3.8

Technology Features
• Implemented in 0.15 CMOS process (0.13 for the 82541xx and 82547GI/EI)
• Packaged in 364 PBGA.
— For the 82544EI, packaged in 416 PBGA.
— For the 82540EP/EM, 82541xx, and 82547GI/EI, packaged in 196 PBGA.

• Implemented in low power (3.3 V dc or 5 V dc compatible PCI signaling) CMOS process

1. Not applicable to the 82541ER.
2. Not applicable to the 82541xx, 82547GI/EI or 82540EP/EM.

Software Developer’s Manual

Introduction

1.4

Conventions
This document uses notes that call attention to important comments:
Note:

1.4.1

Indicates details about the hardware’s operations that are not immediately obvious. Read these
notes to get information about exceptions, unusual situations, and additional explanations of some
PCI/PCI-X Family of Gigabit Ethernet Controller features.

Register and Bit References
This document refers to Ethernet controller register names using all capital letters. To refer to a
specific bit in a register the convention REGISTER.BIT is used. For example, CTRL.ASDE refers
to the Auto-Speed Detection Enable bit in the Device Control Register (CTRL).

1.4.2

Byte and Bit Designations
This document uses “B” to abbreviate quantities of bytes. For example, a 4 KB represents 4096
bytes. Similarly, “b” is used to represent quantities of bits. For example, 100 Mb/s represents 100
Megabits per second.

1.5

Related Documents
• IEEE Std. 802.3, 2000 Edition. Incorporates various IEEE standards previously published
separately.

• PCI Local Bus Specification, Revision 2.2 and 2.3, PCI Local Bus Special Interest Group.

1.6

Memory Alignment Terminology
Some PCI/PCI-X Family of Gigabit Ethernet Controller data structures have special memory
alignment requirements. This implies that the starting physical address of a data structure must be
aligned as specified in this manual. The following terms are used for this purpose:

• BYTE alignment: Implies that the physical addresses can be odd or even. Examples:
0FECBD9A1h, 02345ADC6h.

• WORD alignment: Implies that physical addresses must be aligned on even boundaries. For
example, the last nibble of the address can only end in 0, 2, 4, 6, 8, Ah, Ch, or Eh
(0FECBD9A2h).

• DWORD (Double-Word) alignment: Implies that the physical addresses can only be aligned
on 4-byte boundaries. For example, the last nibble of the address can only end in 0, 4, 8, or Ch
(0FECBD9A8h).

• QWORD (Quad-Word) alignment: Implies that the physical addresses can only be aligned on
8-byte boundaries. For example, the last nibble of the address can only end in 0 or 8
(0FECBD9A8h).

• PARAGRAPH alignment: Implies that the physical addresses can only be aligned on 16-byte
boundaries. For example, the last nibble must be a 0 (02345ADC0h).

Software Developer’s Manual

Architectural Overview

Architectural Overview
2.1

Introduction
This section provides an overview of the PCI/PCI-X Family of Gigabit Ethernet Controllers. The
following sections give detailed information about the Ethernet controller’s functionality, register
description, and initialization sequence. All major interfaces of the Ethernet controllers are
described in detail.
The following principles shaped the design of the PCI/PCI-X Family of Gigabit Ethernet
Controllers:
1. Provide an Ethernet interface containing a 10/100/1000 Mb/s PHY that also supports 1000
Base-X implementations.
2. Provide the highest performance solution possible, based on the following:
— Provide direct access to all memory without using mapping registers
— Minimize the PCI target accesses required to manage the Ethernet controller
— Minimize the interrupts required to manage the Ethernet controller
— Off-load the host processor from simple tasks such as TCP checksum calculations
— Maximize PCI efficiency and performance
— Use mixed signal processing to assure physical layer characteristics surpass specifications
for UTP copper media
3. Provide a simple software interface for basic operations.
4. Provide a highly configurable design that can be used effectively in different environments.
The PCI/PCI-X Family of Gigabit Ethernet Controllers architecture is a derivative of the 82542
and 82543 designs. They take the MAC functionality and integrated copper PHY from their
predecessors and adds SMBus-based manageability and integrated ASF controller functionality to
the MAC1. In addition, the 82546GB/EB features this architecture in an integrated dual-port
solution comprised of two distinct MAC/PHY instances.

Not applicable to the 82544GC/EI or 82541ER.

Software Developer’s Manual

Architectural Overview

2.2

External Architecture
Figure 2-1 shows the external interfaces to the 82546GB/EB.
MDI
Interface A

MDI
Interface B

1000Base-T PHY Interfaces

Design for
Test Interface
External
TBI Interface

LEDs

10/100/1000
PHY
MDIO

10/100/1000
PHY

GMII/
MII

MDIO

Device
Function 0
MAC/Controller
(LAN A)

SMBus
Interface

GMII/
MII

EEPROM
Interface
Flash Interface

Device
Function 1
MAC/Controller
(LAN B)

LEDs
Software
Defined Pins

Software
Defined Pins

PCI (64-bit, 33/66 MHz)/PCI-X (133 MHz)

Figure 2-1. 82546GB/EB External Interface
Figure 2-2 shows the external interfaces to the 82545GM/EM, 82544GC/EI, 82540EP/EM, and
82541xx.
MDI
Interface
1000Base-T PHY Interface

Design for
Test Interface
External
TBI Interface
(82545GM/EM only)

LEDs

10/100/1000
PHY
MDIO

SMBus
Interface

GMII/
MII

EEPROM
Interface
Flash Interface

Device
Function 0
MAC/Controller

Software
Defined Pins

PCI (64-bit, 33/66 MHz)/PCI-X (133 MHz)
Note: 82540EP/EM and 82541xx do not support PCI-X; 82544GC/EI and 82541ER do not support SMBus interface

Figure 2-2. 82545GM/EM, 82544GC/EI, 82540EP/EM, and 82541xx External Interface

Software Developer’s Manual

Architectural Overview

Figure 2-3 shows the external interfaces to the 82547GI/EI.
CSA Port

PCI Core

Slave
Access
Logic

EEPROM

FLASH

DMA Function
Descriptor Management

Control
Status
Logic

RX Filters
(Perfect,
Multicast,
VLAN)

TX/RX MAC
CSMA/CD

40KB
Packet
RAM

VLA
N
Statistics

8 bits

Management
Interface

8 bits

Trellis Viterbi
Encoder/Decoder

Side-stream
Scrambler/
Descrambler
4 bits

PHY
Control

4 bits

ECHO, NEXT,
FEXT
Cancellers

4DPAM5
Encoder

AGC, A/D
Timing
Recovery

Pulse Shaper,
DAC, Filter

Hybrid

Line Driver

Media Dependent Interface

Figure 2-3. 82547GI(EI) External Interface

Software Developer’s Manual

Architectural Overview

2.3

Microarchitecture
Compared to its predecessors, the PCI/PCI-X Family of Gigabit Ethernet Controller’s MAC adds
improved receive-packet filtering to support SMBus-based manageability, as well as the ability to
transmit SMBus-based manageability packets. In addition, an ASF-compliant TCO controller is
integrated into the controller’s MAC for reduced-cost basic ASF manageability.
Note:

The 82544GC/EI and 82541ER do not support SMBus-based manageability.
For the 82546GB/EB, this new functionality is packaged in an integrated dual-port combination.
The architecture includes two instances of both the MAC and PHY along with a single PCI/PCI-X
interface. As a result, each of the logical LAN devices appear as a distinct PCI/PCI-X bus device.
The following sections describe the hardware building blocks. Figure 2-4 shows the internal
microarchitecture.

2.3.1

PCI/PCI-X Core Interface
The PCI/PCI-X core provides a complete glueless interface to a 33/66 MHz, 32/64-bit PCI bus or a
33/66/133 MHz, 32/64 bit PCI-X bus. It is compliant with the PCI Bus Specification Rev 2.2 or 2.3
and the PCI-X Specification Rev. 1.0a. The Ethernet controllers provide 32 or 64 bits of addressing
and data, and the complete control interface to operate on a 32-bit or 64-bit PCI or PCI-X bus. In
systems with a dedicated bus for the Ethernet controller, this provides sufficient bandwidth to
support sustained 1000 Mb/s full-duplex transfer rates. Systems with a shared bus (especially the
32-bit wide interface) might not be able to maintain 1000 Mb/s, but can sustain multiple hundreds
of Mbps.

Host Arbiter

TX
Switch

PCI Interface

PCI/
PCI-X
Core

DMA
Engine
Packet/
Manageability
Filter

TX MAC
(10/100/
1000 Mb)
RX MAC
(10/100/
1000 Mb)

Link I/F
GMII/
MII

MDIO
Packet
Buffer

ASF
Manageability

RMON
Statistics

MDIO

SM Bus

EEPROM

Flash

Figure 2-4. Internal Architecture Block Diagram

Software Developer’s Manual

Architectural Overview

When the Ethernet controller serves as a PCI target, it follows the PCI configuration specification,
which allows all accesses to it to be automatically mapped into free memory and I/O space at
initialization of the PCI system.
When processing transmit and receive frames, the Ethernet controller operates as master on the PCI
bus. As a master, transaction burst length on the PCI bus is determined by several factors, including
the PCI latency timer expiration, the type of bus transfer being made, the size of the data transfer,
and whether the data transfer is initiated by receive or transmit logic.
The PCI/PCI-X bus interfaces to the DMA engine.

2.3.2

82547GI/EI CSA Interface
CSA is derived from the Intel® Hub Architecture. The 82547EI Controller CSA port consists of 11
data and control signals, two strobes, a 66 MHz clock, and driver compensation resistor connections. The operating details of these signals and the packet data protocol that accompanies them are
proprietary. The CSA port has a theoretical bandwidth of 266 MB/s — approximately twice the
peak bandwidth of a 32-bit 33 MHz PCI bus.
The CSA port architecture is invisible to both system software and the operating system, allowing
conventional PCI-like configuration.

2.3.3

DMA Engine and Data FIFO
The DMA engine handles the receive and transmit data and descriptor transfers between the host
memory and the on-chip memory.
In the receive path, the DMA engine transfers the data stored in the receive data FIFO buffer to the
receive buffer in the host memory, specified by the address in the descriptor. It also fetches and
writes back updated receive descriptors to host memory.
In the transmit path, the DMA engine transfers data stored in the host memory buffers to the
transmit data FIFO buffer. It also fetches and writes back updated transmit descriptors.
The Ethernet controller data FIFO block consists of a 64 KB (40 KB for the 82547GI/EI) on-chip
buffer for receive and transmit operation. The receive and transmit FIFO size can be allocated
based on the system requirements. The FIFO provides a temporary buffer storage area for frames
as they are received or transmitted by the Ethernet controller.
The DMA engine and the large data FIFOs are optimized to maximize the PCI bus efficiency and
reduce processor utilization by:

• Mitigating instantaneous receive bandwidth demands and eliminating transmit underruns by
buffering the entire out-going packet prior to transmission

• Queuing transmit frames within the transmit FIFO, allowing back-to-back transmission with
the minimum interframe spacing

• Allowing the Ethernet controller to withstand long PCI bus latencies without losing incoming
data or corrupting outgoing data

• Allowing the transmit start threshold to be tuned by the transmit FIFO threshold. This
adjustment to system performance is based on the available PCI bandwidth, wire speed, and
latency considerations

Software Developer’s Manual

Architectural Overview

• Offloading the receiving and transmitting IP and TCP/UDP checksums
• Directly retransmitting from the transmit FIFO any transmissions resulting in errors (collision
detection, data underrun), thus eliminating the need to re-access this data from host memory

2.3.4

10/100/1000 Mb/s Receive and Transmit MAC Blocks
The controller’s CSMA/CD unit handles all the IEEE 802.3 receive and transmit MAC functions
while interfacing between the DMA and TBI/internal SerDes/MII/GMII interface block. The
CSMA/CD unit supports IEEE 802.3 for 10 Mb/s, IEEE 802.3u for 100 Mb/s and IEEE 802.3z and
IEEE 802.3ab for 1000 Mb/s.
The Ethernet controller supports half-duplex 10/100 Mb/s MII or 1000 Mb/s GMII mode and all
aspects of the above specifications in full-duplex operation. In half-duplex mode, the Ethernet
controller supports operation as specified in IEEE 802.3z specification. In the receive path, the
Ethernet controller supports carrier extended packets and packets generated during packet bursting
operation. The 82554GC/EI, in the transmit path, also supports carrier extended packets and can
be configured to transmit in packet burst mode.
The Ethernet controller offers various filtering capabilities that provide better performance and
lower processor utilization as follows:

• Provides up to 16 addresses for exact match unicast/multicast address filtering.
• Provides multicast address filtering based on 4096 bit vectors. Promiscuous unicast and
promiscuous multicast filtering are supported as well.

• The Ethernet controller strips IEEE 802.1q VLAN tag and filter packets based on their VLAN
ID. Up to 4096 VLAN tags are supported1.

In the transmit path, the Ethernet controller supports insertion of VLAN tag information, on a
packet-by-packet basis.
The Ethernet controller implements the flow control function as defined in IEEE 802.3x, as well as
specific operation of asymmetrical flow control as defined by IEEE 802.3z. The Ethernet controller
also provides external pins for controlling the flow control function through external logic.

2.3.5

MII/GMII/TBI/Internal SerDes Interface Block
The Ethernet controller provides the following serial interfaces:

• A GMII/MII interface to the internal PHY.
• Internal SerDes interface2 (82546GB/EB and 82545GM/EM)/Ten Bit Interface (TBI)2 for the
82544GC/EI: The Ethernet controller implements the 802.3z PCS function, the AutoNegotiation function and 10-bit data path interface (TBI) for both receive and transmit
operations. It is used for 1000BASE-SX, -LX, and -CX configurations, operating only at 1000
Mb/s full-duplex. The on-chip PCS circuitry is only used when the link interface is configured
for TBI mode and it is bypassed in internal PHY modes.

1.
2.

Not applicable to the 82541ER.
Not applicable to the 82544GC/EI, 82540EP/EM, 82541xx, and 82547GI/EI.

Software Developer’s Manual

Architectural Overview

Note:

Refer to the Extended Device Control Register (bits 23:22) for mode selection (see Section 13.4.6).
The link can be configured by several methods. Software can force the link setting to AutoNegotiation by setting either the MAC in TBI mode (internal SerDes for the 82546GB/EB and
82545GM/EM), or the PHY in internal PHY mode.
The speed of the link in internal PHY mode can be determined by several methods:

• Auto speed detection based on the receive clock signal generated by the PHY.
• Detection of the PHY link speed indication.
• Software forcing the configuration of link speed.

2.3.6

10/100/1000 Ethernet Transceiver (PHY)
The Ethernet controller provides a full high-performance, integrated transceiver for 10/100/
1000 Mb/s data communication. The physical layer (PHY) blocks are 802.3 compliant and capable
of operating in half-duplex or full-duplex modes.
Highlights of the PHY blocks are as follows:

• Data stream serializers and encoders. Encoding techniques include Manchester, 4B/5B and
4D/PAM5. These blocks also perform data scrambling for 100/1000 Mb/s transmission as a
technique to minimize radiated Electromagnetic Interference (EMI).

• A multi-mode transmit digital to analog converter, which produces filtered waveforms
appropriate for the 10BASE-T, 100BASE-TX or 1000BASE-T Ethernet standards.

• Receiver Analog-to-Digital Converter (ADC). The ADC uses a 125 MHz sampling rate.
• Receiver decoders. These blocks perform the inverse operations of serializers, encoders and
scramblers.

• Active hybrid and echo canceller blocks. The active hybrid and echo canceller blocks reduce
the echo effect of transmitting and receiving simultaneously on the same analog pairs.

• NEXT canceller. This unit removes high frequency Near End Crosstalk induced among
adjacent signal pairs.

• Additional wave shaping and slew rate control circuitry to reduce EMI.
Because the Ethernet controller is IEEE-compliant, the PHY blocks communicate with the MAC
blocks through an internal GMII/MII bus operating at clock speeds of 2.5 MHz up to 125 MHz.
The Ethernet controller also uses an IEEE-compliant internal Management Data interface to
communicate control and status information to the PHY.

2.3.7

EEPROM Interface
The PCI/PCI-X Family of Gigabit Ethernet Controllers provide a four-wire direct interface to a
serial EEPROM device such as the 93C46 or compatible for storing product configuration
information. Several words of the data stored in the EEPROM are automatically accessed by the
Ethernet controller, after reset, to provide pre-boot configuration data to the Ethernet controller
before it is accessible by the host software. The remainder of the stored information is accessed by
various software modules to report product configuration, serial number and other parameters.

Software Developer’s Manual

Architectural Overview

2.3.8

FLASH Memory Interface
The Ethernet controller provides an external parallel interface to a FLASH device. Accesses to the
FLASH are controlled by the Ethernet controller and are accessible to software as normal PCI
reads or writes to the FLASH memory mapping area. The Ethernet controller supports FLASH
devices with up to 512 KB of memory.
Note:

2.4

The 82540EP/EM provides an external interface to a serial FLASH or Boot EEPROM device. See
Appendix B for more information.

DMA Addressing
In appropriate systems, all addresses mastered by the Ethernet controller are 64 bits in order to
support systems that have larger than 32-bit physical addressing. Providing 64-bit addresses
eliminates the need for special segment registers.
Note:

The PCI 2.2 or 2.3 Specification requires that any 64-bit address whose upper 32 bits are all 0b
appear as a 32-bit address cycle. The Ethernet controller complies with the PCI 2.2 or 2.3
Specification.
PCI is little-endian; however, not all processors in systems using PCI treat memory as little-endian.
Network data is fundamentally a byte stream. As a result, it is important that the processor and
Ethernet controller agree about the representation of memory data. The default is little-endian
mode.
Descriptor accesses are not byte swapped.
The following example illustrates data-byte ordering for little endian. Bytes for a receive packet
arrive in the order shown from left to right.
01 02 03 04 05 06 07 08 09 0a 0b 0c 0d 0e 0f 10 11 12 13 14 15 16 17 18 19 1a 1b 1c 1d 1e

Example 2-1. Byte Ordering
There are no alignment restrictions on packet-buffer addresses. The byte address for the major
words is shown on the left. The byte numbers and bit numbers for the PCI bus are shown across the
top.
Table 2-1. Little Endian Data Ordering

0
7

Byte

Address

Software Developer’s Manual

Architectural Overview

2.5

Ethernet Addressing
Several registers store Ethernet addresses in the Ethernet controller. Two 32-bit registers make up
the address: one is called “high”, and the other is called “low”. For example, the Receive Address
Register is comprised of Receive Address High (RAH) and Receive Address Low (RAL). The least
significant bit of the least significant byte of the address stored in the register (for example, bit 0 of
RAL) is the multicast bit. The LS byte is the first byte to appear on the wire. This notation applies
to all address registers, including the flow control registers.
Figure 2-5 shows the bit/byte addressing order comparison between what is on the wire and the
values in the unique receive address registers.

Preamble & SFD

...55

Destination Address

Source Address

...XXX

Bit 0 of this byte is first on the wire
dest_addr[0]

Destination address stored
internally as shown here

...

Multicast bit

Figure 2-5. Example of Address Byte Ordering
The address byte order numbering shown in Figure 2-5 maps to Table 2-2. Byte #1 is first on the
wire.
Table 2-2. Intel® Architecture Byte Ordering

Note:

IA Byte #

1 (LSB)

6 (MSB)

Byte Value (Hex)

The notation in this manual follows the convention shown in Table 2-2. For example, the address in
Table 2-2 indicates 00_AA_00_11_22_33h, where the first byte (00h_) is the first byte on the wire,
with bit 0 of that byte transmitted first.

Software Developer’s Manual

Architectural Overview

2.6

Interrupts
The Ethernet controller provides a complete set of interrupts that allow for efficient software
management. The interrupt structure is designed to accomplish the following:

• Make accesses “thread-safe” by using ‘set’ and ‘clear-on-read’ rather than ‘read-modify-write’
operations.

• Minimize the number of interrupts needed relative to work accomplished.
• Minimize the processing overhead associated with each interrupt.
Intel accomplished the first goal by an interrupt logic consisting of four interrupt registers. More
detail about these registers is given in sections 13.4.17 through 13.4.21.

• Interrupt Cause ‘Set’ and ‘Read’ Registers
The Read register records the cause of the interrupt. All bits set at the time of the read are autocleared. The cause bit is set for each bit written as a 1b in the Set register. If there is a race
between hardware setting a cause and software clearing an interrupt, the bit remains set. No
race condition exists on writing the Set register. A ‘set’ provides for software posting of an
interrupt. A ‘read’ is auto-cleared to avoid expensive write operations. Most systems have
write buffering, which minimizes overhead, but typically requires a read operation to
guarantee that the write operation has been flushed from the posted buffers. Without autoclear, the cost of clearing an interrupt can be as high as two reads and one write.

• Interrupt Mask ‘Set’ (Read) and ‘Clear’ Registers
Interrupts appear on PCI only if the interrupt cause bit is a 1b, and the corresponding interrupt
mask bit is a 1b. Software can block assertion of the interrupt wire by clearing the bit in the
mask register. The cause bit stores the interrupt event regardless of the state of the mask bit.
The Clear and Set operations make this register more “thread-safe” by avoiding a ‘readmodify-write’ operation on the mask register. The mask bit is set to a 1b for each bit written in
the Set register, and cleared for each bit written in the Clear register. Reading the Set register
returns the current value.
Intel accomplished the second goal (minimizing interrupts) by three actions:

• Reducing the frequency of all interrupts (see Section 13.4.17). Not applicable to the
82544GC/EI.

• Accepting multiple receive packets before signaling an interrupt (see Section 3.2.3)
• Eliminating (or at least reducing) the need for interrupts on transmit (see Section 3.2.7)
The third goal is accomplished by having one interrupt register consolidate all interrupt
information. This eliminates the need for multiple accesses.
Note that the Ethernet controller also supports Message Signaled Interrupts as defined in the PCI
2.2, 2.3, and PCI-X specifications. See Section 4.1.3.1 for details.

Software Developer’s Manual

Architectural Overview

2.7

Hardware Acceleration Capability
The Ethernet controller provides the ability to offload IP, TCP, and UDP checksum for transmit.
The functionality provided by these features can significantly reduce processor utilization by
shifting the burden of the functions from the driver to the hardware.
The checksum offloading feature is briefly outlined in the following sections. More detail about all
of the hardware acceleration capabilities is provided in Section 3.2.9.

2.7.1

Checksum Offloading
The Ethernet controller provides the ability to offload the IP, TCP, and UDP checksum requirements from the software device driver. For common frame types, the hardware automatically
calculates, inserts, and checks the appropriate checksum values normally handled by software.
For transmits, every Ethernet packet might have two checksums calculated and inserted by the
Ethernet controller. Typically, these would be the IP checksum, and either the TCP or UDP
checksum. The software device driver specifies which portions of the packet are included in the
checksum calculations, and where the calculated values are inserted via descriptors (refer to
Section 3.3.5 for details).
For receives, the hardware recognizes the packet type and performs the checksum calculations and
error checking automatically. Checksum and error information is provided to software through the
receive descriptors (refer to Section 3.2.9 for details).

2.7.2

TCP Segmentation
The Ethernet controller implements a TCP segmentation capability for transmits that allows the
software device driver to offload packet segmentation and encapsulation to the hardware. The
software device driver can send the Ethernet controller the entire IP, TCP or UDP message sent
down by the Network Operating System (NOS) for transmission. The Ethernet controller segments
the packet into legal Ethernet frames and transmit them on the wire. By handling the segmentation
tasks, the hardware alleviates the software from handling some of the framing responsibilities. This
reduces the overhead on the CPU for the transmission process thus reducing overall CPU
utilization. See Section 3.5 for details.

2.8

Buffer and Descriptor Structure
Software allocates the transmit and receive buffers, and also forms the descriptors that contain
pointers to, and the status of, those buffers. A conceptual ownership boundary exists between the
driver software and the hardware of the buffers and descriptors. The software gives the hardware
ownership of a queue of buffers for receives. These receive buffers store data that the software then
owns once a valid packet arrives.
For transmits, the software maintains a queue of buffers. The driver software owns a buffer until it
is ready to transmit. The software then commits the buffer to the hardware; the hardware then owns
the buffer until the data is loaded or transmitted in the transmit FIFO.

Software Developer’s Manual

Architectural Overview

Descriptors store the following information about the buffers:

• The physical address
• The length
• Status and command information about the referenced buffer
Descriptors contain an end-of-packet field that indicates the last buffer for a packet. Descriptors
also contain packet-specific information indicating the type of packet, and specific operations to
perform in the context of transmitting a packet, such as those for VLAN or checksum offload.
Section 3 provides detailed information about descriptor structure and operation in the context of
packet transmission and reception.

Software Developer’s Manual

Receive and Transmit Description

Receive and Transmit Description
3.1

Introduction
This section describes the packet reception, packet transmission, transmit descriptor ring structure,
TCP segmentation, and transmit checksum offloading for the PCI/PCI-X Family of Gigabit
Ethernet Controllers.
Note:

3.2

The 82544GC/EI does not support IPv6.

Packet Reception
In the general case, packet reception consists of recognizing the presence of a packet on the wire,
performing address filtering, storing the packet in the receive data FIFO, transferring the data to a
receive buffer in host memory, and updating the state of a receive descriptor.

3.2.1

Packet Address Filtering
Hardware stores incoming packets in host memory subject to the following filter modes. If there is
insufficient space in the receive FIFO, hardware drops them and indicates the missed packet in the
appropriate statistics registers.
The following filter modes are supported:

• Exact Unicast/Multicast — The destination address must exactly match one of 16 stored
addresses. These addresses can be unicast or multicast.

• Promiscuous Unicast — Receive all unicasts.
• Multicast — The upper bits of the incoming packet’s destination address index a bit vector
that indicates whether to accept the packet; if the bit in the vector is one, accept the packet,
otherwise, reject it. The controller provides a 4096 bit vector. Software provides four choices
of which bits are used for indexing. These are [47:36], [46:35], [45:34], or [43:32] of the
internally stored representation of the destination address.

• Promiscuous Multicast — Receive all multicast packets.
• VLAN — Receive all VLAN1 packets that are for this station and have the appropriate bit set
in the VLAN filter table. A detailed discussion and explanation of VLAN packet filtering is
contained in Section 9.3.
Normally, only good packets are received. These are defined as those packets with no CRC error,
symbol error, sequence error, length error, alignment error, or where carrier extension or receive
errors are detected. However, if the store–bad–packet bit is set in the Device Control register
(RCTL.SBP), then bad packets that pass the filter function are stored in host memory. Packet errors
are indicated by error bits in the receive descriptor (RDESC.ERRORS). It is possible to receive all
packets, regardless of whether they are bad, by setting the promiscuous enables (RCTL.UPE/MPE)
and the store–bad–packet bit (RCTL.SBP).
1. Not applicable to the 82541ER.

Software Developer’s Manual

Receive and Transmit Description

If manageability is enabled and if RCMCP is enabled then ARP request packets can be directed
over the SMBus or processed internally by the ASF controller rather than delivered to host memory
(not applicable to the 82544GC/EI or 82541ER.

3.2.2

Receive Data Storage
Memory buffers pointed to by descriptors store packet data. Hardware supports seven receive
buffer sizes:

•
•
•
•

256 B
512 B
1024 B

• 4096 B
• 8192 B
• 16384 B

2048 B

Buffer size is selected by bit settings in the Receive Control register (RCTL.BSIZE &
RCTL.BSEX). See Section 13.4.22 for details.
The Ethernet controller places no alignment restrictions on packet buffer addresses. This is
desirable in situations where the receive buffer was allocated by higher layers in the networking
software stack, as these higher layers may have no knowledge of a specific Ethernet controller’s
buffer alignment requirements.
Although alignment is completely unrestricted, it is highly recommended that software allocate
receive buffers on at least cache-line boundaries whenever possible.

3.2.3

Receive Descriptor Format
A receive descriptor is a data structure that contains the receive data buffer address and fields for
hardware to store packet information. Table 3-1 lists where the shaded areas indicate fields that are
modified by hardware upon packet reception.
Table 3-1. Receive Descriptor (RDESC) Layout

48 47

40 39

32 31

16 15

Buffer Address [63:0]

Special

Errors

Status

Packet Checksum
(See Note)

Length

82544GC/EI only
63

48 47

40 39

0
8

Note:

32 31

16 15

Buffer Address [63:0]
Reserved

Errors

Status

Reserved

Length

The checksum indicated here is the unadjusted “16 bit ones complement” of the packet. A software
assist may be required to back out appropriate information prior to sending it to upper software

Software Developer’s Manual

Receive and Transmit Description

layers. The packet checksum is always reported in the first descriptor (even in the case of multidescriptor packets).
Upon receipt of a packet for Ethernet controllers, hardware stores the packet data into the indicated
buffer and writes the length, Packet Checksum, status, errors, and status fields. Length covers the
data written to a receive buffer including CRC bytes (if any). Software must read multiple
descriptors to determine the complete length for packets that span multiple receive buffers.
For standard 802.3 packets (non-VLAN) the Packet Checksum is by default computed over the
entire packet from the first byte of the DA through the last byte of the CRC, including the Ethernet
and IP headers. Software may modify the starting offset for the packet checksum calculation by
means of the Receive Control Register. This register is described in Section 13.4.22. To verify the
TCP checksum using the Packet Checksum, software must adjust the Packet Checksum value to
back out the bytes that are not part of the true TCP Checksum.

3.2.3.1

Receive Descriptor Status Field
Status information indicates whether the descriptor has been used and whether the referenced
buffer is the last one for the packet. Refer to Table 3-2 for the layout of the status field. Error status
information is shown in Table 3-3.
For multi-descriptor packets, packet status is provided in the final descriptor of the packet (EOP
set). If EOP is not set for a descriptor, only the Address, Length, and DD bits are valid.
Table 3-2. Receive Status (RDESC.STATUS) Layout
7

PIF

IPCS

TCPCS

RSV

IXSM

EOP

Receive
Descriptor Status
Bits

Description

PIF (bit 7)

Passed in-exact filter
Hardware supplies the PIF field to expedite software processing of packets.
Software must examine any packet with PIF set to determine whether to accept
the packet. If PIF is clear, then the packet is known to be for this station, so
software need not look at the packet contents. Packets passing only the
Multicast Vector has PIF set.

IPCS (bit 6)

IP Checksum Calculated on Packet
When Ignore Checksum Indication is deasserted (IXSM = 0b), IPCS bit indicates
whether the hardware performed the IP checksum on the received packet.
0b = Do not perform IP checksum
1b = Perform IP checksum
Pass/Fail information regarding the checksum is indicated in the error bit (IPE) of
the descriptor receive errors (RDESC.ERRORS)
IPv6 packets do not have the IPCS bit set.
Reads as 0b.

Software Developer’s Manual

Receive and Transmit Description

Receive
Descriptor Status
Bits

Note:

3.2.3.2

Description

TCPCS (bit 5)

TCP Checksum Calculated on Packet
When Ignore Checksum Indication is deasserted (IXSM = 0b), TCPCS bit
indicates whether the hardware performed the TCP/UDP checksum on the
received packet.
0b = Do not perform TCP/UDP checksum; 1b = Perform TCP/UDP checksum
Pass/Fail information regarding the checksum is indicated in the error bit (TCPE)
of the descriptor receive errors (RDESC.ERRORS).
IPv6 packets may have this bit set if the TCP/UDP packet was recognized.
Reads as 0b.

RSV (bit 4)

Reserved
Reads as 0b.

VP (bit 3)

Packet is 802.1Q (matched VET)
Indicates whether the incoming packet’s type matches VET (i.e., if the packet is
a VLAN (802.1q) type). It is set if the packet type matches VET and CTRL.VME
is set. For a further description of 802.1q VLANs, see Chapter 9.
Reads as 0b.

IXSM (bit 2)

Ignore Checksum Indication
When IXSM = 1b, the checksum indication results (IPCS, TCPCS bits) should be
ignored.
When IXSM = 0b the IPCS and TCPCS bits indicate whether the hardware
performed the IP or TCP/UDP checksum(s) on the received packet. Pass/Fail
information regarding the checksum is indicated in the status bits as described
below for IPE and TCPE.
Reads as 1b.

EOP (bit 1)

End of Packet
EOP indicates whether this is the last descriptor for an incoming packet.

DD (bit 0)

Descriptor Done
Indicates whether hardware is done with the descriptor. When set along with
EOP, the received packet is complete in main memory.

See Table 3-5 for a description of supported packet types for receive checksum offloading.
Unsupported packet types either have the IXSM bit set, or they don’t have the TCPCS bit set.

Receive Descriptor Errors Field
Most error information appears only when the Store Bad Packets bit (RCTL.SBP) is set and a bad
packet is received. Refer to Table 3-3 for a definition of the possible errors and their bit positions.
The error bits are valid only when the EOP and DD bits are set in the descriptor status field
(RDESC.STATUS)

Software Developer’s Manual

Receive and Transmit Description

Table 3-3. Receive Errors (RDESC.ERRORS) Layout
7

RXE

IPE

TCPE

RSV
CXEa

RSV

SEQ
RSVb

SE
RSVb

a. 82544GC/EI only.
b. 82541xx, 82547GI/EI, and 82540EP/EM only.

Receive
Descriptor Error
bits

Description

RXE (bit 7)

RX Data Error
Indicates that a data error occurred during the packet reception. A data error in TBIa
mode (82544GC/EI)/internal SerDes (82546GB/EB and 82545GM/EM) refers to the
reception of a /V/ code (see Section 8.2.1.3). In GMII or MII mode, the assertion of
I_RX_ER during data reception indicates a data error. This bit is valid only when the
EOP and DD bits are set; it is not set in descriptors unless RCTL.SBP (Store Bad
Packets) control bit is set.

IPE (bit 6)

IP Checksum Error
When set, indicates that IP checksum error is detected in the received packet. Valid
only when the IP checksum is performed on the receive packet as indicated via the
IPCS bit in the RDESC.STATUS field.
If receive IP checksum offloading is disabled (RXCSUM.IPOFL), the IPE bit is set to
0b. It has no effect on the packet filtering mechanism.
Reads as 0b.

TCPE (bit 5)

TCP/UDP Checksum Error
When set, indicates that TCP/UDP checksum error is detected in the received
packet.
Valid only when the TCP/UDP checksum is performed on the receive packet as
indicated via TCPCS bit in RDESC.STATUS field.
If receive TCP/UDP checksum offloading is disabled (RXCSUM.TUOFL), the TCPE
bit is set to 0b.
It has no effect on the packet filtering mechanism.
Reads as 0b.

CXE
RSV (bit 4)

Carrier Extension Error
When set, indicates a packet was received in which the carrier extension error was
signaled across the GMII interface. A carrier extension error is signaled by the PHY
by the encoding of 1Fh on the receive data inputs while I_RX_ER is asserted.
Valid only while working in 1000 Mb/s half-duplex mode of operation.
This bit is reserved for all Ethernet controllers except the 82544GC/EI.

RSV (Bit 3)

Reserved
Reads as 0b.

Software Developer’s Manual

Receive and Transmit Description

Receive
Descriptor Error
bits

Description

SEQ (bit 2)

Sequence Error
When set, indicates a received packet with a bad delimiter sequence (in TBI mode/
internal SerDes). In other 802.3 implementations, this would be classified as a
framing error.
A valid delimiter sequence consists of:
idle start-of-frame (SOF)  data, pad (optional)  end-of-frame (EOF)  fill
(optional)  idle.

SE (bit 1)

Symbol Error
When set, indicates a packet received with bad symbol. Applicable only in TBI mode/
internal SerDes.

CE (bit 0)

CRC Error or Alignment Error
CRC errors and alignment errors are both indicated via the CE bit. Software may
distinguish between these errors by monitoring the respective statistics registers.

a. Not applicable to the 82540EP/EM, 82541xx, or 82547GI/EI.

3.2.3.3

Receive Descriptor Special Field
Hardware stores additional information in the receive descriptor for 802.1q packets. If the packet
type is 802.1q, determined when a packet type field matches the VLAN1 Ethernet Register (VET)
and RCTL.VME = 1b, then the special field records the VLAN information and the four byte
VLAN information is stripped from the packet data storage. The Ethernet controller stores the Tag
Control Information (TCI) of the 802.1q tag in the Special field. Otherwise, the special field
contains 0000h.
Table 3-4. Special Descriptor Field Layout
802.1q Packets
15

13
PRI

CFI

0
VLAN

All Other Packets
15

8 7
00

Receive
Descriptor
Special Field

0
00

Description

VLAN

VLAN Identifier
12 bits that records the packet VLAN ID number

CFI

Canonical Form Indicator
1 bit that records the packet’s CFI VLAN field

PRI

User Priority
3 bits that records the packet’s user priority field.

Not applicable to the 82541ER.

Software Developer’s Manual

Receive and Transmit Description

3.2.4

Receive Descriptor Fetching
The descriptor fetching strategy is designed to support large bursts across the PCI bus. This is made
possible by using 64 on-chip receive descriptors and an optimized fetching algorithm. The fetching
algorithm attempts to make the best use of PCI bandwidth by fetching a cache line (or more)
descriptors with each burst. The following paragraphs briefly describe the descriptor fetch
algorithm and the software control provided.
When the on-chip buffer is empty, a fetch happens as soon as any descriptors are made available
(software writes to the tail pointer). When the on-chip buffer is nearly empty
(RXDCTL.PTHRESH), a prefetch is performed whenever enough valid descriptors
(RXDCTL.HTHRESH) are available in host memory and no other PCI activity of greater priority
is pending (descriptor fetches and write-backs or packet data transfers).
When the number of descriptors in host memory is greater than the available on-chip descriptor
storage, the chip may elect to perform a fetch which is not a multiple of cache line size. The
hardware performs this non-aligned fetch if doing so results in the next descriptor fetch being
aligned on a cache line boundary. This mechanism provides the highest efficiency in cases where
fetches fall behind software.
Note:

The Ethernet controller never fetches descriptors beyond the descriptor TAIL pointer.

No
On-chip
descriptor cache
is empty

Yes
No

Yes

On-chip
descriptor cache <
RDXCTL.PTHRESH

Valid descriptors
in host memory >
RXDCTL.HTHRESH

Descriptors
are available in
host memory

Yes

Pre-fetch (based
on PCI priority)

Fetch

Figure 3-1. Receive Descriptor Fetching Algorithm

Software Developer’s Manual

Receive and Transmit Description

3.2.5

Receive Descriptor Write-Back
Processors have cache line sizes that are larger than the receive descriptor size (16 bytes).
Consequently, writing back descriptor information for each received packet would cause expensive
partial cache line updates. Two mechanisms minimize the occurrence of partial line write backs:

• Receive descriptor packing
• Null descriptor padding
The following sections explain these mechanisms.

3.2.5.1

Receive Descriptor Packing
To maximize memory efficiency, receive descriptors are “packed” together and written as a cache
line whenever possible. Descriptors accumulate and are written out in one of three conditions:

• RXDCTL.WTHRESH descriptors have been used (the specified max threshold of unwritten
used descriptors has been reached)

• The receive timer expires (RADV or RDTR)
• Explicit software flush (RDTR.FPD)
For the first condition, if the number of descriptors specified by RXDCTL.WTHRESH are used,
they are written back, regardless of cacheline alignment. It is therefore recommended that
WTHRESH be a multiple of cacheline sizes.
In the second condition, a timer (RDTR or RADV) expiration causes all used descriptors to be
written back prior to initiating an interrupt.
In the second condition for the 82544GC/EI, a timer (RDTR) is included to force timely write–
back of descriptors. The first packet after timer initialization starts the timer. Timer expiration
flushes any accumulated descriptors and sets an interrupt event (receiver timer interrupt). In
general, the arrival rate is sufficiently fast enough that packing is the common case under load.
For the final condition, software may explicitly flush accumulated descriptors by writing the timer
register with the high order bit set.

3.2.5.2

Null Descriptor Padding
Hardware stores no data in descriptors with a null data address. Software can make use of this
property to cause the first condition under receive descriptor packing to occur early. Hardware
writes back null descriptors with the DD bit set in the status byte and all other bits unchanged.

3.2.6

Receive Descriptor Queue Structure
Figure 3-2 shows the structure of the receive descriptor ring. Hardware maintains a circular ring of
descriptors and writes back used descriptors just prior to advancing the head pointer. Head and tail
pointers wrap back to base when “size” descriptors have been processed.
Software adds receive descriptors by writing the tail pointer with the index of the entry beyond the
last valid descriptor. As packets arrive, they are stored in memory and the head pointer is
incremented by hardware. When the head pointer is equal to the tail pointer, the ring is empty.
Hardware stops storing packets in system memory until software advances the tail pointer, making
more receive buffers available.

Software Developer’s Manual

Receive and Transmit Description

The receive descriptor head and tail pointers reference 16-byte blocks of memory. Shaded boxes in
the figure represent descriptors that have stored incoming packets but have not yet been recognized
by software. Software can determine if a receive buffer is valid by reading descriptors in memory
rather than by I/O reads. Any descriptor with a non-zero status byte has been processed by the
hardware, and is ready to be handled by the software.

Circular Buffer Queues

Base
Head

Owned By
Hardware

Receive
Queue

Tail
Base + Size

Figure 3-2. Receive Descriptor Ring Structure
Note:

The head pointer points to the next descriptor that is written back. At the completion of the
descriptor write-back operation, this pointer is incremented by the number of descriptors written
back. HARDWARE OWNS ALL DESCRIPTORS BETWEEN [HEAD AND TAIL]. Any
descriptor not in this range is owned by software.
The receive descriptor ring is described by the following registers:

• Receive Descriptor Base Address registers (RDBAL and RDBAH)
These registers indicate the start of the descriptor ring buffer. This 64-bit address is aligned on
a 16-byte boundary and is stored in two consecutive 32-bit registers. RDBAL contains the
lower 32-bits; RDBAH contains the upper 32 bits. Hardware ignores the lower 4 bits in
RDBAL.

• Receive Descriptor Length register (RDLEN)
This register determines the number of bytes allocated to the circular buffer. This value must
be a multiple of 128 (the maximum cache line size). Since each descriptor is 16 bytes in
length, the total number of receive descriptors is always a multiple of 8.

• Receive Descriptor Head register (RDH)
This register holds a value that is an offset from the base, and indicates the in–progress
descriptor. There can be up to 64K descriptors in the circular buffer. Hardware maintains a
shadow copy that includes those descriptors completed but not yet stored in memory.

Software Developer’s Manual

Receive and Transmit Description

• Receive Descriptor Tail register (RDT)
This register holds a value that is an offset from the base, and identifies the location beyond the
last descriptor hardware can process. Note that tail should still point to an area in the descriptor
ring (somewhere between RDBA and RDBA + RDLEN). This is because tail points to the
location where software writes the first new descriptor.
If software statically allocates buffers, and uses memory read to check for completed descriptors, it
simply has to zero the status byte in the descriptor to make it ready for reuse by hardware. This is
not a hardware requirement (moving the hardware tail pointer is), but is necessary for performing
an in–memory scan.

3.2.7

Receive Interrupts
The Ethernet controller can generate four receive-related interrupts:

•
•
•
•
3.2.7.1

Receiver Timer Interrupt (ICR.RXT0)
Small Receive Packet Detect (ICR.SRPD)
Receive Descriptor Minimum Threshold (ICR.RXDMT0)
Receiver FIFO Overrun (ICR.RX0)

Receive Timer Interrupt
The Receive Timer Interrupt is used to signal most packet reception events (the Small Receive
Packet Detect interrupt is also used in some cases as described later in this section). In order to
minimize the interrupts per work accomplished, the Ethernet controller provides two timers to
control how often interrupts are generated.

3.2.7.1.1

Receive Interrupt Delay Timer / Packet Timer (RDTR)

The Packet Timer minimizes the number of interrupts generated when many packets are received
in a short period of time. The packet timer is started once a packet is received and transferred to
host memory (specifically, after the last packet data byte is written to memory) and is reinitialized
(to the value defined in RDTR) and started EACH TIME a new packet is received and transferred
to the host memory. When the Packet Timer expires (e.g. no new packets have been received and
transferred to host memory for the amount of time defined in RDTR) the Receive Timer Interrupt is
generated.
Setting the Packet Timer to 0b disables both the Packet Timer and the Absolute Timer (described
below) and causes the Receive Timer Interrupt to be generated whenever a new packet has been
stored in memory.
Writing to RDTR with its high order bit (FPD) set forces an explicit writeback of consumed
descriptors (potentially a partial cache lines amount of descriptors), causes an immediate expiration
of the Packet Timer and generates a Receive Timer Interrupt.
The Packet Timer is reinitialized (but not started) when the Receive Timer Interrupt is generated
due to an Absolute timer expiration or Small Receive Packet Detect Interrupt.
See section Section 13.4.30 for more details on the Packet Timer.

Software Developer’s Manual

Receive and Transmit Description

Initial State

Idle

Packet received &
transferred to host
memory
Restart Count

Other receive
timer interrupt

Restart Count

Running

Generate
Int

Packet received &
transferred to host
memory

Timer expires

Figure 3-3. Packet Delay Timer Operation (State Diagram)

3.2.7.1.2

Receive Interrupt Absolute Delay Timer (RADV)

The Absolute Timer ensures that a receive interrupt is generated at some predefined interval after
the first packet is received. The absolute timer is started once a packet is received and transferred to
host memory (specifically, after the last packet data byte is written to memory) but is NOT
reinitialized / restarted each time a new packet is received. When the Absolute Timer expires (no
receive interrupt has been generated for the amount of time defined in RADV) the Receive Timer
Interrupt is generated.
Setting RADV to 0b or RDTR to 0b disables the Absolute Timer. To disable the Packet Timer only,
RDTR should be set to RADV + 1b.
The Absolute Timer is reinitialized (but not started) when the Receive Timer Interrupt is generated
due to a Packet Timer expiration or Small Receive Packet Detect Interrupt.

Software Developer’s Manual

Receive and Transmit Description

The diagrams below show how the Packet Timer and Absolute Timer can be used together:
C ase A: U sing only an absolute tim er
A bsolute Tim er Value

Interrupt generated due to PKT #1
PKT #1

PKT #2

PKT #3

PKT #4

C ase B: U sing an absolute tim e in conjunction w ith the Packet tim er

A bsolute Tim er Value
A bsolute Tim er Value

PKT #1

PKT #2

PKT #3

PKT #4

PKT #5

PKT #6

1) Packet tim er expires
2) Interrupt generated
3) Absolute tim er reset

...

Interrupt generalted (due to PKT #4)
as absolute tim er expires.
Packet delay tim er disabled untill
next packet is received and
transferred to host m em ory.

C ase C : Packet tim er expiring w hile a packet is transferred to host m em ory.
Illustrates that packet tim er is re-started only after a packet is transferred to host m em ory.

A bsolute Tim er Value
A bsolute Tim er Value

PKT #1

PKT #2

PKT #3

1) Packet tim er expires
2) Interrupt generated
3) Absolute tim er reset

3.2.7.2

PKT #4

PKT #5

PKT #6

...

Interrupt generalted (due to PKT #4)
as absolute tim er expires.
Packet delay tim er disabled untill
next packet is received and
transferred to host m em ory.

Small Receive Packet Detect
A Small Receive Packet Detect interrupt (ICR.SRPD) is asserted when small-packet detection is
enabled (RSRPD is set with a non-zero value) and a packet of (size  RSRPD.SIZE) has been
transferred into the host memory. When comparing the size the headers and CRC are included (if
CRC stripping is not enabled). CRC and VLAN headers are not included if they have been
stripped. A receive timer interrupt cause (ICR.RXT0) is also noted when the Small Packet Detect
interrupt occurs.
For the 82541xx and 82547GI/EI, receiving a small packet does not clear the absolute or packet
delay timers, so one packet might generate two interrupts, one due to small packet reception and
one due to timer expiration.

Software Developer’s Manual

Receive and Transmit Description

3.2.7.3

Receive Descriptor Minimum Threshold (ICR.RXDMT)
The minimum descriptor threshold helps avoid descriptor under-run by generating an interrupt
when the number of free descriptors becomes equal to the minimum amount defined in
RCTL.RDMTS (measured as a fraction of the receive descriptor ring size).

3.2.7.4

Receiver FIFO Overrun
FIFO overrun occurs when hardware attempts to write a byte to a full FIFO. An overrun could
indicate that software has not updated the tail pointer to provide enough descriptors/buffers, or that
the PCI bus is too slow draining the receive FIFO. Incoming packets that overrun the FIFO are
dropped and do not affect future packet reception.

3.2.8

82544GC/EI Receive Interrupts
The presence of new packets is indicated by the following:

• Absolute timer (RDTR) — A predetermined amount of time has elapsed since the first packet
received after the hardware timer was written (specifically, after the last packet data byte was
written to memory); this also flushes any accumulated descriptors to memory. Software can set
the timer value to 0b if it wants to be notified each time a new packet has been stored in
memory.
Writing the absolute timer with its high order bit 1 forces an explicit flush of any partial cache
lines. Hardware writes all used descriptors to memory and updates the globally visible value of
the head pointer.
In addition, hardware provides the following interrupts:

• Receive Descriptor Minimum Threshold (ICR.RXDMT)
The minimum descriptor threshold helps avoid descriptor underrun by generating an interrupt
when the number of free descriptors becomes equal to the minimum. It is measured as a
fraction of the receive descriptor ring size.

• Receiver FIFO Overrun (ICR.RXO)
FIFO overrun occurs when hardware attempts to write a byte to a full FIFO. An overrun could
indicate that software has not updated the tail pointer to provide enough descriptors/buffers, or
that the PCI bus is too slow draining the receive FIFO. Incoming packets that overrun the
FIFO are dropped and do not affect future packet reception.

3.2.9

Receive Packet Checksum Offloading
The Ethernet controller supports the offloading of three receive checksum calculations: the Packet
Checksum, the IP Header Checksum, and the TCP/UDP Checksum.
Note:

IPv6 packets do not have IP checksums.

Software Developer’s Manual

Receive and Transmit Description

The Packet checksum is the one’s complement over the receive packet, starting from the byte
indicated by RXCSUM.PCSS (0b corresponds to the first byte of the packet), after stripping. For
example, for an Ethernet II frame encapsulated as an 802.3ac VLAN packet and with
RXCSUM.PCSS set to 14 decimal, the Packet Checksum would include the entire encapsulated
frame, excluding the 14-byte Ethernet header (DA,SA,Type/Length) and the 4-byte q-tag. The
Packet checksum does not include the Ethernet CRC if the RCTL.SECRC bit is set.
Software must make the required offsetting computation (to back out the bytes that should not have
been included and to include the pseudo-header) prior to comparing the Packet Checksum against
the TCP checksum stored in the packet.
For supported packet/frame types, the entire checksum calculation may be offloaded to the
Ethernet controller. If RXCSUM.IPOFLD is set to 1b, the controller calculates the IP checksum
and indicates a pass/fail condition to software by means of the IP Checksum Error bit
(RDESC.IPE) in the ERROR field of the receive descriptor. Similarly, if the RXCSUM.TUOFLD
is set to 1b, the Ethernet controller calculates the TCP or UDP checksum and indicates a pass/fail
condition to software by means of the TCP/UDP Checksum Error bit (RDESC.TCPE). These error
bits are valid when the respective status bits indicate the checksum was calculated for the packet
(RDESC.IPCS and RDESC.TCPCS).
If neither RXCSUM.IPOFLD nor RXCSUM.TUOFLD is set, the Checksum Error bits (IPE and
TCPE) is 0b for all packets.
Supported Frame Types include:

• Ethernet II
• Ethernet SNAP
Note:

See Table 3-6 for the 82544GC/EI supported receive checksum capabilities.

Table 3-5. Supported Receive Checksum Capabilities
HW IP Checksum
Calculation

Packet Type

HW TCP/UDP Checksum
Calculation

IPv4 packets

Yes

IPv6 packets

No (n/a)

Yes

IPv6 packet with next header options:
Hop-by-Hop options
Destinations options
Routing
Fragment

No (n/a)
No (n/a)
No (n/a)
No (n/a)

Yes
Yes
Yes
No

IPv4 tunnels:
IPv4 packet in an IPv4 tunnel
IPv6 packet in an IPv4 tunnel

No
Yes (IPv4)

No
Yesa

IPv6 tunnels:
IPv4 packet in an IPv6 tunnel
IPv6 packet in an IPv6 tunnel

No
No

Packet is an IPv4 fragment

Yes

Packet is greater than 1552 bytes;
(LPE=1b)b

Yes

Packet has 802.3ac tag

Yes

Software Developer’s Manual

Receive and Transmit Description

Table 3-5. Supported Receive Checksum Capabilities
HW IP Checksum
Calculation

Packet Type

HW TCP/UDP Checksum
Calculation

IPv4 Packet has IP options
(IP header is longer than 20 bytes)

Yes

Packet has TCP or UDP options

Yes

IP header’s protocol field contains a
protocol # other than TCP or UDP.

Yes

a. The IPv6 header portion can include supported extension headers as described in the IPv6 Filter
section.
b.For the 82541xx and 82547GI/EI, frame sizes greater than 2 KB require full-duplex operation.
Table 3-6. 82544GC/EI Supported Receive Checksum Capabilities

HW IP Checksum
Calculation

HW TCP/UDP
Checksum Calculation

IP v4 packets

Yes

IP v6 packets
(no IP checksum in IPv6)

Packet is an IP fragment

Yes

Packet is greater than 1552 bytes; (LPE=1)

Yes

Packet has 802.3ac tag

Yes

Packet has IP options
(IP header is longer than 20 bytes)

Yes

Packet has TCP or UDP options

Yes

IP header’s protocol field contains a protocol
other than TCP or UDP.

Yes

Packet Type

Table 3-5 lists the general details about what packets are processed. In more detail, the packets are
passed through a series of filters (Section 3.2.9.1 through Section 3.2.9.5) to determine if a receive
checksum is calculated.
Note:

3.2.9.1

(Section 3.2.9.1 through Section 3.2.9.5) does not apply to the 82544GC/EI.

MAC Address Filter
This filter checks the MAC destination address to be sure it is valid (IA match, broadcast,
multicast, etc.). The receive configuration settings determine which MAC addresses are accepted.
See the various receive control configuration registers such as RCTL (RTCL.UPE, RCTL.MPE,
RCTL.BAM), MTA, RAL, and RAH.

Software Developer’s Manual

Receive and Transmit Description

3.2.9.2

SNAP/VLAN Filter
This filter checks the next headers looking for an IP header. It is capable of decoding Ethernet II,
Ethernet SNAP, and IEEE 802.3ac headers. It skips past any of these intermediate headers and
looks for the IP header. The receive configuration settings determine which next headers are
accepted. See the various receive control configuration registers such as RCTL (RCTL.VFE), VET,
and VFTA.

3.2.9.3

IPv4 Filter
This filter checks for valid IPv4 headers. The version field is checked for a correct value (4). IPv4
headers are accepted if they are any size greater than or equal to 5 (dwords). If the IPv4 header is
properly decoded, the IP checksum is checked for validity. The RXCSUM.IPOFL bit must be set
for this filter to pass.

3.2.9.4

IPv6 Filter
This filter checks for valid IPv6 headers, which are a fixed size and have no checksum. The IPv6
extension headers accepted are: Hop-by-Hop, Destination Options, and Routing. The maximum
size next header accepted is 16 dwords (64 bytes).
All of the IPv6 extension headers supported by the Ethernet controller have the same header
structure:
Byte 0

Byte 1

Next Header

Hdr Ext Len

Byte 2

Byte 3

• NEXT HEADER is a value that identifies the header type. The supported IPv6 next headers
values are:
— Hop-by-Hop = 00h
— Destination Options = 3Ch
— Routing = 2Bh

• HDR EXT LEN is the 8 byte count of the header length, not including the first 8 bytes. For
example, a value of 3 means that the total header size including the NEXT HEADER and
HDR EXT LEN fields is 32 bytes (8 + 3*8).
— The RXCSUM.IPV6OFL bit must be set for this filter to pass.

3.2.9.5

UDP/TCP Filter
This filter checks for a valid UDP or TCP header. The prototype next header values are 11h and
06h, respectively. The RXCSUM.TUOFL bit must be set for this filter to pass.

3.3

Packet Transmission
The transmission process for regular (non-TCP Segmentation packets) involves:

• The protocol stack receives from an application a block of data that is to be transmitted.

Software Developer’s Manual

Receive and Transmit Description

• The protocol stack calculates the number of packets required to transmit this block based on
the MTU size of the media and required packet headers.

• For each packet of the data block:
— Ethernet, IP and TCP/UDP headers are prepared by the stack.
— The stack interfaces with the software device driver and commands the driver to send the
individual packet.
— The driver gets the frame and interfaces with the hardware.
— The hardware reads the packet from host memory (via DMA transfers).
— The driver returns ownership of the packet to the Network Operating System (NOS) when
the hardware has completed the DMA transfer of the frame (indicated by an interrupt).
Output packets are made up of pointer–length pairs constituting a descriptor chain (so called
descriptor based transmission). Software forms transmit packets by assembling the list of pointer–
length pairs, storing this information in the transmit descriptor, and then updating the on–chip
transmit tail pointer to the descriptor. The transmit descriptor and buffers are stored in host
memory. Hardware typically transmits the packet only after it has completely fetched all packet
data from host memory and deposited it into the on-chip transmit FIFO. This permits TCP or UDP
checksum computation, and avoids problems with PCI underruns.

3.3.1

Transmit Data Storage
Data are stored in buffers pointed to by the descriptors. Alignment of data is on an arbitrary byte
boundary with the maximum size per descriptor limited only to the maximum allowed packet size
(16288 bytes). A packet typically consists of two (or more) descriptors, one (or more) for the
header and one for the actual data. Some software implementations copy the header(s) and packet
data into one buffer and use only one descriptor per transmitted packet.

3.3.2

Transmit Descriptors
The Ethernet controller provides three types of transmit descriptor formats.
The original descriptor is referred to as the “legacy” descriptor format. The two other descriptor
types are collectively referred to as extended descriptors. One of them is similar to the legacy
descriptor in that it points to a block of packet data. This descriptor type is called the TCP/IP Data
Descriptor and is a replacement for the legacy descriptor since it offers access to new offloading
capabilities. The other descriptor type is fundamentally different as it does not point to packet data.
It merely contains control information which is loaded into registers of the controller and affect the
processing of future packets. The following sections describe the three descriptor formats.
The extended descriptor types are accessed by setting the TDESC.DEXT bit to 1b. If this bit is set,
the TDESC.DTYP field is examined to control the interpretation of the remaining bits of the
descriptor. Table 3-7 shows the generic layout for all extended descriptors. Fields marked as NR
are not reserved for any particular function and are defined on a per-descriptor type basis. Notice
that the DEXT and DTYP fields are non-contiguous in order to accommodate legacy mode
operation. For legacy mode operation, bit 29 is set to 0b and the descriptor is defined in Section
3.3.3.

Software Developer’s Manual

Receive and Transmit Description

Table 3-7. Transmit Descriptor (TDESC) Layout

24 23

20 19

Buffer Address [63:0]

3.3.3

DEXT

DTYP

Legacy Transmit Descriptor Format
To select legacy mode operation, bit 29 (TDESC.DEXT) should be set to 0b. In this case, the
descriptor format is defined as shown in Table 3-8. The address and length must be supplied by
software. Bits in the command byte are optional, as are the Checksum Offset (CSO), and
Checksum Start (CSS) fields.
Table 3-8. Transmit Descriptor (TDESC) Layout – Legacy Mode

48 47

40 39 36 35 32 31

0
8

24 23

16 15

Buffer Address [63:0]
Special

CSS

RSV

STA

CMD

CSO

Length

Table 3-9. Transmit Descriptor Legacy Descriptions

Transmit Descriptor
Legacy

Description

Buffer Address

Buffer Address
Address of the transmit descriptor in the host memory. Descriptors with a
null address transfer no data. If they have the RS bit in the command byte
set (TDESC.CMD), then the DD field in the status word (TDESC.STATUS) is
written when the hardware processes them.

Length

Length is per segment.
The maximum length associated with any single legacy descriptor is 16288
bytes. Although a buffer as short as one byte is allowed, the total length of
the packet, before padding and CRC insertion must be at least 48 bytes.
Length can be up to a default value of 16288 bytes per descriptor, and
16288 bytes total. In other words, the length of the buffer pointed to by one
descriptor, or the sum of the lengths of the buffers pointed to by the
descriptors can be as large as the maximum allowed transmit packet.
Descriptors with zero length transfer no data. If they have the RS bit in the
command byte set (TDESC.CMD), then the DD field in the status word
(TDESC.STATUS) is written when the hardware processes them.

CSO

Checksum Offset
The Checksum offset field indicates where, relative to the start of the packet,
to insert a TCP checksum if this mode is enabled. (Insert Checksum bit (IC)
is set in TDESC.CMD). Hardware ignores CSO unless EOP is set in
TDESC.CMD. CSO is provided in unit of bytes and must be in the range of
the data provided to the Ethernet controller in the descriptor. (CSO < length 1).
Should be written with 0b for future compatibility.

Software Developer’s Manual

Receive and Transmit Description

Transmit Descriptor
Legacy

Description

CMD

Command field
See Section 3.3.3.1 for a detailed field description.

STA

Status field
See Section 3.3.3.2 for a detailed field description.

RSV

Reserved
Should be written with 0b for future compatibility.

CSS

Checksum Start Field
The Checksum start field (TDESC.CSS) indicates where to begin computing
the checksum. The software must compute this offset to back out the bytes
that should not be included in the TCP checksum. CSS is provided in units
of bytes and must be in the range of data provided to the Ethernet controller
in the descriptor (CSS < length). For short packets that ar padded by the
software, CSS must be in the range of the unpadded data length. A value of
0b corresponds to the first byte in the packet.
CSS must be set in the first descriptor of the packet.

Special

Special Field
See the notes that follow this table for a detailed field description.

Notes:
1.
2.

Even though CSO and CSS are in units of bytes, the checksum calculation typically works on
16-bit words. Hardware does not enforce even byte alignment.
Hardware does not add the 802.1Q EtherType or the VLAN field following the 802.1Q
EtherType to the checksum. So for VLAN packets, software can compute the values to back
out only on the encapsulated packet rather than on the added fields.
Although the Ethernet controller can be programmed to calculate and insert TCP checksum
using the legacy descriptor format as described above, it is recommended that software use the
newer TCP/IP Context Transmit Descriptor Format. This newer descriptor format allows the
hardware to calculate both the IP and TCP checksums for outgoing packets. See Section 3.3.5
for more information about how the new descriptor format can be used to accomplish this task.

Software Developer’s Manual

Receive and Transmit Description

3.3.3.1

Transmit Descriptor Command Field Format
The CMD byte stores the applicable command and has fields shown in Table 3-10.
Table 3-10. Transmit Command (TDESC.CMD) Layout
7

IDE

VLE

DEXT

RSV
RPSa

IFCS

EOP

a. 82544GC/EI only.

TDESC.CMD

Description

IDE (bit 7)

Interrupt Delay Enable
When set, activates the transmit interrupt delay timer. The Ethernet controller loads
a countdown register when it writes back a transmit descriptor that has RS and IDE
set. The value loaded comes from the IDV field of the Interrupt Delay (TIDV)
register. When the count reaches 0, a transmit interrupt occurs if transmit descriptor
write-back interrupts (IMS.TXDW) are enabled. Hardware always loads the transmit
interrupt counter whenever it processes a descriptor with IDE set even if it is
already counting down due to a previous descriptor. If hardware encounters a
descriptor that has RS set, but not IDE, it generates an interrupt immediately after
writing back the descriptor. The interrupt delay timer is cleared.

VLE (bit 6)

VLAN Packet Enable
When set, indicates that the packet is a VLAN packet and the Ethernet controller
should add the VLAN Ethertype and an 802.1q VLAN tag to the packet. The
Ethertype field comes from the VET register and the VLAN tag comes from the
special field of the TX descriptor. The hardware inserts the FCS/CRC field in that
case.
When cleared, the Ethernet controller sends a generic Ethernet packet. The IFCS
controls the insertion of the FCS field in that case.
In order to have this capability CTRL.VME bit should also be set, otherwise VLE
capability is ignored. VLE is valid only when EOP is set.

DEXT (bit 5)

Extension (0b for legacy mode).
Should be written with 0b for future compatibility.

RPS
RSV (bit 4)

Report Packet Sent
When set, the 82544GC/EI defers writing the DD bit in the status byte
(DESC.STATUS) until the packet has been sent, or transmission results in an error
such as excessive collisions. It is used is cases where the software must know that
the packet has been sent, and not just loaded to the transmit FIFO. The 82544GC/
EI might continue to prefetch data from descriptors logically after the one with RPS
set, but does not advance the descriptor head pointer or write back any other
descriptor until it sent the packet with the RPS set. RPS is valid only when EOP is
set.
This bit is reserved and should be programmed to 0b for all Ethernet controllers
except the 82544GC/EI.

RS (bit 3)

Report Status
When set, the Ethernet controller needs to report the status information. This ability
may be used by software that does in-memory checks of the transmit descriptors to
determine which ones are done and packets have been buffered in the transmit
FIFO. Software does it by looking at the descriptor status byte and checking the
Descriptor Done (DD) bit.

Software Developer’s Manual

Receive and Transmit Description

TDESC.CMD

Description

IC (bit 2)

Insert Checksum
When set, the Ethernet controller needs to insert a checksum at the offset indicated
by the CSO field. The checksum calculations are performed for the entire packet
starting at the byte indicated by the CCS field. IC is ignored if CSO and CCS are out
of the packet range. This occurs when (CSS  length) OR (CSO  length - 1). IC is
valid only when EOP is set.

IFCS (bit 1)

Insert FCS
Controls the insertion of the FCS/CRC field in normal Ethernet packets. IFCS is
valid only when EOP is set.

EOP (bit 0)

End Of Packet
When set, indicates the last descriptor making up the packet. One or many
descriptors can be used to form a packet.

Notes:
1.
2.
3.
4.

3.3.3.2

VLE, IFCS, and IC are qualified by EOP. That is, hardware interprets these bits ONLY when
EOP is set.
Hardware only sets the DD bit for descriptors with RS set.
Descriptors with the null address (0b) or zero length transfer no data. If they have the RS bit
set then the DD field in the status word is written when hardware processes them.
Although the transmit interrupt may be delayed, the descriptor write-back requested by setting
the RS bit is performed without delay unless descriptor write-back bursting is enabled.

Transmit Descriptor Status Field Format
The STATUS field stores the applicable transmit descriptor status and has the fields shown in Table
3-11.
The transmit descriptor status field is only present in cases where RS (or RPS for the 82544GC/EI
only) is set in the command field.
Table 3-11. Transmit Status Layout
3

RSV
TUa

a. 82544GC/EI only.

Software Developer’s Manual

Receive and Transmit Description

TDESC.STATUS

Note:

3.3.4

Description

TU
RSV (bit 3)

Transmit Underrun
Indicates a transmit underrun event occurred. Transmit Underrun might occur if Early
Transmits are enabled (based on ETT.Txthreshold value) and the 82544GC/EI was
not able to complete the early transmission of the packet due to lack of data in the
packet buffer. This does not necessarily mean the packet failed to be eventually
transmitted. The packet is successfully re-transmitted if the TCTL.NRTU bit is
cleared (and excessive collisions do not occur).
This bit is reserved and should be programmed to 0b for all Ethernet controllers
except the 82544GC/EI.

LC (bit 2)

Late Collision
Indicates that late collision occurred while working in half-duplex mode. It has no
meaning while working in full-duplex mode. Note that the collision window is speed
dependent: 64 bytes for 10/100 Mb/s and 512 bytes for 1000 Mb/s operation.

EC (bit 1)

Excess Collisions
Indicates that the packet has experienced more than the maximum excessive
collisions as defined by TCTL.CT control field and was not transmitted. It has no
meaning while working in full-duplex mode.

DD (bit 0)

Descriptor Done
Indicates that the descriptor is finished and is written back either after the descriptor
has been processed (with RS set) or for the 82544GC/EI, after the packet has been
transmitted on the wire (with RPS set).

The DD bit reflects status of all descriptors up to and including the one with the RS bit set (or RPS
for the 82544GC/EI).

Transmit Descriptor Special Field Format
The SPECIAL field is used to provide the 802.1q/802.1ac tagging information.
When CTRL.VME is set to 1b, all packets transmitted from the Ethernet controller that have VLE
set in the TDESC.CMD are sent with an 802.1Q header added to the packet. The contents of the
header come from the transmit descriptor special field and from the VLAN type register. The
special field is ignored if the VLE bit in the transmit descriptor command field is 0b. The special
field is valid only for descriptors with EOP set to 1b in TDESC.CMD.
Table 3-12. Special Field (TDESC.SPECIAL) Layout
15

13
PRI

12
CFI

0
VLAN

TDESC.SPECIAL

Description

PRI

User Priority
3 bits that provide the VLAN user priority field to be inserted in the 802.1Q tag.

CFI

Canonical Form Indicator.

VLAN

VLAN Identifier
12 bits that provide the VLAN identifier field to be inserted in the 802.1Q tag.

Software Developer’s Manual

Receive and Transmit Description

3.3.5

TCP/IP Context Transmit Descriptor Format
The TCP/IP context transmit descriptor provides access to the enhanced checksum offload facility
available in the Ethernet controller. This feature allows TCP and UDP packet types to be handled
more efficiently by performing additional work in hardware, thus reducing the software overhead
associated with preparing these packets for transmission.
The TCP/IP context transmit descriptor does not point to packet data as a data descriptor does.
Instead, this descriptor provides access to an on-chip context that supports the transmit checksum
offloading feature of the controller. A “context” refers to a set of registers loaded or unloaded as a
group to provide a particular function.
The context is explicit and directly accessible via the TCP/IP context transmit descriptor. The
context is used to control the checksum offloading feature for normal packet transmission.
The Ethernet controller automatically selects the appropriate legacy or normal context to use based
on the current packet transmission.
While the architecture supports arbitrary ordering rules for the various descriptors, there are
restrictions including:

• Context descriptors should not occur in the middle of a packet.
• Data descriptors of different packet types (legacy or normal) should not be intermingled
except at the packet level.
All contexts control calculation and insertion of up to two checksums. This portion of the context is
referred to as the checksum context.
In addition to checksum context, the segmentation context adds information specific to the
segmentation capability. This additional information includes the total payload for the message
(TDESC.PAYLEN), the total size of the header (TDESC.HDRLEN), the amount of payload data
that should be included in each packet (TDESC.MSS), and information about what type of protocol
(TCP, IPv4, IPv6, etc.) is used. This information is specific to the segmentation capability and is
therefore ignored for context descriptors that do not have the TSE bit set.
Because there are dedicated resources on-chip for the normal context, the context remains constant
until it is modified by another context descriptor. This means that a context can be used for multiple
packets (or multiple segmentation blocks) unless a new context is loaded prior to each new packet.
Depending on the environment, it may be completely unnecessary to load a new context for each
packet. For example, if most traffic generated from a given node is standard TCP frames, this
context could be set up once and used for many frames. Only when some other frame type is
required would a new context need to be loaded by software. After the “non-standard” frame is
transmitted, the “standard” context would be setup once more by software. This method avoids the
“extra descriptor per packet” penalty for most frames. The penalty can be eliminated altogether if
software elects to use TCP/IP checksum offloading only for a single frame type, and thus performs
those operations in software for other frame types.
This same logic can also be applied to the segmentation context, though the environment is a more
restrictive one. In this scenario, the host is commonly asked to send a message of the same type,
TCP/IP for instance, and these messages also have the same total length and same maximum
segment size (MSS). In this instance, the same segmentation context could be used for multiple
TCP messages that require hardware segmentation. The limitations of this scenario and the
relatively small performance advantage make this approach unlikely; however, it is useful in
understanding the underlying mechanism.

Software Developer’s Manual

Receive and Transmit Description

3.3.6

TCP/IP Context Descriptor Layout
The following section describes the layout of the TCP/IP context transmit descriptor.
To select this descriptor format, bit 29 (TDESC.DEXT) must be set to 1b and TDESC.DTYP must
be set to 0000b. In this case, the descriptor format is defined as shown in Table 3-13.
Note that the TCP/IP context descriptor does not transfer any packet data. It merely prepares the
checksum hardware for the TCP/IP Data descriptors that follow.

Table 3-13. Transmit Descriptor (TDESC) Layout – (Type = 0000b)
63
0

TUCSE

MSS
63

Note:

48 47

40 39

TUCSO

32 31
TUCSS

HDRLEN RSV
48 47

STA

16 15
IPCSE

TUCMD

40 39 36 35 32 31

DTYP

24 23 20 19

IPCSO

0
IPCSS

PAYLEN
0

The first quadword of this descriptor type contains parameters used to calculate the two checksums
which may be offloaded.

Software Developer’s Manual

Receive and Transmit Description

Table 3-14. Transmit Descriptor (TDESC) Layout
Transmit
Descriptor Offload

Description

TUCSE

TCP/UDP Checksum Ending
Defines the ending byte for the TCP/UDP checksum offload feature.
Setting TUCSE field to 0b indicates that the checksum covers from TUCCS to the
end of the packet.

TUCSO

TCP/UDP Checksum Offset
Defines the offset where to insert the TCP/UDP checksum field in the packet data
buffer. This is used in situations where the software needs to calculate partial
checksums (TCP pseudo-header, for example) to include bytes which are not
contained within the range of start and end.
If no partial checksum is required, software must write a value of 0b.

TUCSS

TCP/UDP Checksum Start
Defines the starting byte for the TCP/UDP checksum offload feature.
It must be defined even if checksum insertion is not desired for some reason.
When setting the TCP segmentation context, TUCSS is used to indicate the start
of the TCP header.

IPCSE

IP Checksum Ending
Defines the ending byte for the IP checksum offload feature.
It specifies where the checksum should stop. A 16-bit value supports checksum
offloading of packets as large as 64KB.
Setting IPCSE field to 0b indicates that the checksum covers from IPCCS to the
end of the packet. In this way, the length of the packet does not need to be
calculated.

IPCSO

IP Checksum Offset
The IPCSO field specifies where the resulting IP checksum should be placed. It is
limited to the first 256 bytes of the packet and must be less than or equal to the
total length of a given packet. If this is not the case, the checksum is not inserted.

IPCSS

IP Checksum Start
IPCSS specifies the byte offset from the start of the transferred data to the first
byte in be included in the checksum. Setting this value to 0b means the first byte of
the data would be included in the checksum.
Note that the maximum value for this field is 255. This is adequate for typical
applications.
The IPCSS value needs to be less than the total transferred length of the packet. If
this is not the case, the results are unpredictable.
IPCSS must be defined even if checksum insertion is not desired for some reason.
When setting the TCP segmentation context, IPCSS is used to indicate the start of
the IP header.

MSS

Maximum Segment Size
Controls the Maximum Segment Size. This specifies the maximum TCP or UDP
payload “segment” sent per frame, not including any header. The total length of
each frame (or “section”) sent by the TCP Segmentation mechanism (excluding
802.3ac tagging and Ethernet CRC) is MSS bytes + HRDLEN. The one exception
is the last packet of a TCP segmentation context which is (typically) shorter than
“MSS+HDRLEN”. This field is ignored if TDESC.TSE is not set.

HDRLEN

Header Length
Specifies the length (in bytes) of the header to be used for each frame (or
“section”) of a TCP Segmentation operation. The first HDRLEN bytes fetched from
data descriptor(s) are stored internally and used as a prototype header for each
section, and are pre-pended to each payload segment to form individual frames.
For UDP packets this is normally equal to “UDP checksum offset + 2”. For TCP
packets it is normally equal to “TCP checksum offset + 4 + TCP header option
bytes”. This field is ignored if TDESC.TSE is not set.

Software Developer’s Manual

Receive and Transmit Description

Transmit
Descriptor Offload

Description

RSV

Reserved
Should be programmed to 0b for future compatibility.

STA

TCP/UDP Status field
Provides transmit status indication.
Section 3.3.6.2 provides the bit definition for the TDESC.STA field.

TUCMD

TCP/UDP command field
The command field provides options that control the checksum offloading, along
with some of the generic descriptor processing functions.
Section 3.3.6.1 provides the bit definitions for the TDESC.TUCMD field.

DTYP

Descriptor Type
Set to 0000b for TCP/IP context transmit descriptor type.

PAYLEN

The packet length field (TDESC.PAYLEN) is the total number of payload bytes for
this TCP Segmentation offload context (i.e., the total number of payload bytes that
could be distributed across multiply frames after TCP segmentation is performed).
Following the fetch of the prototype header, PAYLEN specifies the length of data
that is fetched next from data descriptor(s). This field is also used to determine
when “last-frame” processing needs to be performed. Typically, a new data
descriptor is used to denote the start of the payload data buffer(s), but this is not
required. PAYLEN specification should not include any header bytes. There is no
restriction on the overall PAYLEN specification with respect to the transmit FIFO
size, once the MSS and HDRLEN specifications are legal. This field is ignored if
TDESC.TSE is not set. Refer to Section 3.5 for details on the TCP Segmentation
off-loading feature.

Notes:
1.
2.

3.3.6.1

A number of the fields are ignored if the TCP Segmentation enable bit (TDESC.TSE) is
cleared, denoting that the descriptor does not refer to the TCP segmentation context.
Maximum limits for the HDRLEN and MSS fields are dictated by the lengths variables. However, there is a further restriction that for any TCP Segmentation operation, the hardware must
be capable of storing a complete section (completely-built frame) in the transmit FIFO prior to
transmission. Therefore, the sum of MSS + HDRLEN must be at least 80 bytes less than the
allocated size of the transmit FIFO.

TCP/UDP Offload Transmit Descriptor Command Field
The command field (TDESC.TUCMD) provides options to control the TCP segmentation, along
with some of the generic descriptor processing functions.

Software Developer’s Manual

Receive and Transmit Description

Table 3-15. Command Field (TDESC.TUCMD) Layout
7

IDE

RSV

DEXT

RSV

TSE

TCP

TDESC.TUCMD

Description

IDE (bit 7)

Interrupt Delay Enable
IDE activates the transmit interrupt delay timer. Hardware loads a countdown
register when it writes back a transmit descriptor that has the RS bit and the IDE bit
set. The value loaded comes from the IDV field of the Interrupt Delay (TIDV) register.
When the count reaches 0, a transmit interrupt occurs. Hardware always loads the
transmit interrupt counter whenever it processes a descriptor with IDE set even if it is
already counting down due to a previous descriptor. If hardware encounters a
descriptor that has RS set, but not IDE, it generates an interrupt immediately after
writing back the descriptor. The interrupt delay timer is cleared.

RSV (Bit 6)

Reserved. Set to 0b for future compatibility.

DEXT(Bit 5)

Descriptor Extension
Must be 1b for this descriptor type.

RSV (Bit 4)

Reserved. Set to 0b for future compatibility.

RS (Bit 3)

Report Status
RS tells the hardware to report the status information for this descriptor. Because this
descriptor does not transmit data, only the DD bit in the status word is valid. Refer to
Section 3.3.6.2 for the layout of the status field.

TSE (Bit 2)

TCP Segmentation Enable
TSE indicates that this descriptor is setting the TCP segmentation context. If this bit
is not set, the checksum offloading context for normal (non-”TCP Segmentation”)
packets is written. When a descriptor of this type is processed the Ethernet controller
immediately updates the context in question (TCP Segmentation or checksum
offloading) with values from the descriptor. This means that if any normal packets or
TCP Segmentation packets are in progress (a descriptor with EOP set has not been
received for the given context), the results are likely to be undesirable.

IP (Bit 1)

Packet Type (IPv4 = 1b, IPv6 = 0b)
Identifies what type of IP packet is used in the segmentation process. This is
necessary for hardware to know where the IP Payload Length field is located. This
does not override the checksum insertion bit, IXSM.

IP (Bit 1)
82544GC/EI only

Packet Type (IP = 1b)
Identifies the packet as an IP packet. The purpose of this bit is to enable/disable the
updating of the IP header during the segmentation process. This does not override
the checksum insertion bit, IXSM.

TCP (bit 0)

Packet Type (TCP = 1b)
Identifies the packet as either TCP or UDP (non-TCP). This affects the processing of
the header information.

Note:
1.
2.

The IDE, DEXT, and RS bits are valid regardless of the state of TSE. All other bits are ignored
if TSE = 0b.
The TCP Segmentation feature also provides access to a generic block send function and may
be useful for performing “segmentation offload” in which the header information is constant.
By clearing both the TCP and IP bits, a block of data may be broken down into frames of a
given size, a constant, arbitrary length header may be pre-pended to each frame, and two
checksums optionally added.

Software Developer’s Manual

Receive and Transmit Description

3.3.6.2

TCP/UDP Offload Transmit Descriptor Status Field
Four bits are reserved to provide transmit status, although only one is currently assigned for this
specific descriptor type. The status word is only written back to host memory in cases where the RS
is set in the command.
Table 3-16. Transmit Status Layout
3

RSV

TDESC.STA

3.3.7

0
DD

Description

RSV

Reserved
Reserved for future use. Reads as 0b.

DD (bit 0)

Descriptor Done
Indicates that the descriptor is finished and is written back after the descriptor has
been processed.

TCP/IP Data Descriptor Format
The TCP/IP data descriptor is the companion to the TCP/IP context transmit descriptor described
in the previous section. This descriptor type provides similar functionality to the legacy mode
descriptor but also integrates the checksum offloading and TCP Segmentation feature.
To select this descriptor format, bit 29 in the command field (TDESC.DEXT) must be set to 1b and
TDESC.DTYP must be set to 0001b. In this case, the descriptor format is defined as shown in
Table 3-17.

Software Developer’s Manual

Receive and Transmit Description

Table 3-17. Transmit Descriptor (TDESC) Layout – (Type = 0001b)
0

Address [63:0]

8
0

Special
63

POPTS
48 47

RSV

STA

DCMD

40 39 36 35 32 31

Transmit Descriptor

DTYP

DTALEN

24 23 20 19

Description

Address

Data buffer address
Address of the data buffer in the host memory which contains a portion of the
transmit packet.

DTALEN

Data Length Field
Total length of the data pointed to by this descriptor, in bytes.
For data descriptors not associated with a TCP Segmentation operation
(TDESC.TSE not set), the descriptor lengths are subject to the same restrictions
specified for legacy descriptors (the sum of the lengths of the data descriptors
comprising a single packet must be at least 80 bytes less than the allocated size
of the transmit FIFO.)

DTYP

Data Type
Set to 0001b to identify this descriptor as a TCP/IP data descriptor.

DCMD

Descriptor Command Field
Provides options that control some of the generic descriptor processing
features. Refer to Section 3.3.7.1 for bit definitions of the DCMD field.

STA

TCP/IP Status field
Provides transmit status indication.
Section 3.3.7.2 provides the bit definition for the TDESC.STA field.

RSV

Reserved
Set to 0b for future compatibility.

POPTS

Packet Option Field
Provides a number of options which control the handling of this packet. This field
is ignored except on the first data descriptor of a packet.
Section 3.3.7.3 provides the bit definition for the TDESC.POPTS field.

Special

Special field
The Special field is used to provide 802.1q tagging information.
This field is only valid in the last descriptor of the given packet (qualified by the
EOP bit).

Software Developer’s Manual

Receive and Transmit Description

3.3.7.1

TCP/IP Data Descriptor Command Field
The Command field provides options that control checksum offloading and TCP segmentation
features along with some of the generic descriptor processing features.
Table 3-18. Command Field (TDESC.DCMD) Layout
7

IDE

VLE

DEXT

RSV
RPSa

TSE

IFCS

EOP

a. 82544GC/EI only.

TDESC.DCMD

Description

IDE (bit 7)

Interrupt Delay Enable
When set, activates the transmit interrupt delay timer. Hardware loads a countdown
register when it writes back a transmit descriptor that has RS and IDE set. The value
loaded comes from the IDV field of the Interrupt Delay (TIDV) register. When the
count reaches 0, a transmit interrupt occurs if enabled. Hardware always loads the
transmit interrupt counter whenever it processes a descriptor with IDE set even if it is
already counting down due to a previous descriptor. If hardware encounters a
descriptor that has RS set, but not IDE, it generates an interrupt immediately after
writing back the descriptor. The interrupt delay timer is cleared.

VLE (bit 6)

VLAN Enable
When set, indicates that the packet is a VLAN packet and the hardware should add
the VLAN Ethertype and an 802.1q VLAN tag to the packet. The Ethertype should
come from the VET register and the VLAN data comes from the special field of the TX
descriptor. The hardware in that case appends the FCS/CRC.
Note that the CTRL.VME bit should also be set. If the CTRL.VME bit is not set, the
Ethernet controller does not insert VLAN tags on outgoing packets and it sends
generic Ethernet packets. The IFCS controls the insertion of the FCS/CRC in that
case.
VLE is only valid in the last descriptor of the given packet (qualified by the EOP bit).

DEXT (Bit 5)

Descriptor Extension
Must be 1b for this descriptor type

RPS
RSV (bit 4)

Report Packet Sent
RPS is used in cases where software must know that a packet has been sent on the
wire, not just that it has been loaded into the 82544GC/EI controller’s internal packet
buffer.
When set, hardware defers writing the DD bit in the status byte until the packet has
been sent, or transmission results in an error such as excess collisions. Hardware
can continue to pre-fetch data from descriptors logically after the one with RPS set,
but does not advance the head pointer or write back any other descriptors until it has
sent the packet with RPS set.
For a TCP Segmentation context, the RPS bit indicates to the 82544GC/EI that the
descriptor status should only be written back once all packets that make up the given
TCP Segmentation context had been sent.
This bit is reserved and should be programmed to 0b for all Ethernet controllers
except the 82544GC/EI.

