700220_An_Instruction_Set_For_The_18 Bit_PDP K 700220 An Instruction Set For The 18 Bit PDP
User Manual: 700220_An_Instruction_Set_For_The_18-bit_PDP-K
Open the PDF directly: View PDF 
.
Page Count: 33
| Download | |
| Open PDF In Browser | View PDF |
An Instruction Set for the i8-bit PDP-K
Instruction Set
Op
C()d~
Data,
~.ry·p~~s
J).l. s tr i.bu ti or;.
K~ys:
R\~vigion ~
None
Obsolete:
Ncne
Date:
February 20, 1970
~.•
0
i '; . ~r: ( " ,J c: t 1 :' , n
An instruction set for an IB~bit computer~ is proposri].
It cJmbines the best features of the F~P-llts archltectnre .:ind the PDP-olD ~ s instruct~ion set
For ::~t~"e:ral .r'ed50ns f an lB-'bi t (:omput!;;r was considered
s u:>e ~<.('jr; i t solves both t.he Of:) code :tnd addr€-!ss spac. . .~
prob~ernB of: a 16·-bit computer.
In addit.~6.on"it: i;:: a
better data base in two important aredls. Pulse Height
A:lalysis (PHA) programs have proven the need for 18 bits.
1\.1so, thE: 36-bit floating-point rt;prt~sentation has much
wider acceptance, due to its superio~ity of 32-bit for~
mats· ..
1 r "e .. , a computer wi th a 'word len1Jth of 18 bi t~' "
2.0
. L~' ~:~ructi()n
_f~rmat
T.e~rminology
a!?..d
'!'hc ins tr'uction format of most ~inary
(two address)
instruct.ions is shown below. It resembles that of
the PDP-II and has three fields
In s fruG+ Ion
Wi
.J
D
Field
~.~ription
oc:
Operation Code·
Specifies the binary instruction.
s:
Source
. Specifies the Effective AdC:ress
(EA) ,of the
source.
D:
Destination
Specifies- the Effective Addr... (EA) of the
destination ..
The formats of the
shown below.
Sand D fields are identical and
~
.r\,
0 F'/eIJ
,
"..--._._"""
L3 I
..
M
.,
•
3
,,'
...
:::J
R
Field
Description
R:
Reqister
Denotes lout of 8 general reqisters.
Mode
Specifies the addressinq mode in a simila::way to those for tile PDP-li.l
lSee PDP-II Handbook.
conta~ns
dat~
0:
R
1:
@R
~ COl1tains
2:
@(R)+
Autoincrcment, defer
@J>~CR)
Index, defer
(R)+
Autoincrement
5:
- (R)
; Autodecrement
6:
@- (R)
7:
A(R)
: 'R
o.ddress of datu l
J
Aut0decrement, d fer
;Index·
r:['he address as computed from the Rand M f,ields is called
the Effective Address 'tEA".' When M-O, this is from
o to 7. The -tocatio·n of the memory cel1 2 actually
addressed is called the Effective Location "ELM. For
most binary instructions, EL-EA, i.e., the effective
location = the effective address.
In some instructions, the S or D field denotes
an integer number; for example, to specify the number
of shifts in a shift instruction" The format i4 as
follcws:
' - - - . _ _-."
EP ,:; e fJ
-,--l"@·' is used as the "indirect" symbol.
2/\ register is also considered "m~moryu.
l'
DescriJ(tion
Fi.'ld
P:
Register
Denotes lout of 8 general rcgistars.
:1odcs Short
Specifies the addressing rnodco They are
identical to the first 4 modes of the M
I
field:
o:
R
1: .
@R
2:
~
3:
@A{R)
Epl is R
;
EP is
(R)+
@(R)
autoincrement
f
EP is (R)+A
Free Bit
Bit not used
FR:
EP is (R)2
to determine EP.
The Effective Position "EP" can be interpreted as the
number representing the EA when M would be restricted
to the first 4 combinations. The table below shows
the" values EP can have:
Values of EP
Signed Integer
MS
Unsigned Integer
0:
R
-4 to 0; 0 to +3
0 to +7
1:
@R
-2 17 to 0; 0 to 217_1
0 to 2 18
2:
.@ (R)
... 2 17 to 0; 0 to 217_1
0
- ..
