openpower-isa.git
2 years agoset sv_mode to 0b01 in element-strided
Luke Kenneth Casson Leighton [Sat, 24 Sep 2022 16:08:59 +0000 (17:08 +0100)]
set sv_mode to 0b01 in element-strided

2 years agofrickin frick
Luke Kenneth Casson Leighton [Sat, 24 Sep 2022 16:03:58 +0000 (17:03 +0100)]
frickin frick

2 years agoadd assert to stop failfirst+sea
Luke Kenneth Casson Leighton [Sat, 24 Sep 2022 16:00:28 +0000 (17:00 +0100)]
add assert to stop failfirst+sea

2 years agoadd extra RC1 test, without VLI.
Luke Kenneth Casson Leighton [Sat, 24 Sep 2022 16:31:00 +0000 (17:31 +0100)]
add extra RC1 test, without VLI.

2 years agoadd RC1 support to ISACaller.
Luke Kenneth Casson Leighton [Sat, 24 Sep 2022 16:17:55 +0000 (17:17 +0100)]
add RC1 support to ISACaller.
this involves:
* reading Rc=0 and substituting RC1 in its place OR
* for non-Rc instructions just putting RC1 in place of Rc
* reading VLi flag and adding it to srcstep to put into VL on ffirst hit
* setting the cr-bit to test to EQ in RC1 mode

2 years agopower_insn: slightly change table checking style
Dmitry Selyutin [Sat, 24 Sep 2022 13:22:14 +0000 (16:22 +0300)]
power_insn: slightly change table checking style

2 years agoadd extra test_pysvp64dis.py test for ff=~RC1/vli mode
Luke Kenneth Casson Leighton [Sat, 24 Sep 2022 13:25:09 +0000 (14:25 +0100)]
add extra test_pysvp64dis.py test for ff=~RC1/vli mode

2 years agowhoops got mask/match test wrong in power_insn.py
Luke Kenneth Casson Leighton [Sat, 24 Sep 2022 13:22:09 +0000 (14:22 +0100)]
whoops got mask/match test wrong in power_insn.py
should be value & mask == search & mask

2 years agocomment inv,CRbit swap in decode_bo
Luke Kenneth Casson Leighton [Sat, 24 Sep 2022 10:25:04 +0000 (11:25 +0100)]
comment inv,CRbit swap in decode_bo

2 years agosv_binutils: support RS opindex
Dmitry Selyutin [Sat, 24 Sep 2022 09:09:32 +0000 (12:09 +0300)]
sv_binutils: support RS opindex

2 years agopower_insn: reorder mode tables to match the spec
Dmitry Selyutin [Fri, 23 Sep 2022 07:37:47 +0000 (10:37 +0300)]
power_insn: reorder mode tables to match the spec

2 years agopower_insn: rename smr to mr
Dmitry Selyutin [Fri, 23 Sep 2022 07:36:37 +0000 (10:36 +0300)]
power_insn: rename smr to mr

2 years agosv_binutils: provide Boolean class and Rc field
Dmitry Selyutin [Thu, 22 Sep 2022 21:49:46 +0000 (00:49 +0300)]
sv_binutils: provide Boolean class and Rc field

2 years agopower_insn: provide Record.Rc field
Dmitry Selyutin [Thu, 22 Sep 2022 21:49:29 +0000 (00:49 +0300)]
power_insn: provide Record.Rc field

2 years agopower_insn: simplify rsvd naming; drop unused rsvd
Dmitry Selyutin [Thu, 22 Sep 2022 21:30:59 +0000 (00:30 +0300)]
power_insn: simplify rsvd naming; drop unused rsvd

2 years agopower_insn: replace Record.function with Record.mode
Dmitry Selyutin [Thu, 22 Sep 2022 21:30:32 +0000 (00:30 +0300)]
power_insn: replace Record.function with Record.mode

2 years agopysvp64asm: expand vector register macros
Dmitry Selyutin [Wed, 21 Sep 2022 19:35:08 +0000 (22:35 +0300)]
pysvp64asm: expand vector register macros

2 years agosv_binutils: support opcodes offset representation
Dmitry Selyutin [Wed, 21 Sep 2022 15:07:01 +0000 (18:07 +0300)]
sv_binutils: support opcodes offset representation