RS (bit 3)

Report Status
When set, tells the hardware to report the status information for this descriptor as
soon as the corresponding data buffer has been fetched and stored in the controller’s
internal packet buffer.

Software Developer’s Manual

Receive and Transmit Description

TDESC.DCMD

Description

TSE (bit 2)

TCP Segmentation Enable
TSE indicates that this descriptor is part of the current TCP Segmentation context. If
this bit is not set, the descriptor is part of the “normal” context.

IFCS (Bit 1)

Insert IFCS
Controls the insertion of the FCS/CRC field in normal Ethernet packets.
IFCS is only valid in the last descriptor of the given packet (qualified by the EOP bit).

EOP (Bit 0)

End Of Packet
The EOP bit indicates that the buffer associated with this descriptor contains the last
data for the packet or for the given TCP Segmentation context. In the case of a TCP
Segmentation context, the DTALEN length of this descriptor should match the
amount remaining of the original PAYLEN. If it does not, the TCP Segmentation
context is terminated but the end of packet processing may be incorrectly performed.
These abnormal termination events are counted in the TSCTFC statistics register.

Note: The VLE, IFCS, and VLAN fields are only valid in certain descriptors. If TSE is enabled, the VLE,
IFCS, and VLAN fields are only valid in the first data descriptor of the TCP segmentation context.
If TSE is not enabled, then these fields are only valid in the last descriptor of the given packet
(qualified by the EOP bit).

3.3.7.2

TCP/IP Data Descriptor Status Field
Four bits are reserved to provide transmit status, although only the DD is valid1. The status word is
only written back to host memory in cases where the RS bit is set in the command field. The DD bit
indicates that the descriptor is finished and is written back after the descriptor has been processed.
Table 3-19. Transmit Status Layout
3

RSV
TUa

a. 82544GC/EI only.

TDESC.STA
Reserved

Description
Reserved

1. Unless the RPS bit is set in the descriptor (82544GC/EI only).

Software Developer’s Manual

Receive and Transmit Description

TDESC.STA

3.3.7.3

Description

LC (bit2)

Late Collision
Indicates that late collision occurred while working in half-duplex mode.
It has no meaning while working in full-duplex mode.
Note that the collision window is speed dependent: 64 bytes for 10/100 Mb/s and
512 bytes for 1000 Mb/s operation.

EC (bit 1)

Excess Collision
Indicates that the packet has experienced more than the maximum excessive
collisions as defined by TCTL.CT control field and was not transmitted.
Is has no meaning while working in full-duplex mode.

DD (bit 0)

Descriptor Done
Indicates that the descriptor is done and is written back either after the descriptor
has been processed (with RS set), or for the 82554GC/EI only, after the packet has
been transmitted on the wire (with RPS set).

TCP/IP Data Descriptor Option Field
The POPTS field provides a number of options which control the handling of this packet. This field
is ignored except on the first data descriptor of a packet.
Table 3-20. Packet Options Field (TDESC.POPTS) Layout
7

RSV

TXSM

IXSM

TDESC.POPTS

3.3.7.4

Description

RSV (bit 2-7)

Reserved
Should be written with 0b for future compatibility.

TXSM (bit1)

Insert TCP/UDP Checksum
Controls the insertion of the TCP/UDP checksum.
If not set, the value placed into the checksum field of the packet data is not modified,
and is placed on the wire. When set, TCP/UDP checksum field is modified by the
hardware.
Valid only in the first data descriptor for a given packet or TCP Segmentation context.

IXSM (bit 0)

Insert IP Checksum
Controls the insertion of the IP checksum.
If not set, the value placed into the checksum field of the packet data is not modified
and is placed on the wire. When set, the IP checksum field is modified by the
hardware.
Valid only in the first data descriptor for a given packet or TCP Segmentation context.

TCP/IP Data Descriptor Special Field
The SPECIAL field is used to provide the 802.1q/802.3ac tagging information.

Software Developer’s Manual

Receive and Transmit Description

When CTRL.VME is set to 1b, all packets transmitted from the Ethernet controller that has VLE
set in the DCMD field is sent with an 802.1Q header added to the packet. The contents of the
header come from the transmit descriptor special field and from the VLAN type register. The
special field is ignored if the VLE bit in the transmit descriptor command field is 0b. The special
field is valid only when EOP is set.
Table 3-21. Special Field (TDESC.SPECIAL) Layout
15

13
PRI

12
CFI

0
VLAN

TDESC.SPECIAL
PRI

3.4

Description
User Priority
Three bits that provide the VLAN user priority field to be inserted in the 802.1Q tag.

CFI

Canonical Form Indicator

VLAN

VLAN Identifier
12 bits that provide the VLAN identifier field to be inserted in the 802.1Q tag.

Transmit Descriptor Ring Structure
The transmit descriptor ring structure is shown in Figure 3-4. A pair of hardware registers
maintains the transmit queue. New descriptors are added to the ring by writing descriptors into the
circular buffer memory region and moving the ring’s tail pointer. The tail pointer points one entry
beyond the last hardware owned descriptor (but at a point still within the descriptor ring).
Transmission continues up to the descriptor where head equals tail at which point the queue is
empty.
Descriptors passed to hardware should not be manipulated by software until the head pointer has
advanced past them.

Software Developer’s Manual

Receive and Transmit Description

Circular Buffer
Base
Head

Owned By
Hardware

Transmit
Queue

Tail
Base + Size

Figure 3-4. Transmit Descriptor Ring Structure
Shaded boxes in Figure 3-4 represent descriptors that have been transmitted but not yet reclaimed
by software. Reclaiming involves freeing up buffers associated with the descriptors.
The transmit descriptor ring is described by the following registers:

• Transmit Descriptor Base Address registers (TDBAL and TDBAH)
These registers indicate the start of the descriptor ring buffer. This 64-bit address is aligned on
a 16-byte boundary and is stored in two consecutive 32-bit registers. TDBAL contains the
lower 32-bits; TDBAH contains the upper 32 bits. Hardware ignores the lower 4 bits in
TDBAL.

• Transmit Descriptor Length register (TDLEN)
This register determines the number of bytes allocated to the circular buffer. This value must
be 128 byte aligned.

• Transmit Descriptor Head register (TDH)
This register holds a value which is an offset from the base, and indicates the in–progress
descriptor. There can be up to 64K descriptors in the circular buffer. Reading this register
returns the value of “head” corresponding to descriptors already loaded in the output FIFO.

• Transmit Descriptor Tail register (TDT)
This register holds a value which is an offset from the base, and indicates the location beyond
the last descriptor hardware can process. This is the location where software writes the first
new descriptor.
The base register indicates the start of the circular descriptor queue and the length register indicates
the maximum size of the descriptor ring. The lower seven bits of length are hard–wired to 0b. Byte
addresses within the descriptor buffer are computed as follows:
address = base + (ptr * 16), where ptr is the value in the hardware head or tail register.
The size chosen for the head and tail registers permit a maximum of 64 K descriptors, or
approximately 16 K packets for the transmit queue given an average of four descriptors per packet.

Software Developer’s Manual

Receive and Transmit Description

Once activated, hardware fetches the descriptor indicated by the hardware head register. The
hardware tail register points one beyond the last valid descriptor.
Software can determine if a packet has been sent by setting the RS bit (or the RPS bit for the
82544GC/EI only) in the transmit descriptor command field. Checking the transmit descriptor DD
bit in memory eliminates a potential race condition. All descriptor data is written to the IO bus
prior to incrementing the head register, but a read of the head register could “pass” the data write in
systems performing IO write buffering. Updates to transmit descriptors use the same IO write path
and follow all data writes. Consequently, they are not subject to the race condition. Other potential
conditions also prohibit software reading the head pointer.
In general, hardware prefetches packet data prior to transmission. Hardware typically updates the
value of the head pointer after storing data in the transmit FIFO1.
The process of checking for completed packets consists of one of the following:

• Scan memory for descriptor status write-backs.
• Take an interrupt. An interrupt condition can be generated whenever a transmit queue goes
empty (ICR.TXQE). Interrupts can also be triggered in other ways.

3.4.1

Transmit Descriptor Fetching
The descriptor processing strategy for transmit descriptors is essentially the same as for receive
descriptors except that a different set of thresholds are used. As for receives, the number of on-chip
transmit descriptors buffer space is 64 descriptors.
When the on-chip buffer is empty, a fetch happens as soon as any descriptors are made available
(software writes to the tail pointer). When the on-chip buffer is nearly empty
(TXDCTL.PTHRESH), a prefetch is performed whenever enough valid descriptors
(TXDCTL.HTHRESH) are available in host memory and no other DMA activity of greater priority
is pending (descriptor fetches and write-backs or packet data transfers).
The descriptor prefetch policy is aggressive to maximize performance. If descriptors reside in an
external cache, the system must ensure cache coherency before changing the tail pointer.
When the number of descriptors in host memory is greater than the available on-chip descriptor
storage, the chip may elect to perform a fetch which is not a multiple of cache line size. The
hardware performs this non-aligned fetch if doing so results in the next descriptor fetch being
aligned on a cache line boundary. This allows the descriptor fetch mechanism to be most efficient
in the cases where it has fallen behind software.

3.4.2

Transmit Descriptor Write-back
The descriptor write-back policy for transmit descriptors is similar to that for receive descriptors
with a few additional factors. First, since transmit descriptor write-backs are optional (controlled
by RS2 in the transmit descriptor), only descriptors which have one (or both) of these bits set starts
the accumulation of write-back descriptors. Secondly, to preserve backward compatibility with the
82542, if the TXDCTL.WTHRESH value is 0b, the Ethernet controller writes back a single byte of
the descriptor (TDESCR.STA) and all other bytes of the descriptor are left unchanged.

1. With the RPS bit set, the head is not advanced until after the packet is transmitted or rejected due to excess collisions (82544GC/EI only).
2. And RPS for the 82544GC/EI only.

Software Developer’s Manual

Receive and Transmit Description

Since the benefit of delaying and then bursting transmit descriptor write-backs is small at best, it is
likely that the threshold are left at the default value (0b) to force immediate write-back of transmit
descriptors and to preserve backward compatibility.
Descriptors are written back in one of three conditions:

• TXDCTL.WTHRESH = 0b and a descriptor which has RS1 set is ready to be written back
• Transmit Interrupt Delay timer expires
• TXDCTL.WTHRESH > 0b and TXDCTL.WTHRESH descriptors have accumulated
For the first condition, write-backs are immediate. This is the default operation and is backward
compatible. For this case, the Transmit Interrupt delay function works as described in Section
3.4.3.1.
The other two conditions are only valid if descriptor bursting is enabled (see Section 13.4.44). In
the second condition, the Transmit Interrupt Delay timer (TIDV) is used to force timely write–back
of descriptors. The first packet after timer initialization starts the timer. Timer expiration flushes
any accumulated descriptors and sets an interrupt event (TXDW).
For the final condition, if TXDCTL.WTHRESH descriptors are ready for write-back, the writeback is performed.

3.4.3

Transmit Interrupts
Hardware supplies three transmit interrupts. These interrupts are initiated through the following
conditions:

• Transmit queue empty (TXQE) — All descriptors have been processed. The head pointer is
equal to the tail pointer.

• Descriptor done [Transmit Descriptor Write-back (TXDW)] — Set when hardware writes

back a descriptor with RS1 set. This is only expected to be used in cases where, for example,
the streams interface has run out of descriptors and wants to be interrupted whenever progress
is made.

• Transmit Delayed Interrupt (TXDW) — In conjunction with IDE (Interrupt Delay Enable), the
TXDW indication is delayed by a specific time per the TIDV register. This interrupt is set
when the transmit interrupt countdown register expires. The countdown register is loaded with
the value of the IDV field of the TIDV register, when a transmit descriptor with its RS1 bit and
the IDE bit are set, is written back. When a Transmit Delayed Interrupt occurs, the TXDW
interrupt cause bit is set (just as when a Transmit Descriptor Write-back interrupt occurs). This
interrupt may be masked in the same manner as the TXDW interrupt. This interrupt is used
frequently by software that performs dynamic transmit chaining, by adding packets one at a
time to the transmit chain.
Note:

The transmit delay interrupt is indicated with the same interrupt bit as the transmit write-back
interrupt, TXDW. The transmit delay interrupt is only delayed in time as discussed above.

1. Or RPS for the 82544GC/EI only.

Software Developer’s Manual

Receive and Transmit Description

• Link status change (LSC) - Set when the link status changes. When using the internal PHY,
link status changes are determined and indicated by the PHY via a change in its LINK
indication.
When using an external TBI device (82544GC/EI only), the device might indicate a link
status change using its LOS (loss of sync) indication. In this TBI mode, if HW AutoNegotiation is enabled, the MAC can also detect and signal a link status change if the
Configuration Base Page register is received (0b), or if either the LRST or ANE bits are
changed by software.

• Transmit Descriptor Ring Low Threshold Hit (TXD_LOW) (not applicable to the 82544GC/
EI) - Set when the total number of transmit descriptors available (as measured by the
difference between the Tx descriptor ring Head and Tail pointer) hits the low threshold
specified in the TXDCTL.LWTHRESH field.

3.4.3.1

Delayed Transmit Interrupts
This mechanism allows software the flexibility of delaying transmit interrupts until no more
descriptors are added to a transmit chain for a certain amount of time, rather than when the Ethernet
controller’s head pointer catches the tail pointer. This occurs if the Ethernet controller is processing
packets slightly faster than the software, a likely scenario for gigabit operations.
A software driver usually has no knowledge of when it is going to be asked to send another frame.
For performance reasons, it is best to generate only one transmit interrupt after a burst of packets
have been sent.
Refer to Section 3.3.3.1 for specific details.

3.5

TCP Segmentation
Hardware TCP Segmentation is one of the off-loading options of most modern TCP/IP stacks. This
is often referred to as “Large Send” offloading. This feature enables the TCP/IP stack to pass to the
Ethernet controller software driver a message to be transmitted that is bigger than the Maximum
Transmission Unit (MTU) of the medium. It is then the responsibility of the software driver and
hardware to carve the TCP message into MTU size frames that have appropriate layer 2 (Ethernet),
3 (IP), and 4 (TCP) headers. These headers must include sequence number, checksum fields,
options and flag values as required. Note that some of these values (such as the checksum values)
are unique for each packet of the TCP message, and other fields such as the source IP address is
constant for all packets associated with the TCP message.
The offloading of these processes from the software driver to the Ethernet controller saves
significant CPU cycles. The software driver shares the additional tasks to support these options
with the Ethernet controller.
Although the Ethernet controller’s TCP segmentation offload implementation was specifically
designed to take advantage of new “TCP Segmentation offload” features, the hardware
implementation was made generic enough so that it could also be used to “segment” traffic from
other protocols. For instance this feature could be used any time it is desirable for hardware to
segment a large block of data for transmission into multiple packets that contain the same generic
header.

Software Developer’s Manual

Receive and Transmit Description

3.5.1

Assumptions
The following assumption applies to the TCP Segmentation implementation in the Ethernet
controller:

• The RS bit operation is not changed. Interrupts are set after data in buffers pointed to by
individual descriptors is transferred to hardware.

• Checksums are not accurate above a 12 K frame size.
• The function of the RPS1 bit in the Transmit Descriptor is applicable to all of the packets that
make up the “TCP Segmentation” context, not the individual packets segmented by hardware.

3.5.2

Transmission Process
The transmission process for regular (non-TCP Segmentation packets) involves:

• The protocol stack receives from an application a block of data that is to be transmitted.
• The protocol stack calculates the number of packets required to transmit this block based on
the MTU size of the media and required packet headers.

• For each packet of the data block:
• Ethernet, IP and TCP/UDP headers are prepared by the stack.
• The stack interfaces with the software device driver and commands the driver to send the
individual packet.

• The driver gets the frame and interfaces with the hardware.
• The hardware reads the packet from host memory (via DMA transfers).
• The driver returns ownership of the packet to the operating system when the hardware has
completed the DMA transfer of the frame (indicated by an interrupt).
The transmission process for the Ethernet controller TCP segmentation offload implementation
involves:

• The protocol stack receives from an application a block of data that is to be transmitted.
• The stack interfaces to the software device driver and passes the block down with the
appropriate header information.

• The software device driver sets up the interface to the hardware (via descriptors) for the TCP
Segmentation context.

• The hardware transfers the packet data and performs the Ethernet packet segmentation and
transmission based on offset and payload length parameters in the TCP/IP context descriptor
including:
— Packet encapsulation
— Header generation & field updates including IP and TCP/UDP checksum generation
— The driver returns ownership of the block of data to the operating system when the
hardware has completed the DMA transfer of the entire data block (indicated by an
interrupt).

1. 82544GC/EI only.

Software Developer’s Manual

Receive and Transmit Description

3.5.2.1

TCP Segmentation Data Fetch Control
To perform TCP Segmentation in the Ethernet controller, the DMA unit must ensure that the entire
payload of the segmented packet fits into the available space in the on-chip Packet Buffer. The
segmentation process is performed without interruption. The DMA performs various comparisons
between the payload and the Packet Buffer to ensure that no interruptions occur. The TCP
Segmentation Pad & Minimum Threshold (TSPMT) register is used to allow software to program
the minimum threshold required for a TCP Segmentation payload. Consideration should be made
for the MTU value when writing this field. The TSPMT register is also used to program the
threshold padding overhead. This padding is necessary due to the indeterminate nature of the MTU
and the associated headers.

3.5.3

TCP Segmentation Performance
Performance improvements for a hardware implementation of TCP Segmentation offload mean:

• The operating system stack does not need to partition the block to fit the MTU size, saving
CPU cycles.

• The operating system stack only computes one Ethernet, IP, and TCP header per segment,
saving CPU cycles.

• The operating system stack interfaces with the software device driver only once per block
transfer, instead of once per frame.

• Larger PCI bursts are used which improves bus efficiency.
• Interrupts are easily reduced to one per TCP message instead of one per packet.
• Fewer I/O accesses are required to command the hardware.
3.5.4

Packet Format
Typical TCP/IP transmit window size is 8760 bytes (about 6 full size frames). A TCP message can
be as large as 64 KB and is generally fragmented across multiple pages in host memory. The
Ethernet controller partitions the data packet into standard Ethernet frames prior to transmission.
The Ethernet controller supports calculating the Ethernet, IP, TCP, and even UDP headers,
including checksum, on a frame by frame basis.
Ethernet

IPv4

TCP/UDP

DATA

FCS

Figure 3-5. TCP/IP Packet Format
Frame formats supported by the Ethernet controller’s TCP segmentation include:

•
•
•
•
•
•
•

Ethernet 802.3
IEEE 802.1q VLAN (Ethernet 802.3ac)
Ethernet Type 2
Ethernet SNAP
IPv4 headers with options
IPv6 headers with IP option next headers
IPv6 packet tunneled in IPv4

Software Developer’s Manual

Receive and Transmit Description

• TCP with options
• UDP with limitations.
UDP (unlike TCP) is not a “reliable protocol”, and fragmentation is not supported at the UDP
level. UDP messages that are larger than the MTU size of the given network medium are normally
fragmented at the IP layer. This is different from TCP, where large TCP messages can be
fragmented at either the IP or TCP layers depending on the software implementation. The Ethernet
controller has the ability to segment UDP traffic (in addition to TCP traffic). This process has
limited usefulness.
IP tunneled packets are not supported for TCP Segmentation operation1.

3.5.5

TCP Segmentation Indication
Software indicates a TCP Segmentation transmission context to the hardware by setting up a TCP/
IP Context Transmit Descriptor. The purpose of this descriptor is to provide information to the
hardware to be used during the TCP segmentation offload process. The layout of this descriptor is
reproduced in Section 3.3.6.
63

TUCSE

TUCS0

MSS

HDRLEN

TUCSS

RSV

STA

3936

35 32

16
IPCSE

TUCMD
31

IPCS0

DTYP

24 23

0
IPCSS

PAYLEN

20 19

IDE

RSV

DEXT

RSV

TSE

TCP

Figure 3-6. TCP/IP Context Transmit Descriptor & Command Layout
Setting the TSE bit in the Command field to 1b indicates that this descriptor refers to the TCP
Segmentation context (as opposed to the normal checksum offloading context). This causes the
checksum offloading, packet length, header length, and maximum segment size parameters to be
loaded from the descriptor into the Ethernet controller.
The TCP Segmentation prototype header is taken from the packet data itself. Software must
identity the type of packet that is being sent (IP/TCP, IP/UDP, other), calculate appropriate
checksum offloading values for the desired checksums, and calculate the length of the header
which is pre-pended. The header may be up to 240 bytes in length.
Once the TCP Segmentation context has been set, the next descriptor provides the initial data to
transfer. This first descriptor(s) must point to a packet of the type indicated. Furthermore, the data it
points to may need to be modified by software as it serves as the prototype header for all packets
within the TCP Segmentation context. The following sections describe the supported packet types
and the various updates which are performed by hardware. This should be used as a guide to
determine what must be modified in the original packet header to make it a suitable prototype
header.
The following summarizes the fields considered by the driver for modification in constructing the
prototype header:

Software Developer’s Manual

Receive and Transmit Description

• IPv4 Header
— Length should be set to zero
— Identification Field should be set as appropriate for first packet of send (if not already)
— Header Checksum should be zeroed out unless some adjustment is needed by the driver

• IPv6 Header
— Length should be set to zero

• TCP Header
— Sequence Number should be set as appropriate for first packet of send (if not already)
— PSH, and FIN flags should be set as appropriate for last packet of send
— TCP Checksum should be set to the partial pseudo-header checksum as follows:
IP Source Address
IP Destination Address
Zero
Zero
Zeroa

Next Header
Layer 4
Protocola

Zeroa

a. 82544GC/EI only

Figure 3-7. TCP Partial Pseudo-Header Checksum

• UDP Header
— Checksum should be set as in TCP header, above
The Ethernet controller’s DMA function fetches the ethernet, IP, and TCP/UDP prototype header
information from the initial descriptor(s) and save them on-chip for individual packet header generation. The following sections describe the updating process performed by the hardware for each
frame sent using the TCP Segmentation capability.

3.5.6

TCP Segmentation Use of Multiple Data Descriptors
TCP Segmentation enables a packet to be segmented to describe more than one data descriptor. A
large packet contained in a single virtual-address buffer is better described as a series of data
descriptors, each referencing a single physical address page.
The only requirement for this use is if multiple data descriptors for TCP segmentation follows this
guideline:

• If multiple data descriptors are used to describe the IP/TCP/UDP header section, each
descriptor must describe one or more complete headers; descriptors referencing only parts of
headers are not supported.

Software Developer’s Manual

Receive and Transmit Description

Note:

3.5.7

It is recommended that the entire header section, as described by the TCP Context Descriptor
HDRLEN field, be coalesced into a single buffer and described using a single data descriptor.

IP and TCP/UDP Headers
This section outlines the format and content for the IP, TCP and UDP headers. The Ethernet
controller requires baseline information from the software device driver in order to construct the
appropriate header information during the segmentation process.
Header fields that are modified by the Ethernet controller are highlighted in the figures that follow.
The IPv4 header is first shown in the traditional (RFC 791) representation, and because byte and bit
ordering is confusing in that representation, the IP header is also shown in little-endian format. The
actual data is fetched from memory in little-endian format.

0 1 2 3 4 5 6 7
Version

1
8 9 0 1 2 3 4 5

IP Hdr
Length

2
3
6 7 8 9 0 1 2 34 5 6 7 8 9 0 1

TYPE of service

Total length

Identification
Time to Live

Flags

Fragment Offset
Header Checksum

Layer 4 Protocol ID
Source Address
Destination Address
Options

Figure 3-8. IPv4 Header (Traditional Representation)

Software Developer’s Manual

Receive and Transmit Description

Byte 3

Byte 2

7 6 5 4 3 2 1 0

Total length

LSB

Byte 1
7 6 5 4 3 2 1 0
TYPE of service

MSB

R
N M
E
F F
S

Fragment Offset Low

Byte 0

Fragment
Offset High

Version

IP Hdr
Length

Identification

LSB

Header Checksum

7 6 5 4 3 2 1 0

Layer 4 Protocol ID

MSB

Time to Live

Source Address
Destination Address
Options

Figure 3-9. IPv4 Header (Little-Endian Order)
Flags Field Definition:
The Flags field is defined below. Note that hardware does not evaluate or change these bits.

• MF
• NF
• Reserved
Note:

More Fragments
No Fragments

The IPv6 header is first shown in the traditional (RFC 2460), big-endian representation. The actual
data is fetched from memory in little-endian format.
0 1 2 3 4 5 6 7 8 9
Version

0 1 2 3 4 5 6 7 8 9

Traffic Class
Payload Length

0 1 2 3 4 5 6 7 8 9

0 1

Flow Label
Next Header

Hop Limit

Source Address
Destination Address

Figure 3-10. IPv6 TCP Header (Traditional Representation)
A TCP or UDP frame uses a 16 bit wide one’s complement checksum. The checksum word is
computed on the outgoing TCP or UDP header and payload, and on the Pseudo Header. Details on
checksum computations are provided in Section 3.5. TCP requires the use of checksum, where it is
optional for UDP.

Software Developer’s Manual

Receive and Transmit Description

The TCP header is first shown in the traditional (RFC 793) representation. Because byte and bit
ordering is confusing in that representation, the TCP header is also shown in little-endian format.
The actual data is fetched from memory in little-endian format.

0 1

1
9 0

2 3 4 5

2
0 1

Source Port

3
4 5 6 7 8 9 0 1

Destination Port
Sequence Number
Acknowledgement Number

TCP Header
Length

Reserved

U A P R S F
R C S S Y I
G K H T N N

Window

Checksum

Urgent Pointer
Options

Figure 3-11. TCP Header (Traditional Representation)

Byte3

Byte2

7 6 5 4 3 2 1 0

Byte1

7 6 5 4 3 2 1 0

7 6

5 4 3

Destination Port

Byte0
2

7 6 5 4 3 2 1 0

Source Port
Sequence Number

LSB

MSB

Acknowledgement Number
RE
S

Window

Urgent Pointer

U A P R S
R C S S Y
G K H T N

TCP

Header

Reserved

Length

Checksum
Options

Figure 3-12. TCP Header (Little-Endian)
The TCP header is always a multiple of 32 bit words. TCP options may occupy space at the end of
the TCP header and are a multiple of 8 bits in length. All options are included in the checksum.
The checksum also covers a 96-bit pseudo header conceptually prefixed to the TCP Header (see
Figure 3-13 and Figure 3-14). The IPv4 pseudo header contains the IPv4 Source Address, the IPv4
Destination Address, the IPv4 Protocol field, and TCP Length. The IPv6 pseudo header contains
the IPv6 Source Address, the IPv6 Destination Address, the IPv6 Payload Length, and the IPv6
Next Header field. Software pre-calculates the partial pseudo header sum, which includes IPv4 SA,
DA and protocol types, but not the TCP length, and stores this value into the TCP checksum field
of the packet.
The Protocol ID field should always be added the least significant byte (LSB) of the 16 bit pseudo
header sum, where the most significant byte (MSB) of the 16 bit sum is the byte that corresponds to
the first checksum byte out on the wire.
The TCP Length field is the TCP Header Length including option fields plus the data length in
bytes, which is calculated by hardware on a frame by frame basis. The TCP Length does not count
the 12 bytes of the pseudo header. The TCP length of the packet is determined by hardware as:

Software Developer’s Manual

Receive and Transmit Description

TCP Length = Payload + HDRLEN - TUCSS
“Payload” is normally MSS except for the last packet where it represents the remainder of the
payload.
0

31
IP Source Address
IP Destination Address
Zero

Layer 4 Protocol
ID

TCP Length

Figure 3-13. TCP Pseudo Header Content (Traditional Representation)

IP Source Address
IP Destination Address
Upper Layer Packet Length
Zero

Next Header

Figure 3-14. TCP PseudoHeader Content for IPv6
Note:

The IP Destination address is the final destination of the packet. Therefore, if a routing header is
used, the last address in the route list is used in this calculation. The upper-layer packet length is
the length of the TCP header and the TCP payload.
The UDP header is always 8 bytes in size with no options.

0 1 2 3 4 5 6 7

1
8 9 0 1 2 3 4 5

2
6 7 8 9 0 1 2 3

Source Port

Destination Port

Length

Checksum

3
4 5 6 7 8 9 0 1

Figure 3-15. UDP Header (Traditional Representation)

Byte3

Byte2

0 1 2 3 4 5 6 7

Byte1
0 1 2 3 4 5

Byte0
6 7

0 1 2 3 4 5 6 7

Destination Port

Source Port

Checksum

Length

Figure 3-16. UDP Header (Little-Endian Order)

Software Developer’s Manual

Receive and Transmit Description

UDP pseudo header has the same format as the TCP pseudo header. The IPv4 pseudo header
conceptually prefixed to the UDP header contains the IPv4 source address, the IPv4 destination
address, the IPv4 protocol field, and the UDP length (same as the TCP Length discussed above).
The IPv6 pseudo header for UDP is the same as the IPv6 pseudo header for TCP. This checksum
procedure is the same as is used in TCP.
IP Source Address
IP Destination Address
Layer 4
Protocol ID

Zero

UDP Length

Figure 3-17. UDP Pseudo Header Diagram for IPv4

IP Source Address
IP Destination Address
Upper Layer Packet Length
Zero

Next Header

Figure 3-18. UDP PseudoHeader Diagram for IPv6
Note:

The IP Destination Address is the final destination of the packet. Therefore, if a routing header is
used, the last address in the route list is used in this calculation. The upper-layer packet length is
the length of the UDP header and UDP payload.
Unlike the TCP checksum, the UDP checksum is optional. Software must set the TXSM bit in the
TCP/IP Context Transmit Descriptor to indicate that a UDP checksum should be inserted.
Hardware does not overwrite the UDP checksum unless the TXSM bit is set.

3.5.8

Transmit Checksum Offloading with TCP Segmentation
The Ethernet controller supports checksum off-loading as a component of the TCP Segmentation
offload feature and as a standalone capability. Section 3.5.8 describes the interface for controlling
the checksum off-loading feature. This section describes the feature as it relates to TCP
Segmentation.
The Ethernet controller supports IP and TCP/UDP header options in the checksum computation for
packets that are derived from the TCP Segmentation feature. The Ethernet controller is capable of
computing one level of IP header checksum and one TCP/UDP header and payload checksum. In
case of multiple IP headers, the driver has to compute all but one IP header checksum. The
Ethernet controller calculates checksums on the fly on a frame by frame basis and inserts the result
in the IP/TCP/UDP headers of each frame. TCP and UDP checksum are a result of performing the
checksum on all bytes of the payload and the pseudo header.

Software Developer’s Manual

Receive and Transmit Description

Three specific types of checksum are supported by the hardware in the context of the TCP
Segmentation offload feature:

• IPv4 checksum (IPv6 does not have a checksum)
• TCP checksum
• UDP checksum
Each packet that is sent via the TCP segmentation offload feature optionally includes the IPv4
checksum and either the TCP or UDP checksum.
All checksum calculations use a 16-bit wide one’s complement checksum. The checksum word is
calculated on the outgoing data. The checksum field is written with the 16 bit one’s complement of
the one’s complement sum of all 16-bit words in the range of CSS to CSE, including the checksum
field itself.

3.5.9

IP/TCP/UDP Header Updating
IP/TCP/UDP header is updated for each outgoing frame based on the IP/TCP header prototype
which hardware transfers from the first descriptor(s) and stores on chip. The IP/TCP/UDP headers
are fetched from host memory into an on-chip 240 byte header buffer once for each TCP
segmentation context (for performance reasons, this header is not fetched again for each additional
packet that is derived from the TCP segmentation process). The checksum fields and other header
information are later updated on a frame by frame basis. The updating process is performed
concurrently with the packet data fetch.
The following sections define which fields are modified by hardware during the TCP Segmentation
process by the Ethernet controller. Figure 3-19 illustrates the overall data flow.

Software Developer’s Manual

Receive and Transmit Description

PCI FIFO

IP/TCP Header

Header
Update

Checksum
Calculation

Packet Data

TX Packet FIFO

Packet Data

HOST Memory

IP/TCP Header Buf f er

TCP Segmentation Data Flow
Header processing

Descriptors f etch

Header processing

Data Fetch Pause
IP/TCP Header
Packet Data Fetch Checksum Header
Prototy pe f etch
Insertion
Checksum
Calculations

Data Fetch
resume

Data Fetch Pause
Checksum Header
Insertion

Checksum
Calculations

Time

Ev ents Scheduling

Figure 3-19. Overall Data Flow

Software Developer’s Manual

Receive and Transmit Description

3.5.9.1

TCP/IP/UDP Header for the First Frame
The hardware makes the following changes to the headers of the first packet that is derived from
each TCP segmentation context.

• IPv4 Header
— IP Total Length = MSS + HDRLEN – IPCSS
— IP Checksum
— IPv6 Header
— Payload Length = MSS + HDRLEN - IPCSS

• TCP Header
— Sequence Number: The value is the Sequence Number of the first TCP byte in this frame.
— If FIN flag = 1b, it is cleared in the first frame.
— If PSH flag =1b, it is cleared in the first frame.
— TCP Checksum

• UDP Header
— UDP length: MSS + HDRLEN - TUCSS
— UDP Checksum

3.5.9.2

TCP/IP/UDP Header for the Subsequent Frames
The hardware makes the following changes to the headers for subsequent packets that are derived
as part of a TCP segmentation context:

Note:

Number of bytes left for transmission = PAYLEN – (N * MSS). Where N is the number of frames
that have been transmitted.

• IPv4 Header
— IP Identification: incremented from last value (wrap around)
— IP Total Length = MSS + HDRLEN – IPCSS
— IP Checksum

• IPv6 Header
• Payload Length = MSS + HRDLEN - IPCSS
• TCP Header
— Sequence Number update: Add previous TCP payload size to the previous sequence
number value. This is equivalent to adding the MSS to the previous sequence number.
— If FIN flag = 1b, it is cleared in these frames.
— If PSH flag =1b, it is cleared in these frames.
— TCP Checksum

• UDP Header
— UDP Length: MSS + HDRLEN – TUCSS
— UDP Checksum

Software Developer’s Manual

Receive and Transmit Description

3.5.9.3

TCP/IP/UDP Header for the Last Frame
The controller makes the following changes to the headers for the last frame of a TCP
segmentation context:

Note:

Last frame payload bytes = PAYLEN – (N * MSS)

• IPv4 Header
— IP Total Length = (last frame payload bytes + HDRLEN) – IPCSS
— IP Identification: incremented from last value (wrap around)
— IP Checksum

• IPv6 Header
• Payload Length = MSS + HDRLEN - IPCSS
• TCP Header
— Sequence Number update: Add previous TCP payload size to the previous sequence
number value. This is equivalent to adding the MSS to the previous sequence number.
— If FIN flag = 1b, set it in this last frame
— If PSH flag =1b, set it in this last frame
— TCP Checksum

• UDP Header
— UDP length: (last frame payload bytes + HDRLEN) - TUCSS
— UDP Checksum

3.6

IP/TCP/UDP Transmit Checksum Offloading
The previous section on TCP Segmentation offload describes the IP/TCP/UDP checksum
offloading mechanism used in conjunction with TCP Segmentation. The same underlying
mechanism can also be applied as a standalone feature. The main difference in normal packet mode
(non-TCP Segmentation) is that only the checksum fields in the IP/TCP/UDP headers need to be
updated.
Before taking advantage of the Ethernet controller’s enhanced checksum offload capability, a
checksum context must be initialized. For the normal transmit checksum offload feature, this task
is performed by providing the Ethernet controller with a TCP/IP Context Descriptor with TSE = 0b
to denote a non-segmentation context. For additional details on contexts, refer to Section 3.3.5.
Enabling the checksum offloading capability without first initializing the appropriate checksum
context leads to unpredictable results. Once the checksum context has been set, that context, is used
for all normal packet transmissions until a new context is loaded. Also, since checksum insertion is
controlled on a per packet basis, there is no need to clear/reset the context.
The Ethernet controller is capable of performing two transmit checksum calculations. Typically,
these would be used for TCP/IP and UDP/IP packet types, however, the mechanism is general
enough to support other checksums as well. Each checksum operates independently and provides
identical functionality. Only the IP checksum case is discussed as follows.

Software Developer’s Manual

Receive and Transmit Description

Three fields in the TCP/IP Context Descriptor set the context of the IP checksum offloading
feature:

• IPCSS
This field specifies the byte offset form the start of the transferred data to the first byte to be
included in the checksum. Setting this value to 0b means that the first byte of the data is
included in the checksum. The maximum value for this field is 255. This is adequate for
typical applications.
Note:

The IPCSS value needs to be less than the total DMA length to a packet. If this is not the case, the
result will be unpredictable.

• IPCSO
This field specifies where the resulting checksum should be placed. Again, this is limited to
the first 256 bytes of the packet and must be less than or equal to the total length of a given
packet. If this is not the case, the checksum is not inserted.

• IPCSE
This field specifies where the checksum should stop. A 16-bit value supports checksum
offloading of packets as large as 64KB. Setting the IPCSE field to all zeros means End-ofPacket. In this way, the length of the packet does not need to be calculated.
As mentioned above, it is not necessary to set a new context for each new packet. In many cases,
the same checksum context can be used for a majority of the packet stream. In this case, some of
the offload feature only for a particular traffic type, thereby avoiding all context descriptors except
for the initial one.

Software Developer’s Manual

Receive and Transmit Description

Note:

This page intentionally left blank.

Software Developer’s Manual

PCI Local Bus Interface

The PCI/PCI-X Family of Gigabit Ethernet Controllers are PCI 2.2 or 2.3 compliant devices and
implement the PCI-X Addendum to the PCI Local Bus Specification, Revision 1.0.
Note:

4.1

The 82540EP/EM, 82541xx, and 82547GI/EI do not support PCI-X mode.

PCI Configuration
The PCI Specification requires implementation of PCI Configuration registers. After a system
reset, these registers are initially configured by the BIOS, and/or a “Plug and Play” aware
Operating System (OS). Device drivers read these registers to determine what resources (interrupt
number, memory mapping location, etc.) the BIOS and/or OS assigned to the Ethernet controller.
The 82547GI/EI uses a dedicated CSA port for its system bus connection. Logically, it still follows
PCI configuration. However, some configuration parameters, such as cache line, are irrelevant.
Additionally, the 82547GI/EI requires special interrupt configuration in the BIOS (see Section
4.5).
Note:

The 82547GI/EI does not support 64-bit addressing.
Four different regions of the PCI configuration space are used.
Address

Item

00h-3Ch

Description

PCI

Section 2.3.1

DCh-E0h

PCI Power Management

Section 6.3.3

E4h-E8h

PCI-X

F0h-FCh
a.

Section 4.1.1
a

Message Signaled Interrupt

Section 4.1.3.1

Not applicable to the 82541xx and 82547GI/EI.

These spaces are linked into a linked list using the Capabilities Pointer field (Cap_Ptr) in the PCI
Configuration section.
The implementation of the PCI registers for the PCI/PCI-X Family of Gigabit Ethernet Controllers
are listed in Table 4-1:
Table 4-1. Mandatory PCI Registers
Byte Offset

Byte 3

Byte 2

Byte 1

Byte 0

Device ID

Vendor ID

Status Register

Command Register

Class Code (020000h)

Header Type
(00h)

BIST (00h)

Revision ID
Latency
Timer

10h

Base Address 0a

Base Address 1

18h

Base Address 2

Software Developer’s Manual

Cache Line
Size

PCI Local Bus Interface

1Ch

Base Address 3 (unused)

20h

Base Address 4 (unused)

2h4

Base Address 5 (unused)

28h

Cardbus CIS Pointer (not used)

2Ch

Subsystem ID

30h
34h

Reserved

38h
3Ch
a.

Subsystem Vendor ID

Expansion ROM Base Address
Cap_Ptr

Reserved
Max_Latency
(00h)

Min_Grant
(FFh)

Interrupt Pin
(01h)

Interrupt Line

Refer to Table 4-2.

The following list provides explanations of the various PCI registers and their bit fields:
Vendor ID

This uniquely identifies all Intel PCI products. This field may be auto-loaded
from the EEPROM at power on or upon the assertion of PCI_RST#. A value of
8086h is the default for this field upon power up if the EEPROM does not
respond or is not programmed.

Device ID

This uniquely identifies the Ethernet controller. This field may be autoloaded
from the EEPROM at power on or upon the assertion of RST#. The default value
for this field is used upon power up if the EEPROM does not respond or is not
programmed.

Command Reg. The layout is listed in Table 4-3. Shaded bits are not used by this implementation
and are hard wired to 0b.

Status Register

The layout is listed in Table 4-4. Shaded bits are not used by this implementation
and are hard wired to 0b.

Revision

Sequential stepping number starting with 00h for the A0 revision of the Ethernet
controller. Refer to the PCI/PCI-X Family of Gigabit Ethernet Controllers
Specification Update for the latest stepping information.

Class Code

The class code, 020000h identifies the Ethernet controller as an Ethernet adapter.

Software Developer’s Manual

PCI Local Bus Interface

Cache Line Size1 Used to store the cache line size. The value is in units of 4 bytes. A system with a
cache line size of 64 bytes sets the value of this register to 10h. The only sizes
that are supported are 16, 32, 64, and 128 bytes. All other sizes are treated as 0b.
See the information about exceptions in Section 4.4.
Unsupported values affect PCI cache line support. All writes default to using the
memory write (MW) command, and memory read command determination uses
a cache line size of 32 bytes.
Latency Timer

The lower two bits are not implemented and return 0b. The upper six bits are
Read/Write.

Header Type

This is for a normal single function Ethernet controller and reads 00h.

BIST

Built in Self-test is not implemented as supportable from PCI configuration
space in this version of the Ethernet controller.

Base Address Registers
The Base Address Registers (or BARs) are used to map the Ethernet controller’s register space and flash to system memory space. In PCI-X mode
or in PCI mode when the BAR32 bit of the EEPROM is 0b, two registers
are used for each of the register space and the flash memory in order to
map 64-bit addresses. In PCI mode, if the BAR32 bit in the EEPROM is
1b, one register is used for each to map 32-bit addresses.
64-bit BARs

PCI-X mode with BAR32 bit in the EEPROM set to 0b.

Table 4-2. Base Address Registers
BAR

Addr.

31 4

10h

Memory Register Base Address (bits
31:4)

pref.

14h

Memory Register Base Address (bits 63:32)

18h

Memory Flash Base Address (bits
31:4)

1Ch

Memory Flash Base Address (bits 63:32)

20h

IO Register Base Address (bits 31:2)

24h

Reserved (read as all 0b’s)

32-bit BARs

pref.

type

mem

type

mem

Conventional PCI mode with BAR32 bit in the EEPROM set to 1b

BAR

Addr.

10h

Memory Register Base Address

pref.

type

mem

14h

Memory Flash Base Address

pref.

type

mem

18h

IO Register Base Address (bits 31:2)

mem

1Ch

Reserved (read as all 0b’s)

20h

Reserved (read as all 0b’s)

24h

Reserved (read as all 0b’s)

Not applicable to the 82547GI/EI.

Software Developer’s Manual

PCI Local Bus Interface

All base address registers have the following fields:
Field

Mem

Bit(s)

Read/
Write

Initial
Value
0b for
mem

Description
0b indicates memory space. 1b indicates I/O.

1b for I/O

Type

2:1

00b for 32bit

Indicates the address space size.

10b for 64bit

10b = 64-bit

00b = 32-bit

0b = non-prefetchable space
Prefetch

1b = prefetchable space
Ethernet controller implements non-prefetchable space
since it has read side-effects.
The lower bits of the address are hard-wired to 0b. The
upper bits can be written by the system software to set
the base address of the register or flash address space.
The memory register space is 128K bytes. The
Memory Register BAR has:
• Bits 16:4 are hard-wired to 0b.
• Bits 63:17 or 31:17 are read/write.
The size of the flash space can very between 64 KB and
512 KB depending on the FLASH size read from the
EEPROM. The Memory Flash BAR has these
characteristics:

Address

31:0

R/W

0b
Flash Size Valid Bits Zero Bits
(R/W)

(RO)

• 64 KB 63/31:16 15:4
• 128 KB 63/31:17 16:4
• 256 KB 63/31:18 17:4
• 512 KB 63/31:19 18:4
The size of the IO register space is 8 bytes. The I/O
Register BAR has:
• Bit 2 hard-wired to 0b
• Bits 31:3 as read/write

Software Developer’s Manual

PCI Local Bus Interface

Expansion ROM Base Address
This register is used to define the address and size information for boottime access to the optional Flash memory.
31

Expansion Rom Base Address

Bit(s)

Read/
Write

R/W

Reserved

10:1

Field

Reserved

Initial
Value

Description
1b = Enables expansion ROM access.
0b = Disables expansion ROM access.
Always read as 0b. Writes are ignored.
The lower bits of the address are hard-wired to 0b.
The upper bits can be written by the system software
to set the base address of the register or flash
address space.

Address

31:11

R/W

Since the flash is used as the expansion ROM, the
size of the expansion ROM can very between 64 KB
and 512 KB, depending on the FLASH size read
from the EEPROM.

Flash Size Valid Bits Zero Bits:
• 64 KB 63/31:16 15:11
• 128 KB 63/31:17 16:11
• 256 KB 63/31:18 17:11
• 512 KB 63/31:19 18:11

CardBus CIS Pointer (82541PI/GI/EI and 82540EP Only)
When the Enable CLK_RUN# bit of the EEPROM’s Initialization Control
Word 2 and the 64/32 BAR bit of the EEPROM Initialization Control
Word 1 (indicating a 32-bit BAR) are both set to 1b, the Cardbus CIS
Pointer contains a value of 00000022h. Otherwise, it contains a value of
00000000h.
31

3
Offset

Software Developer’s Manual

0
Space

PCI Local Bus Interface

Field

Read/
Write

Bit(s)

Initial
Value

Description
Indicates the address space where the CIS is
located.
0 = Configuration Space
1 = BAR0
2 = BAR1

Space

2:0

R/W

0 or 2

3 = BAR2
4 = BAR3
5 = BAR4
6 = BAR5
7 = Expansion ROM

Offset

Subsystem ID

31:3

0 or 4

Offset within the specified address space,
multiplied by eight. When enabled, the value
indicates that the CIS (Card Information
Structure) is at an offset of 4*8, or 32 bytes into
the Flash memory.

This value can be loaded automatically from the EEPROM upon power-up or
PCI reset. A value of 1008h is the default for this field upon power-up if the
EEPROM does not respond or is not programmed.

Subsystem Vendor ID
This value can be loaded automatically from the EEPROM upon power-up or
PCI reset. A value of 8086h is the default for this field upon power-up if the
EEPROM does not respond or is not programmed.
Cap_Ptr

The Capabilities Pointer field (Cap_Ptr) is an 8-bit field that provides an offset in
the Ethernet controller’s PCI Configuration Space for the location of the first
item in the Capabilities Linked List. The Ethernet controller sets this bit and then
implements a capabilities list to indicate that it supports PCI Power
Management, PCI-X, and Message Signaled Interrupts1. Its value is DCh which
is the address of the first entry: ACPI2 Power Management.
Address

Item

Next Pointer

DCh-E0h

ACPI Power Management

E4h

E4h-E8h

PCI-X

F0h

F0h-FCh

Message Signaled Interrupt

00h

Figure 4-1. Capabilities Linked List
In conventional PCI mode, Message Signaled interrupts can be disabled in the
EEPROM. If disabled, the message signaled interrupts won’t appear on the
linked list and PCI-X’s “Next Pointer” is 0b.
1.
2.

Not applicable to the 82541xx or 82547GI/EI.
Not applicable to the 82541ER.

Software Developer’s Manual

PCI Local Bus Interface

Max_Lat/Min_Gnt1
The Ethernet controller places a very high load on the PCI bus during peak
transmit and receive traffic. In full duplex mode, it has a peak throughput
demand of 250 MB/sec. The peak delivered bandwidth on a 64-bit PCI bus at 33
MHz is 264 MB/sec, so the bus is fully saturated when transmit and receive are
operating simultaneously. In half duplex operation, the Ethernet controller has a
peak throughput demand of 125 MB/sec, which still puts an enormous load on
the PCI bus. Consequently, the Max_Lat should be small and is set to 00h, and
Min_Gnt is set to FFh indicating that the Ethernet controller requires a very high
priority and time slice.
Interrupt Pin
Read only register indicating which interrupt line (INTA# vs. INTB#) the
82546GB/EB uses. A value of 1b indicates that the 82546GB/EB uses INTA#
(as with all single-port Ethernet controllers). A value of 10b indicates that the
82546GB/EB uses INTB#.
For each separate device/function within the Ethernet controller, the value
reported here is based on the EEPROM Initialization Control Word 3 associated
with this controller, as well as whether both device/functions are enabled.
Provided both functions are enabled, then the value reported for each specific
function is based on the Interrupt Pin field of each Ethernet controller’s
Initialization Control Word 3.
If only a single internal device/function is enabled, then the value reported here
is 1b regardless of EEPROM configuration.
Interrupt Line

Read write register programmed by software to indicate which of the system
interrupt request lines this Ethernet controller’s interrupt pin is bound to. See the
PCI definition for more details.

Table 4-3. Command Register Layout
15

10 9
Reserved

0
Command Bits

Bit(s)

Initial Value

Description

I/O Access Enable.

Memory Access Enable.

Enable Mastering. Ethernet controller in PCI-X
mode is permitted to initiate a split completion
transaction regardless of the state of this bit.

Special Cycle Monitoring.

This bit is a don’t care for the 82547GI/EI.

Software Developer’s Manual

PCI Local Bus Interface

Bit(s)

Initial Value

Memory Write and Invalidate Enable (not
applicable to the 82547GI/EI).

Palette Snoop Enable.

Parity Error Response (not applicable to the
82547GI/EI).

Wait Cycle Enable.

SERR# Enable (not applicable to the 82547GI/EI).

Fast Back-to-Back Enable.

10a

Interrupt Disable (INTA# or CSA signaled).

Reserved.

15:10
15:11a
a.

Description

82541xx and 82547GI/EI only.

Table 4-4. Status Register Layout
15

4 3
Status Bits

Bit(s)
3:0

Description

Reserved.

Interrupt Status. This bit is 1b when the Ethernet
controller is generating an interrupt internally.
When Interrupt Disable in the Command Register
is also cleared, the Ethernet controller asserts
INTA# or signal an interrupt over CSA.

New Capabilities: Indicates that an Ethernet
controller implements Extended Capabilities. The
Ethernet controller sets this bit and implements a
capabilities list to indicate that it supports PCI
Power Management, PCI-X Bus, and message
signaled interrupts.

66 MHz Capable (don’t care for the 82547GI/EI).

UDF Supported. Hardwired to 0b for PCI 2.3a.

Fast Back-to-Back CapableThis bit must be
cleared to 0b in PCI-X mode (not applicable to the
82547GI/EI).

Data Parity Reported.

10:9

01b

DEVSEL Timing (indicates medium device). Not
applicable to the 82547GI/EI.

Signaled Target Abort.

2:0a

Initial Value

0
Reserved

Software Developer’s Manual

PCI Local Bus Interface

4.1.1

Bit(s)

Initial Value

Description

Received Target Abort.

Received Master Abort.

Signaled System Error (not applicable to the
82547GI/EI).

Detected Parity Error (not applicable to the
82547GI/EI).

82541xx and 82547GI/EI only.

PCI-X Configuration Registers
The Ethernet controller supports additional configuration registers that are specific to PCI-X.
These registers are visible in conventional PCI and PCI-X modes, although they only affect the
operation of PCI-X mode. The PCI-X registers are linked into the Capabilities linked list.
Note:

The 82540EP/EM, 82541xx, and 82547GI/EI do not support PCI-X mode.

Byte Offset

Byte 3

E4h

Byte 2
PCI-X Command

E8h

Byte 1

Byte 0

Next Capability

PCI-X Capability ID

PCI-X Status

Figure 4-2. PCI-X Capability Registers

4.1.1.1

PCI-X Capability ID
Read/
Write

Bits
7:0

4.1.1.2

Initial
Value

Description
Capability ID - Identifies the PCI-X register set in the capabilities
linked list.

Next Capability
Read/
Write

Bits

7:0
a.

Initial
Value
F0a

Description
Next Capability – points to the next capability in the capabilities
linked list.

In conventional PCI mode, Message Signaled Interrupts can also be disabled in the EEPROM. If disabled, the Message
Signaled Interrupt registers are not visible, and PCI-X’s “Next Capability” pointer is 0b.

Software Developer’s Manual

PCI Local Bus Interface

4.1.1.3

PCI-X Command
15

Read
Write

Max. Split Transactions

Reserved

Bits

Initial
Value

2
Read
Count

Description

Data Parity Error Recovery Enable. If this bit is 1b, the Ethernet
controller attempts to recover from Parity errors. If this bit is 0b, the
Ethernet controller asserts SERR# (if enabled) whenever the Master
Data Parity Error bit (Status Register, bit 8) is set.

Enable Relaxed Ordering. If this bit is set, the Ethernet controller sets
the Relaxed Ordering attribute bit in some transactions.
Maximum Memory Read Byte Count. This register sets the
maximum byte count the Ethernet controller uses for a Memory Read
Sequence. The allowable values are:
Register

3:2

Maximum Byte Count

512

1024

2048

4096

Maximum Outstanding Split Transactions. This register sets the
maximum number of outstanding split transactions that the Ethernet
controller uses. The Ethernet controller is only allowed to have one
outstanding split transaction at any time.
Register

Maximum Outstanding Transactions

0
6:4

15:7

Reserved. Reads as 0b

Software Developer’s Manual

PCI Local Bus Interface

4.1.1.4

PCI-X Status
31 29

28 26

25 23

22 21

Res.

Read
Size

Max.
Split

Rd
Byte

Cplx

USC

SCD

133

64b

Read/
Write

Bits

Intial
Value

8
Bus Number

Device
Number

Func.
Num.

Description

2:0

Function Number. This number forms part of the Requester and
Completer IDs for PCI-X transactions.

7:3

1Fh

Device Number. The system assigns a device number (other than 0b) to
the Ethernet controller. It forms part of the Requester and Completer IDs
for PCI-X transactions. The Ethernet controller updates this register with
the contents of AD[15:11] on any Type 0 Configuration Write cycle.

15:8

FFh

Bus Number. This indicates the bus the Ethernet controller is placed on. It
forms part of the Requester and Completer IDs for PCI-X transactions. The
Ethernet controller updates this register with the contents of AD[7:0] on any
Type 0 Configuration Write cycle.

1ba

64-bit Device. This indicates the Ethernet controller is a 64-bit device. It
does not indicate the current bus width. It is loaded from the EEPROM
Initialization Control Word 2 (see Section 5.6.12).

1ba

133 MHz Capable. A 1b indicates that the Ethernet controller is capable of
operating at 133 MHz in PCI-X mode. A 0b indicates 66 MHz capability.
This bit is loaded from the EEPROM Initialization Control Word 2 (see
Section 5.6.12).

read, write 1b
to clear

Split Completion Discarded. (Write 1b to clear) This bit is set if the
Ethernet controller discards a Split Completion because the requester
would not accept it.

read, write 1b
to clear

Unexpected Split Completion. (Write 1b to clear) This bit indicates
whether the Ethernet controller received an unexpected Split Completion
with its requestor ID.

Device Complexity. A 0b indicates the Ethernet controller is a simple
device. A 1b indicates that the Ethernet controller is a bridge.
Designed Maximum Memory Read Byte Count. Indicates the maximum
memory read byte count the Ethernet controller is designed to generate.
Register

22:21

2ba

Maximum Byte Count

512

1024

2048

4096

The value of this register depends on the Max_Read bit in the EEPROM’s
Initialization Control Word 2 (see Section 5.6.12).
Max_Read = 0b then value = 2 (2 KB)
Max_Read = 1b then value = 3 (4 KB)

Software Developer’s Manual

PCI Local Bus Interface

Read/
Write

Bits

Intial
Value

Description
Designed Maximum Outstanding Split Transactions. A 0b indicates that
the Ethernet controller is designed to have at the most one outstanding
transaction.
Register

25:23

Maximum Outstanding Transactions

Designed Maximum Cumulative Read Size. Indicates a number that is
greater or equal maximum cumulative outstanding bytes to be read at one
time.

(see Description)

28:26

0ba

Maximum Outstanding Bytes

1 KB

2 KB

4 KB

8 KB

16 KB

32 KB

64 KB

128 KB

The value of this register depends on the DMCR_Map and Max_Read bits
in the EEPROM’s Initialization Control Word 2 (see Section 5.6.12).
• DMCR_Map = 0b:
The value of this register reflects the number of bytes programmed in the
Maximum Memory Read Byte Count (MMRBC) field of the PCI-X
Command Register as follows:
• MMRBC = 0 (512) - DMCRS = 0 (1KB)
• MMRBC = 1 (1K) - DMCRS = 0 (1KB)
• MMRBC = 2 (2K) - DMCRS = 1 (2KB)
• MMRBC = 3 (4K) - DMCRS = 2 (4KB)
• DMCR_Map = 1b and Max_Read = 0b: DMCRS = 1 (2KB)
• DMCR_Map = 1b and Max_Read = 1b: DMCRS = 2 (4KB)

Read, write 1b
to clear

Received Split Completion Error Message. This bit is set if the Ethernet
controller receives a Split Completion Message with the Split Completion
Error attribute bit set.

31:30

Reserved. Reads as 0b

a. Loaded from EEPROM.

4.1.2

Reserved and Undefined Addresses
Any PCI or PCI-X register address space not explicitly declared in this specification should be
considered to be reserved, and should not be written. Writing to reserved or undefined
configuration register addresses can cause indeterminate behavior. Reads from reserved or
undefined configuration register addresses can return indeterminate values.

Software Developer’s Manual

PCI Local Bus Interface

Message Signaled Interrupts1

4.1.3

Message Signaled Interrupt (MSI) capability is optional for PCI 2.2 or 2.3, but required for PCI-X.
When Message Signaled Interrupts are enabled, instead of asserting an interrupt pin, the Ethernet
controller generates an interrupt using a memory write command. The address and most of the data
of the command are determined by the system and programmed in configuration registers. This
permits the system to program a different message for each function so it can speed up interrupt
delivery.
To enable Message Signaled Interrupts, the system software writes to the “MSI Enable” bit in the
MSI “Message Control” register. When Message Signaled Interrupts are enabled, the Ethernet
controller no longer asserts its INTA# pin to signal interrupts.
MSI systems allow a function to request up to 32 messages, but does not guarantee that all of them
are allocated. The Ethernet controller supports only a single message. When Message Signaled
Interrupts are enabled, the Ethernet controller generates a message when any of the unmasked bits
in the Interrupt Cause Read register (ICR) are set to 1b. The Ethernet controller does not generate
the message again until the ICR is read and a subsequent interrupt event occurs.
In conventional PCI mode, Message Signaled Interrupts can also be disabled in the EEPROM. If
MSI is disabled, the Message Signaled Interrupt registers is not visible.

4.1.3.1

Message Signaled Interrupt Configuration Registers
Byte Offset

Byte 3

F0h

Byte 2

Message Control

Byte 1

Byte 0

Next Capability

MSI
Capability ID

F4h

Message Address

F8h

Message Upper Address

FCh

Reserved

Message Data

Figure 4-3. Message Signaled Interrupt Configuration Registers

4.1.3.1.1

MSI Capability ID
Read/
Write

Bits
7:0

4.1.3.1.2

Description
Capability ID - Identifies the Message Signaled Interrupt register set in
the capabilities linked list.

05h

Next Capability
Read/
Write

Bits

7:0

Initial
Value

00h

Description
Next Capability – points to the next capability in the capabilities
linked list. Its value is 0b since the Message Signaled Interrupt is the
last item in the list.

Not applicable to the 82541xx or 82547GI/EI.

Software Developer’s Manual

PCI Local Bus Interface

4.1.3.1.3

Message Control
15

8
Reserved

Read/
Write

Bits

7
64b

Multiple
Enable

Initial
Value

Multiple
Capable

0
En

Description
MSI Enable. If 1b, Message Signaled Interruptsa are enabled and the
Ethernet controller generates Message Signaled Interrupts instead of
asserting INTA#.
Multiple Message Capable. Indicates the number of messages
requested. The Ethernet controller only requests one message.
Register

3:1

Number of messages

Reserved

6:4

Multiple Message Enable. Written by the system to indicate the
number of messages allocated. Since the Ethernet controller only
supports one message, the system should never write a value other
than 0b.

64-bit capable. A value of 1b indicates that the Ethernet controller is
capable of generating 64-bit message addresses.

15:8

Reserved. Reads as 0b.

Not applicable to the 82541xx or 82547GI/EI.

Software Developer’s Manual

PCI Local Bus Interface

4.1.3.1.4

Message Address
Read/
Write

Bits

31:0

4.1.3.1.5

Description
Message Address – Written by the system to indicate the lower 32bits of the address to use for the MSI memory write transaction. The
lower two bits are always written as 0b.

Message Upper Address
Read/
Write

Bits

31:0

4.1.3.1.6

Initial
Value

Description
Message Upper Address – Written by the system to indicate the
upper 32-bits of the address to use for the MSI memory write
transaction.

Message Data
Read/
Write

Bits

15:0

4.2

Initial
Value

Description
Message Data – Written by the system to indicate the lower 16 bits of
the data written in the MSI memory write DWORD transaction. The
upper 16 bits of the transaction are written as 0b.

Commands
The Ethernet controller is capable of decoding and encoding commands for both PCI and PCI-X
modes. The difference between PCI and PCI-X commands is noted in Table 4-5.

Table 4-5. PCI and PCI-X Encoding Difference
C/BE
Encoding

PCI Commands

Abr.

PCI-X Commands

Abr.

Interrupt Acknowledge

Special Cycle

I/O Read

IOR

I/O Read

IOR

I/O Write

IOW

I/O Write

IOW

Reserved

Memory Read

Memory Write

8h
9h

Memory Read DWORD

MRD

Reserved

Alias to MRB

AMR

Reserved

Alias to MWB

AMW

Configuration Read

CFR

Configuration Read

CFR

Configuration Write

CFW

Configuration Write

CFW

Memory Read Multiple

MRM

Split Completion

Software Developer’s Manual

PCI Local Bus Interface

Table 4-5. PCI and PCI-X Encoding Difference
C/BE
Encoding

PCI Commands

Abr.

PCI-X Commands

Abr.

Dual Address Cycle

DAC

Dual Address Cycle

DAC

Memory Read Line

MRL

Memory Read Block

MRB

Memory Write & Invalidate

MWI

Memory Write Block

MWB

As a target, the Ethernet controller only accepts transactions that address its BARs or a
configuration transaction in which its IDSEL input is asserted. In PCI-X mode, the Ethernet
controller also accepts split completion for an outstanding memory read command that it has
requested. The Ethernet controller does not respond to Interrupt Acknowledge or Special Cycle in
either mode.
Table 4-6. Accepted PCI/PCI-X Command as a Target
Transaction Target

PCI Commands

PCI-X Commands

MR,MRL,MRM,IOR

MRD, MRB, AMR,IOR

MW, MWI,IOW

MW, MWB, AMW,IOW

Configuration Read

CFR

Configuration Write

CFW

Memory Read Completion

N/A

As a master, the Ethernet controller generates Read and Write commands for different causes as
listed in Table 4-7. The addresses of these transactions are programmed either by system software
or the software driver. The Ethernet controller always expects that they are claimed by one of the
devices on the bus segment. The Ethernet controller never generates Interrupt Acknowledge,
Special Cycle, I/O commands, or Configuration Commands.
Table 4-7. Generated PCI/PCI-X as a Master
Transaction Cause

PCI Commands

PCI-X Commands
CMD

Tx Descriptor Read

MR,MRL,MRM

MRB

Tx Descriptor Write back

MW,MWI

MWB

Tx Data Read

MR, MRL,MRM

MRB

Rx Descriptor Read

MR,MRL,MRM

MRB

Rx Descriptor Write back

MW,MWI

MWB

Rx Data Write

MW,MWI

MWB

Message Signaled Interrupt

MWB

Split Completion

N/A

Not applicable to the 82541xx or 82547GI/EI.

Transaction burst length on PCI is determined by several factors, including the PCI latency timer
expiration, the type of bus transfer (descriptor read/write or data read/write) made, the size of the
data transfer (for data transfers), and whether the cycle is initiated by the receive or transmit logic.

Software Developer’s Manual

PCI Local Bus Interface

Following are a few specific rules:

• For descriptor fetches, the burst length is always equal to the multiple of cache line sizes set by
the transmit and receive descriptor fetch threshold fields. (See Section 3.2.4 and Section 3.4.1)
For descriptor writes, the transfer size ranges from 8 bytes to N cache line's worth of data.
Cache line sizes are: 16, 32, 64, and 128 bytes.

• For transmit data fetches, the burst length is generally equal to the block of data being fetched,
in other words, a descriptor's worth of data.

• For receive data writes, the burst size is typically equal to the packet length (rounded up to the
next 8 bytes) or the buffer size, whichever is smaller.

4.3

PCI/PCI-X Command Usage
The Ethernet controller optimizes the use of PCI/PCI-X bus cycles to maximize throughput. The
following sections describe this behavior.

4.3.1

Memory Write Operations
Memory write command usage has been implemented in the Ethernet controller to improve PCI
performance. As noted below, cache line size has a significant impact on the usage of memory
write commands. Specifically, cache line size entries which are unsupported causes hardware to
default to the Memory Write (MW) command for all master write transactions. Also, all writes
default to MW if the Memory Write and Invalidate (MWI) enable bit in the PCI configuration
command register is 0b. MWI is the preferred write command and is used when the circumstances
allow it.
Figure 4-4 depicts a behavioral state-machine representation of the command usage algorithm for
master write operations.
Upon EACH master write access, the hardware evaluates the address alignment and the amount of
data to be transferred. The following guidelines are used for command determination:

• If the address is cache line aligned and there is at least one cache line of data, then hardware
uses the MWI command.

• If the address is aligned but there is not at least one cache line of data, or the address is not
aligned, or if the MWI enable bit is set to 0b, then hardware uses the MW command.
During the burst, regardless of which command was originally issued, the hardware evaluates the
remaining amount of data each time the write burst comes to a cache line boundary, or when the
transaction is terminated due to a target disconnect or latency timer expiration.

Software Developer’s Manual

PCI Local Bus Interface

IDLE

Wr_Req || (Cnt_Rmn != 0)

Write Command
Determination

Aligned && (Count >= CLS)

MWI
Burst*

!Aligned || (Aligned && (Count < CLS)) || !MWI_Enable

MW
Burst

Cnt_Rmn = 0
Cnt_Rmn = 0

Terminate
Transaction
Boundary || Terminate
MWI_Enable &&
d && Cnt_Rmn >= CLS)

Terminate
MWI
Boundary
Evaluation

!MWI_Enable ||
(!Aligned || Cnt_Rmn < CLS)

* Either the initiation or continuation of the MWI Burst

Count = Amount of data for XFR
Cnt_Rmn = Remaining data for XFR
Wr_Req = Initial request for master write
CLS = Cache line size
Boundary = At cache line boundary?
Aligned = Address aligned to cache line boundary
Terminate = Target disconnect or latency timer ex

Figure 4-4. Master Write Command Usage Algorithm

4.3.1.1

MWI Bursts
• If there is at least one cache line of data remaining, then the Ethernet controller continues the
MWI burst.

• If there is not at least one cache line of data remaining, then the Ethernet controller terminates
the transaction on the boundary, re-acquires the bus, and issues a MW command for the
remainder of data.

• If the transaction is terminated prematurely due to a target disconnect or latency time-out, the
Ethernet controller re-evaluates command usage based on the new start address and the
amount of remaining data.

Software Developer’s Manual

PCI Local Bus Interface

4.3.1.2

MW Bursts
• The Ethernet controller always continues the burst until the end. If the system is concerned
about MWI usage, it disconnects at the cache line boundary. The Ethernet controller then
restarts the transaction and re-evaluates command usage.

Note:

4.3.2

The algorithm described above defaults to the MW case when the MWI enable bit in the
Configuration Register is set to 0b.

Memory Read Operations
For all read commands, the hardware evaluates the amount of data to be read with respect to the
cache line size register and the read address alignment for command determination. The following
rules apply:

Table 4-8. Rules for Memory Read Operations
Amount of
Data Requested

Number of Cache Line
Boundaries Crossed

Command
Used by Hardware

> 2 Cache Lines

n/a

MRM

>= 1 Cache Line

>= 2

MRM

<= 1 Cache Line

0 or 1

MRL

< 1 Cache Line

In other words, read command usage depends on the number of cache lines from which the data
must be read from the target device.
As mentioned above, unsupported values in the cache line size field default to a size of 32 bytes for
the memory read command usage algorithm.
Note:

4.3.2.1

MRL should be used for a single cache line of data that is cache line aligned.

PCI-X Command Usage
In PCI-X mode, the Ethernet controller takes advantage of split transaction protocol to minimize
retries and eliminate delayed read transactions.
Target Split Responses
When the Ethernet controller responds to a Memory Read or I/O Read command it determines if
the data can be returned within 16 clock cycles. If not, it signals a split response and returns data
later through the split completion protocol. If the Ethernet controller already has a command in its
completion register it retries the requested read until the register is empty. Target posted writes and
split completion are still accepted during that period. The internal register reads that cause splits
are:

• General Registers: CTRL, STATUS, EECD, CTRL_EXT, FCAL, FCAH, FCT, VET, FCTTV,
TXCW, RXCW, PBA,

• Interrupt Registers: ICR, ICS, IMS, IMC
• Transmit Registers: TCTL

Software Developer’s Manual

PCI Local Bus Interface

Outstanding Memory Read
When the Ethernet controller masters a memory read and is responded to with a split response it
waits for the completion of the data as a target. The Ethernet controller allows one outstanding
memory read command at any time. The Ethernet controller continues to master posted memory
writes and split completions if there are any.
Relaxed Ordering

• The Ethernet controller takes advantage of the relaxed ordering rules in PCI-X. By setting the
RO bit for some of its master transactions, the Ethernet controller allows the system to
optimize performance in the following cases:
— Relaxed ordering for descriptor and data reads: When the Ethernet controller masters a
read transaction its split completion has no relationship with the writes from the CPUs
(same direction). It should be allowed to bypass the writes from the CPUs.
— Relaxed ordering for receiving data writes: When the Ethernet controller masters receive
data writes it also allows them to bypass each other in the path to system memory because
the software does not process this data until their associated descriptor writes are done.

• The Ethernet controller cannot relax ordering for descriptor writes or an MSI write.
No Snoop Setting
The Ethernet controller always clears this bit in all of its master transactions because it cannot
guarantee that the memory locations between transaction addresses are not cached in the system.

4.4

Cache Line Information1
The cache line size PCI configuration register is programmed by the BIOS and/or OS after a
system reset. The value in the cache line size register corresponds to the cache line size that the
system supports.
The value programmed into the cache line size register affects the DMA operations of the Ethernet
controller. In general, the hardware attempts to fetch descriptors on a cache line basis. It also
attempts to write back descriptors when a cache line of descriptors has been filled.
The size of the cache line register also has an effect on the Ethernet controller’s usage of the MWI
PCI command, because the use of this command requires that at least a whole cache line of data is
written. The memory read commands are also affected as discussed in Section 4.3.2.
In PCI-X mode, the cache line size does not affect the commands used. However it does affect the
descriptor transfer. If an unsupported cache line size larger than 128 bytes is programmed, the
Ethernet controller acts as if a cache line size of 128 bytes was programmed.

Not applicable to the 82547GI/EI.

Software Developer’s Manual

PCI Local Bus Interface

4.4.1

Target Transaction Termination
When the Ethernet controller accepts a transaction as a target it always disconnects the transaction
after a single data phase by following the “Master Completion Termination” in PCI 2.2, 2.3, or
“Single data phase disconnect termination” in PCI-X. The “memory” in the Ethernet controller is
actually a set of registers and is marked as “non-prefetchable”. This is also the case for FLASH
memory.

4.5

Interrupt Assignment (82547GI/EI Only)
During a Power-On Self-Test (POST), the system BIOS must assign an Interrupt Request (IRQ) for
the 82547GI(EI). The 82547GI(EI) generates an interrupt by sending a hub interface message
through the CSA port.
In a typical system, the 82547GI(EI) component is Device 1 on the bus behind the CSA bridge
component. When the GMCH component receives an interrupt message, it forwards the interrupt
through a PIRQ programmed into the CSA Interface Interrupt Control Register (CSAINTC). Use
the following information to program this register:
Address Offset - 48h
Default Value - 04h
Access - R/W
Size - 8 bits
Bit Field - [2:1]
Description - PCI Interrupt type for CSA generated Interrupt - R/W
00
01
10
11

Reserved
INT#B
INT#C (default)
INT#D

Interrupts are communicated with a special CSA bus cycle that causes the MCH to issue an
interrupt to the interrupt controller in the ICH.

4.6

LAN Disable
For LAN designs, it is often desirable to program the BIOS setup to selectively enable or disable
LOM devices. This capability gives the end user more control over system resource management
and avoids conflicts with add-in boards.
Device presence or absence must be established early during BIOS execution to ensure that
resource allocation (interrupts and memory) is performed correctly. This task is frequently
accomplished using a BIOS CVDR (Configuration Values Driven on Reset) mechanism.
The 82541xx and 82547GI/EI LAN disable function resides on the FLSH_SO pin. This pin should
be driven by a port on the system Super IO device so that BIOS can control it dependably.

Software Developer’s Manual

PCI Local Bus Interface

4.7

CardBus Application (82541PI/GI/EI Only)
The 82541PI/GI/EI has some features to facilitate its use in a CardBus application, following
revision 7 of the PC Card specification.
To use the 82541PI/GI/EI on CardBus, an external flash memory is required. Configure the Base
Address Registers to 32-bit (required for CardBus) and enable CLKRUN in EEPROM. Setting
these bits also enables the CardBus Information Space (CIS) pointer in the PCI Configuration
Space. When enabled, the CIS pointer starts the first tuple of the CIS chain at byte address 20h in
the Flash device. The PC Card specification requires that this tuple be a CISTPL_LINKTARGET
tuple. The tuple chain can then continue within the memory.
When operating as a CardBus card, the 82541PI/GI/EI supports full functionality with the
exception of wake up.

Software Developer’s Manual

EEPROM Interface

EEPROM Interface
5.1

General Overview
The PCI/PCI-X Family of Gigabit Ethernet Controllers uses an EEPROM device for storing
product configuration information. The EEPROM is divided into four general regions:

• Hardware accessed – loaded by the Ethernet controller after power-up, PCI Reset
deassertion, D3->D0 transition, or software commanded EEPROM reset
(CTRL_EXT.EE_RST).

• ASF accessed – loaded by the Ethernet controller in ASF mode after power-up, ASF Soft
Reset (ASF FRC_RST), or software commanded ASF EEPROM read (ASF FRC_EELD).

• Software accessed – used by software only. The meaning of these registers as listed here is a
convention for the software only and is ignored by the Ethernet controller.

• External BMC (TCO) accessed – loaded by an external BMC (TCO) from the SMBus after
power up.
Note:

The 82544GC/EI and 82541ER do not support ASF, SMBus, or an external BMC (TCO).
Several words of the EEPROM are accessed automatically by the Ethernet controller after reset to
provide pre-boot configuration data before it is accessed by host software. The remainder of the
stored information is available to software for storing the MAC address, serial numbers, and
additional configuration information.
Intel has a software utility called EEUPDATE, which can be used to program EEPROM images in
development or production line environments. To obtain a copy of this program, contact your Intel
representative.

Note:

Since the 82546GB/EB is a dual port device, there are portions of the EEPROM and Flash that
control one or both ports. Special considerations due to this feature are noted in this section.

Software Developer’s Manual

EEPROM Interface

5.2

Component Identification Via Programming
Interface
Ethernet controller stepping is identified by the following register contents.
Table 5-1. Component Identification
Stepping

Vendor ID

Device ID

Description

82547EI-A0

8086h

1019h

Copper

82547EI-A1

8086h

1019h

Copper

82547EI-B0

8086h

1019h

Copper

82547EI-B0

8086h

101Ah

Mobile

82547GI-B0

8086h

1019h

Copper

82546EB-A1

8086h

1010h

82546EB-A1

8086h

1012h

Fiber; Dual Port

82546EB-A1

8086h

101Dh

Copper; Quad Port

82546GB-B0

8086h

1079h

Copper; Dual Port

82546GB-B0

8086h

107Ah

Fiber; Dual Port

82546GB-B0

8086h

107Bh

SerDes; Dual Port

82545EM-A

8086h

100Fh

Copper

82545EM-A

8086h

1011h

Fiber

82545GM-B

8086h

1026h

82545GM-B

8086h

1027h

Fiber

82545GM-B

8086h

1028h

SerDes

82544EI-A4

8086h

1107h

82544GC-A4

8086h

1112h

82541EI-A0

8086h

1013h

82541EI-A0

8086h

1013h

82541EI-B0

8086h

1013h

82541EI-B0

8086h

1018h

Copper; Dual Port
MAC Default

Copper
MAC Default

Copper
MAC Default
Copper
MAC Default
Cooper
MAC Default
Cooper
MAC Default
Cooper
MAC Default
Mobile

Software Developer’s Manual

EEPROM Interface

Table 5-1. Component Identification
Stepping

Note:

5.3

Vendor ID

Device ID

Description

82541GI-B1

8086h

1076h

Cooper

82541GI-B1

8086h

1077h

Mobile

82541PI-C0

8086h

1076h

Cooper

82541ER-C0

8086h

1078h

Cooper

82540EP-A

8086h

1017

Desktop

82540EP-A

8086h

1016

Mobile

82540EM-A

8086h

100E

Desktop

82540EM-A

8086h

1015

Mobile

These Ethernet controllers also provide identification data through the Test Access Port (TAP).

EEPROM Device and Interface
The EEPROM access algorithm, programmed into the Ethernet controller, is compatible with most,
but not all, commerically available 3.3 V dc Microwire* interfaces and serial EEPROM devices
with a 1 MHz speed rating. Ethernet controllers are compatible with two sizes of 4-wire serial
EEPROM devices1. If ASF mode functionality is desired, a 4096-bit serial NM93C66 compatible
EEPROM can be used. Otherwise, a 1024-bit serial NM93C46 compatible EEPROM can be used.
Both EEPROMs are accessed in 16-bit words; the larger has 256 words while the smaller has 64
words. Refer to the appropriate Ethernet controller’s design guide for recommended EEPROM
manufacturers.
An Ethernet controller automatically determines which EEPROM it is connected to and sets the
EEPROM SIZE field of the EEPROM/FLASH Control and Data Register (EEC.EE_SIZE) field
appropriately. Software can use this field to determine how to access the EEPROM using direct
access. Note that different EEPROM sizes have different numbers of address bits and therefore
must be accessed with a slightly different serial protocol. Software must be aware of this if it
accesses the EEPROM using direct access.

The 82544GC/EI only supports one size of EEPROM. Refer to the 82544GC Gigabit Ethernet Controller Datasheet and Hardware Design
Guide (AP-427) for more information.

Software Developer’s Manual

EEPROM Interface

The EEPROM interface trace routing is not critical because the interface runs at a very slow speed.
Note:

5.3.1

For the 82544GC/EI, 82540EP/EM, 82541xx, and 82547GI/EI, the EEPROM access algorithm
drives extra pulses on the shift clock at the beginnings and ends of read and write cycles. the extra
pulses might violate the timing specifications of some EEPROM devices. In selecting a serial
EEPROM, choose a device that specifies “don’t care” shift clock states between accesses.

Software Access
The Ethernet controller provides two different methods for software access to the EEPROM.
Software can either use the built-in controller to read the EEPROM, or access the EEPROM
directly using the EEPROM’s 4-wire interface.
Software can use the EEPROM Read register (EERD) to cause the Ethernet controller to read a
word from the EEPROM that the software can then use. To do this, software writes the address to
read the Read Address (EERD.ADDR) field and then simultaneously writes a 1b to the Start Read
bit (EERD.START). The Ethernet controller then reads the word from the EEPROM, sets the Read
Done bit (EERD.DONE), and puts the data in the Read Data field (EERD.DATA). Software can
poll the EEPROM Read register until it sees the EERD.DONE bit set, then use the data from the
EERD.DATA field. Any words read this way are not written to hardware’s internal registers.
Software can also directly access the EEPROM’s 4-wire interface through the EEPROM/FLASH
Control Register (EEC). It can use this for reads, writes, or other EEPROM operations.
To directly access the EEPROM, software should follow these steps:
1. Write a 1b to the EEPROM Request bit (EEC.EE_REQ).
2. Read the EEPROM Grant bit (EEC.EE_GNT) until it becomes 1b. It remains 0b as long as the
hardware is accessing the EEPROM.
3. Write or read the EEPROM using the direct access to the 4-wire interface as defined in the
EEPROM/FLASH Control & Data Register (EEC). The exact protocol used depends on the
EEPROM placed on the board and can be found in the appropriate data sheet.
4. Write a 0b to the EEPROM Request bit (EEC.EE_REQ).
Software can cause the Ethernet controller to re-read the hardware accessed fields of the EEPROM
(setting hardware’s internal registers appropriately) by writing a 1b to the EEPROM Reset bit of
the Extended Device Control Register (CTRL_EXT.EE_RST). This action will also cause a reset.

5.4

Signature and CRC Fields
The Ethernet controller uses the Signature and CRC fields to determine if an EEPROM is present
by attempting to read the EEPROM. The Ethernet controller first reads the Initialization Control
Word 1 at address 0Ah and then checks the received value for bits 15 and 14. If bit 15 is 0b and bit
14 is 1b, the Ethernet controller considers the EEPROM to be present and valid. It then reads the
additional EEPROM words and programs its internal registers based on the values read. Otherwise,
it ignores the values it read from the Initialization Control Word 1 and does not read any other
words.

Software Developer’s Manual

EEPROM Interface

In ASF Mode1, the Ethernet controller's ASF function reads the ASF CRC word to determine if the
EEPROM is valid. If the CRC is not valid, the ASF Configuration registers retain their default
value. This CRC does not affect any of the remaining Ethernet controller's configuration, including
the Management Control Register.

5.5

EEUPDATE Utility
The EEUPDATE utility meets the two basic requirements for an in-circuit programming utility.
First, the utility can be used to update EEPROM images as part of an end-of-line production tool.
Secondly, it can be used as a standalone development tool. The tool uses the two basic data files
outlined in the following section (static data file and IA address file). To obtain a copy of this
program, contact your Intel representative.
The EEUPDATE utility is flexible and can be used to update the entire EEPROM image or update
only the IA address of the Ethernet controller.

5.5.1

Command Line Parameters
The DOS command format is a follows:
EEUPDATE Parameter_1 Parameter_2
where:
Parameter_1 = filename or /D
Parameter_2 = filename or /A
Parameter 1, above, is file1.eep, which contains the complete EEPROM image in a
specific format that is used to update the complete EEPROM. All comments in the .eep file must
be preceded by a semicolon (;).
Parameter 1 can also be a switch /D. the switch /D implies: do not update the complete
EEPROM image.
Parameter 2, above, is file2.dat, which contains a list of IA addresses. the EEUPDATE
utility picks up the first unused address from this file and uses it to update the EEPROM. An
address is marked as used by following the address with a date stamp. When the utility uses a
specific address, it updates that address as used in a log file called eelog.dat. This file should then
be used as the .dat file for the next update.
Note:

Refer to the appropriate Ethernet controller’s EEPROM map and programming information for
sample EEPROM images.

The 82544GC/EI and 82541ER do not support ASF.

Software Developer’s Manual

EEPROM Interface

EEPROM Address Map1

5.6

Table 5-2 lists the EEPROM address map for the Ethernet controllers. Each word listed is
described in the sections that follow.
Note:

The “LAN A/B” column in Table 5-2 is only applicable to the 82546GB/EB.

Table 5-2. Ethernet Controller Address Map
Word

Used
By

Bit
15 - 8

Bit
7-0

Image
Value

LAN
A/B

00h

Ethernet Address Byte 2

Ethernet Address Byte 1

IA(2,1)

01h

Ethernet Address Byte 4

Ethernet Address Byte 3

IA(4,3)

02h

Ethernet Address Byte 6a

Ethernet Address Byte 5

IA(6,5)

LAN
A/B
(both)

Compatibility High

Compatibility Low

0000h

both

FFFFh

both

0000h

N/A

03h

SerDes Configuration
04h

05h

Note: Not applicable to the 82540EP/EM, 82541xx, and
82547GI/EI
EEPROM Image Version
Note: Applicable to the 82541xx and 82547GI/EI only

05h

Compatibility High

Compatibility Low

0000h

06h

(Words 06h and 07h reserved
for the 82541xx and 82547GI/
EI)

0000h

07h

(Words 06h and 07h reserved
for the 82541xx and 82547GI/
EI)

08h

PBA, byte 1

PBA, byte 2

09h

PBA, byte 3

PBA, byte 4

both

0000h

4408h

0Ah

Init Control 1

0Bh

Subsystem ID (Vendor)

0Ch

Subsystem Vendor ID

640Ah for
the
82541xx
and
82547GI/EI

both

see Table
5-1 for
specific
image
values

both

8086h

both

Refer to Table 5-3 for the 82544GC/EI and 82541ER EEPROM address map.

