Google Groups no longer supports new Usenet posts or subscriptions. Historical content remains viewable.
Dismiss
Simulator progress
112 views
Skip to first unread message
Harry Reed
Feb 25, 2013, 10:55:54 AM2/25/13
to
For all interested parties,
My simulator for the dps-8/m is progressing along rather well, albeit
somewhat slower than I would like. All multi-word EIS instructions have
been implemented and tested (to a point) - including decimal arithmetic
instructions. (The decimal instructions are, at this time, rather
fragile since the decNumber package expects it's inputs w/o error. Need
to put boilerplate code in to ensure that decNumber doesn't have fits
with bad numeric descriptors.)
After much debugging, I am able to run a slightly modified erpt (sans
IOM functionality) from BOS for output formatting/editing.
Still need to implement faults & interrupts and need to start figuring
out how I might be able to simulate the IOM.
For more information see http://sourceforge.net/projects/dps8m/
Onwards and Upwards,
Harry Reed
My simulator for the dps-8/m is progressing along rather well, albeit
somewhat slower than I would like. All multi-word EIS instructions have
been implemented and tested (to a point) - including decimal arithmetic
instructions. (The decimal instructions are, at this time, rather
fragile since the decNumber package expects it's inputs w/o error. Need
to put boilerplate code in to ensure that decNumber doesn't have fits
with bad numeric descriptors.)
After much debugging, I am able to run a slightly modified erpt (sans
IOM functionality) from BOS for output formatting/editing.
Still need to implement faults & interrupts and need to start figuring
out how I might be able to simulate the IOM.
For more information see http://sourceforge.net/projects/dps8m/
Onwards and Upwards,
Harry Reed
doo...@gmail.com
Jan 7, 2014, 5:11:30 PM1/7/14
to
Always, Onwards and Upwards,
Harry Reed
Hristo Asenov
Jan 29, 2014, 2:54:41 PM1/29/14
to
dps8 hctrl-hsa01% make
(cd ../simh; make)
make[1]: Entering directory `/home/hasenov/code-0/trunk/dps8m/src/simh'
make[1]: Nothing to be done for `all'.
make[1]: Leaving directory `/home/hasenov/code-0/trunk/dps8m/src/simh'
(cd ../decNumber; make)
make[1]: Entering directory `/home/hasenov/code-0/trunk/dps8m/src/decNumber'
make[1]: Nothing to be done for `all'.
make[1]: Leaving directory `/home/hasenov/code-0/trunk/dps8m/src/decNumber'
make dps8
make[1]: Entering directory `/home/hasenov/code-0/trunk/dps8m/src/dps8'
clang -g -O0 -I../decNumber -I../simh -std=c99 -U__STRICT_ANSI__ -D_GNU_SOURCE -DUSE_READER_THREAD -DHAVE_DLOPEN=so -Wall -DUSE_INT64 -DQUIET_UNUSED -c -o dps8_addrmods.o dps8_addrmods.c
dps8_addrmods.c:99:9: warning: implicit declaration of function 'READOP' is invalid in C99 [-Wimplicit-function-declaration]
if (READOP(i))
^
dps8_addrmods.c:117:9: warning: implicit declaration of function 'WRITEOP' is invalid in C99 [-Wimplicit-function-declaration]
if (WRITEOP(i))
^
dps8_addrmods.c:138:9: warning: implicit declaration of function 'TRANSOP' is invalid in C99 [-Wimplicit-function-declaration]
if (TRANSOP(i))
^
dps8_addrmods.c:146:56: error: no member named 'info' in 'struct DCDstruct'
if (!READOP(i) && !WRITEOP(i) && !TRANSOP(i) && i->info->flags & PREPARE_CA)
~ ^
dps8_addrmods.c:160:12: error: no member named 'info' in 'struct DCDstruct'
if (i->info->ndes > 0)
~ ^
dps8_addrmods.c:172:66: error: use of undeclared identifier 'Ypair'
sprintf(temp, "CYpair[0]=%012llo CYpair[1]=%012llo", Ypair[0], Ypair[1]);
^
dps8_addrmods.c:183:1: error: unknown type name 'modificationContinuation'
modificationContinuation _modCont, *modCont = &_modCont;
^
dps8_addrmods.c:186:18: error: unknown type name 'modificationContinuation'
char *modContSTR(modificationContinuation *i)
^