@A(R~
-2 17 to 0; 0 to 217_1
0 to 2 18
.
+
lttEpn ..:: Effective Position
to 2 18
3.0
('1,Hupatibili·ty_
Introducing a different word length will cause some
compatibility problems.
3.1
Peripheral COffipatibilit:i
A separate memorandum will be devoted to this problem.
The incompatibility can be reduced by r.aving the same
bus
ing
3.2
st~ucture for
consider~dc
P~ogram
the PDP-K as the PDP-Il.
This is be-
Compatibility
Two aspects have to be considered.
3.2*1 Word Length Compatibilit:t
This can be done ,by hardware by having a 16- and an IS-bit
mode; by software;! through a conversion pr)grarn simi lar
to that for converting PDP-8 to ~DP-9/l5 ?rograms leaving
certain portions to be recoded "by hand" 'e.g., shift
and rotate instruct-ions).
3.2.2 Instruction Set compatibilitl
This can be accomplished through microproJramming.
,
Because of the PDP-Res lS-bit word lenqth, microprogramming becomes very attractive ,because the lOP-IO can be
emulated.
-(j-
4.
a
Propos~~d
I?DP-K Instructio,!l Sc t
The proposed instruction ~et is shown in Appendix A.
Only tte major instructions ar~ shown. These are the
essential ones or thos~ requiring lots of op code 'space.
It is .;\ssur"".cd that the reader has some knowledge of the
PDP-II instruction set.
The instructions operate on 5 data types.
4.1
Bit, "En~
A bit is a Boolean quantity which is true itT" or
false 'F".
4•2
.::1.:
Byte,·
A byte is a character
4. 3
..
Word, ...:!!.:.
A wore is:
.....
I
2.
3.
4. 4
~Jle
A Boolean Array with 18 elements
signed integer (2's comple~ent)
A~ unsigned integer
A
Word, "0"
A double word is a single precision, floating-point
num.)er.
4.5
Qu<,druple Word, "0"
A '~uadruple word "is a double-precision, floating-point
number.
lDenotes abbreviat~.6n for the particular data type.
By tC$ are handled in a \vay simi lar t.o th,(~ PDP -1 0 ,
as described in Appendix B. Few instructions operd tf~
on byte because bytes ~re considered a riata ~ormat for
characters only.
Most instructions operate on words as the word is considered the data format for program control ani integer
numbers.
It is felt that higher level languages
(FORTRAN, ALGOL, etc.) use integers T.rtostly for subscript-
ing and program control an~, therefo:e, a singlt. IS-bi t
integer is considered sufficient.
The condition· code "ce" is handled in a way as described
in Appendix c.
-
-8-
5.0
!?Et.scr!I?tion of InstructiQns'
d~scribes the instruction formats and the
interpretation of the fields 'of the format.
Appendix D
The data ty.pe of the instruction will be indicated by
a letter following. the mnemonic of the instruction .
The letters are, as defined before: B~; bi t, Y :;: byte I
W = word or no letter (default) I 0 = double word and
Q - quadrupl.e word. Henc'e, MOV can be designated by
MOVY I MOW
or MaV, MOVD and
MOVO •
. The ,operation. of the individual instl'uct.ions is qiven
below.
Format ,"
------..-------~------------------Move----------------ISOl
MNEM
Operation
MOV
(8)+0
•
H·Ule
( (S),~)2+Cl
(tS) .0) '+«:2
({S»O).t.c3,
(D) ~'(S)
.CQM .
(r
.'
-Compare
.so
Com.pare
'SD
<0) 3-+<:1
(r=O) -+<:2
(r>O)+C3
~carry-O) "C
(Overflow-l)·Y
(D) - (S) , (8+1.)4_(1»'
,. .
wi.th. Limits
«(D)~(S»' «S+n»)(D»)~2
«D) «&» +Cl..
«O)-(S+n»..c3
(D)+(S)+D
'SD
(r O)+C3·
,:( Carr:y-l ) -+c
~ (OVerfl()W=ll) ~V
"~FO~ in.t~ction format see Appendix 0 *
'.
~,:~.=:".( (5) O)+C3
( Ir i? 211) :2 .. V
DIV
(D) , (D+~j) / (S) -~ D , D+ 1
(q --0) 3-+Cl
.