Software Developer’s Manual

EEPROM Interface

Table 5-2. Ethernet Controller Address Map
Word

Used
By

Bit
15 - 8

Bit
7-0

Image
Value

LAN
A/B

LAN A

both

0Dh

Device ID

see Table
5-1 for
specific
image
values

0Eh

Vendor ID

8086h
3040h for
the
82545GM/
EM and
82540EP/
EM

0Fh

both

Init Control 2
B080h for
the
82541xx
and
82547GI/EI

10h
11h

PHY Registers

00BAh

82541xx and 82547GI/EI only

0000h

N/A

Software Defined Pins Control (82546GB/EB only)
10h

11h

12h

13h
1Eh

14h

16h

Device ID
82546GB/EB only

XXXXh

LAN B

see Table
5-1 for
specific
image
values

LAN B

EEPROM Size

N/A

82541xx and 82547GI/EI only
Common Power

13h

15h

Note: Words 10h, 11h, and 13h through 1Fh are reserved for the
82545GM/EM and 82540EP/EM

both

PHY Registers

00BAh

82541xx and 82547GI/EI only

0000h

Management Control

LAN B

Init Control 3

SMBus Addressb

XXXXh

IPv4 Address Byte 2

IPv4 Address Byte 1

IP(2,1)

IPv4 Address Byte 4

IPv4 Address Byte 3

IP(4,3)

Software Developer’s Manual

N/A

LAN B

EEPROM Interface

Table 5-2. Ethernet Controller Address Map
Word

Used
By

Bit
15 - 8

Bit
7-0

Image
Value

17h

IPv6 Address Byte 2

IPv6 Address Byte 1

IP(2,1)

18h

IPv6 Address Byte 4

IPv6 Address Byte 3

IP(4,3)

19h

IPv6 Address Byte 6

IPv6 Address Byte 5

IP(6,5)

1Ah

IPv6 Address Byte 8

IPv6 Address Byte 7

IP(8,7)

IPv6 Address Byte 10

IPv6 Address Byte 9

IP(10,9)

1Ch

IPv6 Address Byte 12

IPv6 Address Byte 11

IP(12,11)

1Dh

IPv6 Address Byte 14

IPv6 Address Byte 13

IP(14,13)

1Eh

IPv6 Address Byte 16

IPv6 Address Byte 15

IP(16,15)

1Bh

CSA Port Config 1 (82547GI/EI only)

1Fh

Note: This word is reserved for all remaining Ethernet controllers

0000h

LAN
A/B

LAN B

N/A

XXXXh
20h

Software Defined Pins Control

21h

CSA Port Config 2 (82547GI/EI only)

000Ch for
the
82541xx
and
82547GI/EI
0002h

LAN A

N/A

7863h
21h

Circuit Control

7061h for
the
82540EP/
EM

both

280Ch
22h

D0 Power

D3 Power

280Bh for
the
82541xx
and
82547GI/EI

both

XXC8h
23h

Management Control

XXXXh for
the
82541xx
and
82547GI/EI

LAN A

XXXXh

24h

Init Control 3

SMBus Addressb

XXC8h for
the
82545GM/
EM and
82540EP/
EM

LAN A

001Ch for
the
82541xx
and
82547GI/EI

100

Software Developer’s Manual

EEPROM Interface

Table 5-2. Ethernet Controller Address Map
Word
25h

Used
By

Bit
15 - 8

Bit
7-0

Image
Value

IPv4 Address Byte 2

IPv4 Address Byte 1

IP(2,1)

IPv4 Address Byte 4

IPv4 Address Byte 3

IP(4,3)

27h

IPv6 Address Byte 2

IPv6 Address Byte 1

IP(2,1)

28h

IPv6 Address Byte 4

IPv6 Address Byte 3

IP(4,3)

29h

IPv6 Address Byte 6

IPv6 Address Byte 5

IP(6,5)

2Ah

IPv6 Address Byte 8

IPv6 Address Byte 7

IP(8,7)

IPv6 Address Byte 10

IPv6 Address Byte 9

IP(10,9)

2Ch

IPv6 Address Byte 12

IPv6 Address Byte 11

IP(12,11)

2Dh

IPv6 Address Byte 14

IPv6 Address Byte 13

IP(14,13)

2Eh

IPv6 Address Byte 16

IPv6 Address Byte 15

IP(16,15)

26h

2Bh

2Fh

LEDCTL Default

0602h

LAN
A/B
LAN A

LAN A

both

Intel Boot Agent Configuration
30h
31h
32h
33h

Firmware

34h

…
Note: Words 34h and 35h are not applicable to the 82545GM/EM
and 82540EP/EM

...
3Eh

3Fh

Software Checksum, words 00h through 3Fh

40h
...

ASF

Controlled by the ASF Agent

F7h
F8h
...

Free for Software

FFh
a.
b.

The lower bit of the last byte is complemented for LAN B.
The SMBus Address is a 7-bit value that is found in bits 7 through 1 of this byte. Bit 0 should be 0b.

Software Developer’s Manual

101

EEPROM Interface

Table 5-3. 82544GC/EI and 82541ER EEPROM Address Map

Word
Address

Description
(Hi Byte)

Description
(Low Byte)

Default
Image Value
(hex)

00h

Yes

IA Byte 2

IA Byte 1

IA(2,1)

01h

Yes

IA Byte 4

IA Byte 3

IA(4,3)

02h

Yes

IA Byte 6

IA Byte 5

IA(6,5)

03h

Compatibility high

Compatibility low

0000h

04h

05h

06h
07h
08h

102

HW Access

Reserved
EEPROM Image Version
Note: Word 05h is reserved for the 82544GC/EI

No
No

0000h

Reserved
PBA, byte 1

0000h
0000h

PBA, byte 2

09h

PBA, byte 3

PBA, byte 4

0Ah

Yes

Init Control 1, high byte

Init Control 1, low byte

See Text

0Bh

Yes

Subsystem_ID,
high byte

Subsystem_ID,
low byte

1005h

0Ch

Yes

Subsystem_Vendor,

high byte

low byte

0Dh

Yes

Device ID, high

Device ID, low

0Eh

Yes

Vendor ID, high

Vendor ID, low

8086h

0Fh

Yes

Init Control 2, high byte

Init Control 2, low byte

See Text

10h - 1Fh

OEM Reserved

0000h

20h

Yes

Software Defined Pins
Control, high byte

Software Defined Pins
Control, low byte

See Text

21h

Yes

Circuit Control, high

Circuit Control, low

0021h

8086h
1008h

22h

Yes

D0 Power

D3 Power

See Text

23h - 2Eh

Reserved

0000h

2Fh

Yes

30h - 33h

Firmware

34h - 3Eh

Fixed

Reserved

0000h

3Fh

Checksum, high byte

Checksum, low byte

Checksum of
words 00h - 3Eh

LEDCTRL Default
Note: Word 2Fh is reserved for the 82544GC/EI
Intel Boot Agent
Note: Words 30 - 33h are reserved for the 82541ER

0602h
0000h

Software Developer’s Manual

EEPROM Interface

5.6.1

Ethernet Address (Words 00h-02h)
The Ethernet Individual Address (IA) is a six-byte field that must be unique for each Ethernet port
(and unique for each copy of the EEPROM image). The first three bytes are vendor specific. The
value from this field is loaded into the Receive Address Register 0 (RAL0/RAH0). For a MAC
address of 12-34-56-78-90-AB, words 2:0 load as follows (note that these words are byteswapped):
Word 0 = 3412
Word 1 = 7856
Word 2 - AB90
Note:

5.6.2

Since the 82546GB/EB is a dual-port device, the Ethernet Address in these words are assigned to
LAN A. The Ethernet Address for LAN B is the Ethernet Address for LAN A with its least
significant bit inverted.

Software Compatibility Word (Word 03h)
This is the third word read by the Ethernet controller and contains additional initialization values
that:

• Sets defaults for some internal registers
• Enables/disables specific features
Note:

For the 82544GC/EI, typical values are 0000h for a fiber-based design and 0400h for a copperbased design.

Table 5-4. Software Compatibility Word (Word 03h)
Bit

Name

Description

15:12

Reserved

Reserved for future use.

LOM

LAN on Motherboard (LOM#). Set this bit to 1b (default) to enable LOM#;
set to 0b to disable LOM#.

SRV

Server card. Set this bit to 1b (default) to enable server card; set to 0b to
disable server card.
0b is the default setting for the 82541xx and 82547GI/EI.

CLI

Client card. Set this bit to 0b (default) to disable client card; set to 1b to
enable client card.
1b is the default setting for the 82541xx and 82547GI/EI.

OEM

OEM card. Set this bit to 1b (default) to enable OEM card; set to 0b to
disable OEM card.

7:6

Reserved

Reserved for future use. Set these bits to 0b.

Reserved

Reserved for future use.
Set this bit to 1b.
Set this bit to 0b for the 82540EP/EM, 82541xx and 82547GI/EI.
4

SMBa

Software Developer’s Manual

SMBus. Set this bit to 1b (default) to enable SMBus; set to 0b to disable
SMBus.

103

EEPROM Interface

Table 5-4. Software Compatibility Word (Word 03h)
3

Reserved

Reserved for future use. Set this bit to 0b.

BOB

PCI bridge. Set this bit to 0b (default) to disable PCI bridge; set to 1b to
enable PCI bridge.
1b is the default setting for the 82540EP/EM.

1:0
a.

5.6.3

Reserved

Reserved for future use. Set these bits to 0b.

Not applicable to the 82544GC/EI or 82541ER.

SerDes Configuration (Word 04h)
If this word has a value of other than FFFFh, software programs its value into the Extended PHY
Specific Control Register 2, located at address 26d in the PHY register space (see Table 13-47).
Note:

5.6.4

SerDes Configuration (Word 04h) is a reserved area for the 82544GC/EI, 82540EP/EM, 82541xx,
and 82547GI/EI.

EEPROM Image Version (Word 05h)
Word 05h determines the EEPROM image version for the 82541xx and the 82547GI/EI.

Bits

5.6.5

Name

Value

15:12

EEPROM major version.

0000h

11:8

EEPROM minor version.

0000h

7:0

EEPROM fix.

00000000h

Compatibility Fields (Word 05h - 07h)
These areas are reserved for compatibility information and are used by software drivers.

5.6.6

PBA Number (Word 08h, 09h)
A nine-digit Printed Board Assembly (PBA) number, used for Intel manufactured adapter cards,
are stored in a four-byte field. Other hardware manufacturers can use these fields as they wish.
Software device drivers should not rely on this field to identify the product or its capabilities.

104

Software Developer’s Manual

EEPROM Interface

5.6.7

Initialization Control Word 1 (Word 0Ah)
The first word read by the Ethernet controller contains initialization values that:

• Sets defaults for some internal registers
• Enables/disables specific features
• Determines which PCI configuration space values are loaded from the EEPROM
Table 5-5. Initialization Control Word 1 (Word 0Ah)
Bit

Name

Description
The Signature field represents a signature of 01b (default), indicating to the
Ethernet controller that there is a valid EEPROM present.

15:14

Signature

64/32 BAR

If the Signature field is not 01b, the other bits in this word are ignored, no
further EEPROM read is performed, and default values are used for the
configuration space IDs.
When set to 0b (default), enables 64-bit memory mapping.
When set to 1b, disables 64-bit memory mapping
Note: Set to 1b for the 82540EP/EM.
When set to 0b (default), does not invert the Power State Output bit 0
(CTRL_EXT[14]).

IPS0

When set to 1b, inverts the Power State Output invert bit 0
(CTRL_EXT[14]).
Note: Reserved bit for the 82541xx and 82547GI/EI (set to 0b).
Force Speed bit in the Device Control register (CTRL[11]).
When set to 0b (default), does not force speed.

FRCSPD

When set to 1b, forces speed (default for the 82540EP/EM).
For 10/100/1000 Mb/s systems using TBI mode (82544GC/EI)/internal
SerDes (82546GB/EB and 82545GM/EM)a, set this bit to 0b.
Note: Reserved bit for the 82541xx and 82547GI/EI (set to 0b).
Full Duplex (mapped to CTRL[0] and TXCW[5]).
When set to 1b (default), enables full duplex (TBI mode/internal SerDes
only).

When set to 0b, disables full duplex (TBI mode only/internal SerDes).
Note: Reserved bit for the 82541PI/GI/EI and 82547GI/EI (set to 1b).
Note: Reserved bit for the 82541ER (set to 0b).
Link Reset (mapped to CTRL[3]).

LRST

When set to 0b, enables Auto-Negotiation at power up or when asserting
RST# without driver intervention.
When set to 1b, disables Auto-Negotiation at power up or when asserting
RST# without driver intervention.
Note: Reserved bit for the 82541xx and 82547GI/EI (set to 0b).
When set to 0b (default), does not invert the Power State Output bit 1
(CTRL_EXT[16]).

IPS1

When set to 1b, inverts the Power State Output invert bit 1
(CTRL_EXT[16]).
Note: Reserved bit for the 82541xx and 82547GI/EI (set to 0b).

Software Developer’s Manual

105

EEPROM Interface

Table 5-5. Initialization Control Word 1 (Word 0Ah)
82541xx and 82547GI/EI Only
Internal VREG Power
Down Control

This bit is used to define the usage of the internal 1.2 V dc and 1.8 V dc
regulators to supply power.
0b = Yes (default).
1b = No (external regulators used).
Note: Reserved bit for all other Ethernet controllers.

Bit

Name

Description

6:5

Reserved

Reserved for future use. Set these bits to 0b.

Reserved

Reserved for copper PHY. Set this bit to 0b.
When set to 1b (default), enables full support for power management.

Power Management

When set to 0b, the Power Management Registers set is read only. The
Ethernet controller does not execute a hardware transition to D3.
Note: Reserved bit for the 82541PI/GI/EI and 82547GI/EI (set to 1b).
Note: Reserved bit for the 82541ER (set to 0b).
When set to 0b (default), indicates that the PCICLK is not required for
PME# output.

PME Clock

When set to 1b, indicates that the PCICLK is required for PME# output.
Note: Reserved bit for the 82541xx and 82547GI/EI (set to 0b).

Load Subsystem IDs

When set to 1b (default), indicates that the Ethernet controller is to load its
PCI Subsystem ID and Subsystem Vendor ID from the EEPROM (words
0Bh, 0Ch).
When set to 0b, indicates that the Ethernet controller is to load the default
PCI Subsystem ID and Subsystem Vendor ID.

5.6.8

Load Vendor/Device
IDs

When set to 0b (default), indicates that the Ethernet controller is to load the
default values for PCI Vendor and Device IDs.
When set to 1b (default for the 82541xx and 82547GI/EI only), indicates
that the Ethernet controller is to load its PCI Vendor and Device IDs from
the EEPROM (words 0Dh, 0Eh).

Not applicable to the 82541xx, 82547GI/EI or 82540EP/EM.

Subsystem ID (Word 0Bh)
If the signature bits (15:14) and bit 1 (Load Subsystem IDs) of word 0Ah are valid, this word is
read in to initialize the Subsystem ID.

5.6.9

Subsystem Vendor ID (Word 0Ch)
If the signature bits (15:14) and bit 1 (Load Subsystem IDs) of word 0Ah are valid, this word is
read in to initialize the Subsystem Vendor ID.

106

Software Developer’s Manual

EEPROM Interface

5.6.10

Device ID (Word 0Dh, 11h1)
If the signature bits (15:14) and bit 1 (Load Subsystem IDs) of word 0Ah are valid, this word is
read in to initialize the Subsystem ID.
For the 82546GB, the Device ID must be forced to 107Bh for SerDes-SerDes interface operation.
For the 82545GM, the Device ID should be 1028h. This ensures proper functionality with Intel
drivers and boot agent.

Note:

5.6.11

Since the 82546GB/EB is a dual-port device, the Device ID in 0Dh corresponds to LAN A and the
Device ID in 11h corresponds to LAN B.

Vendor ID (Word 0Eh)
If the signature bits (15:14) and bit 1 (Load Subsystem IDs) of word 0Ah are valid, this word is
read in to initialize the Subsystem ID.

5.6.12

Initialization Control Word 2 (Word 0Fh)
This is the second word read by the Ethernet controller and contains additional initialization values
that:

• Sets defaults for some internal registers
• Enables/disables specific features
Table 5-6. Initialization Control Word 2 (Word 0Fh)
Bit

Name

Description
Initial value of the Assert PME On APM Wakeup bit in the Wakeup Control
Register (WUC.APMPME).

APM PME# Enable

When set to 0b (default), deasserts PME# on wakeup.
Set this bit to 1b for Intel LAN controller cards.

ASDE

When set to 0b (default), indicates the initial value of the Auto-Speed
Detection Enable (ASDE) bit of the Device Control Register (CTRL).
When set to 1b, enables 10/100/1000 Mb/s systems.
Pause Capability - Mapped to TXCW[8:7].

13:12

PAUSE Capability

When set to 1b (default), enables desired PAUSE capability for an
advertised configuration base page.
When set to 0b, disables desired PAUSE capability for an advertised
configuration base page.

10:9

ANE

Auto-Negotiation Enable. Mapped to TXCW[31].

Reserved bit for the
82541xx and
82547GI/EI (set to
0b).

Set this bit to 1b to automatically enable Auto-Negotiation.

FLASH Size Indication

Set this bit to 0b (default) to automatically disable Auto-Negotiation.
Note: Fiber implementations do not support this function.
Indicates FLASH size. 00b = 64 KB (default); 01b = 128 KB; 10b = 256 KB;
11b = 512 KB.
These bits also impact the requested memory space for the FLASH and
Expansion ROM BARs in PCI configuration space.

Word 11h only applicable to the 82546GB/EB.

Software Developer’s Manual

107

EEPROM Interface

Table 5-6. Initialization Control Word 2 (Word 0Fh)
Bit

Name

Description
82541PI/GI Only.
0b = MAC runs at full speed.

MAC Clock Speed

1b = MAC runs at 1/4 speed on any drop from 1000 Mb/s.
Note: Reserved bit for all other Ethernet controllers (set to 0b). Formally
FLASH Disable, now located in Initialization Control Word 3, bit 3.
When set to 0b (default), enables Message Signalled Interrupts (MSI) in
standard PCI mode.

MSI Disable

When set to 1b, disables Message Signalled Interrupts (MSI) in standard
PCI mode.
Note: Reserved bit for the 82541xx and 82547GI/EI (set to 1b).
When set to 1b (default), maps the 133 MHz Capable bit of the PCI-X
Status Register (PCIXS).

133 MHz Capable

When set to 0b, does not map the 133 MHz Capable bit of the PCI-X Status
Register (PCIXS).
Note: Reserved bit for the 82541xx, 82547GI/EI, and 82540EP/EM (set to
0b).
Indicates how the Designed Maximum Cumulative Read size bits in the
PCI-X Status register are mapped.

DMCR_Map

When set to 1b (default), the DMCR value reflects the hard-coded design
capability as indicated by the Max_Read bit (bit 4).
When set to 0b, the DMCR is mapped directly to the Maximum Memory
Read Byte Count indicated in the PCI-X Command register.
Note: Reserved bit for the 82541xx, 82547GI/EI, and 82540EP/EM (set to
0b).
Indicates the maximum read value as advertised in the Designed Maximum
Memory Read Byte Count field in the PCI-X Status Register.
When set to 0b (default), or if there is no EEPROM, the advertised
maximum read is 2 KB.

Max_Read

When set to 1b, the advertised maximum read is 4 KB. Note that it is not
recommended to set Max_Read to 1b because transmit FIFO overruns are
possible under specific operating conditions.
Note: Reserved bit for the 82541xx, 82547GI/EI, and 82540EP/EM (set to
0b).
When set to 1b (default), loads the 64-bit Device field of the PCI-X Status
Register.

64-bit

When set to 0b, does not load the 64-bit Device field of the PCI-X Status
Register.
Note: Reserved bit for the 82541xx, 82547GI/EI, and 82540EP/EM (set to
0b).

Reserved

Reserved for future use (set to 0b). Formerly APM Enable, now located in
Initialization Control Word 3, bit 2.
Note: Set to 1b for the 82541xx and 82547GI/EI.
When set to 0b (default), certain critical registers are decoded for non-split
access.

Force CSR Read Split

When set to 1b, forces all Ethernet controller control/status register-reads
to be split when operating in a PCI-X environment.
Note: Reserved bit for the 82541xx and 82547GI/EI (set to 0b).

108

Reserved

Reserved for future use (set to 0b).

Software Developer’s Manual

EEPROM Interface

5.6.13

PHY Register Address Data (Words 10h, 11h, and 13h - 1Eh)
These settings are specific to individual platform configurations for the 82541xx and 82547GI/EI
and should not be altered from the reference design unless instructed to do so. Future Intel Ethernet
controllers might use this space differently.

5.6.14

OEM Reserved Words (Words 10h, 11h, 13h - 1Fh)
Words 10h, 11h, and 13h through 1Fh of the EEPROM are reserved areas for general OEM use for
all Ethernet controllers except the 82546GB/EB.

5.6.15

EEPROM Size (Word 12h)
This word is only applicable to 82541xx and 82547GI/EI Ethernet controllers that use SPI
EEPROMs. Unused bits are reserved and should be programmed to 0b. Bits 15:13 and 8:0 are
reserved (see Table 5-7).
Table 5-7. SPI EEPROM Sizes
Bits 12:10

5.6.16

Bit 9

EEPROM Size (Bits)

EEPROM Size (Bytes)

000

1 Kb

128 Bytes

001

4 Kb

512 Bytes

010

8 Kb

1 KB

011

16 Kb

2 KB

100

32 Kb

4 KB

101

64 Kb

8 KB

110

128 Kb

16 KB

111

Reserved

Common Power (Word 12h)
For all Ethernet controllers except the 82541xx and 82547GI/EI, if the signature bits are valid and
Power Management is not disabled, the value in this field is used in the PCI Power Management
Data Register when the Data_Select field of the Power Management Control/Status Register
(PMCSR) is set to 8. This setting indicates the power usage and heat dissipation of the common
logic that is shared by both functions in tenths of a watt.

5.6.17

Software Defined Pins Control (Word 10h1, 20h)
This field contains initial settings for the Software Defined Pins (SPD). The default value for the
upper byte (bits 15:8) is DFh; the default value for the lower byte (bits 7:0) is DEh.

Applicable to the 82546GB/EB only.

Software Developer’s Manual

109

EEPROM Interface

Table 5-8. Software Defined Pins Control (Word 10h, 20h)
Bit

Name

Description
SDP7(3) Pin - Initial Direction.

SDPDIR[7]
15

SDPDIR[3] for the 82541xx
and 82547GI/EI

Set this bit to 0b (default) to configure the initial hardware value of
the SDP7(3)_IODIR bit in the Extended Device Control Register
(CTRL_EXT) following power up.
Set this bit to 1b if not connected on a board or if used as an
output.
SDP6(2) Pin - Initial Direction.

SDPDIR[6]
14

13:10

SDPDIR[2] for the 82541xx
and 82547GI/EI

Reserved

Set this bit to 0b (default) to configure the initial hardware value of
the SDP6(2)_IODIR bit in the Extended Device Control Register
(CTRL_EXT) following power up.
Set this bit to 1b if not connected on a board or if used as an
output.
Set these bits to 0b.
SDP1 Pin - Initial Direction.

SDPDIR[1]

Set this bit to 0b (default) to configure the initial hardware value of
the SDP1_IODIR bit in the Extended Device Control Register
(CTRL_EXT) following power up.
Set this bit to 1b if not connected on a board or if used as an
output.
SDP0 Pin - Initial Direction.

SDPDIR[0]

Set this bit to 0b (default) to configure the initial hardware value of
the SDP0_IODIR bit in the Extended Device Control Register
(CTRL_EXT) following power up.
Set this bit to 1b if not connected on a board or if used as an
output.
SDP7(3) Pin - Initial Output Value.

SDPVAL[7]
7

SDPVAL[3] for the 82541xx
and 82547GI/EI

Set this bit to 0b (default) to configure the initial power-on value
output on SDP7(3) (when configured as an output) by configuring
the initial hardware value of the SDP7(3)_DATA bit in the Extended
Device Control Register (CTRL_EXT) after power up.
Set this bit to 1b if used as an output.
SDP6(2) Pin - Initial Output Value.

SDPVAL[2] for the 82541xx
and 82547GI/EI

Set this bit to 0b (default) to configure the initial power-on value
output on SDP6(2) (when configured as an output) by configuring
the initial hardware value of the SDP6(2)_DATA bit in the Extended
Device Control Register (CTRL_EXT) after power up.

5:4

Reserved

Set these bits to 0b.

SDPVAL[6]

Set this bit to 1b if used as an output.

Set this bit to 1b (default) to configure the initial hardware default
value of this bit in the Device Control Register (CTRL) following
power up.
3

110

EN_PHY_PWR_MGMT

Set this bit to 0b to not configure the initial hardware default value
of this bit in the Device Control Register (CTRL) following power
up.

Software Developer’s Manual

EEPROM Interface

Table 5-8. Software Defined Pins Control (Word 10h, 20h)
Bit

Name

Description

D3COLD_WAKEUP_ADV_EN

Set this bit to 1b (default) to configure the initial hardware default
value of the ADVD3WUC bit in the Device Control Register (CTRL)
following power up.
Set this bit to 0b to not configure the initial hardware default value
of the ADVD3WUC bit in the Device Control Register (CTRL)
following power up.
SDP1 Pin - Initial Output Value.

Set this bit to 0b (default) to configure the initial power-on value
output on SDP1 (when configured as an output) by configuring the
initial hardware value of the SDP1_DATA bit in the Extended
Device Control Register (CTRL_EXT) after power up.

SDPVAL[1]

Set this bit to 1b if used as an output.
SDP0 Pin - Initial Output Value.
0

Set this bit to 0b (default) to configure the initial power-on value
output on SDP0 (when configured as an output) by configuring the
initial hardware value of the SDP0_DATA bit in the Extended
Device Control Register (CTRL_EXT) after power up.

SDPVAL[0]

Set this bit to 1b if used as an output.

Note:

5.6.18

Since the 82546GB/EB is a dual-port device, the SDP control in 10h corresponds to LAN B, and
the SDP control in 20h corresponds to LAN A.

CSA Port Configuration 2 (Word 21h)
For the 82547GI/EI only, this word controls the CSA port configuration and must be programmed
to 93A7h for regular operation (see Table 5-9).
Table 5-9. CSA Port Configuration 2 (Word 21h)
Bit

Description

Default

15:13

Reserved.

Set to 100b.

Reserved.

Set to 1b.

11:2

Reserved.

Set to 0011101001b.

Dock/Undock Polarity.

Reserved.

Software Developer’s Manual

1b = Indicates Docked (default).
0b = Indicates Undocked.
Set to 1b.

111

EEPROM Interface

5.6.19

Circuit Control (Word 21h)
This word is loaded into the Circuit Control Register (CIRC) for setting PCI-X driver strength. See
Table 5-2 and Table 5-3 for suggested values.

Note:

5.6.20

PCI-X is not applicable to the 82540EP/EM, 82541xx, and 82547GI/EI.

D0 Power (Word 22h high byte)
If the signature bits are valid and Power Management is not disabled, the value in this field is used
in the PCI Power Management Data Register when the Data_Select field of the Power
Management Control/Status Register (PMCSR) is set to 0 or 4. This indicates the power usage and
heat dissipation of the networking function (including the Ethernet controller and any other devices
managed by the Ethernet controller) in tenths of a watt. For example:
If word22h = 290E, POWER CONSUMPTION (in 1/10W, hex), then:
bits 15:8 = 29h Power in D0a, 29h = 4.1W
bits 7:0 = 0Eh Power in D3h, 0Eh = 1.4W

5.6.21

D3 Power (Word 22h low byte)
If the signature bits are valid and Power Management is not disabled, the value in this field is used
in the PCI Power Management Data Register when the Data_Select field of the Power
Management Control/Status Register (PMCSR) is set to 3 or 7. This indicates the power usage and
heat dissipation of the networking function (including the Ethernet controller and any other devices
managed by the Ethernet controller) in tenths of a watt as described in Section 5.6.20.

5.6.22

Reserved Words (23h - 2Eh)
Words 23h through 2Eh of the EEPROM are reserved areas for the 82541ER.

5.6.23

Reserved Words (23h - 2Fh)
Words 23h through 2Fh of the EEPROM are reserved areas for the 82544GC/EI.

112

Software Developer’s Manual

EEPROM Interface

5.6.24

Management Control (Word 13h1, 23h2)
The following table lists the initial settings for the Management Control Register as well as valid
bits for the IPv4 Address and the IPv6 Address.
Table 5-10. Initial Management Control Register Settings
Bit

Name

Description
This bit controls the initial value of the MANC.RSP_EN bit.

Enable ARP Response
Filtering

0b: ARP response packets are delivered to host memory.

Reserved

Reserved. Set this bit to 0b.

1b: ARP response packets are delivered to the Ethernet controller
for automatic ARP reply or forwarded to the BMC.

This bit controls the initial value of the MANC.ARP_EN bit.
13

Enable ARP Request Filtering

0b: ARP request packets are delivered to host memory.
1b: ARP request packets are delivered to the Ethernet controller for
automatic ARP reply or forwarded to the BMC.

12:10

Reserved

Set these bits to 0b.
This bit is controlled by the ASF agent. Manually set this bit for
TCO mode.

Enable RMCP 0298h Filtering

Set this bit to 0b (default) to control the initial value of the
MANC.0298_EN bit, which enables classifying UDP packets of port
0298h as Management Packets for delivery to the SMBus or ASF
controller.
Set this bit to 0b if the SMBus is disabled.
This bit is controlled by the ASF agent. Manually set this bit for
TCO mode.

Enable RMCP 026F Filtering

Set this bit to 1b (default) to control the initial value of the
MANC.0298_EN bit, which enables classifying UDP packets of port
026Fh as Management Packets for delivery to the SMBus or ASF
controller.
Set this bit to 0b if the SMBus is disabled.

IPv6 Address Valid

IPv6 Address in the IP Address EEPROM register is valid. This is
written to bit 16 of the IP Address Valid (IPAV[16]) register.

IPv4 Address Valid

IPv4 Address in the IP Address EEPROM register is valid. This is
written to bit 0b of the IP Address Valid (IPAV[0]) register.
82541PI/GI/EI and 82547GI/EI Only.
This bit enables the flexible filter loaded from the EEPROM.

Flex Filter Enable

0b = Disable flex filter.
1b = Enable flex filter.
Note: Reserved bit for all other Ethernet controllers (set to 0b).

4:3

1.
2.

Reserved

Set these bits to 0b.

Applicable to the 82546GB/EB only.
Not applicable to the 82544GC/EI or 82541ER.

Software Developer’s Manual

113

EEPROM Interface

Table 5-10. Initial Management Control Register Settings
Bit

Name

Description
This bit is controlled by the ASF agent. Manually set this bit for
TCO mode.

Reset on Force TCO

Reset the Ethernet controller on a ForceTCO SMBus Command
with the “Force” bit set to 1b (default) in TCO mode, or on various
conditions in ASF mode.
Set this bit to 0b if the SMBus is disabled.
This bit is controlled by the ASF agent.

ASF Mode

Set this bit to 1b to enable ASF mode.
Set this bit to 0b (default) for TCO mode and Intel’s ASF Agent
implementation.
This bit is controlled by the ASF agent. Manually set this bit for
TCO mode.

SMBus Enable

Set this bit to 1b (default) to enable SMBus functionality.
Set this bit to 0b to enable ASF mode if its being routed by LAN A
and if LAN A is the active interface for ASF to the BIOS.

Note:

5.6.25

Since the 82546GB/EB is a dual-port device, the Management Control in 13h corresponds to LAN
B, and the Management Control in 23h corresponds to LAN A.

SMBus Slave Address (Word 14h1 low byte, 24h low byte)
The following table lists the SMBus slave address for TCO mode.
Table 5-11. SMBus Slave Address
Bit

Name

7:1

SMBus Slave Address

Reserved

Description
These bits are controlled by the ASF agent. Manually set these bits for
TCO mode.
Contains the SMBus slave address for TCO mode. This address must
be 1100b 100b for ASF mode.
Set this bit to 0b.

Note:

This byte must be C8h for ASF mode. For example, to program an address of 0011_001b, the byte
should be set to 0011_0010b. When this address is used on the SMBus, the address byte is
0011_0010b for writes and 0011_0011b for reads.

Note:

Since the 82546GB/EB is a dual-port device, the SMBus Slave Address in 14h corresponds to
LAN B, and the SMBus Slave Address in 24h corresponds to LAN A.

Applicable to the 82546GB/EB only.

114

Software Developer’s Manual

EEPROM Interface

5.6.26

Initialization Control 3 (Word 14h1 high byte, 24h high byte)
This word controls the general initialization values.
Table 5-12. Initialization Control 3
Bit
7:5

Name
Reserved

Description
Reserved. Set these bits to 0b.
Controls the value advertised in the Interrupt Pin field of the PCI
Configuration header for this device/function.

Interrupt Pin

A value of 0b (default), reflected in the Interrupt Pin field, indicates that
the 82546GB/EB uses INTA#; a value of 1b indicates that the
82546GB/EB uses INTB#.
If only a single device/function of the Ethernet controller is enabled, this
value is ignored and the Interrupt Pin field of the enabled device reports
INTA# usage.
Set this bit to 0b (default) to enable the FLASH logic.

FLASH Disable

APM Enable

Set this bit to 1b to disable the FLASH logic. Note that the Expansion
ROM & secondary FLASH access BARs in PCI configuration space are
also disabled.
Initial value of Advanced Power Management Wakeup Enable in the
Wakeup Control Register (WUC.APME).
The default for this bit is 0b.
Initial value of Link Mode bits of the Extended Device Control Register
(CTRL_EXT.LINK_MODE), specifying which link interface and protocol
is used by the MAC.
For Address 24h (High Byte) / LAN A
00b = MAC operates in GMII/MII mode with internal copper PHYa
01b = External GMII/MII mode

1:0

Link Mode

10b = Internal SerDes mode (not applicable to the 82540EP/EM)
11b = MAC operates in TBI mode using external TBI interface
For Address 14h (High Byte) / LAN B
00b = MAC operates in GMII/MII mode with internal copper PHY
01b = Reserved
10b = Internal SerDes mode (not applicable to the 82540EP/EM)
11b = MAC operates in TBI mode using external TBI interface

Note:

For the 82540EP/EM, 82541PI/GI/EI, and 82547GI/EI to properly communicate with the internal copper PHY, this value
must be set to 00b.

Since the 82546GB/EB is a dual-port device, the Initialization Control Word 3 bit assignments are
port specific.

Applicable to the 82546GB/EB only.

Software Developer’s Manual

115

EEPROM Interface

5.6.27

IPv4 Address (Words 15h - 16h1 and 25h - 26h)
The following table lists the initial values for the IPv4 addresses.
Table 5-13. IPv4 Addresses
Bit

Name

Description
The initial value of IPv4 Address Table entry 0. (IP4AT[0]).

31:0

Note:

5.6.28

IPv4 Address

Refer to the EEPROM Address Map listed in Table 5-2 for an indication
of how the bytes are stored.

Since the 82546GB/EB is a dual-port device, the IPv4 Address in 15h-16h corresponds to LAN B,
and the IPv4 Address in 25h-26h corresponds to LAN A.

IPv6 Address (words 17h - 1Eh1 and 27h - 2Eh)
The following table lists the initial values for the IPv6 addresses.
Table 5-14. IPv6 Address
Bit

Name

Description
The initial value of IPv6 Address Table entry 0. (IP6AT[0])

127:0

Note:

IPv6 Address

Refer to the EEPROM Address Map listed in Table 5-2 for an indication
of how the bytes are stored.

Since the 82546GB/EB is a dual-port device, the IPv6 Address in 17h-1Eh corresponds to LAN B,
and the IPv6 Address in 27h-2Eh corresponds to LAN A.

Applicable to the 82546GB/EB only.

116

Software Developer’s Manual

EEPROM Interface

5.6.29

LED Configuration Defaults (Word 2Fh)1
This EEPROM word specifies the hardware defaults for the LEDCTL register fields controlling the
LED0/(LINK_UP#) and LED2/LINK100 output behaviors. Refer to Table 13-60 for the LED
Control bit descriptions and Table 13-61 for the Mode Encodings.

Note:

A value of 0602h is used to configure default hardware LED behavior equivalent to 82544-based
Copper adapters (LED0/LINK_UP#, LED1/ACTIVITY# (blinking), LED2/LINK100#, and
LED3/LINK1000#).
Table 5-15. LED Configuration Defaults

Bit

5.6.30

Name

Description
Initial value of the LED0_MODE field specifying what event/state/
pattern will be displayed on LED0 (LINK_UP) output. A value of
0010 (0x2) causes this to indicate LINK_UP state.

3:0

LED0 Mode

5:4

Reserved

LED0 Invert

LED0 Blink

11:8

LED2 Mode

Initial value of the LED2_MODE field specifying what event/state/
pattern will be displayed on LED2 (LINK_100) output. A value of
0110b (0x6) causes this to indicate 100 Mb/s operation.

13:12

Reserved

Reserved. Set as 0b.

LED2 Invert

LED2 Blink

Reserved.
Set as 0b.
Initial value of LED0_IVRT field.
0b = Active-low output.
Initial value of LED0_BLINK field.
0b = Non-blinking.

Initial value of LED2_IVRT field.
0b = Active-low output.
Initial value of.LED2_BLINK field.
0b = Non-blinking

Boot Agent Main Setup Options (Word 30h)
The boot agent software configuration is controlled by the EEPROM with the main setup options
stored in word 30h. These options are those that can be changed by using the Control-S setup menu
or by using the IBA Intel Boot Agent utility.

Note:

The 82541ER does not support the Intel Boot Agent functionality.

Not applicable to the 82544GC/EI.

Software Developer’s Manual

117

EEPROM Interface

Table 5-16. Boot Agent Main Setup Options
Bit

Name

Description
PXE Presence.
Setting this bit to 0b Indicates that the image in the FLASH contains a
PXE image.
Setting this bit to 1b indicates that no PXE image is contained.

PPB

The default for this bit is 0b in order to be backwards compatible with
existing systems already in the field.
If this bit is set to 0b, EEPROM word 32h (PXE Version) is valid. When
EPB is set to 1b and this bit is set to 0b, indicates that both images are
present in the FLASH.
EFI Presence.
Setting this bit to 1b Indicates that the image in the FLASH contains an
EFI image.
Setting this bit to 0b indicates that no EFI image is contained.

EPB

The default for this bit is 0b in order to be backwards compatible with
existing systems already in the field.
If this bit is set to 1b, EEPROM word 33h (EFI Version) is valid. When
PPB is set to 0b and this bit is set to 1b, indicates that both images
(PXE and EFI) are present in the FLASH.

Reserved

FDP

Reserved for future use. Set this bit to 0b.
Force Full Duplex.
Set this bit to 0b for half duplex; set to 1b for full duplex.
Note that this bit is a don’t care unless bits 10 and 11 are set.
These bits determine speed. 01b = 10 Mbs, 10b = 100 Mbs, 11b = Not
allowed.

11:10

FSP

All zeros indicate Auto-negotiate (the current bit state).
Note that bit 12 is a don’t care unless these bits are set.

Reserved

Reserved for future use. Set this bit to 0b.
Display Setup Message.

DSM

If this bit is set to 1b, the "Press Control-S" message appears after the
title message.
The default for this bit is 1b.
Prompt Time. These bits control how long the "Press Control-S" setup
prompt message appears, if enabled by DIM.
00b = 2 seconds (default)

7:6

01b = 3 seconds
10b = 5 seconds
11b = 0 seconds
Note that the Ctrl-S message does not appear if 0 seconds prompt time
is selected.

118

LBS

Local Boot Selection (OBSOLETE). In previous versions of the agent,
this bit enables or disables local boot, if the DBS bit selects it.
The default for this bit is 1b; enable local booting. The boot agent, at
runtime, no longer uses this bit.

Software Developer’s Manual

EEPROM Interface

Table 5-16. Boot Agent Main Setup Options
Bit

Name

Description
Default Boot Selection. These bits select which device is the default
boot device. These bits are only used if the agent detects that the BIOS
does not support boot order selection or if the MODE field of word 31h
is set to MODE_LEGACY.

4:3

DBS

00b = Network boot, then local boot
01b = Local boot, then network boot
10b = Network boot only
11b = Local boot only

BBS

BIOS Boot Specification (OBSOLETE). In previous versions of the
agent, this bit enables or disables use of the BBS to determine boot
order. If set to 1b, the BIOS boot order is used, and the DBS bits are
ignored. The boot agent at runtime no longer uses this bit. The runtime
checks for BBS/PnP and the setting in the MODE field of word 31h are
used instead.
Protocol Select. These bits select the boot protocol.

1:0

00b = PXE (default value)
01b = RPL protocol
Other values are undefined.

5.6.31

Boot Agent Configuration Customization Options (Word
31h)
Word 31h contains settings that can be programmed by an OEM or network administrator to
customize the operation of the software. These settings cannot be changed from within the ControlS setup menu or the IBA Intel Boot Agent utility. The lower byte contains settings that would
typically be configured by a network administrator using the Intel Boot Agent utility; these settings
generally control which setup menu options are changeable. The upper byte are generally settings
that would be used by an OEM to control the operation of the agent in a LOM environment,
although there is nothing in the agent to prevent their use on a NIC implementation

Software Developer’s Manual

119

EEPROM Interface

.
Table 5-17. Boot Agent Configuration Customization Options (Word 31h)
Bit

Name

Description

15:14

SIG

Signature. These bits must be set to 1b to indicate that this word has
been programmed by the agent or other configuration software.

13:11

Reserved

Reserved for future use. Set these bits to 0b.
Selects the agent's boot order setup mode. This field changes the
agent's default behavior in order to make it compatible with systems
that do not completely support the BBS and PnP Expansion ROM
standards. Valid values and their meanings are:
000b - Normal behavior. The agent attempts to detect BBS and PnP
Expansion ROM support as it normally does.
001b - Force Legacy mode. The agent does not attempt to detect BBS
or PnP Expansion ROM supports in the BIOS and assumes the BIOS is
not compliant. The BIOS boot order can be changed in the Setup
Menu.

10:8

MODE

010b - Force BBS mode. The agent assumes the BIOS is BBScompliant, even though it may not be detected as such by the agent's
detection code. The BIOS boot order CANNOT be changed in the
Setup Menu.
011b - Force PnP Int18 mode. The agent assumes the BIOS allows
boot order setup for PnP Expansion ROMs and hooks interrupt 18h (to
inform the BIOS that the agent is a bootable device) in addition to
registering as a BBS IPL device. The BIOS boot order CANNOT be
changed in the Setup Menu.
100b - Force PnP Int19 mode. The agent assumes the BIOS allows
boot order setup for PnP Expansion ROMs and hooks interrupt 19h (to
inform the BIOS that the agent is a bootable device) in addition to
registering as a BBS IPL device. The BIOS boot order CANNOT be
changed in the Setup Menu.
101b - Reserved for future use. If specified, treated as value 000b.
110b - Reserved for future use. If specified, treated as value 000b.
111b - Reserved for future use. If specified, treated as value 000b.

7:6

Reserved

Reserved for future use. Set these bits to 0b.
Disable FLASH Update.

DFU

If set to 1b, no updates to the FLASH image using PROSet is allowed.
The default for this bit is 0b; allow FLASH image updates using
PROSet.
Disable Legacy Wakeup Support.

DLWS

If set to 1b, no changes to the Legacy OS Wakeup Support menu
option is allowed.
The default for this bit is 0b; allow Legacy OS Wakeup Support menu
option changes.
Disable Boot Selection.

DBS

If set to 1b, no changes to the boot order menu option is allowed.
The default for this bit 0b; allow boot order menu option changes.

120

Software Developer’s Manual

EEPROM Interface

Table 5-17. Boot Agent Configuration Customization Options (Word 31h)
Bit

Name

Description
Disable Protocol Select.

DPS

If set to 1b, no changes to the boot protocol is allowed.
The default for this bit is 0b; allow changes to the boot protocol.
Disable Title Message.

If set to 1b, the title message displaying the version of the boot agent is
suppressed; the Control-S message is also suppressed. This is for
OEMs who do not wish the boot agent to display any messages at
system boot.

DTM

The default for this bit is 0b; allow the title message that displays the
version of the boot agent and the Control-S message.
Disable Setup Menu.
0

If set to 1b, no invoking the setup menu by pressing Control-S is
allowed. In this case, the EEPROM can only be changed via an
external program.

DSM

The default for this bit is 0b; allow invoking the setup menu by pressing
Control-S.

5.6.32

Boot Agent Configuration Customization Options (Word
32h)
Word 32h is used to store the version of the boot agent that is stored in the FLASH image. When
the Boot Agent loads, it can check this value to determine if any first-time configuration needs to
be performed. The agent then updates this word with its version. Some diagnostic tools to report
the version of the Boot Agent in the FLASH also read this word. This word is only valid if the PPB
is set to 0b. Otherwise the contents might be undefined.
Table 5-18. Boot Agent Configuration Customization Options (Word 32h)
Bit

Name

Description

15:12

MAJOR

PXE boot agent major version. The default for these bits is 0b.

11:8

MINOR

PXE boot agent minor version. The default for these bits is 0b.

7:0

BUILD

PXE boot agent build number. The default for these bits is 0b.

Software Developer’s Manual

121

EEPROM Interface

5.6.33

IBA Capabilities (Word 33h)
Word 33h is used to enumerate the boot technologies that have been programmed into the FLASH.
It is updated by IBA configuration tools and is not updated or read by IBA.
Table 5-19. IBA Capabilities
Bit

Name

Description

15:14

SIG

Signature. These bits must be set to 1b to indicate that this word has
been programmed by the agent or other configuration software.

13:5

Reserved

Reserved for future use. Set these bits to 0b.

SAN

SAN capability is present in FLASH.
0b = The SAN capability is not present (default).
1b = The SAN capability is present.
EFI UNDI capability is present in FLASH.
3

EFI

0b = The RPL code is not present (default).
1b = The RPL code is present.
RPL capability is present in FLASH.

RPL

1b = The RPL code is present (default).
0b = The RPL code is not present.
PXE/UNDI capability is present in FLASH.

UNDI

1b = The PXE base code is present (default).
0b = The PXE base code is not present.
PXE base code is present in FLASH.

0b = The PXE base code is present (default).
1b = The PXE base code is not present.

5.6.34

IBA Secondary Port Configuration (Words 34h-35h)
These words provide a unique configuration for the second port of the 82546GB/EB. The format is
the same as that used in words 30h and 31h for LAN A.

122

Software Developer’s Manual

EEPROM Interface

5.6.35

Checksum Word Calculation (Word 3Fh)
The Checksum word (3Fh) should be calculated such that after adding all the words (00h-3Fh),
including the Checksum word itself, the sum should be BABAh. The initial value in the 16-bit
summing register should be 0000h and the carry bit should be ignored after each addition. This
checksum is not accessed by the Ethernet controller. If CRC checking is required, it must be
performed by software.

Note:

5.6.36

Hardware does not calculate checksum word 3Fh during EEPROM write; it must be calculated by
software independently and included in the EEPROM write data. Hardware does not compute a
checksum over words 00h-0Fh during EEPROM reads in order to determine the validity of the
EEPROM image; this field is provided strictly for software verification of EEPROM validity. All
hardware configuration based on word 00h-0Fh content is based on the validity of the Signature
field of EEPROM Initialization Control Word 1. Signature must be 01b.

82546GB/EB Dual-Channel Fiber Wake on LAN (WOL) Mode
and Functionality (Word 0Ah, 20h)
Four bits and two words determine dual-channel fiber WOL mode and functionality. In addition to
the power management bits, one of the SDP’s must be set in order for the “A” laser to remain on
when the system goes into D3. Three states are defined for the EEPROM image:

• Never - The Ethernet controller has WOL disabled and cannot be put into WOL mode.
• Possible - The Ethernet controller can be WOL enabled, but currently has the feature turned
off. This is the normal shipping configuration for WOL-capable server adapter cards.

• On - The Ethernet controller is set with WOL functionality enabled.
Table 5-20. WOL Mode and Functionality (Word 0Ah)
State

Bit 2

Bit 7

Resulting Word

Never

4C03

Possible

4C0B

4C2B

Table 5-21. WOL Mode and Functionality (Word 20h)
State

5.6.37

Bit 2

Bit 7

Resulting Word

Never

C109

Possible

C10D

C18D

EEPROM Images
Refer to the appropriate Ethernet controller’s EEPROM map and programming information for
sample EEPROM images.

Software Developer’s Manual

123

EEPROM Interface

5.7

Parallel FLASH Memory
All Ethernet controllers except the 82540EP/EM provide an external parallel interface to an
optional FLASH or boot EEPROM device. Accesses to the FLASH memory are controlled by the
Ethernet controllers, but are accessible to host software as normal PCI reads or writes to the
FLASH memory mapping range. Software developers can also map FLASH memory to I/O space.
The Ethernet controllers support 8-bit wide parallel FLASH memory up to 4 Mb (512 KB); an
appropriate size for typical applications would be 1 Mb (128 KB). The size of the FLASH
implemented in the design can be encoded into bits in the EEPROM. FLASH and expansion ROM
base address registers are reconfigured based on these EEPROM settings.
Representative FLASH memory devices that have been found to work satisfactorily with the
Ethernet controllers are listed in Table 5-22:
Table 5-22. FLASH Memory Manufacturers
Manufacturer

Atmel

Manufacturer’s Part
Number
AT49LV010
AT49BV002AN-70J1
SST39V512

Silicon Storage
Technology

39VF020-90-4I-NH
39VF020-70-4C-NH

The FLASH memory interface trace routing is not critical because the interface runs at a very slow
speed. In a tightly space-constrained design, the FLASH memory device is a good choice for
placement in relative isolation from the Ethernet controllers.
Note:

124

The 82540EP/EM provides an external interface to a serial FLASH or Boot EEPROM device. See
Appendix B for more information.

Software Developer’s Manual

EEPROM Interface

Note:

This page intentionally left blank.

Software Developer’s Manual

125

EEPROM Interface

126

Software Developer’s Manual

FLASH Memory Interface

All Ethernet controllers (except the 82540EP/EM) provide an external parallel interface to a
FLASH, or boot ROM, device such as the Atmel AT49LV0101. All accesses to this device are
managed by the Ethernet controller and are accessible to software as normal PCI reads or writes to
the FLASH memory mapping range. The Ethernet controller supports parallel FLASH devices
with up to 4 Mb (512 KB) of memory. The size of the FLASH implemented with the Ethernet
controller can be encoded into bits in the EEPROM. The FLASH and Expansion ROM Base
Address Registers are reconfigured based on these EEPROM settings.
Note:

7.1

Though the Ethernet controller supports devices with up to 512 KB of memory, smaller devices can
also be used. Accesses to memory beyond the FLASH device size results in access wrapping as
only the lower address bits are utilized by the FLASH.

FLASH Interface Operation
The FLASH is read from, or written to, each time the host processor performs a read or a write
operation to a memory location that is within the FLASH address mapping or upon boot through
accesses in the space indicated by the Expansion ROM Base Address Register (see Section 4.1).
All accesses to the FLASH, except read accesses, require the appropriate command sequence for
the device used. Refer to the specific FLASH data sheet for more details on reading from or writing
to FLASH. Accesses to the FLASH are based on a direct decode of processor accesses to a memory
window defined in either the Ethernet controller's FLASH Base Address Register (PCI Control
Register at offset 14h or 18h) or the Expansion ROM Base Address Register (PCI Control Register
at offset 30h).
FLASH accesses must always be assembled or disassembled by the Ethernet controller to or from
the FLASH whenever the access is greater than a byte-wide access. Due to slow access times to a
typical FLASH, word (32-bit) accesses are not recommended for any cycles that occur after system
initialization in order to avoid violating PCI hold specifications. Boot ROM shadowing is an
exception to the 16 clock rule and can incur in excess of 25 wait states. The Ethernet controller byte
reads to the FLASH take approximately of 256 ns. The Ethernet controller issues retry accesses
during this time.

7.2

FLASH Control and Accesses
Write control of the FLASH is controlled by the FWE bits in the EEPROM/FLASH Control and
Data Register (EECD.FWE). See Section 13.4.3 for details.
Processor accesses to the FLASH are very slow. The Ethernet controller always issues a targetdisconnect at the first data cycle. The Ethernet controller asserts the STOP# signal to indicate the
target-disconnect. Section 7.2.1 and Figure 7.2.2 show read and write accesses to the FLASH. Note
that burst accesses to the FLASH address space are not allowed.

The 82540EP/EM provides an external interface to a serial FLASH or Boot EEPROM device. See Appendix B for more information.

Software Developer’s Manual

127

FLASH Memory Interface

7.2.1

Read Accesses
Upon reads to the FLASH address space, the Ethernet controller uses the TRDY# signal to insert
target wait states until valid data can be read from the FLASH device and presented on the data
lines. When TRDY# is asserted, the Ethernet controller drives valid data on the data lines. The
processor master can then complete normal data read cycle by asserting IRDY# when it is ready.

CLK

FRAME#
AD

ADDRESS

CBE#

MEM-RD

DATA

BE#s

IRDY#
TRDY#
DEVSEL#
STOP#

Figure 7-1. FLASH Buffer Read Cycles

7.2.2

Write Accesses
The processor, as the initiator, drives the address lines AD[63:0], the command and byte enable
lines [7:0]#, and the control lines IRDY# and FRAME#. It also provides the Ethernet controller
with valid data immediately after asserting TRDY#. The Ethernet controller controls the TRDY#
signal and deasserts it for a certain number of clocks until valid data is written to the FLASH
buffer. By asserting TRDY#, the Ethernet controller signals the processor that the current data
phase has completed.

128

Software Developer’s Manual

FLASH Memory Interface

CLK

FRAME#
AD
CBE#

ADDRESS

DATA

MEM-W R

BE#s

IRDY#
TRDY#
DEVSEL#
STOP#

Figure 7-2. FLASH Buffer Write Cycle

Software Developer’s Manual

129

FLASH Memory Interface

Note:

130

This page is intentionally left blank.

Software Developer’s Manual

Power Management

Power Management
6.1

Introduction to Power Management
The PCI/PCI-X Family of Gigabit Ethernet Controllers support the Advanced Configuration and
Power Interface (ACPI) specification as well as Advanced Power Management (APM). This
section describes how Power Management is implemented in the Ethernet controllers.
Note:

The 82541ER does not support ACPI or APM wakeup.
Power Management can be disabled via bits in the Initialization Control Word 2 which is loaded
from the EEPROM during power-up reset. See the EEPROM description in Section 5.6.12 for
further details. Even when disabled, the Power Management register set is still present.
The Ethernet controller supports the following Power Management related features:

• Power states of D0 & D3hot with optional D3cold support
• Power(D3) < Power(D0)
• Wakeup

6.2

Assumptions
The following assumptions apply to the implementation of Power Management for the Ethernet
controller.

• Any time LAN_PWR_GOOD is asserted all power supplies are stable, RST# is stable, and the
clock is stable.

• Prior to transition from D0 to the D3 state, the operating system ensures that the software
device driver is disabled and all pending bus transactions are complete or cleanly terminated.

• The driver sets up the filters prior to the system transitioning the Ethernet controller to the D3
state.

• The system never deactivates the PCI clock in 66 MHz PCI mode or any PCI-X1 mode without
asserting RST#.

• No wakeup capability, except APM Wakeup if enabled in the EEPROM, is required after the
system asserts, then de-asserts RST#.

• No wakeup capability, except APM Wakeup if enabled in the EEPROM, is required after the
system puts the Ethernet controller in the D3 state and then returns it to D0.

• If the APMPME bit (bit 3) in the Wakeup Control Register (WUC.APMPME) is set to 1b, it is
permissible to assert PME# even when PME_En is 0b.

• The deassertion (rising) edge of RST# puts the Ethernet controller in the D0u state.

The 82540EP/EM, 82541xx, and 82547GI/EI do not support PCI-X mode.

Software Developer’s Manual

131

Power Management

6.3

D3cold support
If the AUX pin is connected to logic 1b, the Ethernet controller advertises D3cold Wakeup support.
The amount of power required for this function (which includes the entire Ethernet port circuitry)
is advertised in the Power Management Data Register which is loaded from the EEPROM.
If D3cold is supported, the PME_En and PME_Status bits of the Power Management Control/Status
Register (PMCSR), as well as their shadow bits in the Wakeup Control Register (WUC) are not
reset by RST#. If D3cold Wakeup is not supported, PMCSR and WUC is reset on the deassertion
(rising edge) of RST#.
The only effect of setting AUX to 1b is advertising D3cold Wakeup support and changing the reset
function of PME_En and PME_Status. The 82541PI/GI can enter a fully-disabled low-power state
in D3cold if an enable bit is set in the EEPROM. All remaining Ethernet controllers do nothing
different in D3cold compared to D3hot. AUX_POWER is level sensitive, and any changes are
immediately reflected in the D3cold Wakeup advertisements and the PME_En and PME_Status
reset function.

6.3.1

Power States
The Ethernet controller supports D0 and D3 power states defined in the PCI Power Management
Specification. D0 is divided into two sub-states: D0u, and D0a. In addition, it supports a Dr state
that is entered when RST# is asserted. Dr behaves the same as D3 except that the PCI bus is
isolated. Figure 6-1 illustrates the power states and the conditions that cause transitions from state
to state.
LAN_POWER_GOOD
assertion

RST#
deassertion

Dr*

D0u
RST# assertion

RST#
assertion

Enable Memory
Access
Write 00b
to Power
State

RST#
assertion

Write 11b to
Power State

D0a

*equivalent to D3 except PCI pins are floated

Figure 6-1. Power State Transitions

132

Software Developer’s Manual

Power Management

6.3.1.1

Dr State
At initial boot-up, once LAN_PWR_GOOD is asserted, the Ethernet controller reads the
EEPROM. If the APM Mode bit in the EEPROM’s Initialization Control Word 2 is set then APM
Wakeup is enabled.
The system may maintain RST# asserted for an arbitrary time. During this time, and for up to 1 ms
afterwards, the Ethernet controller does not assert any PCI signals except PME#.
During operation, the system may assert RST# at any time. In particular, if the system wishes an
Ethernet controller to enter the D3cold state it must assert RST# before dropping main power. Any
time RST# is asserted, the Ethernet controller transitions to the Dr state. It also floats all PCI
signals except PME# and remains in the “reset” state until no more than 1 ms after the deassertion
of RST#.
Internally, the Ethernet controller treats the reset state equivalently to D3. Any Wakeups enabled
before entering reset is maintained. For power savings, the Ethernet controller shuts down some
internal clocks and registers and deasserts PWR_STATE1. If Wakeup is not enabled, the Ethernet
controller also deasserts PWR_STATE0. As a result, the Ethernet controller won’t transmit any
frames in Dr state or send idles in TBI mode (82544GC/EI)/internal SerDes (82546GB/EB and
82545GM/EM)1.
The deassertion (rising edge) of RST# causes a transition to D0u.

6.3.1.2

D0u State
The D0u state is a low-power state used after RST# is deasserted, or when coming out of D3, but
before the Ethernet controller is initialized.
When entering D0u, the Ethernet controller disables Wakeups, resets the PHY, and then re-reads
the EEPROM. If the APM Mode bit in the EEPROM’s Initialization Control Word 2 is set, then
APM Wakeup is enabled.
Internally, D0u is treated like D3 and some internal clocks and registers are shut down. The D0u
state is exited when the system enables memory space access to the Ethernet controller by writing a
1b to the Memory Access Enable bit of the PCI Command Register.

Note:

In order for hardware to transition from D3 to the D0 state properly, BIOS should not alter the
Memory Access Enable or the I/O Access Enable bit of the PCI Command Register. Also, the PCI
configuration space must be programmed when hardware transitions out of D3 to D0.

Not applicable to the 82541xx, 82547GI/EI, or 82540EP/EM.

Software Developer’s Manual

133

Power Management

6.3.1.3

D0a (D0 active)
Once memory space is enabled, all internal clocks are activated, the Ethernet controller enters an
active state, and can then transmit and receive packets if properly configured by the software
driver. The controller also signals the PHY (if using the internal PHY) to indicate full speed/
power1. If APM Wakeup was activated it remains active. The software driver can deactivate APM
Wakeup by writing to the Wakeup Control Register (WUC), or activate other Wakeup Filters by
writing to the Wakeup Registers.

6.3.1.4

D3
Prior to transition from D0 to the D3 state, the software driver must ensure the Ethernet controller
transmit and receive functions have been disabled and all pending bus transactions are complete or
cleanly terminated. If Wakeup capability is needed, the software driver needs to set up the
appropriate Wakeup registers and the system needs to write a 1b to the PME_En bit of the Power
Management Control / Status Register (PMCSR) prior to the transition to D3.
When the system writes a 11b to the PowerState field of the Power Management Control/Status
Register (PMCSR) the Ethernet controller transitions to D3. Any Wakeups that are enabled remain
enabled. Upon transitioning to D3 the Ethernet controller clears the Memory Access Enable or the
I/O Access Enable bit of the PCI Command Register, which disables memory access decode. In
D3, the Ethernet controller only responds to PCI configuration accesses. It won’t generate master
cycles, transmit any frames on the TBI/internal SerDes2/internal PHY interface, or transmit idles in
TBI mode/internal SerDes if Wakeup is enabled.
For power savings the Ethernet controller shuts down some internal clocks and registers.
To transition back to D0u, the system writes a 00b to the Power State field of the Power
Management Control/Status Register (PMCSR).

6.3.2

Timing
The following sections give detailed timing for the state transitions. In the diagrams the dotted
connecting lines represent the Ethernet controller’s requirements, while the solid connecting lines
represent the Ethernet controller’s guarantees.
Note:

The following timing diagrams are not to scale. The clocks edges are shown to indicate running
clocks only and are not used to indicate the actual number of cycles for any operation.
If CLK_RUN# functionality is enabled in the EEPROM, then the 82541PI/GI/EI and 82540EP
Ethernet controllers assert the CLK_RUN# pin when it requires the PCI clock. Otherwise, the
clock is not required and the system might shut the PCI clock off.

1.
2.

Not applicable to the 82541xx or 82547GI/EI.
Not applicable to the 82541xx, 82547GI/EI, or 82540EP/EM.

134

Software Developer’s Manual

Power Management

6.3.2.1

Power Up (Off to Dr to D0u to D0a)

Power
1

LAN_POWER_GOOD

CLK#
5
4

RST#

8
2

Reading EEPROM

Memory Access

Read EEPROM

PCI Pins

Running

Wakeup Enabled

PWR_STATE[1:0]

00b

APM Wakeup

00b if wakeup is disabled, 10b if wakeup is enabled

11b

82544GC/EI Only
DState

D0u

D0a

Figure 6-2. Startup Timing

Diagram #

Notes

LAN_PWR_GOOD must not be asserted until all power supplies are good and the clock is stable.

An EEPROM read starts on the rising edge of LAN_PWR_GOOD and RST#.

APM Wakeup mode can be enabled based on what is read from the EEPROM.

The system can delay an arbitrary time before deasserting RST#.

The PCI 2.2 or 2.3 specification requires the clock to be active 100 µs before deasserting RST#. (Tclk-rst
parameter)

The deassertion edge of RST# causes the EEPROM to be re-read and Wakeup disabled.

Synchronizing the clock generators and circuit adjustments require up to 512 PCI clocks before the Ethernet
controller drives PCI signals and responds to PCI transactions.

The system can delay an arbitrary time before enabling Memory Access.

Writing a 1b to the Memory Access Enable or I/O Access Enable bit in the PCI Command Register transitions the
Ethernet controller from D0u to D0 state.
For the 82544GC/EI, writing a 1b to the Memory Access Enable or I/O Access Enable bit in the PCI Command
Register transitions the Ethernet controller from D0u to D0 state and asserts both PWR_STATE outputs.

Software Developer’s Manual

135

Power Management

6.3.2.2

Transition From D0a to D3 and Back Without PCI Reset

I_PCI_CLK

RST#

D0 Write
2

Memory Access Enable 7

Reading EEPROM

Read EEPROM

D3 write
1

PCI Pins

Running

Wakeup Enabled

Any mode

APM only

4
5

PWR_STATE[1:0]
82544GC/EI Only
DState

11b

D0a

00b if wakeup is disabled, 01b if wakeup is enabled

00b/01b

11b

D0u

Figure 6-3. Transition from D0a to D3 and Back Without PCI Reset

Diagram #

Notes

Writing a 11b to the Power State field of the Power Management Control/Status Register (PMCSR) transitions
the Ethernet controller to D3.

The system can keep the Ethernet controller in D3 state for an arbitrary amount of time.

To exit D3 state the system writes 00b to the Power State field of the Power Management Control/Status
Register (PMCSR).

APM Wakeup mode can be enabled based on what is read in the EEPROM.

For the 82544GC/EI, PWR_STATE[1:0] is set to 01b if APM Wakeup is enabled, 00b otherwise.

The system can delay an arbitrary time before enabling memory access.

136

Writing a 1b to the Memory Access Enable or I/O Access Enable bit in the PCI Command Register transitions
the Ethernet controller from D0u to D0 state.
For the 82544GC/EI, writing a 1b to the Memory Access Enable or I/O Access Enable bit in the PCI Command
Register transitions the Ethernet controller from D0u to D0 state and asserts both PWR_STATE outputs.

Software Developer’s Manual

Power Management

6.3.2.3

Transition From D0a to D3 and Back with PCI Reset

CLK#
3

RST#

9
6
2

Memory Access Enable 10

Reading EEPROM

Read EEPROM

D3 write
7

PCI Pins

Running

Wakeup Enabled

Any mode

APM Wakeup
8

PWR_STATE[1:0]

82544GC/EI Only
DState

11b

D0a

00b if wakeup is disabled, 01b if wakeup is enabled

00b/01b

D0u

11b

D0a

Figure 6-4. Transition From D0a to D3 and Back with PCI Reset

Diagram #

Notes

Writing a 11b to the Power State field of the Power Management Control/Status Register (PMCSR) transitions
the Ethernet controller to D3.

The system can delay an arbitrary amount of time between setting D3 mode and asserting RST#.

In 66 MHz or PCI-Xa modes the system must assert RST# before stopping the PCI clock. It may assert RST#
without stopping the clock.
For the 82541PI/GI/EI and 82540EP, If CLK_RUN# is enabled, then they do not require a continuous clock
during this time, but does require that the system drive the clock in response to CLK_RUN# assertion.

Upon assertion of RST# the Ethernet controller floats all PCI pins except PME# and goes to “Dr” state.

In 66 MHz or PCI-X modes, the PCI 2.2 and 2.3 specification requires the system to start the PCI clock 100 µs
before deassertion of RST#. In 33 MHz systems the PCI clock can start and stop at any time independent of
RST#.

The deassertion edge of RST# causes the EEPROM to be re-read and Wakeup disabled.

Synchronizing the clock circuits and circuit adjustments require up to 512 PCI clocks before the Ethernet
controller drives PCI signals and responds to PCI transactions.

For the 82544GC/EI, O_PWR_STATE is set to 01b if APM Wakeup is enabled, 00b otherwise.

The system can delay an arbitrary time before enabling memory access.

Writing a 1b to the Memory Access Enable or I/O Access Enable bit in the PCI Command Register transitions
the Ethernet controller from D0u to D0 state.
For the 82544GC/EI, writing a 1b to the Memory Access Enable or I/O Access Enable bit in the PCI Command
Register transitions the Ethernet controller from D0u to D0 state and asserts both PWR_STATE outputs.
Not applicable to the 82541xx, 82547GI/EI, or 82540EP/EM.

Software Developer’s Manual

137

Power Management

6.3.2.4

PCI Reset Without Transition to D3

CLK#
3

RST#
7

Memory Access Enable 8

Reading EEPROM

Read EEPROM

PCI Pins

Running

Wakeup Enabled

Running

Any mode

APM Wakeup
6

PWR_STATE[1:0]

11b

DState

D0a

82544GC/EI Only

00b if wakeup is disabled, 01b if wakeup is enabled

00b/01b

D0u

11b

D0a

Figure 6-5. PCI Reset Sequence

Diagram #

Notes
PCI-Xa

In 66 MHz or
modes, the system must assert RST# before stopping the PCI clock. It may assert RST#
without stopping the clock.

Upon assertion of RST# the Ethernet controller floats all PCI pins except PME# and goes to “Dr” state.

In 66 MHz or PCI-X modes the system must assert RST# before stopping the PCI clock. It may assert RST#
without stopping the clock.
For the 82541PI/GI/EI and 82540EP, If CLK_RUN# is enabled, then they do not require a continuous clock
during this time, but does require that the system drive the clock in response to CLK_RUN# assertion.

The deassertion edge of RST# caused the EEPROM to be re-read and Wakeup disabled.

Synchronizing the clock circuits and circuit adjustments require up to 512 PCI clocks before the Ethernet
controller drives PCI signals and responds to PCI transactions.

For the 82544GC/EI, PWR_STATE[1:0] is set to 01b if APM Wakeup is enabled, 00b otherwise.

The system can delay an arbitrary time before enabling memory access.
Writing a 1b to the Memory Access Enable or I/O Access Enable bit in the PCI Command Register transitions
the Ethernet controller from D0u to D0 state.

138

For the 82544GC/EI, writing a 1b to the Memory Access Enable or I/O Access Enable bit in the PCI
Command Register transitions the Ethernet controller from D0u to D0 state and asserts both PWR_STATE
outputs.
Not applicable to the 82541xx, 82547GI/EI, or 82540EP/EM.

Software Developer’s Manual

Power Management

6.3.3

PCI Power Management Registers
Power Management registers are part of the capabilities linked list pointed to by the Capabilities
Pointer (Cap_Ptr) in the PCI configuration space. Refer to Section 4.1.
All fields are reset by LAN_PWR_GOOD. All of the fields except PME_En and PME_Status are
reset by the deassertion (rising edge) of RST#. If AUX_POWER = 0b, the PME_En and
PME_Status fields also reset by the deassertion (rising edge) of RST#.
The following table lists the organization of the PCI Power Management Register Block:

Byte Offset

Byte 3

Byte 2

Power Management Capabilities

DCh

(PMC)

E0h

Data

PMCSR_BSE
Bridge Support
Extensions

Byte 1

Byte 0

Next Item Ptr

Capability ID

Power Management Control / Status
Register (PMCSR)

The following sections describe the register definitions, whether they are required or optional for
compliance, and how they are implemented in the Ethernet controller. Complete details can be
found in the PCI Power Management Interface specification.
Note:

6.3.3.1

The offset indicated is the byte-offset from the position indicated by Cap_Ptr in the Configuration
Space Header.

Capability ID
1 Byte Offset = 0 (RO)
Bits
07:00

6.3.3.2

Default
01h

R/W
Read
Only

Description
ID – The Ethernet controller returns a value of 01h for this field, indicating
the linked list item as being the PCI Power Management Registers.

Next Item Pointer
1 Byte Offset = 1 (RO)
Bits
07:00
a.

Default
E4h

R/W
Read
Only

Description
Next Item Pointer - This field provides an offset into the function’s PCI
Configuration Space pointing to the location of next item in the function’s
capability list. Its value of E4h points to the PCI-Xa capability.

Not applicable to the 82541xx, 82547GI/EI, or 82540EP/EM.

Software Developer’s Manual

139

Power Management

6.3.3.3

Power Management Capabilities - (PMC)
2 Bytes Offset = 2 (RO)
Bits

Default

R/W

Description
PME_Support – This 5-bit field indicates the power states in which the
function may assert PME#a. A value of 0b for any bit indicates that the
function is not capable of asserting the PME# signal while in that power
state.
bit (11) (XXXX1)b – PME# can be asserted from D0
bit (12) (XXX1X)b – PME# can be asserted from D1
bit (13) (XX1XX)b – PME# can be asserted from D2

15:11

See text

Read
Only

bit (14) (X1XXX)b – PME# can be asserted from D3hot
bit (15) (1XXXX)b – PME# can be asserted from D3cold
If Power Management is not disabled in the EEPROM, the Ethernet
controller supports PME# generation from D0 and D3hot states. If Power
Management is not disabled and AUX_POWER = 1b, the Ethernet
controller also supports the D3cold state.
Value
Power Management disabled in EEPROM

AUX_POWER = 01001b

Power Management enabled,

AUX_POWER = 11001b

Power Management enabled,

D2_Support - If this bit is set to 1b, supports the D2 Power Management
State.

Read
Only

D1_Support - If this bit is set to 1b, supports the D1 Power Management
State. The Ethernet controller returns a value of 0b for this bit indicating
that it does not support D1.

08:06

000b

Read
Only

AUX Current – Specifies the auxiliary power current required for PME#
generation from D3cold if the Data Register is not implemented.

The Ethernet controller returns a value of 0b for this bit indicating that it
does not support D2 and cannot handle the PCI clock stopping in PCI
66 MHz mode (or PCI-Xb mode) without RST# being asserted.

Read
Only

DSI – The Device Specific Initialization bit indicates whether special
initialization of this function is required (beyond the standard PCI
configuration header) before the generic class device driver is able to use
it. The Ethernet controller returns a value of 1b for this bit indicating that
it’s device driver must be executed following transition to the D0
uninitialized state.

Read
Only

Reserved

Loaded
from
EEPROM

Read
Only

PME_Clock - When this bit is a 1b it indicates that the function relies on
the presence of the PCI clock for PME# operation. The controller loads
this bit from the EEPROM. Otherwise, it returns a 0b.

02:00

010b

Read
Only

Version - A value of 010b indicates that this function complies with the
Revision 1.1 of the PCI Power Management Interface Specification.

a.
b.

140

Condition
00000b

Not applicable to the 82541ER.
Not applicable to the 82541xx, 82547GI/EI, or 82540EP/EM.

Software Developer’s Manual

Power Management

6.3.3.4

Power Management Control / Status Register - (PMCSR)
2 Bytes Offset = 4 (RO)

Bits

Default

0b (see
description)

R/W

Read/
Write 1b
to clear

Description
PME_Status – This bit is set when the function would normally assert the
PME# signal independent of the state of the PME_En bit. The Ethernet
controller returns a value of 1b for this bit if a Wakeup condition has been
detected.
Writing a 1b clears this bit and deasserts PME#a.
If the AUX_POWER input is 1b, the PME_Status field is only reset by
LAN_PWR_GOOD. If AUX_POWER is 0b, PME_Status is also reset on the
deassertion (rising edge) of RST#.

00b
14:13

01b if
Manageability is
enabled

Read
Only

Data_Scale - This 2- bit read-only field indicates the scaling factor to be
used when interpreting the value of the Data register. This field outputs 01b
(to indicate units of 0.1 watt) when Manageability is enabled in the
EEPROM and the Data_Select field is set to 0, 3, 4, or 7, and 00b
otherwise.

Read/
Write

Data_Select - This 4-bit field is used to select which data is to be reported
through the Data register and Data_Scale field. These bits are only writable
when Power Management is enabled via EEPROM.

(see description)
12:09

0000b

PME_En – If Power Management is not disabled in the EEPROM, writing a
1b to this register enables Wakeup and causes the Ethernet controller to
assert PME# when it receives a Wakeup event enabled in the Wakeup Filter
Control Register (WUFC).
Note: This bit cannot be set for the 82541ER.
0b on Power-On
reset

Read/
Write

If Power Management is disabled in the EEPROM, writing a 1b to this bit
has no affect, and does not set the bit to 1b.
If the AUX_POWER input is 1b, the PME_En field is only reset by
LAN_PWR_GOOD. If AUX_POWER is 0b, it is also reset on the
deassertion (rising edge) of RST#.
Note: If APM Wakeup is enabled, the PME# pin can be asserted even if
PME_En is 0b. See Section 6.4.1 for details.

07:02

000000b

Read
Only

Reserved - The Ethernet controller returns a value of 000000b for this field.
PowerState - This 2-bit field is used both to determine the current power
state of a function and to set the function into a new power state. The
definition of the field values is as follows:
00b - D0
01b - D1 (ignored if written with this value)

01:00

00b

Read/
Write

10b - D2 (ignored if written with this value)
11b - D3
If software attempts to write an unsupported state to this field, 00b or 10b, or
if Power Management is disabled in the EEPROM, then the Ethernet
controller completes the write operation normally on the bus, however the
data is discarded and no state change occurs.
These bits are cleared and the power state is returned to D0 after the trailing
edge of RST#.

Not applicable to the 82541ER.

This register is used to control and monitor power management events in the Ethernet controller. If
auxiliary power is present, as indicated by AUX_POWER = 1b, a PCI reset does not clear
PME_En and PME_Status.

Software Developer’s Manual

141

Power Management

6.3.3.5

PMCSR_BSE Bridge Support Extensions
1 Byte Offset = 6 (RO)

This register indicates support for PCI bridge specific functions. Note that these functions are not
implemented in the Ethernet controller and the values are set to 00h.

6.3.3.6

Data Register
1 Byte Offset = 7 (RO)
Bits

07:00

Default
00h (loaded
from
EEPROM)

R/W
Read
Only

Description
Data returned.
See the following explanation.

This register is used to report power consumption and heat dissipation. Its value and meaning is
determined by the value programmed in the Data_Select field of the Power Management Control/
Status Register (PMCSR).
Data Select

Meaning

D0 Power Consumed

D1 Power Consumed

D2 Power Consumed

D3 Power Consumed

D0 Power Dissipated

D1 Power Dissipated

D2 Power Dissipated

D3 Power Dissipated

Common power consumption of multi-function devices

9-15

Reserved

The units are defined by the Data_Scale field of the Power Management Control/Status Register.
Data Select

Meaning

Unknown (used for unsupported states)

0.1 Watts (used by Ethernet controller for supported
states)

0.01 Watts

0.001 Watts

If power management is disabled in the EEPROM, then the data register always reads 0b.

142

Software Developer’s Manual

Power Management

If power management is not disabled and when the Data_Select field is programmed to 0 or 4, the
Ethernet controller sets the Data Register to the D0 Power value in the EEPROM. When the
Data_Select field is programmed to 3 or 7, the Ethernet controller sets the Data Register to the D3
Power value in the EEPROM. Otherwise it returns 0b.

6.4

Wakeup
The Ethernet controller supports two types of wakeup mechanisms:

• Advanced Power Management (APM) Wakeup
• ACPI Power Management Wakeup
Note:

The 82541ER contains power management logic, but is not spec-compliant, because it does not
assert PME# for Magic Packets, Network Wakeup Packets, or link change status.
The ACPI Power Management Wakeup uses the PME# pin to wake up the system. The Advanced
Power Management Wakeup uses the PME# pin.

6.4.1

Advanced Power Management Wakeup
“Advanced Power Management Wakeup”, or “APM Wakeup”, was previously known as “Wake on
LAN”. The basic premise is to receive a broadcast or unicast packet with an explicit data pattern,
and then to assert a signal to wake up the system. In the earlier generations of the Ethernet
controller, this was accomplished by using special signal. The Ethernet controller would assert the
signal for approximately 50 ms to signal a wakeup. The 82544GC/EI uses the APM_WAKEUP
pin for this function. For the remaining Ethernet controllers, the PCI PME# signal has been used to
wake up the system.
On power-up, the Ethernet controller reads the APM Enable bits from the EEPROM Initialization
Control Word 2 into the APM Enable (APME) bits of the Wakeup Control Register (WUC). These
bits control enabling of APM Wakeup.
When APM Wakeup is enabled, the Ethernet controller checks all incoming packets for “Magic
Packets”. See Section 6.4.3.1.4 for a definition of “Magic Packets*”.
Once the Ethernet controller receives a matching magic packet, it:

• Sets the PME_Status bit in the Power Management Control / Status Register (PMCSR) and
asserts PME#. If the Assert PME On APM Wakeup (APMPME) bit is set in the Wakeup
Control Register (WUCR).

•
•
•
•

Stores the first 128 bytes of the packet in the Wakeup Packet Memory (WUPM).
Sets the Magic Packet Received bit in the Wakeup Status Register (WUS).
Sets the packet length in the Wakeup Packet Length Register (WUPL).
Asserts PME# until the driver clears the Magic Packet Received AMAG bit in the Wakeup
Status Register (WUS), the driver clears the Assert PME On APM Wakeup (APMPME) bit in
the Wakeup Control Register (WUC), or the driver disables APM Wakeup.

• For the 82544GC/EI only, asserts APM_WAKEUP for 50 ms. For purposes of
APM_WAKEUP assertion, the 82544GC/EI ignores any additional magic packets received
during that 50 ms. If the 82544GC/EI receives another magic packet afterwards, it reasserts
APM_WAKEUP for another 50 ms.

Software Developer’s Manual

143

Power Management

• Maintains the first magic packet received in the Wakeup Packet Memory (WPM) until the
driver writes a 0b to the Magic Packet Received MAG bit in the Wakeup Status Register
(WUS).
“APM Wakeup” is supported in all power states and only disabled if a subsequent EEPROM read
results in the APM Wakeup bit being cleared or software explicitly writes a 0b to the APM Wakeup
(APM) bit of the WUC register.

6.4.2

ACPI Power Management Wakeup
The Ethernet controller supports ACPI Power Management based wakeups. It generates system
wakeup events from three sources:

• Reception of a “Magic Packet”.
• Reception of a Network Wakeup Packet.
• Detection of a link change of state.
Note:

The 82541ER does not support ACPI wakeup events.
Activating ACPI Power Management Wakeup requires:

• The software driver to program the Wakeup Filter Control Register (WUFC). This indicates
the packets the driver wishes to wake up and supplies the necessary data to the IP Address
Table (IPAT) and the Flexible Filter Mask Table (FFMT), Flexible Filter Length Table (FFLT),
and the Flexible Filter Value Table (FFVT). The driver can also set the Link Status Change
Wakeup Enable (LNKC) bit in the Wakeup Filter Control Register (WUFC) to cause wakeup
when the link changes state.

• The OS to write a 1b to the Pme_En bit of the Power Management Control / Status Register
(PMCSR).
Normally, after enabling wakeup, the OS sets the Ethernet controller to D3 (low-power mode).
Once wakeup is enabled, the Ethernet controller monitors incoming packets, first filtering them
according to its standard address filtering method, then filtering them with all of the enabled
wakeup filters. If a packet passes both the standard address filtering and at least one of the enabled
wakeup filters, the Ethernet controller:

• Sets the PME_Status bit in the Power Management Control / Status Register (PMCSR)
• Assert PME#. If the PME_En bit in the Power Management Control / Status Register
(PMCSR) is set.

• Stores the first 128 bytes of the packet in the Wakeup Packet Memory.
• Sets one or more of the “Received” bits in the Wakeup Status Register (WUS). Note that the
Ethernet controller sets more than one bit if a packet matches more than one filter.

• Sets the packet length in the Wakeup Packet Length Register (WUPL).
If enabled, a link state change wakeup causes similar results. For example, setting PME_Status,
asserting PME#, and setting the Link Status Changed (LNKC) bit in the Wakeup Status Register
(WUSR) when the link goes up or down.
PME# remains asserted until the OS either writes a 1b to the PME_Status bit of the PMCSR
register or writes a 0b to the Pme_En bit.

144

Software Developer’s Manual

Power Management

After receiving a wakeup packet, the Ethernet controller ignores any subsequent wakeup packets
until the driver clears all of the “Received” bits in the Wakeup Status Register (WUS). It also
ignores link change events until the driver clears the Link Status Changed (LNKC) bit in the
Wakeup Status Register (WUSR).

6.4.3

Wakeup Packets
The Ethernet controller supports various wakeup packets using two types of filters:

• Pre-defined Filters
• Flexible Filters
Each of these filters are enabled if the corresponding bit in the Wakeup Filter Control Register
(WUFC) is set to 1b.

6.4.3.1

Pre-Defined Filters
The following packets are supported by the Ethernet controller’s pre-defined filters:

•
•
•
•
•

Directed Packet (including exact, multicast indexed, and broadcast)
Magic Packet* (not applicable to the 82541ER)
ARP/IPv4 Request Packet (ARP Request Packet for the 82544GC/EI)
Directed IPv4 Packet (Directed IP Packet for the 82544GC/EI)
Directed IPv6 Packet1

Each of these filters are enabled if the corresponding bit in the Wakeup Filter Control Register
(WUFC) is set to 1b.
The following explanation of each filter includes a table listing which bytes at which offsets are
compared to determine if the packet passes the filter. Note that both VLAN2 frames and LLC/Snap
can increase the given offsets if they are present (see Section 9.3 for details).
For the 82541PI/GI/EI and 82547GI/EI, various tables can also include a reference to a possible
VLAN Tag and LLC/SNAP Header. These Ethernet controllers detect VLAN and LLC/Snap
frames by checking the initial size/type field. They first check for a VLAN header by comparing
the size/type field to the value programmed in the VLAN EtherType register. If the field matches,
the Ethernet controllers consider the frame a VLAN frame. They then check the VLAN ID against
the values programmed in the VLAN Filter Table Array. If the ID matches the packet, processing
continues. If the ID doesn’t match, or the CTRL.VME bit is 0b, and the VLAN Tag is listed as
compare in the table, the packet is not considered a wakeup packet.
After processing a possible VLAN Tag, the 82541PI/GI/EI and 82547GI/EI Ethernet controllers
check for an LLC/SNAP Header. If the size/type field is less than or equal to 1500 bytes, they
check the following 6 bytes for the pattern of AAAA03000000h. If the pattern matches, then the
packet processing continues. If the pattern doesn’t match, and the LLC/SNAP Header is listed as
compare or check in the table, the packet is not considered a wakeup packet.
6.4.3.1.1

1.
2.

Directed Exact Packet

Not applicable to the 82544GC/EI.
Not applicable to the 82541ER.