dps8_addrmods.c:457:104: error: use of undeclared identifier 'Ypair'
sim_debug(DBG_APPENDING, &cpu_dev, "ITP Pair: PRNUM=%o BITNO=%o WORDNO=%o MOD=%o\n", GET_ITP_PRNUM(Ypair), GET_ITP_WORDNO(Ypair), GET_ITP_BITNO(Ypair), GET_ITP_MOD(Ypair));
^
./dps8.h:213:43: note: expanded from:
#define GET_ITP_PRNUM(Ypair) ((word3)((Ypair[0] >> 33) & 3))
^
../simh/scp.h:154:102: note: expanded from:
#define sim_debug(dbits, dptr, ...) if (sim_deb && ((dptr)->dctrl & dbits)) _sim_debug (dbits, dptr, __VA_ARGS__); else (void)0
^
dps8_addrmods.c:520:20: error: use of undeclared identifier 'illproc_fault'
doFault(i, illproc_fault, ill_mod, "Incorrect address modifier");
^
dps8_addrmods.c:547:39: error: use of undeclared identifier 'OPERAND_READ'
Read(i, address + 1, &itxPair[1], OPERAND_READ, i->a);
^
dps8_addrmods.c:561:8: error: conflicting types for 'doComputedAddressFormation'
t_stat doComputedAddressFormation(DCDstruct *i)
^
./dps8.h:862:6: note: previous declaration is here
void doComputedAddressFormation(DCDstruct *, eCAFoper action);
^
dps8_addrmods.c:564:9: warning: implicit declaration of function 'RMWOP' is invalid in C99 [-Wimplicit-function-declaration]
if (RMWOP(i))
^
dps8_addrmods.c:565:20: error: use of undeclared identifier 'rmwCY'
operType = rmwCY; // r/m/w cycle
^
dps8_addrmods.c:581:12: error: no member named 'info' in 'struct DCDstruct'
if (i->info->flags & NO_TAG) // for instructions line STCA/STCQ
~ ^
dps8_addrmods.c:606:9: warning: implicit declaration of function 'sim_printf' is invalid in C99 [-Wimplicit-function-declaration]
sim_printf("doComputedAddressFormation(startCA): unknown Tm??? %o\n", GET_TM(rTAG));
^
dps8_addrmods.c:635:47: error: use of undeclared identifier 'rmwCY'
if (operType == readCY || operType == rmwCY)
^
dps8_addrmods.c:637:13: warning: implicit declaration of function 'ReadOP' is invalid in C99 [-Wimplicit-function-declaration]
ReadOP(i, TPR.CA, OPERAND_READ, i->a); // read appropriate operand(s)
^
dps8_addrmods.c:637:31: error: use of undeclared identifier 'OPERAND_READ'
ReadOP(i, TPR.CA, OPERAND_READ, i->a); // read appropriate operand(s)
^
dps8_addrmods.c:642:48: error: use of undeclared identifier 'rmwCY'
if (operType == writeCY || operType == rmwCY)
^
dps8_addrmods.c:661:24: error: use of undeclared identifier 'illproc_fault'
doFault(i, illproc_fault, ill_mod, "RI_MOD: Td == TD_DU || Td == TD_DL");
^
dps8_addrmods.c:679:35: warning: implicit conversion from enumeration type 'enum _processor_cycle_type' to different enumeration type 'enum eMemoryAccessType' [-Wconversion]
Read(i, TPR.CA, &indword, INDIRECT_WORD_FETCH, i->a); //TM_RI);
~~~~ ^~~~~~~~~~~~~~~~~~~
dps8_addrmods.c:701:9: error: use of undeclared identifier 'cu'
cu.CT_HOLD = Td;
^
dps8_addrmods.c:702:69: error: use of undeclared identifier 'cu'
sim_debug(DBG_ADDRMOD, &cpu_dev, "IR_MOD: CT_HOLD=%o %o\n", cu.CT_HOLD, Td);
^
../simh/scp.h:154:102: note: expanded from:
#define sim_debug(dbits, dptr, ...) if (sim_deb && ((dptr)->dctrl & dbits)) _sim_debug (dbits, dptr, __VA_ARGS__); else (void)0
^
dps8_addrmods.c:712:35: warning: implicit conversion from enumeration type 'enum _processor_cycle_type' to different enumeration type 'enum eMemoryAccessType' [-Wconversion]
Read(i, TPR.CA, &indword, INDIRECT_WORD_FETCH, i->a);
~~~~ ^~~~~~~~~~~~~~~~~~~
dps8_addrmods.c:731:90: error: use of undeclared identifier 'cu'
sim_debug(DBG_ADDRMOD, &cpu_dev, "IR_MOD(TM_IT): Td=%02o => %02o\n", Td, cu.CT_HOLD);
^
../simh/scp.h:154:102: note: expanded from:
#define sim_debug(dbits, dptr, ...) if (sim_deb && ((dptr)->dctrl & dbits)) _sim_debug (dbits, dptr, __VA_ARGS__); else (void)0
^
dps8_addrmods.c:738:40: error: use of undeclared identifier 'f2_fault'
doFault(i, f2_fault, 0, "IT_F2");
^
fatal error: too many errors emitted, stopping now [-ferror-limit=]
8 warnings and 20 errors generated.