(q-O)...,C2
(q> 0) -+C3
(Jql~217) .. v
IMUlr
(D). * ($.) -• .D
I.
~>
For CC flee MDl",
ID'IV
"_"
q tf:/
(B.) I (~ ...~),
Por
CC·: ....:,;·DIV .
'".
J
BXCH
Divide
.,:-,.,."<,~. '., ''''.
".
.,
•
(S)+t"':';]J»~S, (temp.)·,.i)
.
. ' ,~'.': .~;.!.
~
~
Exchange
ISD/
¢ospare
Loqical
'SD
. [fa) '' (S) ). (S)~cc>
«(D)'(S».(S)~O)
All ,1 11 • • in (S) are 0 in 7 (D)
SON -l 's in ( S ) are 0 in -( D )
Some l's in (S) are 1. :in (D)
All lis in (S) are 1 in (D)
(D) , {S) ·...D
.'~>9i:eal
«D) & (5)-0) .
«((D)&(S».(S)+O)
( (D)" (S) .0)
AND
'+CC
AND
(r0) +C3
2Jrl
lq
lIlI\
at
absolute value
quotient. of
or: r.
divi.aion~
'SD
10-
ANDes (0)&(5) $.0
LOC,li cal
~.ND
l'''or CC s ((~e AND
~
I SO
wit.h
Complem.:-:nted
Source
Logical
.SO
Inclusive
·IOR
OR
For CC see AND
lORes
(D)! (S)
LO'g i c Z/.l# S D
v -.0
Inclusive
with
OR
XOP.
{D)8(S)+D
cal
'SD
Bxcl ua l. ·"e
OR
For CC see'AND
XORCS
(D), (5) '.+D
Ex<:,l'Jsive
th
FADD
(I)} + (5)
-~D
F
AOti
n9'~
(x- <0) ~'::l
(r-O) -+C2
(r>O) ~~3
(Overflow-i) ,..,..v ana 'trap
(Underflow-l) ..U
~~40.
F'SUB
(D)
<- (S)"D
Floating'
Subtract
ISD
?l·oatlng
tso
.For CC .ee FADD
F~'10L
( D ) ., (S» -t;·,D
Mult.iply
POI"
F'D!V
CC see
FJ\DD
)/{S)~D
CC
..
'
~"1.oat.in9
Dl~lide
see «PAnD'
t
-1:-
ftU~Elw1
- . - - - . . - .....
Operation
~~-
AOS
__
(D)
.-..-~~""_
..... _ ,_ _
-.-.t~~
+l~'D
___
Name
.''''''''''JW~;~~
__
Q~1f.9
:F'ol~mat
_______
Add One
~
__
''-~~,,
__
~._
#CSKI'
and Skip
1. f
(CC=T), then
(PC)
+R·~PC
When skip condition is satisfi.ed# the PC
wi th th(~ va.lue in the R field (0 t;.., 7
Slmtract
tCSl(P
One and
Skip
if (CC=T), then- (PC) +R.. PC
Test and
fCSKP
Skip
if (CC=T), then (PC)+R+PC
~
i.£
SOJ
Add One
and Jump
'C,JblP
subtract
fCJMP
One and
Jump
if (CC:.:T), then (D) -+PC
1f STJ
Test and
(R)+J\
tCJMP
(;Jutnp
if
(CC-T)~
. LSH
then (D)·.PC
Shift . (D) +0
Loqical
fEPD f
iLSH
Shift
The shif't direction· and the nu.raber of shifts depend on
the siqn and absolute value of the number determined by
the EP in the S field of the instruction.
(r <0) ·..·Cl
. (r-O) +C2
(r.>O) -+C3
(last bit shifted out)+C
(OvE.~rflow=l) l .. v
LSHC
Shif t Combined (O)
·,.D ,D+1
i!
(D-+ 1)
Logical
fEPD, 'LSHC
Shift
Combined.
For explanation and CC
see LSH'
lOverflow c,'ccurs (on le"ft shifts and rotat,es onl~{) wheneveI;':
the value of the two most significant bit.s of (D) become
unequal.. Once V is set, it stays set.~ On a rlqh.t shift ()l'
rotate, V is cleared ..
~
M}iEt"'l. Operatior:..