Software Developer’s Manual

145

Power Management

The Ethernet controller generates a wakeup event after receiving any packet whose destination
address matches one of the 16 valid programmed Receive Addresses if the Directed Exact Wakeup
Enable bit is set in the Wakeup Filter Control Register (WUFC.EX).
# of
bytes

Offset
0

6.4.3.1.2

Field

Value

Destination Address

Action

Comment
Match any preprogrammed address

Compare

Directed Multicast Packet

For multicast packets, the upper bits of the incoming packet’s destination address indexes a bit
vector (Multicast Table Array) that indicates whether to accept the packet. If the Directed Multicast
Wakeup Enable bit set in the Wakeup Filter Control Register (WUFC.MC) and the indexed bit in
the vector is one then the Ethernet controller generates a wakeup event. The exact bits used in the
comparison are programmed by software in the Multicast Offset field of the Receive Control
Register (RCTL.MO).
Offset

# of
Bytes

Field

Destination Address

6.4.3.1.3

Value

Action

Comment

Compare

See Section 6.4.3.1.2.

Broadcast
If the Broadcast Wakeup Enable bit in the Wakeup Filter Control Register (WUFC.BC) is set, the
Ethernet controller generates a wakeup event when it receives a broadcast packet.

Offset

# of
bytes

Field

Value

Action

Destination Address

FF*6

Compare

6.4.3.1.4

Comment

Magic Packet*1

Magic Packet* technology is defined at: http://www.amd.com/products/npd/overview/20212.html.
The Ethernet controller expects the destination address to:
1. Be the broadcast address (FF.FF.FF.FF.FF.FF)
2. Match the value in Receive Address Register 0 (RAH0, RAL0). This is initially loaded from
the EEPROM but can be changed by the software driver.
3. Match any other address filtering enabled by the software driver.
The Ethernet controller searches for the contents of Receive Address Register 0 (RAH0, RAL0) as
the embedded IEEE address. It considers any non FF byte after a series of at least 6 FFs to be the
start of the address for comparison purposes (for example, it catches the case of 7 FFs followed by
the address). As soon as one of the first 96 bytes after a string of FFs doesn’t match, it continues to
search for anther set of at least 6 FFs followed by the 16 copies of the IEEE address later in the
packet. Note that this definition precludes the first byte of the destination address from being FF.

Not applicable to the 82541ER.

146

Software Developer’s Manual

Power Management

# of
bytes

Offset

Field

Value

Action

Destination Address

Compare

Source Address

Skip

12a

Possible LLC/SNAP Header

Skip

12a

Possible VLAN Tag

Skip

12a

Type

any

Synchronizing Stream

FF*6+

Compare

any+6

16 copies of Node Address

A*16

Compare

Comment
MAC Header –
processed by main
address filter

Skip

Compared to Receive
Address Register 0
(RAH0, RAL0)

Not applicable to the 82541PI/GI/EI and 82547GI/EI.

A Magic Packet’s destination address must match the address filtering enabled in the configuration
registers with the exception that broadcast packets are considered to match even if the Broadcast
Accept bit of the Receive Control Register (RCTL.BAM) is 0b. If APM Wakeup is enabled in the
EEPROM, Ethernet controller starts up with the Receive Address Register 0 (RAH0, RAL0)
loaded from the EEPROM. This is enables it to accept packets with the matching IEEE address
before the driver comes up.
Note:

Accepting broadcast magic packets for wakeup purposes when the Broadcast Accept bit of the
Receive Control Register (RCTL.BAM) is 0b is a change from the 82544GC/EI Ethernet
controller, which initialized RCTL.BAM to 1b if APM was enabled in the EEPROM, but then
required that bit to be 1b to accept broadcast Magic Packets, unless broadcast packets passed
another perfect or multicast filter.

Software Developer’s Manual

147

Power Management

ARP/IPv4 Request Packet1

6.4.3.1.5

The Ethernet controller supports receiving ARP Request packets for wakeup if the ARP bit is set in
the Wakeup Filter Control Register (WUFC). Four IPv4 addresses are supported which are
programmed in the IPv4 Address Table (IPv4AT)2. A successfully matched packet must contain a
broadcast MAC address, a Protocol Type of 0806h, an ARP OPCODE of 01h, and one of the four
programmed IPv4 addresses. The Ethernet controller also handles ARP Request packets that have
VLAN tagging on both Ethernet II and Ethernet SNAP types.
# of
bytes

Offset

Field

Value

Action

Destination Address

Compare

Source Address

Skip

12
12 + Sa

Possible LLC/SNAP Header

Type<= 1500 and
AAAA_0300_0000ha

Skip

Possible VLAN Tag

8100h and check IDa

Skip

S = (0/4)a

12
12 + D + Sa

Type

0806h

Compare

14
14 + D + Sa

HW Type

0001h

Compare

16
16 + D + Sa

Protocol Type

0800h

Compare

18
18 + D + Sa

Hardware Size

06h

Compare

19
19 + D + Sa

Protocol Address Length

04h

Compare

20
20 + D + Sa

Operation

0001h

Compare

22
22 + D + Sa

Sender HW Address

Ignore

Sender IP Address

Ignore

32
32 + D + Sa

Target HW Address

Ignore

38
38 + D + Sa

Target IP Address

IPv4ATb

Compare

28
28 + D + Sa

a.
b.

1.
2.

D = (0/8)a

Checka

Comment

MAC Header –
processed by main
address filter

Checka
ARP

May match any of 4
values in IPv4ATb

82541PI/GI/EI and 82547GI/EI only.
IPAT for the 82544GC/EI.

ARP Request Packet for the 82544GC/EI.
Four IP addresses are supported which are programmed in the IP Address Table (IPAT) for the 82544GC/EI.

148

Software Developer’s Manual

Power Management

6.4.3.1.6

Directed IPv4 Packet1
The Ethernet controller supports receiving Directed IPv42 packets for wakeup if the IPv4 bit is set
in the WakeUp Filter Control Register (WUFC). Four IPv4 addresses are supported which are
programmed in the IPv4 Address Table (IPv4AT). A successfully matched packet must contain the
station’s MAC address, a Protocol Type of 0800h, and one of the four programmed IPv4 addresses.
The Ethernet controller also handles Directed IPv4 packets that have VLAN tagging on both
Ethernet II and Ethernet SNAP types.

Offset

Field

Destination Address

Source Address

12
12 + Sa

Value

Action

Comment

Compare
Skip
Ignorea
MAC Header –
processed by main
address filter

Possible LLC/SNAP Header

Type<= 1500
and
AAAA_0300_
0000ha

Possible VLAN Tag

8100h and
check IDa

Skip

S = (0/4)a

12
12 + D + Sa

Type

0800h

Compare

14
14 + D + Sa

Version/HDR length

4Xh

Compare

Check IPv4b and
header length

15
15 + D + Sa

Type of Service

Ignore

16
16 + D + Sa

Packet Length

Ignore

18
18 + D + Sa

Identification

Ignore

20
20 + D + Sa

Fragment Info

Ignore

22
22 + D + Sa

Time to live

Ignore

23
23 + D + Sa

Protocol

Ignore

24
24 + D + Sa

Header Checksum

Ignore

26
26 + D + Sa

Source IP Address

Ignore

30
30 + D + Sa

Destination IP Address

IPv4ATb

Compare

a.
b.

D = (0/8)a
4

1.
2.

# of bytes

Skip
Checka

Checka

May match any of four
values in IPv4ATb

82541PI/GI/EI and 82547GI/EI only.
IP instead of IPv4 for the 82544GC/EI.

Directed IP Packet for the 82544GC/EI.
IP instead of IPv4 for the 82544GC/EI.

Software Developer’s Manual

149

Power Management

6.4.3.2

Directed IPv6 Packet1
The Ethernet controller supports receiving Directed IPv6 packets for wakeup if the IPv6 bit is set in
the Wakeup Filter Control Register (WUFC). One IPv6 address is supported and it is programmed
in the IPv6 Address Table (IPv6AT). A successfully matched packet must contain the station’s
MAC address, a Protocol Type of 0800h, and the programmed IPv6 address. The Ethernet
controller also handles Directed IPv6 packets that have VLAN tagging on both Ethernet II and
Ethernet SNAP types.
.

Offset

# of bytes

Field

Destination Address

Source Address

12
12 + Sa

Value

Skip
Ignorea

Possible LLC/SNAP Header

S = (0/4)a

Possible VLAN Tag

8100h and
check IDa

12
12 + D + Sa

Type

14
14 + D + Sa

15
15 + D + Sa

0800h

MAC Header –
processed by main
address filter

Skip
Checka
Skip
Checka

86DDa

Compare

Version/Traffic Classa

6Xh

Compare

Traffic Classa/Flow Label

Ignore

18
18 + D + Sa

Payload Length

Ignore

20
20 + D + Sa

Next Header

IPv6 Next
Header
Typesa

Ignore

21
21 + D + Sa

Hop Limit

Ignore

22
22 + D + Sa

Source IP Address

Ignore

38
38 + D + Sa

Destination IP Address

IPv6AT

Compare

Version/ Priority

Comment

Compare

Type<=
1500 and
AAAA_030
0_0000ha

D = (0/8)a

Action

IP
IPv6a
Check IPv6

Match value in IPv6AT

82541PI/GI/EI and 82547GI/EI only.

Not applicable to the 82544GC/EI.

150

Software Developer’s Manual

Power Management

6.4.3.3

Flexible Filter
The Ethernet controller supports a total of four flexible filters. Each filter is configured to
recognize any arbitrary pattern within the first 128 bytes of the packet. To configure the flexible
filter, the software driver must mask values into the Flexible Filter Mask Table (FFMT), the
required values into the Flexible Filter Value Table (FFVT), and the minimum packet length into
the Flexible Filter Length Table (FFLT). These contain separate values for each filter. The software
driver must also enable the filter in the Wakeup Filter Control Register (WUFC) as well as the
overall wakeup functionality by setting PME_En in the Power Management Control Status
Register or the Wakeup Control Register.
Once enabled, the flexible filters scan incoming packets for a match. If the filter encounters any
byte in the packet where the mask bit is 1b and the byte doesn’t match the byte programmed in the
Flexible Filter Value Table (FFVT), then the filter failed that packet. If the filter reaches the
required length without failing the packet, it passes the packet and generates a wakeup event. It
ignores any mask bits set to 1b beyond the required length. (the wakeup packet is stored, see
Section 6.4.3.5).
For the 82541xx and 82547GI/EI, the flexible filter does not have any way to automatically skip
VLAN or LLC/SNAP headers. If such headers are included, the offsets of the subsequent fields
must be adjusted accordingly.

Note:

6.4.3.3.1

This following flexible packet filters are listed for reference only.

IPX Diagnostic Responder Request Packet Example1

Offset

# of bytes

Field

Value

Action

Destination Address

Compare

Source Address

Skip

S = (0/4)

Possible VLAN Tag

Compare or
Skip

12 + S

D = (0/8)

Possible LLC/SNAP Header

Compare or
Skip

12 + D + S

Type

8137h

Compare

14 + D + S

Some IPX Stuff

Ignore

30 + D + S

IPX Diagnostic Socket

0456h

Compare

Comment
MAC Header –
processed by
main address filter

IPX

82541xx and 82547GI/EI only.

Software Developer’s Manual

151

Power Management

6.4.3.3.2

Directed IPX Packet Example
A valid Directed IPX Packet contains the station’s MAC address, a Protocol Type of 8137h, and an
IPX Node Address that equals to the station’s MAC address. It can include LLC/SNAP Headers
and VLAN Tags. Since filtering this packet relies on the flexible filters, which use offsets specified
by the OS directly, the OS must account for the extra offset LLC/SNAP Headers and VLAN tags.
# of
bytes

Offset

Value

Action

Destination Address

Compare

Source Address

Skip

12
12 + Sa

Possible LLC/SNAP Header

Possible VLAN Tag

D = (0/8)

MAC Header –
processed by main
address filter

Compare or
Skipa
Skip

S = (0/4)

Compare or
Skipa

12
12 + D + Sa

Type

8137h

Compare

14
14 + D + Sa

Some IPX Stuff

Ignore

24
24 + D + Sa

IPX Node Address

Receive
Address 0

Compare

Comment

Skip
a

6.4.3.4

Field

IPX

Must match Receive
Address 0

82541PI/GI/EI and 82547GI/EI.

IPv6 Neighbor Discovery Filter1
In IPv6, a Neighbor Discovery packet is used for address resolution. A flexible filter can be used to
check for a “Neighborhood Discovery Packet”.
82541xx and 82547GI/EI Only
# of
Bytes

Offset

Field

Value

Action

Comment

Destination Address

Compare

Source Address

Skip

Possible VLAN Tag

Compare or
Skip

Possible LLC/SNAP Header

Compare or
Skip

12+D+S

Type

86DDh

Compare

14+D+S

Version/ Traffic Class

6Xh

Compare

Check IPv6

15+D+S

Traffic Class/Flow Label

Ignore

MAC Header –
processed by main
address filter, or
broadcast

Not applicable to the 82544GC/EI.

152

Software Developer’s Manual

Power Management

Offset
18+D+S

6.4.3.5

# of
Bytes
2

Field

Value

Action

Payload Length

Check

Comment

Ignore

20+D+S

Next Header

3Ah, 00h,
2Bh, or
3Ch

21+D+S

Hop Limit

FFh

Check

22+D+S

Source IP Address

Ignore

38+D+S

Destination IP Address

ICMP, or IPv6 next
headers:
+ routing (2Bh)
+ dest options (3Ch)
+ hop-by-hop (00h)

Ignore

54+D+S

Possible IPv6 Next Headers

Check

Process headers to
get next header.
Header type must be
routing, destination
options, or hop-byhop.

54+D+S+N

Type

87h

Check

Neighbor Solicitation

55+D+S+N

Code

00h

Check

56+D+S+N

ICMP Header Checksum

Ignore

58+D+S+N

Reserved

Ignore

62+D+S+N

Target Address

Check

78+D+S+N

Possible source link-layer
address

Ignore

...

any

Ignore

Packet data

last 4

CRC

Compare

Validate correct

Match IPV6AT[0]

Wakeup Packet Storage
The Ethernet controller saves the first 128 bytes of the wakeup packet in its internal buffer, which
can be read through the Wakeup Packet Memory (WUPM) after system wakeup.

Software Developer’s Manual

153

Power Management

Note:

154

This page intentionally left blank.

Software Developer’s Manual

Ethernet Interface

Ethernet Interface
8.1

Introduction
The PCI/PCI-X Family of Gigabit Ethernet Controllers provide a complete CSMA/CD function
supporting IEEE 802.3 (10Mb/s), 802.3u (100Mb/s), 802.3z and 802.3ab (1000Mb/s)
implementations. They perform all of the functions required for transmission, reception and
collision handling called out in the standards.
The internal Gigabit Media Independent Interface/Media Independent Interface (GMII/MII)
supports the onboard 10/100/1000 BASE-T transceivers full duplex operation and supports the
onboard 10/100 BASE-T transceivers in full or half duplex operation.
Note:

8.2

The 82541xx, 82547GI/EI, and 82540EP/EM do support SerDes functionality.

Link Interfaces Overview
82546GB/EB and 82545GM/EM Ethernet controllers contain an internal 10-bit Fibre Channel
Interface (TBI), as specified in IEEE 802.3z, for full-duplex operation with a SerDes transceiver.
This configuration is applicable to 1000BASE-SX, -LX, or –CX links. When in internal SerDes
mode, they provide the full Physical Coding Sub-layer implementation including Auto-Negotiation
as called out in IEEE 802.3z.
82544GC/EI Ethernet controllers support a 10-bit TBI, as specified in IEEE 802.3z, for fullduplex operation with a SerDes transceiver. This configuration is applicable to 1000BASE-SX, LX, or –CX links. When in TBI mode, they provide the full Physical Coding Sub-layer
implementation including Auto-Negotiation as called out in IEEE 802.3z.

Selection between the various configurations is programmable via each MAC’s Extended Device
Control Register (CTRL_EXT.LINK_MODE bits) and defaulted via EEPROM settings. Note that
the external TBI interface is a single resource that can only be associated with a single MAC.
The GMII/MII mode used to communicate between the MAC and the internal PHY supports 10/
100/1000 Mbps operation, with both half- and full-duplex operation at 10/100 Mbps, and fullduplex operation at 1000 Mbps.
Note:

The Ethernet controllers are optimized for full-duplex operation in 1000 Mbps mode. Half-duplex
1000 Mbps operation is NOT supported and is not recommended.
The internal copper PHY features 10/100/1000-BaseT signalling and is capable of performing
intelligent power-management based on both the system power-state and LAN energy-detection
(detection of unplugged cables). Power management includes ability to shut-down to extremely
low (powered-down) state when not needed, as well as ability to auto-negotiate to lower-speed
(and less power-hungry) 10/100 Mbps operation when the system is in low power-states.

Software Developer’s Manual

155

Ethernet Interface

8.2.1

Internal SerDes Interface/TBI Mode– 1Gb/s1
The 82546GB/EB and 82545GM/EM Ethernet controllers contain one or two internal SerDes
devices (depending whether or not they support one or two ports). The MAC communicates with
the SerDes over a TBI interface. Normally, this interface is not exposed externally.
For the 82554GC/EI, TBI mode is selectable via an external pin TBI-MODE. Software cannot
override this pin. This interface has 125 Mb/s 10-bit data paths for both receive and transmit. The
clock at the transmit interface operates at 125 MHz; the receive interface has two clocks running at
62.5 MHz that are 180 degrees out of phase as follows:

• RX_DATA: 10-bit receive data bus
• TX_DATA: 10-bit transmit data bus
• RBC0/I_RBC1: Receive clocks (62.5 MHz; 180 degree phase shift between I_RBC0 and
I_RBCI)

• GTX_CLK: Transmit clock (125 MHz)
8.2.1.1

Gigabit Physical Coding Sub-Layer (PCS) for the Internal SerDes2
The Ethernet controller integrates the 802.3z PCS function on-chip. The on-chip PCS circuitry is
used when the link interface is configured for internal SerDes mode and is bypassed in internal
PHY mode.
The packet encapsulation is based on the Fibre Channel physical layer (FC0/FC1) and uses the
same coding scheme to maintain transition density and DC balance. The physical layer device is a
SerDes and is used for 1000BASE-SX, -LX, or -CX configurations.

8.2.1.2

8B10B Encoding/Decoding
The Gigabit PCS circuitry uses the same transmission coding scheme used in the Fibre Channel
physical layer specification. The 8B10B coding scheme was chosen by the IEEE standards
committee in order to provide a balanced, continuous stream with sufficient transition density to
allow for clock recovery at the receiving station. There is a 25 percent overhead for this
transmission code which accounts for the data signaling rate of 1250 Mb/s with 1000 Mb/s of
actual data.

1.
2.

TBI (10-Bit Interface) - 1GB/s for the 82554GC/EI.
Gigabit Physical Coding Sub-Layer (PCS) for TBI (82554GC/EI).

156

Software Developer’s Manual

Ethernet Interface

8.2.1.3

Code Groups and Ordered Sets
Code group and ordered set definitions are defined in clause 36 of the IEEE 802.3z standard. These
represent special symbols used in the encapsulation of Gigabit Ethernet packets. Table 8-1 lists a
brief description of defined ordered sets for informational purposes only.
Table 8-1. Code Group and Ordered Set Usage
Code

8.2.2

Ordered_Set

# of Code
Groups

Usage

/C/

Configuration

General reference to configuration ordered sets, either
/C1/ or /C2/, which is used during Auto Negotiation to
advertise & negotiate link operation information between
link partners. Last two code groups contain config base
and next page registers.

/C1/

Configuration 1

See /C/. Differs from /C2/ in second code group for
maintaining proper signaling disparity.

/C2/

Configuration 2

See /C/. Differs from /C1/ in second code group for
maintaining proper signaling disparity.

/I/

IDLE

General reference to IDLE ordered sets. IDLE characters
are continually transmitted by the end stations and are
replaced by encapsulated packet data. The transitions in
the IDLE stream allow the SerDes to maintain clock and
symbol synchronization between to link partners.

/I1/

IDLE 1

See /I/. Differs from /I2/ in second code group for
maintaining proper signaling disparity.

/I2/

IDLE 2

See /I/. Differs from /I1/ in second code group for
maintaining proper signaling disparity.

/R/

Carrier_Extend

This ordered set is used to indicate carrier extension to
the receiving PCS. It is also used as part of the
end_of_packet encapsulation delimiter as well as IPG for
packets in a burst of packets.

/S/

Start_of_Packet

The SPD (start_of_packet delimiter) ordered set is used to
indicate the starting boundary of a packet transmission.
This symbol replaces the last byte of the preamble
received from the MAC layer.

/T/

End_of_Packet

The EPD (end_of_packet delimiter) is comprised of three
ordered sets. The /T/ symbol is always the first of these
and indicates the ending boundary of a packet.

/V/

Error_Propagation

The /V/ ordered set is used by the PCS to indicate error
propagation between stations. This is normally intended to
be used by repeaters to indicate collisions.

GMII – 1 Gb/s
The internal Gigabit Media Independent Interface (GMII) is similar to the 10/100 Mb/s Media
Independent Interface (MII). The GMII uses the MII management interface and registers. These
common elements of operation allow the Ethernet controller to determine PHY capabilities for any
supported speed of operation and configuration of the hardware based on those capabilities.
Most of the MII and GMII signals use the same names, but the width of the RX and TX data busses
and the semantics of the associated control signals differ between MII and GMII operation. The
GMII transmit path clocking also differs significantly from MII clocking.

Software Developer’s Manual

157

Ethernet Interface

8.2.3

MII – 10/100 Mb/s
The internal MII implementation for the Ethernet controller provides full IEEE 802.3 and IEEE
802.3u compliant operation for 10Mb/s and 100Mb/s operation in conjunction with the onboard
MII compliant PHY.
The MII uses a clocked, nibble-wide (4-bit) data path in each direction. The clock rate for Fast
Ethernet operation is 25 MHz with data transfer speed of 4 bits x 25 MHz = 100 Mb/s. For
10 Mb/s operation the clock rate is 2.5 MHz and also uses the nibble-wide data path.

8.3

Internal Interface1
The Ethernet controller supports the IEEE 802.3 MII Management Interface also known as the
Management Data Input/Output (MDI/O) Interface. This interface allows upper-layer devices to
monitor and control the state of the PHY.
For the 82546GB/EB, 82545GM/EM, 82541xx, 82540EP/EM, and 82547GI/EI, the MDI/O
interface consists of an internal connection, a special protocol that runs across the connection, and
an internal set of addressable registers. For the 82541xx and 82547GI/EI, the physical interface
between the MAC and PHY is not available externally.
For the 82544GC/EI, the MDI/O interface consists of a physical connection, a special protocol that
runs across the connection, and an internal set of addressable registers. The physical interface
consists of a (B_MDIO) data line and a clock line (O_MDC).

• O_MDC:
Management Data Clock, used by the PHY as a clock timing reference for information transfer
on the B_MDI/O signal. The O_MDC is not a continuous signal and can be frozen by the
Ethernet controller when no management data is transferred. The O_MDC signal has a
maximum operating frequency of 2.5 MHz.

• B_MDO:
Management Data I/O, a bidirectional data signal used to transfer control information and
status between the Ethernet controller and the PHY (read and write PHY management
registers). The B_MDO signal is sampled by the rising edge of the O_MDC signal.
Software can use MDI/O to read and write registers in a internal PHY by accessing the Ethernet
controller’s MDIC register.

8.4

Duplex Operation
The 82546GB/EB and 82545GM/EM supports half-duplex and full-duplex 10/100 Mb/s mode.
Half-duplex in 1000 Mb/s mode using either the Internal SerDes or GMII interface is NOT
supported.
The 82544GC/EI, 82540EP/EM, 82541xx, and 82547GI/EI, support half-duplex and full-duplex
10/100 Mb/s mode or 1000 MB/s mode. However, only full-duplex mode is supported when the
82544GC/EI TBI interface option is used.

MDIO/MDC Interface for the 82544GC/EI, 82540EP/EM, 82541xx, and 82547GI/EI.

158

Software Developer’s Manual

Ethernet Interface

Configuration of the duplex operation of the Ethernet controller can be forced or determined via
the Auto-Negotiation process. See Section 8.6 for details on link configuration setup and
resolution.

8.4.1

Full Duplex
All aspects of the IEEE 802.3, 802.3u, 802.3z, and 802.3ab specifications are supported in full
duplex operation. Full duplex operation is enabled by several mechanisms depending on the speed
configuration of the Ethernet controller and the specific capabilities of the PHY used in the
application. During full duplex operation, the Ethernet controller may transmit and receive packets
simultaneously across the link interface.
In Internal Serdes mode for the 82546GB/EB and 82545GM/EM (TBI mode for the 82544GC/
EI), the transmission and reception of packets is indicated by symbols embedded in the data
stream. These symbols delineate the packet encapsulation and the protocol does not rely on other
control signals. See Section 8.2.1.3 for details.

8.4.2

Half Duplex
Note:

The Ethernet controller operates in half duplex mode only when configured for internal PHY
mode. For the 82546GB/EB and 82545GM/EM, internal SerDes mode does not support half
duplex operation.
In half duplex mode, the Ethernet controller attempts to avoid contention with other traffic on the
wire, by monitoring the carrier sense signal provided by the internal PHY, and deferring to passing
traffic. When the Internal Carrier Sense signal is deasserted or after sufficient InterPacket Gap
(IPG) has elapsed after a transmission, frame transmission can begin.
In the case of a collision, the internal PHY asserts a collision signal. Transmission of the frame
stops within four clock times and then the Ethernet controller sends a JAM sequence onto the link.
After the end of a collided transmission, the Ethernet controller backs off and attempts to
retransmit per the standard CSMA/CD method. Note that the retransmission is done from the data
stored internally in the Ethernet controller transmit packet buffer. The Ethernet controller does not
access data in host memory again.
In the case of a successful transmission, the Ethernet controller is ready to transmit any other
frames queued in its transmit FIFO within the minimum Inter Frame Spacing (IFS) of the link.
The internal carrier sense signal is expected to be asserted before one slot time has elapsed;
however, the transmission completes successfully even if internal carrier sense is not asserted. If
internal carrier sense is not asserted within the slot time window, the PHY is not behaving properly
and can either be configured incorrectly or be in a link down situation. Note that this event is
counted in the Transmit Without CRS statistic register (see Section 13.7.12).
Half duplex reception is as indicated for full duplex in Section 8.4.1 except for 1000 Mb/s specific
operation, as described in Section 8.4.2.1 and Section 8.4.2.2.
The Ethernet controller does not provide support for half-duplex operation as specified in the IEEE
802.3z specification when operating at 1000 Mb/s in internal PHY mode.

Software Developer’s Manual

159

Ethernet Interface

For receives, the Ethernet controller supports carrier extended packets and packets generated
during packet bursting operations (see Section 8.4.2.1 and Section 8.4.2.2). The Ethernet controller
can be configured to transmit in packet burst mode via the TCTL.PBE bit in the Transmit Control
register (see Section 13.4.46).
Carrier extension is only defined in the IEEE 802.3z standard for half-duplex operation for
operation frequencies above 100 Mb/s (Gigabit Ethernet).

8.4.2.1

Carrier Extension (1000 Mb/s Only)
One of the objectives of the IEEE 802.3z standard development was to support a maximum
collision domain of 200 m and retain the IEEE 802.3 Ethernet frame format. The scaling of the line
transfer rate by 10x to 1 Gb/s reduced the bit time by 10 and effectively reduced the theoretical
collision domain to an unusable size with the minimum packet size of 64 bytes. To overcome this,
the 802.3z specification development added the notion of carrier extension to the standard.
Carrier extension provides a method to increase the duration of the carrier event to a minimum
usable duration in order to meet the collision domain objective. Packets that are signaled from the
CSMA/CD layer that do not meet the minimum slot time of 512 bytes have extension bytes
appended to them in order to meet this minimum slot time requirement. The extension bytes are
defined within the context of the frame encapsulation discussion of the 802.3z standard and are
recognized by 802.3z compliant devices (see Figure 8-1).

Preamble

SFD

T/L

Data/Pad

FCS

Extension

Minimum Frame Size
Slot Time
Duration of Carrier Event

Figure 8-1. Carrier Extended Frame Format

The Ethernet controller supports the reception and transmission of carrier extended packets. Carrier
extension is implemented via specifying the collision distance parameter, COLD, in the Transmit
Control register (TCTL). Note that this field is evaluated whether in full- or half- duplex operation.

8.4.2.2

Packet Bursting
In an attempt to recover some of the lost overhead encountered with short duration packets using
carrier extension, the IEEE 802.3z standard incorporates the implementation of packet bursting.
Packet bursting is a mechanism that allows a transmitting device to “own-the-wire” for a longer
duration and “pack” extra packets in a burst without relinquishing ownership of the medium. A
burst length timer is implemented which allows the Ethernet controller to continue to send packets
until the timer expires (if packets are available for transmission).
In the case where a transmitting station has more than one packet to send, it can transmit the first
packet (extending to 512 bytes if necessary) and then begin the transmission of subsequent packets.
Packet transmission can continue until either there are no more packets ready for transmission, or
the burst timer has expired. The burst timer limit is specified as 8 KB.

160

Software Developer’s Manual

Ethernet Interface

The normal rules for IPG are followed during packet bursting after the first packet has met the
minimum slot time requirements, with the exception that the Inter Frame Content (IFC) is
extension symbols rather than IDLEs. Under some circumstances, it might be desirable to extend
this IPG time during a burst. This can be done via the AIFS field in the AIT register. See Section
13.4.35.

8.5

Auto-Negotiation and Link Setup1
Configuration of the link can be accomplished by several methods ranging from software forcing
the link settings to Auto-Negotiation by the internal PHY. Section 8.6.1 describes the process of
bringing the link up including configuration of the MAC and PHY, as well as the various methods
of determining duplex and speed configuration.
The PHY performs Auto-Negotiation per 802.3ab clause 40 and extensions to clause 28. Link
resolution is obtained from the PHY after the link has been established via the MDIO interface, by
the controller via specific input signals from the PHY or by the controller’s specific auto detection
functions.
Upon power up, or device reset via the RST# input, the controller initiates Auto-Negotiation based
on the default settings in the Device Control and Transmit Configuration Word registers, as well as
settings read from the EEPROM. If enabled in the EEPROM, the Ethernet controller will immediately perform Auto-Negotiation.

8.6

Auto-Negotiation and Link Setup2
Configuration of the link can be accomplished by several methods ranging from software forcing
the link settings to Auto-Negotiation by either the MAC (Internal Serdes mode) or the internal
PHY (GMII/MII mode). The following sections describe the process of bringing the link up
including configuration of the MAC and PHY, as well as the various methods of determining
duplex and speed configuration.
The process of determining link configuration differs slightly depending on the Ethernet controller
type and version. In Internal Serdes mode, the MAC performs Auto-Negotiation per clause 37 of
the 802.3z standard. The transceiver used in this mode (the SerDes) does not participate in the
Auto-Negotiation process as all aspects of Auto-Negotiation are controlled by the MAC.
For internal PHY mode, the PHY performs Auto-Negotiation per 802.3ab clause 40 and extensions
to clause 28. Link resolution is obtained from the PHY after the link has been established via the
MDI/O interface, by the Ethernet controller via specific input signals from the PHY, or by the
Ethernet controller’s specific auto detection functions.
The method for configuring the link between two link partners is highly dependent on the mode of
operation as well as the functionality provided by the specific physical layer device (PHY or
SerDes). For Internal Serdes mode, the Ethernet controller provides the complete 802.3z PCS
function on-chip. For GMII/MII mode, the PCS and Auto-Negotiation functions are maintained
within the PHY.

1.
2.

82541xx, 82547GI/EI, and 82540EP/EM only.
Applicable to the 82546GB/EB, 82545GM/EM, and 82544GC/EI only.

Software Developer’s Manual

161

Ethernet Interface

The following section describes the link configuration process in the Internal Serdes for the
82546GB/EB and 82545GM/EM (TBI mode for the 82544GC/EI) and internal PHY modes.

8.6.1

Link Configuration in Internal Serdes/TBI Mode1
Internal Serdes for the 82546GB/EB and 82545GM/EM (TBI for the 82544GC/EI) Mode link
configuration can be performed via the on-chip PCS function in the Ethernet controller. The
hardware supports both hardware and software Auto-Negotiation methods for determining the link
configuration, as well as allowing for manual configuration to force the link.
Hardware Auto-Negotiation is the preferred method.

8.6.1.1

Link Speed
Internal Serdes for the 82546GB/EB and 82545GM/EM (TBI for the 82544GC/EI) Mode is only
defined for 1000 Mb/s operation. Other link speeds are not supported.
When the 82546GB/EB and 82545GM/EM is in internal Serdes mode, the speed determination
function is disabled and the Device Status register bits (STATUS.SPEED) bits indicate a value of
10b for 1000 Mb/s.
For the 82544GC/EI, when the TBI_MODE input is asserted for TBI mode, the speed
determination function is disabled and the Device Status register bits (STATUS.SPEED) bits
indicate a value of 10b for 1000 Mb/s.

8.6.1.2

Auto-Negotiation
At power up, or Ethernet controller reset via the RST# input, it initiates Auto-Negotiation based on
the default settings in the Device Control and Transmit Configuration Word registers, as well as
settings read from the EEPROM. If enabled in the EEPROM, the Ethernet controller immediately
performs Auto-Negotiation.
TBI Mode Auto-Negotiation, as defined in clause 37 of the IEEE 802.3z standard, provides a
protocol for two Ethernet controllers to advertise and negotiate a common operational mode across
a Gigabit Ethernet link. The Ethernet controller fully supports the IEEE 802.3z Auto-Negotiation
function when using the internal Serdes mode for the 82546GB/EB and 82545GM/EM or when
using the TBI and on-chip PCS for the 82544GC/EI.
TBI Mode Auto-Negotiation is used to determine the following information:

• Duplex resolution
• Flow control configuration
Speed for Internal Serdes mode (TBI mode for the 82544GC/EI) is fixed at 1000 Mb/s, so speed
settings in the Device Control register are unaffected by the Auto-Negotiation process.
There are two implementations accessible in the design:
1. A full hardware Auto-Negotiation implementation that does not require software intervention
in order to successfully reach a negotiated link configuration.
2. Software driven negotiation.
1.

TBI mode for the 82544GC/EI.

162

Software Developer’s Manual

Ethernet Interface

A set of registers is provided to facilitate either hardware or software Auto-Negotiation.
The hardware supports both hardware and software Auto-Negotiation methods for determining
link configuration as well as allowing for manual configuration to force the link. The IEEE 802.3z
specification defines a set of resources that software can use to control a hardware implementation
of Auto-Negotiation, but this definition is sub-optimal for Internal Serdes mode (TBI mode for the
82544GC/EI) and hardware Auto-Negotiation is the preferred method.
In addition, it specifies optional resources that exist only to support the exchange of “Next Pages”,
something that is not required for the Ethernet controller. The hardware defined in this
specification accepts and exchanges next pages in Internal Serdes mode (TBI mode for the
82544GC/EI), but does so by dropping all incoming next pages and sending only null next pages.
The Ethernet controller can only send null next pages when in hardware Auto-Negotiation. A full
next page exchange can take place if software performs Auto-Negotiation.
The Ethernet controller fully complies with IEEE 802.3z with respect to next page exchange in that
both link partners must request next page exchange in order to do so.

8.6.1.3

Hardware Auto-Negotiation
Hardware supports negotiation of the link configuration per clause 37 of the 802.3z standard. This
is accomplished by the exchange of /C/ ordered sets that contain the txConfigWord register values
from TXCW in the third and fourth symbols of the ordered sets.
Bits FD and LU of the Device Status register (STATUS), and ANC of the RXCW register provide
status information regarding the negotiated link.
Auto-Negotiation can be initiated by the following:

•
•
•
•
•

LRST transition from 1b to 0b in CTRL register
ANE transition from 0b to 1b in TXCW register
Receipt of /C/ ordered set during normal operation
Receipt of different value of the /C/ ordered set during the negotiation process
Transition from loss of synchronization to synchronized state (if ANE is enabled)

Resolution of the negotiated link determines device operation with respect to flow control
capability and duplex settings. These negotiated capabilities override advertised and S/W
controlled device configuration.
Software must configure the TXCW.txConfigWord field to the desired advertised base page. The
bits in the Device Control register are not mapped to the txConfigWord field in hardware until after
Auto-Negotiation completes. The Figure 8-2 and Figure 8-3 show txConfigWord and the mapping
to the Config_reg Base Page encoding per clause 37 of the standard. Table 8-2 lists the bit contents.
15

RS
V

Figure 8-2. TXCW.txConfigWord
15
Np

14
Ack

13
Rf2

Software Developer’s Manual

Rf1

RS
V

8
Ps2

7
Ps1

6
Hd

RS
V

163

Ethernet Interface

Figure 8-3. 802.3z Advertised Base Page Mapping
Table 8-2. Bits Content in TXCW.txConfigWord
Bit

Description

Next Page Indication
When set indicates a request for next page exchange

Asymmetric Pause Connection is Desired
When set, results in independent enabling/disabling of the flow control
receive and transmit. When cleared, results in symmetric enabling/disabling
of the flow control receive and transmit

Pause Function
When set, indicates that the Ethernet controller is capable and intends to
stop upon reception of 802.3x flow control Pause packets.
When cleared, indicates that the Ethernet controller is not capable, or does
not intend to stop upon reception of flow control Pause packets.

Half-Duplex Ability
When set, indicates that the Ethernet controller is capable of working in
half-duplex mode of operation

Full Duplex Ability
When set, indicates that the Ethernet controller is capable of working in fullduplex mode of operation

RSV

Reserved
Should be written as 0b

The reserved bits should be written as zero. The remote fault bits [13:12] can be set by software to
indicate remote fault type to the link partner if desired. The AS and PS bits are used for
advertisement of PAUSE frame operation. Refer to clause 37 of the 802.3z specification for details.

8.6.1.4

Software Auto-Negotiation
Auto-Negotiation can also be performed by software with TXCW.ANE set to 0b. Data stored in the
txConfigWord field is transmitted during the configuration process. Software should not (in
general) read back the contents of this register.
If hardware loses receive synchronization, the contents of the TXCW register changes and during
the time of the change, the value read back can be inconsistent. In the absence of loss of
synchronization, the value read back is stable and equal to the last written value.
Software controls the negotiation process by writing the appropriate values to the txConfigWord
and transmitting /C/ ordered sets by setting txConfig (in TXCW) to 1b. Software must monitor the
RXCW register for status of the negotiation process and respond via writes to the TXCW register
appropriately.
The software algorithm must follow the state machine implementation of sub-clause 37.3.1.5 of
IEEE 802.3z, Figure 37-6. The link timer specification is 10 ms (+10 ms/-0 ms). In some systems,
response time for the S/W implementation can make it difficult to meet this requirement if system
utilization is high due to latencies on the PCI bus.
For more information, refer to the register definitions for TXCW and RXCW in Sections 13.4.13
and 13.4.14, respectively.

164

Software Developer’s Manual

Ethernet Interface

8.6.1.5

Forcing Link
In cases where the Ethernet controller is connected to a non-Auto-Negotiating link partner, the
hardware allows for manual configuration of the link via the Device Control register (CTRL).
Forcing link can be accomplished by software writing a 1b to CTRL.SLU which forces the TBI
PCS logic into a link up state if the LOS input is not asserted. Setting the SLU bit enables the MAC
to communicate with the internal SerDes and allows recognition of the LOS signal. If autonegotiation is enabled (TXCW.ANE = 1b) Set Link Up is ignored. The LINK UP# output, as well
as internal status logic, indicates link status.
The TXCW.ANE bit must be set to logic 0b to allow for forcing link. When link is forced via the
CTRL.SLU bit, the link cannot come up unless the LOS input is deasserted, implying there is a
valid signal being received by the optics or the SerDes.
An interrupt bit, RXCFG, flags software that the hardware is receiving configuration symbols (/C/
codes). Software should mask (enable) this interrupt when forcing link. When the link is forced, the
link partner can begin to Auto-Negotiate based due to a reset or enabling of Auto-Negotiation. The
reception of /C/ codes causes an interrupt to software and the proper hardware configuration can be
set.

8.6.2

Internal GMII/MII Mode
Link configuration in GMII/MII mode is generally determined by the PHY via Auto-Negotiation.
The software driver must intervene in cases where a successful link is not negotiated or a
programmer desires to manually configure the link. The following sections discuss the methods of
link configuration for internal PHY mode.

8.6.2.1

Auto-Negotiation
In GMII/MII mode, the PHY performs the Auto-Negotiation function. The operational details of
this function are described in the IEEE P802.3ab draft standard.
Auto-Negotiation provides a method for two link partners to exchange information in a systematic
manner in order to establish a link configuration providing the highest common level of
functionality supported by both partners. Once configured, the link partners exchange
configuration information to resolve link settings such as:

• Speed: 10/100/1000 Mb/s
• Duplex: Full- or Half• Flow Control Operation
PHY specific information required for establishing the link is also exchanged, but is not relevant to
the operation of the Ethernet controller.
If flow control is enabled in the MAC, the settings for the desired flow control behavior must also
be made by software in the PHY and Auto-Negotiation must be restarted. After Auto-Negotiation
completes, the software driver must read the MII registers in the PHY to determine the resolved
flow control behavior of the link and reflect these parameters in the Ethernet controller register
(CTRL.TFCE and CTRL.RFCE).

Software Developer’s Manual

165

Ethernet Interface

Once PHY Auto-Negotiation is complete, the PHY asserts the link indication signal. Software
MUST set the “set link up” bit in the Device Control Register (CTRL.SLU) before the Ethernet
controller recognizes the link. Setting the SLU bit permits the MAC to recognize the LINK signal
from the PHY, which indicates the PHY has gotten the link up, and to receive and transmit data.

8.6.2.2

Link Speed
The speed of the link in GMII/MII mode can be determined by several methods with the Ethernet
controller. These include:

• Software forced configuration of link speed
• Automatically detecting the Auto-Negotiated speed from the PHY
• Direct indication of speed configuration from the PHY
These methods are discussed in the following sections.
8.6.2.2.1

Forcing Speed

There can be circumstances when the software driver must force the link speed of the Ethernet
controller. This can occur when the link is manually configured.
The software driver can force speed in the MAC by setting the CTRL.FRCSPD (force-speed) bit to
1b, and then setting the speed bits in the Device Control register (CTRL.SPEED) to the desired
speed setting. See Section 13.4.1 for details.
When forcing the Ethernet controller to a specific speed configuration, the driver must also ensure
the PHY is configured to a speed setting consistent with the MAC. This statement implies that
software accesses to the PHY either force the speed, or read the MII management status register
bits that indicate link speed within the PHY itself.
Forcing the speed setting with CTRL.SPEED also can be accomplished by setting the
CTRL_EXT.SPD_BYPS bit. This bit bypasses the internal clock switching logic, and gives
complete control to the driver when the speed setting takes place. The CTRL.FRCSPD bit uses the
internal clock switching logic, which delays the effect of the speed change.
8.6.2.2.2

Using Auto-Speed Detection (ASD)

The Ethernet controller provides a method in hardware for automatically sensing the speed of the
link by observing the receive clock signal generated by the PHY once the link is established. The
Auto-Speed Detection (ASD) function is enabled via the ASDE bit in the Device Control register
(CTRL.ASDE). ASD provides a method of determining the link speed without the need for
software accesses to the MII management registers. ASD is not supported in Internal Serdes mode
for the 82546GB/EB and 82545GM/EM or TBI mode for the 82544GC/EI.
In internal PHY mode, the internal receive clock input operates at the byte rate of the link interface.
By sensing this clock, the Ethernet controller makes a determination of the link speed and sets the
proper configuration in the control registers without software intervention.
The ASD function is initiated upon the assertion of a valid link by the PHY via an internal signal
input. After the speed is detected, the Device Control and Device Status register bits are set and
reflect the speed of the link. As described earlier, software must set the CTRL.SLU bit to allow the
speed selection to take effect.

166

Software Developer’s Manual

Ethernet Interface

STATUS.ASDV [9:8], provides the results of speed status indication for diagnostics purposes
regardless of whether the Auto-Speed Detection feature is enabled. This function is initiated with a
write to the CTRL_EXT.ASDCHK bit. See Section 13.4.6 for details.
8.6.2.2.3

Automatic Detection of Link Speed using SPD-IND

With the CTRL register configure as CTRL.FRCSPD = 0, the speed is reconfigured automatically
each time a new linkup event is detected. This configuration is recommended why the PHY is
configured for Auto-Negotiation.

8.6.2.3

Duplex
The duplex configuration of the link is also resolved during the Auto-Negotiation process. As
previously mentioned, the Ethernet controller supports both full- and half-duplex operation in
internal PHY mode. When the PHY asserts its link signal to the MAC, it also communicates the
duplex setting.
Software can override the duplex setting via the CTRL.FD bit when the CTRL.FRCDPLX (force
duplex) bit is set. If CTRL.FRCDPLX is 0b, the CTRL.FD bit is ignored.

8.6.2.4

MII Management Registers
The software driver is required under some circumstances to read from, or write to, the MII
management registers in the PHY. These accesses are performed via the MDIC registers. The MII
registers allow the software driver to have direct control over the PHY’s operation, which includes:

•
•
•
•
•
•

Resetting the PHY
Setting preferred link configuration for advertisement during the Auto-Negotiation process
Restarting the Auto-Negotiation process
Reading Auto-Negotiation status from the PHY
Forcing the PHY to a specific link configuration
Extended capabilities

The standard set of PHY management registers can be found in the IEEE P802.3ab standard.

8.6.2.5

Comments Regarding Forcing Link
Forcing link in GMII/MII mode requires the software driver to configure both the MAC and the
PHY in a consistent manner with respect to each other as well as the link partner. After
initialization, the software driver configures the desired modes in the MAC, then accesses the PHY
MII registers to set the PHY to the same configuration.
In internal PHY mode, setting the CTRL.SLU bit forces a link up condition in the MAC. The
duplex setting at this point should be forced by software on the CTRL.FD bit.

Software Developer’s Manual

167

Ethernet Interface

8.6.3

Internal SerDes Mode1 Control Bit Resolution
Tables 8-3, 8-4, and 8-52 list how on-chip Auto-Negotiation affects control bits in the Ethernet
controller. Table 8-5 lists the case where software Auto-Negotiation is not performed and link is
forced.
Table 8-3. Internal Serdes Mode1 – Hardware Enabled

TXCW.ANE = 1b
Control Bit

Effect on Control Bits

CTRL.FD

Ignored; duplex is set by priority resolution of TXCW and RXCW

CTRL.SLU

Ignored; it is not possible to force link configuration (ANE takes precedence)

CTRL.RFCE

Set by priority resolution (read only)

CTRL.TFCE

Set by priority resolution (read only)

CTRL.SPEED

No impact; speed always 1000 Mb/s in Internal Serdesa mode

STATUS.FD

Set by priority resolution

STATUS.LU

Duplicate of RXCW.ANC (Auto-Negotiation complete)

STATUS.SPEED

Internal SerDesa Mode is always 1000 Mb/s; fixed at 10b

TBI for the 82544GC/EI.

Table 8-4. Internal Serdes1 Mode – Software Enabled

TXCW.ANE = 0b
Control Bit

Effect on Control Bits

CTRL.FD

Duplex is set by software priority resolution

CTRL.SLU

Set by software when Auto-Negotiation is complete.

CTRL.RFCE

Set by software as a result of software priority resolution

CTRL.TFCE

Set by software as a result of software priority resolution

CTRL.SPEED

No impact; speed always 1000 Mb/s in Internal SerDesa mode

STATUS.FD

Reflects the value of CTRL.FD

STATUS.LU

Reflects CTRL.SLU and internal link indication

STATUS.SPEED Internal Serdesa Mode is always 1000 Mb/s; fixed at 10b
a.

1.
2.

TBI for the 82544GC/EI.

TBI Mode for the 82544GC/EI.
Not applicable to the 82541xx, 82547GI/EI, or 82540EP/EM.

168

Software Developer’s Manual

Ethernet Interface

Table 8-5. Internal Serdes Mode1 – Auto-Negotiation Skipped

TXCW.ANE = 0b
Control Bit
CTRL.FD

Duplex is set by software for the desired mode of operation

CTRL.SLU

Set by software

CTRL.RFCE

Set by software for the desired mode of operation

CTRL.TFCE

Set by software for the desired mode of operation

CTRL.SPEED

No impact; speed always 1000 Mb/s in Internal SerDesa mode

STATUS.FD

Reflects the value of CTRL.FD

STATUS.LU

Reflects CTRL.SLU and internal link indication

STATUS.SPEED

Internal SerDesa Mode is always 1000 Mb/s; fixed at 10b

8.6.4

Effect on Control Bits

TBI for the 82544GC/EI.

Internal PHY Mode Control Bit Resolution
Tables 8-6, 8-7, 8-8, and 8-9 list how Auto-Negotiation affects control bits in the Ethernet
Controller.
Refer to IEEE 802.3z, clause 37 for information related duplex and flow control link resolution per
the 802.3z Auto-Negotiation method. The Ethernet controller fully complies to the specified
resolution functions.
Table 8-6. GMII/MII Mode – PHY Speed Indication

CTRL.FRCSPD = CTRL.ASDE = CTRL.FRCDPLX = 0b
Control Bit

Effect on Control Bits

CTRL.FD

Duplex is set per internal signal after link up assertion by PHY.

CTRL.SLU

Software should set to allow PHY to control.

CTRL.RFCE

Must be set by software after reading flow control resolution from MII
registers.

CTRL.TFCE

Must be set by software after reading flow control resolution from MII
registers.

CTRL.SPEED

Ignored; no impact on speed.

STATUS.FD

Reflects the value of CTRL.FD as above.

STATUS.LU

Reflects link status and SLU set.

STATUS.SPEED

Speed status bits reflect speed resolved from speed indication inputs from
PHY.

TBI for the 82544GC/EI.

Software Developer’s Manual

169

Ethernet Interface

Table 8-7. GMII/MII Mode – Auto-Speed Detection

CTRL.FRCSPD = CTRL.FRCDPLX = 0b; CTRL.ASDE = 1b
Control Bit

Effect on Control Bits

CTRL.FD

Duplex is set per internal duplex indication after link up assertion by PHY.

CTRL.SLU

Software should set to allow PHY to control.

CTRL.RFCE

Must be set by software after reading flow control resolution from MII
registers.

CTRL.TFCE

Must be set by software after reading flow control resolution from MII
registers.

CTRL.SPEED

Ignored; no impact on speed.

STATUS.FD

Reflects the value of CTRL.FD as above.

STATUS.LU

Reflects internal link status

STATUS.SPEED

Speed status bits reflect speed resolved from ASD function.

Table 8-8. GMII/MII Mode – Force Speed

CTRL.FRCSPD = 1b; CTRL.FRCDPLX = 0b; CTRL.ASDE = X
Control Bit

Effect on Control Bits

CTRL.FD

Duplex is set per internal duplex indicates input after link up assertion by PHY.

CTRL.SLU
CTRL.RFCE

Must be set by software after reading flow control resolution from MII
registers.

CTRL.TFCE

Must be set by software after reading flow control resolution from MII
registers.

CTRL.SPEED

Set by software to set speed of the MAC; must match PHY speed settings.

STATUS.FD

Reflects the value of CTRL.FD.

STATUS.LU

Reflects internal link status.

STATUS.SPEED

170

Software should set to allow PHY to control.

Speed status bits reflect speed forced by CTRL.SPEED.

Software Developer’s Manual

Ethernet Interface

Table 8-9. GMII/MII Mode – Force Link

CTRL.FRCSPD = CTRL.FRCDPLX = CTRL.SLU = 1b
Control Bit

8.6.5

Effect on Control Bits

CTRL.FD

Set by software.

CTRL.SLU

Set by software, and assumed PHY is also forced to assert link.

CTRL.RFCE

Set by software for the desired mode of operation.

CTRL.TFCE

Set by software for the desired mode of operation.

CTRL.SPEED

Set by software.

STATUS.FD

Reflects the value of CTRL.FD.

STATUS.LU

Reflects CTRL.SLU set and internal link status.

STATUS.SPEED

Reflects CTRL.SPEED.

Loss of Signal/Link Status Indication
For the 82546GB/EB and 82545GM/EM, the internal LOS signal allows for indication of physical
link status to the Ethernet controller’s MAC.
For the 82544GC/EI, the LOS input is provided to allow for indication of physical link status to
the Ethernet controller. When the 82544GC/EI is configured in TBI mode, the input is typically
connected to the loss-of-signal connection from the optics while in internal PHY mode.
If the LSC (Link Status Change) interrupt is enabled, the hardware posts an interrupt to be serviced
by the software driver when the link goes up or down. See Section 3.4.3 for more details.

8.6.5.1

Internal Serdes Mode 1
When asserted, LOS indicates there is no activity on the fiber due to either an unplugged cable or a
defective optical device. An assertion on LOS implies the link is not available and the hardware is
disabled. This is true whether the link is forced by the Set Link Up bit (CTRL.SLU) or if the
Ethernet controller is configured to perform Auto-Negotiation. When Auto-Negotiation is enabled,
the Ethernet controller is forced to restart Auto-Negotiation but does not complete the negotiation
process until the LOS is deasserted.

8.6.5.2

Internal PHY Mode
While in internal PHY mode, an internal signal provides status of the physical link as indicated by
the PHY. Indication that the link is not up disables MAC operation. Upon determination of a valid
link, the assertion of the internal link signal asserts the LSC interrupt (if enabled) to indicate to the
software driver to check the link status.

TBI mode for the 82544GC/EI.

Software Developer’s Manual

171

Ethernet Interface

8.7

10/100 Mb/s Specific Performance Enhancements

8.7.1

Adaptive IFS1
The Ethernet controller supports back-to-back transmit Inter-Frame-Spacing (IFS) of 960 ns in 100
Mb/s operation and 9.6 s in 10 Mb/s operation. Although back-to-back transmission is normally
desirable, sometimes it can actually hurt performance in half-duplex environments due to excessive
collisions. Excessive collisions are likely to occur in environments where one station is attempting
to send large frames back-to-back, while another station is attempting to send acknowledge (ACK)
packets.
The Ethernet controller contains an Adaptive IFS Throttle - AIT register (see Section 13.4.35) that
enables the implementation of a driver-based adaptive IFS algorithm for collision reduction.
Adaptive IFS throttles back-to-back transmissions in the transmit MAC and delays their transfer to
the CSMA/CD transmit function. Normally, this register should be set to zero. However, if
additional delay is desired between back-to-back transmits, then this register can be set with a
value greater than zero. By setting this register with a higher value, collisions can be reduced in
certain half-duplex environments, because the adapter is less aggressive in acquiring the wire, and
therefore less likely to collide with another adapter that attempts to transmit after minimum IFS.
Note:

IFS and IPG (inter-packet gap) are equivalent terms and may be used interchangeably in this
manual.
The AIFS field provides a similar function to the IGPT field in the TIPG register (see Section
13.4.34). However this Adaptive IFS throttle register counts in units of transmit clocks (which are
8 ns, 80 ns, 800 ns for 10, 100, 1000 Mb/s mode respectively), and is 16 bits wide, thus providing a
greater maximum delay value.
Using values lower than a certain minimum (determined by the ratio of transmit clock to link
speed), has no effect on back-to-back transmission. This is because the Ethernet controller does not
start transmission until the minimum IEEE IFS (9.6 us at 10 Mb, 960 ns at 100 Mb, and 96 ns at 1
Gb) has been met regardless of the value of Adaptive IFS. For example, if the Ethernet controller is
configured for 100 Mb/s operation, the minimum IEEE IFS at 100 Mb/s is 960 ns. Setting AIFS to
a value of 10 (decimal) would not affect back-to-back transmission time on the wire, because the
800 ns delay introduced (10 * 80n s = 800 ns) is less than the minimum IEEE IFS delay of 960 ns.
However, setting this register with a value of 20 (decimal), which corresponds to 1600 ns for the
above example, would delay back-to-back transmits because the ensuing 1600 ns delay is greater
than the minimum IFS time of 960 ns.
It is important to note that this register has no effect on transmissions that occur immediately after
receives or on transmissions that are not back-to-back (unlike the IPGR1 and IPGR2 values in the
TIPG register described in Section 13.4.34). In addition, Adaptive IFS also has no effect on retransmission timing (re-transmissions occur after collisions). Therefore, AIFS is only enabled in
back-to-back transmission. The AIFS value is NOT additive to the TIPG.IPGT value; instead, the
actual IPG equals the larger of AIFS and TIPG.IPGT.

Not applicable to the 82541xx or 82547GI/EI.

172

Software Developer’s Manual

Ethernet Interface

8.7.2

Flow Control
Flow control as defined in IEEE specification 802.3x, as well as the specific operation of
asymmetrical flow control defined by 802.3z, are supported. The following registers are defined
for the implementation of flow control:
Table 8-10. Flow Control Registers
Register Name

Description

Flow Control Address Low, High (FCAL/H)

6-byte flow control multicast address

Flow Control Receive Thresh Hi (FCRTH)

13-bit high water mark indicating receive
buffer fullness

Flow Control Transmit Timer Value (FCTTV)

16 bit timer value to include in transmitted
PAUSE frame

Flow Control Type (FCT)

16-bit field to indicate flow control type

Flow Control Receive Thresh Lo (FCRTL)

13-bit low water mark indicating receive
buffer emptiness

Flow control is implemented as a means of reducing the possibility of receive buffer overflows
which result in the dropping of received packets, and allows for local control of network
congestion levels. This can be accomplished by sending an indication to a transmitting station of a
nearly-full receive buffer condition at a receiving station.
The implementation of asymmetric flow control allows for one link partner to send flow control
packets while being allowed to ignore their reception. For example, not required to respond to
PAUSE frames.
For the 82541xx and 82547GI/EI, there are two forms of flow control that can be established via
auto-negotiation: symmetric and asymmetric. Symmetric flow control is for point-to-point links;
asymmetric for hub-to-end-node connections. Symmetric flow control allows either node to flowcontrol the other. Asymmetric flow control allows a repeater or switch to flow-control a DTE, but
not vice versa

8.7.3

MAC Control Frames & Reception of Flow Control Packets
Three comparisons are used to determine the validity of a flow control frame:
1. A match on the 6-byte multicast address for MAC Control Frames or to the station address of
the device (Receive Address Register 0).
2. A match on the type field.
3. A comparison of the MAC Control Opcode field.
Standard 802.3x defines the MAC Control Frame multicast address as 01_80_C2_00_00_01h. This
address must be loaded into the Flow Control Address Low/High registers (FCAL/H).
The Flow Control Type register (FCT) contains a 16-bit field that is compared against the flow
control packet’s type field to determine if it is a valid flow control packet: XON or XOFF. 802.3x
reserves this value as 8808h. This number must be loaded into the Flow Control Type (FCT)
register.

Software Developer’s Manual

173

Ethernet Interface

The final check for a valid PAUSE frame is the MAC Control Opcode. At this time only the
PAUSE control frame opcode is defined. It has a value of 0001h.
Frame based flow control differentiates XOFF from XON based on the value of the PAUSE timer
field. Non-zero values constitute XOFF frames while a value of zero constitutes an XON frame.
Values in the timer field are in units of slot time. A “slot time” is hard wired to 64 byte times, or
512 ns.

Note:

“S” is the Start-of-Packet delimiter and “T” is the first part of the End-ofPacket delimiters for 802.3z encapsulation.

Figure 8-4. 802.3x MAC Control Frame Format

The receiver is enabled to receive flow control frames if flow control is enabled through the RFCE
bit in the Device Control register (CTRL). Software sets this bit consistently with the advertised
capability in the Transmit Configuration Word Register (TXCW).

174

Software Developer’s Manual

Ethernet Interface

Flow control capability must be negotiated between link partners via the Auto-Negotiation process.
The Auto-Negotiation process can modify the value of these bits based on the resolved capability
between the local device and the link partner.
Once the receiver has validated the reception of an XOFF, or PAUSE frame, the Ethernet controller
performs the following:

•
•
•
•

Increment the appropriate statistics register(s)
Set the TXOFF bit in the Device Status Register (STATUS)
Initialize the pause timer based on the packet’s PAUSE timer field
Disable packet transmission or schedule the disabling of transmission after the current packet
completes.

Resumption of transmission can occur under the following conditions:

• Expiration of the PAUSE timer
• Reception of on XON frame (a frame with its PAUSE timer set to 0b)
Either condition clears the STATUS.TXOFF bit and transmission can resume. Hardware records
the number of received XON frames in the XONRXC counter.

8.7.4

Discard PAUSE Frames and Pass MAC Control Frames
Note:

When receive flow control is enabled (CTRL.RFCE) is 1b, the following special filtering is
performed on PAUSE and MAC Control frames. When receive flow control is disabled, these
frames are filtered like any other frames and the rest of this section can be ignored.
Two bits in the Receive Control register (RCTL) are implemented specifically for control over
receipt of PAUSE and MAC control frames. These bits are Discard PAUSE Frames (DPF) and Pass
MAC Control Frames (PMCF). See Section 13.4.22 for DPF and PMCF bit definitions.
The DPF bit forces the discarding of any valid PAUSE frame addressed to the Ethernet controller’s
station address. If the packet is a valid PAUSE frame and is addressed to the station address
(receive address [0]), the Ethernet controller does not pass the packet to host memory if the DPF bit
is set to logic high. The DPF bit does not affect pause frames that are addressed to the MAC control
frame multicast address (01-80-C2-00-00-01). These frames are DMA’ed if they pass standard
address filtering, including receive address 1 to 15, multicast hash filtering, or the Multicast
Promiscuous bit is enabled. TheDPF has no affect on PAUSE operation, only the DMA function.
The PMCF bit allows for the passing of any valid MAC control frames to the system which do not
have a valid PAUSE opcode. In other words, the frame can have the correct MAC control frame
multicast address (or the MAC station address) as well as the correct type field match with the FCT
register, but does not have the defined PAUSE opcode of 0001h. Frames of this type are transferred
to host memory when PMCF is logic high. The results of this filter are logically ORed into the
standard filters, so even if PMCF is 0b, any MAC control frame that isn't a PAUSE frame that
passes standard address filtering is DMA’ed

Software Developer’s Manual

175

Ethernet Interface

8.7.5

Transmission of PAUSE Frames
Transmitting PAUSE frames is enabled by software writing a 1b to the CTRL.TFCE bit. This bit is
mapped to bit 8 of the TXCW txConfigWord field. (ASM_DIR bit).
Similar to the reception flow control packets described earlier, XOFF packets can be transmitted
only if this configuration has been negotiated between the link partners via the Auto-Negotiation
process. In other words, the setting of this bit indicates the desired configuration. The resolution of
the Auto Negotiation process is discussed in Sections 8.6.3 and 8.6.4.
The contents of the Flow Control Receive Threshold High register (FCRTH) determine at what
point hardware transmits a PAUSE frame. Hardware monitors the fullness of the receive FIFO and
compares it with the contents of FCRTH. When the threshold is reached, hardware sends a PAUSE
frame with its pause time field equal to FCTTV. Once the receive buffer fullness reaches the low
water mark, hardware sends an XON message (a PAUSE frame with a timer value of 0b). Software
enables this capability with the XONE field of the FCRTL.
Hardware sends one more PAUSE frames if it has previously sent one and the FIFO overflows (so
the threshold must not be set greater than the FIFO size). This function is intended to minimize the
number of packets dropped if the first PAUSE frame does not reach its target.
Transmitting Flow Control frames should only be enabled in full duplex mode per the IEEE 802.3
standard. Software should ensure that the transmission of flow control packets is disabled when the
Ethernet controller is operating in half-duplex mode.

8.7.6

Software Initiated PAUSE Frame Transmission
The Ethernet controller has the added capability to transmit an XOFF frame through software. This
function is accomplished by software writing a 1b to the SWXOFF bit of the Transmit Control
register (TCTL). Once this bit is set, hardware initiates the transmission of a PAUSE frame in a
manner similar to that automatically generated by hardware.
The SWXOFF bit is self clearing after the PAUSE frame has been transmitted.
The state of the CTRL.TFCE bit or the negotiated flow control configuration does not affect
software generated PAUSE frame transmission.
Software sends an XON frame by programming a zero in the PAUSE timer field of the FCTTV
register.

Caution:

8.7.7

Use of SWXOFF is not recommended due to security concerns.

External Control of Flow Control Operation1
Transmitting XOFF and XON frames can be triggered by external pins. When enabled through
FCRTH.XFCE, the XOFF and XON inputs can be used to provide external effective threshold
information that initiate XOFF and XON transmission, respectively.

82544GC/EI only.

176

Software Developer’s Manual

Ethernet Interface

When the XOFF signal is asserted high, the device transmits a single XOFF frame. The assertion of
I_XON (after deassertion of XOFF) initiates an XON frame transmission if enabled by
FCRTL.XONE. The assertion/deassertion of XON is required between assertions of XOFF in order
to send another XOFF frame, providing a built-in hysteresis mechanism.
Output signals are also provided from the 82544GC/EI to indicate the device is either above the
programmed flow control high threshold or below the flow control low threshold (ABV_HI and
BLW_LOW respectively).
Flow control transmission must also be enabled through CTRL.TFCE.

Software Developer’s Manual

177

Ethernet Interface

Note:

178

This page intentionally left blank.

Software Developer’s Manual

802.1q VLAN Support

The PCI/PCI-X Family of Gigabit Ethernet Controllers provide several specific mechanisms to
support 802.1q VLANs:

• Optional adding (for transmits) and stripping (for receives) of IEEE 802.1q VLAN tags
• Optional ability to filter packets belonging to certain 802.1q VLANs
Note:

9.1

The 82541ER Ethernet controller does not support VLAN tags.

802.1q VLAN Packet Format
Table 9-1 compares the format of an untagged 802.3 Ethernet packet with an 802.1q VLAN tagged
packet. The CRC for the 802.1q tagged frame is re-computed, so that it covers the entire tagged
frame including the 802.1q tag header.
Table 9-1. VLAN Packet Format Comparison
802.3 Packet

#Octets

802.1q VLAN
Packet

#Octets

Type/Length

8021.q Tag

Data

46-1500

Type/Length

CRC

Data

46-1500

CRC*

Maximum frame size for a standard 803.3ac (802.1q VLAN and/or 802.1p priority) packet is 1522
octets as opposed to 1518 octets for a normal 802.3 Ethernet packet. If jumbo frames are used,
enabling 802.3ac adds 4 bytes to the packet to accommodate the q-tag. If multiple descriptors are
required for a transmit, the q-tag information for the packet is extracted from only the last
descriptor of the packet. VLAN tagging is supported independently of packet size.

9.1.1

802.1q Tagged Frames
For 802.1q, the Tag Header field consists of four octets containing the Tag Protocol Identifier
(TPID) and Tag Control Information (TCI), each utilizing 2 octets. The first 16 bits of the tag
header make up the TPID. It contains the “protocol type” which identifies the packet as a valid
802.1q tagged packet.
The two octets making up the TCI contain three fields (see Table 9-2 for details):

• User Priority (UP)
• Canonical Form Indicator (CFI). The CFI should be 0b for transmits. For receives, the
Ethernet controller has the capability to filter out packets that have this bit set. See the CFIEN
and CFI bits in the RCTL as described in Section 13.4.22.

• VLAN Identifier (VID)

Software Developer’s Manual

179

802.1q VLAN Support

Table 9-2. 802.1q Tagged Frames

Octet 1
UP

9.2

Octet 2
CFI

VID

Transmitting and Receiving 802.1q Packets
Since the 802.1q tag is only four bytes, adding and stripping of tags can done completely in
software. (For transmits, software inserts the tag into packet data before it builds the transmit
descriptor list, and for receives, software strips the four byte tag from the packet data before
delivering the packet to upper layer software.)
However, because adding and stripping of tags in software results in more overhead for the host,
the Ethernet controller has additional capabilities to add and strip tags in hardware, as discussed in
the following two sections.

9.2.1

Adding 802.1q Tags on Transmits
Software can command the Ethernet controller to insert an 802.1q VLAN tag on a per packet basis.
If CTRL.VME is set to 1b, and the VLE bit in the transmit descriptor is set to 1b, then the Ethernet
controller inserts a VLAN tag into the packet that it transmits over the wire. The Tag Protocol
Identifier (TPID) field of the 802.1q tag comes from the VLAN Ether Type (VET) register, and the
Tag Control Information (TCI) of the 802.1q tag comes from the special field of the transmit
descriptor (TDESC.SPECIAL).

9.2.2

Stripping 802.1q Tags on Receives
Software can instruct the Ethernet controller to strip 802.1q VLAN tags from received packets. If
the CTRL.VME bit is set to 1b, and the incoming packet is an 802.1q VLAN packet (its Ethernet
Type field matched the VET register), then the Ethernet controller strips the 4-byte VLAN tag from
the packet, and stores the TCI in the Special field of the receive descriptor.
The Ethernet controller also sets the VP bit in the receive descriptor to indicate that the packet had
a VLAN tag that was stripped. If the CTRL.VME bit is not set, the 802.1q packets can still be
received if they pass the receive filter. In this case, the VLAN tag is not stripped and the VP bit is
not set. Refer to Table 9-3 for more information regarding receive packet filtering.

9.3

802.1q VLAN Packet Filtering
VLAN filtering is enabled by setting the RCTL.VFE bit to 1b. If enabled, hardware compares the
type field of the incoming packet to a 16-bit field in the VLAN EtherType (VET) register. If the
VLAN type field in the incoming packet matches the VET register, the 802.1q VLAN packet is
then compared against the VLAN Filter Table Array (VFTA) for acceptance.
The Virtual LAN ID field indexes a 4096 bit vector. If the indexed bit in the vector is 1b, there is a
Virtual LAN match. Software can set the entire bit vector to 1b’s if the node does not implement
802.1q filtering.

180

Software Developer’s Manual

802.1q VLAN Support

In summary, the 4096 bit vector is composed of 128 32-bit registers. Matching to this bit vector
follows the same algorithm as indicated in Section 13.5.1 for Multicast Address filtering. The
VLAN Identifier (VID) field consists of 12 bits. The upper 7 bits of this field are decoded to
determine the 32-bit register in the VLAN Filter Table Array to address and the lower 5 bits
determine which of the 32 bits in the register to evaluate for matching.
Two other bits in the Receive Control register (see Section 13.4.22), CFIEN and CFI, are also used
in conjunction with 802.1q VLAN filtering operations. CFIEN enables the comparison of the value
of the CFI bit in the 802.1q packet to the Receive Control register CFI bit as an acceptance criteria
for the packet.
Note:

The VFE bit does not affect whether the VLAN tag is stripped. It only affects whether the VLAN
packet passes the receive filter.
Table 9-3 lists reception actions according to control bit settings.
Table 9-3. Packet Reception Decision Table
Is
packet
802.1q?

CTRL.
VME

RCTL.
VFE

Normal packet reception.

Yes

Receive a VLAN packet if it passes the standard filters (only). Leave the
packet as received in the data buffer. Clear the VP bit in the receive
descriptor.

Yes

Receive a VLAN packet if it passes the standard filters and the VLAN
filter table. Leave the packet as received in the data buffer (the VLAN
tag is not stripped). Clear the VP bit in the receive descriptor.

Yes

Receive a VLAN packet if it passes the standard filters (only). Strip off
the VLAN information (four bytes) from the incoming packet and store in
the descriptor. Set the VP bit in the receive descriptor.

Yes

Receive a VLAN packet if it passes the standard filters and the VLAN
filter table. Strip off the VLAN information (four bytes) from the incoming
packet and store in the descriptor. Set the VP bit in the receive
descriptor.

Software Developer’s Manual

ACTION

181

802.1q VLAN Support

Note:

182

This page intentionally left blank.

Software Developer’s Manual

Configurable LED Outputs

Configurable LED Outputs
10.1

Configurable LED Outputs1
The PCI/PCI-X Family of Gigabit Ethernet Controller’s MAC implements four output drivers
intended for driving external LED circuits. Each MAC’s four LED outputs can be individually
configured to select the particular event, state, or activity that is indicated on that output. In
addition, each LED can be individually configured for output polarity as well as for blinking vs.
non-blinking (steady-state) indication.
The configuration for LED outputs is specified via the LEDCTL register. In addition, the
hardware-default configuration for two of the LED outputs, LED0/LINK_UP# and LED2/
LINK100# can be specified via EEPROM fields, thereby supporting LED displays configurable to
a particular OEM preference.

10.1.1

Selecting an LED Output Source
Each of the four LED indications can be independently configured. The LEDCTL register MODE
field corresponding to each LED selects the expression generating the LED output. The LED
outputs are, by default, active low; it is assumed they are connected to the negative side (cathode)
of an external LED. They will, by default, output a low value upon the assertion of the event (such
as COLLISION) or state (such as LINK1000#) selected. Note that the active sense of the LED
outputs can be inverted). See Section 10.1.2 for details.

LINK_UP
ACTIVITY
COLLISION

VCC/LED_OFF

(OPTIONAL)
POLARITY
INVERSION

(OPTIONAL)
BLINK
CONTROL
CIRCUIT

LED
OUTPUT
DRIVER

EXTERNAL
LED

GND/LED_ON

Figure 10-1. Selecting an LED Output Source

Section 10 does not apply to the 82544GC/EI.

Software Developer’s Manual

183

Configurable LED Outputs

LED outputs can be based on the following expressions:

•
•
•
•
•
•
•
•
•
•
•
•
•

LINK_UP is asserted while link of any speed is maintained
LINK_10 indicates link at 10 Mbps
LINK_100 indicates link at 100 Mbps
LINK_1000 indicates link at 1000 Mbps
LINK_100/1000 indicates link at either 100 or 1000 Mbps
LINK_10/1000 indicates link at either 10 or 1000 Mbps
ACTIVITY is asserted when link is established and packets are being transmitted or received
LINK/ACTIVITY is asserted when link is established but there is NO transmit or receive
activity
COLLISION is asserted each time a collision is observed
PAUSED is asserted while the Ethernet controller’s transmitter is paused due to flow control
PCIX_MODE is asserted when the Ethernet controller is in PCI-X mode (versus PCI mode)
FULL_DUPLEX is asserted when the link is configured for full duplex operation
BUS_SPEED is asserted in PCI 66 MHz or PCI-X 133 MHz configurations (high-speed
operation)

• BUS_SIZE is asserted in 64-bit PCI or PCI-X configurations
• LED_ON is always asserted (low); LED_OFF is always deasserted (high)

10.1.2

Polarity Inversion
The LEDCTL.IVRT field enables the selected LED source to be optionally inverted. This can be
used to drive external circuitry where an active high indication of one of the selectable states/
events is required (such as multi-color LED circuits).

Note:

10.1.3

Polarity inversion (LEDCTL.IVRT = 1b) and blinking (LEDCTL.BLINK = 1b) at the same time
for a given LED is not recommended. Introducing additional polarity inversion on a selected state/
event while blink-control is also enabled can produce nonsensical LED behavior (such as blinking
LED’s during periods of NO activity or when link is down).

Blink Control
Each LED’s output circuitry also includes a blink-control circuit that can additionally be enabled.
The blink control circuitry turns its output sequentially on (low) for 200 ms, then off for another
200 ms, each time its input is active/asserted. The LEDCTL.BLINK field controls whether a blink
circuit is enabled for an LED output.
The blink control is especially useful for ensuring that certain brief events, such as momentary
ACTIVITY or COLLISION events, cause LED transitions which are sufficiently visible to a
human eye. The circuit re-evaluates after each on/off blink cycle, ensuring a continuous blink
pattern throughout periods of continuous event/state assertion (such as heavy ACTIVITY periods
or long PAUSED times).

184

Software Developer’s Manual

Configurable LED Outputs

Note:

It is especially important to note with respect to the blink-control circuit that:

• the blink circuit, when enabled, exists as the LAST stage of the LED circuitry, after any
(optional) signal inversion

• the blink sequence occurs when the circuit input is asserted low
As a result, it is possible to select combinations of IVRT and BLINK which do not make sense or
produce unexpected results, such as examples previously noted. It is recommended that BLINK
only be selected for indicating ACTIVITY, COLLISION, PAUSED, or the combination LINK/
ACTIVITY signal. events/states, and that IVRT = 0b when blink is selected.
Note:

Selecting the LEDCTL.MODE = LINK/ACTIVITY with BLINK = 1b selects a unique LED
output expression (this configuration is meaningful ONLY when IVRT inversion is disabled). In
this configuration, the LED is off (output high) if there is no LINK, on if there is LINK but no
ACTIVITY, and blinking if there is LINK with ACTIVITY.
LINK_UP

(LINK & NO ACTIVITY)
TOGGLING DURING ACTIVITY

ACTIVITY

BLINK
CONTROL
CIRCUIT

LED
OUTPUT
DRIVER

EXTERNAL
LED

Figure 10-2. Blink Control

Software Developer’s Manual

185

Configurable LED Outputs

Note:

186

This page is intentionally left blank.

Software Developer’s Manual

PHY Functionality and Features

PHY Functionality and Features
11.1

Auto-Negotiation
Auto-Negotiation between the PCI/PCI-X Family of Gigabit Ethernet Controllers and its link
partner is performed by the PHY. Under normal, expected operating conditions, the MAC
automatically establishes common speed and duplex settings via the PHY. This section details PHY
configuration features involved in the auto-negotiation process.

11.1.1

Overview
Auto-Negotiation by the PHY is initiated upon any of the following conditions:

•
•
•
•

Power-up reset (copper and fiber)

•
•
•
•

PHY hardware reset asserted using the MAC CTRL.PHY_RST bit (copper only)

PHY detects loss of link (copper and fiber)
PHY detects re-appearance of energy on the link (copper and fiber)
MAC control of PHY power-management is enabled (CTRL.EN_PHY_PWR_MGMT = 1b
and MAC transitions to low power state (D3) where continued PHY operation required for
wakeup/manageability (copper and fiber)

PHY soft-reset initiated via the PHY Control Register (bit 15, copper only)
Explicit Auto-Negotiation Re-Start initiated via the PHY Control Register (bit 9, copper only)
Explicit transition of PHY from internal IEEE power-down to normal mode via the PHY
Control Register (bit 11, copper only)

• Explicit transition of PHY from internal IEEE power-down to normal mode via the PHY
Control Register by setting CTRL.LRST = 1 and TXCW.ANE = 1 (fiber only)
Hardware defaults for the PHY configurations enable the PHY to advertise its full 1000BASE-T
and 1000BASE-X capability, and to auto-negotiate to the best possible operation1 without any
software intervention required. If the remote device does not have Auto-Negotiation capability, the
Ethernet controller PHY uses the parallel detect function to determine the speed of the remote
device for 100BASE-TX and 10BASE-T modes. Under certain circumstances, it might be
desirable to configure auto-negotiation options to restrict certain behavior. For example, operate in
half-duplex mode only.2
Note:

1.
2.

Any PHY auto-negotiation options configured by software are only persistent while the LAN
power (indicated by LAN_PWR_GOOD) remains available. Following a complete loss of power,
the PHY reverts to auto-negotiation using its hardware-defaults.

1000 half-duplex not supported.
TXCW and RXCW registers are used for fiber auto-negotiation advertising. For fiber, the MAC can be forced to 1000 full-duplex when
connected to a non-auto-negotiating fiber switch.

Software Developer’s Manual

187

PHY Functionality and Features

11.1.2

Next Page Exchanges
If 1000BASE-T mode is advertised, then the Ethernet controller PHY automatically sends the
appropriate next pages to advertise the capability and negotiate master/slave mode of operation. If
a developer does not want to transmit additional next pages, the next page bit (PCI-Config Register
bit 15) can be set to 0b and the software need take no further action.
If next pages in addition to the ones required for 1000BASE-T are needed, then the software can
set Auto-Negotiation Expansion Register bit 15 to 1b, and send and receive additional next pages
via the Next Page Transmit Register (NPT) and Link Partner Next Page Register (LPN),
respectively. The PHY stores the previous results from the Link Partner Next Page Register (LPN)
in internal registers so that new next pages can overwrite the Link Partner Next Page Register
(LPN).

Note:

11.1.3

1000BASE-T next page exchanges are automatically handled without any software intervention,
regardless of whether or not additional next pages are sent.

Register Update
Changes to PHY Control Register bits 6, 8, 12, and 13, and PHY Specific Control Register bits 3,
4, 6:5, 9:8 and 11, do not take effect unless one of the following takes place (copper only):

•
•
•
•

PHY soft reset (PHY Control Register bit 15)
Restart Auto-Negotiation (PHY Control Register bit 9)
Transition of PHY from IEEE power-down to normal operation (PHY Control Register bit 11)
The link goes down

To enable or disable Auto-Negotiation, PHY Control Register bit 12 should be changed
simultaneously with either PHY Control Register bits 15 or 9. For example, to disable AutoNegotiation and force 10BASE-T half-duplex mode, the PHY Control Register should be written
with 8000h.
To disable Auto-Negotiation (fiber only), set TXCW.ANE = 0.
The Auto-Negotiation Expansion Register and the 1000BASE-T Control Register are internally
latched once every time the Auto-Negotiation enters the Ability Detect state in the arbitration state
machine. As a result, a write to the Auto-Negotiation Expansion Register or the 1000BASE-T
Control Register has no effect once the PHY begins to transmit Fast Link Pulses (FLPs). This
guarantees that sequences of FLPs transmitted are consistent with one another.
The Next Page Transmit Register is treated similarly to the Auto-Negotiation Expansion Register
and the 1000BASE-T Control Register during additional next page exchanges.

188

Software Developer’s Manual

PHY Functionality and Features

11.1.4

Status
Once the PHY completes auto-negotiation, it updates the various statuses in the PHY Status
Register, Link Partner Ability Register (Base Page), Auto-Negotiation Expansion Register, and
1000BASE-T Status Register. For 1000BASE-T operation, the Auto-Negotiation Expansion
Register and the Link Partner Ability Register (Base Page) are updated. Speed, duplex, page
received, and Auto-Negotiation completion statuses are also available in the PHY Specific Status
Register (PSTATUS) and the PHY Interrupt Status Register (PINTS).
For fiber, the CTRL.STATUS register will reflect link status.
Assuming normal MAC configuration, the MAC status register STATUS reports bits SPEED, FD
(duplex/half indication), and LU (link up status) shortly after the PHY (or MAC, for fiber)
completes auto-negotiation.

11.2

MDI/MDI-X Crossover (copper only)
The Ethernet controller PHY automatically determines whether or not it needs to cross over
between pairs as shown in the following table so that an external crossover cable is not required. If
the PHY interoperates with a device that cannot automatically correct for crossover, the Ethernet
controller PHY makes the necessary adjustment prior to commencing Auto-Negotiation. If the
PHY operates with a device that implements MDI/MDI-X crossover, a random algorithm as
described in IEEE 802.3 clause 40.4.4 determines which device performs the crossover.
When the Ethernet controller PHY interoperates with legacy 10BASE-T devices that do not
implement Auto-Negotiation, the PHY follows the same algorithm as described above since link
pulses are present. However, when interoperating with legacy 100BASE-TX devices that do not
implement Auto-Negotiation (link pulses are not present), the Ethernet controller PHY uses signaldetection to determine whether to crossover.
Auto MDI/MDI-X crossover is the default hardware configuration, but can be disabled via the
PHY Specific Control Register bits 6:5 (PSCON).
The pin mapping in MDI/MDI-X modes are as follows:
Pin

MDI

MDIX

1000BASE-T

100BASE-TX

10BASE-T

1000BASE-T

100BASE-TX

10BASE-T

MDI[0]+/-

BI_DA +/-

TX +/-

BI_DB +/-

RX +/-

MDI[1]+/-

BI_DB +/-

RX +/-

BI_DA +/-

TX +/-

MDI[2]+/-

BI_DC +/-

unused

BI_DD +/-

unused

MDI[3]+/-

BI_DD +/-

unused

BI_DC +/-

unused

Software Developer’s Manual

189

PHY Functionality and Features

11.2.1

Polarity Correction (copper only)
The Ethernet controller PHY automatically corrects for polarity errors on the receive pairs in
1000BASE-T and 10BASE-T modes. In 100BASE-TX mode, the polarity does not matter.
In 1000BASE-T mode, receive polarity errors are automatically corrected based on the sequence of
the symbols. Once the descrambler is locked, the polarity is also locked on all pairs. The polarity
becomes unlocked only when the receiver loses lock.

11.2.2

10/100 Downshift (82540EP/EM Only)
Gigabit speed operation requires a 4-pair cable to operate. Some existing cables have only two
pairs. Other cables might have 4 pairs, but one might be broken, leaving three working pairs. Over
two or three pairs, two gigabit link partners might be able to successfully auto-negotiate 1000
Mbps speed, but then be unable to achieve link.
The downshift feature enables the 82540EP/EM PHY to auto-negotiate with another gigabit link
partner using a two or three pair cable and downshift to link at 100 Mbps or 10 Mbps, whichever is
the highest speed below gigabit that the link partner is capable of.
By default, downshift is turned on. Refer to Table 13-39 to disable the downshift feature.

190

Software Developer’s Manual

PHY Functionality and Features

11.3

Cable Length Detection (copper only)
In 100/1000 Mbps operation, the Ethernet controller PHY attempts to indicate the approximate
length of the CAT 5 cable attached. The estimated cable length is reported as one of the following
ranges:

•
•
•
•
•

<= 50 m
50 – 80 m
80 – 110 m
110 – 140 m
>= 140 m

The estimated cable length can be obtained by reading the PHY Specific Status Register bits 9:7.