make[1]: *** [dps8_addrmods.o] Error 1
make[1]: Leaving directory `/home/hasenov/code-0/trunk/dps8m/src/dps8'
make: *** [locallibs_then_dps8] Error 2
Could someone help me to fix this?
Hristo Asenov
Jan 29, 2014, 11:36:24 PM1/29/14
to
Michael Mondy
Jan 30, 2014, 1:40:13 AM1/30/14
to
emul/wiki. One notable feature is that it displays the PL/1 source
lines that were used to create the boot tape. (This is possible because
the boot tape contains inter-procedure linkage information and the
available compiler listings provide a mapping between machine
instruction offsets and the corresponding PL/1 source lines.) My
emulator reads much of the boot tape and executes a couple million
instructions before crashing during disk initialization due to
unimplemented features.
See also recent discussions on the multicians mailing list about both
emulators and about system initialization.
-
Michael Mondy
Michael Mondy
Jan 30, 2014, 1:47:35 AM1/30/14
to
On Thu, 30 Jan 2014 06:40:13 +0000, Michael Mondy wrote:
> On Wed, 29 Jan 2014 20:36:24 -0800, Hristo Asenov wrote:
>
>> On Monday, February 25, 2013 10:55:54 AM UTC-5, Harry Reed wrote:
>>> For all interested parties,
>>>
>>>
>>>
>>> My simulator for the dps-8/m is progressing along rather well, albeit
>>> [ ... ]
> On Wed, 29 Jan 2014 20:36:24 -0800, Hristo Asenov wrote:
>
>> On Monday, February 25, 2013 10:55:54 AM UTC-5, Harry Reed wrote:
>>> For all interested parties,
>>>
>>>
>>>
>>> My simulator for the dps-8/m is progressing along rather well, albeit
>
> Please see also my emulator at
> http://github.com/MichaelMondy/multics-emul/wiki.
> One notable feature is that it displays the PL/1 source
> lines that were used to create the boot tape. (This is possible
> because the boot tape contains inter-procedure linkage information and
> the available compiler listings provide a mapping between machine
> instruction offsets and the corresponding PL/1 source lines.) My
> emulator reads much of the boot tape and executes a couple million
> instructions before crashing during disk initialization due to
> unimplemented features.
>
> See also recent discussions on the multicians mailing list about both
> emulators and about system initialization.
>
> -
> Michael Mondy
Thanks pam for the auto-wrap break at the hyphen... Corrected URL:
> Please see also my emulator at
> http://github.com/MichaelMondy/multics-emul/wiki.
> One notable feature is that it displays the PL/1 source
> lines that were used to create the boot tape. (This is possible
> because the boot tape contains inter-procedure linkage information and
> the available compiler listings provide a mapping between machine
> instruction offsets and the corresponding PL/1 source lines.) My
> emulator reads much of the boot tape and executes a couple million
> instructions before crashing during disk initialization due to
> unimplemented features.
>
> See also recent discussions on the multicians mailing list about both
> emulators and about system initialization.
>
> -
> Michael Mondy
http://github.com/MichaelMondy/multics-emul/wiki
0 new messages