-~"'~.>eto_"''''''~~''''-~''''Ar~~_~~_'l'''''·',)Y''>l!I...,.,..,_-.._.~_~
Nan1€~
F;'o:emrtt
....w;..'~·~'_'':'l.>"""~'_"_'''''W'''''''''~'_·'''~''''·-''_.'.#_'''''_':l;'''_, __ .__
Rotate {D)~D
Rotate
fEPO, tROT
The rotate direction
r
bit
ons
'rotated depend on the sign and absolute
the.
number {l,;:termined by t:he E;P
t.h.t; S
str'Uction ..
(rO)-+C3
(last bit rotated
out)~C
(Overflow=l)l-1V
ROTC
Hota.te Camhi
-~ D ,D+ 1
(n~+"l)
t:e
Combint:~d
FCir explanation and CC;f
see Rt)T ..
Shift Arithmetically (,0)'''4D
ASHe
tnllle; t.! c
For explanatit)n and CC I'
see SH ..
Shi.ft
Shift A.t'itJunetici"lly C<)m-'
bined (D), (r,+ 1) -+t~ i D+l
IA.::: i t.tlmE: ti C'
Shift Com ...,
bine.d
J."or explana.t.ion a.n.d. CC i'
see LSH.
BIS
1+EBL2.
Bit Set
tEPD
The EBL is determined lUS follt)w8;' t.he EA
of the instructic)ft. is t.aken, starti.'!l9 from the
of the wbrd -denoted by' EA, EP bit. locations
Note:
EP
allowed ,to be b,:tqqer thAn 18"
~,
.
) ..:~,
) ~l>Cl
(BaL)')! (C) .... C2
(EBL) 3f; (C)..c3
" (EBL) J
~C
Bit Clear tEPD
SIeL
j{or explarLation andCC"
see SIS<\>
left shifts and rotates only) whenever
Overflow occ\..u:~
two most significant bi ts of (D) hecolr.e
t ..he value of
is set it stays set. On it .tight
oX:'
llnequal
ared ..
.t'otat:e t 'l
'"
bi t. loc ..... (8P) ..
'
stopped when (Re-O)! (CC=T)..
(CC=T)'
(RC~O),
: BPD
Repeat
, Repeat next two instructiOlu;.
UPS. '
JMP
'J~
XCT
tREP
Dcnlble
f~()r explanation, see
if (Cc-T) the (D) "PC
Jump
'JMP
takes place when jump condition ,is satisfied ..
if (CC-Tf,<,then Execute
Execut.e
~
fJMP
:When condition satisfied, the instruction denoted by (
~~!:ta·. executed.
xc:ru.
"'.'.'
-''\!''\'.,
if (CC~') the Execute
Undisturbed
'
Execute
Undis-
JJMP
,
turbe4,
.
·1Iben:condi tfon aati.fied, the;iinst~uctt.on denoted by
1~~
is executed undisturbed, l~.~i·"',~tre.ult of the
"Qpe:ration i8 not st.ore3 only ~., ,CC' is aet.
-
-15-
6" 0
Registe£._~eV~!!
General'register "R7" is used in the PDP-li as the PC
(program Gounter)* Because of this, certain addressing
modes are no~ advisable or lead to "selfdestruction" of the program. 'The table below shows,
this.
ADDRESSING MODES FOR' R7
WI,.,
r
............
....
............
------------~-.---,.
Source
Destination
----~~~-~-~.-.---------------------
R7
OK
R1
OK
@R7
OK
@R7
Error
,eR7)+
OK
,,:a7) +
OK
fA (It?)'
OK
'A(R7)
OK
(R7)+
OK.
eR7)+
-- (It"
Error
-(R1)
Error
@- (lti)
.Error
@-'(R7)
Error
A (1t7)
OJ(
At.') - Oft
- ..... ..
"
J
a
r"<.''-·
Of'.
J
lilt
;
~
~tt. suq. . . tedllOt"cmlY" to.p:.;."eJrt ~,fJle proqrattmter
!Jr 9l' , -&king these ,~~ror., ba~,.l.oto,turn these
'~~lty
com.biJl-atione into
.ome~in9
useful.