11.4

PHY Power Management (copper only)
The Ethernet controller PHY supports power-management based on either link status, MAC
power-state, or both. During link-down states, the PHY utilizes its energy-detection capabilities to
consume the least amount of power while still being capable of resuming link-up automatically. the
Ethernet controller can be configured to automatically reduce PHY power during certain MAC D3
states by re-negotiating a low-speed link (the default behavior, but this can be disabled).

11.4.1

Link Down – Energy Detect (copper only)
When the link is operational and the Ethernet controller PHY detects loss of link, it initiates an
Auto-Negotiation session. Failing to re-establish link via auto-negotiation, the PHY reverts to an
Energy Detect power-down state. Following link loss, the PHY monitors receive energy on the
wire. If the PHY detects energy on the wire, it starts to initiate Auto-Negotiation sending FLPs for
five seconds. If at the end of five seconds the Auto-Negotiation has not completed, then the PHY
stops sending FLPs and returns to monitoring receive energy. While monitoring receive energy, the
PHY sends out a single 10 Mbps NLP (Normal Link Pulse) every one second in an attempt to wake
up a connected device.
The above behavior is considered to be an advanced Energy-Detect (Energy Detect+) mode of
PHY operation. The PHY can also be configured for a regular Energy Detect mode, which behaves
similarly, except does not send out NLPs while monitoring receive energy. In this configuration,
the Ethernet controller PHY can be woken by a connected device, but does not wake up the
connected device itself.
The PHY Specific Control Register bits 9 and 8 described in Section 13.4.7.1.13 are used to
configure the specific Energy-Detect mode of behavior for the PHY.

Software Developer’s Manual

191

PHY Functionality and Features

11.4.2

D3 State, No Link Required (copper only)
Each time the MAC transitions to a D3 or D0u power-state with no link required (wakeup disabled
and no manageability enabled), the PHY enters its IEEE power-down mode, consuming the least
amount of power possible. When powered-down, the PHY does not perform any form of Energy
Detection, and does not generate any energy (NLPs) on the wire itself.
MAC transitions back to D0 power-states, either through explicit system/software mechanisms or
by hardware reset operations, return the PHY back to a functional power-state. This will also reinitiate an auto-negotiation attempt advertising all speeds possible (10/100/1000 Mbps)1, reverting
to an Energy Detect state if unsuccessful in establishing link-up.

11.4.3

D3 Link-Up, Speed-Management Enabled (copper only)
If the MAC is configured for PHY power management (CTRL.EN_PHY_PWR_MGMT = 1b) and
the PHY is linked at 1000 Mbps, then upon MAC transitions to D3 or D0u power-states where link
IS required (either wakeup or manageability are enabled), the PHY re-initiates an AutoNegotiation operation, advertising only 10/100 Mbps capability. This results in D3 operation at a
lower-speed link and a reduced power level.
If a wakeup, management operation, or other system event causes the MAC to revert to fullyoperational D0 state, the PHY initiates another Auto-Negotiation operation, advertising all 10/100/
1000 Mbps speed capability, in order to return to maximum-speed operation.

11.4.4

D3 Link-Up, Speed-Management Disabled (copper only)
If the MAC is configured for no PHY power management (CTRL.EN_PHY_PWR_MGMT = 0b),
and the MAC transitions to D3 power-states where link is required for either wakeup or
manageability, then the PHY simply remains operational at its current line speed, without initiating
a new Auto-Negotiation operation. This configuration is not recommended, since D3 power
consumption at 1000 Mbps exceeds 20 mA Vaux.

Half duplex not supported.

192

Software Developer’s Manual

PHY Functionality and Features

11.5

Initialization
Note:

Section 11.5 through Section 11.14 apply only to the 82541xx and 82547GI/EI Ethernet
controllers.
At power-up or reset, the PHY core performs the initialization as shown in Figure 11-1. The
software driver has access to the PHY register 0d, bits 15 and 11 for PHY reset and PHY Power
Down control, respectively.

Power-up
or
Reset

Read H/W
Control
Interface

Initialize
MDIO
Registers

Pass control
to
MDIO Interface

Figure 11-1. PHY Initialization Sequence

11.5.1

MDIO Control Mode
In the MDIO Control mode, the PHY uses the Hardware Control Interface to set up initial (default)
values of the MDIO registers. Once initial values are set, bit control reverts to the MDIO interface.
The PHY can perform some low level initializations such as DSP configuration based upon
EEPROM settings. The details of those initializations are reserved.

Software Developer’s Manual

193

PHY Functionality and Features

11.6

Determining Link State
The PHY and its link partner determine the type of link established through one of three methods:

• Auto-Negotiation
• Parallel Detection
• Forced Operation
Auto-Negotiation is the only method allowed by the 802.3ab standard for establishing a
1000BASE-T link, although forced operation could be used for test purposes. For 10/100 links, any
of the three methods can be used. The sections that follow discuss each in greater detail.
Figure 11-2 provides an overview of link establishment. First the PHY checks if Auto-Negotiation
is enabled. By default, the PHY supports Auto-Negotiation (PHY register 0, bit 12). If not, the
PHY forces operation as directed. If Auto-Negotiation is enabled, the PHY begins transmitting
Fast Link Pulses (FLPs) and receiving FLPs from its link partner. If FLPs are received by the PHY,
Auto-Negotiation proceeds. It also can receive 100BASE-TX MLT3 and 10BASE-T Normal Link
Pulses (NLPs). If either MLT3 or NLPs are received, it aborts FLP transmission and immediately
brings up the corresponding half-duplex link.

Power-Up, Reset,
Link Failure

Start

A/N
Enabled
?

0 = 10M

Speed
0.13
?

Yes

Send FLP

1 = 100M
No

Send NLP

Send IDLES

Detect
10Mbps
?

Detect
FLP

Auto
Negotiate

Parallel
Detect

Detect
IDLES

Yes

Link
Down

Link
Up

Take
Link
Down

Take
Link
Up

Yes

100M
Half-Duplex
Link

Yes

10M
Half-Duplex
Link

Detect
NLP

Figure 11-2. Overview of Link Establishment

194

Software Developer’s Manual

PHY Functionality and Features

11.6.1

False Link
When the PHY is first powered on, reset, or encounters a link down state, it must determine the line
speed and operating conditions to use for the network link.
The PHY first checks the MDIO registers (initialized via the Hardware Control Interface or written
by software) for operating instructions. Using these mechanisms, programmers can command the
PHY to do one of the following:

• Force twisted-pair link operation to:
— 1000T Full Duplex
— 1000T Half Duplex
— 100TX, Full Duplex
— 100TX, Half Duplex
— 10BASE-T, Full Duplex
— 10BASET, Half Duplex

• Allow Auto-Negotiation/parallel-detection.
In the first six cases (forced operation), the PHY immediately begins operating the network
interface as commanded. In the last case, the PHY begins the Auto-Negotiation/parallel-detection
process.

11.6.2

Forced Operation
Forced operation can be used to establish 10 and 100 links, and 1000 links for test purposes. In this
method, Auto-Negotiation is disabled completely and the link state of the PHY is determined by
PHY register 0d.

Note:

When speed is forced, the MDI/MDI-X crossover feature is not functional.
In forced operation, the programmer sets the link speed (10, 100, or 1000) and duplex state (full or
half). For Gigabit (1000) links, the programmer must explicitly designate one side as the Master
and the other as the Slave. Table 10-1 summarizes link establishment procedures.
Table 10-1. Determining Duplex State Via Parallel Detection
Configuration

Result

Both sides set for Auto-Negotiate.

Link is established via Auto-Negotiation.

Both sides set for forced operation.

No problem as long as duplex settings match.

One side set for Auto-Negotiation and the other for
forced, half-duplex.

Link is established via parallel detect.

One side set for Auto-Negotiation and the other for
forced full-duplex.

Link is established; however, sides disagree, resulting
in transmission problems. Forced side is full-duplex,
Auto-Negotiation side is half-duplex.

Software Developer’s Manual

195

PHY Functionality and Features

11.6.3

Auto Negotiation
The PHY supports the IEEE 802.3u Auto-Negotiation scheme with next page capability. Next Page
exchange uses PHY register 7d to send information and PHY register 8d to receive them. Next
Page exchange can only occur if both ends of the link advertise their ability to exchange Next
Pages.

11.6.4

Parallel Detection
Parallel detection can only be used to establish 10 and 100 links. It occurs when the PHY tries to
negotiate (transmit FLPs to its link partner), but instead of sensing FLPs from the link partner, it
senses 100BASE-TX MLT3 code or 10BASE-T Normal Link Pulses (NLPs) instead. In this case,
the PHY immediately stops Auto-Negotiation (terminates transmission of FLPs) and immediately
brings up whatever link corresponds to what it has sensed (MLT3 or NLPs). If the PHY senses both
of the technologies together, a parallel detection fault is detected and the PHY continues sending
FLPs
With parallel detection, it is impossible to determine the true duplex state of the link partner, and
the IEEE standard requires the PHY to assume a half-duplex link. Parallel detection also does not
allow exchange of flow-control ability (PAUSE and ASM_DIR) or Master/Slave relationship
required by 1000BASE-T. For this reason, parallel detection cannot be used to establish Gigabit
Ethernet links.

11.7

Link Criteria
Once the link state is determined—via Auto-Negotiation, parallel detection or forced operation—
the PHY and its link partner bring up the link.

11.7.1

1000BASE-T
For 1000BASE-T links, the PHY and its link partner enter a training phase. They exchange idle
symbols and use the information gained to set their adaptive filter coefficients.
Either side indicates completion of the training phase to its link partner by changing the encoding
of the idle symbols it transmits. When both sides so indicate, the link is up. Each side continues
sending idle symbols whenever it has no data to transmit. The link is maintained as long as valid
idle, data, or carrier extension symbols are received.

11.7.2

100BASE-TX
For 100BASE-TX links, the PHY and its link partner immediately begin transmitting idle symbols.
Each side continues sending idle symbols whenever it has no data to transmit. The link is
maintained as long as valid idle symbols or data is received.
In 100Mbps mode, the PHY establishes a link whenever the scrambler becomes locked and
remains locked. Link will remain up unless the descrambler receives idles at less than a specified
rate.

196

Software Developer’s Manual

PHY Functionality and Features

11.7.3

10BASE-T
For 10BASE-T links, the PHY and its link partner begin exchanging Normal Link Pulses (NLPs).
The PHY transmits an NLP every 16 ms, and expects to receive one every 10 to 20 ms. The link is
maintained as long as normal link pulses are received.

11.8

Link Enhancements
The PHY offers two enhanced link functions, each of which are discussed in the sections that
follow:

• SmartSpeed
• Flow Control

11.8.1

SmartSpeed
SmartSpeed is an enhancement to Auto-Negotiation that enables the PHY to react intelligently to
network conditions that prohibit establishment of a 1000BASE-T link, such as cable problems.
Such problems might enable Auto-Negotiation to complete, but then inhibit completion of the
training phase. Normally, if a 1000BASE-T link fails, the PHY returns to the Auto-Negotiation
state with the same speed settings indefinitely. With SmartSpeed enabled, after five failed attempts,
the PHY automatically downgrades the highest ability it advertises to the next lower speed: from
1000 to 100 to 10. Once a link is established, and if it is later broken, the PHY automatically
upgrades the capabilities advertised to the original setting.

11.8.1.1

Using SmartSpeed
SmartSpeed is enabled by setting PHY register 16d, bit 7 to 1b. When SmartSpeed downgrades the
PHY advertised capabilities, it sets bit 5 of PHY register 19. When link is established, its speed is
indicated in PHY register 17, bits 15:14. SmartSpeed automatically resets the highest-level AutoNegotiation abilities advertised, if link is established and then lost for more than two seconds.

11.8.2

Flow Control
Flow control enables congested nodes to pause traffic. MACs indicate their ability to implement
flow control during Auto-Negotiation.
The PHY transparently supports MAC-to-MAC advertisement of flow control through its AutoNegotiation process. Prior to Auto-Negotiation, the MAC indicates its flow control capabilities via
PHY register 4d, bit 10 (Pause) and PHY register 4d, bit 11 (ASM_DIR). After Auto-Negotiation,
the link partner’s flow control capabilities are indicated in PHY register 5d, bits 11:10.

Software Developer’s Manual

197

PHY Functionality and Features

Table 10-2 lists the intended operation for the various settings of ASM_DIR and Pause. This
information is provided for reference only; it is the responsibility of the MAC to implement the
correct function. The PHY merely enables the two MACs to communicate their abilities to each
other.
Table 10-2. Pause And Asymmetric Pause Settings
ASM_DIR Settings
Local
(PHY Register 4d, Bit
10) and Remote (PHY
Register 5d, Bit 10)

Pause Setting Local (PHY
Register 4d, Bit
9)

Pause Setting Remote (PHY
Register 5d, Bit
9)

Result

Symmetric - Either side can flow control
the other

Asymmetric - Remote can flow control
local only

Asymmetric - Local can flow control
remote

No flow control

Symmetric - Either side can flow control
the other

Both ASM_DIR = 1b

Either or both
ASM_DIR = 0b
Either or both = 0b

11.9

No flow control

Management Data Interface
The PHY supports the IEEE 802.3 MII Management Interface also known as the Management
Data Input/Output (MDIO) Interface. The MDIO interface consists of a physical connection to the
MAC, a specific protocol which runs across the connection, and a 16-bit MDIO register set.
PHY Registers 0d through 10d and 15d are required and their functions are specified by the IEEE
802.3 specification. Additional registers are included for expanded functionality.

11.10

Low Power Operation
The Ethernet controller can be get into a low-power state according to MAC control (Power
Management controls) or via PHY register 0d. In either power down mode, the Ethernet controller
is not capable of receiving or transmitting packets.

198

Software Developer’s Manual

PHY Functionality and Features

11.10.1

Powerdown via the PHY Register
The PHY can be powered down using the control bit found in PHY register 0d, bit 11. This bit
powers down a significant portion of the port but clocks to the register section remain active. This
enables the PHY management interface to remain active during power-down. The power-down bit
is active high. When the PHY exits software power-down (PHY register 0d, bit 11 = 1b), it reinitializes all analog functions, but retains its previous configuration settings.

11.10.2

Smart Power-Down
Smart Power-Down (SPD) is a link-disconnect capability applicable to all power management
states, and is intended for mobile applications. Smart powerdown combines a power saving
mechanism with the fact that link might disappear and resume.
SPD is enabled by PHY register 20d, bit 5 or by the SPD Enable bit in the EEPROM, and is
entered when the PHY detects link lost. Auto-Negotiation must also be enabled. While in the SPD
state, the PHY powers down circuits and clocks that are not required for detection of link activity.
The PHY is still able to detect link pulses (including parallel detect) and wake up to engage in link
negotiation. The PHY does not send link pulses (NLP) while in the SPD state. Register accesses
are still possible.
Connecting a member of the family to another system with the SPD feature can lead to link failures
if both ports are allowed to enter the SPD state.

11.11

1000 Mbps Operation

11.11.1

Introduction
This section provides an overview of 1000BASE-T functions, followed by discussion and review
of the internal functional blocks shown in Figure 11-3.

Software Developer’s Manual

199

PHY Functionality and Features

MAC Interface

8
8
Side-stream
Scrambler /
Descrambler

Trellis Viterbi
Encoder/ Decoder
4
DSP

ECHO, NEXT,
FEXT Cancellers

4DPAM5
Encoder

AGC, A/D,
Timing Recovery

Pulse Shaper,
DAC, Filter

Hybrid

Line Driver

Line
Interface

Figure 11-3. 1000 Base-T PHY Functions Overview

200

Software Developer’s Manual

PHY Functionality and Features

11.11.2

Transmit Functions
This section describes functions used when the Media Access Controller (MAC) transmits data
through the PHY and out onto the twisted-pair connection.

11.11.2.1

Scrambler
The scrambler randomizes the transmitted data. The purpose of scrambling is two fold:
1. Scrambling eliminates repeating data patterns from the 4DPAM5 waveform to reduce EMI.
2. Each channel (A, B, C, D) gets a unique signature that the receiver uses for identification.
The scrambler is driven by a Linear Feedback Shift Register (LFSR), which is randomly loaded at
power-up. The LFSR function used by the Master differs from that used by the Slave, giving each
direction its own unique signature. The LFSR, in turn, generates uncorrelated outputs. These
outputs randomize the inputs to the 4DPAM5 and Trellis encoders and randomize the sign of the
4DPAM5 outputs.

11.11.3

Transmit FIFO
The transmit FIFO re-synchronizes data transmitted by the MAC to the transmit reference used by
the PHY.

11.11.3.1

Transmit Phase-Locked Loop PLL
This function generates the 125 MHz timing reference used by the PHY to transmit 4DPAM5
symbols. When the PHY is the Master side of the link, the crystal input is the reference for the
transmit PLL. When the PHY is the Slave side of the link, the recovered receive clock is the
reference for the transmit PLL.

11.11.3.2

Trellis Encoder
The Trellis Encoder uses the two high-order bits of data and its previous output to generate a ninth
bit, which determines if the next 4DPAM5 pattern should be even or odd. This function provides
forward error correction and enhances the signal-to-noise (SNR) ratio by a factor of 6 dB.

11.11.3.3

4DPAM5 Encoder
The 4DPAM5 encoder translates 8B codes transmitted by the MAC into 4DPAM5 symbols. The
encoder operates at 125 Mhz, which is both the frequency of the MAC interface and the baud rate
used by 1000BASE-T.
Each 8B code represents one of 256 data patterns. Each 4DPAM5 symbol consists of one of five
signal levels (-2,-1,0,1,2) on each of the four twisted pair (A,B,C,D) representing 54 or 625
possible patterns per baud period. Of these, 113 patterns are reserved for control codes, leaving 512
patterns for data. These data patterns are divided into two groups of 256 even and 256 odd data
patterns. As a result, each 8B octet has two possible 4DPAM5 representations—one even and one
odd pattern.

Software Developer’s Manual

201

PHY Functionality and Features

11.11.3.4

Spectral Shaper
This function causes the 4DPAM5 waveform to have a spectral signature that is very close to that
of the MLT3 waveform used by 100BASE-TX. This enables 1000BASE-T to take advantage of
infrastructure (cables, magnetics) designed for 100BASE-TX.
The shaper works by transmitting 75% of a 4DPAM5 code in the current baud period, and adding
the remaining 25% into the next baud period.

11.11.3.5

Low-Pass Filter
To aid with EMI, this filter attenuates signal components more than 180 Mhz. In 1000BASE-T, the
fundamental symbol rate is 125 Mhz.

11.11.3.6

Line Driver
The line driver drives the 4DPAM5 waveforms onto the four twisted-pair channels (A, B, C, D),
adding them onto the waveforms that are simultaneously being received from the link partner.

11.11.3.7

Transmit/Receive Flow

D0
D1
D2
D3

8
Scrambler

Trellis
Encoder

4D
PAM-5

PAM-5
Encoded Output
to 4-Pair UPT Line

D5
D6
D7

Scrambler Polynomials:
1 + x13 + x33 (Master PHY Mode)
1 + x20 + x33 (Slave PHY Mode)
Figure 11-4. 1000BASE-T Transmit Flow And Line Coding Scheme

202

Software Developer’s Manual

PHY Functionality and Features

D0
D1
D2
D3
GMII

Polynomial
8 bits

Descrambler

8 bits

D4
D5

DSP
Receiver

PAM-5
Encoded Input
from 4-Pair UTP Line

D6
D7

Figure 11-5. 1000BASE-T Receive Flow

11.11.4

Receive Functions
This section describes function blocks that are used when the PHY receives data from the twisted
pair interface and passes it back to the MAC.

11.11.4.1

Hybrid
The hybrid subtracts the transmitted signal from the input signal, allowing the use of simple
100BASE-TX compatible magnetics.

11.11.4.2

Automatic Gain Control
The Automatic Gain Control (AGC) normalizes the amplitude of the received signal, adjusting for
the attenuation produced by the cable.

11.11.4.3

Timing Recovery
This function re-generates a receive clock from the incoming data stream which is used to sample
the data. On the Slave side of the link, this clock is also used to drive the transmitter.

11.11.4.4

Analog-to-Digital Converter
The Analog-to-Digital (ADC) function converts the incoming data stream from an analog
waveform to digitized samples for processing by the DSP core.

11.11.4.5

Digital Signal Processor
The Digital Signal Processor (DSP) provides per-channel adaptive filtering, which eliminates
various signal impairments including:

• Inter-symbol interference (equalization).
• Echo caused by impedance mismatch of the cable.
• Near-end crosstalk (NEXT) between adjacent channels (A, B, C, D).

Software Developer’s Manual

203

PHY Functionality and Features

• Far-end crosstalk (FEXT)
• Propagation delay variations between channels of up to 120 ns.
• Extraneous tones that have been coupled into the receive path.
The adaptive filter coefficients are initially set during the training phase. They are continuously
adjusted (adaptive equalization) during operation through the decision-feedback loop.

11.11.4.6

Descrambler
The descrambler identifies each channel by its characteristic signature, removing the signature and
re-routing the channel internally. In this way, the receiver can correct for channel swaps and
polarity reversals. The descrambler uses the same base LFSR used by the transmitter on the other
side of the link.
The descrambler requires approximately 15 s. to lock, normally accomplished during the training
phase.

11.11.4.7

Viterbi Decoder/Decision Feedback Equalizer (DFE)
The Viterbi decoder generates clean 4DPAM5 symbols from the output of the DSP. The decoder
includes a Trellis encoder identical to the one used by the transmitter. The Viterbi decoder simultaneously looks at the received data over several baud periods. For each baud period, it predicts
whether the symbol received should be even or odd, and compares that to the actual symbol
received. The 4DPAM5 code is organized in such a way that a single level error on any channel
changes an even code to an odd one and vice versa. In this way, the Viterbi decoder can detect
single-level coding errors, effectively improving the Signal-To-Noise (SNR). When an error
occurs, this information is quickly fed back into the equalizer to prevent future errors.

11.11.4.8

4DPAM5 Decoder
The 4DPAM5 decoder generates 8B data from the output of the Viterbi decoder.

11.12

100 Mbps Operation
The MAC passes data to the PHY over the MII. The PHY encodes and scrambles the data, then
transmits it using MLT-3 for 100TX over copper. The PHY descrambles and decodes MLT-3 data
received from the network. When the MAC is not actively transmitting data, the PHY sends out
idle symbols on the line.

11.13

10 Mbps Operation
The PHY operates as a standard 10 Mbps transceiver. Data transmitted by the MAC as 4-bit nibbles
is serialized, Manchester-encoded, and transmitted on the MDI[0]+/- outputs. Received data is
decoded, de-serialized into 4-bit nibbles and passed to the MAC across the internal MII. The PHY
supports all the standard 10 Mbps functions.

204

Software Developer’s Manual

PHY Functionality and Features

11.13.1

Link Test
In 10 Mbps mode, the PHY always transmits link pulses. If the Link Test Function is enabled, it
monitors the connection for link pulses. Once it detects 2 to 7 link pulses, data transmission is
enabled and remains enabled as long as the link pulses or data reception continues. If the link
pulses stop, the data transmission is disabled.
If the Link Test function is disabled, the PHY might transmit packets regardless of detected link
pulses. Setting PHY register 16d, bit 14 can disable the Link Test function.

11.13.2

10Base-T Link Failure Criteria and Override
Link failure occurs if Link Test is enabled and link pulses stop being received. If this condition
occurs, the PHY returns to the Auto-Negotiation phase if Auto-Negotiation is enabled. Setting
PHY register 16d, bit 14 disables the Link Integrity Test function, then the PHY transmits packets,
regardless of link status.

11.13.3

Jabber
If the MAC begins a transmission that exceeds the jabber timer, the PHY disables the Transmit and
loopback functions and asserts collision indication to the MAC. The PHY automatically exits
jabber mode after 250-750 ms. This function can be disabled by setting PHY register 16d, bit 10 to
1b.

11.13.4

Polarity Correction
The PHY automatically detects and corrects for the condition where the receive signal
(MDI_PLUS[0]/MDI_MINUS[0]) is inverted. Reversed polarity is detected if eight inverted link
pulses, or four inverted end-of-frame markers, are received consecutively. If link pulses or data are
not received for 96-130 ms, the polarity state is reset to a non-inverted state.

11.13.5

Dribble Bits
The PHY device handles dribble bits for all of its modes. If between one to four dribble bits are
received, the nibble is passed across the interface. The data passed across is padded with 1b’s if
necessary. If between five to seven dribble bits are received, the second nibble is not sent onto the
internal MII bus to the MAC. This ensures that dribble bits between 1-7 do not cause the MAC to
discard the frame due to a CRC error.

11.14

PHY Line Length Indication
The PHY has a mechanism to deliver coefficient data for use in measuring cable length. If this
capability is required, please contact your Intel representative for details.

Software Developer’s Manual

205

PHY Functionality and Features

Note:

206

This page is intentionally left blank.

Software Developer’s Manual

Dual Port Characteristics

Dual Port Characteristics
12.1

Introduction1
The 82546GB/EB architecture includes two instances of both the MAC and PHY (see Figure 2-1).
With both MAC/PHY pairs operating, the Ethernet controller appears as a multi-function PCI
device containing two identically-functioning devices. To avoid confusion, each MAC (when
combined with either an internal PHY or an internal TBI transceiver/SerDes) is referred to as
“LANx”, where x = “A” or x = “B” to refer to each logical LAN device (LAN A or LAN B).
This section details specific features common to each MAC or PHY, resources/interfaces for which
dedicated independent hardware/software interfaces exists for each LAN, as well as resources
which are shared by both LAN devices.
The Ethernet controller normally appears to the system as a single, multi-function PCI device. It
provides the ability to selectively disable one of the internal LAN functions, thereby allowing it to
appear to the system as a single-function, single-LAN device. The mechanisms for controlling this
behavior and the resulting appearance to the system are described in Section 12.5 entitled, “LAN
Disable”.

12.2

Features of Each MAC
The Ethernet controller is designed to have the capability to appear as two independent instances of
a gigabit controller. The following section details major features that can be considered to be
distinct features available to each Ethernet controller MAC independently.

12.2.1

PCI/PCI-X interface
The Ethernet controller contains a single physical PCI/PCI-X interface. The Ethernet controller is
designed so that each of the logical LAN devices (LAN A and LAN B) appear as a distinct PCI/
PCI-X bus device implementing, along with other registers, the following PCI device header space:
Byte Offset

Byte 0

Byte 1

Byte 3

Device ID

Vendor ID

Status Register

Command Register

8h
Ch

Byte 2

Class Code 020000h
BIST 00h

Header Type 00h

Revision ID 00h
Latency Timer

10h

Base Address 0

14h

Base Address 1

1h8

Base Address 2

1Ch

Base Address 3

20h

Base Address 4

Cache Line Size

Section 12 only applies to the 82546GB/EB.

Software Developer’s Manual

207

Dual Port Characteristics

Byte Offset

Byte 0

Byte 1

Byte 2

24h

Base Address 5

28h

Cardbus CIS Pointer (not used)

2Ch

Subsystem ID

30h

Subsystem Vendor ID
Expansion ROM Base Address

34h

Reserved

38h
3Ch

Byte 3

Cap_Ptr
Reserved

Max_Latency 00h

Min_Grant
FFh

Interrupt Pin
01h or 00h)

Interrupt Line
00h

Many of the fields of the PCI header space contain hardware default values that are either fixed or
can be overridden using EEPROM, but cannot be independently specified for each logical LAN
device. The following fields are considered to be common to both LAN devices:
Vendor ID

The Vendor ID of the Ethernet controller can be specified via EEPROM, but
only a single value can be specified. The value is reflected identically for both
LAN devices.

Revision

The revision number of the Ethernet controller is reflected identically for both
LAN devices.
This field indicates if a device is single function or multifunction. The value
reflected in this field is reflected identically for both LAN devices, but the actual
value reflected depends on LAN Disable configuration.

Header Type

When both Ethernet controller LAN ports are enabled, both PCI headers return
80h in this field, acknowledging being part of a multi-function device. LAN A
exists as device “function 0”, while LAN B exists as device “function 1”.
If one of the LAN ports is disabled, then only a single-function device is
indicated (this field returns a value of 00h), and the LAN exists as device
“function 0”.

208

Subsystem ID

The Subsystem ID of the Ethernet controller can be specified via EEPROM,
but only a single value can be specified. The value is reflected identically for
both LAN devices.

Subsystem Vendor ID

The Subsystem Vendor ID of the Ethernet controller can be specified via
EEPROM, but only a single value can be specified. The value is reflected
identically for both LAN devices.

Class Code,
Cap_Ptr,
Max Latency,
Min Grant

These fields reflect fixed values that are constant values reflected for both LAN
devices.

Software Developer’s Manual

Dual Port Characteristics

The following fields are implemented unique to each LAN device:
Device ID

The Device ID reflected for each LAN device can be independently specified
via EEPROM.

Command,
Status

Each LAN device implements its own command/status registers.

Latency Timer,
Cache Line Size

Each LAN device implements these registers uniquely. The system should
program these fields identically for each LAN to ensure consistent behavior
and performance of each device.

Memory BAR,
Flash BAR,
IO BAR,

Each LAN device implements its own Base Address registers, allowing each
device to claim its own address region(s).

Expansion ROM BAR

Interrupt Pin

12.2.2

Each LAN device independently indicates which interrupt pin (INTA# or INTB#)
is used by that Ethernet controller’s MAC to signal system interrupts. The
value for each LAN device can be independently specified via EEPROM, but
only if both LAN devices are enabled.

MAC Configuration Register Space
All device control/status registers detailed in Section 13.4, Main Register Descriptions, are
implemented per-LAN device. Each LAN device can be accessed using memory or I/O cycles,
depending on the specific BAR setting(s) established for that LAN device.
Register accesses to each MAC instance are independent. In PCI bus operation, a register access to
one LAN which is retried as a delayed-read requires subsequent accesses to that LAN to retry the
read identically until complete. An outstanding delayed-read for one LAN device does not impact
the Ethernet controller’s ability to accept a register access to the other LAN. Similarly, in PCI-X
bus operation, and register access resulting in a split & split-completion by one LAN device in no
way prevents the other LAN device from accepting and servicing (or splitting) an access to its
register space.

12.2.3

SDP, LED, INT# output
Each LAN device provides an independent set of LED outputs and software-programmable I/O
pins (SDP). Four LED outputs and four SDP pins are provided per LAN device. These pins and
their function are bound to a specific LAN device (eight SDP pins cannot be associated with a
single LAN device, for example).
Each LAN device can use a dedicated pin for signalling interrupts to the system. Two pins, INTA#
and INTB#, exist on the Ethernet controller to signal interrupts by the different LAN devices. The
specific pin used by each LAN is configurable when both LAN devices are enabled.

Software Developer’s Manual

209

Dual Port Characteristics

12.3

Shared EEPROM
The Ethernet controller uses a single EEPROM device to configure hardware default parameters
for both LAN devices, including Ethernet Individual Addresses (IA), LED behaviors, receive
packet-filters for manageability and wakeup capability, etc. Certain EEPROM words are used to
specify hardware parameters which are LAN device-independent (such as those that affect circuits
behavior). Other EEPROM words are associated with a specific LAN device. LAN A and LAN B
accesses the EEPROM to obtain their respective configuration settings.

12.3.1

EEPROM Map
The EEPROM map identifies those words configuring both LAN devices or the entire Ethernet
controller component as “LAN A/B Shared”. Those words configuring a specific LAN device
parameters are identified as either “LAN A” or “LAN B”.
The following EEPROM words warrant additional notes specifically related to dual-LAN support:

Ethernet Address (IA)
(LAN A/B shared)

Initialization Control 1,
Initialization Control 2
(LAN A/B shared)

12.3.2

The EEPROM specifies the IA associated with the LAN A device and used as
the hardware default of the Receive Address Registers for that device. The
hardware-default IA for the LAN B device is automatically determined by the
same EEPROM word, and is set to the value of {Ethernet IA LAN A with its
least significant bit inverted}.
These EEPROM words specify hardware-default values for parameters that
apply a single value to both LAN devices, such as link configuration
parameters required for auto-negotiation, wakeup settings, PCI/PCI-X bus
advertised capabilities, etc.

Initialization Control 3
(LAN A, LAN B unique)

This EEPROM word configures default values associated with each LAN
device’s hardware connections, including which link mode (internal PHY,
external TBI SerDes) is used with this LAN device. Because a separate
EEPROM word configures the defaults for each LAN, extra care must be taken
to ensure that the EEPROM image does not specify a resource conflict. For
example, multiple LAN devices both attempting to utilize the external TBI
transceiver interface at once.

Management Control

This EEPROM word configures manageability parameters. Note that this word
controls whether an internal ASF controller is enabled/disabled for this LAN,
and whether the SMBus is enabled/disabled for this LAN. Extra care must be
taken to ensure that the EEPROM image does not specify a resource conflict –
if an internal ASF controller is being used, it can only be enabled for a single
LAN device. The SMBus can only be enabled for a single LAN device.

EEPROM Arbitration
The Ethernet controller uses a single EEPROM to store hardware configuration words for both
LAN devices. The words used by each specific LAN device are noted in the EEPROM map. Each
LAN device obtains its EEPROM configuration parameters by performing its own independent
EEPROM read. Each LAN device reads the entire EEPROM image, verifying the EEPROM
signature, and applying the word(s) appropriate for the specific LAN device. The Ethernet
controller internally arbitrates between EEPROM access by the two LAN devices, to ensure that
each device is able to perform a complete, uninterrupted EEPROM read sequence.

210

Software Developer’s Manual

Dual Port Characteristics

The result of multiple LAN devices’ reading EEPROM is that power-on and reset-initiated
EEPROM read sequences might appear slightly differently from the sequences illustrated during
the discussion of power-state transitions (Section 6.3.2). Those illustrations indicate EEPROM
read periods without distinguishing between reads by LAN A versus LAN B devices. At initial
power-on, both LAN devices always execute an EEPROM read sequence. However, since the
enabling/disabling of a particular LAN device occurs on the deassertion of PCI reset, the post-reset
EEPROM read sequence(s) are only performed by LAN device(s) that are enabled. The following
illustration more clearly illustrates the EEPROM read sequence for a scenario where a single LAN
device is enabled:

Power
tppg
1
LAN_PWR_GOOD
ttxpg
125 MHz XTAL IN

CLK
tpgrst
RST#

2
Reading EEPROM

tee
EEPROM read,
LAN A device

6 tclkpr

tpgee

EEPROM read,
LAN B device

tpree 3 tee
EEPROM read,
LAN A device
9 teepci

PCI Pins

12.4

Running

Shared FLASH
The Ethernet controller provides an interface to an external FLASH/ROM memory device, as
described in Section 7. This FLASH/ROM device can be mapped into memory and/or I/O address
space for each LAN device through the use of PCI Base Address Registers (BARs). Bit 3 of the
EEPROM Initialization Control Word 3 associated with each LAN device selectively disables/
enables whether the FLASH can be mapped for each LAN device by controlling the BAR register
advertisement and writeability.

12.4.1

FLASH Access Contention
Unlike the shared EEPROM implementation, the Ethernet controller does NOT implement any
internal arbitration between FLASH accesses initiated through the LAN A device and those
initiated through the LAN B device. If accesses from both LAN devices are initiated during the
same approximate time window, access contention can occur. If contention occurs, the external
FLASH addresses can be corrupted or unstable throughout the access. During writes to FLASH,
contention can result in corrupt or unstable data values; contention during reads can result in
erroneous read data being returned.

Software Developer’s Manual

211

Dual Port Characteristics

Note:

Access contention to FLASH by both LAN devices is more than likely to result in indeterminate
data results (during read transactions), corrupted FLASH (during write transactions), or other
unpredictable behavior.
To avoid this contention, accesses from both LAN devices MUST be synchronized using external
software synchronization of the memory or I/O transactions responsible for the access. It might be
possible to ensure contention-avoidance simply by nature of software sequentially.

12.5

LAN Disable
For a LOM design, it might be desirable for the system to provide BIOS-setup capability for
selectively enabling or disabling LOM devices. This might allow an end-user more control over
system resource-management, avoid conflicts with add-in NIC solutions, etc. The Ethernet
controller provides support for selectively enabling or disabling one or both LAN device(s) in the
system.

12.5.1

Overview
Device presence (or non-presence) must be established early during BIOS execution in order to
ensure that BIOS resource-allocation (of interrupts, of memory or IO regions) is done according to
devices that are present only. This is frequently accomplished using a BIOS CVDR (Configuration
Values Driven on Reset) mechanism. The Ethernet controller LAN-disable mechanism is
implemented in order to be compatible with such a solution. The Ethernet controller samples two
pins (FLASH data pins, bits 1 and 0) on reset to determine the LAN-enable configuration.
When a particular LAN is disabled, all internal clocks to that LAN are disabled, the device is held
in reset, and the internal PHY for that LAN is powered-down. The device does not respond to PCI
configuration cycles. Effectively, the LAN device becomes invisible to the system from both a
configuration and power-consumption standpoint.

Note:

12.5.2

Since the LAN-disable mechanisms is implemented using the FLASH data pins, this mechanism
can only be used when no FLASH device is present (FLASH disabled). An Ethernet controllerbased NIC built with support for a FLASH device always enables both LAN devices.

Values Sampled on Reset
The Ethernet controller samples values from the pins FLSH_DATA[1] and FLSH_DATA[0] on the
rising edge of LAN_PWR_GOOD and RST#. Based on the values sampled, the LAN devices are
enabled/disabled according to the following table:

212

Pin sampled

LAN device controlled

FLSH_DATA[0]

LAN A device

FLSH_DATA[1]

LAN B device

Enable/Disable
Vcc/logic 1b = enabled
Vss/logic 0b = disabled
???

Software Developer’s Manual

Dual Port Characteristics

12.5.3

Multi-Function Advertisement
If one of the LAN devices is disabled, the Ethernet controller no longer is a multi-function device.
It normally reports a 01h in the PCI Configuration Header field Header Type, indicating multifunction capability. However, if a LAN id disabled, it reports a 0h in this filed to signify singlefunction capability.

12.5.4

Interrupt Use
When both LAN devices are enabled, the Ethernet controller uses both the INTA# and INTB# pins
for interrupt-reporting. The EEPROM Initialization Control Word 3 (bit 4) associated with each
LAN device controls which of these two pins is used for each LAN device. The specific interrupt
pin used is reported in the PCI Configuration Header Interrupt Pin field associated with each LAN
device.
However, if either LAN device is disabled, then the INTA# be used for the remaining LAN device,
regardless of the EEPROM configuration. Under these circumstances, the Interrupt Pin field of the
PCI Header always reports a value of 1h, indicating INTA# usage.

12.5.5

Power Reporting
When both LAN devices are enabled, the PCI Power Management Register Block has the
capability of reporting a Common Power value. The Common Power value is reflected in the data
field of the PCI Power Management registers. The value reported as Common Power is specified
via EEPROM, and is reflected in the data field each time the Data_Select field has a value of 8h
(8h = Common Power Value Select).
When either LAN is disabled and the Ethernet controller appears as a single-function device, the
Common Power value, if selected, reports 0h (undefined value), as Common Power is undefined
for a single-function device.

Software Developer’s Manual

213

Dual Port Characteristics

12.5.6

Summary
The following table lists the various LAN enabled/disabled configurations possible:
CVDR values sampled-on-reset
FLSH_DATA[1]

FLSH_DATA[0]

LAN
device

Enabled/
Disabled

PCI
function

Interrupt Line Used


(enabled)

INTA# or INTB# (specified by LAN A
EEPROM InitCtrl3 value)


(enabled)

INTA# or INTB# (specified by LAN B
EEPROM InitCtrl3 value)

INTA#

n/a

A
B

214

X
(disabled)

n/a

INTA#

n/a

Software Developer’s Manual

Introduction
This section details the state inside the PCI/PCI-X Family of Gigabit Ethernet Controllers that are
visible to the programmer. In some cases, it describes hardware structures invisible to software in
order to clarify a concept.
The address space within the Ethernet controller is divided up into eight main categories:
• PCI
• General Configuration and Wakeup
• Interrupt
• MAC Receive
• MAC Transmit
• PHY Receive, Transmit and Special Function
• Statistics
• Diagnostic State (not used in normal operation)
The Ethernet controller’s address space is mapped into four regions with PCI Base Address
Registers described in Table 13-2. These regions are shown as follows.
Internal registers and memories (including PHY)

Memory

128 KB

Flash (optional)

Memory

64 - 512 KB

Expansion ROM (optional)

Memory

64 - 512 KB

Internal registers and memories, Flash (optional)

I/O

8 Bytes

Both the Flash an Expansion ROM Base Address Registers map the same Flash memory. The
internal registers and memories and Flash can be access through I/O space by doing a level of
indirection, as explained later.
Note:

13.2

The PHY registers are accessed indirectly through the MDI/O interface described in Section 8.2.

Register Conventions
All registers in the Ethernet controller are defined to be 32 bits, should be accessed as 32-bit double
words, and are aligned on a 64-bit boundary. There are exceptions to this rule:
• PCI configuration registers
• I/O space registers (IOADDR and IODATA) are aligned on 32-bit boundaries
• Register pairs where two 32-bit registers make up a larger logical size
• Accesses to Flash memory (through Expansion ROM space or secondary Base Address
Register space) can be byte, word, double word or quadword accesses.

• Reserved bit positions. Some registers contain certain bits that are marked as “reserved.”
These bits should never be set to a value of 1b by software. Reads from registers containing
reserved bits can return indeterminate values in the reserved bit positions unless read values
are explicitly stated. When read, these reserved bits should be ignored by software.

Software Developer’s Manual

215

• Reserved and/or undefined addresses. Any register not explicitly declared in this
specification should be considered to be reserved and should not be written. Writing to
reserved or undefined register addresses can cause indeterminate behavior. Reads from
reserved or undefined configuration register addresses can return indeterminate values unless
read values are explicitly stated for specific addresses.

• Initial values. Most registers define the initial hardware values prior to being programmed. In
some cases, hardware initial values are undefined and are listed as such via the text
“undefined,” “unknown,” or “X.” Some such values might need setting through EEPROM
configuration or software in order for proper operation to occur; this need is dependent on the
function of the bit. Other registers might cite a hardware default that is overridden by a higher
precedence operation. Operations that might supersede hardware defaults can include a valid
EEPORM load, completion of a hardware operation (such as hardware Auto-Negotiation), or
writing of a different register whose value is then reflected in another bit.
For registers that should be accessed as 32-bit double words, partial writes (less than a 32-bit
double word) is ignored. Partial reads return all 32 bits of data regardless of the byte enables.
Partial reads to read-on-clear registers (for example, ICR) can have unexpected results since all 32
bits are actually read regardless of the byte enables. Partial reads should not be performed.
All statistics registers are implemented as 32-bit registers. 64-bit accesses to these registers must
have the upper byte enables deasserted. 32-bit registers with addresses not on a quadword
boundary cannot be accessed through a 64-bit access.
Note:

The PHY registers are accessed indirectly through the MDI/O interface.

13.2.1

Memory and I/O Address Decoding

13.2.1.1

Memory-Mapped Access to Internal Registers and Memories
The internal registers and memories can be accessed as direct memory-mapped offsets from the
base address register (BAR0 or BAR0/BAR1, see Section 4.1). Refer to Table 13-2 for the
appropriate offset for each specific internal register.

13.2.1.2

Memory-Mapped Access to FLASH
The external Flash can be accessed using direct memory-mapped offsets from the Flash base
address register (BAR1 or BAR2/BAR3, see Section 4.1). The Flash is only accessible if enabled
through the EEPROM Initialization Control Word, and if the Flash Base Address register contains
a valid (non-zero) base memory address. For accesses, the offset from the Flash BAR corresponds
to the offset into the flash actual physical memory space.

13.2.1.3

Memory-Mapped Access to Expansion ROM
The external Flash can also be accessed as a memory-mapped expansion ROM. Accesses to offsets
starting from the Expansion ROM Base address (see Section 4.1) reference the Flash provided that
access is enabled through the EEPROM Initialization Control Word, and if the Expansion ROM
Base Address register contains a valid (non-zero) base memory address.

216

Software Developer’s Manual

13.2.2

I/O-Mapped Internal Register, Internal Memory, and Flash1
To support pre-boot operation (prior to the allocation of physical memory base addresses), all
internal registers, memories, and Flash can be accessed using I/O operations. I/O accesses are
supported only if an I/O Base Address is allocated and mapped (BAR2 or BAR4, see Section 4.1),
the BAR contains a valid (non-zero value), and I/O address decoding is enabled in the PCI/PCIX
configuration.
When an I/O BAR is mapped, the I/O address range allocated opens a 32-byte window in the
system I/O address map. Within this window, two I/O addressable registers are implemented:
IOADDR and IODATA. The IOADDR register is used to specify a reference to an internal register,
memory, or Flash, and then the IODATA register is used as a window to the register, memory or
Flash address specified by IOADDR:
Offset

13.2.2.1

Abbreviation

Name

00000000h

IOADDR

Internal Register, Internal Memory, or Flash Location
Address
00000h - 1FFFFh — Internal Registers and Memories
20000h - 7FFFFh — Undefined
80000h - FFFFFh — Flash

00000004h

IODATA

Data field for reads or writes to the Internal Register Internal
Memory, or Flash location as identified by the current value
in IOADDR. All 32 bits of this register are read/write-able.

Size

4 bytes

IOADDR
The IOADDR register must always be written as a DWORD access (for example, the C/BE#[3:0]
byte enables must all be enabled). Writes that are less than 32 bits are ignored. Reads of any size
return a DWORD of data. However, the chipset or CPU can only return a subset of that DWORD.
For Intel architecture programmers, the IN and OUT instructions must be used to cause I/O cycles
to be used on the PCI bus. Since writes must be to a 32-bit quantity, the source register of the OUT
instruction must be EAX (the only 32-bit register supported by the OUT command). For reads, the
IN instruction can have any size target register, but it is recommended that the 32-bit EAX register
be used.
Since only a particular range is addressable, the upper bits of this register are hard coded to 0b. Bits
31 through 20 are not write-able and always read back as 0b.
At hardware reset (LAN_PWR_GOOD) or PCI Reset, this register value resets to 00h. Once
written, the value is retained until the next write or reset.

13.2.2.2

IODATA
The IODATA register must always be written as a DWORD access when the IOADDR register
contains a value for the Internal Register and Memories (00000h - 1FFFCh). In this case, writes
less than 32 bits are ignored.

Not applicable to the 82547GI/EI.

Software Developer’s Manual

217

The IODATA register can be written as a byte, word, or Dword access when the IOADDR register
contains a value for the Flash (80000h - FFFFFh). In this case, the value in IOADDR must be
properly aligned to the data value. Additionally, the lower 2 bits of the IODATA PCI-X access must
correspond to the byte, word, or Dword access.Table 13-1 lists the supported configurations:
Table 13-1. IODATA Register Configurations
Access Type

Ethernet Controller
IOADDR Register Bits
[1:0]

PCI-X1 IODATA Access
AD[1:0] Bits in Address
Phase

PCI-X IODATA Access
AD C/BE#[3:0] Bits in
Data Phase

00b

1110b

01b

1101b

10b

1011b

11b

0111b

00b

1100b

10b

0011b

00b

0000b

BYTE (8 bits)

WORD (16 bits)

DWORD (32 bits)
1.

The 82540EP/EM does not support PCI-X.

Software might need to implement special code to access the Flash memory at a byte or word at a
time. Example code that reads a Flash byte is shown here to illustrate the impact of Table 13-1:
char *IOADDR;
char *IODATA;
IOADDR = IOBASE + 0;
IODATA = IOBASE + 4;
*(IOADDR) = Flash_Byte_Address;
Read_Data = *(IODATA + (Flash_Byte_Address % 4));

Reads to IODATA of any size returns a Dword of data. However, the chipset or CPU can only
return a subset of that Dword.
For Intel architecture programmers, the IN and OUT instructions must be used to cause I/O cycles
to be used on the PCI bus. Where 32-bit quantities are required on writes, the source register of the
OUT instruction must be EAX (the only 32-bit register supported by the OUT command).
Writes and reads to IODATA when the IOADDR register value is in an undefined range (20000h 7FFFCh) should not be performed. Results are indeterminate.
There are no special software timing requirements on accesses to IOADDR or IODATA. All
accesses are immediate except when data is not readily available or acceptable. In this case, the
Ethernet controller delays the results through normal bus methods.

218

Software Developer’s Manual

Table 13-2. Ethernet Controller Register Summary
Category

Offset

Abbreviation

Name

General

00000h

CTRL

Device Control

General

00008h

STATUS

General

00010h

EECD

R/W

Page

R/W

224

Device Status

229

EEPROM/Flash Control/Data

R/W

232

R/W

234

General

00014h

EERD

EEPROM Read (not applicable to the
82544GC/EI)

General

0001Ch

FLA

Flash Access (applicable to the 82541xx and
82547GI/EI only)

R/W

236

General

00018h

CTRL_EXT

Extended Device Control

R/W

237

General

00020h

MDIC

MDI Control

R/W

242

General

00028h

FCAL

Flow Control Address Low

R/W

283

General

0002Ch

FCAH

Flow Control Address High

R/W

283

General

00030h

FCT

Flow Control Type

R/W

284

General

00038h

VET

VLAN EtherType

R/W

284

General

00170h

FCTTV

Flow Control Transmit Timer Value

R/W

285
286

General

00178h

TXCW

Transmit Configuration Word (not applicable to
R/W
the 82540EP/EM, 82541xx and 82547GI/EI)

General

00180h

RXCW

Receive Configuration Word (not applicable to
R
the 82540EP/EM, 82541xx and 82547GI/EI)

287

General

00E00h

LEDCTL

LED Control (not applicable to the 82544GC/
EI)

R/W

289

DMA

01000h

PBA

Packet Buffer Allocation

R/W

292

Interrupt

000C0h

ICR

Interrupt Cause Read

293

Interrupt Throttling (not applicable to the
82544GC/EI)

R/W

295

Interrupt

000C4h

ITR

Interrupt

000C8h

ICS

Interrupt Cause Set

296

Interrupt

000D0h

IMS

Interrupt Mask Set/Read

R/W

297

Interrupt

000D8h

IMC

Interrupt Mask Clear

298

Receive

00100h

RCTL

Receive Control

R/W

300

Receive

02160h

FCRTL

Flow Control Receive Threshold Low

R/W

304

Receive

02168h

FCRTH

Flow Control Receive Threshold High

R/W

305

Receive

02800h

RDBAL

Receive Descriptor Base Low

R/W

306

Receive

02804h

RDBAH

Receive Descriptor Base High

R/W

306

Receive

02808h

RDLEN

Receive Descriptor Length

R/W

307

Receive

02810h

RDH

Receive Descriptor Head

R/W

307

Receive

02818h

RDT

Receive Descriptor Tail

R/W

308

Receive

02820h

RDTR

Receive Delay Timer

R/W

308

R/W

309

R/W

310

Receive

0282Ch

RADV

Receive Interrupt Absolute Delay Timer (not
applicable to the 82544GC/EI)

Receive

02C00h

RSRPD

Receive Small Packet Detect Interrupt (not
applicable to the 82544GC/EI)

Transmit

00400h

TCTL

Transmit Control

R/W

310

Transmit

00410h

TIPG

Transmit IPG

R/W

312

Transmit

00458h

AIFS

Adaptive IFS Throttle - AIT

R/W

314

Transmit

03800h

TDBAL

Transmit Descriptor Base Low

R/W

315

Transmit

03804h

TDBAH

Transmit Descriptor Base High

R/W

316

Transmit

03808h

TDLEN

Transmit Descriptor Length

R/W

316

Transmit

03810h

TDH

Transmit Descriptor Head

R/W

317

Transmit

03818h

TDT

Transmit Descriptor Tail

R/W

318

Transmit

03820h

TIDV

Transmit Interrupt Delay Value

R/W

318

Software Developer’s Manual

219

Category

Abbreviation

Name

R/W

Page

TX DMA

03000h

TXDMAC

TX DMA Control (applicable to the 82544GC/
EI only)

TX DMA

03828h

TXDCTL

Transmit Descriptor Control

R/W

319

TADV

Transmit Absolute Interrupt Delay Timer (not
applicable to the 82544GC/EI)

R/W

321

TX DMA

220

Offset

0282Ch

R/W

319

TX DMA

03830h

TSPMT

TCP Segmentation Pad and Threshold

R/W

322

RX DMA

02828h

RXDCTL

Receive Descriptor Control

R/W

324

RX DMA

05000h

RXCSUM

Receive Checksum Control

R/W

325

Receive

05200h053FCh

MTA[127:0]

Multicast Table Array (n)

R/W

327

Receive

05400h05478h

RAL(8n)

Receive Address Low (n)

R/W

329

Receive

05404h0547Ch

RAH(8n)

Receive Address High (n)

R/W

329

Receive

05600h057FCh

VFTA[127:0]

VLAN Filter Table Array (n)
Not applicable to the 82541ER

R/W

330

Wakeup

05800h

WUC

Wakeup Control

R/W

331

Wakeup

05808h

WUFC

Wakeup Filter Control

R/W

332

Wakeup

05810h

WUS

Wakeup Status

333

Wakeup

05838h

IPAV

IP Address Valid

R/W

335

Wakeup

05840h05858h

IP4AT
IPv4 Address Table
IPAT (82544GC/
IP Address Table (82544GC/EI)
EI)

R/W

336

Wakeup

05880h0588Ch

IP6AT

IPv6 Address Table (not applicable to the
82544GC/EI)

R/W

337

Wakeup

05900h

WUPL

Wakeup Packet Length

R/W

338

Wakeup

05A00h05A7Ch

WUPM

Wakeup Packet Memory

R/W

338

Wakeup

05F00h05F18h

FFLT

Flexible Filter Length Table

R/W

338

Wakeup

09000h093F8h

FFMT

Flexible Filter Mask Table

R/W

339

Wakeup

09800h09BF8h

FFVT

Flexible Filter Value Table

R/W

340

Statistics

04000h

CRCERRS

CRC Error Count

341

Statistics

04004h

ALGNERRC

Alignment Error Count

341

Statistics

04008h

SYMERRS

Symbol Error Count

342

Statistics

0400Ch

RXERRC

RX Error Count

342

Statistics

04010h

MPC

Missed Packets Count

343

Statistics

04014h

SCC

Single Collision Count

343

Statistics

04018h

ECOL

Excessive Collisions Count

344

Statistics

0401Ch

MCC

Multiple Collision Count

344

Statistics

04020h

LATECOL

Late Collisions Count

345

Statistics

04028h

COLC

Collision Count

345

Statistics

04030h

Defer Count

346

Statistics

04034h

TNCRS

Transmit - No CRS

346

Statistics

04038h

SEC

Sequence Error Count

347

Statistics

0403Ch

CEXTERR

Carrier Extension Error Count

347

Statistics

04040h

RLEC

Receive Length Error Count

348

Statistics

04048h

XONRXC

XON Received Count

348

Statistics

0404Ch

XONTXC

XON Transmitted Count

349

Statistics

04050h

XOFFRXC

XOFF Received Count

349

Software Developer’s Manual

Category

Offset

Abbreviation

Name

R/W

Page

Statistics

04054h

XOFFTXC

XOFF Transmitted Count

349

Statistics

04058h

FCRUC

FC Received Unsupported Count

R/W

350

Statistics

0405Ch

PRC64

Packets Received (64 Bytes) Count

R/W

350

Statistics

04060h

PRC127

Packets Received (65-127 Bytes) Count

R/W

351

Statistics

04064h

PRC255

Packets Received (128-255 Bytes) Count

R/W

351

Statistics

04068h

PRC511

Packets Received (256-511 Bytes) Count

R/W

352

Statistics

0406Ch

PRC1023

Packets Received (512-1023 Bytes) Count

R/W

352

Statistics

04070h

PRC1522

Packets Received (1024-Max Bytes)

R/W

353

Statistics

04074h

GPRC

Good Packets Received Count

353

Statistics

04078h

BPRC

Broadcast Packets Received Count

354

Statistics

0407Ch

MPRC

Multicast Packets Received Count

354

Statistics

04080h

GPTC

Good Packets Transmitted Count

355

Statistics

04088h

GORCL

Good Octets Received Count (Low)

355

Statistics

0408Ch

GORCH

Good Octets Received Count (Hi)

355

Statistics

04090h

GOTCL

Good Octets Transmitted Count (Low)

356

Statistics

04094h

GOTCH

Good Octets Transmitted Count (Hi)

356

Statistics

040A0h

RNBC

Receive No Buffers Count

356

Statistics

040A4h

RUC

Receive Undersize Count

357

Statistics

040A8h

RFC

Receive Fragment Count

357

Statistics

040ACh

ROC

Receive Oversize Count

358

Statistics

040B0h

RJC

Receive Jabber Count

358

359

Statistics

040B4h

MGTPRC

Management Packets Received Count (not
applicable to the 82544GC/EI or 82541ER)

Statistics

040B8h

MGTPDC

Management Packets Dropped Count (not
applicable to the 82544GC/EI or 82541ER)

360

Statistics

040BCh

MGTPTC

Management Pkts Transmitted Count (not
applicable to the 82544GC/EI or 82541ER)

360

Statistics

040C0h

TORL

Total Octets Received (Lo)

360

Statistics

040C4h

TORH

Total Octets Received (Hi)

360

Statistics

040C8h

TOTL

Total Octets Transmitted (Lo)

361

Statistics

040CCh

TOTH

Total Octets Transmitted (Hi)

361

Statistics

040D0h

TPR

Total Packets Received

362

Statistics

040D4h

TPT

Total Packets Transmitted

362

Statistics

040D8h

PTC64

Packets Transmitted (64 Bytes) Count

363

Statistics

040DCh

PTC127

Packets Transmitted (65-127 Bytes) Count

363

Statistics

040E0h

PTC255

Packets Transmitted (128-255 Bytes) Count

364

Statistics

040E4h

PTC511

Packets Transmitted (256-511 Bytes) Count

364

Statistics

040E8h

PTC1023

Packets Transmitted (512-1023 Bytes) Count

365

Statistics

040ECh

PTC1522

Packets Transmitted (1024 Bytes or Greater)
Count

Statistics

040F0h

MPTC

Multicast Packets Transmitted Count

366

Statistics

040F4h

BPTC

Broadcast Packets Transmitted Count

366

Statistics

040F8h

TSCTC

TCP Segmentation Context Transmitted Count R

367

Statistics

040FCh

TSCTFC

TCP Segmentation Context Tx Fail Count

367

Diagnostic

02410h

RDFH

Receive Data FIFO Head

R/W

368

Diagnostic

02418h

RDFT

Receive Data FIFO Tail

R/W

368

Diagnostic

02420h

RDFHS

Receive Data FIFO Head Saved Register

R/W

369

Diagnostic

02428h

RDFTS

Receive Data FIFO Tail Saved Register

R/W

369

Diagnostic

02430h

RDFPC

Receive Data FIFO Packet Count

R/W

370

Diagnostic

03410h

TDFH

Transmit Data FIFO Head

R/W

370

Software Developer’s Manual

221

Category
Diagnostic

Note:

Abbreviation
TDFT

Name

R/W

Page

Transmit Data FIFO Tail

R/W

371

Diagnostic

03420h

TDFHS

Transmit Data FIFO Head Saved Register

R/W

371

Diagnostic

03428h

TDFTS

Transmit Data FIFO Tail Saved Register

R/W

372

Diagnostic

03430h

TDFPC

Transmit Data FIFO Packet Count

R/W

372

Diagnostic

10000h1FFFCh

PBM

Packet Buffer Memory (n)

R/W

373

The PHY registers are accessed indirectly through the MDI/O interface described in Section 8.2.

Category

MDI
Register

Abbreviation

Name

R/W

Page

PHY

00d

PCTRL

PHY Control Register

R/W

245

PHY

01d

PSTATUS

PHY Status Register

248

PHY

02d

PID

PHY Identifier (LSB)

250

PHY

03d

EPID

Extended PHY Identifier (MSB)

250

PHY

04d

ANA

Auto-Negotiation Advertisement Register

R/W

251

PHY

05d

LPA

Link Partner Ability Register (Base Page)

255

PHY

06d

ANE

Auto-Negotiation Expansion Register

258

PHY

07d

NPT

Next Page Transmit Register

R/W

259

PHY

08d

LPN

Link Partner Next Page Register

260

PHY

09d

GCON

1000BASE-T Control Register

R/W

261

PHY

10d

GSTATUS

1000BASE-T Status Register

262

PHY

15d

EPSTATUS

Extended PHY Status Register

263

PHY

16d

PSCON

PHY Specific Control Register

R/W

264

PHY

17d

PSSTAT

PHY Specific Status Register

267

PHY

18d

PINTE

PHY Interrupt Enable

R/W

270

PHY

19d

PINTS

PHY Interrupt Status

272

PHY

20d

EPSCON1
EPSCON (82544GC/
EI)

Extended PHY Specific Control 1

R/W

275

PHY

21d

PREC

PHY Receive Error Counter

277

279

PHY

23d1

PGSTAT

PHY Global Status
SPEED_TEN_LED and LINK_ACT_LED2

PHY

24d1

PLED

PHY LED Control
SPEED_100_LED and
SPEED_1000_LED2

R/W

280

PHY

26d

EPSCON2

Extended PHY Specific Status (not
applicable to the 82540EP/EM or 82544GC/ R/W
EI)

281

PHY

27d3

EPSSTAT

Extended PHY Specific Status

R/W

281

PHY

29d

R30PS

PHY

30d

R30AW

282

PHY

31d

PPAGE

Page Select (82541xx and 82547GI/EI
only)

283

1.
2.
3.

222

Offset
03418h

R/W

Applicable to the 82544GC/EI, 82541xx, and 82547GI/EI only.
Applicable to the 82541xx, and 82547GI/EI only.
Applicable to the 82544GC/EI only.

Software Developer’s Manual

13.3

PCI-X Register Access Split1
The PCI-X specification states that accesses to internal device memory spaces must complete
within a specific target initial latency, or else the device should signal that it completes the
transaction later using a split-completion operation. Due to internal access latencies, read accesses
to most device registers in the Ethernet controller exceeds target initial-access latencies, and
therefore are split.
Once a register read operation has been split, the device may, as part of normal operation, initiate a
large inbound or outbound transmit or receive data burst transaction. The split completion for the
pending register read might be forced to wait until the data burst completes. Therefore, the read
access-delay for most registers can be indeterminate (although is generally bounded by the nature
or normal burst transactions).
A small subset of the internal register space has been identified as most critical for high-performance driver execution. The variable completion delay for access to some registers could
potentially limit the performance of such critical routines as Interrupt Service Routines (ISRs). To
help minimize potential critical routine performance, read accesses to a small subset of internal
register space will instead complete without being split. These registers are listed as follows:
Category

Offset

Abbreviation

Name

General

00000h

CTRL

Device Control Register

General

00008h

STATUS

Device Status Register

General

00010h

EECD

EEPROM/Flash Control/Data Register

General

00018h

CTRL_EXT

Extended Device Control Register

General

00020h

MDIC

MDI Control Register

General

00028h

FCAL

Flow Control Address Low

General

0002Ch

FCAH

Flow Control Address High

General

00030h

FCT

Flow Control Type

General

00038h

VET

VLAN Ether Type

General

00170h

FCTTV

Flow Control Transmit Timer Value

General

00178h

TXCW

Transmit Configuration Word

General

00180h

RXCW

Receive Configuration Word

General

01000h

PBA

Packet Buffer Allocation

Interrupt

000C0h

ICR

Interrupt Cause Read

Interrupt

000C8h

ICS

Interrupt Cause Set

Not applicable to the 82540EP/EM, 82541xx, or 82547GI/EI.

Software Developer’s Manual

223

Category

Offset

Abbreviation

Name

Interrupt

000D0h

IMS

Interrupt Mask Set/Read

Interrupt

000D8h

IMC

Interrupt Mask Clear

Transmit

00400h

TCTL

Transmit Control

The EEPROM configuration bit “Force CSR Read Split” (Initialization Control Word 2, word 0Fh)
provides the ability to configure the device to split all internal register accesses, rather than
providing non-split behavior for the registers listed.

13.4

Main Register Descriptions
This section contains detailed register descriptions for general purpose, DMA, interrupt, receive,
and transmit registers. These registers correspond to the main functions of the Ethernet controller.

13.4.1

Device Control Register
CTRL (00000h; R/W)
This register and the Extended Device Control register (CTRL_EXT) control the major operational
modes for the Ethernet controller.
While software writes to this register to control device settings, several bits (such as FD and
SPEED) can be overridden depending on other bit settings and the resultant link configuration
determined by the PHY’s Auto-Negotiation resolution.

Note:

224

TBI Mode is used only by the 82544GC/EI Ethernet controller. Internal SerDes mode is used only
by the 82546GB/EB and 82545GM/EM Ethernet controllers.

Software Developer’s Manual

Table 13-3. CTRL Register Bit Description
31

0
Device Control Bits

Field

Reserved

Bit(s)

Initial
Value

Description
Full-Duplex
Enables software to override the hardware Auto-Negotiation
function. The FD sets the duplex mode only if CTRL.FRCDPLX
is set.
When cleared, the Ethernet controller operates in half-duplex;
when set, the Ethernet controller operates in full-duplex.
When the Ethernet controller operates in TBI mode/internal
SerDes mode, and the AN Hardware is enabled, this bit is
ignored. When the Ethernet controller operates in TBI mode/
internal SerDes, and the AN Hardware is disabled, or the link is
forced, this bit should be set by software.
When the Ethernet controller operates in internal PHY mode,
the FD bit is set by software based on AN and data rate
resolution.
Configurable through the EEPROM.

0b1

2:1

These bits are reserved and should be set to 00b.

LRST

Link Reset (not applicable to the 82540EP/EM, 82541xx, or
82547GI/EI)
0b = Normal; 1b = Link Reset
Applicable only in TBI mode/internal SerDes of operation. Used
to reset the link control logic and restart the Auto-Negotiation
process, when TXCW.ANE is set and TBI mode/internal SerDes
is enabled.
When set, transmission and reception are halted regardless of
TBI mode/internal SerDes setting. A transition to 0b initiates the
Auto-Negotiation function. Configurable from the EEPROM,
allowing initiation of Auto-Negotiation function at power up.

Reserved

Reserved
Factory use only. Should be written with 0b.

Auto-Speed Detection Enable.
When set, the Ethernet controller automatically detects the
resolved speed of the link by sampling the link in internal PHY
mode and self-configures the appropriate status and control
bits. Software must also set the SLU bit for this operation. This
function is ignored in TBI mode/internal Serdes. The ASD
feature provides a method of determining the link speed without
the need for software accesses to the MII management
registers.

ASDE

Software Developer’s Manual

225

Field

SLU

ILOS

Initial
Value

Description

Set Link Up
In TBI mode/internal SerDes, provides manual link
configuration. When set, the Link Up signal is forced high once
receiver synchronization is achieved (LOS not asserted) using
CTRL.FD to determine the duplex mode. This operation
bypasses the link configuration process. If Auto-Negotiation is
enabled (TXCW.ANE equals 1b), then Set Link Up is ignored.
In internal PHY mode, this bit must be set to 1b to permit the
Ethernet controller to recognize the I_LOS/I_LIND link signal
from the PHY.
The "Set Link Up" is normally initialized to 0b. However, if either
the APM Enable or SMBus Enable bits are set in the EEPROM
then it is initialized to 1b, ensuring MAC/PHY communication
during preboot states (for example, the 82547EI and 82541EI).
Driver software sets this bit when the driver software initializes,
therefore LED indications (link, activity, speed) are not active
until the software driver loads even though the PHY has autonegotiated and established link with a partner on the Ethernet.
See Section 8.6 for more information about Auto-Negotiation
and link configuration in the various modes.
Configurable through the EEPROM.

Invert Loss-of-Signal (LOS).
0b = do not invert (active high input signal); 1b = invert signal
(active low input signal).
If using the internal PHY, this bit should be set to 0b to ensure
proper communication with the MAC. If using an external TBI
device, this bit can be set if the Ethernet controller provides a
link loss indication with negative polarity.
Note: This is a reserved bit for the 82541xx and 82547GI/EI.

SPEED

9:8

10b

Speed selection.
These bits determine the speed configuration and are written by
software after reading the PHY configuration through the MDI/O
interface. These signals are ignored in TBI mode/internal
Serdes or when Auto-Speed Detection (CTRL.ASDE) is
enabled. See Section 8.6 for details.
00b 10 Mb/s
01b 100 Mb/s
10b 1000 Mb/s
11b not used

Reserved

Reserved
Should be written with 0b to ensure future compatibility.

Force Speed
When set, the Ethernet controller speed is configured by
CTRL.SPEED bits. The PHY device must resolve to the same
speed configuration or software must manually set it to the
same speed as the Ethernet controller.
When cleared, this allows the PHY device or ASD function
(CTRL.ASDE is set) to set the Ethernet controller speed.
This bit is superseded by the CTRL_EXT.SPD_BYPS bit, which
has a similar function.
Applicable only in internal PHY mode of operation and is
configurable through EEPROM.

FRCSPD

226

Bit(s)

Software Developer’s Manual

Bit(s)

Initial
Value

Description

FRCDPLX

Force Duplex
When set, software can override the duplex indication from the
PHY which is in internal PHY mode. When set the CTRL.FD bit
sets duplex. When cleared, the CTRL.FD is ignored.

Reserved

17:13

Reserved
Should be written with 0b to ensure future compatibility.
Read as 0b

0b2

SDP0 Data Value. Used to read (write) value of softwarecontrollable IO pin SDP0. If SDP0 is configured as an output
(SDP0_IODIR=1b), this bit controls the value driven on the pin
(initial value EEPROM-configurable). If SDP0 is configured as
an input, reads return the current value of the pin.

0b2

SDP1 Data Value. Used to read (write) value of softwarecontrollable IO pin SDP1. If SDP1 is configured as an output
(SDP1_IODIR=1b), this bit controls the value driven on the pin
(initial value EEPROM-configurable). If SDP1 is configured as
an input, reads return the current value of the pin.

0b2

D3Cold Wakeup Capability Advertisement Enable. When set,
D3Cold wakeup capability is advertised based on whether the
AUX_PWR pin advertises presence of auxiliary power (yes if
AUX_PWR is indicated, no otherwise). When 0b, however,
D3Cold wakeup capability is not advertised even if AUX_PWR
presence is indicated. Formerly used as SDP2 pin data value,
initial value is EEPROM-configurable.
Note: Not applicable to the 82541ER.

Field

SDP0_DATA

SDP1_DATA

ADVD3WUC

EN_PHY_
PWR_MGMT

SDP0_IODIR

SDP1_IODIR

Reserved

Software Developer’s Manual

0b2
1b1

PHY Power-Management Enable. When set, the PHY is
informed of power-state transitions and attempts to autonegotiate advertising lower line speeds only (10 or 100 Mb/sec)
when entering D3 or D0u power states with wakeup or
manageability enabled. It again re-negotiates, advertising full
speed capabilities (10/100/1000 Mbps) when transitioning back
to full D0 operational state. If this bit is clear, the PHY automatic
speed/power management capability is disabled, and the PHY
remains operational at its current line speed through powerstate transitions. Formerly used as SDP3 pin data value, initial
value is EEPROM-configurable.

0b2

SDP0 Pin Directionality. Controls whether software-controllable
pin SDP0 is configured as an input or output (0b = input, 1b =
output). Initial value is EEPROM-configurable. This bit is not
affected by software or system reset, only by initial power-on or
direct software writes.

0b2

SDP1 Pin Directionality. Controls whether software-controllable
pin SDP1 is configured as an input or output (0b = input, 1b =
output). Initial value is EEPROM-configurable. This bit is not
affected by software or system reset, only by initial power-on or
direct software writes.

25:24

Reserved. Formerly used as SDP2and SDP3 pin input/output
direction control.

227

Field

RST

RFCE

Initial
Value

Description

Device Reset
0b = normal; 1b = reset. Self clearing.
When set, it globally resets the entire Ethernet controller with
the exception of the PCI configuration registers. All registers
(receive, transmit, interrupt, statistics, etc.), and state machines
are set to their power-on reset values. This reset is equivalent to
a PCI reset, with the one exception being that the PCI
configuration registers are not reset.
To ensure that global device reset has fully completed and that
the Ethernet controller responds to subsequent access, wait
approximately 1 s after setting and before attempting to check
to see if the bit has cleared or to access any other device
register.

Receive Flow Control Enable.
When set, indicates that the Ethernet controller responds to the
reception of flow control packets. Reception and responding to
flow control packets requires matching the content of the
Ethernet controller’s FCAL/H and FCT registers. If AutoNegotiation is enabled, this bit is set to the negotiated flow
control value. See Section 8.6 for more information about AutoNegotiation.

TFCE

Transmit Flow Control Enable.
When set, indicates that the Ethernet controller transmits flow
control packets (XON and XOFF frames) based on the receive
FIFO fullness, or when triggered to do so based on external
control pins (XOFF XON pins when FCTRH.XFCE is set). If
Auto-Negotiation is enabled, this bit is set to the negotiated flow
control value. See Auto-Negotiation for more information.

Reserved

Reserved.
Should be written with 0b to ensure future compatibility. Read as
0b.

VLAN Mode Enable
When set to 1b, all packets transmitted from the Ethernet
controller that have VLE bit set in their descriptor is sent with an
802.1Q header added to the packet. The contents of the header
come from the transmit descriptor and from the VLAN type
register. On receive, VLAN information is stripped from 802.1Q
packets and is loaded to the packet’s descriptor. See Section
9.2 for more details.
Reserved. Should be written with 0b to ensure future
compatibility.
Note: Not applicable to the 82541ER.

PHY Reset
0b = Normal.
1b = Assert hardware reset to the internal PHY.
The technique is to set the bit, wait approximately 3 s, then
clear the bit. For the 82547GI/82541GI (B1 stepping), this
register must be used instead of a PHY register.
Note: For the 82546GB, when resetting the PHY through the
MAC, the PHY should be held in reset for a minimum of 10 ms
before releasing the reset signal.

VME

PHY_RST

1.
2.

228

Bit(s)

82541xx and 82547GI/EI only.
If the signature bits of the EEPROM’s Initialization Control Word 1 match (01b), these bits are read from the EEPROM.

Software Developer’s Manual

The ADVD3WUC bit (Advertise D3Cold Wakeup Capability Enable control) allows the
AUX_PWR pin to determine whether D3Cold support is advertised. If full 1 Gb/s operation in D3
state is desired but the system’s power requirements in this mode would exceed the D3Cold
Wakeup-Enabled specification limit (375 mA at 3.3 V dc), this bit can be used to prevent the
capability from being advertised to the system.
EEPROM settings allow the default PHY behavior to re-negotiate a lower functional link speed in
D3 and D0u states, when PHY operation is still needed for manageability or wakeup capability.
The EN_PHY_PWR_MGMT bit allows this capability to be disabled, in case full 1Gb/s speed is
desired in these states. The PHY is always powered-down in D3 states when unneeded for either
manageability or wakeup support.
Table 13-4. Little-Endian Data Ordering
BEM = 0 (64-bit mode; Little-Endian)
63

13.4.2

0
08

Device Status Register
STATUS (00008h; R)
This register provides software status indication about the Ethernet controller’s settings and modes
of operation.

Note:

TBI Mode is used only by the 82544GC/EI Ethernet controller. Internal SerDes mode is used only
by the 82546GB/EB and 82545GM/EM Ethernet controllers.

Software Developer’s Manual

229

Table 13-5. Status Register Bit Description
31

13 12
Reserved

Field

Function ID

TXOFF

TBIMODE

230

Initial
Value

Bit(s)

3:2

0
Status

Description

Link Full Duplex configuration Indication
When cleared, the Ethernet controller operates in half-duplex;
when set, the Ethernet controller operates in Full duplex.
The FD provides the duplex setting status of the Ethernet
controller as set by either Hardware Auto-Negotiation function,
or by software.

Link Up Indication
0b = no link config; 1b = link config.
For TBI mode/internal SerDes operation: If Auto-Negotiation is
enabled, this bit is set if a valid link is negotiated. If link is forced
through CTRL.SLU, it reflects the status of this control bit.
For internal PHY mode of operation: Reflects the status of the
internal link signal indicating a transition to a Link Up.
See Section 8.6 for more information about Auto-Negotiation.

Function ID.
Provides software a mechanism to determine the Ethernet
controller function number (LAN identifier) for this MAC. Read
as: [0b,0b] LAN A, [0b,1b] LAN B.
Note: These settings are only applicable to the 82546GB/EB.
For all other Ethernet controllers, set these bits to 0b.

Transmission Paused
When set, Indicates the transmit function is in Pause state due
to reception of an XOFF pause frame when symmetrical flow
control is enabled. It is cleared upon expiration of the pause
timer, or receipt of an XON frame. Applicable only while working
in full-duplex flow-control mode of operation.

TBI Mode/internal SerDes Indication
When set, the Ethernet controller is configured to work in TBI
mode/internal SerDes of operation.
When clear, the Ethernet controller is configured to work in
internal PHY mode.
Note: For the 82544GC/EI, reflects the status of the TBI_MODE
input pin.
For all other Ethernet controllers, set this bit to 0b.

Software Developer’s Manual

Field

SPEED

Initial
Value

Bit(s)

7:6

Description

Link speed setting
Indicates the configured speed of the link.
These bits are either forced by software when forcing the link
speed through the CTRL.SPEED control bits, automatically set
by hardware when Auto-Speed Detection is enabled or reflect
the internal indication inputs from the PHY.
When Auto-Speed Detection is enabled, the Ethernet
controller’s speed is configured only once after the internal link
is asserted.
Speed indication is mapped as follows:
00b = 10 Mb/s
01b = 100 Mb/s
10b = 1000 Mb/s
11b = 1000 Mb/s
These bits are not valid in TBI mode/internal SerDes.

ASDV

9:8

Auto Speed Detection Value
Indicates the speed sensed by the Ethernet controller from the
internal PHY. The ASDV status bits are provided for diagnostics
purposes. The ASD function can be initiated by software writing
a logic 1b to the CTRL_EXT.ASDCHK bit. The resultant speed
detection is reflected in these bits. See Section 13.4.6 for
details.

Reserved

Reserved
Reads as 0b.

PCI Bus speed indication
When set, indicates that the PCI bus is running at 66 MHz.
Reflects the M66EN input pin.
Note: Not applicable to the 82547GI/EI.

PCI66

PCI Bus Width indication
When set, indicates that the Ethernet controller is sitting on a
64-bit PCI/PCI-X bus. BUS64 is determined by REQ64#
assertion.

PCIX_MODE1

PCI-X Mode indication

BUS64

When set to 1b, the Ethernet controller is operating in PCI-X
Mode; otherwise, the Ethernet controller is operating in
conventional PCI Mode.
PCIXSPD1

15:14

PCI-X Bus Speed Indication
Attempts to indicate the speed of the bus when operating in a
PCI-X bus. Only valid when STATUS.PCIX_Mode = 1b.
00b = 50-66 MHz
01b = 66-100 MHz
10b = 100-133 MHz
11b = Reserved

Reserved

31:16

Reserved
Reads as 0b.

Not applicable to the 82540EP/EM, 82541xx, or 82547GI/EI.

Software Developer’s Manual

231

13.4.3

EEPROM/Flash Control & Data Register
EECD (00010h; R/W)
This register provides a simplified interface for software accesses to the EEPROM. Software
controls the EEPROM by successive writes to this register.
Data and address information is clocked into the EEPROM by software toggling the EECD.SK bit
(2) of this register with EECD.CS set to 1b.
Data output from the EEPROM is latched into bit 3 of this register and can be accessed by software
through reads of this register.
See Section 5 for more detailed EEPROM information.
Table 13-6. EECD Register Bit Description
31 - 10

Reserved

SIZE

PRES

GNT

REQ

FWE

D
O

DI CS

0
SK

82544GC/EI Only
31

6 5
Reserved

Field

Description

Clock input to the EEPROM
The EESK output signal is mapped to this bit and provides the
serial clock input to the EEPROM. Software clocks the EEPROM
by means of toggling this bit with successive writes to EECD.

Chip select input to the EEPROM
The EECS output signal is mapped to the chip select of the
EEPROM device. Software enables the EEPROM by writing a 1b
to this bit.

Data input to the EEPROM
The EEDI output signal is mapped directly to this bit. Software
provides data input to the EEPROM through writes to this bit.

Data output bit from the EEPROM
The EEDO input signal is mapped directly to this bit in the register
and contains the EEPROM data output. This bit is read-only from
the software perspective – writes to this bit have no effect.

01b

Flash Write Enable Control
These two bits, control whether writes to Flash memory are
allowed.
00b = Not allowed
01b = Flash writes disabled
10b = Flash writes enabled
11b = Not allowed

FWE

232

Initial
Value

Bit

4
FWE

5:4

Software Developer’s Manual

Field

Bit

Initial
Value

Description

EE_REQ

Request EEPROM Access
The software must write a 1b to this bit to get direct EEPROM
access. It has access when EE_GNT is 1b. When the software
completes the access it must write a 0b.

EE_GNT

Grant EEPROM Access
When this bit is 1b the software can access the EEPROM using the
SK, CS, DI, and DO bits.

1b
0b2

EEPROM Present
This bit indicates that an EEPROM is present by monitoring the
EEDO input for a active-low acknowledge by the serial EEPROM
during initial EEPROM scan. 1b = EEPROM present.

EEPROM Size
0b = 1024-bit (64 word) NM93C46 compatible EEPROM
1b = 4096-bit (256 word) NM93C66 compatible EEPROM
This bit indicates the EEPROM size, based on acknowledges seen
during EEPROM scans of different addresses. This bit is read-only.
Note: This is a reserved bit for the 82541xx and 82547GI/EI.

EE_PRES

EE_SIZE

101

EEPROM Size (82541xx and 82547GI/EI)
For Microwire EEPROMs:
0b = 6-bit addressable (64 words).
1b = 8-bit addressable (256 words).
For SPI EEPROMs:
0b = 8-bit addressable.
1b = 16-bit addressable.

Reserved

12:11

00b

Reserved
Should be written with 0b to ensure future compatibility. Reads as
0b.

EE_TYPE

131

EEPROM Type: Reflects the EE_MODE pin. (82541xx and
82547GI/EI)
0b = Microwire.
1b = SPI.

Reserved

31:14

Reserved
Should be written with 0b to ensure future compatibility. Reads as
0b.

1.
2.

Not applicable to the 82544GC/EI.
82541xx and 82547GI/EI only.

This register provides software direct access to the EEPROM. Software can control the EEPROM
by successive writes to this register. Data & address information is clocked into the EEPROM by
software toggling the EESK bit (2) of this register with EECS set to 1b. Data output from the
EEPROM is latched into bit 3 of this register via the internal 62.5 MHz clock and can be accessed
by software via reads of this register.
Note:

Attempts to write to the FLASH device when writes are disabled (FEW = 01b) should not be
attempted. Behavior after such an operation is undefined, and can result in component and/or
system hangs.

Software Developer’s Manual

233

13.4.4

EEPROM Read Register1
EERD (00014h; RW)
Table 13-7. EEPROM Read Register Bit Description
31

16
Data

Field

Bit(s)

8
Address

Initial
Value

RSV.

DONE

1
RSV.

0
START

Description

START

Start Read
Writing a 1b to this bit causes the EEPROM to read a (16-bit) word at
the address stored in the EE_ADDR field and then storing the result in
the EE_DATA field. This bit is self-clearing.

Reserved

3:1

Reserved. Reads as 0b.

DONE

Read Done
Set to 1b when the EEPROM read completes.
Set to 0b when the EEPROM read is in progress.
Writes by software are ignored.

Reserved

7:5

Reserved. Reads as 0b.

ADDR

15:8

Read Address
This field is written by software along with Start Read to indicate the
word to read.

DATA

31:16

Read Data. Data returned from the EEPROM read.

Not applicable to the 82544GC/EI.

234

Software Developer’s Manual

Table 13-8. EEPROM Read Register Bit Description (82541xx and 82547GI/EI)
31

16
Data

Field

START

Bit(s)

Initial
Value

2
Address

DONE

START

Description

Start Read
Writing a 1b to this bit causes the EEPROM to read a (16-bit) word at
the address stored in the EE_ADDR field and then storing the result in
the EE_DATA field. This bit is self-clearing.

DONE

Read Done
Set to 1b when the EEPROM read completes.
Set to 0b when the EEPROM read is in progress.
Writes by software are ignored.

ADDR

15:2

Read Address
This field is written by software along with Start Read to indicate the
word to read.

DATA

31:16

Read Data. Data returned from the EEPROM read.

This register is used by software to cause the Ethernet controller to read individual words in the
EEPROM. To read a word, software writes the address to the Read Address field and simultaneously writes a 1b to the Start Read field. The Ethernet controller reads the word from the EEPROM
and places it in the Read Data field, setting the Read Done filed to 1b. Software can poll this
register, looking for a 1b in the Read Done filed, and then using the value in the Read Data field.
When this register is used to read a word from the EEPROM, that word is not written to any of
Ethernet controller’s internal registers even if it is normally a hardware accessed word.
Note:

If software has requested direct pin control of the EEPROM using the EEC register, an access
through the EERD register mechanism can stall until the EEC control has been released. Software
should ensure that EEC.EE_REQ = 0b and that EEC.EE_GNT = 0b as well before attempting to
use EERD to access the EEPROM.