2 years agosv_binutils: fix fields traversal
Dmitry Selyutin [Wed, 21 Sep 2022 15:06:32 +0000 (18:06 +0300)]
sv_binutils: fix fields traversal

2 years agopower_insn: sort database finally
Dmitry Selyutin [Wed, 21 Sep 2022 15:04:53 +0000 (18:04 +0300)]
power_insn: sort database finally

2 years agopower_insn: provide missing cr_in2 properties
Dmitry Selyutin [Wed, 21 Sep 2022 10:21:46 +0000 (13:21 +0300)]
power_insn: provide missing cr_in2 properties

2 years agosv_binutils: generate svp64_cr_in2 opindices
Dmitry Selyutin [Wed, 21 Sep 2022 08:39:00 +0000 (11:39 +0300)]
sv_binutils: generate svp64_cr_in2 opindices

2 years agosv_binutils: generate BA opindex
Dmitry Selyutin [Wed, 21 Sep 2022 08:37:46 +0000 (11:37 +0300)]
sv_binutils: generate BA opindex

2 years agopower_fields: restore class-oriented traversal
Dmitry Selyutin [Tue, 20 Sep 2022 20:50:43 +0000 (23:50 +0300)]
power_fields: restore class-oriented traversal

2 years agopcdec. works!
Jacob Lifshay [Sat, 24 Sep 2022 00:12:47 +0000 (17:12 -0700)]
pcdec. works!

2 years agocheck svstate (vl) in failfirst test
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 23:00:20 +0000 (00:00 +0100)]
check svstate (vl) in failfirst test

2 years agogrr annoying recurrence of svshape bug, mscale starts with 6 bits
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 22:15:23 +0000 (23:15 +0100)]
grr annoying recurrence of svshape bug, mscale starts with 6 bits

2 years agoadd data-dependent fail-first mode, Rc=1 variant not RC1 yet
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 21:44:14 +0000 (22:44 +0100)]
add data-dependent fail-first mode, Rc=1 variant not RC1 yet
first unit test passes, not Vertical-First Mode

2 years agowhoops consistent inversion of inv,CRbit was CRbit,inv
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 21:42:49 +0000 (22:42 +0100)]
whoops consistent inversion of inv,CRbit was CRbit,inv

2 years agoextra failfirst dis tests
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 20:17:52 +0000 (21:17 +0100)]
extra failfirst dis tests

2 years agoremove barse-ackwardsness, use SelectableInt() in decode_bo
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 19:20:39 +0000 (20:20 +0100)]
remove barse-ackwardsness, use SelectableInt() in decode_bo

2 years agoput back the barse-ackward decode_bo inversion of {inv||CR_bit}
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 19:12:05 +0000 (20:12 +0100)]
put back the barse-ackward decode_bo inversion of {inv||CR_bit}

2 years agoremove need for explicit-hack for "pcdec." - rc column in minor_4.csv file
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 17:40:33 +0000 (18:40 +0100)]
remove need for explicit-hack for "pcdec." - rc column in minor_4.csv file
can be set "rc=ONE" which tells ISACaller (and PowerDecoder2) to
*always* write to CR0

2 years agolots of really bad hacks, here
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 17:38:34 +0000 (18:38 +0100)]
lots of really bad hacks, here
https://bugs.libre-soc.org/show_bug.cgi?id=933
1) rename to "pcdec." because it always sets CR0. following the convention
   set by "stbcx." etc.
2) hacked ISACaller into submission because this is the first instruction
   supported with "." at the end which is not Rc=1
3) handle_comparison was bypassed when CR0 is detected as explicitly
   an output: there is no point computing Rc=1 EQ/LT/GT/SO when CR0
   is supplied by the pseudocode
4) the test case case_pcdec_simple() was not making explicit deepcopy()
   of the registers, which causes problems
5) various places in actually getting the instruction from the insn
   dictionary, have to special-case "pcdec."
6) sv/trans/svp64.py updated to name "pcdec."
.