'.'';'. :...ff
;.Use
the destination mode .("')+
i40 not store the result of the
~l-ornary instructions .become
'structions ~
tn.'
nOhlal way exc~pt
operation. This way
-te.t iaaediate" in-
6 .. 2
Use the dest.inat.ion modes -(R7) and @-(R7) as flags
indicating the
follO\riin9~
Consider the instruction a st.ack operation with the
stack (i .. e,. f there where R6 o(;)in'ts to) as the d~;stination
and as source the contents of "(R5) +EN" • The E:.: fecti ve
, Number "EN" is the contents of the S field of tI\€ Instruc ....
Elan -rnterpreted as an unsigned integex (i. e,,; fr'om
to 63.) ... The binary instructions look like:
o
...
(SP) Operation
«R5)+EN).-+ (SP)
6.2.2 @-(R?) C~.~~
Operation 'similar to the - (R7) cas(~ excent ILS source
the contents of «(RS) +EN) is taken..
Bin~iry instructions
look like:
(SP) Operation@«R5}+EN)-+(SP)
lOW
c
double
20W .. quadruple word
':ot
}
-S+n
= next
data word
~'c~~-·_.ct
_._.~c;. ___...____• ___ •. ___. _ _ _ _ _
,SG\UD~~.a.atJ;>uction
AOJ
1
1
1
SOJ
(R)-l+R, jump?
(R)-..R,
Bit
-----~--.~- -~
.... ---.-
Byte
jump?
(BBL) .... -(SP) 4
(SP)++EBL
O+IBL
1 ...£81,1
B.lt Set
Bit Cleat;
BIeN
(EBL) '+EBL
(C)2-+EBL
BIT
(BBL) +ce l
Bit Complement
Bit Copy
tlI'l'C
(BBL) '-+CC
Bit Test
Bit Test Complement
SHOY
MMOV
. (0) +- (SP) 4
(SP)++D
step
....__
,irl~
.
\
~
~'L
"
"
,:
.
ASHe
BIMS
BIMM
BIS
BIeL
c~·-
.:
Logical Shitt
Logical Shift Combined
Rotate
Rotate Combined
Arithmetic Shift
Arithmetic Shift Combined
Bit Move to S'tack
Bit Move to Memory
\'Jord
.;
Add One and Jump
Subtract One and Jump
Te.t and Jump
(R) +l-;.R, jump?
ASH
1/2
3
3
J;tescription
TSTJ
-- -..-- .-. - .-----.- ---.--.-.> ...--...-.
c
.-- --.. --- ....- - . - - - - - - -...- - -
LSD
LSHC
ROT
RCT£
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
1/2
.
.;
I
I
...
I
I
I
'I
,"
"
I
.;
I
.;
.;
Stack.Move, Multiple Indexed
Memory Move, . Multiple Indexed
.;
.;
./
,I
,
~
Bit
Z
JSR,
1
ANAL '
1/8
l/8
1/4
1/4
1/4
REPD
JMP
XCT
XCTU
TST
.,""'!>:;1I!'C';:.;._...;;.~._
-"
.,.... ,., .t!!k.!.t ..•.... ,,,.
.~.
$",
Bit
Byte
._--_.
Word
.-
-:
~C
' - - - ' ---.--
..
...
._,
Q\.'
I:~~
~
-'
O+D
l ....D
-l+D
Set to plus One
set to Minus one
(D)+(C) ....D
Add Carry
Set to all Zeros
4/64
1/64
1/64
SETZ
SETPO
1/64
1/64
ADDe
suac
(D)-(C)+D
1/64
3/64
roc
(D)
'rrc
(D) '+l+D
1/64
1/64
1/64
1/64
1/64
'1/64
CIFS
CIFD
(0)+-(8P)
(D)+-(SP)
1/64
1/64
1/256
1/256
1/4096
3/64
1/64
C
;truction . Description
Count
-
-----_. '--_'
." . . . . . . . . _ _. _ ._____
SETMO
D
t ....
. Df:SI
(D)+~(SP)
.. CPSD
(D)+-(SP)
(D)+-(SP)
(D)"-(SP)
CFOI
C!o'DS
.
I
l
t'
.I
Subtract Carry
i
.;
Take One's Complement
"Take Two'. Complement
convert
Convert
Convert
Convert
Convert
Convert
I
I
I
.;
/
InteCjer to Float. Single
Intec}er to Ploat. Double
Float. Single to Integer
Float. Sin91e to Float. D.