Software Developer’s Manual

235

13.4.5

Flash Access1
FLA (0001Ch; R/W)
This register provides software direct access to the Flash memory. Software can control the Flash
device by successive writes to this register. Data and address information is clocked into the Flash
memory by software toggling the FL_SCK bit (0) of this register with FL_CE set to 1b. Data
output from the Flash memory is latched into bit three of this register via the internal 125 MHz
clock and is accessed by software via reads of this register.
Table 13-9. Flash Access – FLA
31

FL
ER

FL
BS

Field

29 - 6

Reserved

FL
GNT

FL
REQ

FL
SO

FL
SI

FL
CS

FL
SCK

Initial
Value

Bit(s)

Description

FL_SCK

Clock Input to the FLASH.
When FL_GNT is 1b, the FL_SCK out signal is mapped to this bit and
provides the serial clock input to the Flash device. Software clocks the
Flash memory via toggling this bit with successive writes.

FL_CE

Chip FL Input to the FLASH.
When FL_GNT is 1b, the FL_CS output signal is mapped to the chip
select of the device. Software enables the FLASH by writing a 0b to this
bit.

FL_SI

Data Input to the FLASH.
When FL_GNT is 1b, the FL_DI output signal is mapped directly to this
bit. Software provides data input to the FLASH via writes to this bit.

FL_SO

Data Output Bit from the FLASH.
The FL_SO input signal is mapped directly to this bit in the register and
contains the Flash memory serial data output. This bit is read only from
the software perspective — writes to this bit have no effect.

FL_REQ

Request FLASH Access.
The software must write a 1b to this bit to get direct Flash memory
access. It has access when FL_GNT is 1b. When the software
completes the access it must write a 0b.

FL_GNT

Grant FLASH Access.
When this bit is 1b, the software can access the Flash memory using
the FL_SCK, FL_CE, FL_SI, and FL_SO bits.

Reserved

29:6

Reserved.
Reads as 0b.

FL_BUSY

Flash Busy.
This bit is set to 1b while a write or an erase to the Flash memory is in
progress. While this bit is clear (read as 0b) software can access to
write a new byte to the Flash device.

FL_ER

Flash Erase Command.
This command will be sent to the Flash component only if bits 5:4 are
also set. This bit is automatically cleared and read as 0b.

Applicable to the 82541xx and 82547GI/EI only.

236

Software Developer’s Manual

13.4.6

Extended Device Control Register
CTRL_EXT (00018h, R/W)
This register and the Device Control register (CTRL) controls the major operational modes for the
Ethernet controller. CTRL_EXT provides extended control of the Ethernet controller functionality
over the Device Control register (CTRL).

Note:

See Table 13-12 and Table 13-13 for the 82544GC/EI.
Table 13-10. CTRL_EXT Register Bit Description
31 - 24

23 - 16 15

Reserved

Field

Reserved

GPI_EN

Extended Device Control Bits

Bit(s)

1:0

3:0

Initial
Value

Description

Reserved bits for the 82541xx and 82547GI/EI. Should be written
as 0b to ensure future compatibility.

General Purpose Interrupt Enables
These bits determine whether the upper three software definable
pins SDP[7:6] and SDP[4] are mapped to the ICR.GPI interrupt
bits. These mappings are enabled only when the SDP[7:6] and
SDP[4] pins are configured as inputs through
CTRL_EXT.SWDPIOHI. Refer to Table 13-11 for SDP to ICR.GPI
bit mapping.

GPI_EN

3:2

General Purpose Interrupt Enables for the 82541xx and 82547GI/
EI.
These bits determine whether the upper software definable pins
SDP[3:2] are mapped to the ICR.GPI interrupt bits. These
mappings are enabled only when the SDP[3:2] pins are configured
as inputs through CTRL_EXT.SWDPIOHI. Refer to Table 13-11 for
SDP to ICR.GPI bit mapping.

Reserved

Reserved. Formally used as SDP4 pin data value. Reads as 0b.

PHYINT

PHY Interrupt Value. When read, returns the current value of the
PHY internal interrupt status PHYINT.
Note: This is a reserved bit for the 82541xx and 82547GI/EI.

0b1

SDP6[2] Data Value. Used to read (write) value of softwarecontrollable IO pin SDP6[2]. If SDP6[2] is configured as an output
(SDP6[2]_IODIR = 1b), this bit controls the value driven on the pin
(initial value EEPROM-configurable). If SDP6[2] is configured as
an input, reads return the current value of the pin.

SDP6_DATA
SDP2_DATA
(82541xx and
82547GI/EI)

SDP7_DATA
SDP3_DATA
(82541xx and
82547GI/EI)

0b1

SDP7[3] Data Value. Used to read (write) value of softwarecontrollable IO pin SDP7[3]. If SDP7[3] is configured as an output
(SDP7[3]_IODIR = 1b), this bit controls the value driven on the pin
(initial value EEPROM-configurable). If SDP7[3] is configured as
an input, reads return the current value of the pin.

Reserved

9:8

01b

Reserved
Should be written as 01b to ensure future compatibility.

Software Developer’s Manual

237

Field

Bit(s)

Initial
Value

SDP6_IODIR
SDP2_IODIR
(82541xx and
82547GI/EI)

0b1

SDP6[2] Pin Directionality. Controls whether software-controllable
pin SDP6[2] is configured as an input or output (0b = input, 1b =
output). Initial value is EEPROM-configurable. This bit is not
affected by software or system reset, only by initial power-on or
direct software writes.

0b1

SDP7[2] Pin Directionality. Controls whether software-controllable
pin SDP7[2] is configured as an input or output (0b = input, 1b =
output). Initial value is EEPROM-configurable. This bit is not
affected by software or system reset, only by initial power-on or
direct software writes.

ASD Check
Initiate an Auto-Speed-Detection (ASD) sequence to sense the
frequency of the PHY receive clock. The results are reflected in
STATUS.ASDV. This bit is self-clearing.
This functionality is provided for diagnostic purposes, regardless of
whether the Auto Speed Detection feature is enabled. This bit is
applicable only for internal PHY mode of operation.

SDP7_IODIR
SDP2_IODIR
(82541xx and
82547GI/EI)

ASDCHK

238

Description

EE_RST

EEPROM Reset
When set, initiates a reset-like event to the EEPROM function. This
causes the EEPROM to be read as if a RST# assertion had
occurred. All device functions should be disabled prior to setting
this bit. This bit is self-clearing.

Reserved

0b1

Reserved. Should be set to 0b.

SPD_BYPS

Speed Select Bypass
When set to 1b, all speed detection mechanisms are bypassed,
and the Ethernet controller is immediately set to the speed
indicated by CTRL.SPEED. This can be used to override the
hardware clock switching circuitry and give full control to software.
SPD_BYPS differs from the CTRL.FRCSPD function in that
FRCSPD uses the internal clock switching circuitry rather than an
immediate forcing function of the speed settings, as does
SPD_BYPS.

Reserved

0b1

Reserved. Should be set to 0b.

RO-DIS

Relaxed Ordering Disabled
When set to 1b, the Ethernet controller does not request any
relaxed ordering transactions in PCI-X mode regardless of the
state of bit 1 in the PCI-X command register. When this bit is clear
and bit 1 of the PCI-X command register is set, the Ethernet
controller requests relaxed ordering transactions.
Note: This is a reserved bit for the 82540EP/EM, 82541xx, and
82547GI/EI. Set to 0b.

Reserved

20:18

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

Software Developer’s Manual

Field

VREG POWER
DOWN

Initial
Value

Bit(s)

Description

Voltage Regulator Power Down (82541xx and 82547GI/EI only)
0b = Normal operation.
1b = Voltage regulators power down.
This bit is initialized from the EEPROM.
Note: This is a reserved bit for all remaining Ethernet controllers.
Set to 0b.

LINK_MODE

23:22

Link Mode. This controls which interface is used to talk to the link.
00b = Direct copper (1000Base-T) interface (GMII/MII internal PHY
mode)
01b = Reserved
10b = Direct Fiber interface (using internal SerDes)
11b = external TBI interface
Note: These are reserved bits for the 82540EP/EM, 82541xx, and
82547GI/EI. Set to 00b

Reserved

31:24

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

These bits are read from the EEPROM

The Ethernet controller allows for up to two externally controlled interrupts. The upper two
software-definable pins, SDP[7:6] (SDP[3:2] for the 82541xx and 82547GI/EI), can be mapped
for use as GPI interrupt bits. These mappings are enabled by the SDPx_GPIEN bits only when
these signals are also configured as inputs via SDPx_IODIR. When configured to function as
external interrupt pins, a GPI interrupt is generated when the corresponding pin is sampled in an
active-high state.
The bit mappings are shown in Table 13-11 for clarity.

Table 13-11. GPI to SDP Bit Mappings
SDP pin to be used as
GPI

CTRL_EXT field settings
Directionality

Enable as GPI interrupt

SDP7_IODIR

SDP7_GPIEN

SDP2_IODIR1

SDP2_GPIEN1

SDP6_IODIR

SDP6_GPIEN

SDP3_IODIR1

SDP3_GPIEN1

Resulting ICR bit
(GPI)

82541xx and 82547GI/EI only

Software Developer’s Manual

239

Table 13-12. 82544GC/EI CTRL_EXT Register Bit Description
31

16 15
Reserved

Field

GPI_EN

SWDPINSHI

SWDPIOHI

ASDCHK

EE_RST

IPS

240

Initial
Value

Bit(s)

3:0

7:6, 4

11:10, 8

0
Extended Device Control Bits

Description

General Purpose Interrupt Enables
These bits determine whether the upper three software
definable pins SDP[7:6] and SDP[4] are mapped to the ICR.GPI
interrupt bits. These mappings are enabled only when the
SDP[7:6] and SDP[4] pins are configured as inputs through
CTRL_EXT.SWDPIOHI. Refer to Table 13-13 for SDP to
ICR.GPI bit mapping.

Software Defined Pins – high nibble
These three bits allow direct control of SDP[7:6] and SDP[4].
These pins can be either input pins or output pins as
determined by the SWDPIOHI bits. The initial direction of the
software defined pins is read out of the EEPROM.
Note: SDP[5] and its associated bit is not used. It should be
programmed to 0b for future compatibility.

Software Defined Pins Input or Output
These three bits control whether each of the high nibble
software defined pins SDP[7:6] and SDP[4] is used as an input
or an output.
0b = inputs; 1b = outputs.
This field is not affected by assertion of software reset
(CTRL.RST).
Configurable through EEPROM.
Note: SDP[5] and its associated bit is not used. It should be
programmed to 0b for future compatibility.

ASD Check
Initiate an Auto-Speed-Detection (ASD) sequence to sense the
frequency of the PHY receive clock. The results are reflected in
STATUS.ASDV. This bit is self-clearing.
This functionality is provided for diagnostic purposes,
regardless of whether the Auto Speed Detection feature is
enabled. This bit is applicable only for internal PHY mode of
operation.

EEPROM Reset
When set, initiates a “reset-like” event to the EEPROM function.
This causes the EEPROM to be read as if a RST# assertion had
occurred. All Ethernet controller functions should be disabled
prior to setting this bit. This bit is self-clearing.

Invert Power State Bit 0
When set to 1b, inverts the assertion polarity of the
PWR_STATE bit 0 output. When cleared to 0b, PWR_STATE is
logic high in normal operation.
Configurable through EEPROM.

Software Developer’s Manual

Field

Initial
Value

Bit(s)

Description

SPD_BYPS

Speed Select Bypass
When set to 1b, all speed detection mechanisms are bypassed,
and the Ethernet controller is immediately set to the speed
indicated by CTRL.SPEED. This might be used to override the
hardware clock switching circuitry and give full control to
software. SPD_BYPS differs from the CTRL.FRCSPD function
in that FRCSPD uses the internal clock switching circuitry rather
than an immediate forcing function of the speed settings, as
does SPD_BYPS.

IPS1

Invert Power State Bit 1
Inverts the polarity of bit 1 of the PWR_STATE signal when set
to 1b.
Configurable through the EEPROM.

RO-DIS

Relaxed Ordering Disabled
When set to 1b, the Ethernet controller does not request any
relaxed ordering transactions in PCI-X mode regardless of the
state of bit 1 in the PCI-X command register. When this bit is
clear and bit 1 of the PCI-X command register is set, the
Ethernet controller requests relaxed ordering transactions as
described.

Reserved

31:18

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

Table 13-13. 82544GC/EI GPI to SDP Bit Mapping

B_SDP

CTRL_EXT
(SWDPINHI)

CTRL_EXT
(GPI_EN)

ICR
(GPI)

Reserved

Software Developer’s Manual

241

13.4.7

MDI Control Register
MDIC (00020h; R/W)
Software uses this register to read or write Management Data Interface (MDI) registers in the
internal PHY.
To read a location in the PHY, first perform an MDI write cycle with the following bit settings:

•
•
•
•
•

Ready = 0b
Interrupt Enable programmed to 1b or 0b
Opcode = 10b (read)
PHYADD = PHY address from the MDI register
REGADD = Register address within the PHY to be read

When the serial data transfer from PHY to MAC is complete, the Ethernet controller issues an
interrupt if the MDI interrupt is enabled. The Ethernet controller also sets the Ready bit. This
indication tells the system that the read data is available from the 16-bit data field in the MDI
Control register. Perform a second read operation to the register at this time to recover the data.
To write a location in the PHY, perform an MDI write cycle with the following bit settings:

•
•
•
•
•
•

Ready = 0b
Interrupt Enable programmed to 1b or 0b
Opcode = 01b (write)
PHYADD = PHY address from the MDI register
REGADD = Register address within the PHY to be written
Data = Specific data for PHY operation

If enabled, the Ethernet controller issues an interrupt when the write completes. The Ethernet
controller also sets the Ready bit, denoting that a subsequent operation can be carried out.
Note:

242

The internal PHY register bit descriptions follow Table 13-14.

Software Developer’s Manual

Table 13-14. MDI Control Register Bit Description
31

RSV

Field

27 26 25 21 20 16 15
OP

PHY

Initial
Value

Bit(s)

REG

DATA

Description

DATA

15:0

Data
In a Write command, software places the data bits and the
Ethernet controller shifts them out to the PHY. In a Read
command, the Ethernet controller reads these bits serially from
the PHY and software can read them from this location.

REGADD

20:16

PHY Register Address: Reg. 0, 1, 2, ...31

PHYADD

25:21

PHY Address
The Internal PHY’s MDI address for each MAC is 0001b

27:26

Opcode
01b = MDI Write
10b = MDI Read
All other values are reserved.

Ready Bit
Set to 1b by the Ethernet controller at the end of the MDI
transaction (for example, indication of a Read or Write
completion). It should be reset to 0b by software at the same
time the command is written.

Interrupt Enable
When set to 1b by software, it causes an Interrupt to be
asserted to indicate the end of an MDI cycle.

Error
This bit is set to 1b by hardware when it fails to complete an
MDI read. Software should make sure this bit is clear (0b)
before issuing an MDI read or write command.

Reserved

Reserved
Reads as 0b.

Software Developer’s Manual

243

13.4.7.1

PHY Registers
This document uses a special nomenclature to define the read/write mode of individual bits in each
register. See Table 13-15.
For all binary equations appearing in the register map, the symbol “|” is equivalent to a binary OR
operation.
Table 13-15. PHY Register Bit Mode Definitions
Register Mode

244

Description

Latched High. Event is latched and erased when read.

Latched Low. Event is latched and erased when read. For
example, Link Loss is latched when the PHY Control
Register bit 2 = 0b. After read, if the link is good, the PHY
Control Register bit 2 is set to 1b.

Read Only.

R/W

Read and Write.

Self-Clear. The bit is set, automatically executed, and then
reset to normal operation.

Clear after Read. For example, 1000BASE-T Status
Register bits 7:0 (Idle Error Counter).

Update

Value written to the register bit does not take effect until
software PHY reset is executed.

Software Developer’s Manual

13.4.7.1.1

PHY Control Register
PCTRL (00d; R/W)

Table 13-16. PHY Control Register Bit Description
Field
Reserved

Speed Selection
(MSB)

Collision Test

Duplex Mode

Bit(s)

Description

HW Rst

SW Rst

5:0

These bits are reserved and should be
set to 000000b.

RO
RW1

Always
000000b

Speed Selection is determined by bits 6
(MSB) and 13 (LSB) as follows.
11b = Reserved
10b = 1000 Mbps
01b = 100 Mbps
00b = 10 Mbps
A write to these bits do not take effect
until a software reset is asserted,
Restart Auto-Negotiation is asserted, or
Power Down transitions from power
down to normal operation.
82544GC/EI only:
The Speed Selection bits take on the
values set by external pins ANEG[3:0]
on hardware reset only.
Bit 6: ANEG[3] | ANEG[2] | (MODE[3:0]
is one of xx01b, 1x00b, 001xb, 0111b).
Bit 13: (ANEG[3:1] = 001b) and
(MODE[3:0] is not any of xx01b, 1x00b,
001xb, 0111b).

R/W

1b = Enable COL signal test.
0b = Disable COL signal test.

R/W

1b = Full Duplex.
0b = Half Duplex.
82544GC/EI only:
The Duplex bit takes on the value set by
external pins ANEG[3:2, 0] on hardware
reset only.
Bit 8: ANEG[3:2,0] = 001 | ANEG[3].
A write to this bit does not take effect
until a software reset is asserted,
Restart Auto-Negotiation is asserted, or
Power Down transitions from power
down to normal power.

R/W

R/W,
SC

Self
Clear

R/W

Restart AutoNegotiation

1b = Restart Auto-Negotiation Process.
0b = Normal operation.
Auto-Negotiation automatically restarts
after hardware or software reset
regardless of whether or not the restart
bit is set.

Isolate

1b = Isolate.
0b = Normal operation.

Software Developer’s Manual

Mode

Update

ANEG[3:2]
MODE[3:0]

1b
Update
ANEG[3:2,
0]

245

Table 13-16. PHY Control Register Bit Description
Field

Power Down

246

Bit(s)

Description

Mode

1b = Power down.
0b = Normal operation.
Power down shuts down the Ethernet
controller except for the MAC interface if
the MAC interface power down bit is set
to 1b. If it equals 0b, then the MAC
interface also shuts down. For the
82544GC/EI, power down has no effect
on the 125CLK output if the Disable
125CLK bit is set to 0b.
NOTE: Setting this bit to 1b will prevent
wakeup by detecting circuitry
on the CAT5 cable. To enable
wakeup, this bit must be written
back to 0b.
82544GC/EI only:
If bit 12 is set to 0b and speed is
manually forced to 1000 Mb/s in bits 13
and 6, then Auto-Negotiation is still
enabled and only 1000BASE-T full
duplex is advertised if bit 8 is set to 1b.
1000BASE-T half duplex is advertised if
bit 8 is cleared (0b). Duplex settings in
other registers are ignored. AutoNegotiation is required by IEEE for
proper operation in 1000BASE-T.

R/W

HW Rst

SW Rst

Software Developer’s Manual

Table 13-16. PHY Control Register Bit Description
Field

Auto-Negotiation
Enable

Bit(s)

Description
1b = Enable Auto-Negotiation Process.
0b = Disable Auto-Negotiation Process.
A write to this bit does not take effect
until a software reset is asserted,
Restart Auto-Negotiation is asserted, or
Power Down transitions from power
down to normal operation.
When the port is switched from power
down to normal operation, software
reset and restart Auto-Negotiation are
performed even if bits Reset and Restart
Auto-Negotiation are not set by the
programmer.
If bit 12 is set to 0b and speed is
manually forced to 1000 Mbps in bits 13
and 6, then Auto-Negotiation is still
enabled and only 1000BASE-T full
duplex is advertised if bit 8 is set to 1b.
1000BASE-T half duplex is advertised if
bit 8 is cleared (0b). Duplex settings in
other registers are ignored. AutoNegotiation is required by IEEE for
proper operation in 1000BASE-T.
82544GC/EI only:
Auto-Negotiation enable takes on the
value set by external pins ANEG[3:0] on
hardware reset only.
Bit 12: ANEG[3:2] = 11b.
If MODE[3:0] equals 001xb or 0111b,
where x equals either 0b or 1b, then the
ANE bit determines whether
1000BASE-X Auto-Negotiation is on or
off. Otherwise ANE determines whether
10/100/1000BASE-T Auto-Negotiation
is on or off.

Mode

HW Rst

SW Rst

R/W

Update

R/W

ANEG[3:2]
ENA_XC

Update

R/W

0b
ANEG[3:1]2
MODE[3:0]

Update

Speed Selection
(LSB)

See Speed Selection (MSB), bit 6.

Loopback

1b = Enable loopback.
0b = Disable loopback.

R/W

Reset

1b = PHY reset.
0b = Normal operation.

R/W,
SC

Self
Clear

1.
2.

82541xx and 82547GI/EI only.
82544GC/EI only.

Software Developer’s Manual

247

13.4.7.1.2

PHY Status Register
PSTATUS (01d; R)

Table 13-17. PHY Status Register Bit Description
Field

Description

Mode

HW Rst

SW Rst

Extended Capability

1b = Extended register capabilities.

Always 1b

Jabber Detect

1b = Jabber condition detected.
0b = Jabber condition not detected.

RO,L
H

Link Status

1b = Link is up.
0b = Link is down.
This register indicates whether link was
lost after the last read. For the current
link status, either read this register
back-to-back or read the Link Real Time
bit 17 in the PHY Specific Status
Register.

RO,
LL

Auto-Negotiation
Ability

1b = PHY able to perform AutoNegotiation.

Always 1b

Remote Fault

1b = Remote fault condition detected.
0b = Remote fault condition not
detected.

RO,L
H

Auto-Negotiation
Complete

1b = Auto-Negotiation process
complete.
0b = Auto-Negotiation process not
complete.

MF Preamble
Suppression

1b = PHY accepts management frames
with preamble suppressed.

Always 1b
(0b for the 82541xx
and 82547GI/EI)

Reserved

Reserved. Should be set to 0b

Always 0b

Extended Status

1b = Extended status information in the
Extended PHY Status Register (15d).

Always 1b

100BASE-T2 Half
Duplex

0b = PHY not able to perform half
duplex 100BASE-T2.

Always 0b

100BASE-T2 Full
Duplex

0b = PHY not able to perform full duplex
100BASE-T2.

Always 0b

1b = PHY able to perform half duplex
10BASE-T.
0b = PHY not able to perform half
duplex 10BASE-T.
82544GC/EI only:
Bit 14 = Bit 13 = Bit 12 = Bit 11 =
(MODE[3:0] is not any of xx01b, 1x00b,
001xb, 0111b).

MODE[3:0]

10 Mb/s Half Duplex

248

Bit(s)

Software Developer’s Manual

Table 13-17. PHY Status Register Bit Description
Field

10 Mb/s Full Duplex

100BASE-X Half
Duplex

100BASE-X Full
Duplex

100BASE-T4

Software Developer’s Manual

Bit(s)

Description
1b = PHY able to perform full duplex
10BASE-T.
0b = PHY not able to perform full duplex
10BASE-T.
82544GC/EI only:
Bit 14 = Bit 13 = Bit 12 = Bit 11 =
(MODE[3:0] is not any of xx01b, 1x00b,
001xb, 0111b).
1b = PHY able to perform half duplex
100BASE-X.
0b = PHY able to perform half duplex
100BASE-X.
82544GC/EI only:
Bit 14 = Bit 13 = Bit 12 = Bit 11 =
(MODE[3:0] is not any of xx01b, 1x00b,
001xb, 0111b).
1b = PHY able to perform full duplex
100BASE-X.
0b = PHY not able to perform full duplex
100BASE-X.
82544GC/EI only:
Bit 14 = Bit 13 = Bit 12 = Bit 11 =
(MODE[3:0] is not any of xx01b, 1x00b,
001xb, 0111b).

Mode

HW Rst

MODE[3:0]

0b = PHY not able to perform 100BASERO
T4.

SW Rst

Always 0b

249

13.4.7.1.3

PHY Identifier Register (LSB)
PID (02d; R)

Table 13-18. PHY Identifier Bit Description
Field

Bit(s)

Organizationally
Unique Identifier Bit
18:31

1.
2.

13.4.7.1.4

Description

Mode

0000_0001_0100_0001b
OUI is 005043h.
15:0

HW Rst

SW Rst

Always 0141h

The PHY identifier composed of bits 3
through 18 of the OUI (Organizationally
Unique Identifier)2

RO
Always 02A8h2

PHY ID number for the 82541xx and 82547GI/EI only.
82541xx and 82547GI/EI only.

Extended PHY Identifier Register (MSB)
EPID (03d; R)

Table 13-19. Extended PHY Identifier Bit Description
Field

Bit(s)

Description

Mode

3:0

4 bits containing the manufacturer’s
revision number.1

Organizationally
Unique Identifier Bit
19:24
(PHY ID Number for
the 82541xx and
82547GI/EI)
1.

250

9:4

15:10

82544GC/EI only:
000011b = 10/100/1000 Copper PHY.

00h1

Always 000010b

000010b = 10/100/1000 Copper PHY.

Model Number

SW Rst

Always static

Contains the current revision number.
Revision Number

HW Rst

82544GC/EI only:
Always 000011b

6 bits containing the manufacturer’s part
number1

Always 111000b1

000011b.

Always 000011b

The PHY identifier composed of bits 19
through 24 of the OUI1

0h1

82541xx and 82547GI/EI only.

Software Developer’s Manual

13.4.7.1.5

Auto-Negotiation Advertisement Register
ANA (04d; R/W)

Table 13-20. Auto-Negotiation Advertisement Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

00001b = 802.3
For the 82541xx and 82547GI/EI:
Other combinations are reserved.
Selector Field

4:0

Unspecified or reserved combinations
should not be transmitted.

Always 00001b

Note: Setting this field to a value other
than 00001b can cause auto
negotiation to fail.
1b = Advertise.

0b = Not advertised.
Values programmed in the AutoNegotiation Advertisement Register
have no effect unless AutoNegotiation is restarted (PHY Control
Register) or link goes down.

10BASE-TX Half
Duplex
5
10Base-T (82541xx
and 82547GI/EI)

This bit can be overridden by the PHY
Control Register.

R/W

Retain

82544GC/EI only:

ANEG[3:1]

Bit 5: (ANEG[3:0] = 0000b |
ANEG[3:2] = 11b) and (MODE[3:0] is
not any of xx01b, 1x00b, 001xb,
0111b).

MODE[3:0]

82541xx and 82547GI/EI only:

1b = DTE is 10BASE-T capable.
0b = DTE is not 10BASE-T capable.
1b = Advertise.

0b = Not advertised.
Values programmed in the AutoNegotiation advertisement register
have no effect unless AutoNegotiation is restarted (PHY Control
Register) or link goes down.
10BASE-TX Full
Duplex

This bit can be overridden by the PHY
Control Register.
6

10Base-T Full Duplex
(82541xx and
82547GI/EI)

R/W

Retain

82544GC/EI only:
Bit 5: (ANEG[3:0] = 0001b |
ANEG[3:2] = 11b) and (MODE[3:0] is
not any of xx01b, 1x00b, 001xb,
0111b).
82541xx and 82547GI/EI only:

ANEG[3:0]
MODE[3:0]
1b

1b = DTE is 10BASE-T full duplex
capable.
0b = DTE is not 10BASE-T full duplex
capable.

Software Developer’s Manual

251

Table 13-20. Auto-Negotiation Advertisement Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

1b = Advertise.
1b

0b = Not advertised.
Values programmed in the AutoNegotiation advertisement register
have no effect unless AutoNegotiation is restarted (PHY Control
Register) or link goes down.

100BASE-TX Half
Duplex

This bit can be overridden by the PHY
Control Register.
7

100Base-TX
(82541xx and
82547GI/EI)

R/W

Retain

82544GC/EI only:
Bit 5: (ANEG[3:0] = 0010b |
ANEG[3:2] = 11b) and (MODE[3:0] is
not any of xx01b, 1x00b, 001xb,
0111b).

ANEG[3:1]

82541xx and 82547GI/EI only1:

MODE[3:0]

1b = DTE is 100BASE-TX capable.
0b = DTE is not 100BASE-TX
capable.
1b = Advertise.

0b = Not advertised.
Values programmed in the AutoNegotiation advertisement register
have no effect unless AutoNegotiation is restarted (PHY Control
Register) or link goes down.
This bit can be overridden by the PHY
Control Register.
100BASE-TX Full
Duplex

R/W

82544GC/EI only:

Retain
ANEG[3:0]

Bit 8: (ANEG[3:0] = 0011b |
ANEG[3:2] = 11b) and (MODE[3:0] is
not any of xx01b, 1x00b, 001xb,
0111b).

MODE[3:0]

82541xx and 82547GI/EI only:1

1b = DTE is 100BASE-TX full duplex
capable.
0b = DTE is not 100BASE-TX full
duplex capable.
100BASE-T4

0b = Not capable of 100BASE-T4.

Always 0b

1b = MAC PAUSE implemented.
0b = MAC PAUSE not implemented.

PAUSE

Values programmed in the AutoNegotiation advertisement register
have no effect unless AutoNegotiation is restarted (PHY Control
Register) or link goes down.
82541xx and 82547GI/EI only:
Advertise to Partner that Pause
operation (as defined in 802.3x) is
desired.

252

R/W

1b for the
(82541xx
and
82547GI/
EI)

Retain

Software Developer’s Manual

Table 13-20. Auto-Negotiation Advertisement Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

1b = Asymmetric Pause.
0b = No asymmetric Pause.
Asymmetric Pause
ASM_DIR for the
(82541xx and
82547GI/EI)

Values programmed in the AutoNegotiation advertisement register
have no effect unless AutoNegotiation is restarted (PHY Control
Register) or link goes down.

R/W

82541xx and 82547GI/EI only:
Advertise Asymmetric Pause direction
bit. This bit is used in conjunction with
PAUSE.

1b for the
(82541xx
and
82547GI/
EI)

Retain

This bit is reserved and equals 0b.

Reserved

Values programmed in the AutoNegotiation advertisement register
have no effect unless AutoNegotiation is restarted (PHY Control
Register) or link goes down.

R/W

Retain

R/W

Retain

Always 0b

R/W

Reserved bit is R/W to allow for
forward compatibility with future IEEE
standards.
1b = Set Remote Fault bit.
0b = Do not set Remote Fault bit.
Remote Fault

Reserved

Values programmed in the AutoNegotiation advertisement register
have no effect unless AutoNegotiation is restarted (PHY Control
Register) or link goes down.
Reserved. Should be set to 0b.
1b = Advertise.
0b = Not advertised.
Values programmed in the AutoNegotiation advertisement register
have no effect unless AutoNegotiation is restarted (PHY Control
Register) or link goes down.

If 1000BASE-T is advertised then the
required next pages are automatically
transmitted. The Next Page bit should
equal 0 if no additional next pages are
needed.

Retain

82541xx and 82547GI/EI only:
1b = Manual control of Next Page
(Software).
0b = Ethernet controller control of Next
Page (Auto).
1.

For the 82541xx and 82547GI/EI, if EEPROM bit ADV10LU is asserted, then the default is set to 0b. Otherwise, the default
is 1b.

Software Developer’s Manual

253

82544GC/EI Only:
Table 13-21. Auto-Negotiation Advertisement Register Bit Description
(MODE[3:0] is one of 001xb, 0111b)
Field

Description

Mode

HW Rst

SW Rst

4:0

Values programmed in this register have
no effect unless Auto-Negotiation is
restarted (PHY Control Register) or the
link goes down.
Reserved bit is R/W to allow forward
compatibility with future IEEE standards.

R/W

00000b

Retain

1b = Advertise
0b = Not advertised
Values programmed in this register have
no effect unless Auto-Negotiation is
restarted (PHY Control Register) or the
link goes down.
Bit 5: ANEG[3] = 1b

R/W

ANEG[3]

Retain

1000BASE-X Half
Duplex

1b = Advertise
0b= Not advertised
Values programmed in this register have
no effect unless Auto-Negotiation is
restarted (PHY Control Register) or the
link goes down.
Bit 6: ANEG[2] = 1b

R/W

ANEG[2]

Retain

Pause

Values programmed in this register have
no effect unless Auto-Negotiation is
restarted (PHY Control Register) or the
link goes down.

R/W

Retain

Asymmetric Pause

Values programmed in this register have
no effect unless Auto-Negotiation is
restarted (PHY Control Register) or the
link goes down.

R/W

000b

Retain

11:9

R/W

Retain

Values programmed in this register have
no effect unless Auto-Negotiation is
restarted (PHY Control Register) or the
link goes down.
Ethernet controller has no ability to
detect remote fault.

R/W

Retain

RF2

Values programmed in this register have
no effect unless Auto-Negotiation is
restarted (PHY Control Register) or the
link goes down.
Ethernet controller has no ability to
detect remote fault.

R/W

Retain

Reserved

Reserved. Should be set to 0b.

Always 0b

Reserved

1000BASE-X Full
Duplex

Reserved

RF1

254

Bit(s)

Software Developer’s Manual

Table 13-21. Auto-Negotiation Advertisement Register Bit Description
(MODE[3:0] is one of 001xb, 0111b)
Field

Bit(s)

13.4.7.1.6

Description
0b = Not advertised
Values programmed in this register have
no effect unless Auto-Negotiation is
restarted (PHY Control Register) or the
link goes down.
Next Page is not supported in
1000BASE-X mode.

Mode

HW Rst

SW Rst

Always 0b

Link Partner Ability Register (Base Page)
LPA (05d; R)

Table 13-22. Link Partner Ability Register (Base Page) Bit
Description1
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

Selector Field

4:0

Received Code Word Bit 4:0.

00000b

Technology Ability
Field

12:5

Received Code Word Bit 12:5.

00h

Remote Fault

Received Code Word Bit 13.

Acknowledge

Received Code Word Bit 14.

Received Code Word Bit 15.

(MODE[3:0] is one of xx00b, 1x01b, 101xb, 1111b) for the 82544GC/EI only.

Software Developer’s Manual

255

82544GC/EI Only:
Table 13-23. Link Partner Ability Register (Base Page) Bit
Description1
Field

Description

Mode

HW Rst

SW Rst

Reserved

4:0

Reserved. Should be set to 00000b.

00000b

10BASE-TX Half
Duplex

1b = 10 Base-TX half duplex is
available.
0b = 10 Base-TX half duplex is not
available.

10BASE-TX Full
Duplex

1b = 10 Base-TX full duplex is available.
0b = 10 Base-TX full duplex is not
available.

100BASE-TX Half
Duplex

1b = 100 Base-TX half duplex is
available.
0b = 100 Base-TX half duplex is not
available.

100BASE-TX Full
Duplex

1b = 100 Base-TX full duplex is
available.
0b = 100 Base-TX full duplex is not
available.

100BASE-T4

0b = Not capable of 100BASE-T4.

Pause

1b = Pause operation is available.
0b = Pause operation is not available.

Asymmetric Pause

1b = Asymmetric Pause operation is
available.
0b = Asymmetric Pause operation is not
available.

Reserved

Reserved. Should be set to 0b.

Remote Fault

Indicates a remote fault.

Reserved

Reserved. Should be set to 0b.

1b = Link partner is Next Pagable.
0b = Link partner is not Next Pagable.

256

Bit(s)

Always 0b

(MODE[3:0] is one of 001xb, 0111b).

Software Developer’s Manual

82541xx and 82547GI/EI Only:
Table 13-24. PHY Link Page Ability Bit Description1
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

Selector Field [4:0]

4:0

<00001> = IEEE 802.3
Other combinations are reserved.
Unspecified or reserved combinations
shall not be transmitted.
If field does not match PHY Register 4,
bits 4:0, the AN process does not
complete and no HCD is selected.

10BASE-T

1b = Link Partner is 10BASE-T capable.
0b = Link Partner is not 10BASE-T
capable.

N/A

10BASE-T Full
Duplex

1b = Link Partner is 10BASE-T full
duplex capable.
0b = Link Partner is not 10BASE-T full
duplex capable.

N/A

1b = Link Partner is 100BASE-TX
capable.
0b = Link Partner is not 100BASE-TX
capable.

N/A

100BASE-TX Full
Duplex

1b = Link Partner is 100BASE-TX full
duplex capable.
0b = Link Partner is not 100BASE-TX
full duplex capable.

N/A

100BASE-T4

1b = Link Partner is 100BASE-T4
capable.
0b = Link Partner is not 100BASE-T4
capable.

N/A

LP Pause

Link Partner uses Pause Operation as
defined in 802.3x.

N/A

LP ASM_DIR

Asymmetric Pause Direction Bit
1b = Link Partner is capable of
asymmetric pause.
0b = Link Partner is not capable of
asymmetric pause.

N/A

Reserved

Reserved. Should be set to 0b.

Remote Fault

1b = Remote fault.
0b = No remote fault.

N/A

Acknowledge

1b = Link Partner has received Link
Code Word from the PHY.
0b = Link Partner has not received Link
Code Word from the PHY.

N/A

1b = Link Partner has ability to send
multiple pages.
0b = Link Partner has no ability to send
multiple pages.

N/A

100BASE-TX

N/A

PHY register 8d stores the Auto-Negotiation Link Partner Received Next Pages. PHY register 5d is not used to store Next
Pages. It contains the information from the last Base Page correctly received.

Software Developer’s Manual

257

13.4.7.1.7

Auto-Negotiation Expansion Register
ANE (06d; R)

Table 13-25. Auto-Negotiation Expansion Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

1b = Link Partner is Auto-Negotiation
able.
0b = Link Partner is not AutoNegotiation able.

Page Received

1b = A New Page has been received.
0b = A New Page has not been
received.
82541xx and 82547GI/EI only:
If PHY register 16, bit 1 (Alternate NP
Feature) is set, the Page Received bit
also clears when mr_page_rx = false or
transmit_disable = true.

RO/
LH1

Local Next Page Able

1b = Local Device is Next Pageable.

1b = Link Partner is Next Page able.
0b = Link Partner is not Next Page able.
82544GC/EI only:
Bit 2 = 1 if MODE[3:0] is not 001xb or
0111b.

Link Partner AutoNegotiation Able

Link Partner Next
Page Able

1b (82541xx and
82547GI/EI)

MODE[3:0]

MODE
[3:0]

1b = A fault has been detected via the
Parallel Detection function.
0b = A fault has not been detected via
the Parallel Detection function.

RO/
LH1

Base Page

82541xx and 82547GI/EI only:
This bit indicates the status of the AutoNegotiation variable, base page. If flags
synchronization with the AutoNegotiation state diagram enabling
detection of interrupted links. This bit is
only used if PHY register 16, bit 1
(Alternate NP Feature) is set.
1b = base_page = true.
0b = base_page = false.
Note: This is a reserved bit for all
remaining Ethernet controllers.

RO/
LH1

Reserved

15:6

0000000000b

Always 000h

Parallel Detection
Fault

82541xx and 82547GI/EI only.

NOTE: The ANE Register is not valid until the Auto-Negotiation complete bit in the PHY Status Register
indicates completion of the Auto-Negotiation process.

258

Software Developer’s Manual

13.4.7.1.8

Next Page Transmit Register
NPT (07d; R/W)

Table 13-26. Next Page Transmit Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

10:0

Transmit Code Word Bit 10:0.
82541xx and 82547GI/EI only:
11-bit message code field.

R/W

001h

Transmit Code Word Bit 11.
82541xx and 82547GI/EI only:
1b = Previous value of the transmitted
Link Code Word = logical 0.
0b = Previous value of the transmitted
Link Code Word = logical 1.

Acknowledge 2

Transmit Code Word Bit 12.
82541xx and 82547GI/EI only:
1b = Complies with message.
0b = Cannot comply with message.

R/W

Message Page

Transmit Code Word Bit 13.
82541xx and 82547GI/EI only:
1b = Message page.
0b = Unformatted page.

R/W

1b
0b1

Reserved

Transmit Code Word Bit 14.
82541xx and 82547GI/EI only:
Always read as 0b. Write to 0b for
normal operation.

Transmit Code Word Bit 15.
A write to the NPT Register implicitly
sets a variable in the Auto-Negotiation
state machine.
82541xx and 82547GI/EI only:
1b = Additional next pages follow.
0b = Last page.

R/W

Message/
Unformatted Field

Toggle

82541xx and 82547GI/EI only.

Software Developer’s Manual

259

13.4.7.1.9

Link Partner Next Page Register
LPN (08d; R)

Table 13-27. Link Partner Next Page Register Bit Description
Bit(s)

10:0

260

Field

Description

Mode

HW Rst

SW Rst

Message/
Unformatted Field

Received Code Word Bit 10:0.
82541xx and 82547GI/EI only:
11-bit message code field.

000h

Toggle

Received Code Word Bit 11.
82541xx and 82547GI/EI only:
1b = Previous value of the transmitted
Link Code Word = a logic 0.
0b = Previous value of the transmitted
Link Code Word = a logic 1.

Acknowledge 2

Received Code Word Bit 12.
82541xx and 82547GI/EI only:
1b = Link Partner complies with the
message.
0b = Link Partner cannot comply with
the message.

Message Page

Received Code Word Bit 13.
82541xx and 82547GI/EI only:
1b = Page sent by the Link Partner is a
Message Page.
0b = Page sent by the Link Partner is an
Unformatted Page.

Acknowledge

Received Code Word Bit 14.
82541xx and 82547GI/EI only:
1b = Link Partner has received Link
Code Word from the PHY.
0b = Link Partner has not received Link
Code Word from the PHY.

Received Code Word Bit 15.
82541xx and 82547GI/EI only:
1b = Link Partner has additional next
pages to send.
0b = Link Partner has no additional next
pages to send.

Software Developer’s Manual

13.4.7.1.10

1000BASE-T Control Register
GCON (09d; R/W)

Table 13-28. 1000BASE-T Control Register Bit Description
Bit(s)
7:0

Field

Description

Reserved

Reserved. Should be set to 00000000b.

1000BASE-T Half
Duplex

1b = Advertise.
0b = Not advertised.
82544GC/EI only:
Bit 8 = ANEG[2].
82541xx and 82547GI/EI only:
1b = DTE is 1000BASE-T capable.
0b = DTE is not 1000BASE-T capable.
This bit is used by Smart Negotiation.

1000BASE-T Full
Duplex

1b = Advertise.
0b = Not advertised.
82544GC/EI only:
Bit 9 = ANEG[3].
82541xx and 82547GI/EI only:
1b = DTE is 1000BASE-T full duplex
capable.
0b = DTE is not 1000BASE-T full duplex
capable. This bit is used by Smart
Negotiation (see Note 3).

Port Type

1b = Prefer multi-port device
(MASTER).
0b = Prefer single port device (SLAVE)1.
82544GC/EI only:
Bit 10 =!ANEG[0].
Bit 10 is ignored if bit 12 = 1b.

MASTER/SLAVE
Configuration Value

1b = Manual configure as MASTER3.
0b = Manual configure as SLAVE3.
82544GC/EI only:
Bit 11 =!ANEG[0].
Bit 11 is ignored if bit 12 = 0b.

Mode

HW Rst

SW Rst

R/W

0b
ANEG[2]

Retain

R/W

1b
ANEG[3]

Retain

R/W

1b2
ANEG[0]

Retain

R/W

1b2
ANEG[0]

Retain

R/W

1b2
ANEG[1]

Retain

1b = Manual MASTER/SLAVE
MASTER/SLAVE
12

Manual Configuration
Enable

configuration.
0b = Automatic MASTER/SLAVE
configuration.
82544GC/EI only:
Bit 12 =!ANEG[1].

Software Developer’s Manual

261

Table 13-28. 1000BASE-T Control Register Bit Description
Bit(s)

Field

Description

Mode

HW Rst

SW Rst

000b = Normal Mode.
001b = Test Mode 1 - Transmit
Waveform Test.
15:13

010b = Test Mode 2 - Transmit Jitter
Test (MASTER mode).

Test mode

R/W

011b = Test Mode 3 - Transmit Jitter
Test (SLAVE mode).

000b

100b = Test Mode 4 - Transmit
Distortion Test.
101b, 110b, 111b = Reserved.
1.
2.
3.

For the 82541xx and 82547GI/EI, only when PHY register 9, bit 12 is set to a logical 0.
0b for the 82541xx and 82547GI/EI only
For the 82541xx and 82547GI/EI, only when PHY register 9, bit 12 is set to a logical 1.

NOTES:
1. Values programmed in bits 12:8 of the 1000BASE-T Control Register have no effect unless Auto-Negotiation
is restarted (PHY Control Register, bit 9) or the link goes down. These bits can also be overridden by the PHY
Control Register.
2. The symbol “!” is equivalent to logical “not.”
3. For the 82541xx and 82547GI/EI, the default for bit 9 is affected by configuration bits in the EEPROM. If
EEPROM ANI1000DIS is asserted, then the default is set to 0b. If EEPROM ADV10LU is asserted, then the
default is set to 0b.

13.4.7.1.11

1000BASE-T Status Register
GSTATUS (10d; R)

Table 14-29. 1000BASE-T Status Register Bit Description
Field

Description

Mode

HW Rst

SW Rst

Idle Error Count

7:0

Idle Error Counter.
The counter stops at 1111b 1111b and
does not roll over.

RO,
SC

0000b
0000b

Reserved

9:8

Reserved. Should be set to 00b

00b

1b = Link Partner is capable of
1000BASE-T half duplex.
0b = Link Partner is not capable of
1000BASE-T half duplex.
Values in bits 11:10 are not valid until
the ANE Register Page Received bit
equals 1b.

Link Partner
1000BASE-T
Full Duplex Capability

1b = Link Partner is capable of
1000BASE-T full duplex.
0b = Link Partner is not capable of
1000BASE-T full duplex.
Values in bits 11:10 are not valid until
the ANE Register Page Received bit
equals 1b.

Remote Receiver
Status

1b = Remote Receiver OK.
0 b = Remote Receiver Not OK.

Link Partner
1000BASE-T
Half Duplex
Capability

262

Bit(s)

Software Developer’s Manual

Table 14-29. 1000BASE-T Status Register Bit Description
Field

Bit(s)

Local Receiver Status

MASTER/SLAVE
Configuration
Resolution

MASTER/SLAVE
Configuration Fault

13.4.7.1.12

Description

Mode

HW Rst

SW Rst

1b = Local Receiver OK.
0b = Local Receiver Not OK.

1b = Local PHY configuration resolved
to MASTER.
0b = Local PHY configuration resolved
to SLAVE.

1b = MASTER/SLAVE configuration
fault detected.
0b = No MASTER/SLAVE configuration
fault detected.

RO,
LH,
SC

Extended PHY Status Register
EPSTATUS (15d; R)

Table 13-30. Extended PHY Status Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

11:0

Reserved. Should be set to
000000000000b.

000h

1000BASE-T Half
Duplex

1b = 1000BASE-T half duplex capable.
0b = not 1000BASE-T half duplex
capable.
Note: 1000Mb-Half duplex is NOT
supported.

1b1

1000BASE-T Full
Duplex

1b = 1000BASE-T full duplex capable.
0b = Not 1000BASE-T full duplex
capable.

1b1

1000BASE-X Half
Duplex

1b =1000BASE-X half duplex capable.
0b = Not 1000BASE-X half duplex
capable.
Note: 1000Mb-Half Duplex is NOT
supported.

0b1

1000BASE-X Full
Duplex

1b =1000BASE-X full duplex capable.
0b = Not 1000BASE-X full duplex
capable.

0b1

Reserved

MODE[3:0] 82544GC/EI only.

NOTES:
1. 1000BASE-X Half Duplex only applicable to the 82544GC/EI.
2. Bit 12 = bit 13 = 1b if MODE[3:0] does not = 001xb or 0111b.
3. Bit 14 = bit 15 = 1b if MODE[3:0] = 001xb or 0111b.

Software Developer’s Manual

263

13.4.7.1.13

PHY Specific Control Register
PSCON (16d; R/W)

Table 13-31. PHY Specific Control Register Bit Description
Description
Field

Bit(s)

Mode
1000BASE-T

HW Rst

SW Rst

10/100BASE-T

1b = Disable jabber function.
Disable Jabber

0b = Enable jabber function.

R/W

Retain

R/W

Retain

R/W

Retain

R/W

Update

R/W

DIS_
125CLK1

Update

Jabber has effect only in 10BASE-T half
duplex mode.
1b = Polarity Reversal Disabled.
Polarity Reversal

0b = Polarity Reversal Enabled.
If polarity is disabled, then the polarity is
forced to be normal in 10BASE-T.
1b = SQE test enabled.

SQE Test

0b = SQE test disabled.
Jabber has effect only in 10BASE-T half
duplex mode.
1b = Always power up.
0b = Can power down.

MAC Interface Power
Down

This bit determines whether the MAC
interface powers down when register
PHY Control Register bit 11 is used to
power down the Ethernet controller or
when the PHY enters the energy detect
state.
1b = 125CLK Low.

Disable 125CLK1
Reserved

0b = 125CLK Toggling.
4

Bit 4 = ENA_XC.
This bit is reserved for all Ethernet
controllers except the 82544GC/EI.
Should be set to 0b.

00b = Manual MDI configuration.
01b = Manual MDI-X configuration.
10b = Reserved.
MDI Crossover Mode

6:5

11b = Enable automatic crossover for all
modes.

R/W

11b

Update

R/W

Retain

82544GC/EI only:
Bit 6 = DIS_125.
Bit 5 = ENA_XC.
1b = Lower 10BASE-T receive
threshold.
Enable Extended
Distance

264

0b = Normal 10BASE-T receive
threshold.
When a cable longer than 100 m is
used, the 10BASE-T receive threshold
must be lowered in order to detect
incoming signals.

Software Developer’s Manual

Table 13-31. PHY Specific Control Register Bit Description
Description
Field

Bit(s)

Mode
1000BASE-T

HW Rst

SW Rst

10/100BASE-T

Energy Detect.
0xb = Off.
Energy Detect

9:8

10b = Sense only on receive.

R/W

Retain

R/W

Retain

R/W

Retain

R/W

00b

Retain

R/W

00b

Retain

11b = Sense and periodically transmit
NLP.
1b = Force link good.
0 b = Normal operation.
Force Link Good

Assert CRS on
Transmit

If link is forced to be good, the link state
machine is bypassed and the link is
always up. In 1000BASE-T mode this
has no effect.
1b = Assert on transmit.
0b = Never assert on transmit.
This bit has no effect in full duplex.

Receive FIFO depth

13:12

00b = 16 Bits

00b = 8 Bits

01b = 24 Bits

01b = 12 Bits

10b = 32 Bits

10b = 16 Bits

11b = 40 Bits

11b = 20 Bits

Receive FIFO is enabled in internal
SerDes mode2 or serial interface mode.

Transmit FIFO depth

15:14

00b = 16 Bits

00 b= 8 Bits

01b = 24 Bits

01b = 12 Bits

10b = 32 Bits

10b = 16 Bits

11b = 40 Bits

11b = 20 Bits

Transmit FIFO is enabled in 1000BASET mode or serial interface mode.
1.
2.

82544GC/EI only.
TBI Mode for the 82544GC/EI.

PHY Port Configuration Register (82541xx and 82547GI/EI Only)
PPCONF (16d; R/W)

Table 13-32. PHY Port Configuration Register Bit Description
Field
Reserved

Bit(s)
0

Description
Write to 0b for normal operation.

Mode

HW Rst

SW Rst

R/W

1b = Enable alternate Auto-Negotiate
next page feature.
Alternate NP Feature

0b = Disable alternate Auto-Negotiate
next page feature.
If polarity is disabled, then the polarity is
forced to be normal in 10BASE-T.

Reserved

Software Developer’s Manual

3:2

Write to 0b for normal operation.

265

Table 13-32. PHY Port Configuration Register Bit Description
Field

Auto MDIX Parallel
Detect Bypass

Bit(s)

Description

Mode

Auto_MDIX Parallel Detect Bypass.
Bypasses the fix to IEEE auto-MDIX
algorithm for the case where the PHY is
in forced-speed mode and the link
partner is auto-negotiating.
1b = Strict 802.3 Auto-MDIX algorithm.
0b = Auto-MDIX algorithm handles
Auto-Negotiation disabled modes. This
is accomplished by lengthening the
auto-MDIX switch timer before
attempting to swap pairs on the first time
out.

HW Rst

SW Rst

R/W

0b1

R/W

Preamble Enable
PRE_EN

0b = Set RX_DV high coincident with
SFD.
1b = Set RX_DV high and RXD =
preamble (after CRS is asserted).

Reserved

Smart Speed

TP Loopback
(10BASE-T)

Reserved

Jabber (10BASE-T)

Bypass 4B5B
(100BASE-TX)

Bypass Scramble
(100BASE-TX)

Transmit Disable

Write to 0b for normal operation.
1b = Smart Speed selection enabled.
0b = Smart Speed selection disabled.
1b = Disable TP loopback during halfduplex operation.
0b = Normal operation.
Write to 0b for normal operation.
1b = Disable jabber.
0b = Enable jabber.
1b = Bypass4B5B encoder and
decoder.
0b = Normal operation.
1b = Bypass scrambler and
descrambler.
0b = Normal operation.
1b = Disable twisted-pair transmitter.
0b = Normal operation.
1b = Force link pass
0b = Normal operation

Link Disable

Reserved

16:15

266

For 10BASE-T, this bit forces the link
signals to be active. In 100BASE-T
mode, setting this bit should force the
Link Monitor into it’s LINKGOOD state.
For Gigabit operation, this merely
bypasses Auto-Negotiation—the link
signals still correctly indicate the
appropriate status.
Always read as 0b. Write 0b for normal
operation.

The default for this bit is determined by EEPROM bit SSPEED.

Software Developer’s Manual

13.4.7.1.14

PHY Specific Status Register
PSSTAT (17d; R)

Table 13-33. PHY Specific Status Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

Jabber (real time)

1b = Jabber.
0 = No jabber.

Retain

Polarity (real time)

1b = Reversed.
0b = Normal.

1b = Receive pause enabled.
0b = Receive pause disabled.
The Receive Pause Enable bit is valid
only after the Speed and Duplex
Resolved bit (11) is set. This occurs
when Auto-Negotiation is completed or
Auto-Negotiation is disabled.
This is a reflection of the MAC pause
resolution.

Transmit Pause
Enabled

1b = Transmit pause enabled.
0b = Transmit pause disabled.
The Transmit Pause Enable bit is valid
only after the Speed and Duplex
Resolved bit (11) is set. This occurs
when Auto-Negotiation is completed or
Auto-Negotiation is disabled.

Energy Detect Status

0b = Active
1b = Sleep

Downshift Status

1b = Downshift.
0b = No Downshift.

1b = MDI-X.
0b = MDI.
The MDI Crossover Status bit is valid
only after the Speed and Duplex
Resolved bit (11) is set. This occurs
when Auto-Negotiation is completed or
Auto-Negotiation is disabled.

Cable Length
(100/1000 modes
only)

9:7

000b = < 50 m.
001b = 50  80 m.
010b = 80  110 m.
011b = 110  140 m.
100b = >140 m.
Cable length measurement is only a
rough estimate. The actual value
depends on the attenuation of the cable,
output levels of the remote transceiver,
connector impedance, etc.

000b

Link (real time)

1b = Link up.
0b = Link down.

Receive Pause
Enable

MDI Crossover
Status

Software Developer’s Manual

267

Table 13-33. PHY Specific Status Register Bit Description
Field

Description

Mode

HW Rst

SW Rst

Speed and Duplex
Resolved

1b = Resolved.
0b = Not resolved.
Speed, Duplex, MDI Crossover Status,
Transmit Pause Enable, and Receive
Pause Enable bits are valid only after
the Speed and Duplex Resolved bit (11)
is set. This occurs when AutoNegotiation is completed or AutoNegotiation is disabled.

Page Received

1b = Page received.
0b = Page not received.

RO,
LH

1b = Full duplex.
0b = Half duplex.
The Duplex bit is valid only after the
Speed and Duplex Resolved bit (11) is
set. This occurs when Auto-Negotiation
is completed or Auto-Negotiation is
disabled.

Retain

15:14

11b = Reserved.
10b = 1000 Mb/s.
01b = 100 Mb/s.
00b = 10 Mb/s.
The Speed bit is valid only after the
Speed and Duplex Resolved bit (11) is
set. This occurs when Auto-Negotiation
is completed or Auto-Negotiation is
disabled.

00b

Retain

Duplex

Speed

268

Bit(s)

Software Developer’s Manual

PHY Port Status 1 Register (82541xx and 82547GI/EI Only)
PPSTAT (17d; R)

Table 13-34. PHY Status 1 Register Bit Description
Field

LFIT Indicator

Bit(s)

Description
Status bit indicating the AutoNegotiation Link Fail Inhibit Timer has
expired. This indicates that the AutoNegotiation process completed page
exchanges but was unable to bring up
the selected MAU’s link.
1b = Auto-Negotiation has aborted Link
establishment following normal page
exchange.
0b = Auto-Negotiation has either
completed normally, or is still in
progress.
This bit is cleared when read or when
one of the following occurs:
Link comes up (PHY register 17, bit 10 =
1b).
Auto-Negotiation is disabled (PHY
register 0, bit 12 = 0b).

Mode

HW Rst

SW Rst

RO/
LH/
SC

Auto-Negotiation is restarted (PHY
register 0, bit 9 = 1b).
Polarity Status

Reserved

8:2

Duplex Mode

Link

1b = 10BASE-T polarity is reversed.
0b = 10BASE-T polarity is normal.
Write to 0b for normal operation.
1b = Full duplex.
0b = Half duplex.
Indicates the current status of the link.
Differs from PHY register 1, bit 2 in that
this bit changes anytime the link status
changes. PHY register 1, bit 2 latches
low and stays low until read regardless
of link status.
1b = Link is currently up.
0b = Link is currently down.

MDI-X Status

Status indicator of the current MDI/MDIX state of the twisted pair interface. This
status bit is valid regardless of the MAU
selected.
1b = PHY has selected MDI-X (crossed
over).
0b = PHY has selected MDI (NOT
crossed over).
1b = PHY currently receiving a packet.

Receive Status

0b = PHY receiver is IDLE.
When in loopback, this bit reads as 0b.

Software Developer’s Manual

269

Table 13-34. PHY Status 1 Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

1b = PHY currently transmitting a
packet.
Transmit Status

0b = PHY transmitter is IDLE.

When in loopback, this bit reads as 0b.
00b = Reserved.
01b = PHY operating in 10BASE-T
mode.
Data Rate

15:14

10b = PHY operating in 100BASE-TX
mode.
11b = PHY operating in 1000BASE-T
mode.

13.4.7.1.15

PHY Interrupt Enable Register
PINTE (18d; R/W)

Table 13-35. PHY Interrupt Enable Bit Description
Field

270

Bit(s)

Description

Mode

HW Rst

SW Rst

Jabber Interrupt
Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

Polarity Changed
Interrupt Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

Reserved

3:2

Always write 00b.

R/W

Retain

Energy Detect

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

Downshift Interrupt
Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

MDI Crossover
Changed
Interrupt Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

FIFO Over/Underflow
Interrupt Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

False Carrier
Interrupt Enable

1b = Interrupt enable.
0 b= Interrupt disable.

R/W

Retain

Symbol Error
Interrupt Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

Link Status Changed
Interrupt Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

Auto-Negotiation
Completed
Interrupt Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

Software Developer’s Manual

Table 13-35. PHY Interrupt Enable Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

Page Received
Interrupt Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

Duplex Changed
Interrupt Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

Speed Changed
Interrupt Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

Auto-Negotiation
Error
Interrupt Enable

1b = Interrupt enable.
0b = Interrupt disable.

R/W

Retain

PHY Port Control Register (82541xx and 82547GI/EI Only)
PPCONT (18d; R/W)

Table 13-36. PHY Port Control Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

Reserved

3:0

Always read as 0b. Write to 0b for
normal operation.

R/W

TP Loopback

Allow gigabit loopback on twisted pairs.

R/W

0b1

R/W

82541EI/82547GI (B0 stepping):
Reserved, write to 0b.
Fast Downshift
Enable

82541/GI/ER and 82547GI (B1
stepping): Fast 1000 Mb to 100 Mb
downshift enable.
0b = Downshift after 16 seconds.
1b = Downshift after 10 seconds.

Reserved

8:6

Non-Compliant
Scrambler
Compensation

TEN_CRS_Select

Reserved

Auto-MDI-X

Software Developer’s Manual

Always read as 0b. Write to 0b for
normal operation.
1b = Detect and correct for noncompliant scrambler.
0b = Detect and report non-compliant
scrambler.
1b = Extend CRS to cover GMII latency
and RX_DV.
0b = Do not extend CRS (RX_DV can
continue past CRS).
Always write as 1b for normal operation.
Auto-MDI-X algorithm enable.
1b = Enable Auto-MDI-X mode.
0b = Disable Auto-MDI-X mode (manual
mode).
Note: When forcing speed to 10Base-T
or 100Base-T, use manual mode. Clear
the bit and set PHY register 18, bit 13
according to the required MDI-X mode.

271

Table 13-36. PHY Port Control Register Bit Description
Field

Bit(s)

MDI-X Mode

Description

Mode

Force MDI-X mode. Valid only when
operating in manual mode. (PHY
register 18, bit 12 = 0b.

HW Rst

SW Rst

R/W

1b = MDI-X (cross over).
0b = MDI (no cross over).
Reserved

Always read as 0b. Write to 0b for
normal operation.
This configuration bit is used to enable
the Ethernet controller to drive its
differential transmit clock out through
the appropriate Analog Test (ATEST+/-)
output pads. This feature is required in
order to demonstrate conformance to
the IEEE Clause 40 jitter specification.
When high, it sends Jitter Test Clock
out.

Jitter Test Clock

This bit works in conjunction with
internal PHY register 18, bit 15. In order
to have the clock probed out, it is
required to perform the following write
sequence:
PHY register 18, bit15 = 1b
PHY register 31 = 4000h (page select)
PHY register 17 = 0080h
PHY register 31 = 0000h (page select)

13.4.7.1.16

The default for this bit is determined by EEPROM configuration bits. If EEPROM bit NCSCRAMB is asserted, then the
default is set to 1b.

PHY Interrupt Status Register
PINTS (19d; R)

Table 13-37. PHY Interrupt Status Bit Description
Field

272

Bit(s)

Description

Mode

HW Rst

SW Rst

Jabber

1b = Jabber.
0b = No jabber.

RO,
LH

Polarity Changed

1b = Polarity Changed.
0b = Polarity not changed.

RO,
LH

Reserved

3:2

Reserved. Should be set to 00b.

Always 00b

Energy Detect

1b = Energy Detect state changed
0b = No state change detected

RO,
LH

Downshift Detected

1b = Downshift detected.
0b = No down shift.

RO,
LH

MDI Crossover
Changed

1b = Crossover changed.
0b = Crossover not changed.

RO,
LH

Software Developer’s Manual

Table 13-37. PHY Interrupt Status Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

FIFO Over/Underflow

1b = Over/Underflow Error.
0b = No FIFO Error.

RO,
LH

False Carrier

1b = False carrier.
0b = No false carrier.

RO,
LH

Symbol Error

1b = Symbol error.
0b = No symbol error.

RO,
LH

Link Status Changed

1b = Link status changed.
0b = Link status not changed.

RO,
LH

Auto-Negotiation
Completed

1b = Auto-Negotiation completed.
0b = Auto-Negotiation not completed.

RO,
LH

Page Received

1b = Page received.
0b = Page not received.

RO,
LH

Duplex Changed

1b = Duplex changed.
0b = Duplex not changed.

RO,
LH

Speed Changed

1b = Speed changed.
0b = Speed not changed.

RO,
LH

1b = Auto-Negotiation Error.
0b = No Auto-Negotiation Error.
An error occurs if MASTER/SLAVE
does not resolve, parallel detect fault,
no common HCD, or link does not
validate after negotiation has
completed.

RO,
LH

Auto-Negotiation
Error

PHY Link Health Register (82541xx and 82547GI/EI Only)
PLINK (19d; R)

Table 13-38. PHY Link Health Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

Valid Channel A

The channel A DSP had converged to
incoming data.

Valid Channel B

The channel B DSP had converged to
incoming data.

Valid Channel C

The channel C DSP had converged to
incoming data.

Valid Channel D

The channel D DSP had converged to
incoming data.

Auto-Negotiation
Active

Auto-Negotiate is actively deciding
HCD.

Reserved

Always read as 0b.

Software Developer’s Manual

273

Table 13-38. PHY Link Health Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

Auto-Negotiation
Fault

Auto-Negotiate Fault: This is the logical
OR of PHY register 1, bit 4, PHY
register 6, bit 4, and PHY register 10, bit
15.

Reserved

Always read as 0b.

RO/
LH

Mode:
Data Err[0]

10: 10 Mbps polarity error.
100: Symbol error.
1000: Gig idle error.
Mode:

Data Err[1]

10: Reserved.
100: Scrambler unlocked.
1000: Local receiver not OK.

Count Overflow

The idle error counter has overflowed.

RO/
LH

Gigabit Rem Rcvr
NOK

Gig has detected a remote receiver
status error. This is a latched high
version of PHY register 10, bit 12.

RO/
LH

Gig has resolved to master. This is a
duplicate of PHY register 10, bit 14.
Programmers must read PHY register
10, bit 14 to clear this bit.

A fault has occurred with the gig master/
slave resolution process. This is a copy
of PHY register 10, bit 15.
Programmers must read PHY register
10, bit 15 to clear this bit.

Gigabit Scrambler
Error

1b indicates that the PHY has detected
gigabit connection errors that are most
likely due to a non-IEEE compliant
scrambler in the link partner.
0b = Normal scrambled data.
Definition is: If an_enable is true and in
Gigabit mode, on the rising edge of
internal signal link_fail_inibit
timer_done, the dsp_lock is true but
loc_rcvr_OK is false.

SS Downgrade

Smart Speed has downgraded the link
speed from the maximum advertised.

RO/
LH

Gigabit Master
Resolution

Gigabit Master Fault

274

Software Developer’s Manual

13.4.7.1.17

Extended PHY Specific Control Register 11
EPSCON1 (20d; R/W)

Table 13-39. Extended PHY Specific Control 1 Bit Description1
Field

Description

Mode

HW Rst

SW Rst

Reserved

1:0

00b

R/W

00b

Retain

Reserved

R/W

6:4

Reserved. Should be set to 0b.
Changes to this bit are disruptive to the
normal operation; any change to this
register must be followed by software
reset to take effect.

R/W

110b

Update

Reserved. Should be set to 0b.
Changes to this bit are disruptive to the
normal operation; any change to this
register must be followed by software
reset to take effect.

R/W

Update

9:8

00b = disable. (10/100 downshift)
01b = 1x.
10b = 2x.
11b = 3x.
Changes to this bit are disruptive to the
normal operation; any change to this
register must be followed by software
reset to take effect.
Bits 11:10 have no effect unless bits 1:0
are set to their default values.

R/W

01b

Update

Master downshift
counter

11:10

00b = 1x.
01b = 2x.
10b = 3x.
11b = 4x.
Number of times that the PHY attempts
to achieve gigabit link before
downshifting to the next speed in Master
Mode.
Changes to this bit are disruptive to the
normal operation; any change to this
register must be followed by software
reset to take effect.
Bits 11:10 have no effect unless bits 1:0
are set to their default values.

R/W

11b

Update

Reserved

Reserved. Should be set to 0b.

R/W

Reserved

Reserved. Should be set to 0b.

R/W

Reserved

Reserved. Should be set to 0b.

R/W

Reserved

Reserved. Should be set to 0b.

R/W

Reserved

Slave downshift
counter

Bit(s)

Extended PHY Specific Control Register - EPSCON for the 82544GC/EI only.

Software Developer’s Manual

275

GMII FIFO Register (82541xx and 82547GI/EI Only)
PFIFO (20d; R/W)

Table 13-40. GMII FIFO Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

3:0

An unsigned integer that stipulates the
number of write clocks to delay the read
controller after internal GMII’s tx_en is
first asserted. This “buffer” protects from
underflow at the expense of latency.
The maximum value that can be set is
13d or Dh.

R/W

0101b

When set, the PHY advertises higher
speed than 10Base-T after reconnect of
the cable, even if the software
advertised only 10Base-T speed.

R/W

Note 2

Power Down On Link
Disconnect

When set, the PHY optimizes the power
consumption when the cable is
disconnected. The PHY gets back to
normal operation reconnect of the cable,
supporting Auto-Negotiation and parallel
detection.

R/W

Note 1

Reserved

7:6

Always read as 0b. Write to 0b for
normal operation.

R/W

00b

R/W

00b

Buffer Size

Enable Speed-Up
Upon Cable
Reconnect

FIFO Out Steering

9:8

00b, 01b: Enable the output data bus
from GMII FIFO to transmitters, drives
zeros on the output loop-back bus from
GMII FIFO to external application and to
DSP RX-FIFOs in test mode.
10b: Drive zeros on output bus from
GMII FIFO to transmitters, enable data
on the output loop-back bus from GMII
FIFO to external application and to DSP
RX-FIFOs in test mode.
11b: Enable the output data bus from
GMII FIFO to both transmitters and
loop-back bus.

Disable Error Out

When set, disables the addition of
under/overflow errors to the output data
stream on internal GMII’s tx_error.

R/W

Reserved

13:11

Always read as 0b. Write to 0b for
normal operation.

R/W

FIFO Overflow

Status bit set when read clock that is
faster than internal GMII’s gtx_clk
empties the FIFO mid packet. Increase
the buffer size.

RO/
LH

FIFO Underflow

Status bit set when read clock that is
slower than internal GMII’s gtx_clk has
allowed the FIFO to fill to capacity mid
packet. Decrease buffer size.

RO/
LH

NOTES:
1. The default is determined by EEPROM bit SPD_EN.
2. The default is determined by EEPROM bit ADV10LU.

276

Software Developer’s Manual

13.4.7.1.18

PHY Receive Error Counter
PREC (21d; R)

Table 13-41. PHY Receive Error Counter Bit Description
Field

Bit(s)

Receive Error Count

15:0

Description
Error Count.

Mode
RO,SC

HW Rst
0000h

SW RST
0000h

NOTE: The counter stops at FFFFh and does not roll over.

PHY Channel Quality Register (82541xx and 82547GI/EI Only)
PCHAN (21d; R)

Table 13-42. PHY Channel Quality Register Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

MSE D

3:0

The converged mean square error for
Channel D.

MSE C

7:4

The converged mean square error for
Channel C.

MSE B

11:8

The converged mean square error for
Channel B.

MSE A

15:12

The converged mean square error for
Channel A. This field is only meaningful
in gigabit, or in 100BASE-TX if this is
the receive pair.
Use of this field is complex and needs
interpretation based on the chosen
threshold value.

13.4.7.1.19

SPEED_TEN_LED and LINK_ACT_LED Control (82541xx and 82547GI/EI Only)
(23d; R/W)

Table 13-43. SPEED_TEN_LED and LINK_ACT_LED Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

LED Source Select

3:0

MUX the designated input to
SPEED_TEN_LED.

R/W

0000b

LED Blink Disable

Disable the SPEED_TEN_LED Blink
Logic.
0b = Enable logic.
1b = Disable logic.

R/W

Software Developer’s Manual

277

Table 13-43. SPEED_TEN_LED and LINK_ACT_LED Bit Description

LED Stretch Disable

Disable the SPEED_TEN_LED
Extension Logic.
0b = Enable logic.
1b = Disable logic.
Note: Only when both the stretch and
blink are disabled the input bypasses
the blink logic and is muxed out with no
sampling (only combinational logic).

LED Source Select

9:6

Mux the designated input to
LED_ACT_LED.

R/W

0001b

LED Blink Disable

Disable the LINK_ACT_LED Blink
Logic.
0b = Enable logic.
1b = Disable logic.

R/W

LED Stretch Disable

Disable the LINK_ACT_LED Extension
Logic.
0b = Enable logic.
1b = Disable logic.
Note: Only when both the stretch and
blink are disabled the input bypasses
the blink logic and is muxed out with no
sampling (only combinational logic).

R/W

Invert Select

When set to 1b, all LED outputs are
inverted.

R/W

Reserved

14:13

Always read as 0b. Write to 0b for
normal operation

R/W

00b

This bit is used to disable special power
saving in 10BASE-T mode and parallel
detection. When set to 1b, power
reduction features of 10BASE-10 are
disabled (reserved for customers).

R/W

Disable 10 Power
Saving

278

R/W

Software Developer’s Manual

13.4.7.1.20

PHY Global Status (82544GC/EI Only)
PGSTAT (23d; R)

Table 13-44. PHY Global Status Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

Port 0 Interrupt

0b = No Interrupt on Port.
1b = Interrupt on Port.

Port 1 Interrupt

0b = No Interrupt on Port.
1b = Interrupt on Port.

Port 2 Interrupt

0b = No Interrupt on Port.
1b = Interrupt on Port.

Port 3 Interrupt

0b = No Interrupt on Port.
1b = Interrupt on Port.

Reserved

15:4

Reserved. Should be set to 0b.

NOTE: Bits 3:0 remain high until the active corresponding interrupt bits are cleared on a read of the PHY
Interrupt Status Register.

13.4.7.1.21

SPEED_100_LED and SPEED_1000_LED Control (82541xx and 82547GI/EI Only)
(24d; R/W)

Table 13-45. SPEED_100_LED and SPEED_1000_LED Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

3:0

MUX the designated input to
SPEED_100_LED.

R/W

0011b

Disable the SPEED_100_LED Blink
Logic.
0b = Enable logic.
1b = Disable logic.

R/W

LED Stretch Disable

Disable the SPEED_100_LED
Extension Logic.
0b = Enable logic.
1b = Disable logic.
Note: Only when both the stretch and
blink are disabled the input bypasses
the blink logic and is muxed out with no
sampling (only combinational logic).

R/W

LED Source Select

9:6

Mux the designated input to
SPEED_1000_LED.

R/W

0100b

LED Blink Disable

Disable the SPEED_1000_LED Blink
Logic.
0b = Enable logic.
1b = Disable logic.

R/W

LED Source Select

LED Blink Disable

Software Developer’s Manual

279

Table 13-45. SPEED_100_LED and SPEED_1000_LED Bit Description

13.4.7.1.22

LED Stretch Disable

Disable the SPEED_1000_LED
Extension Logic.
0b = Enable logic.
1b = Disable logic.
Note: Only when both the stretch and
blink are disabled the input bypasses
the blink logic and is muxed out with no
sampling (only combinational logic).

Reserved

15:12

Always read as 0b. Write to 0b for
normal operation

R/W

00b

PHY LED Control Register (82544GC/EI Only)
PLED (24d; R/W)

Table 13-46. PHY LED Control Bit Description
Field

Bit(s)

LED_TX control

LED_RX control

Reserved

LED_LINK control

4:3

Description
1b = Activity/Link.
0b = Transmit activity.
1b = Receive activity/Link.
0b = Receive activity.
Reserved. Should be set to 0b.

Mode

HW Rst

SW Rst

R/W

Retain

R/W

Retain

R/W

Retain

R/W

00b

Retain

R/W

000b

Retain

R/W

001b

Retain

R/W

Retain

R/W

100b

Retain

R/W

Retain

1xb = Link[2:1], Link.
01b = Link, Speed[1:0].
00b = Link[2:0].
Reserved

7:5

Reserved. Should be set to 000b.
000b = 42 ms.
001b = 84 ms.

Blink Rate

10:8

010b = 170 ms.
011b = 340 ms.
100b = 670 ms.
101b to 111b = Reserved.

Force INT# to Assert

0b = Do not force INT# assertion.
1b = Force INT# assertion.
000b = no pulse stretching.
001b = 21 ms to 42 ms.
010b = 42 ms to 84 ms.

Pulse stretch duration

14:12

01b1 = 84 ms to 170 ms.
100b = 170 ms to 340 ms.
101b = 340 ms to 670 ms.
110b = 670 ms to 1.3 s.
111b = 1.3s to 2.7 s.

Disable LED

280

0b = Enable.
1b = Disable.

Software Developer’s Manual

13.4.7.1.23

Extended PHY Specific Control Register 2
EPSCON2 (26d; R/W)

Table 13-47. Extended PHY Specific Control Register 2 Bit Description
Field

Bit(s)

Description

Mode

HW Rst

SW Rst

Fiber Output
Amplitude

2:0

111b = 1.2, 100b = 0.9, 001b = 0.6
110b = 1.1, 011b = 0.8, 000b = 0.5
101b = 1.0, 010b = 0.7

R/W

100b

Retain

Reserved

Reserved. Should be set to 1b.

R/W

Reserved

Reserved. Should be set to 0b.

R/W

Fiber Output
Impedance

1b= 75 ohm.
0b = 50 ohm.

R/W

Update

Fiber Input
Impedance

1b = 75 ohm.
0b = 50 ohm.

R/W

Update

Reserved

15:7

Reserved. Should be set to 0b.

R/W

000h

NOTE: Not applicable to the 82540EP/EM, 82544GC/EI, 82541xx, or 82547GI/EI.

13.4.7.1.24

Extended PHY Specific Status Register (82544GC/EI Only)
EPSSTAT (27d; R)

Table 13-48. Extended PHY Specific Status Bit Description
Field

13.4.7.1.25

Bit(s)

Description

Mode

HW Rst

SW Rst

MODE[3:0]

27.3:0

MODE[3:0].

MODE[3:0]

Retain

Reserved

27.15:4

Reserved. Should be set to
000000000000b.

MDI Register 30 Page Select1
R30PS (29d; WO)

Table 13-49. MDI Register 30 Page Select Bit Description
Field
Register 30 Page
Select

Bit(s)
15:0

Description
Selects the register accessible via the
“window” at MDI register 30.

Mode
R/W

HW Rst
0000h

SW Rst
0000h

Not applicable to the 82544GC/EI, 82541xx, or 82547GI/EI.

Software Developer’s Manual

281

13.4.7.1.26

MDI Register 30 Access Window1
R30AW (30d; R/W)

Table 13-50. MDI Register 30 Page Select Bit Description
Field

Bit(s)

13.4.7.1.27

15:0

Description

Mode

Provides read/write capability for
register selected via MDI register 29.

R/W

HW Rst
0000h

SW Rst
0000h

Documented MDI Register 30 Operations1

Unless otherwise specified, no reset operations are required in order for the following operations to
take effect.

Table 13-51. MDI Register 30 Operations
To Perform
Power down SerDes (optimize
power for copper PHY operation)1

Operation
30_31.15:0 <= x2001h
30_5.8 <= 0b

MDI Read/Write Sequence
• Write MDI register 29 <= 31d
• Write MDI register 30 <= 2001h
• Write MDI register 29 <= 5d
• Read MDI register 30
• Change bit 8 to 0
• Write result to MDI register 30

Tune VCO on SerDes Rx for
optimal Bit Error Ratio (BER)1
30_4.11 <= 1b

• Write MDI register 29 <= 4d
• Read MDI register 30
• Change bit 11 to 1
• Write result to MDI register 30

Set PHY output drivers into Class A
mode (Class AB is default after
reset)
1.

30_11.15:0 = 8004h

• Write MDI register 29 = 11d
• Write MDI register 30 = 8004h

Not applicable to 82540EP/EM

NOTE: Any time the PHY is reset it returns to Class AB drive mode.

Not applicable to the 82544GC/EI, 82541xx, or 82547GI/EI.

282

Software Developer’s Manual

13.4.7.1.28

PHY Page Select Register (82541xx and 82547GI/EI Only)
PPAGE (31d; R/W)

Table 13-52. PHY Page Select Register Bit Description
Field

Bit(s)

PAGE_SEL

13.4.8

Description

Mode

This register is used to swap out the
Base Page containing the IEEE
registers for Intel reserved test and
debug pages residing within the
Extended Address space.

15:0

HW Rst

SW Rst

Flow Control Address Low
FCAL (00028h; R/W)
Flow control packets are defined by IEEE 802.3x to be either a unique multicast address or the
station address with the EtherType field indicating PAUSE. The FCAL, FCAH registers provide
the value hardware compares incoming packets against to determine that it should PAUSE its
output, and hardware use when transmit PAUSE packets to its remote node when flow control is
activated.
The FCAL register contains the lower bits of the internal 48-bit Flow Control Ethernet address. All
32 bits are valid. Software can access the High and Low registers as a register pair if it can perform
a 64-bit access to the PCI bus. This register should be programmed with 00_C2_80_01h. The
complete flow control multicast address is: 01_80_C2_00_00_01h; where 01h is the first byte on
the wire, 80h is the second, etc.
Table 13-53. FCAL Register Bit Description
31

0
FCAL

Field

FCAL

13.4.9

Initial
Value

Bit(s)

31:0

Description
Flow Control Address Low
Should be programmed with 00_C2_80_01h

Flow Control Address High
FCAH (0002Ch; R/W)
This register contains the upper bits of the 48-bit Flow Control Ethernet address. Only the lower 16
bits of this register have meaning. The complete Flow Control address is {FCAH, FCAL}. This
register should be programmed with 01_00h. The complete flow control multicast address is:
01_80_C2_00_00_01h; where 01h is the first byte on the wire, 80h is the second, etc.

Software Developer’s Manual

283

Table 13-54. FCAH Register Bit Description
31

16 15

Reserved

Field

13.4.10

FCAH

Initial
Value

Bit(s)

Description

FCAH

15:0

Flow Control Address High
Should be programmed with 0100h.

Reserved

31:16

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

Flow Control Type
FCT (00030h; R/W)
This register contains the type field that hardware matches to recognize a flow control packet and
that hardware uses when transmitting a PAUSE packet to its remote node. Only the lower 16 bits of
this register have meaning. This register should be programmed with 88_08h. The upper byte is
first on the wire FCT[15:8].
Table 13-55. FCT Register Bit Description
31

16 15

Reserved

Field

13.4.11

Initial
Value

Bit(s)

FCT

Description

FCT

15:0

Flow Control Type
Should be programmed with 88_08h.

Reserved

31:16

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

VLAN Ether Type
VET (00038h; R/W)
This register contains the type field hardware matches against to recognize an 802.1Q (VLAN)
Ethernet packet and uses when add and transmit VLAN Ethernet packets. To be compliant with the
802.3ac standard, this register should be programmed with the value 8100h. For VLAN
transmission the upper byte is first on the wire (VET[15:8]).

284

Software Developer’s Manual

Table 13-56. VET Register Bit Description
31

16 15

Reserved

Field

13.4.12

VET

Initial
Value

Bit(s)

Description

VET

15:0

VLAN EtherType
Should be programmed with 8100h.

Reserved

31:16

Reserved
Reads as 0b.

Flow Control Transmit Timer Value
FCTTV (00170h; R/W)
Provides the Pause slot time value to be included in the transmitted XOFF Pause packets.
The slot time value that is used is a fixed slot of 64-byte time.
Table 13-57. FCTTV Register Bit Description
31

16 15
Reserved

Field

Initial
Value

Bit(s)

0
TTV

Description

TTV

15:0

Transmit Timer Value
Slot time value (Slot time value is 64-byte time) to be inserted into
the transmitted Pause frame. If software wishes to send XON
frame, it should set the FCTTV. TTV value to 0b prior to initializing
the Pause frame.

Reserved

31:16

Reserved
Reads as 0b.
Should be written to 0b for future compatibility.

Software Developer’s Manual

285

13.4.13

Transmit Configuration Word Register1
TXCW (00178h; R/W)
This register is applicable to the TBI mode/internal SerDes mode of operation. For internal PHY
operation, program the register to 0000h. For example, clear this register in MMI mode.
This register has two meanings, depending on the state of Auto-Negotiation: one as the “AN
advertise register” defined by IEEE 802.3z, and the other as a register for software control of the
Auto-Negotiation process.
When performing hardware Auto-Negotiation, it fulfills the function defined by sub-clause
37.3.6.1.3 of IEEE 802.3z.
Table 13-58. TXCW Register Bit Description
31

Field

TxConfigWord

16 15

Reserved

Bit(s)

15:0

Initial
Value

TxConfigWord

Description
Data transmitted during Auto-Negotiation process.
When performing hardware Auto-Negotiation (TXCW.ANE is
set), the value of TxConfigWord is encoded as two 10bit symbols and sent as the “config_word” field for the /C/
ordered set. When the Ethernet controller performs software
Auto-Negotiation, TxConfig and TxConfigWord are used to
negotiate with the link partner. Data stored in TxConfigWord is
transmitted during the Auto-Negotiation process. Software
should not read back the contents of this field as content might
change during the software Auto-Negotiation process. In the
absence of loss of synchronization, the value read back is stable
and equal to the value written.
The mapping of the TxConfigWord is as follows:
• TxConfigWord[15] Next page request
• TxConfigWord[14] Reserved (write as 0b; ignore on read)
• TxConfigWord[13:12] Remote fault indication
• TxConfigWord[11:9] Reserved (write as 0b; ignore on read)
• TxConfigWord[8:7] Pause
• TxConfigWord[6] Half-duplex
• TxConfigWord[5] Full-duplex
• TxConfigWord[4:0] Reserved (write as 0b; ignore on read)
Bits 5,7 & 8 of TxConfigWord are loadable from the EEPROM
upon power-up, or chip reset.

Reserved

29:16

Reserved
Reads as 0b.
Should be written to 0b for future compatibility.

Not applicable to the 82541xx, 82547GI/EI, or 82540EP/EM.

286

Software Developer’s Manual

Field

TxConfig

ANE

Note:

13.4.14

Bit(s)

Initial
Value

Description

Transmit Config Control bit
0b = Transmit data/idle
1b = Transmit /C/ ordered sets
Setting the TxConfig bit causes transmission of /C/ ordered set in
a software controlled Auto-Negotiation process
(TXCW.ANE=0b).

Auto-Negotiation Enable.
1b = Enable the hardware Auto-Negotiation state machine.
0b = Disable the hardware Auto-Negotiation state machine.
This bit has the same function as bit 0.12 defined in sub-clause
22.2.4.1.4 of the 802.3z standard. Since this bit is a “static”
value, a pulse is generated by hardware in response to writing
this bit with a 1b. This pulse is used to restart the AutoNegotiation state machine.
When ANE is set, a transition from loss of synchronization to
synchronized state restarts the Auto-Negotiation as well.
If ANE is cleared, then the Ethernet controller is performing
software Auto-Negotiation. In that case TxConfig and
TxConfigWord are used to negotiate with the link partner.
The ANE is loadable from the EEPROM upon power up or chip
reset.

Careful attention to the IEEE 802.3z standard is required in order to meet specified timing
requirements for timing during a software negotiated link.

Receive Configuration Word Register1
RXCW (00180h; R)
This register has meaning only in TBI/internal SerDes mode of operation. The RXCW register
records the partner abilities and provides indications about its Auto-Negotiation status.
The contents of this register depend on the state of TXCW.ANE. If ANE is set, then this register
records the 16-bit defined in IEEE 802.3z. When performing software Auto-Negotiation, software
should look for RXCW.ANC. When RXCW.ANC is set, the contents of RXCW.RxConfigWord are
valid, when RXCW.ANC is cleared, then the content of this register is undefined.

Note:

While in internal SerDes mode (82546GB/EB and 82545GM/EM only), software might be
required to inspect or monitor the results of RXCW to generate a link up/down indication.

Not applicable to the 82541xx, 82547GI/EI, or 82540EP/EM.

Software Developer’s Manual

287

Table 13-59. RXCW Register Bit Description

Field

Description

RxConfigWord

15:0

Data received during Auto-Negotiation process.
When performing hardware Auto-Negotiation
(TXCW.ANE = 1b), the “AN link partner ability base page
register” is recorded in the RxConfigWord.
When TXCW.ANE is clear, then this register is used by
software to perform software based Auto-Negotiation. In
that capacity, RxConfigWord records the raw values
returned from the Auto-Negotiation process.

Reserved

25:16

Reserved
Reads as 0b.
Should be written with 0b for future compatibility.

Carrier Sense Indication
0b = Ethernet controller not receiving idle characters
(Carrier sense is true)
1b = Ethernet controller receiving idle characters (Carrier
sense is false)
Hardware sets RxConfigNoCarrier when it is receiving idle
characters (and when receiving link configuration
information or port status).
Software can use this bit to determine that idles have been
seen prior to exiting the link start-up procedure.
82544GC/EI only:
Valid only in software Auto-Negotiation mode
(TXCW.ANE = 0b).

Invalid Symbol during configuration process
0b = Have not received an invalid symbol during AutoNegotiation process.
1b = Have received an invalid symbol during AutoNegotiation process (bit is LH).
If the Ethernet controller detects an invalid symbol at any
time, it sets the RxConfigInvalid bit. The bit is latched high
until read by software. Software is expected to restart the
configuration process when the Ethernet controller receives
an invalid symbol at any time during the Auto-Negotiation
process.
82544GC/EI only:
Valid only in software Auto-Negotiation mode
(TXCW.ANE = 0b).

Change to the RxConfigWord indication
0b = RxConfigWord has changed since last read.
1b = RxConfig is unchanged since last read (LH)
Indicates that the Ethernet controller interface has seen a
change to the RxConfigWord. This bit is latched high until
read by software.
82544GC/EI only:
Valid only in software Auto-Negotiation mode
(TXCW.ANE = 0b).

RxConfigNoCarrier

RxConfigInvalid

RxConfigChange

288

Initial
Value

Bit(s)

Software Developer’s Manual

Field

RxConfig

RxSynchronize

ANC

13.4.15

Bit(s)

Initial
Value

Description

/C/ order set reception indication
0b = Receive idle/data stream.
1b = Receiving /C/ order sets.
Provides an indication as to whether the interface is
receiving /C/ order set, or normal idle/data stream.
82544GC/EI only:
Valid only in software Auto-Negotiation mode
(TXCW.ANE = 0b).

Lost bit synchronization indication
0b = Lost synchronization.
1b = Bit synchronization (bit is LL).
Used to qualify all other bits in the register (when in
software Auto-Negotiation).
Each time the Ethernet controller loses bit synchronization,
this bit becomes ‘0b’ and stays ‘0b’ until read by software.
82544GC/EI only:
Valid only in software Auto-Negotiation mode
(TXCW.ANE = 0b).

Auto Negotiation Complete
0b = The contents of the register are undefined in hardware
Auto-Negotiation mode (TXCW.ANE = 1b)
1b = The contents of the register are valid in hardware AutoNegotiation mode.
Reports the status as required in sub-clause 22.2.4.2.11.
This bit remains cleared from the time Auto-Negotiation is
reset until Auto-Negotiation reaches the “LINK_OK” state.
It remains set until Auto-Negotiation is disabled or restarted.
This bit has meaning only if TXCW.ANE = 1b. If software
Auto-Negotiation is used, it always reads 0b and should be
ignored.

LED Control1
LEDCTL (00E00h; RW)

31 - 24

23 - 16

15 - 8

7-0

LED3 (LINK1000#)

LED2 (LINK100#)

LED1 ACTIVITY#)

LED0 (LINK_UP#)

Not applicable to the 82544GC/EI.

Software Developer’s Manual

289

Table 13-60. LED Control Bit Description1
Field

Initial
Value

Description

LED0_MODE

3:0

0010b1

LED0/LINK# Mode. This field specifies the control source for the
LED0 output. An initial value of 0010b selects LINK_UP# indication.

Reserved

5:4

00b

Reserved. Read-only as 0b. Write as 0b for future compatibility.

LED0_IVRT

0b1

LED0/LINK# Invert. This field specifies the polarity/ inversion of the
LED source prior to output or blink control.
0b = do not invert LED source.
1b = invert LED source.

LED0_BLINK

0b1

LED0/LINK# Blink. This field specifies whether to apply blink logic to
the (possibly inverted) LED control source prior to the LED output.
0b = do not blink asserted LED output.
1b = blink asserted LED output.

LED1_MODE

11:8

0011b

LED1/ACTIVITY# Mode. This field specifies the control source for the
LED1 output. An initial value of 0011b selects ACTIVITY# indication.

Reserved

13:12

00b

Reserved. Read-only as 0b. Write as 0b for future compatibility.

LED1_IVRT

LED1/ACTIVITY# Invert.

LED1_BLINK

LED1/ACTIVITY# Blink.

LED2_MODE

19:16

0110b1

LED2/LINK100# Mode. This field specifies the control source for the
LED2 output. An initial value of 0011b selects LINK100# indication.

Reserved

21:20

00b

Reserved. Read-only as 0b. Write as 0b for future compatibility.

LED2_IVRT

0b1

LED2/LINK100# Invert.

LED2_BLINK

0b1

LED2/LINK100# Blink.

LED3_MODE

27:24

0111b

LED3/LINK1000# Mode. This field specifies the control source for the
LED3 output. An initial value of 0111b selects LINK1000# indication.

Reserved

29:28

00b

Reserved. Read-only as 0b. Write as 0b for future compatibility.

LED3_IVRT

LED3/LINK1000# Invert.

LED3_BLINK

LED3/LINK1000# Blink.

13.4.15.1

Bit

These bits are read from the EEPROM.

MODE Encodings for LED Outputs1
The Table 13-61 lists the MODE encodings used to select the desired LED signal source for each
LED output. Refer to Section 10.1.1 to ensure proper understanding of expression polarity and
resulting LED output polarity.

Note:

All 16 modes listed are functional.

Not applicable to the 82544GC/EI.

290

Software Developer’s Manual

Table 13-61. Mode Encodings for LED Outputs1
Mode

Pneumonic

State / Event Indicated

0000b

LINK_10/1000

Asserted when either 10 or 1000 Mbps link is
established and maintained.

0001b

LINK_100/1000

Asserted when either 100 or 1000 Mbps link is
established and maintained.

0010b

LINK_UP

Asserted when any speed link is established and
maintained.

0011b

ACTIVITY

Asserted when link is established and packets are
being transmitted or receive activity that passes
filtering.

0100b

LINK/ACTIVITY

Asserted when link is established and when there is
no transmit or receive activity that passes filtering.

0101b

LINK_10

Asserted when a 10 Mbps link is established and
maintained.

0110b

LINK_100

Asserted when a 100 Mbps link is established and
maintained.

0111b

LINK_1000

Asserted when a 1000 Mbps link is established and
maintained.

1000b

PCIX_MODE

Asserted when Ethernet controller is in PCI-X mode
(deasserted in PCI mode).

1001b

FULL_DUPLEX

Asserted when the link is configured for full duplex
operation (deasserted in half-duplex).

1010b

COLLISION

Asserted when a collision is observed.

1011b

BUS_SPEED

Asserted when the Ethernet controller is operating
in a PCI 66 MHz or a PCI-X 133 MHz configuration
(high-speed PCI operation), deasserted for 33 MHz
PCI and 66 MHz PCI-X (as determined by pins
sampled at PCI reset).

1100b

BUS_SIZE
Reserved for the
82547GI/EI only)

Asserted when the Ethernet controller is operating
as a 64-bit PCI or PCI-X device, deasserted for 32bit configuration.

1101b

PAUSED

Asserted when the Ethernet controller’s transmitter
is flow controlled.

1110b

VCC/LED_ON

Always high. Assuming no optional inversion
selected, causes output pin high / LED ON for
typical LED circuit.

Not applicable to the 82544GC/EI.

Software Developer’s Manual

291

Mode

Pneumonic

1111b

13.4.16

GND/LED_OFF

State / Event Indicated
Always low. Assuming no optional inversion
selected, causes output pin low / LED OFF for
typical LED circuit.

Packet Buffer Allocation
PBA (01000H; R/W)
This register sets the on-chip receive and transmit storage allocation ratio. The receive allocation
value is read/write for the lower seven bits. The receive allocation value must be a multiple of eight
(multiple of two for the 82547GI/EI B1 stepping). The transmit allocation is read-only and is
calculated based on PBA.RXA.
Table 13-62. PBA Register Bit Description
31

16 15
TXA

Field

Bit(s)

Initial Value

0030h1
6:0
RXA

RXA_R2

15:0 (82541xx
and 82547GI/EI)

0016h (82547EI A0-B0
steppings)
001Eh (82547GI B1
stepping)

15:7

0010h
31:16
TXA

1.
2.

292

15:00 (82541xx
and 82547GI/EI)

0
RXA

0012h (82547EI A0-B0
steppings)
000Ah (82547GI B1
stepping)

Description
Receive Packet Buffer Allocation in KB
Sets the size of the receive packet buffer.
The value of this field must be a multiple of eight.
The upper nine bits are read only as 0b. Default is
48 KB.
For the 82541xx and 82547GI/EI, the upper unused
bits are read only as 0b. The default is 48 KB for the
82541xx, 24 KB for the 82547EI, and 20 KB for the
82547GI.
Receive Packet Buffer Allocation – Upper Bits
Provides the upper nine bits of the receive packet
buffer allocation.
Read only bits - Read as 0b.
Transmit Packet Buffer Allocation
Provides the size of the transmit packet buffer. The
value is in units of KB. These bits are read only.
TXA is calculated based on RXA value: TXA = 64 –
RXA.
For the 82547GI/EI, TXA is calculated based on
RXA value: TXA = 40 – RXA.

Not applicable to the 82547GI/EI.
Not applicable to the 82541xx or 82547GI/EI.

Software Developer’s Manual

13.4.17

Interrupt Cause Read Register
ICR (000C0H; R)
This register contains all interrupt conditions for the Ethernet controller. Each time an interrupt
causing event occurs, the corresponding interrupt bit is set in this register. A PCI interrupt is
generated each time one of the bits in this register is set, and the corresponding interrupt is enabled
through the Interrupt Mask Set/Read IMS Register (see Section 13.4.20).
All register bits are cleared upon read. As a result, reading this register implicitly acknowledges
any pending interrupt events. Writing a 1b to any bit in the register also clears that bit. Writing a 0b
to any bit has no effect on that bit.
Table 13-63. ICR Register Bit Description
31 - 17

16 14
Reserved

Field

Bit(s)

0
ICR Bits

Initial
Value

Description

TXDW

Transmit Descriptor Written Back
Set when hardware processes a transmit descriptor with the RS
bit set (and possibly IDE set). If using delayed interrupts (IDE set),
the interrupt occurs after the timer expires.

TXQE

Transmit Queue Empty
Set when the last descriptor block for a transmit queue has been
used.

Link Status Change
This bit is set each time the link status changes (either from up to
down, or from down to up). This bit is affected by the internal link
indication when configured for internal PHY mode.

LSC

RXSEQ

Receive Sequence Error
In TBI mode/internal SerDes1, incoming packets with a bad
delimiter sequence set this bit. In other 802.3 implementations,
this would be classified as a framing error. A valid sequence
consists of:
idle  SOF  data  pad (opt) EOF  fill (opt)  idle.
This is a reserved bit for the 82541xx, 82547GI/EI, and 82540EP/
EM. Set to 0b.

RXDMT0

Receive Descriptor Minimum Threshold Reached
Indicates that the minimum number of receive descriptors are
available and software should load more receive descriptors.

Reserved

Reserved
Reads as 0b.

RXO

Receiver Overrun
Set on receive data FIFO overrun. Could be caused either
because there are no available receive buffers or because PCI
receive bandwidth is inadequate.

Software Developer’s Manual

293

Field

Bit(s)

Initial
Value

RXT0

Receiver Timer Interrupt
Set when the receiver timer expires.
The receiver timer is used for receiver descriptor packing. Timer
expiration flushes any accumulated descriptors and sets an
interrupt event when enabled.

Reserved

Reserved
Reads as 0b.

MDAC

MDI/O Access Complete
This bit is set when the MDI/O access is completed.

RXCFG

Receiving /C/ ordered sets
Mapped to RXCW.RxConfig.
Sets when the hardware receives configuration symbols (/C/
codes). Software should enable this interrupt when forcing link.
When the link is forced, the link partner can begin to AutoNegotiate based, due to a reset or enabling of Auto-Negotiation.
The reception of /C/ codes causes an interrupt to software and the
proper hardware configuration might be set.
See Section 13.4.14 for details. Only valid in internal SerDes
mode (TBI mode for the 82544GC/EI).
This is a reserved bit for the 82541xx and 82547GI/EI. Set to 0b.

Reserved

Reserved. Set this bit to 0b.
Not applicable to the 82544GC/EI.

PHY Interrupt (not applicable to the 82544GC/EI)
Set when the PHY generates an interrupt.
If bit 1 (PHYINT_EN) of the CTRL_EXT register (00018h) is set,
then this bit gets set.
This is a reserved bit for the 82541xx and 82547GI/EI. Set to 0b.

General Purpose Interrupt on SDP6[2]. If GPI interrupt detection is
enabled on this pin (via CTRL_EXT), this interrupt cause is set
when the SDP6[2] is sampled high.
Not applicable to the 82544GC/EI.

General Purpose Interrupt on SDP7[3]. If GPI interrupt detection is
enabled on this pin (via CTRL_EXT), this interrupt cause is set
when the SDP7[3] is sampled high.
Not applicable to the 82544GC/EI.

GPI

14:13,
11

General Purpose Interrupts (82544GC/EI only)
These bits are mapped to the upper three SDP pins when they are
configured as inputs. Refer to Section 13.4.6.

TXD_LOW2

Transmit Descriptor Low Threshold hit.
Indicates that the descriptor ring has reached the threshold
specified in the Transmit Descriptor Control register.

Small Receive Packet Detected.
Indicates that a packet of size  RSRPD.SIZE register has been
detected and transferred to host memory. The interrupt is only
asserted if RSRPD.SIZE register has a non-zero value.

PHYINT

GPI_SDP6
GPI_SDP2
(82541xx and
82547GI/EI)
GPI_SDP7
GPI_SDP3
(82541xx and
82547GI/EI)

SRPD2

294

Description

Software Developer’s Manual

Field

Reserved
1.
2.

Note:

13.4.18

Initial
Value

Bit(s)

31:17

Description
Reserved
Reads as 0b.

The 82540EP/EM, 82541xx, or 82547GI/EI do not support SerDes functionality.
Not applicable to the 82544GC/EI.

The 82547GI/EI signals interrupts over the CSA port, not a dedicated interrupt pin.

Interrupt Throttling Register1
ITR (000C4h; R/W)

Field

31 - 16

15 - 0

Reserved

INTERVAL

Initial
Value

Bit(s)

Description

INTERVAL

15:0

Minimum inter-interrupt interval. The interval is specified in 256 ns
increments. Setting this bit to 0b disables interrupt throttling logic.

Reserved

31:16

Reserved. Should be written with 0b to ensure future compatibility.

Software can use this register to pace (or even out) the delivery of interrupts to the host CPU. This
register provides a guaranteed inter-interrupt delay between interrupts asserted by the Ethernet
controller, regardless of network traffic conditions. To independently validate configuration
settings, software can use the following algorithm to convert the inter-interrupt interval value to the
common interrupts/sec performance metric:
–9

interrupts/ sec ond =  256 10

 interval 

–1

For example, if the interval is programmed to 500d, the Ethernet controller guarantees the CPU is
not interrupted by the Ethernet controller for 128 sec from the last interrupt. The maximum
observable interrupt rate from the Ethernet controller must never exceed 7813 interrupts/sec.
Inversely, inter-interrupt interval value can be calculated as:
–9

inter-interrupt interval =  256 10

 interrupts/sec 

–1

The optimal performance setting for this register is very system and configuration specific. A
initial suggested range is 651-5580 (28Bh - 15CCh).

Not applicable to the 82544GC/EI.

Software Developer’s Manual

295

13.4.19

Interrupt Cause Set Register
ICS (000C8h; W)
Software uses this register to set an interrupt condition. Any bit written with a 1b sets the
corresponding interrupt. This results in the corresponding bit being set in the Interrupt Cause Read
Register (see Section 13.4.17). A PCI interrupt is generated if one of the bits in this register is set
and the corresponding interrupt is enabled through the Interrupt Mask Set/Read Register (see
Section 13.4.20).
Bits written with 0b are unchanged.
Table 13-64. ICS Register Bit Description
31

17 16

Reserved

Field

296

ICS Bits

Initial
Value

Bit(s)

Description

TXDW

Sets Transmit Descriptor Written Back Interrupt.

TXQE

Sets Transmit Queue Empty Interrupt.

LSC

Sets Link Status Change Interrupt.

RXSEQ

Sets Receive Sequence Error Interrupt.
This is a reserved bit for the 82541xx and 82547GI/EI. Set to
0b.

RXDMT0

Sets Receive Descriptor Minimum Threshold Reached
Interrupt.

Reserved

Reserved
Should be written with 0b to ensure future compatibility.

RXO

Sets on Receiver FIFO Overrun Interrupt.

RXT0

Sets Receiver Timer Interrupt.

Reserved

Reserved
Should be written with 0b to ensure future compatibility.

MDAC

Sets MDI/O Access Complete Interrupt.

RXCFG

Sets Receiving /C/ ordered sets Interrupt.
This is a reserved bit for the 82541xx and 82547GI/EI. Set to
0b.

Reserved

Reserved
Should be written with 0b to ensure future compatibility.
Not applicable to the 82544GC/EI.

PHYINT

Sets PHY interrupt. Not applicable to the 82544GC/EI.
This is a reserved bit for the 82541xx and 82547GI/EI. Set to
0b.

Software Developer’s Manual

Field

13.4.20

Initial
Value

Bit(s)

Description

GPI

14:11

Sets General Purpose Interrupts (82544GC/EI only).

GPI

14:13

Sets General Purpose Interrupts.

TXD_LOW

Transmit Descriptor Low Threshold Hit. Not applicable to the
82544GC/EI.

SRPD

Small Receive Packet Detected and Transferred. Not applicable
to the 82544GC/EI.

Reserved

31:17

Reserved
Should be written with 0b to ensure future compatibility.

Interrupt Mask Set/Read Register
IMS (000D0h; R/W)
An interrupt is enabled if its corresponding mask bit is set to 1b, and disabled if its corresponding
mask bit is set to 0b. A PCI interrupt is generated each time one of the bits in this register is set and
the corresponding interrupt condition occurs. The occurrence of an interrupt condition is reflected
by having a bit set in the Interrupt Cause Read Register (see Section 13.4.17).
A particular interrupt can be enabled by writing a 1b to the corresponding mask bit in this register.
Any bits written with a 0b are unchanged. As a result, if software desires to disable a particular
interrupt condition that had been previously enabled, it must write to the Interrupt Mask Clear
Register (see Section 13.4.21) rather than writing a 0b to a bit in this register.
Reading this register returns bits that have an interrupt mask set.

Note:

For the 82547GI/EI, programmers need to first write (clear) the IMS and IMC registers due to a
Hub Link bus being occupied. This results in an interrupt de-assertion message that can’t to be sent
out. When a future interrupt assertion message is generated, two messages are re-ordered and sent
out. This signals APIC that the 82547GI/EI is in a de-asserted state when it is actually in an
asserted state, which causes a system dead lock. To avoid a system dead lock, first clear the IMS
and IMC registers by writing FFFFh and then re-assert IRQ enable.
Table 13-65. IMS Register Bit Description
31

17 16
Reserved

Field

0
IMS Bits

Initial
Value

Bit(s)

Description

TXDW

Sets mask for Transmit Descriptor Written Back.

TXQE

Sets mask for Transmit Queue Empty.

LSC

Sets mask for Link Status Change.

Software Developer’s Manual

297

Field

13.4.21

Initial
Value

Bit(s)

Description

RXSEQ

Sets mask for Receive Sequence Error.
This is a reserved bit for the 82541xx and 82547GI/EI. Set to 0b.

RXDMT0

Sets mask for Receive Descriptor Minimum Threshold hit.

Reserved

Reserved
Should be written with 0b to ensure future compatibility.

RXO

Sets mask for on Receiver FIFO Overrun.

RXT0

Sets mask for Receiver Timer Interrupt.

Reserved

Reserved
Should be written with 0b to ensure future compatibility.

MDAC

Sets mask for MDI/O Access Complete Interrupt.

RXCFG

Sets mask for Receiving /C/ ordered sets.
This is a reserved bit for the 82541xx and 82547GI/EI. Set to 0b.

Reserved

Reserved
Should be written with 0b to ensure future compatibility (not
applicable to the 82544GC/EI).

PHYINT

Sets mask for PHY Interrupt (not applicable to the 82544GC/EI).
This is a reserved bit for the 82541xx and 82547GI/EI. Set to 0b.

GPI

14:11

Sets mask for General Purpose Interrupts (82544GC/EI only).

GPI

14:13

Sets mask for General Purpose Interrupts.

TXD_LOW

Sets the mask for Transmit Descriptor Low Threshold hit (not
applicable to the 82544GC/EI).

SRPD

Sets mask for Small Receive Packet Detection (not applicable to
the 82544GC/EI).

Reserved

31:17

Reserved
Should be written with 0b to ensure future compatibility.

Interrupt Mask Clear Register
IMC (000D8h; W)
Software uses this register to disable an interrupt. Interrupts are presented to the bus interface only
when the mask bit is set to 1b and the cause bit set to 1b. The status of the mask bit is reflected in
the Interrupt Mask Set/Read Register (see Section 13.4.20), and the status of the cause bit is
reflected in the Interrupt Cause Read Register (see Section 13.4.17).
Software blocks interrupts by clearing the corresponding mask bit. This is accomplished by writing
a 1b to the corresponding bit in this register. Bits written with 0b are unchanged (their mask status
does not change).

298

Software Developer’s Manual

Software should write a 1b to the reserved bits to ensure future compatibility. Since this register
masks interrupts when 1b is written to the corresponding (defined) bits, then writing 1b to the
reserved bits ensures that the software is never called to handle an interrupt that the software is not
aware exists.
Note:

Reserved

Field

Initial
Value

Bit(s)

IMC Bits

Description

TXDW

Clears mask for Transmit Descriptor Written Back.

TXQE

Clears mask for Transmit Queue Empty.

LSC

Clears mask for Link Status Change.

RXSEQ

Clears mask for Receive Sequence Error.
This is a reserved bit for the 82541xx and 82547GI/EI. Set to 0b.

RXDMT0

Sets mask for Receive Descriptor Minimum Threshold hit.

Reserved

Reserved: Should be written with 1b to ensure future
compatibility.

RXO

Clears mask for on Receiver FIFO Overrun.

RXT0

Clears mask for Receiver Timer Interrupt.

Reserved

Reserved: Should be written with 1b to ensure future
compatibility.

MDAC

Clears mask for MDI/O Access Complete Interrupt.

RXCFG

Clears mask for Receiving /C/ ordered sets.
This is a reserved bit for the 82541xx and 82547GI/EI. Set to 0b.

Reserved

Reserved: Should be written with 1b to ensure future
compatibility (not applicable to the 82544GC/EI).

PHYINT

Clears PHY Interrupts (not applicable to the 82544GC/EI).
This is a reserved bit for the 82541xx and 82547GI/EI. Set to 0b.

GPI

14:11

Clears General Purpose Interrupts (82544GC/EI only).

GPI

14:13

Clears General Purpose Interrupts.

Software Developer’s Manual

299

Field

13.4.22

Initial
Value

Bit(s)

Description

TXD_LOW

Clears the mask for Transmit Descriptor Low Threshold hit (not
applicable to the 82544GC/EI).

SRPD

Clears mask for Small Receive Packet Detect Interrupt (not
applicable to the 82544GC/EI).

Reserved

31:17

Reserved
Should be written with 1b to ensure future compatibility.

Receive Control Register
RCTL (00100h; R/W)
This register controls all Ethernet controller receiver functions.
Table 13-67. RCTL Register Bit Description
31

Reserved

Field

Reserved

SBP

UPE

300

0
Receive Control Bits

Bit(s)

Initial
Value

Description

Reserved
Write to 0b for future compatibility.

Receiver Enable
The receiver is enabled when this bit is 1b. Writing this bit to 0b
stops reception after receipt of any in-progress packets. Data
remains in the receive FIFO until the device is re–enabled.
Disabling or re-enabling the receiver does not reinitialize the packet
filter logic that demarcates packet start and end locations in the
FIFO; Therefore the receiver must be reset before re-enabling it.

Store Bad Packets
0b = do not store.
1b = store bad packets.
When set, the Ethernet controller stores bad packets (CRC error,
symbol error, sequence error, length error, alignment error, short
packets or where carrier extension or RX_ERR errors) that pass the
filter function in host memory. When the Ethernet controller is in
promiscuous mode, and SBP is set, it might possibly store all
packets.

Unicast Promiscuous Enabled
0b = Disabled.
1b = Enabled.
When set, passes without filtering out all received unicast packets.
Otherwise, the Ethernet controller accepts or rejects unicast
packets based on the received packet destination address match
with 1 of the 16 stored addresses.

Software Developer’s Manual

Field

MPE

LPE

LBM

Bit(s)

7:6

Initial
Value

Description

Multicast Promiscuous Enabled
0b = Disabled.
1b = Enabled.
When set, passes without filtering out all received multicast packets.
Otherwise, the Ethernet controller accepts or rejects a multicast
packet based on its 4096-bit vector multicast filtering table.

Long Packet Reception Enable
0b = Disabled.
1b = Enabled.
LPE controls whether long packet reception is permitted. When LPE
is cleared, the Ethernet controller discards packets longer than
1522 bytes. When LPE is set, the Ethernet controller discards
packets that are longer than 16384 bytes.
For the 82541xx and 82547GI/EI, packets larger than 2 KB require
full duplex operation.

Loopback mode.
Controls the loopback mode of the Ethernet controller.
00b = No loopback.
01b = Undefined.
10b = Undefined.
11b = PHY or external SerDes loopback.
All loopback modes are only allowed when the Ethernet controller is
configured for full-duplex operation. Receive data from transmit
data looped back internally to the SerDes or internal PHY. In TBI
mode (82544GC/EI), the EWRAP signal is asserted.
Note: The 82540EP/EM, 82541xx, and 82547GI/EI do not support
SerDes functionality.

RDMTS

9:8

Receive Descriptor Minimum Threshold Size
The corresponding interrupt ICR.RXDMT0 is set each time the
fractional number of free descriptors becomes equal to RDMTS.
The following table lists which fractional values correspond to
RDMTS values. The size of the total receiver circular descriptor
buffer is set by RDLEN. See Section 13.4.27 for details regarding
RDLEN.
00b = Free Buffer threshold is set to 1/2 of RDLEN.
01b = Free Buffer threshold is set to 1/4 of RDLEN.
10b = Free Buffer threshold is set to 1/8 of RDLEN.
11b = Reserved.

Reserved

11:10

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

13:12

Multicast Offset
The Ethernet controller is capable of filtering multicast packets
based on 4096-bit vector multicast filtering table. The MO
determines which bits of the incoming multicast address are used in
looking up the 4096-bit vector.
00b = bits [47:36] of received destination multicast address.
01b = bits [46:35] of received destination multicast address.
10b = bits [45:34] of received destination multicast address.
11b = bits [43:32] of received destination multicast address.

Reserved

Reserved
Should be written with 0 to ensure future compatibility
Reads as 0

Software Developer’s Manual

301

Field

BAM

BSIZE

VFE

17:16

Initial
Value

Description

Broadcast Accept Mode.
0 = ignore broadcast; 1 = accept broadcast packets.
When set, passes and does not filter out all received broadcast
packets. Otherwise, the Ethernet controller accepts, or rejects a
broadcast packet only if it matches through perfect or imperfect
filters.

Receive Buffer Size
Controls the size of the receive buffers, allowing the software to
trade off between system performance and storage space. Small
buffers maximize memory efficiency at the cost of multiple
descriptors for bigger packets.
RCTL.BSEX = 0b:
00b = 2048 Bytes.
01b = 1024 Bytes.
10b = 512 Bytes.
1b1 = 256 Bytes.
RCTL.BSEX = 1b:
00b = Reserved; software should not program this value.
01b = 16384 Bytes.
10b = 8192 Bytes.
11b = 4096 Bytes.

VLAN Filter Enable1
0b = Disabled (filter table does not decide packet acceptance).
1b = Enabled (filter table decides packet acceptance for 802.1Q
packets).
Three bits control the VLAN filter table. RCTL.VFE determines
whether the VLAN filter table participates in the packet acceptance
criteria. RCTL.CFIEN and RCTL.CFI are used to decide whether
the CFI bit found in the 802.1Q packet’s tag should be used as part
of the acceptance criteria.

CFIEN

Canonical Form Indicator Enable
0b = Disabled (CFI bit found in received 802.1Q packet’s tag is not
compared to decide packet acceptance).
1b = Enabled (CFI bit found in received 802.1Q packet’s tag must
match RCTL.CFI to accept 802.1Q type packet.

CFI

Canonical Form Indicator bit value
If RCTL.CFIEN is set, then 802.1Q packets with CFI equal to this
field is accepted; otherwise, the 802.1Q packet is discarded.

Reserved

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

Discard Pause Frames
0 = incoming pause frames subject to filter comparison.
1 = incoming pause frames are filtered out even if they match filter
registers.
DPF controls the DMA function of flow control PAUSE packets
addressed to the station address (RAH/L[0]). If a packet is a valid
flow control packet and is addressed to the station’s address, it is
not transferred to host memory if RCTL.DPF = 1b. However, it is
transferred when DPF is set to 0b.

DPF

302

Bit(s)

Software Developer’s Manual

Field

Initial
Value

Bit(s)

Description

PMCF

Pass MAC Control Frames
0b = Do not (specially) pass MAC control frames.
1b = Pass any MAC control frame (type field value of 8808h) that
does not contain the pause opcode of 0001h.
PMCF controls the DMA function of MAC control frames (other than
flow control). A MAC control frame in this context must be
addressed to either the MAC control frame multicast address or the
station address, match the type field and NOT match the PAUSE
opcode of 0001h. If PMCF = 1b then frames meeting this criteria are
transferred to host memory. Otherwise, they are filtered out.

Reserved

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

BSEX

Buffer Size Extension
When set to one, the original BSIZE values are multiplied by 16.
Refer to the RCTL.BSIZE bit description.

SECRC

Strip Ethernet CRC from incoming packet
0b = Do not strip CRC field.
1b = Strip CRC field.
Controls whether the hardware strips the Ethernet CRC from the
received packet. This stripping occurs prior to any checksum
calculations. The stripped CRC is not transferred to host memory
and is not included in the length reported in the descriptor.

Reserved

31:27

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

Not applicable to the 82541ER.

Software Developer’s Manual

303

13.4.23

Flow Control Receive Threshold Low
FCRTL (02160h; R/W)
This register contains the receive threshold used to determine when to send an XON packet. It
counts in units of bytes. Each time the receive FIFO crosses the receive high threshold
FCRTH.RTH (filling up), and then crosses the receive low threshold FCRTL.RTL, with
FCRTL.XONE enabled, hardware transmits an XON frame.
Flow control reception/transmission are negotiated capabilities by the Auto-Negotiation process.
When the Ethernet controller is manually configured, flow control operation is determined by the
CTRL.RFCE and CTRL.TFCE bits.
Table 13-68. FCRTL Register Bit Description
31

16 15

XONE1
1.

3 2
RTL

0
0

82544GC/EI, 82541xx, and 82547GI/EI only.

Field

Reserved

Bit(s)

2:0

Initial
Value

Description

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

RTL

15:3

Receive Threshold Low.
FIFO low water mark for flow control transmission. Each time the
receive FIFO crosses the receive high threshold FCRTH.RTH and
later crosses the receive low threshold FCRTL.RTL with
FCRTL.XONE enabled, hardware transmits an XON frame (a
PAUSE frame with a timer value of 0b).
RTL is provided in units of 8 bytes.

Reserved

31:16

Reserved
Should be written with 0b for future compatibility.
Reads as 0b.

XON Enable (82544GC/EI, 82541xx, and 82547GI/EI only)
0b = Disabled.
1b = Enabled.
When set, enables the Ethernet controller to transmit XON packets
based on receive FIFO crosses FCRTL.RTL threshold value, or
based on external pins XOFF and XON.

XONE

304

Reserved

Software Developer’s Manual

13.4.24

Flow Control Receive Threshold High
FCRTH (02168h; R/W)
This register contains the receive threshold used to determine when to send an XOFF packet. It
counts in units of bytes. Each time the receive FIFO reaches the fullness indicated by FCRTH,
hardware transmits a PAUSE frame if the transmission of flow control frames is enabled
(CTRL.TFCE).
Flow control reception/transmission are negotiated capabilities by the Auto-Negotiation process.
When the Ethernet controller is manually configured, flow control operation is determined by the
CTRL.RFCE & CTRL.TFCE bits.
Table 13-69. FCRTH Register Bit Description
31

16 15

XFCE1
1.

Reserved

3 2
RTH

0
0

82544GC/EI only.

Field

Reserved

Initial
Value

Bit(s)

2:0

Description

Reserved
Should be written with 0 for future compatibility
Reads as 0

RTH

15:3

Receive Threshold High.
FIFO high water mark for flow control transmission.
Each time the receive FIFO reaches the fullness indicated by RTH,
the Ethernet controller transmits a Pause packet if enabled to do so.

Reserved

31:16

Reserved
Should be written with 0b for future compatibility.
Reads as 0b.

External Flow Control Enabled (82544GC/EI only)
0b = Disabled.
1b = Enabled.
Allows the Ethernet controller to send XOFF and XON frames based
on external pins XOFF and XON. The transmission of pause frames
must be also enabled through the CTRL.TFCE control bit. When the
XOFF signal is asserted high, the Ethernet controller transmits a
single XOFF frame. The assertion of XON (after deassertion of
XOFF) initiates an XON frame transmission, if enabled by
FCRTL.XONE. The assertion/deassertion of XON is required
between assertions of XOFF in order to send another XOFF frame.
This behavior also provides a built-in hysteresis mechanism.

XFCE

Software Developer’s Manual

305

13.4.25

Receive Descriptor Base Address Low
RDBAL (02800h;R/W)
This register contains the lower bits of the 64-bit descriptor base address. The four low-order
register bits are always ignored. The Receive Descriptor Base Address must point to a 16-byte
aligned block of data.
Table 13-70. RDBAL Register Bit Description
31

4
RDBAL

Field

13.4.26

0
0

Initial
Value

Bit(s)

Description

3:0

Ignored on writes. Returns 0b on reads.

RDBAL

31:4

Receive Descriptor Base Address Low.

Receive Descriptor Base Address High
RDBAH (02804h; R/W)
This register contains the upper 32 bits of the 64-bit Descriptor Base Address.
Table 13-71. RDBAH Register Bit Description
31

0
RDBAH

Field
RDBAH

306

Initial
Value

Bit(s)
31:0

Description
Receive Descriptor Base Address [63:32]

Software Developer’s Manual

13.4.27

Receive Descriptor Length
RDLEN (02808h; R/W)
This register determines the number of bytes allocated to the circular receive descriptor buffer. This
value must be 128-byte aligned (the maximum cache line size). Since each descriptor is 16 bytes in
length, the total number of receive descriptors is always a multiple of eight.
Table 13-72. RDLEN Register Bit Description
31

20 19

7 6

Reserved

Field

13.4.28

LEN

Initial
Value

Bit(s)

0
0

Description

Zero

6:0

Zero value
Ignore on write. Reads back as 0b.

LEN

19:7

Receive Descriptor length
Provides the number of receive descriptors (in a multiple of eight).

Reserved

31:20

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

Receive Descriptor Head
RDH (02810h; R/W)
This register contains the head pointer for the receive descriptor buffer. The register points to a 16byte datum. Hardware controls the pointer. The only time that software should write to this register
is after a reset (RCTL.RST or CTRL.RST) and before enabling the receiver function (RCTL.EN).
If software were to write to this register while the receive function was enabled, the on-chip
descriptor buffers can be invalidated and other indeterminate operations might result. Reading the
descriptor head to determine which buffers are finished is not reliable.
Table 13-73. RDH Register Bit Description
31

16 15
Reserved

Field

Bit(s)

0
RDH

Initial
Value

Description

RDH

15:0

Receive Descriptor Head.

Reserved

31:16

Reserved. Should be written with 0b for future compatibility.
Reads as 0b.

Software Developer’s Manual

307

13.4.29

Receive Descriptor Tail
RDT (02818h;R/W)
This register contains the tail pointers for the receive descriptor buffer. The register points to a 16byte datum. Software writes the tail register to add receive descriptors to the hardware free list for
the ring.
Table 13-74. RDT Register Bit Description
31

16 15
Reserved

Field

13.4.30

Bit(s)

0
RDT

Initial
Value

Description

RDT

15:0

Receive Descriptor Tail.

Reserved

31:16

Reserved
Reads as 0b.
Should be written with 0b for future compatibility.

Receive Delay Timer Register
RDTR (02820h; R/W)
This register is used to delay interrupt notification for the receive descriptor ring. Delaying
interrupt notification helps maximize the number of receive packets serviced by a single interrupt.

Warning:

It is strongly recommended that the Delay Timer field of this register not be used. For any
application requiring an interrupt moderation mechanism, it is recommended that the Interrupt
Throttling Register (ITR) be used instead. ITR provides a more direct interrupt solution than
RDTR. In addition, Intel software device drivers use ITR instead of RDTR. Refer to the 82546EB
Gigabit Ethernet Controller Specification Update or 82546GB Gigabit Ethernet Controller
Specification Update for additional details.
Table 13-75. RDTR Register Bit Description
31

16 15

FPD

Reserved

Field

308

Bit(s)

0
Delay Timer

Initial
Value

Description

Delay Timer

15:0

Receive delay timer measured in increments of 1.024 s.

Reserved

30:16

Reserved. Reads as 0b.

FPD

Flush partial descriptor block when set to 1b; ignore otherwise.
Reads 0b (self-clearing).

Software Developer’s Manual

This feature operates by initiating a countdown timer upon successfully receiving each packet to
system memory. If a subsequent packet is received BEFORE the timer expires, the timer is reinitialized to the programmed value and re-starts its countdown. If the timer expires due to NOT
having received a subsequent packet within the programmed interval, pending receive descriptor
writebacks are flushed and a receive timer interrupt is generated.
Setting the value to 0b represents no delay from a receive packet to the interrupt notification, and
results in immediate interrupt notification for each received packet.
Writing this register with FPD set initiates an immediate expiration of the timer, causing a
writeback of any consumed receive descriptors pending writeback, and results in a receive timer
interrupt in the ICR.
Receive interrupts due to a Receive Absolute Timer (RADV) expiration cancels a pending RDTR
interrupt. The RDTR countdown timer is reloaded but halted, so as to avoid generation of a
spurious second interrupt after the RADV has been noted, but might be restarted by a subsequent
received packet.

13.4.31

Receive Interrupt Absolute Delay Timer1
RADV (0282Ch; RW)

Warning:

It is strongly recommended that the Delay Timer field of this register not be used. For any
application requiring an interrupt moderation mechanism, it is recommended that the Interrupt
Throttling Register (ITR) be used instead. ITR provides a more direct interrupt solution than
RADV. In addition, Intel software device drivers use ITR instead of RADV. Refer to the 82546EB
Gigabit Ethernet Controller Specification Update or 82546GB Gigabit Ethernet Controller
Specification Update for additional details.

16
Reserved

Field

Bit(s)

0
Delay Timer

Initial
Value

Description

Delay Timer

15:0

Receive absolute delay timer measured in increments of 1.024 s (0b =
disabled).

Reserved

31:16

Reserved. Reads as 0b.

If the packet delay timer is used to coalesce receive interrupts, the Ethernet controller ensures that
when receive traffic abates, an interrupt is generated within a specified interval of no receives.
During times when receive traffic is continuous, it may be necessary to ensure that no receive
remains unnoticed for too long an interval. This register can be used to ENSURE that a receive
interrupt occurs at some predefined interval after the first packet is received.

Not applicable to the 82544GC/EI.

Software Developer’s Manual

309

When this timer is enabled, a separate absolute countdown timer is initiated upon successfully
receiving each packet to system memory. When this absolute timer expires, pending receive
descriptor writebacks are flushed and a receive timer interrupt is generated.
Setting this register to 0b disables the absolute timer mechanism (the RDTR register should be
used with a value of 0b to cause immediate interrupts for all receive packets).
Receive interrupts due to a Receive Packet Timer (RDTR) expiration cancels a pending RADV
interrupt. If enabled, the RADV countdown timer is reloaded but halted, so as to avoid generation
of a spurious second interrupt after the RDTR has been noted.

13.4.32

Receive Small Packet Detect Interrupt1
RSRPD (02C00h; R/W)

Reserved

Field

13.4.33

SIZE

Initial
Value

Bit(s)

Description

SIZE

11:0

If the interrupt is enabled, any receive packet of size  SIZE asserts
an Interrupt. SIZE is specified in bytes and includes the headers and
the CRC. It does not include the VLAN header in size calculation if it
is stripped.

Reserved

31:12

Reserved. Reads as 0b.

Transmit Control Register
TCTL (00400h;R/W)
This register controls all transmit functions for the Ethernet controller.

Not applicable to the 82544GC/EI.

310

Software Developer’s Manual

Table 13-76. TCTL Register Bit Description
31

26 25
Reserved

22 21

12 11

CNTL Bits

Field

Reserved

Bit(s)

4 3

COLD

Initial
Value

0
CNTL Bits

Description

Reserved
Write as 0b for future compatibility.

Transmit Enable
The transmitter is enabled when this bit is set to 1b. Writing 0b to
this bit stops transmission after any in progress packets are sent.
Data remains in the transmit FIFO until the device is re-enabled.
Software should combine this operation with reset if the packets in
the TX FIFO should be flushed.

Reserved

Reserved
Reads as 0b.
Should be written to 0b for future compatibility.

Pad Short Packets
0b = Do not pad.
1b = Pad short packets.
Padding makes the packet 64 bytes long. The padding content is
data.
When the Pad Short Packet feature is disabled, the minimum
packet size the Ethernet controller can transfer to the host is 32
bytes long.
This feature is not the same as Minimum Collision Distance
(TCTL.COLD).

Collision Threshold
This determines the number of attempts at re-transmission prior to
giving up on the packet. The Ethernet back–off algorithm is
implemented and clamps to the maximum value after 16 retries. It
only has meaning in half-duplex operation. Recommended value –
0Fh.

PSP

COLD

11:4

21:12

Collision Distance
Specifies the minimum number of byte times that must elapse for
proper CSMA/CD operation. Packets are padded with special
symbols, not valid data bytes. Hardware checks this value and
padded packets even in full-duplex operation.
Recommended value:
Half-Duplex – 512-byte time (200h)
Full-Duplex – 64-byte time (40h)
Note: 10/100 half-duplex - 64 - 68 (40h to 44h) byte times for the
82541xx and 82547GI/EI only.

SWXOFF

Software Developer’s Manual

Software XOFF Transmission
When set to 1b, the Ethernet controller schedules the transmission
of an XOFF (PAUSE) frame using the current value of the PAUSE
timer (FCTTV.TTV). This bit self-clears upon transmission of the
XOFF frame. This bit is valid only in Full-Duplex mode of
operation. Software should not set this bit while the Ethernet
controller is configured for half-duplex operation.

311

Field

Initial
Value

Bit(s)

Reserved

RTLC

Description

Reserved
Read as 0b.
Should be written with 0b for future compatibility.

Re-transmit on Late Collision
When set, enables the Ethernet controller to re-transmit on a late
collision event.
The collision window is speed dependent. For example, 64 bytes
for 10/100 Mb/s and 512 bytes for 1000Mb/s operation. If a late
collision is detected when this bit is disabled, the transmit function
assumes the packet is successfully transmitted.
The RTLC bit is ignored in full-duplex mode.

NRTU1
Reserved

No Re-transmit on underrun (82544GC/EI only)
If this bit is set, the 82544GC/EI does not re-transmit packets that
initially had an underrun.
This function is accomplished by waiting for the entire packet to be
buffered in the transmit FIFO before the controller attempts to retransmit a packet that previously encountered an underrun. This
operation guarantees only one underrun can occur per packet.
This is a reserved bit for all other Ethernet controllers and should
be written with 0b for future compatibility.

Reserved

31:26

Reserved
Read as 0.
Should be written with 0b for future compatibility.

82544GC/EI only.

For the 82541xx and 82547GI/EI, carrier extension (through the TCTLCOLD field) provides a
method to increase the duration of the carrier event to a minimum usable duration in order to meet
a 200 m collision domain objective, even though half-duplex operation is impractical at Gigabit.
Packets that are signaled from the CSMA/CD layer that do not meet the minimum slot time of 512
bytes have extension bytes appended to them in order to meet this minimum slot time requirement.
The extension bytes are defined within the context of the frame encapsulation discussion of the
802.3z standard and are recognized by 802.3z compliant devices. Refer to Figure 13-1

Preamble

SFD

T/L

Data/Pad

FCS

Extension

Minimum Frame Size
Slot Time
Duration of Carrier Event

Figure 13-1. Carrier Extended Frame Format (82541xx and 82547GI/EI)

13.4.34

Transmit IPG Register
TIPG (00410;R/W)
This register controls the IPG (Inter Packet Gap) timer for the Ethernet controller.

312

Software Developer’s Manual

Table 13-77. TIPG Register Bit Description
31

Reserved

IPGR2

Field

IPGR1

Initial
Value

Bit(s)

0
IPGT

Description
IPG Transmit Time
Specifies the IPG time for back-to-back packet transmissions
Measured in increments of the MAC clock:
• 8 ns MAC clock when operating @ 1 Gbps.
• 80 ns MAC clock when operating @ 100 Mbps.

IPGT

IPGR1

9:0

19:10

• 800 ns MAC clock when operating @ 10 Mbps.
To calculate the actual IPG value for TBI applications, a value of
10 (6 for the 82544GC/EI) should be added to the IPGT value.
For the IEEE 802.3 standard IPG value of 96-bit time, the value
that should be programmed into IPGT is 10 (6 for the 82544GC/
EI).
To calculate the IPG value for 10/100/1000BASE-T applications,
a value of four should be added to the IPGT value as four clocks
are used by the MAC as internal overhead. The value that
should be programmed into IPGT is 10 (8 for the 82544GC/EI).
These values are recommended to assure that the minimum
IPG gap is met under all synchronization conditions.
IPG Receive Time 1
Specifies the length of the first part of the IPG time for non backto-back transmissions. During this time, the internal IPG counter
restarts if any carrier event occurs. Once the time specified in
IPGR1 has elapsed, carrier sense does not affect the IPG
counter.
According to the IEEE802.3 standard, IPGR1 should be 2/3 of
IPGR2 value.
Measured in increments of the MAC clock:
• 8 ns MAC clock when operating @ 1 Gbps (82544GC/EI
only).
• 80 ns MAC clock when operating @ 100 Mbps
• 800 ns MAC clock when operating @ 10 Mbps.
For IEEE 802.3 minimum IPG value of 96-bit time, the value that
should be programmed into IPGR1 is eight.
IPGR1 is significant only in half-duplex mode of operation.

Software Developer’s Manual

313

Field

Initial
Value

Bit(s)

Description
IPG Receive Time 2
Specifies the total length of the IPG time for non back-to-back
transmissions.
Measured in increments of the MAC clock:
• 8 ns MAC clock when operating @ 1 Gbps (82544GC/EI
only).
• 80 ns MAC clock when operating @ 100 Mbps

13.4.35

IPGR2

29:20

Reserved

31:30

• 800 ns MAC clock when operating @ 10 Mbps.
In order to calculate the actual IPG value, a value of six should
be added to the IPGR2 value as six MAC clocks are used by the
MAC for synchronization and internal engines.
For the IEEE 802.3 standard IPG value of 96-bit time, the value
that should be programmed into IPGR2 is six (total IPG delay of
12 MAC clock cycles) According to the IEEE802.3 standard,
IPGR1 should be 2/3 of IPGR2 value.
IPGR2 is significant only in half-duplex mode of operation.
Reserved
Read as 0b.
Should be written with 0b for future compatibility.

Adaptive IFS Throttle - AIT
AIFS (00458;R/W)
This register throttles back-to-back transmissions in the transmit packet buffer and delays their
transfer to the CSMA/CD transmit function. As a result, it can be used to delay the transmission of
back-to-back packets on the wire.
For the 82544GC/EI, this register can be used to increase the IPG value between transmitting
back-to-back packets on the wire and between frames while in half-duplex mode.

314

Software Developer’s Manual

Table 13-78. AIFS Register Bit Description
31

Reserved

Field

13.4.36

Bit(s)

Adaptive IFS

Initial
Value

Description

AIFS

15:0

Adaptive IFS Value (82544GC/EI only)
Adaptive IFS throttles back-to-back transmissions in the
transmit packet buffer and delays their transfer to the CSMA/CD
transmit function. Normally, this register should be set to 0b.
However, if additional delay is desired between back-to-back
transmit packets, then this register can be set with a value
greater than zero (0). This feature can be helpful in high collision
half-duplex environments.
In order for AIFS to take effect it should be larger than the
minimum IFS value defined in IEEE 802.3 standard.
AIFS has no effect on transmissions that occur immediately
after receives or transmissions that are not back-to-back. In
addition, it has no effect on re-transmission timing (retransmission after collisions).
The AIFS programming value is in units of 8 ns (TX_CLK), and
is 16 bits wide, thus providing greater flexibility and maximum
delay value comparing TIPG.IPGT. The AIFS value is additive to
the TIPG.IPGT value.
Adaptive IFS Value (all remaining Ethernet controllers)
This value is in units of 8 ns.

Reserved

31:16

Reserved
Should be written with 0b.

Transmit Descriptor Base Address Low
TDBAL (03800h; R/W)
This register contains the lower bits of the 64-bit transmit Descriptor base address. The base
register indicates the start of the circular transmit descriptor queue. Since each descriptor is 16 bits
in length, the lower four bits are ignored as the Transmit Descriptor Base Address must point to a
16-byte aligned block of data.
Table 13-79. TDBAL Register Bit Description
31

4 3
TDBAL

Software Developer’s Manual

0
0

315

Field

13.4.37

Initial
Value

Bit(s)

Description

ZERO

3:0

Zero Value
This field is ignored on write and reads as 0b.

TDBAL

31:4

Transmit Descriptor Base Address Low [31:4]
This register indicates lower 32 bits of the start address for the
transmit descriptor ring buffer.

Transmit Descriptor Base Address High
TDBAH (03804h; R/W)
This register contains the upper 32 bits of the 64-bit transmit Descriptor base address.
Table 13-80. TDBAH Register Bit Description
31

0
TDBAH

Field

TDBAH

13.4.38

Initial
Value

Bit(s)

31:0

Description
Transmit Descriptor Base Address [63:32]
This register indicates upper 32 bits of the start address for the
transmit descriptor ring buffer.

Transmit Descriptor Length
TDLEN (03808h; R/W)
This register determines the number of bytes allocated to the transmit descriptor circular buffer.
This value must be a multiple of 128 bytes (the maximum cache line size). Since each descriptor is
16 bits in length, the total number of receive descriptors is always a multiple of eight.
Table 13-81. TDLEN Register Bit Description
31

20 19
Reserved

Field
ZERO

316

Bit(s)
6:0

7 6
LEN

Initial
Value
0b

0
0

Description
Ignore on write. Reads back as 0b.

Software Developer’s Manual

13.4.39

LEN

19:7

Descriptor Length.
Number of bytes allocated to the transmit descriptor circular
buffer.

Reserved

31:20

Reserved
Reads as 0b. Should be written with 0b for future compatibility.

Transmit Descriptor Head
TDH (03810h; R/W)
This register contains the head pointer for the transmit descriptor ring. It holds a value that is an
offset from the base, and indicates the in–progress descriptor. It points to a 16-byte datum.
Hardware controls this pointer. The only time that software should write to this register is after a
reset (TCTL.RST or CTRL.RST) and before enabling the transmit function (TCTL.EN). If
software were to write to this register while the transmit function was enabled, the on-chip
descriptor buffers can be invalidated and indeterminate operation can result. Reading the transmit
descriptor head to determine which buffers have been used (and can be returned to the memory
pool) is not reliable.
Table 13-82. TDH Register Bit Description
31

16 15
Reserved

Software Developer’s Manual

0
TDH

317

Field

13.4.40

Initial
Value

Bit(s)

Description

TDH

15:0

Transmit Descriptor Head

Reserved

31:16

Reserved
Reads as 0b. Should be written with 0b for future compatibility.

Transmit Descriptor Tail
TDT (03818h; R/W)
This register contains the tail pointer for the transmit descriptor ring. It holds a value that is an
offset from the base, and indicates the location beyond the last descriptor hardware can process.
This is the location where software writes the first new descriptor. It points to a 16-byte datum.
Software writes the tail pointer to add more descriptors to the transmit ready queue. Hardware
attempts to transmit all packets referenced by descriptors between head and tail.
Table 13-83. TDT Register Bit Description
31

16 15
Reserved

Field

13.4.41

Bit(s)

0
TDT

Initial
Value

Description

TDT

15:0

Transmit Descriptor Tail

Reserved

31:16

Reserved
Reads as 0b. Should be written to 0b for future compatibility.

Transmit Interrupt Delay Value
TIDV (03820h; R/W)
This register contains the transmit interrupt delay value. It determines the amount of time that
elapses between writing back a descriptor that has Report Status (RS) [or Report Packet Sent (RPS)
for the 82544GC/EI] and Interrupt Delay Enable (IDE) set in the transmit descriptor and when the
IMS.TXDW (Transmit Descriptor Written Back) interrupt bit is set. Counts are in units of 1.024
s. A value of 0b is not allowed. The TIDV value is used to force timely write–back of descriptors
as well. The first packet after timer initialization starts the timer. Timer expiration flushes any
accumulated descriptors and sets an interrupt event (TXDW).

318

Software Developer’s Manual

Table 13-84. TIDV Register Bit Description
31

16 15
Reserved

Field

13.4.42

Initial
Value

Bit(s)

0
IDV

Description

IDV

15:0

Interrupt Delay Value
Counts in units of 1.024 s. A value of 0bis not allowed.

Reserved

31:16

Reserved
Reads as 0b. Should be written to 0b for future compatibility.

TX DMA Control (82544GC/EI only)
TXDMAC (03000h; R/W)
This register controls the transmit DMA pre-fetching and preemption abilities.
Table 11-85. TXDMAC Register Bit Description
31

1
Reserved

Field

13.4.43

Initial
Value

Bit(s)

0
DPP

Description

DPP

Disable packet prefetching
When set, prevents the Ethernet controller from starting a
transmit descriptor data fetch before it has finished processing
the previous descriptor. In general, performance increases
when this bit is set to 0b.

Reserved

31:1

Reserved
Reads as 0b. Should be written to 0b for future compatibility.

Transmit Descriptor Control
TXDCTL (03828h; R/W)
This register controls the fetching and write back of transmit descriptors. The three threshold
values provided are used to determine when descriptors are read from and written to host memory.
The values can be in units of cache lines or descriptors (each descriptor is 16 bytes).

Software Developer’s Manual

319

Table 13-86. TXDCTL Register Bit Description
31

LWTHRESH
RSV1
1.

GRAN

22 21

RSV

16 15
WTHRESH

14 13

RSV

8 7
HTHRESH

6 5
RSV

0
PTHRESH

82544GC/EI only.

Field

Bit(s)

Initial
Value

Description

PTHRESH

5:0

Prefetch Threshold
Used to control when a pre-fetch of descriptors is considered.
This threshold refers to the number of valid, unprocessed
transmit descriptors the Ethernet controller has in its on-chip
buffer. If this number drops below PTHRESH, the algorithm
considers prefetching descriptors from host memory. This fetch
does not happen unless there are at least TXDCTL.HTHRESH
valid descriptors in host memory to fetch. Value of PTHRESH
can be in either cache line units, or based on number of
descriptors based on TXDCTL.GRAN.

Reserved

7:6

Reserved
Reads as 0b. Should be written as 0b for future compatibility.

HTHRESH

15:8

Host Threshold
Provides the threshold of the valid descriptors in host memory A
descriptor prefetch is performed each time enough valid
descriptors (TXDCTL.HTHRESH) are available in host memory,
no other DMA activity of greater priority is pending (descriptor
fetches and write backs or packet data transfers) and the number
of transmit descriptors the Ethernet controller has on its on-chip
buffers drops below TXDCTL.PTHRESH. The value of
HTHRESH can be in either cache line units, or based on number
of descriptors based on TXDCTL.GRAN.

Reserved

15:14

Reserved
Reads as 0b. Should be written as 0b for future compatibility.

Write Back Threshold
WTHRESH controls the write back of processed transmit
descriptors. This threshold refers to the number of transmit
descriptors in the Ethernet controller’s on-chip buffer which are
ready to be written back to host memory. In the absence of
external events (explicit flushes), the write back occurs only after
more than WTHRESH descriptors are available for write back.
WTHRESH must contain a non-zero value to take advantage of
the write back bursting capabilities of the Ethernet controller. A
value of 0b causes the descriptors to be written back as soon as
they are processed.
The value of WTHRESH can be in either cache line units, or
based on number of descriptors based on RXDCTL.GRAN.

WTHRESH

320

21:16

Software Developer’s Manual

Field

Reserved

Initial
Value

Bit(s)

23:22

Description

Reserved
Reads as 0b. Should be written as 0b for future compatibility.

GRAN

Granularity
Set the values of PTHRESH, HTHRESH and WTHRESH in units
of cache lines or descriptors (each descriptor is 16 bytes)
1b = Descriptor granularity.
0b = Cache line granularity.

LWTHRESH1

31:25

Transmit descriptor Low Threshold
Interrupt asserted when the number of descriptors pending
service in the transmit descriptor queue (processing distance
from the TDT) drops below this threshold.

Not applicable to the 82544GC/EI.

Since write back of transmit descriptors is optional (under the control of RS bit in the descriptor),
not all processed descriptors are counted with respect to WTHRESH. Descriptors start accumulating after a descriptor with RS (or RPS for the 82544GC/EI) is set. Furthermore, with transmit
descriptor bursting enabled, some descriptors are written back that did not have RS (or RPS for the
82544GC/EI) set in their respective descriptors.
LWTHRESH (not applicable to the 82544GC/EI) controls the number of pre-fetched transmit
descriptors at which a transmit descriptor-low interrupt (ICR.TXD_LOW) is reported. This can
enable software to operate more efficiently by maintaining a continuous addition of transmit work,
interrupting only when the hardware nears completion of all submitted work. LWTHRESH
specifies a multiple of eight descriptors. An interrupt is asserted when the number of descriptors
available transitions from (threshold level=8*LWTHRESH)+1 to (threshold
level=8*LWTHRESH). Setting this value to 0b causes this interrupt to be generated only when the
transmit descriptor cache becomes completely empty.

13.4.44

Transmit Absolute Interrupt Delay Value1
TADV (0382Ch; RW)

Reserved

Field

IDV

Bit(s)

Initial
Value

Description

IDV

15:0

Interrupt Delay Value
Counts in units of 1.024 s. (0b = disabled)

Reserved

31:16

Reads as 0b. Should be written to 0b for future compatibility.

Not applicable to the 82544GC/EI.

Software Developer’s Manual

321

The transmit interrupt delay timer (TIDV) can be used to coalesce transmit interrupts. However, it
might be necessary to ensure that no completed transmit remains unnoticed for too long an interval
in order ensure timely release of transmit buffers. This register can be used to ENSURE that a
transmit interrupt occurs at some predefined interval after a transmit is completed. Like the
delayed-transmit timer, the absolute transmit timer ONLY applies to transmit descriptor operations
where (a) interrupt-based reporting is requested (RS set) and (b) the use of the timer function is
requested (IDE is set).
This feature operates by initiating a countdown timer upon successfully transmitting the buffer.
When the timer expires, a transmit-complete interrupt (ICR.TXDW) is generated. The occurrence
of either an immediate (non-scheduled) or delayed transmit timer (TIDV) expiration interrupt halts
the TADV timer and eliminates any spurious second interrupts.
Setting the value to 0b disables the transmit absolute delay function. If an immediate (nonscheduled) interrupt is desired for any transmit descriptor, the descriptor IDE should be set to 0b.

13.4.45

TCP Segmentation Pad And Minimum Threshold
TSPMT (03830h; RW)
This register specifies fields affecting the Ethernet controller behavior during TCP Segmentation
operations. Values are specified in bytes. For normal (non TCP Segmentation) operations, the
Ethernet controller’s transmit DMA never begins servicing an individual data descriptor unless the
transmit Packet Buffer has sufficient room to accept all of the data associated with the descriptor.
However, for TCP Segmentation operations, it might be desirable to use a data descriptor that
refers to a larger contiguous buffer in host memory than is actually allocated for the transmit Packet
Buffer. For TCP segmentation, then, the transmit DMA is able to initiate smaller transfers than the
entire descriptor’s data length field.

16
TSPBP

Field

322

Bit(s)

0
TSMT

Initial
Value

Description

TSMT

15:0

0400h

TCP Segmentation Minimum Transfer

TSPBP

31:16

0100h

TCP Segmentation Packet Buffer Padding

Software Developer’s Manual

When performing TCP segmentation, the packet prototype header initially transferred by DMA is
stored internally and updated as each packet of the TCP segmentation operation is composed. As
data for subsequent TCP segments is DMA’d into the Ethernet controller, the frame header for each
segment is dynamically inserted in front of the frame payload data stream prior to being written to
the packet buffer. In order to obtain the most efficient use of burst DMA operations, the transmit
DMA attempts to fetch as much data from a descriptor as possible, rather than limiting itself to
bursting each data segment individually. However, to do this, sufficient packet buffer space must
be reserved to account for all headers are inserted into the fetched data stream, as the burst might
span multiple data segments. The calculation of how much packet buffer space should be reserved
is dependent on the MSS being used, the maximum-sized data buffer pointed to by a descriptor,
and the current header size. This calculation is left to software to pre-calculate for the worst-case
usage. The TSPBP register allows configuration of this buffer space that must be reserved as “pad”
for worst-case header insertion. To ensure that this value does not prevent descriptors from being
serviced at all, it is necessary that the transmit packet buffer allocation should be larger than the
sum of (maximum TCP HDRLEN + maximum MSS + TMPBP + 80 bytes).
Because the DMA attempts to issue burst fetches for a much data as possible, it is possible for the
transmit DMA to cause the transmit packet buffer to approach fullness (less the pad specified).
However, if the packet buffer empties slightly, the transmit DMA could initiate a series of small
transfers. To further optimize the efficiency of the transmit DMA during TCP segmentation
operation, the TSMT register allows configuration of the minimum number of bytes that the DMA
should attempt to transfer in a single burst operation. The transmit DMA uses this value to refrain
from issuing a burst read until at least TSMT bytes of data from the current data descriptor can be
stored in the packet buffer. This check is ignored if, after a series of DMA operations, the
descriptor contains a smaller number of unfetched data bytes. To ensure that this minimum
threshold does not prevent descriptors from being serviced at all, it is necessary that the transmit
packet buffer allocation should be larger than the sum of (TSMT + TSPBP + 80 bytes). To ensure
compliance with PCI-X specifications, this value should not be programmed to exceed 4 KB (the
largest single-burst transfer allowed by PCI-X).
The transmit DMA further refrains from initiating service of a new data descriptor unless sufficient
packet buffer space exists to at least fetch a full data segment or complete a partially-fetched
segment.

Software Developer’s Manual

323

13.4.46

Receive Descriptor Control
RXDCTL (02828h; R/W)
This register controls the fetching and write-back of receive descriptors. The three threshold values
are used to determine when descriptors are read from and written to host memory. The values can
be in units of cache lines or descriptors (each descriptor is 16 bytes) based on the GRAN flag. If
GRAN = 0b (specifications are in cache-line granularity), the thresholds specified (based on the
cacheline size specified in the PCI header CLS field) must not represent greater than 31
descriptors.
Table 13-87. RXDCTL Register Bit Description
31

Reserved

Field

324

GRAN

22 21

RSV

Bit(s)

16 15
WTHRESH

Initial
Value

14 13

RSV

8 7
HTHRESH

6 5

RSV

0
PTHRESH

Description

PTHRESH

5:0

Prefetch Threshold
Used to control when a prefetch of descriptors is considered. This
threshold refers to the number of valid, unprocessed receive
descriptors the Ethernet controller has in its on-chip buffer. If this
number drops below PTHRESH, the algorithm considers
prefetching descriptors from host memory. This fetch does not
happen unless there are at least RXDCTL.HTHRESH valid
descriptors in host memory to fetch. Value of PTHRESH can be in
either cache line units, or based on number of descriptors based
on RXDCTL.GRAN.

RSV

7:6

Reserved
Reads as 0b.
Should be written as 0b for future compatibility.

HTHRESH

13:8

Host Threshold
Provides the threshold of the valid descriptors in host memory.
A descriptors prefetch is performed each time enough valid
descriptors (TXDCTL.HTHRESH) are available in host memory,
no other DMA activity of greater priority is pending (descriptor
fetches and write backs or packet data transfers) and the number
of receive descriptors the Ethernet controller has on its on-chip
buffers drops below RXDCTL.PTHRESH. Value of HTHRESH
can be in either cache line units, or based on number of
descriptors based on RXDCTL.GRAN.

RSV

15:14

Reserved
Reads as 0b.
Should be written as 0b for future compatibility.

Software Developer’s Manual

Field

13.4.47

Bit(s)

Initial
Value

Description

WTHRESH

21:16

Write Back Threshold
WTHRESH controls the write back of processed receive
descriptors. This threshold refers to the number of receive
descriptors in the Ethernet controller’s on-chip buffer which are
ready to be written back to host memory. In the absence of
external events (explicit flushes), the write back occurs only after
more than WTHRESH descriptors are available for write back.
WTHRESH must contain a non-zero value to take advantage of
the write back bursting capabilities of the Ethernet controller.
A value of 1b causes the descriptors to be written back as soon
as one cache line is available. A value of WTHRESH can be in
either cache line units, or based on number of descriptors based
on RXDCTL.GRAN.

RSV

23:22

Reserved
Reads as 0b.
Should be written as 0b for future compatibility.

GRAN

Granularity
Set the values of PTHRESH, HTHRESH and WTHRESH in units
of cache lines or descriptors (each descriptor is 16 bytes)
0b = Cache line granularity.
1b = Descriptor granularity.

Reserved

31:25

Reserved
Reads as 0b.
Should be written as 0b for future compatibility.

Receive Checksum Control
RXCSUM (05000h; R/W)
The Receive Checksum Control register controls the receive checksum offloading features of the
Ethernet controller. The Ethernet controller supports the offloading of three receive checksum
calculations: the Packet Checksum, the IP Header Checksum, and the TCP/UDP Checksum.
The frame types that are supported:

• Ethernet II
• Ethernet SNAP
Table 13-88. RXCSUM Register Bit Description
31:11
RSV
1.

10
IPV6OFL1

TUOFLD IPOFLD

0
PCSS

Not applicable to the 82544GC/EI.

Software Developer’s Manual

325

Field

PCSS

Reserved

1.
2.

326

IP Checksum Off-load Enable
RXCSUM.IPOFLD is used to enable the IP Checksum
offloading feature. If RXCSUM.IPOFLD is set to 1b, the
Ethernet controller calculates the IP checksum and indicates
a pass/fail indication to software through the Checksum Error
bit (CSE) in the ERROR field to the receive descriptor. If both
RXCSUM.IPOFLD and RXCSUM.TUOFLD are set, the
Checksum Error bit (CSE) is set if either checksum was
incorrect. If neither RSCSUM.IPOFLD nor
RXCSUM.TUOFLD is set, the Checksum Error bit (CSE) is
be 0b for all packets.

TCP/UDP Checksum Off-load Enable
RXCSUM.TUOFL is used to enable the TCP/UDP Checksum
off-loading feature. When set to 1b, the Ethernet controller
calculates the TCP or UDP checksum and indicate a pass/fail
indication to software through the Checksum Error bit (CSE).
If both RXCSUM.TUOFLD and RXCSUM.TUOFLD are set,
the Checksum Error bit (CSE) is set if either checksum was
incorrect. If neither RXCSUM.IPOFLD nor
RXCSUM.TUOFLD is set, the Checksum Error bit (CSE) is
0b for all packets.

IPv6 Checksum Offload Enable
If IPV6OFL is set to 1b, hardware parses IPv6 headers when
parsing a receive packet. This applies to checksum offloading
only.

31:11

Reserved
Reads as 0b. Should be written with 0b for future
compatibility.

Description

TUOFLD

Initial
Value

Packet Checksum Start
Controls the starting byte for the Packet Checksum
calculation. The Packet Checksum is the one’s complement
over the receive packet, starting from the byte indicated by
RXCSUM.PCSS (0b corresponds to the first byte of the
packet), after stripping. For example, for an Ethernet II frame
encapsulated as an 802.3ac VLAN1 packet and with
RXCSUM.PCSS set to 14, the packet checksum would
include the entire encapsulated frame, excluding the 14-byte
Ethernet header (DA,SA,Type/Length) and the 4-byte VLAN
tag. The Packet Checksum does not include the Ethernet
CRC if the RCTL.SECRC bit is set. Software must make the
required offsetting computation (to back out the bytes that
should not have been included and to include the pseudoheader) prior to comparing the Packet Checksum against the
TCP checksum stored in the packet.

7:0

IPOFLD

IPV6OFL

Bit(s)

Not applicable to the 82541ER.
Not applicable to the 82544GC/EI.

Software Developer’s Manual

13.5

Filter Registers
This section contains detailed descriptions for those registers associated with the Ethernet
controller’s address filter capabilities.

13.5.1

Multicast Table Array
MTA[127:0] (05200h-053FCh; R/W)
The multicast table array is a way to extend address filtering beyond the 16 perfect in the Receive
Address Register (RAR). Note that the MTA is an imperfect filter that allows you to filter on 4096
similar addresses using a much smaller data structure than would be required to store all 4096
addresses in a linear table such as a perfect filter.
The Ethernet controller provides a 4096-bit vector multicast table array that is used when all the 16
perfect filters in the Receive Address Registers (RAR) are used. There is one register per 32 bits of
the Multicast Address Table for a total of 128 registers (thus the MTA[127:0] designation). The
size of the word array depends on the number of bits implemented in the multicast address table.
Software must mask to the desired bit on reads and supply a 32-bit word on writes. Accesses to this
table must be 32-bit.

Table 13-89. MTA Register Bit Description
Field

MC Bit Vector

Bit(s)

31:0

Initial
Value
X

Description
Multicast bit vector specifying 32 bits in the multicast address filter
table.

The operating system provides a list of addresses that it would like to respond to. The driver fills in
the Receive Address Registers (RAR) first, as these are exact matching addresses. If the OS
provides more than the 16 addresses available in RARs, the overflow is put into the MTA. The
MTA does not match the exact address, but a subset of the address. Each address filtered on is
represented by a single bit within the MTA table. Software needs to do the same calculations that
hardware does when checking against the MTA, so it can program the appropriate bit in the MTA.
When the hardware receives an address, it goes through the RARs, and if it does not find a match,
it does the same calculations that are described below on the address that it was given and only
checks one bit in the MTA. If that bit is set, it allows the packet to pass. If that bit is not set, it drops
the packet.
The calculation to find that bit is as follows (using the example of 12:34:56:78:9A:BCh):
Check the RCTL bits 13:12 to see what they are set to. In this example it is 00h that means that
we only look at bits 47:36. This corresponds to 0BC9h in the example address (assuming that
in your example 12 is the least significant byte and 0BCh is the most significant byte). The
way the address is stored in memory is the same that it would be going out on the wire, which
is the least significant byte is the first on the wire, so it looks like this:
BC:9A:78:56:34:12h so that the LSB (12) goes on the wire first. Breaking 0BC9h down into a
word:
0BC9h = 0000_1011_1100_1001b

Software Developer’s Manual

327

Of the 16 bits, look at bits 11:5, starting from zero. These seven bits corresponds to the row
within the MTA table (the MTA has 128 rows which require seven bits to define). In the
example, bits 11:5 are 1011110b. This corresponds to row 94.
Of these 16 bits, count out the first five bits, again starting from bit zero. These first five bits
correspond to the bit within the row (the MTA is 32 bits wide which require five bits to
define). In the example this is 01001b. This corresponds to bit nine. This is the offset within
the row.
Therefore, software needs to set bit nine of row 94 in the MTA. If the OS removes this address
from the filter list, software would need to clear this bit. This is the same bit that the hardware
would check if it received a packet with an address of xx:xx:xx:xx:9x:BCh.

Destination Address
47:40 39:32 31:24 23:16

15:8

7:0

bank[1:0]
Multicast Table
Array 32 x 128
(4096-bit vector)

word

?
pointer[11:5]

...
...
bit

pointer[4:0]

Figure 13-2. Multicast Table Array

328

Software Developer’s Manual

13.5.2

Receive Address Low
RAL (05400h + 8*n; R/W)
16 registers contain the lower bits of the 48-bit Ethernet address. All 32 bits are valid. Software can
access the High and Low registers as a register pair if it can perform a 64-bit access to the PCI bus.
The addresses stored in these registers are used for unicast/multicast address filtering.
The first receive address register (RAL0, RAH0) is also used for exact match PAUSE frame
checking (Valid PAUSE packet that is addressed to the station’s address). Therefore, RAL0 and
RAH0 always should be used to store the individual Ethernet MAC address of the Ethernet
controller.
Table 13-90. RAL Register Bit Description
31

0
RAL

Field

RAL

13.5.3

Initial
Value

Bit(s)

31:0

Description
Receive address low
Contains the lower 32-bit of the 48-bit Ethernet address.
RAL0 should be used to store the lower 32-bit of the Ethernet
controller’s Ethernet MAC address.

Receive Address High
RAH (05404h + 8n; R/W)
16 registers contain the upper bits of the 48-bit Ethernet address. The complete address is {RAH,
RAL}. Software can access the High and Low registers as a register pair if it can perform a 64-bit
access to the PCI bus. The addresses stored in these registers are used for unicast/multicast address
filtering.
The first receive address register (RAL0, RAH0) is also used for exact match Pause frame
checking (Valid Pause packet that is addressed to the station’s address). Therefore, RAL0 and
RAH0 always should be used to store the individual Ethernet MAC address of the Ethernet
controller.

Note:

When writing to this register, always write low-to-high. When clearing this register, always clear
high-to-low.

Software Developer’s Manual

329

Table 13-91. RAH Register Bit Description
31

18 17

Reserved

Field

RAH

0
RAH

Description

Receive address High
Contains the upper 16 bits of the 48-bit Ethernet address.
RAH0 should be used to store the upper 16-bit of the Ethernet
controller’s Ethernet MAC address.

17:16

Address Select
Selects how the address is to be used in the address filtering.
00b = Destination address (required for normal mode)
01b = Source address
10b = Reserved
11b = Reserved

Reserved

30:18

Reserved
Should be written with 0b to ensure future compatibility.
Reads as 0b.

Address Valid
Determines whether this address is compared against the
incoming packet. When set, the address is valid and is
compared against the incoming packet. When cleared, the
address is invalid and is not compared against the received
packet. AV is only cleared by a PCI reset or software reset. This
bit is unchanged by rx_reset.

13.5.4

Initial
Value

Bit(s)

15:0

16 15
AS

VLAN Filter Table Array1
VFTA[127:0] (05600h – 057FCh; R/W)
The Ethernet controller provides a 4096-bit vector VLAN Filter table array. There is one register
per 32 bits of the VLAN Filter Table, for a total of 128 registers (thus the VFTA[127:0]
designation). The size of the word array depends on the number of bits implemented in the VLAN
Filter table. Software must mask to the desired bit on reads, and supply a 32-bit word on writes.
Accesses to this table must be 32-bit.
The algorithm for VLAN filtering using the VFTA is identical to that used for the Multicast Table
Array. Refer to Section 13.5.1 for a block diagram of the algorithm. If VLANs are not used, there is
no need to initialize the VFTA.

Not applicable to the 82541ER.

330

Software Developer’s Manual

Table 13-92. VFTA[127:0] Bit Description
31

0
VLAN Filter Bit Vector

Field

Bit Vector

Initial
Value

Bit(s)

31:0

Description
Double-word wide bit vector specifying 32 bits in the VLAN
Filter table.

13.6

Wakeup Registers

13.6.1

Wakeup Control Register
WUC (05800h; R/W)
This register is reset any time LAN_PWR_GOOD is set to 0b. When AUX_POWER equals 0b,
this register is also reset by de-asserting (rising edge) RST#.

4
Reserved

Field

APME

PME_En

PME_Status

Initial
Value

Bit(s)

Software Developer’s Manual

APMPME

PME_S

PME_EN

APME

Description

Advance Power Management Enable
If set to 1b, APM Wakeup is enabled.
Note: Always 0b for the 82541ER.

PME_En
This read/write bit is used by the driver to access the PME_En
bit of the Power Management Control / Status Register
(PMCSR) without writing to the PCI configuration space.
Note: Do not set this bit for the 82541ER.

PME_Status
This bit is set when the Ethernet controller receives a wakeup
event. It is the same as the PME_Status bit in the Power
Management Control / Status Register (PMCSR). Writing a 1b
to this bit clears the PME_Status bit in the PMCSR.

331

Field

Description

APMPME

Assert PME On APM Wakeup
If set to 1b, the Ethernet controller sets the PME_Status bit in
the Power Management Control / Status Register (PMCSR)
and asserts PME# when APM Wakeup is enabled and the
Ethernet controller receives a matching Magic Packet. This
field value is loaded from the EEPROM.
Note: Not applicable to the 82541ER.

Reserved

27:4

Reserved
Reads as 0b.

Dynamic
Powerdown1

Dynamic Powerdown Mode
When programmed to 1b, enables dynamic powerdown
operation.

Auto Freq
Select1

30:29

0b2

Automatic Frequency Select
Determines automatic reduction of MAC frequency in 82541xx
only. Reserved for the 82547GI/EI.
Bit 29 controls MAC speed at 1000 MB. When cleared,
enables the MAC to run at full speed. When set and PCI is
configured for 33 MHz (82541xx), allows the MAC to run at
half speed.
Bit 30 controls MAC speed at other Ethernet rates. When
cleared, allows the MAC to run at full speed. When set, it
allows the MAC to run at quarter speed.
If both bits are cleared, the MAC frequency select bits control
MAC frequency.

SPM1

Smart Powerdown MAC
When programmed to 0b, the MAC operates normally. When
programmed to 1b, the MAC enters smart powerdown mode.

Reserved3

31:28

Reserved
Reads as 0b.

1.
2.
3.

13.6.2

Initial
Value

Bit(s)

82541xx and 82547GI/EI only.
Loaded from the EEPROM.
Not applicable to the 82541xx and 82547GI/EI.

Wakeup Filter Control Register
WUFC (05808h; R/W)
This register is used to enable each of the pre-defined and flexible filters for wakeup support. A
value of 1b means the filter is turned on, and a value of 0b means the filter is turned off.
This register is reset any time LAN_PWR_GOOD is 0b. When AUX_POWER equals 0b, this
register is also reset by deasserting (rising edge) RST#.

332

Software Developer’s Manual

31
Reserved
1.
2.
3.

FLX3

FLX2

FLX1

FLX0

ITCO1

14
Reserved

IPv62

IPv43

ARP

MAG

LNKC

82541xx and 82547GI/EI only.
Not applicable to the 82544GC/EI.
IP for the 82544GC/EI.

Field

Initial Value

Description

LNKC

Link Status Change Wakeup Enable.

MAG

Magic Packet Wakeup Enable.

Directed Exact Wakeup Enable.

Directed Multicast Wakeup Enable.

Broadcast Wakeup Enable.

ARP

ARP Request Packet Wakeup Enable.

IPv41

Directed IPv4 Packet Wakeup Enable.

IPv62

Directed IPv6 Packet Wakeup Enable.

Reserved

14:8

Reserved. Set these bits to 0b.

ITCO3

Ignore TCO/management packets for wakeup.

FLX0

Flexible Filter 0 Enable.

FLX1

Flexible Filter 1 Enable.

FLX2

Flexible Filter 2 Enable.

FLX3

Flexible Filter 3 Enable.

1.
2.
3.

13.6.3

Bit(s)

IP for the 82544GC/EI.
Not applicable to the 82544GC/EI.
82541xx and 82547GI/EI only.

Wakeup Status Register
WUS (05810h; R)
This register is used to record statistics about all wakeup packets received. If a packet matches
multiple criteria then multiple bits could be set. Writing a 1b to any bit clears that bit.
This register is not cleared when RST# is asserted. It is only cleared when LAN_PWR_GOOD is
de-asserted or when cleared by the driver software.

Software Developer’s Manual

333

Reserved
1.
2.

FLX3

FLX2

FLX1

FLX0

Reserved

IPv61

IPv42

ARP

MAG

LNKC

Not applicable to the 82544GC/EI.
IP for the 82544GC/EI.

Field

Bit(s)

Initial Value

Description

LNKC

Link Status Change.

MAG

Magic Packet Received.

Directed Exact Packet Received
The packet’s address matched one of the 16 pre-programmed
exact values in the Receive Address registers.

Directed Multicast Packet Received
The packet was a multicast packet whose hashed to a value
that corresponded to a 1 bit in the Multicast Table Array.

Broadcast Packet Received.

ARP

ARP Request Packet Received.

IPv41

Directed IPv4 Packet Received.

IPv62

Directed IPv6 Packet Received.

FLX0

Flexible Filter 0 Match.

FLX1

Flexible Filter 1 Match.

FLX2

Flexible Filter 2 Match.

FLX3

Flexible Filter 3 Match.

1.
2.

334

IP for the 82544GC/EI.
Not applicable to the 82544GC/EI.

Software Developer’s Manual

13.6.4

IP Address Valid
IPAV (5838h; R/W)
The IP Address Valid indicates whether the IP addresses in the IP Address Table are valid.
The valid bits are reset any time LAN_PWR_GOOD is 0b. When AUX_POWER equals 0b, the
valid bits are also reset by deasserting (rising edge) RST#.

16
v601

Reserved
1.
2.

Field

Reserved

V43

V42

V41

V40

V32

V22

V12

V02

Not applicable to the 82544GC/EI.
82544GC/EI only.

Bit(s)

Initial Value

Description

V40
V0

0b1

IPv4 Address 0 Valid
IP Address 0 Valid2

V41
V1

IPv4 Address 1 Valid
IP Address 1 Valid2

V42
V2

IPv4 Address 2 Valid
IP Address 2 Valid2

V43
V3

IPv4 Address 3 Valid
IP Address 3 Valid2

V603

IPv6 Address 0 Valid

1.
2.
3.

The initial value is loaded from the IP Address Valid bit of the EEPROM’s Management Control Register.
82544GC/EI only.
Not applicable to the 82544GC/EI.

Software Developer’s Manual

335

13.6.5

IPv4 Address Table1
IP4AT (05840h - 05858h; R/W)2
The IPv4 Address Table is used to store the four IP addresses for ARP Request packet and Directed
IP packet wakeup for IPv4.

Note:

This table is not cleared by any reset.

DWORD#

5840h

IPV4ADDR01

5848h

IPV4ADDR1

5850h

IPV4ADDR2

5858h

IPV4ADDR3

IPA for the 82544GC/EI.

Field

Dword #

Address

Bit(s)

Initial Value

Description

IPV4ADDR01

5840h

31:0

IPv4 Address 0

IPV4ADDR1

5848h

31:0

IPv4 Address 1

IPV4ADDR2

5850h

31:0

IPv4 Address 2

IPV4ADDR3

5858h

31:0

IPv4 Address 3

1.
2.

Address

IPA for the 82544GC/EI.
The first entry is loaded from the EEPROM if the IP Address Valid field of the EEPROM’s Management Control word is 1b
and the IP Address Type field is 0b (IPv4). Otherwise, the value of this register is undefined after reset.