Float. D. to Integer
Float. D • to Float. s.
v
.
.;
t
I
I
I
.
.Inore.nt Byte Pointer
!NCBP
Dec:r_nt Byte Pointer
DEtSP
MCCS
(CC)+-(SP)
MCCC
(ee) +C
MSCC
I
. (S,)+ ...C
NEeH
(D)~(D+n)
LOCK
( (D) -0 )::a.( 81') ++D
I
I
Move CC tb Stack
Move CC to C Bit
Move
.
I
~
,.!)
Stack to C Bit·
"'.
Next ••change
Lock
I
I
J~
I
APPENDIX B
PDP-K Byte
Handlin~
The PDP-K will handle bytes in the same manner as the PDP-lO.
The format of the byte instructions will be similar to all
other instructions.
I nst roc t.on
..--~
[
.I..-....,-J
OC.
'rhe possible
oct.
]
6
b
b
."---r-J
•
S
J
"
areiMOVY, COMY,' and COMLY.
The Sand D fieids are'identical in format and define the.;,
locations of the source and. destination byte pointers "sn
a.nd DYP". The Sand Dfields art interpreted the same way
as the EP field, described in.¥ction 2.0 and a8 shown belOfl.·
if,'
~.
~
s ", 0........
,.
.....,..,
~~
.
..a
1>1 , 2
FA
;
F,e;Id
•
IfS
.f
[
\
3
t.
~
________
~*~
",
I
II J
______
.
JP
The locations of "the SYP (source byte pointer) and the DD
(d. . t.laation byte pointer) .~e 4eter.mine4)~ tbecontents
fiel~ of the i •• tructions.
The
fzwe bi t.s "Fa- are . .ad· to alloW for iacrementinq the byte
of the EP' s of th. • and D
poinur.
'the formats of til.
id•• ti~l
sn ana ow'.....
.,.jYP., Oyp
~.....
r
'
..
fMl4 shown
bel~'.
-2~.-
Field
Description
yp
;t'he position of the first bit of .the byte in the
double word addressed by YL.
YS
The length of the byte in bits.
YL
YL is interpreted as a reqular destination ani denotes the location of the double word contain:ng the
byte.
-22-
APPENDIX C
Condition Codes
The PDP-K condition code differs from the PDP-ll because
of the specitl require~nts imposed by the single bit diddling
ins,tructions of POP-It. The. instructions making use of th~
condition code have 4 bits to specify the ,condition. The
function of 4 of the condition code flip-flops will be disct.ssed
below.
c.~ FI;
Condi·h'on
I
,.-
(1
,
£
indicate.
indicate.
ell
C2: .
indicates
1· C2 I
p- Flops
~
C3
J
C ]
?
•
J
in arithmetic: operations
.~ft,le bit operatioas
"til!"
-6-
in
-.-
~ arl~tlc
operations
"indicates "I" in sinlle bit operations
e31'
indicates .. ,. .... in .r1~tic operations
indicates ••• in 8ingle bit operation.
Ci .
carq bit also "sed
operation.
.~
.t.eat bit iA single bit
In hlthMt:ic operations the flip-flops el, e2, C3 and C are
used. a8 listed in the table below and interpretttd as follows.
Cl-1 'When the renlt i . 4(); .C2-1 when result -0# C3-1 when
r ••ult.,i1!':O, and C.l when there i . a carry or when there is no
borrow.
Xatbeca.e of bit diddling, .the flip-flops are used
follOWS:
&8
!:;.'
(BBL) 2, (C) J+Cl
(EBL) I (C) +c2
(EBL).(C)..-C3
(EBL)"-C .
.
,
'''Ii - - - - - - - - -
. ls.e Appendix 'A iDatructio68 alS, BIeL, BleD, BIT, BltfC and BICP.
't
2EaL -
cont.ents· of EffectiVe Bit Locatioa ,complemente.c! wheD
the 8TC (bit test ca.plement) instruction ia used.
3(C) ,= contents of the carry flip-flop.
-23-
The operation above allows all 16 boolean operators between
2 variables directly and allows domplex boolean equations
to be evaluated easily.
The interpretation of the contents of" the flip-flops el, c2
and C3 for signed arithmetic and bit didlinq is shown below
and required 8 "condition code combinations" out of the 16
total.