IP Address Table for the 82544GC/EI.
IPAT for the 82544GC/EI.

336

Software Developer’s Manual

13.6.6

IPv6 Address Table1
IP6AT (05880h - 0588Ch; R/W)
The IPv6 Address Table is used to store the IPv6 addresses for ARP Request packet and Directed
IP packet wakeup for IPv6.

Note:

This table is not cleared by any reset.

DWORD#

Address

5880h

5884h

5888h

588Ch

IPV6ADDR0

Field

Dword #

Address

Bit(s)

Initial Value1

Description

5880h

31:0

IPv6 Address 0, bytes 1-4

5884h

31:0

IPv6 Address 0, bytes 5-8

5888h

31:0

IPv6 Address 0, bytes 9-12

588Ch

31:0

IPv6 Address 0, bytes 16-13

IPV6ADDR0

This table is loaded from the EEPROM if the IP Address Valid field of the EEPROM’s Management Control word is 1b and
the IP Address Type field is 0b (IPv4). Otherwise, the value of this register is undefined after reset.

Not applicable to the 82544GC/EI.

Software Developer’s Manual

337

13.6.7

Wakeup Packet Length
WUPL (05900h; R/W)
This register indicates the length of the first wakeup packet received. It is valid if one of the bits in
the Wakeup Status Register (WUSR) is set. It can be written for diagnostic purposes and is not
cleared by any reset.

Reserved

Field
LEN

13.6.8

Bit(s)
11:0

0
Length

Initial Value

Description
Length of wakeup packet. (If jumbo frames is enabled and the
packet is longer than 2047 bytes this field is 2047.)

Wakeup Packet Memory (128 Bytes)
WUPM (05A00h - 05A7Ch; R/W)
This register is read-only and it is used to store the first 128 bytes of the wakeup packet for
software retrieval after system wakeup. It can be written for diagnostic purposes and is not cleared
by any reset.

0
WUPD

Field
WUPD

13.6.9

Bit(s)
31:0

Initial
Value
X

Description
Wakeup Packet Data

Flexible Filter Length Table
FFLT (05F00h - 05F18h; R/W)
The Flexible Filter Length Table stores the minimum packet lengths required to pass each of the
Flexible Filters. Any packets that are shorter than the programmed length won’t pass that filter.
Each Flexible Filter considers a packet that doesn’t have any mismatches up to that point to have
passed the Flexible Filter when it reaches the required length. It does not check any bytes past that
point.
All reserved fields read as 0b’s and ignore writes.

338

Software Developer’s Manual

Before writing to the Flexible Filter Length Table the driver must first disable the flexible filters by
writing 0b’s to the Flexible Filter Enable bits of the Wakeup Filter Control Register
(WUFC.FLXn).

13.6.10

Reserved

Length 0

Reserved

Length 1

Reserved

Length 2

Reserved

Length 3

Field

Dword #

Address

Bit(s)

Initial Value

Description

LEN0

5F00h

10:0

Minimum Length for Flexible Filter 0

LEN1

5F08h

10:0

Minimum Length for Flexible Filter 1

LEN2

5F10h

10:0

Minimum Length for Flexible Filter 2

LEN3

5F18h

10:0

Minimum Length for Flexible Filter 3

Flexible Filter Mask Table
FFMT (09000h - 093F8h; R/W)
The Flexible Filter Mask and Table is used to store the four 1-bit masks for each of the first 128
data bytes in a packet, one for each Flexible Filter. If the mask bit is set to 1b, the corresponding
Flexible Filter compares the incoming data byte at the index of the mask bit to the data byte stored
in the Flexible Filter Value Table.
Before writing to the Flexible Filter Mask Table the driver must first disable the flexible filters by
writing 0b’s to the Flexible Filter Enable bits of the Wakeup Filter Control Register
(WUFC.FLXn).

Software Developer’s Manual

Reserved

Byte 0 Mask

Reserved

Byte 1 Mask

Reserved

Byte 2 Mask

Reserved

Byte 126 Mask

Reserved

Byte 127 Mask

339

Field

Dword #

Address

Bit(s)

Initial Value

Description

MASK0

9000h

15:0

Mask for Filter [3:0] for Byte 0

MASK1

9008h

15:0

Mask for Filter [3:0] for Byte 2

MASK2

9010h

15:0

Mask for Filter [3:0] for Byte 3

254

93F8h

15:0

Mask for Filter [3:0] for Byte 127

...
MASK127

13.6.11

Flexible Filter Value Table
FFVT (09800h - 09BF8h; R/W)
The Flexible Filter Value and Table is used to store the one value for each byte location in a packet
for each flexible filter. If the corresponding mask bit is set to 1b, the Flexible Filter compares the
incoming data byte to the values stored in this table.
Before writing to the Flexible Filter Value Table the driver must first disable the flexible filters by
writing 0b’s to the Flexible Filter Enable bits of the Wakeup Filter Control Register
(WUFC.FLXn).

Reserved

Byte0: Value3

Value2

Value1

Value0

Reserved

Byte1: Value3

Value2

Value1

Value0

Reserved

Byte2: Value3

Value2

Value1

Value0

Reserved

Byte127: Value3

Value2

Value1

Value0

Field

Dword #

Address

Bit(s)

Initial Value

Description

MASK0

9800h

15:0

Mask for Filter [3:0] for Byte 0

MASK1

9808h

15:0

Mask for Filter [3:0] for Byte 2

MASK2

9810h

15:0

Mask for Filter [3:0] for Byte 3

254

9BF8h

15:0

Mask for Filter [3:0] for Byte 127

...
MASK127

13.7

Statistics Registers
All statistics registers are implemented as 32-bit registers. 64-bit accesses to these registers must
have the upper byte enables de-asserted. 32-bit registers with addresses not on a quadword
boundary cannot be accessed through a 64-bit access.
Registers that count octets make up 64-bit registers.

340

Software Developer’s Manual

All Statistics registers reset when read. 64-bit registers reset whenever the upper 32 bits are read. In
addition, they stick at FFFFh_FFFFh when the maximum value is reached.
The Statistics registers are not hardware initialized. Their default value is unknown. Software
should read the contents of all registers in order to clear them prior to enabling the receive and
transmit channels.
Note:

13.7.1

For the receive statistics, it should be noted that a packet is indicated as “received” if it passes the
device filters, and it is placed in the packet buffer memory. A packet does not have to be transferred
to host memory in order to be counted as “received.”

CRC Error Count
CRCERRS (04000h; R)
Counts the number of receive packets with CRC errors. In order for a packet to be counted in this
register, it must pass address filtering and must be 64 bytes or greater (from
through , inclusively) in length. If receives are not enabled, then this register does not
increment.
Table 13-93. CRCERRS Register Bit Description
31

0
CEC

13.7.2

Field

Bit(s)

Initial
Value

CEC

31:0

Description
CRC error count

Alignment Error Count
ALGNERRC (04004h; R)
Counts the number of receive packets with alignment errors (the packet is not an integer number of
bytes in length). In order for a packet to be counted in this register, it must pass address filtering
and must be 64 bytes or greater (from through , inclusively) in
length. If receives are not enabled, then this register does not increment. This register is valid only
in MII mode during 10/100 Mb/s operation.

Software Developer’s Manual

341

Table 13-94. ALGNERRC Register Bit Description
31

0
AEC

13.7.3

Field

Bit(s)

Initial
Value

AEC

31:0

Description
Alignment error count

Symbol Error Count
SYMERRS (04008h; R)
Counts the number of symbol errors between reads. The count increases for every bad symbol
received, whether or not a packet is currently being received and whether or not the link is up. This
register only increments in internal SerDes mode (TBI mode for the 82544GC/EI).
Table 13-95. SYMERRS Register Bit Description
31

0
SYMERRS

13.7.4

Field

Bit(s)

Initial
Value

SYMERRS

31:0

Description
Symbol Error Count

RX Error Count
RXERRC (0400Ch; R)
Counts the number of packets received in which I_RX_ER was asserted by the PHY. In order for a
packet to be counted in this register, it must pass address filtering and must be 64 bytes or greater
(from through , inclusively) in length. If receives are not enabled,
then this register does not increment. In internal SerDes mode (TBI mode for the 82544GC/EI),
this register increments on the reception of
/V/ codes.

342

Software Developer’s Manual

Table 13-96. RXERRC Register Bit Description
31

0
RXEC

13.7.5

Field

Bit(s)

Initial
Value

RXEC

31:0

Description
RX error count

Missed Packets Count
MPC (04010h; R)
Counts the number of missed packets. Packets are missed when the receive FIFO has insufficient
space to store the incoming packet. This can be caused because of too few buffers allocated, or
because there is insufficient bandwidth on the PCI bus. Events setting this counter cause RXO, the
Receiver Overrun Interrupt, to be set. This register does not increment if receives are not enabled.
These packets are also counted in the Total Packets Received register as well as in Total Octets
Received.
Table 13-97. MPC Register Bit Description
31

0
MPC

13.7.6

Field

Bit(s)

Initial
Value

MPC

31:0

Description
Missed Packets Count

Single Collision Count
SCC (04014h; R)
This register counts the number of times that a successfully transmitted packet encountered a single
collision. This register only increments if transmits are enabled and the Ethernet controller is in
half-duplex mode.

Software Developer’s Manual

343

Table 13-98. SCC Register Bit Description
31

0
SCC

13.7.7

Field

Bit(s)

Initial
Value

SCC

31:0

Description
Number of times a transmit encountered a single collision.

Excessive Collisions Count
ECOL (04018h; R)
When 16 or more collisions have occurred on a packet, this register increments, regardless of the
value of collision threshold. If collision threshold is set below 16, this counter won’t increment.
This register only increments if transmits are enabled and the Ethernet controller is in half-duplex
mode.
Table 13-99. ECOL Register Bit Description
31

0
ECC

13.7.8

Field

Bit(s)

Initial
Value

ECC

31:0

Description
Number of packets with more than 16 collisions.

Multiple Collision Count
MCC (0401Ch; R)
This register counts the number of times that a transmit encountered more than one collision but
less than 16. This register only increments if transmits are enabled and the Ethernet controller is in
half-duplex mode.

344

Software Developer’s Manual

Table 13-100. MCC Register Bit Description
31

0
MCC

13.7.9

Field

Bit(s)

Initial
Value

MCC

31:0

Description
Number of times a successful transmit encountered multiple
collisions.

Late Collisions Count
LATECOL (04020h; R)
Late collisions are collisions that occur after 64-byte time into the transmission of the packet while
working in 10-100 Mb/s data rate, and 512 byte time into the transmission of the packet while
working in the 1000 Mb/s data rate. This register only increments if transmits are enabled and the
device is in half-duplex mode.
Table 13-101. LATECOL Register Bit Description
31

0
LCC

13.7.10

Field

Bit(s)

Initial
Value

LCC

31:0

Description
Number of packets with late collisions.

Collision Count
COLC (04028h; R)
This register counts the total number of collisions that are not late collisions seen by the transmitter.
This register only increments if transmits are enabled and the Ethernet controller is in half-duplex
mode.
Table 13-102. COLC Register Bit Description
31

0
CCC

Field

Bit(s)

Initial
Value

CCC

31:0

Software Developer’s Manual

Description
Total number of collisions experienced by the transmitter.

345

13.7.11

Defer Count
DC (04030h; R)
This register counts defer events. A defer event occurs when the transmitter cannot immediately
send a packet due to the medium being busy either because another device is transmitting, the IPG
timer has not expired, half-duplex deferral events, reception of XOFF frames, or the link is not up.
This register only increments if transmits are enabled. This counter does not increment for
streaming transmits that are deferred due to TX IPG.
Table 13-103. DC Register Bit Description
31

0
CDC

13.7.12

Field

Bit(s)

Initial
Value

CDC

31:0

Description
Number of defer events.

Transmit with No CRS
TNCRS (04034h; R)
This register counts the number of successful packet transmissions in which the internal carrier
sense signal from the PHY was not asserted within one slot time of start of transmission.
The PHY should assert the internal carrier sense signal during every transmission. Failure to do so
may indicate that the link has failed, or the PHY has an incorrect link configuration. This register
only increments if transmits are enabled. This register is not valid in internal SerDes1 mode (TBI
mode for the 82544GC/EI), and is only valid when the Ethernet controller is operating at full
duplex.
Table 13-104. TNCRS Register Bit Description
31

0
TNCRS

Field
TNCRS

Bit(s)

Initial
Value

Description

31:0

Number of transmissions without a CRS assertion from the PHY.

The 82540EP/EM, 82541xx, and 82547GI/EI do not support SerDes functionality.

346

Software Developer’s Manual

13.7.13

Sequence Error Count
SEC (04038h; R)
This register counts sequence error events. The proper sequence of 8b/10b symbols is as follows:
idle, start-of-frame (SOF), data, pad (optional), end-of-frame (EOF), fill (optional), idle. Hardware
increments this counter for any illegal sequence of delimiters. If the link is not up, this register does
not increment. This register is only valid in internal SerDes mode (TBI mode for the 82544GC/
EI).
Table 13-105. SEC Register Bit Description
31

0
SEC

13.7.14

Field

Bit(s)

Initial
Value

SEC

31:0

Description
Number of sequence error events.

Carrier Extension Error Count
CEXTERR (0403Ch; R)
This register counts the number of packets received in which the carrier extension error was
signaled across the internal PHY interface. The PHY propagates carrier extension errors to the
MAC when an error is detected during the carrier extended time of a packet reception. An
extension error is signaled by the PHY by the encoding of 1Fh on the receive data inputs. This
register only increments if receives are enabled and the Ethernet controller is operating at
1000 Mb/s. For example, internal PHY or internal SerDes modes (internal PHY or TBI modes for
the 82544GC/EI).
Table 13-106. CEXTERR Register Bit Description
31

0
CEXTERR

Field

Bit(s)

Initial
Value

CEXTERR

31:0

Software Developer’s Manual

Description
Number of packets received with a carrier extension error.

347

13.7.15

Receive Length Error Count
RLEC (04040h; R)
This register counts receive length error events. A length error occurs if an incoming packet passes
the filter criteria but is undersized or oversized. Packets less than 64 bytes are undersized. Packets
over 1522 bytes are oversized if LongPacketEnable is 0b (RCTL.LPE). If LongPacketEnable
(LPE) is 1b, then an incoming packet is considered oversized if it exceeds 16384 bytes.
If receives are not enabled, this register does not increment. These lengths are based on bytes in
the received packet from through , inclusively.
Table 13-107. RLEC Register Bit Description
31

0
RLEC

13.7.16

Field

Bit(s)

Initial
Value

RLEC

31:0

Description
Number of packets with receive length errors.

XON Received Count
XONRXC (04048h; R)
This register counts the number of valid XON packets received. XON packets can use the global
address, or the station address. This register only increments if receives are enabled.
Table 13-108. XONRXC Register Bit Description
31

0
XONRXC

348

Field

Bit(s)

Initial
Value

XONRXC

31:0

Description
Number of XON packets received.

Software Developer’s Manual

13.7.17

XON Transmitted Count
XONTXC (0404Ch; R)
This register counts the number of XON packets transmitted. These can be either due to a full
queue or due to software initiated action (using TCTL.SWXOFF). This register only increments if
transmits are enabled.
Table 13-109. XONTXC Register Bit Description
31

0
XONTXC

13.7.18

Field

Bit(s)

Initial
Value

XONTXC

31:0

Description
Number of XON packets transmitted.

XOFF Received Count
XOFFRXC (04050h; R)
This register counts the number of valid XOFF packets received. XOFF packets can use the global
address or the station address. This register only increments if receives are enabled.
Table 13-110. XOFFRXC Register Bit Description
31

0
XOFFRXC

13.7.19

Field

Bit(s)

Initial
Value

XOFFRXC

31:0

Description
Number of XOFF packets received.

XOFF Transmitted Count
XOFFTXC (04054h; R)
This register counts the number of XOFF packets transmitted. These can be either due to a full
queue or due to software initiated action (using TCTL.SWXOFF). This register only increments if
transmits are enabled.

Software Developer’s Manual

349

Table 13-111. XOFFTXC Register Bit Description
31

0
XOFFTXC

13.7.20

Field

Bit(s)

Initial
Value

XOFFTXC

31:0

Description
Number of XOFF packets transmitted.

FC Received Unsupported Count
FCRUC (04058h; R)
This register counts the number of unsupported flow control frames that are received.
The FCRUC counter increments when a flow control packet is received that matches either the
reserved flow control multicast address (in FCAH/L) or the MAC station address, and has a
matching flow control type field match (to the value in FCT), but has an incorrect opcode field.
This register only increments if receives are enabled.
Table 13-112. FCRUC Register Bit Description
31

0
FCRUC

13.7.21

Field

Bit(s)

Initial
Value

FCRUC

31:0

Description
Number of unsupported flow control frames received.

Packets Received (64 Bytes) Count
PRC64 (0405Ch; R)
This register counts the number of good packets received that are exactly 64 bytes (from
through , inclusively) in length. Packets that are counted in the
Missed Packet Count register are not counted in this register. This register does not include
received flow control packets and increments only if receives are enabled.

350

Software Developer’s Manual

Table 13-113. PRC64 Register Bit Description
31

0
PRC64

13.7.22

Field

Bit(s)

Initial
Value

PRC64

31:0

Description
Number of packets received that are 64 bytes in length.

Packets Received (65-127 Bytes) Count
PRC127 (04060h; R)
This register counts the number of good packets received that are 65-127 bytes (from through , inclusively) in length. Packets that are counted in the Missed Packet
Count register are not counted in this register. This register does not include received flow control
packets and increments only if receives are enabled.
Table 13-114. PRC127 Register Bit Description
31

0
PRC127

13.7.23

Field

Bit(s)

Initial
Value

PRC127

31:0

Description
Number of packets received that are 65-127 bytes in length.

Packets Received (128-255 Bytes) Count
PRC255 (04064h; R)
This register counts the number of good packets received that are 128-255 bytes (from
through , inclusively) in length. Packets that are counted in the
Missed Packet Count register are not counted in this register. This register does not include
received flow control packets and increments only if receives are enabled.

Software Developer’s Manual

351

Table 13-115. PRC225 Register Bit Description
31

0
PRC255

13.7.24

Field

Bit(s)

Initial
Value

PRC255

31:0

Description
Number of packets received that are 128-255 bytes in length.

Packets Received (256-511 Bytes) Count
PRC511 (04068h; R)
This register counts the number of good packets received that are 256-511 bytes (from
through , inclusively) in length. Packets that are counted in the
Missed Packet Count register are not counted in this register. This register does not include
received flow control packets and increments only if receives are enabled.
Table 13-116. PRC551 Register Bit Description
31

0
PRC511

13.7.25

Field

Bit(s)

Initial
Value

PRC511

31:0

Description
Number of packets received that are 256-511 bytes in length.

Packets Received (512-1023 Bytes) Count
PRC1023 (0406Ch; R)
This register counts the number of good packets received that are 512-1023 bytes (from
through , inclusively) in length. Packets that are counted in the
Missed Packet Count register are not counted in this register. This register does not include
received flow control packets and increments only if receives are enabled.

352

Software Developer’s Manual

Table 13-117. PRC1023 Register Bit Description
31

0
PRC1023

13.7.26

Field

Bit(s)

Initial
Value

Description

PRC1023

31:0

Number of packets received that are 512-1023 bytes in length.

Packets Received (1024 to Max Bytes) Count
PRC1522 (04070h; R)
This register counts the number of good packets received that are from 1024 bytes to the maximum
(from through , inclusively) in length. The maximum is dependent
on the current receiver configuration and the type of packet being received. If a packet is counted in
Receive Oversized Count, it is not counted in this register (see Section 13.7.36). This register does
not include received flow control packets and only increments if the packet has passed address
filtering and receives are enabled.
Due to changes in the standard for maximum frame size for VLAN tagged frames in IEEE
Standard 802.3, these Ethernet controllers accept packets which have a maximum length of 1522
bytes. The RMON statistics associated with this range has been extended to count 1522-byte long
packets.
Table 13-118. PRC1522 Register Bit Description
31

0
PRC1522

13.7.27

Field

Bit(s)

Initial
Value

Description

PRC1522

31:0

Number of packets received that are 1024-Max bytes in length.

Good Packets Received Count
GPRC (04074h; R)
This register counts the number of good packets received of any legal length. The legal length for
the received packet is defined by the value of LongPacketEnable (CTRL.LPE) (see Section
13.7.15). This register does not include received flow control packets and only counts packets that
pass filtering. This register only increments if receives are enabled. This register does not count
packets counted by the Missed Packet Count (MPC) register.

Software Developer’s Manual

353

Table 13-119. GPRC Register Bit Description
31

0
GPRC

13.7.28

Field

Bit(s)

Initial
Value

GPRC

31:0

Description
Number of good packets received (of any length).

Broadcast Packets Received Count
BPRC (04078h; R)
This register counts the number of good (no errors) broadcast packets received. This register does
not count broadcast packets received when the broadcast address filter is disabled. This register
only increments if receives are enabled.
Table 13-120. BPRC Register Bit Description
31

0
BPRC

13.7.29

Field

Bit(s)

Initial
Value

BPRC

31:0

Description
Number of broadcast packets received.

Multicast Packets Received Count
MPRC (0407Ch; R)
This register counts the number of good (no errors) multicast packets received. This register does
not count multicast packets received that fail to pass address filtering nor does it count received
flow control packets. This register only increments if receives are enabled. This register does not
count packets counted by the Missed Packet Count (MPC) register.

354

Software Developer’s Manual

Table 13-121. MPRC Register Bit Description
31

0
MPRC

13.7.30

Field

Bit(s)

Initial
Value

MPRC

31:0

Description
Number of multicast packets received.

Good Packets Transmitted Count
GPTC (04080h; R)
This register counts the number of good (no errors) packets transmitted. A good transmit packet is
considered one that is 64 or more bytes in length (from through ,
inclusively) in length. This does not include transmitted flow control packets. This register only
increments if transmits are enabled.
Table 13-122. GPTC Register Bit Description
31

0
GPTC

Field
GPTC

13.7.31

Bit(s)
31:0

Initial
Value
0b

Description
Number of good packets transmitted.

Good Octets Received Count
GORCL (04088h; R)/GORCH (0408Ch; R)
These registers make up a 64-bit register that counts the number of good (no errors) octets
received. This register includes bytes received in a packet from the field
through the field, inclusively. This register resets each time the upper 32 bits are read
(GORCH).
In addition, it sticks at FFFFh_FFFFh_FFFFh_FFFFh when the maximum value is reached. Only
octets of packets that pass address filtering are counted in this register. This register only
increments if receives are enabled.
These octets do not include octets of received flow control packets.

Software Developer’s Manual

355

Table 13-123. GORCL and GORCH Register Bit Description
31

GORCH

13.7.32

GORCL

Field

Bit(s)

Initial
Value

GORCL

31:0

Number of good octets received – lower 4 bytes.

GORCH

31:0

Number of good octets received – upper 4 bytes.

Description

Good Octets Transmitted Count
GOTCL (04090h; R)/ GOTCH (04094; R)
These registers make up a 64-bit register that counts the number of good (no errors) octets
transmitted. This register resets each time the upper 32 bits are read (GOTCH).
In addition, it sticks at FFFF_FFFF_FFFF_FFFFh when the maximum value is reached. This
register includes bytes transmitted in a packet from the field through the
field, inclusively. This register counts octets in successfully transmitted packets that are 64
or more bytes in length. This register only increments if transmits are enabled.
These octets do not include octets in transmitted flow control packets.
Table 13-124. GOTCL and GOTCH Register Bit Description
31

0
GOTCH

Field

13.7.33

Bit(s)

0
GOTCL

Initial
Value

Description

GOTCL

31:0

Number of good octets transmitted – lower 4 bytes.

GOTCH

31:0

Number of good octets transmitted – upper 4 bytes.

Receive No Buffers Count
RNBC (040A0h; R)
This register counts the number of times that frames were received when there were no available
buffers in host memory to store those frames (receive descriptor head and tail pointers were equal).
The packet is still received if there is space in the FIFO. This register only increments if receives
are enabled.

356

Software Developer’s Manual

This register does not increment when flow control packets are received.
Table 13-125. RNBC Register Bit Description
31

0
RNBC

Field
RNBC

13.7.34

Bit(s)
31:0

Initial
Value
0b

Description
Number of receive no buffer conditions.

Receive Undersize Count
RUC (040A4h; R)
This register counts the number of received frames that passed address filtering, and were less than
minimum size (64 bytes from through , inclusively), and had a
valid CRC. This register only increments if receives are enabled.
Table 13-126. RUC Register Bit Description
31

0
RUC

Field
RUC

13.7.35

Bit(s)
31:0

Initial
Value
0b

Description
Number of receive undersize errors.

Receive Fragment Count
RFC (040A8h; R)
This register counts the number of received frames that passed address filtering, and were less than
minimum size (64 bytes from through , inclusively), but had a bad
CRC (this is slightly different from the Receive Undersize Count register). This register only
increments if receives are enabled.

Software Developer’s Manual

357

Table 13-127. RFC Register Bit Description
31

0
RFC

13.7.36

Field

Bit(s)

Initial
Value

RFC

31:0

Description
Number of receive fragment errors.

Receive Oversize Count
ROC (040ACh; R)
This register counts the number of received frames with valid CRC field that passed address
filtering, and were greater than maximum size. Packets over 1522 bytes are oversized if LongPacketEnable (RCTL.LPE) is 0b. If LongPacketEnable is 1b, then an incoming packet is considered
oversized if it exceeds 16384 bytes.
If receives are not enabled, this register does not increment. These lengths are based on bytes in the
received packet from through , inclusively.
Table 13-128. ROC Register Bit Description
31

0
ROC

Field
ROC

13.7.37

Bit(s)
31:0

Initial
Value
0b

Description
Number of receive oversize errors.

Receive Jabber Count
RJC (040B0h; R)
This register counts the number of received frames that passed address filtering, and were greater
than maximum size and had a bad CRC (this is slightly different from the Receive Oversize Count
register).
Packets over 1522 bytes are oversized if LongPacketEnable (RCTL.LPE) is 0b.
If LongPacketEnable is 1b, then an incoming packet is considered oversized if it exceeds 16384
bytes.
If receives are not enabled, this register does not increment. These lengths are based on bytes in the
received packet from through , inclusively.

358

Software Developer’s Manual

Table 13-129. RJC Register Bit Description
31

0
RJC

13.7.38

Field

Bit(s)

Initial
Value

RJC

31:0

Description
Number of receive jabber errors.

Management Packets Received Count1
MGTPRC (040B4h; R)
This register counts the total number of packets received that pass the management filters as
described in the appropriate Total Cost of Ownership (TCO) System Management Bus Interface
Application Notes. Management packets include RMCP and ARP packets. Any packets with errors
are not counted, except that packets dropped because the management receive FIFO is full or the
packet is longer than 200 bytes is counted.
31

0
MGTPRC

Field
MGTPRC

Initial
Value

Bit(s)
31:0

Description
Number of management packets received.

Not applicable to the 82544GC/EI or 82541ER.

Software Developer’s Manual

359

13.7.39

Management Packets Dropped Count1
MGTPDC (040B8h; R)
This register counts the total number of packets received that pass the management filters as
described in the appropriate Total Cost of Ownership (TCO) System Management Bus Interface
Application Notes and then are dropped because the management receive FIFO is full or the packet
is longer than 200 bytes. Management packets include RMCP and ARP packets.
31

0
MGTPDC

Field
MGTPDC

13.7.40

Initial
Value

Bit(s)
31:0

Description
Number of management packets dropped.

Management Pkts Transmitted Count1
MGTPTC (040BCh; R)
This register counts the total number of packets that are transmitted that are either received over the
SMBus or are generated by the Ethernet controller’s ASF function.
31

0
MGTPTC

Field
MGTPTC

13.7.41

Initial
Value

Bit(s)
31:0

Description
Number of management packets transmitted.

Total Octets Received
TORL (040C0h; R) / TORH (040C4h; R)
These registers make up a 64-bit register that counts the total number of octets received. This
register resets each time the upper 32 bits are read (TORH). In addition, it sticks at
FFFF_FFFF_FFFF_FFFFh when the maximum value is reached.

Not applicable to the 82544GC/EI or 82541ER.

360

Software Developer’s Manual

All packets received have their octets summed into this register, regardless of their length, whether
they are erred, or whether they are flow control packets. This register includes bytes received in a
packet from the field through the field, inclusively. This register
only increments if receives are enabled.
Table 13-130. TORL and TORH Register Bit Descriptions
31

TORH

13.7.42

TORL

Field

Bit(s)

Initial
Value

TORL

31:0

Number of total octets received – lower 4 bytes.

TORH

31:0

Number of total octets received – upper 4 bytes.

Description

Total Octets Transmitted
TOTL (040C8h; R/W / TOTH (040CCh; R)
These registers make up a 64-bit register that counts the total number of octets transmitted. This
register resets each time the upper 32 bits are read (TOTH). In addition, it sticks at
FFFF_FFFF_FFFF_FFFFh when the maximum value is reached.
All transmitted packets have their octets summed into this register, regardless of their length or
whether they are flow control packets. This register includes bytes transmitted in a packet from the
field through the field, inclusively.
Octets transmitted as part of partial packet transmissions (collisions in half-duplex mode) are not
included in this register. This register only increments if transmits are enabled.
Table 13-131. TOTL and TOTH Register Bit Descriptions
31

0
TOTH

0
TOTL

Field

Bit(s)

Initial
Value

TOTL

31:0

Number of total octets transmitted – lower 4 bytes.

TOTH

31:0

Number of total octets transmitted – upper 4 bytes.

Software Developer’s Manual

Description

361

13.7.43

Total Packets Received
TPR (040D0h; R)
This register counts the total number of all packets received. All packets received are counted in
this register, regardless of their length, whether they have errors, or whether they are flow control
packets. This register only increments if receives are enabled.
Table 13-132. TPR Register Bit Description
31

0
TPR

Field
TPR

13.7.44

Bit(s)
31:0

Initial
Value
0b

Description
Number of all packets received.

Total Packets Transmitted
TPT (040D4h; R)
This register counts the total number of all packets transmitted. All packets transmitted are counted
in this register, regardless of their length, or whether they are flow control packets.
Partial packet transmissions (collisions in half-duplex mode) are not included in this register. This
register only increments if transmits are enabled. This register counts all packets, including
standard packets, secure packets, packets received over the SMBus1, and packets generated by the
ASF function.
Table 13-133. TPT Register Bit Description
31

0
TPT

Field

Bit(s)

Initial
Value

TPT

31:0

Description
Number of all packets transmitted.

The 82544GC/EI and the 82541ER do not support SMBus or ASF functionality.

362

Software Developer’s Manual

13.7.45

Packets Transmitted (64 Bytes) Count
PTC64 (040D8h; R)
This register counts the number of packets transmitted that are exactly 64 bytes (from through , inclusively) in length. Partial packet transmissions (collisions in halfduplex mode) are not included in this register. This register does not include transmitted flow
control packets (which are 64 bytes in length). This register only increments if transmits are
enabled. This register counts all packets, including standard packets, secure packets, packets
received over the SMBus, and packets generated by the ASF function.
Table 13-134. PTC64 Register Bit Description
31

0
PTC64

13.7.46

Field

Bit(s)

Initial
Value

PTC64

31:0

Description
Number of packets transmitted that are 64 bytes in length.

Packets Transmitted (65-127 Bytes) Count
PTC127 (040DCh; R)
This register counts the number of packets transmitted that are 65-127 bytes (from through , inclusively) in length. Partial packet transmissions (collisions in halfduplex mode) are not included in this register. This register only increments if transmits are
enabled. This register counts all packets, including standard packets, secure packets, packets
received over the SMBus1, and packets generated by the ASF function.
Table 13-135. PTC127 Register Bit Description
31

0
PTC127

Field

Bit(s)

Initial
Value

Description

PTC127

31:0

Number of packets transmitted that are 65-127 bytes in length.

The 82544GC/EI does not support SMBus or ASF functionality.

Software Developer’s Manual

363

13.7.47

Packets Transmitted (128-255 Bytes) Count
PTC255 (040E0h; R)
This register counts the number of packets transmitted that are 128-255 bytes (from through , inclusively) in length. Partial packet transmissions (collisions in halfduplex mode) are not included in this register. This register only increments if transmits are
enabled. This register counts all packets, including standard packets, secure packets, packets
received over the SMBus, and packets generated by the ASF function.
Table 13-136. PTC255 Register Bit Description
31

0
PTC255

13.7.48

Field

Bit(s)

Initial
Value

PTC255

31:0

Description
Number of packets transmitted that are 128-255 bytes in
length.

Packets Transmitted (256-511 Bytes) Count
PTC511 (040E4h; R)
This register counts the number of packets transmitted that are 256-511 bytes (from through , inclusively) in length. Partial packet transmissions (collisions in halfduplex mode) are not included in this register. This register only increments if transmits are
enabled. This register counts all packets, including standard packets, secure packets, packets
received over the SMBus1, and packets generated by the ASF function.
Table 13-137. PTC511 Register Bit Description
31

0
PTC511

Field

Bit(s)

Initial
Value

PTC511

31:0

Description
Number of packets transmitted that are 256-511 bytes in
length.

The 82544GC/EI does not support SMBus or ASF functionality.

364

Software Developer’s Manual

13.7.49

Packets Transmitted (512-1023 Bytes) Count
PTC1023 (040E8h; R)
This register counts the number of packets transmitted that are 512-1023 bytes (from through , inclusively) in length. Partial packet transmissions (collisions in halfduplex mode) are not included in this register. This register only increments if transmits are
enabled.This register counts all packets, including standard packets, secure packets, packets
received over the SMBus, and packets generated by the ASF function.
Table 13-138. PTC1023 Register Bit Description)
31

0
PTC1023

13.7.50

Field

Bit(s)

Initial
Value

Description

PTC1023

31:0

Number of packets transmitted that are 512-1023 bytes in length.

Packets Transmitted (1024 Bytes or Greater) Count
PTC1522 (040ECh; R)
This register counts the number of packets transmitted that are 1024 or more bytes (from
through , inclusively) in length. Partial packet transmissions
(collisions in half-duplex mode) are not included in this register. This register only increments if
transmits are enabled.
Due to the maximum frame size for VLAN tagged frames in IEEE Standard 802.3ac, these
Ethernet controllers transmit packets that have a maximum length of 1522 bytes. The RMON
statistics associated with this range has been extended to count 1522 byte long packets. This
register counts all packets, including standard packets, secure packets, packets received over the
SMBus1, and packets generated by the ASF function.
Table 13-139. PTC1522 Register Bit Description
31

0
PTC1522

Field

Bit(s)

Initial
Value

Description

PTC1522

31:0

Number of packets transmitted that are 1024 or more bytes in
length.

The 82544GC/EI does not support SMBus or ASF functionality.

Software Developer’s Manual

365

13.7.51

Multicast Packets Transmitted Count
MPTC (040F0h; R)
This register counts the number of multicast packets transmitted. This register does not include
flow control packets and increments only if transmits are enabled. Counts clear as well as secure
traffic.
Table 13-140. MPTC Register Bit Description
31

0
MPTC

13.7.52

Field

Bit(s)

Initial
Value

MPTC

31:0

Description
Number of multicast packets transmitted.

Broadcast Packets Transmitted Count
BPTC (040F4h; R)
This register counts the number of broadcast packets transmitted. This register only increments if
transmits are enabled.Counts clear as well as secure traffic. (Management packets are never more
than 200 bytes).
Table 13-141. BPTC Register Bit Description
31

0
BPTC

366

Field

Bit(s)

Initial
Value

BPTC

31:0

Description
Number of broadcast packets transmitted count.

Software Developer’s Manual

13.7.53

TCP Segmentation Context Transmitted Count
TSCTC (040F8h; R)
This register counts the number of TCP segmentation offload transmissions and increments once
the last portion of the TCP segmentation context payload is segmented and loaded as a packet into
the Ethernet controller’s on-chip transmit buffer. Note that this is not a measurement of the number
of packets sent out (covered by other registers). This register only increments if transmits and TCP
Segmentation offload are enabled.

0
TSCTC

Field
TSCTC

13.7.54

Bit(s)
31:0

Initial
Value
0b

Description
Number of TCP Segmentation contexts transmitted count.

TCP Segmentation Context Transmit Fail Count
TSCTFC (040FCh; R)
This register counts the number of TCP segmentation offload requests to the hardware that failed to
transmit all data in the TCP segmentation context payload. There is no indication by hardware of
how much data was successfully transmitted. Only one failure event is logged per TCP
segmentation context. Failures can be caused by excessive collisions or PAYLEN errors. This
register only increments if transmits are enabled.

0
TSCTFC

Field

TSCTFC

Software Developer’s Manual

Bit(s)

31:0

Initial
Value
0b

Description
Number of TCP Segmentation contexts where the Ethernet
controller failed to transmit the entire data payload.

367

13.8

Diagnostics Registers
The Ethernet controller contains several diagnostic registers. These registers enable software to
directly access the contents of the Ethernet controller’s internal Packet Buffer Memory (PBM), also
referred to as FIFO space. These registers also give software visibility into what locations in the
PBM that the hardware currently considers to be the “head” and “tail” for both transmit and receive
operations.

13.8.1

Receive Data FIFO Head Register
RDFH (02410h; R/W)
This register stores the head of the Ethernet controller’s on–chip receive data FIFO. Since the
internal FIFO is organized in units of 64-bit words, this field contains the 64-bit offset of the
current Receive FIFO Head. So a value of “8h” in this register corresponds to an offset of 8
quadwords into the Receive FIFO space. This register is available for diagnostic purposes only, and
should not be written during normal operation.
Table 13-142. RDFH Register Bit Description
31

13 12

Reserved

13.8.2

FIFO Head

Field

Bit(s)

Initial
Value

FIFO Head

12:0

Receive FIFO Head pointer.

Reserved

31:13

Reads as 0b. Should be written to 0b for future compatibility.

Description

Receive Data FIFO Tail Register
RDFT (02418h; R/W)
This register stores the tail of the Ethernet controller’s on–chip receive data FIFO. Since the
internal FIFO is organized in units of 64-bit words, this field contains the 64-bit offset of the
current Receive FIFO Tail. So a value of “8h” in this register corresponds to an offset of eight
quadwords or into the Receive FIFO space. This register is available for diagnostic purposes only,
and should not be written during normal operation.

368

Software Developer’s Manual

Table 13-143. RDFT Register Bit Description
31

13 12

Reserved

13.8.3

FIFO Tail

Field

Bit(s)

Initial
Value

FIFO Tail

12:0

Receive FIFO Tail pointer.

Reserved

31:13

Reads as 0b. Should be written to 0b for future compatibility.

Description

Receive Data FIFO Head Saved Register
RDFHS (02420h; R/W)
This register stores a copy of the Receive Data FIFO Head register in case the internal register
needs to be restored. This register is available for diagnostic purposes only, and should not be
written during normal operation.
Table 13-144. RDFHS Register Bit Description
31

13 12

Reserved

13.8.4

FIFO Head

Field

Bit(s)

Initial
Value

FIFO Head

12:0

A “saved” value of the Receive FIFO Head pointer.

Reserved

31:13

Reads as 0b. Should be written to 0b for future compatibility.

Description

Receive Data FIFO Tail Saved Register
RDFTS (02428h; R/W)
This register stores a copy of the Receive Data FIFO Tail register in case the internal register needs
to be restored. This register is available for diagnostic purposes only, and should not be written
during normal operation.

Software Developer’s Manual

369

Table 13-145. RDFTS Register Bit Description
31

13 12
Reserved

13.8.5

0
FIFO Tail

Field

Bit(s)

Initial
Value

FIFO Tail

12:0

A “saved” value of the Receive FIFO Tail pointer.

Reserved

31:13

Reads as 0b. Should be written to 0b for future compatibility.

Description

Receive Data FIFO Packet Count
RDFPC (02430h; R/W)
This register reflects the number of receive packets that are currently in the Receive FIFO. This
register is available for diagnostic purposes only, and should not be written during normal
operation.
Table 13-146. RDFPC Register Bit Description
31

13 12
Reserved

13.8.6

0
FIFO Tail

Field

Bit(s)

Initial
Value

RX FIFO
Packet Count

12:0

The number of received packets currently in the RX FIFO.

Reserved

31:13

Reads as 0b. Should be written to 0b for future compatibility.

Description

Transmit Data FIFO Head Register
TDFH (03410h; R/W)
This register stores the head of the Ethernet controller’s on–chip transmit data FIFO. Since the
internal FIFO is organized in units of 64-bit words, this field contains the 64-bit offset of the
current Transmit FIFO Head. So a value of “8h” in this register corresponds to an offset of 8
quadwords into the Transmit FIFO space. This register is available for diagnostic purposes only,
and should not be written during normal operation.

370

Software Developer’s Manual

Table 13-147. TDFH Register Bit Description)
31

11 10
Reserved

13.8.7

0
FIFO Head

Field

Bit(s)

Initial
Value

FIFO Head

10:0

Transmit FIFO Head pointer.

Reserved

31:11

Reads as 0b. Should be written to 0b for future compatibility.

Description

Transmit Data FIFO Tail Register
TDFT (03418h; R/W)
This register stores the head of the Ethernet controller’s on–chip transmit data FIFO. Since the
internal FIFO is organized in units of 64-bit words, this field contains the 64-bit offset of the
current Transmit FIFO Tail. So a value of “8h” in this register corresponds to an offset of 8
quadwords into the Transmit FIFO space. This register is available for diagnostic purposes only,
and should not be written during normal operation.
Table 13-148. TDFT Register Bit Description
31

11 10
Reserved

13.8.8

0
FIFO Tail

Field

Bit(s)

Initial
Value

FIFO Tail

10:0

Transmit FIFO tail pointer.

Reserved

31:11

Reads as 0b. Should be written to 0b for future compatibility.

Description

Transmit Data FIFO Head Saved Register
TDFHS (03420h; R/W)
This register stores a copy of the Transmit Data FIFO Head register in case the internal register
needs to be restored. This register is available for diagnostic purposes only, and should not be
written during normal operation.

Software Developer’s Manual

371

Table 13-149. TDFHS Register Bit Description
31

13 12

Reserved

13.8.9

FIFO Head

Field

Bit(s)

Initial
Value

FIFO Head

12:0

A “saved” value of the Transmit FIFO Head pointer.

Reserved

31:13

Reads as 0b. Should be written to 0b for future compatibility.

Description

Transmit Data FIFO Tail Saved Register
TDFTS (03428h; R/W)
This register stores a copy of the Transmit Data FIFO Tail register in case the internal register
needs to be restored. This register is available for diagnostic purposes only, and should not be
written during normal operation.
Table 13-150. TDFTS Register Bit Description
31

13 12

Reserved

13.8.10

FIFO Tail

Field

Bit(s)

Initial
Value

FIFO Tail

12:0

A “saved” value of the Transmit FIFO Tail pointer.

Reserved

31:13

Reads as 0b. Should be written to 0b for future compatibility.

Description

Transmit Data FIFO Packet Count
TDFPC (03430h; R/W)
This register reflects the number of packets to be transmitted that are currently in the Transmit
FIFO. This register is available for diagnostic purposes only, and should not be written during
normal operation.

372

Software Developer’s Manual

Table 13-151. TDFPC Register Bit Description
31

13 12

Reserved

13.8.11

FIFO Tail

Field

Bit(s)

Initial
Value

FIFO Tail

12:0

The number of packets to be transmitted that are currently in
the TX FIFO.

Reserved

31:13

Reads as 0b. Should be written to 0b for future compatibility.

Description

Packet Buffer Memory
PBM (10000h - 1FFFCh; R/W)
All PBM (FIFO) data is available to diagnostics. Locations can be accessed as 32-bit or 64-bit
words. The internal PBM is 64 KB (40 KB for the 82547GI/EI) in size. Software can configure the
amount of PBM space that is used as the transmit FIFO versus the receive FIFO. The default is 16
KB of transmit FIFO space and 48 KB of receive FIFO space. For the 82547GI/EI, the default is
18 KB of transmit FIFO space and 22 KB of receive FIFO space.
Regardless of the individual FIFO sizes that software configures, the RX FIFO is located first in
the memory mapped PBM space. So for the default FIFO configuration, the RX FIFO occupies
offsets 10000h - 1BFFFh of the memory mapped space, while the TX FIFO occupies offsets
1C000h - 1FFFFh of the memory mapped space.
Table 13-152. PBM Bit Description
31

0
FIFO Data

Field

Bit(s)

Initial
Value

FIFO Data

31:0

Software Developer’s Manual

Description
Packet Buffer Data

373

Note:

374

This page intentionally left blank.

Software Developer’s Manual

General Initialization and Reset Operation

General Initialization and Reset
Operation
14.1

Introduction
This section lists all necessary initializations and describes the reset commands for the PCI/PCI-X
Family of Gigabit Ethernet Controllers.
Note:

14.2

TBI mode is used by the 82544GC/EI. Internal SerDes is used by the 82546GB/EB and
82545GM/EM.

Power Up State
At power up, the Ethernet controller is not automatically configured by the hardware for normal
operation. Software initialization is required before normal operation can continue. In general, the
Ethernet controller is considered non-functional until the software driver successfully loads and
sets up the hardware. However, Auto-Negotiation can start at power up or upon receipt of an
assertion of PCI reset if configured to do so by the EEPROM.

14.3

General Configuration
Several values in the Device Control Register (CTRL) need to be set upon power up or after an
Ethernet controller reset for normal operation.

• Speed and duplex are determined via Auto-Negotiation by the PHY, Auto-Negotiation by the
MAC for internal SerDes1 mode, or forced by software if the link is forced. In internal PHY
mode, the Ethernet controller can be configured automatically by hardware or forced by
software to the same configuration as the PHY.

• In internal PHY mode, the Auto-Speed Detection Enable (CTRL.ASDE) bit, when set to 1b,
detects the resolved speed and duplex of the link and self-configure the MAC appropriately.
This bit should be set in conjunction with the Set Link Up (CTRL.SLU) bit.

• The MAC can also be forced to a specific Speed/Duplex combination. This is accomplished by
setting the Set Link Up (CTRL.SLU), Force Speed (CTRL. FRCSPD) and Force Duplex
(CTRL.FRCDPLX) bits. Once speed and duplex are determined (either via Auto-Negotiation
or forced by software), speed is forced by setting the appropriate Speed Selection
(CTRL.SPEED) bits and duplex is forced by updating the Full Duplex (CTRL.FD) bit.

• For the 82541xx and 82547GI/EI, configure the LED behavior through LEDCTRL.
• Link Reset (CTRL.LRST) should be set to 0b (normal). The Ethernet controller defaults to
LRST = 1b which disables Auto-Negotiation. A transition to 0b initiates the Auto-Negotiation
function. LRST can be defined in the EEPROM. This bit is only valid in internal SerDes mode
and has no effect in internal PHY mode.
1.

The 82540EP/EM, 82541xx, and 82547GI/EI do not support any SerDes functionality.

Software Developer’s Manual

375

General Initialization and Reset Operation

• PHY Reset (CTRL.PHY_RST) should be set to 0b. Setting this bit to 1b resets the PHY
without accessing the PHY registers. This bit is ignored in internal SerDes mode.

• CTRL.ILOS should be set to 0b (not applicable to the 82541xx and 82547GI/EI).
• If Flow Control is desired, program the FCAH, FCAL, FCT and FCTTV registers. If not, they
should be written with 0b. To enable XON frame transmission, the XON Enable
(FCTRL.XONE) bit must be set. Advertising Flow Control capabilities during the AutoNegotiation process is dependent on whether the Ethernet controller is operating in internal
SerDes or internal PHY mode. In internal SerDes mode, the TXCW register must be set up
prior to starting the Auto-Negotiation process. In internal PHY mode, the appropriate PHY
registers must be set up properly to advertise desired capabilities prior to starting or re-starting
the Auto-Negotiation process. The Receive Flow Control Enable (CTRL.RFCE) and Transmit
Flow Control Enable (CTRL.TFCE) bits need to be explicitly set by software in internal PHY
mode (because Auto-Negotiation is managed by PHY rather than the MAC), or when a fiber
connection is desired but link was forced rather than Auto-Negotiated.

• If VLANs are not used, software should clear VLAN Mode Enable (CTRL.VME) bit. In this
instance, there is no need then to initialize the VLAN Filter Table Array (VFTA). If VLANs
are desired, the VFTA should be both initialized and loaded with the desired information.

• For the 82541xx and 82547GI/EI, clear all statistical counters.

14.4

Receive Initialization
Program the Receive Address Register(s) (RAL/RAH) with the desired Ethernet addresses.
RAL[0]/RAH[0] should always be used to store the Individual Ethernet MAC address of the
Ethernet controller. This can come from the EEPROM or from any other means (for example, on
some machines, this comes from the system PROM not the EEPROM on the adapter port).
Initialize the MTA (Multicast Table Array) to 0b. Per software, entries can be added to this table as
desired.
Program the Interrupt Mask Set/Read (IMS) register to enable any interrupt the software driver
wants to be notified of when the event occurs. Suggested bits include RXT, RXO, RXDMT,
RXSEQ, and LSC. There is no immediate reason to enable the transmit interrupts.
If software uses the Receive Descriptor Minimum Threshold Interrupt, the Receive Delay Timer
(RDTR) register should be initialized with the desired delay time.
Allocate a region of memory for the receive descriptor list. Software should insure this memory is
aligned on a paragraph (16-byte) boundary. Program the Receive Descriptor Base Address
(RDBAL/RDBAH) register(s) with the address of the region. RDBAL is used for 32-bit addresses
and both RDBAL and RDBAH are used for 64-bit addresses.
Set the Receive Descriptor Length (RDLEN) register to the size (in bytes) of the descriptor ring.
This register must be 128-byte aligned.
The Receive Descriptor Head and Tail registers are initialized (by hardware) to 0b after a power-on
or a software-initiated Ethernet controller reset. Receive buffers of appropriate size should be
allocated and pointers to these buffers should be stored in the receive descriptor ring. Software
initializes the Receive Descriptor Head (RDH) register and Receive Descriptor Tail (RDT) with the
appropriate head and tail addresses. Head should point to the first valid receive descriptor in the
descriptor ring and tail should point to one descriptor beyond the last valid descriptor in the
descriptor ring.

376

Software Developer’s Manual

General Initialization and Reset Operation

Program the Receive Control (RCTL) register with appropriate values for desired operation to
include the following:

• Set the receiver Enable (RCTL.EN) bit to 1b for normal operation. However, it is best to leave
the Ethernet controller receive logic disabled (RCTL.EN = 0b) until after the receive
descriptor ring has been initialized and software is ready to process received packets.

• Set the Long Packet Enable (RCTL.LPE) bit to 1b when processing packets greater than the
standard Ethernet packet size. For example, this bit would be set to 1b when processing Jumbo
Frames.

• Loopback Mode (RCTL.LBM) should be set to 00b for normal operation.
• Configure the Receive Descriptor Minimum Threshold Size (RCTL.RDMTS) bits to the
desired value.

• Configure the Multicast Offset (RCTL.MO) bits to the desired value.
• Set the Broadcast Accept Mode (RCTL.BAM) bit to 1b allowing the hardware to accept
broadcast packets.

• Configure the Receive Buffer Size (RCTL.BSIZE) bits to reflect the size of the receive buffers
software provides to hardware. Also configure the Buffer Extension Size (RCTL.BSEX) bits if
receive buffer needs to be larger than 2048 bytes.

• Set the Strip Ethernet CRC (RCTL.SECRC) bit if the desire is for hardware to strip the CRC
prior to DMA-ing the receive packet to host memory.

• For the 82541xx and 82547GI/EI, program the Interrupt Mask Set/Read (IMS) register to
enable any interrupt the driver wants to be notified of when the even occurs. Suggested bits
include RXT, RXO, RXDMT, RXSEQ, and LSC. There is no immediate reason to enable the
transmit interrupts. Plan to optimize interrupts later, including programming the interrupt
moderation registers TIDV, TADV, RADV and IDTR.

• For the 82541xx and 82547GI/EI, if software uses the Receive Descriptor Minimum
Threshold Interrupt, the Receive Delay Timer (RDTR) register should be initialized with the
desired delay time.

14.5

Transmit Initialization
Allocate a region of memory for the transmit descriptor list. Software should insure this memory is
aligned on a paragraph (16-byte) boundary. Program the Transmit Descriptor Base Address
(TDBAL/TDBAH) register(s) with the address of the region. TDBAL is used for 32-bit addresses
and both TDBAL and TDBAH are used for 64-bit addresses.
Set the Transmit Descriptor Length (TDLEN) register to the size (in bytes) of the descriptor ring.
This register must be 128-byte aligned.
The Transmit Descriptor Head and Tail (TDH/TDT) registers are initialized (by hardware) to 0b
after a power-on or a software initiated Ethernet controller reset. Software should write 0b to both
these registers to ensure this.
Initialize the Transmit Control Register (TCTL) for desired operation to include the following:

• Set the Enable (TCTL.EN) bit to 1b for normal operation.
• Set the Pad Short Packets (TCTL.PSP) bit to 1b.

Software Developer’s Manual

377

General Initialization and Reset Operation

• Configure the Collision Threshold (TCTL.CT) to the desired value. Ethernet standard is 10h.
This setting only has meaning in half duplex mode.

• Configure the Collision Distance (TCTL.COLD) to its expected value. For full duplex
operation, this value should be set to 40h. For gigabit half duplex, this value should be set to
200h. For 10/100 half duplex, this value should be set to 40h.
Program the Transmit IPG (TIPG) register with the following decimal values to get the minimum
legal Inter Packet Gap:

378

Fiber

Copper

Fiber
(82544GC/EI

Copper
(82544GC/EI

IPGT

IPGR1

IPGR2

Applicable to the 82541xx and 82547GI/EI.

Software Developer’s Manual

General Initialization and Reset Operation

Note:

IPGR1 and IPGR2 are not needed in full duplex, but are easier to always program to the values
shown.
Table 14-1. Signal Descriptions
Signal
LOS / LINK

Ball

Name and Function

A10

Loss of Signal (TBI) / Link Indication. Loss of signal (high for lost
signal) from the optical transceiver when LINK_MODE equals 11b;
active high link indication from PHY in GMII/MII mode.

C7
TX_DATA[9] / TX_ER
TX_DATA[8] / TX_EN
TX_DATA[7]
TX_DATA[6]
TX_DATA[5]
TX_DATA[4]
TX_DATA[3]
TX_DATA[2]
TX_DATA[1]
TX_DATA[0]

D7
E6
B5
E5
C5
E4
C4

Transmit Data.
TBI: TX_DATA[9:0] for transmit data bus.
GMII: TX_DATA[7:0] for transmit data bus.
TX_ER forces propagation of transmit errors and is used for carrier
extension. TX_EN is asserted to indicate transmission of data on the
interface.
MII: TX_DATA[3:0] for transmit data bus.
TX_ER is not used. TX_EN is used for transmit enable signal.

D5
D4

GTX_CLK

EWRAP

COL

Software Developer’s Manual

Transmit Clock.
TBI: 125 MHz transmit clock.
GMII: Operates at 125 MHz.
MII: Undefined.

E10

Enable Wrap.
TBI: EWRAP is low in normal operation. When it is high, the SerDes
device is forced to transceiver loopback the serialized transmit data
to the receiver.
This pin is tri-stated during EEPROM read. In order to avoid a
floating input in an external SerDes, a weak external pull-down
should be connected to this pin.
GMII / MII: Not used.

Collision.
TBI: Undefined.
GMII / MII: This signal indicates that a collision was detected on the
medium by the PHY. This signal remains asserted while the collision
persists. For half-duplex transceivers, this signal indicates
simultaneous transmission and reception. This signal is ignored in
full-duplex mode.
Normal Mode: This signal must be connected to VSS except for test
mode.

379

General Initialization and Reset Operation

Signal

CRS

Ball

Name and Function

Carrier Sense.
TBI: Undefined.
GMII / MII: This signal indicates traffic activity on the cable, either
incoming or outgoing. This signal is driven by the PHY. CS is not
required to transition synchronously with respect to the RX or TX
clocks. This signal is ignored in full-duplex mode.
Normal Mode: This signal must be connected to VSS except for test
mode.

A9
RX_DATA[9] / RX_ER
RX_DATA[8] / RX_DV
RX_DATA[7]
RX_DATA[6]
RX_DATA[5]
RX_DATA[4]
RX_DATA[3]
RX_DATA[2]
RX_DATA[1]
RX_DATA[0]

D10
B9
C9
D9
E9
E8
C8

Receive Data.
TBI: RX_DATA[9:0] for receive data bus
GMII: RX_DATA[7:0] for receive data bus.
RX_ER signals a receive error. RX_DV is asserted to indicate data is
valid on the interface.
MII: RX_DATA[3:0] for receive data bus.
RX_ER signals a receive error. RX_DV indicates data is valid on the
interface.

A7
B7

RBC0 / RX_CLK

14.5.1

C11

Receive Clock 0.
TBI: RBC0 is receive clock (62.5 Mbps).
GMII: RX_CLK is receive clock (125 Mbps).
MII: RX_CLK is receive clock for 100 Mbps operation (25 Mbps) and
for 10 Mbps operation (2.5 Mbps).

Signal Interface
The external GMII/MII interface is similar in function to the interface used to communicate
between the MAC and internal PHY. As with use of the internal PHY, the external GMII/MII
interface supports 10/100/1000 Mbps operation, with both half- and full-duplex operation at 10/
100 Mbps, and full-duplex operation at 1000 Mbps. Unlike the communication path to the internal
PHY, the external interface does not provide certain additional control/status interfaces for
automatic hardware link setup and/or power-management
Table 14-2lists the signals, functions, and pins used to provide this interface.

380

Software Developer’s Manual

General Initialization and Reset Operation

Table 14-2. Signal Functions
Signal

Function

GMII (1000 Mbps) Operations
CRS

Carrier Sense

CRS

COL

Collision Detect

COL

TX_ER

Transmit Code Error

TX_DATA[9]/TX_ER

TX_EN

Transmit Enable

TX_DATA[8]/TX_EN

GTX_CLK

Transmit Data Clock (125 MHz)

GTX_CLK

TX_DATA

Transmit Data

TX_DATA[7:0]

RX_CLK

Receive Data Clock (125 MHz)

RBC0/RX_CLK

RX_DATA

Receive Data

RX_DATA[7:0]

RX_ER

Receive Error

RX_DATA[9]/RX_ER

RX_DV

Receive Data Valid

RX_DATA[8]/RX_DV

LINK

PHY Link Indication

LOS/LINK

MII (10/100 Mbps) Differences
MTX_CLK

14.5.2

Transmit Data Clock (25/2.5 MHz)

RBC1/MTX_CLK

TX_DATA

Transmit Data

TX_DATA[3:0]

RX_CLK

Receive Data Clock (25/2.5 MHz)

RBC0/RX_CLK

RX_DATA

Receive Data

RX_DATA[3:0]

GMII/MII Features not Supported
Table 14-3 lists the signals and functions not provided by this interface.

Software Developer’s Manual

381

General Initialization and Reset Operation

Table 14-3. Signal Functions Not Supported
Signal

Function

Ramifications

MII Management Interface (PHY Register Access)
MDC

Management Data Clock

MDI/O

Management Data I/O

No support/access to MII register set.
Direct PHY Indications to MAC

14.5.3

FDX

PHY-negotiated full/half duplex indication

Can limit use to specific known duplex
setting.

SPD_IND

PHY-negotiated speed (10/100/1000 Mbps)

Can limit use to specific known speed or
require use of auto-speed detection.

Avoiding GMII Test Mode(s)
Note that the Ethernet controller contains a set of test modes that use this interface for component
manufacturing and/or diagnostic test. To avoid accidental engagement of unexpected test mode(s)
when using the external GMII (or TBI), the TEST_GMII[2:0] test pins must remain de-asserted
(low) and the TEST_DM_N pin must remain de-asserted (high).

14.5.4

MAC Configuration
The Ethernet controller MAC operates in a GMII/MII mode when operating with the internal PHY;
this mode is similar to the GMII/MII mode of the standalone 82543 MAC components and others.
In GMII/MII mode, the MAC operates assuming use of a GMII/MII interface communication,
variable duplex & speed configuration (unless forced or auto-detected). For the Ethernet
controller, to use this external interface as a GMII/MII interface and have the MAC operate in this
GMII/MII Mode, the LINK_MODE must be set to 01b.
It is likely that the MAC might be required to be configured in a forced-duplex configuration, as no
means is provided (either the MDI/O access or direct PHY-to-MAC signaling) of any duplex
configuration that might be negotiated between the attached Ethernet controller/transceiver and its
link partner.
The MAC can further be required to be configured in a forced-speed configuration, as no direct
speed indication is available via the external interface (compared to the SPD_IND signals provided
by the internal PHY). The Auto-Speed detection (ASD) can be potentially useful in automatically
calculating and configuring a speed setting based in the interface signals that are provided.
The MAC is unable to provide any access to MII Management registers through the MDIC register,
as no explicit MDI/O signals are included in this interface. However, it is possible that softwaredefinable pins (SDP) can be capable of providing the necessary access capability.

382

Software Developer’s Manual

General Initialization and Reset Operation

14.5.5

Link Setup
The following examples are provided as suggestions for configuring common settings between the
MAC and an Ethernet controller attached in the GMII/MII mode.
 MAC duplex and speed settings forced by software based on resolution of PHY
(CTRL.FRCDPLX = 1b, CTRL.FRCSPD = 1b, CTRL.ASDE = don’t care)
CTRL.FD Set by software based on reading PHY status register after
PHY has autonegotiated a successful link-up.
CTRL.SLU Must be set to 1b by software to enable communications
between MAC and PHY
CTRL.RFCE Must be set by S/W after reading flow control resolution from
PHY registers
CTRL.TFCE - Must be set by S/W after reading flow control resolution from PHY registers
CTRL.SPEED Set by software based on reading PHY status register after
PHY has autonegotiated a successful link-up.
STATUS.FD Reflects the MAC forced duplex setting written to CTRL.FD
STATUS.LU Reflects link indication (LINK) from PHY qualified with
CTRL.SLU (set to 1b)
STATUS.SPEED Reflects MAC forced speed setting written in
CTRL.SPEED


MAC duplex setting forced by software based on resolution of PHY; speed auto-detected by
MAC
(CTRL.FRCDPLX = 1b, CTRL.FRCSPD = 0b, CTRL.ASDE = 1b)
CTRL.FD Set by software based on reading PHY status register after
PHY has autonegotiated a successful link-up.
CTRL.SLU Must be set to 1b by software to enable communications
between MAC and PHY
CTRL.RFCE Must be set by S/W after reading flow control resolution from
PHY registers
CTRL.TFCE ..............
Must be set by S/W after reading flow control resolution from PHY
registers
CTRL.SPEED Don’t care; speed setting is calculated by the MAC based
on signals from the PHY after PHY has autonegotiated a successful
link-up
STATUS.FD Reflects the MAC forced duplex setting written to CTRL.FD
STATUS.LU Reflects link indication (LINK) from PHY qualified with
CTRL.SLU (set to 1b)
STATUS.SPEED Reflects actual speed setting calculated by MAC ASD
function

Software Developer’s Manual

383

General Initialization and Reset Operation



MAC/PHY duplex and speed settings both forced by software (fully-forced link setup)
(CTRL.FRCDPLX = 1b, CTRL.FRCSPD = 1b, CTRL.SLU = 1b)
CTRL.FD ...................Set by software to desired full/half duplex operation (must match
duplex setting of PHY)
CTRL.SLU.................Must be set to 1b by software to enable communications between MAC
and PHY. PHY must also be forced/configured to indicate positive link
indication (LINK) to the MAC
CTRL.RFCE ..............Must be set by S/W to desired flow-control operation (must match flowcontrol settings of PHY)
CTRL.TFCE ..............Must be set by S/W to desired flow-control operation (must match flowcontrol settings of PHY)
CTRL.SPEED ............Set by software to desired link speed (must match speed setting of
PHY)
STATUS.FD ...............Reflects the MAC duplex setting written by software to CTRL.FD
STATUS.LU...............Reflects 1b (positive link indication LINK from PHY qualified with
CTRL.SLU). Note: since both CTRL.SLU and the PHY link indication
LINK are forced, this bit set does not GUARANTEE that operation of
the link has been truly established.
STATUS.SPEED ........Reflects MAC forced speed setting written in CTRL.SPEED

Note:

14.6

It is important to note that for the Ethernet controller’s link indication (LINK) to be noted by the
MAC, the MAC control bit CTRL.SLU must be set to 1b. Normal MAC/PHY speed and duplex
configuration are based on observing events on this link indication from the Ethernet controller.

PHY Initialization (10/100/1000 Mb/s Copper Media)
Software needs to determine the PHY address at which the PHY actually resides. This number can
be anywhere from 0 to 31.The PHY address is programmable. Board designers can then choose at
what PHY address the PHY resides. Software needs to identify the PHY address so that the PHY
can be accessed successfully.
To accomplish read and write access to any of the PHY registers, software must program the MDI
Control Register (MDIC) with the appropriate data. A PHY is reset at power-up and is enabled to
Auto-Negotiate by default. Typically in most environments, by the time the software driver is
loaded, the Auto-Negotiation process has completed. However, the PHY might or might not
advertise the appropriate capabilities desired by the design. In this instance, it is up to the software
to insure that the PHY registers are set up properly to advertise the appropriate Ethernet controller
capabilities. For example, by default the Ethernet controller advertises no flow control capabilities
in its Auto-Negotiation Advertisement Register (MII Register 4). In order to advertise TX and/or
RX Pause capabilities, this register must be modified and Auto-Negotiation re-started to advertise
these capabilities to the link partner.
The MII Status Register (PHY Register 1) should be used to check link status.
Software can also force the speed/duplex of a PHY via MII/GMII register access. Note that forcing
gigabit speed in a copper environment is not allowed per IEEE specification. Only 10/100 speed
and duplex should be forced in the PHY.

384

Software Developer’s Manual

General Initialization and Reset Operation

Once link is achieved by the PHY, software is notified when a Link Status Change (LSC) interrupt
is generated by the Ethernet controller. This only occurs if software enabled the LSC bit in the
Interrupt Mask Set/Read (MS) Register.

14.7

Reset Operation
The following reset signals affect the Ethernet controller in different ways. RST# is the only
external signal. Other reset events are asserted by performing slave writes to specific bits in the
control registers.
Values indicated as “?” imply the default value is either unknown or is read from the EEPROM.
Note:

In situations where the TX block is reset, the TX data lines are forced to all 0b’s. This causes a
substantial number of symbol errors to be detected by the link partner. In TBI mode (82544GC/
EI)/internal SerDes (82546GB/EB and 82545GM/EM), if the duration is long enough, the link
partner can restart the Auto-Negotiation process by sending “break-link” (/C/ codes with the
configuration register value set to all 0b’s).
LAN_PWR_GOOD:

Deasserting LAN_PWR_GOOD resets all resettable registers in the Ethernet controller. The signal
is level-sensitive, and the Ethernet controller is held in reset until LAN_PWR_GOOD is asserted.
While asserted, all PCI signals are forced to a high impedance state.
General Registers:
Interrupt Registers:
Receive Registers:

Transmit Registers:
Statistics Registers:
Wakeup Registers:

Diagnostic Registers:
PCI Config Space:
PHY:

Reset to power-on values.
Reset to power-on values.
Reset to power-on values (exceptions are the RAH/RAL, MTA, VFTA
and RDBAH/RDBAL registers, which are not reset to any preset
value. The valid bit of the RAH register is cleared).
Reset to power-on values (exceptions are the TDBAH/TDBAL registers, which are not reset to any preset value).
Reset to power-on values.
The WUC (except for the PME_En and PME_Status bits if
AUX_POWER = 1b), WUFC, IPAV, and FFLT registers are reset to
their default value.
Reset to power-on values (exception is the PBM memory, which is not
reset to any preset value).
Context Lost; requires initialization.
RST# is asserted to reset the PHY while LAN_PWR_GOOD is deasserted.

In addition, the Ethernet controller automatically reads certain values from the EEPROM and
configures itself to use those EEPROM settings.

Software Developer’s Manual

385

General Initialization and Reset Operation

RST#:

When asserted, all PCI signals are forced to a high impedance state. Upon deassertion, the Ethernet
controller’s internal registers, excluding the following exceptions, are reset.
General Registers:
Reset to power-on values.
Interrupt Registers:
Reset to power-on values.
Receive Registers:
Reset to power-on values (exceptions are the RAH/RAL, MTA,
VFTA and RDBAH/RDBAL registers, which are not reset to any
preset value. The valid bit of the RAH register is cleared).
Transmit Registers:
Reset to power-on values (exceptions are the TDBAH/TDBAL,
and TIPG registers, which is not reset to any preset value).
Statistics Registers:
Reset to power-on values.
Wakeup Registers:
The WUC (except for the PME_En and PME_Status bits if
AUX_POWER = 1b), WUFC, IPAV, and FFLT registers are reset
to their default value.
Diagnostic Registers:
Reset to power-on values (exception is the PBM memory, which is
not reset to any preset value).
PCI Config Space:
Context Lost; requires initialization. If AUX_POWER = 1b then
the PME_En and PME_Status bits of the Power Management
Control/Status Register are preserved.
PHY:
RST# is asserted for 400 ns after deassertion of RST#.
Asserting RST# puts the Ethernet controller into the “Dr” Power Management state. See Section
6.3.1.1 for details on the power states, and Section 6.3.2.4 for reset related timing.
Deasserting RST# also causes the EEPROM to be re-read and the registers that get values from the
EEPROM to be re-loaded.
Global Reset:

Bit 26 of the Device Control Register (CTRL.RST) performs an Ethernet controller reset of all
functions to their equivalent power on state similar to asserting RST#, except that the state of the
PCI core and PCI configuration space is not affected.
General Registers:
Interrupt Registers:
Receive Registers:

Transmit Registers:
Statistics Registers:
Wakeup Registers:
Diagnostic Registers:
PCI Config Space:
PHY:

386

Reset to power-on values.
Reset to power-on values.
Reset to power-on values (exceptions are the RAH/RAL, MTA, VFTA
and RDBAH/RDBAL registers, which are not reset to any preset
value. The valid bit of the RAH register is cleared).
Reset to power-on values (exceptions are the TDBAH/TDBAL, and
TIPG registers).
Reset to power-on values.
The WUC (except for the PME_En and PME_Status bits), WUFC,
IPAV, and FFLT registers are reset to their default value.
Reset to power-on values (exception is the PBM memory, which is not
reset to any preset value).
No Change.
No effect.

Software Developer’s Manual

General Initialization and Reset Operation

Default values for certain bits of the Device Control Register must be read out of the EEPROM and
appropriately set by software if an EEPROM is used.
Global Reset does NOT affect the direction of the software programmable pins.
Link_Reset:

When LRST (bit 3 of the Device Control register) is written as a logic 1b, the Ethernet controller is
forced into a link reset state. When LRST is set to 1b the Auto-Negotiation function is disabled.
The Auto-Negotiation logic is initiated/restarted when LRST is transitions to 0b. A link reset is
only relevant in TBI mode/internal SerDes (not applicable to the 82540EP/EM, 82541xx and
82547GI/EI).
The transmitter sends /C/ ordered_sets when LRST is asserted.
General Registers:
No change.
Interrupt Registers:
No change.
Receive Registers:
The RXCW register is cleared.
Transmit Registers:
No change.
Statistics Registers:
No change.
Wakeup Registers:
No change.
Diagnostic Registers:
No change.
PHY:
No effect.
EE_RST (Extended Device Control Register):

EEPROM reset bit. Initiates a “reset-like” event to the EEPROM function that causes the
EEPROM to be read again. Control registers bits are not affected other than those read from the
EEPROM.
PHY_RST (Device Control Register):

PHY reset bit in the Device Control Register. By writing a 1b to this bit the software forces the
assertion of an internal signal output to reset the PHY device without accessing the PHY registers
through the MII management interface (MDI/O & MDC). Internal states of the Ethernet controller
are not impacted. To release the PHY reset the software must write a 0b to the bit.
In situations where the Ethernet controller is reset using the software reset CTRL.RST, the TX data
lines are forced to all 0b’s. This causes a substantial number of symbol errors to be detected by the
link partner. In TBI mode/internal SerDes, if the duration is long enough, the link partner can
restart the Auto-Negotiation process by sending “break-link” (/C/ codes with the config register
value set to all 0b’s).
Some registers mentioned above within the Ethernet controller are treated specially. The RAH/
RAL[n], MTA[n], VFTA[n], WUPM[n], FFMT[n], FFVT[n], TDBAH/TDBAL, and RDBAH/
RDBAL registers have no default value and if the functions associated with the registers are
enabled they must be programmed by software. Once programmed, their value is preserved
through all resets as long as power is applied to the Ethernet controller. Bit 31, the valid bit, of the
RAH[n] registers is the exception and is reset with the LAN_PWR_GOOD and RST# and software
reset (CTRL.RST) bit.

Software Developer’s Manual

387

General Initialization and Reset Operation

Driver accessible Wakeup Status registers are excluded from all resets except for
LAN_PWR_GOOD. This includes:

• Wakeup Status Register.
• Wakeup Packet Length.
• Wakeup Packet Memory.
Finally, the “Wakeup Context” as defined in the PCI Bus Power Management Interface
Specification is reset on LAN_PWR_GOOD, and is also reset on the deassertion of RST# if
AUX_POWER = 0b. This includes:

• PME_En bit of the Power Management Control/Status Register (PMCSR).
• PME_Status bit of the Power Management Control/Status Register (PMCSR).
The shadow copies of these bits in the Wakeup Control Register are treated identically.

14.8

Initialization of Statistics
Statistics registers are hardware-initialized to values as detailed in each particular register’s
description. The initialization of these registers begins upon transition to D0active power state
(when internal registers become accessible, as enabled by setting the Memory Access Enable of the
PCI Command register), and is guaranteed to be completed within 1 µs of this transition. Access to
statistics registers prior to this interval can return indeterminate values. Given typical system boot
times and the software driver’s Ethernet controller initialization routines, no initialization of these
registers through software should be necessary.

388

Software Developer’s Manual

Diagnostics and Testability

Diagnostics and Testability
15.1

Diagnostics
This section explains the registers provided for diagnostic access.
These registers enable system level integration and debugging, including the ability to access all
internal memories. This information is often critical in determining failure modes and in
developing software workarounds.
At a diagnostic level, all of the major internal data structures visible to and controllable by
software, including all of the FIFO space. However, interlocks are not provided for any operations,
so diagnostic accesses need to be performed under very controlled circumstances.

15.1.1

FIFO State
The internal data FIFO pointers are visible through the head and tail diagnostic data FIFO registers
(see Section 13.8). Diagnostics software uses these FIFO pointers to confirm correct operation and
to directly write packets into, or directly read out of, the FIFO.
These registers are available for diagnostic purposes only and should not be written during normal
operation.

15.1.2

FIFO Data
All of the FIFO data is visible through the PBM register. Locations can be accessed as 32-bit or 64bit words. Refer to Section 13.8.11 for details.

15.1.3

Loopback
One loopback mode is provided in the Ethernet controller to assist with system and device debug.
This loopback mode is enabled via RCTL.LBM control bits. The Ethernet controller must be
operating in full-duplex mode for loopback.

Software Developer’s Manual

389

Diagnostics and Testability

15.1.3.1

Internal Loopback
This loopback mode internally loops back the transmit to receive path in the PHY, exercising the
internal GMII/MII bus. Programming both MAC and PHY is required. Following is the flow:
/* Auto-MDI/MDIX Off */
e1000_write_phy_reg(16, 0x0808);
/* reset to update Auto-MDI/MDIX */
e1000_write_phy_reg(0, 0x9140);
/* autoneg off */
e1000_write_phy_reg(0, 0x8140);
/* force 1000, set loopback */
e1000_write_phy_reg(0, 0x4140);
/* Now set up the MAC to the same speed/duplex as the PHY. */
ctrl_reg = E1000_READ_REG(CTRL);
ctrl_reg &= ~E1000_CTRL_SPD_SEL;

/* Clear the speed sel bits */

ctrl_reg |= (E1000_CTRL_FRCSPD |

/* Set the Force Speed Bit */

E1000_CTRL_FRCDPLX |

/* Set the Force Duplex Bit */

E1000_CTRL_SPD_1000 |

/* Force Speed to 1000 */

E1000_CTRL_FD);

/* Force Duplex to FULL */

/* Set the ILOS bit on the fiber Nic is half duplex link is detected. */
stat_reg = E1000_READ_REG(STATUS);
if((stat_reg & E1000_STATUS_FD) = 0)
ctrl_reg |= (E1000_CTRL_ILOS | E1000_CTRL_SLU);
E1000_WRITE_REG(CTRL, ctrl_reg);

15.2

Testability
The Ethernet controller uses full Boundary Scan/IEEE 1149.1 JTAG standard test methods. The
TAP controller supports EXTEST, SAMPLE/PRELOAD, IDCODE, USERCODE, and BYPASS
instructions.

390

Software Developer’s Manual

Diagnostics and Testability

15.2.1

EXTEST Instruction
This instruction allows testing of off-chip circuitry and board level interconnections. Data is
typically loaded onto the latched parallel outputs of the boundary-scan shift register stages using
the SAMPLE/PRELOAD instruction prior to selection of the EXTEST instruction.

15.2.2

SAMPLE/PRELOAD Instruction
This mandatory instruction allows a snapshot of the normal operation of the component to be taken
and examined. It also allows data values to be loaded onto the latched parallel outputs of the
boundary-scan shift register prior to selection of the other boundary-scan test instructions.

15.2.3

IDCODE Instruction
The IDCODE instruction provides information on the base component. When an Ethernet
controller identification register is included in a component design, the IDCODE instruction is
forced into the instruction register’s parallel output latches.
For example, the 82546EB controller’s ID is determined and derived from the manufacturer as
follows:
Component
Product Code
82546EB

15.2.4

Ver

0001

Product
1

001001

Gen
0010

Model
00100

Manf ID
00000001001

1
1

ID Code
(hex)
19244013

BYPASS Instruction
This instruction is the only instruction defined by the standard that causes operation of the bypass
register. The bypass register contains a single-shift register stage and is used to provide a minimum
length serial path between the TDI and TDO pins of a component when no test operation of that
component is required. This allows more rapid movement of test data to and from other
components on a board that are required to perform test operations.

Software Developer’s Manual

391

Diagnostics and Testability

Note:

392

This page intentionally left blank.

Software Developer’s Manual

Appendix (Changes From 82544EI/82544GC)

Appendix (Changes From 82544EI/82544GC)
A.1

Introduction
This section describes the new features that have been added to the PCI/PCI-X Family of Gigabit
Ethernet Controllers from its predecessor, the 82544EI/82544GC and highlights its registers that
have been changed.

A.2

New Features
Following is a list of the new features in the Ethernet controller, along with sections in this manual
that describe these features in detail:

• Integrated dual-port solution. The 82546GB/EB architecture includes two instances of both
the MAC and PHY. The Ethernet controller contains a single PCI/PCI-X interface so that each
of the logical LAN devices appear as a distinct PCI/PCI-X bus device (see Chapter 12).

• IPv6 Support. The Ethernet controller supports IP/TCP/UDP receive checksum offload for
IPv6 packets, IPv6 wakeup filters, and IPv6 TCP segmentation (see Chapter 3).

• Improved Interrupts. The Ethernet controller has the following new interrupt generation
features to enhance driver performance:
— Packet timers and absolute delay timers for transmit and receive (see Section 3.2.7).
— Short packet detection interrupts (see Section 3.2.7).
— Transmit descriptor low interrupts (see Section 3.4.3).
— Interrupt throttling control to limit maximum interrupt rate (see Section 13.4.8).
— Acknowledge interrupts by writing 1b’s to the ICR (see Section 13.4.17).

• EEPROM access. Because the 82546GB/EB has two MAC/PHY instances that could
potentially access the EEPROM at the same time, a semaphore has been added to gate access.
In addition, the EERD now allows easy EEPROM access (see Section 12.3.2).

• Manageability. The Ethernet controller introduces manageability for ASF 1.0 and AOL 2.0.
• Configurable LED. The Ethernet controller enables software to customize LED displays (see
Chapter 10).

• Power Management. The internal copper PHY features 10/100/1000-BaseT signalling and is
capable of performing intelligent power-management based on both the system power-state
and LAN energy-detection (detection of unplugged cables). See Chapter 6.

Software Developer’s Manual

393

Appendix (Changes From 82544EI/82544GC)

A.3

Table A-1. Register Changes
Register

394

Offset

CTRL

00000h

STATUS

00008h

EEC

00010h

EERD

00014h

CTRL_EXT

00018h

LEDCTL

00E00h

ICR

000C0h

ITR

000C4h

ICS

000C8h

IMS

000D0h

IMC

000D8h

RDTR

02820h

RXDCTL

02828h

RADV

0282Ch

RSRPD

02C00h

RXCSUM

05000h

TXDCTL

03828h

TADV

0382Ch

TPT

040D4h

PTC64

040D8h

PTC127

040DCh

PTC255

040E0h

PTC511

040E4h

PTC1023

040E8h

PTC1522

040ECh

MPTC

040F0h

BPTC

040F4h

Software Developer’s Manual

Appendix (82540EP/EM and 82545GM/EM Differences)

Appendix (82540EP/EM and 82545GM/EM
Differences)

B.1

Introduction
This section describes the differences between the 82546GB/EB, the 82540EP/EM and the
82545GM/EM. All three of these Ethernet controllers come from the same family so their register
sets are essentially the same. The sections that follow describe the differences between the
82546GB/EB and the 82540EP/EM or 82545GM/EM, and resulting register differences and
developer impact.

B.2

82540EP/EM Differences
Below are the differences between the 82540EP/EM and the 82546GM/EM:

• Serial FLASH interface. The 82540EP/EM does not support a parallel FLASH interface.
• No TBI/internal SerDes interface. The 82540EP/EM provides internal GMII / MII
interfaces only

• Single-port functionality. The dual-port functionality of the 82546GB/EB is not included.
• 32-bit PCI support only. The 82540EP/EM does not support 64-bit PCI or PCI-X.
• Internal SerDes. The 82540EP/EM does not support internal SerDes.
The impact to registers and the developer are outlined in the following sections.

B.2.1

Serial FLASH Interface
The 82540EP/EM provides an external interface to a serial Flash or Boot EPROM device.
Hardware implements a serial command set compatible with the Atmel AT25-series devices in the
512 Kb (64 KB) thru 1024 Kb (128 KB) sizes. The size of the FLASH used with the 82540EP/EM
should be encoded into bits in EEPROM to configure the amount of address space required when
mapped. All accesses to this device are controlled by hardware and are accessible to software as
normal PCI reads or writes to the FLASH memory mapping range. The FLASH and Expansion
ROM BARs are reconfigured based on these EEPROM settings.

Note:

The 82540EP/EM serial FLASH controller supports reads from programmed FLASH devices, and
writes to erased FLASH devices. Chip and sector-erase commands are not supported. It is
recommended that FLASH devices be socketed to enable removal for re-programming if
necessary.

Software Developer’s Manual

395

Appendix (82540EP/EM and 82545GM/EM Differences)

Note:

Though the 82540EP/EM supports devices with up to 512 KB of memory, smaller devices may
also be used. Accesses to memory beyond the FLASH device size results in access wrapping as
only the lower address bits are utilized by the FLASH.
The 82540EP/EM does not provide an interface for performing an “Erase” operation to the serial
Flash device. Flash write operations must be performed to an initialized or pre-erased Flash device.
If in-circuit erase is required, an external source (such as hardware's software-definable pins) can
be used to drive the Flash pins to perform the erase operation.

B.2.2

No TBI/Internal SerDes Interface
The 82540EP/EM does not support a TBI/internal Serdes interface. As a result, all TBI/internal
SerDes-related registers (for example, RXCW) should not be used.

B.2.3

Single-Port Functionality
The 82540EP/EM and the 82545GM/EM do not have the dual-port functionality of the
82546GB/EB. As a result, the 82540EP/EM and the 82545GM/EM appear as single PCI bus
devices.

B.2.4

32-Bit PCI Support
No developer impact.

396

Software Developer’s Manual

Source Exif Data:

File Type                       : PDF
File Type Extension             : pdf
MIME Type                       : application/pdf
PDF Version                     : 1.6
Linearized                      : No
Encryption                      : Standard V4.4 (128-bit)
User Access                     : Print, Copy, Extract, Print high-res
Author                          : LAD Information Engineering
Create Date                     : 2009:03:11 12:43:07Z
Modify Date                     : 2016:08:12 16:36:55+05:30
XMP Toolkit                     : Adobe XMP Core 5.6-c015 84.158975, 2016/02/13-02:40:29
Producer                        : Acrobat Distiller 9.0.0 (Windows)
Creator Tool                    : FrameMaker 8.0
Metadata Date                   : 2016:08:12 16:36:55+05:30
Format                          : application/pdf
Title                           : 8254x Family of Gigabit Ethernet Controllers Software Developer’s Manual
Creator                         : LAD Information Engineering
Document ID                     : uuid:987ca466-217d-4df2-a9de-f6b2963cdec4
Instance ID                     : uuid:8722a2c1-06da-4da8-a912-483178f71d22
Page Mode                       : UseOutlines
Page Count                      : 410

EXIF Metadata provided by EXIF.tools

8254x Family Of Gigabit Ethernet Controllers Software Developer’s Manual Pci X Gbe Dev

Navigation menu

Versions of this User Manual:

Views

Navigation