TABLE Cl
,
1
<
=
•
C2
C3
CI
>
Siqned
Arithmetic
Interpretation
Bit
Diddling
Interpretation
.
O·
0
0
0
PaIse
BHOT
1
0
0
1
>
BG'l'
2
0
1
0
•
BEO
.
0
1
1
~
BGE'
4
1
0
0
<
BLT
5
1
0
1
6
7
.
1
I
1
1
'.
BXOR
BIOR
;
3
.
.,•
BNB
0
.s
BLE
1
True
BRA
,"
SHAND
.'
BHIOR
I •
True
BAND
BNXOR
BRA
The remaining 8 combinations are used as shown in the table
below. Toqether with the \BEO and BHE conditions fr~m above
they contain all conditions for unsigned arithmetic.
-24-
TABLE
. --.--- -
-
C2
---~-
-.~~
._-----_ _.......
. -.
...
-
Unsigned
Arithmetic
Interpretation
Speciz..l
Condition
Interpretation
r--'-
Repeat count
0
:::I:
2
OVerflow
BOV
3
No Carry
BSCA
Carry
BCA
No Overflow
BNOV
5
I
6
1
Repeat count
~
0
_J
BZR
0
1
4
1
•
I
BNZR
.
>
BHI
~
BHIE
<
BLO
~
BLOE
J
-25-
APPENDIX D
Instruction Formats
..
0.1
~ormat lSD, Source Destination
Instruction has 3 fields of 6 bits
instruction
1\
6
I
"
6
""
'v
v
OC
"
I
A.
S
6
I
I
v
I
or
S
3
.V
For.a.~
SUle
forsu.t as ahown below.
field
I\,
M
0.2
D
operatior~
0
The S and D fields have the
,
code
source
5 D - destination
OC a
"
I
3
V
I.
R
M
a'" reqi8ter j:ield
• mode fi~la
R
'CSKP, Conditional Skip
Instruction has 4 fields. '!'he SC field (skip ,
. condition) i8 interpreted as in Table Cl of Appendix c.
The R field contains the number of words to be skipped
(from 0 to 7).
]
UJ• r
•
6
v
OC
6
\.
sc
v
D
-
D.3
Porm~t
26~-
#CJMP,
__. ._____ Conditional Jumc
_.-'_~I'f
~",.'4
_ _ _ _ _ _ _ _ ' _ _ _ _ "_' ••
"-,,,.......L.-,
'1' i i. i
sin s t r 11 C' t. ion has 4 fie 1. d 5 •
'1" h e J C fie 1 d con t a ins
t'1f:: jump condition, interprete,d as shown in Table Cl
of Appendix c~
The R field denotes the register to be
tested after an increment (decrement or test).
oc
JC
OC
. ...1
I
~~
'L
= jump
EP
=
condition
D
R
[ 6"JLGJ
-_.---v".- -
JC
"----v-- I
E~
effective positior
field
D
FR • free bit used to extend the OC field
D is a regular destination field, EP is a regular effective position field.
D.5
Format 'LSH, L02ical Sh.i.t~
(D)
A
,---
Q~[hi9h
II
18
------""
:
lOW: ""1--
-Q
Format tLSHC A Logical
D06
Shi~~ ~ombine~
(D)
,..-
A.
.~igh
18
f-:l-- r
!
0
L_-1
D. 7
Format
l---'--
*¥
Source Exif Data:
File Type : PDF
File Type Extension : pdf
MIME Type : application/pdf
PDF Version : 1.3
Linearized : No
XMP Toolkit : Adobe XMP Core 4.2.1-c041 52.342996, 2008/05/07-21:37:19
Producer : Adobe Acrobat 9.0 Paper Capture Plug-in
Modify Date : 2018:03:23 18:05:47-07:00
Create Date : 2018:03:23 18:05:47-07:00
Metadata Date : 2018:03:23 18:05:47-07:00
Format : application/pdf
Document ID : uuid:b4743c21-cf34-8c40-89eb-371fcf563b25
Instance ID : uuid:ba61239d-bfd9-8543-8d1d-d8a41445db60
Page Layout : SinglePage
Page Mode : UseNone
Page Count : 33
EXIF Metadata provided by EXIF.tools