2 years agofix/hack some bugs in prefix_codes_cases
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 17:05:48 +0000 (18:05 +0100)]
fix/hack some bugs in prefix_codes_cases

2 years agoadd (sigh) to the hack-job get_pdecode_idx_out2() in ISACaller
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 16:49:45 +0000 (17:49 +0100)]
add (sigh) to the hack-job get_pdecode_idx_out2() in ISACaller
really should be relying on PowerDecoder2 but hey

2 years agochange variablename dec2.use_svp64_fft to implicit_rs
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 16:43:31 +0000 (17:43 +0100)]
change variablename dec2.use_svp64_fft to implicit_rs

2 years agoadd match on implicit_rc for pcdec
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 16:43:13 +0000 (17:43 +0100)]
add match on implicit_rc for pcdec

2 years agorename all "fft" variables in PowerDecoder2 because they are
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 16:35:16 +0000 (17:35 +0100)]
rename all "fft" variables in PowerDecoder2 because they are
to be used for pcdec as well.
https://bugs.libre-soc.org/show_bug.cgi?id=933

2 years agoadd sv.maddld/mr unit test example with expected results
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 16:21:00 +0000 (17:21 +0100)]
add sv.maddld/mr unit test example with expected results

2 years agoadd expected results for sv.maddld in openpower/test/mul_cases.py
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 16:12:56 +0000 (17:12 +0100)]
add expected results for sv.maddld in openpower/test/mul_cases.py

2 years agoreduce field name lengths (not in use)
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 10:21:40 +0000 (11:21 +0100)]
reduce field name lengths (not in use)

2 years agowhoops offset-tracking on 3-in 2-out supposed to be by MAXVL not VL
Luke Kenneth Casson Leighton [Fri, 23 Sep 2022 09:13:39 +0000 (10:13 +0100)]
whoops offset-tracking on 3-in 2-out supposed to be by MAXVL not VL

2 years agoremoved unneeded file
Konstantinos Margaritis [Fri, 23 Sep 2022 07:49:56 +0000 (07:49 +0000)]
removed unneeded file

2 years agoupdate Makefile
Konstantinos Margaritis [Fri, 23 Sep 2022 07:33:44 +0000 (07:33 +0000)]
update Makefile

2 years agoadd new SVP64 function
Konstantinos Margaritis [Fri, 23 Sep 2022 07:33:37 +0000 (07:33 +0000)]
add new SVP64 function

2 years agoreduce iterations, taking too long in the simulator
Konstantinos Margaritis [Fri, 23 Sep 2022 07:33:01 +0000 (07:33 +0000)]
reduce iterations, taking too long in the simulator

2 years agorename reference functions with _c suffix, add header
Konstantinos Margaritis [Fri, 23 Sep 2022 07:32:42 +0000 (07:32 +0000)]
rename reference functions with _c suffix, add header

2 years agorenamed variance_svp64.c to variancefuncs_svp64.c
Konstantinos Margaritis [Fri, 23 Sep 2022 07:31:58 +0000 (07:31 +0000)]
renamed variance_svp64.c to variancefuncs_svp64.c

2 years agoadd pcdec -- doesn't yet work due to broken ISACaller RT/RS output handling
Jacob Lifshay [Fri, 23 Sep 2022 03:05:35 +0000 (20:05 -0700)]
add pcdec -- doesn't yet work due to broken ISACaller RT/RS output handling

2 years agofix maddld pseudo-code
Jacob Lifshay [Fri, 23 Sep 2022 03:22:45 +0000 (20:22 -0700)]
fix maddld pseudo-code

2 years agoadd missing minor_4 decoder
Jacob Lifshay [Fri, 23 Sep 2022 03:02:02 +0000 (20:02 -0700)]
add missing minor_4 decoder

2 years agofix 'write reg ' log call
Jacob Lifshay [Fri, 23 Sep 2022 03:00:52 +0000 (20:00 -0700)]
fix 'write reg ' log call

2 years agoadd RC input to isa/caller.py
Jacob Lifshay [Fri, 23 Sep 2022 02:59:40 +0000 (19:59 -0700)]
add RC input to isa/caller.py

2 years agoformat code
Jacob Lifshay [Fri, 23 Sep 2022 02:56:11 +0000 (19:56 -0700)]
format code

2 years agomaddhd[u]/maddld are official ops
Jacob Lifshay [Fri, 23 Sep 2022 01:02:59 +0000 (18:02 -0700)]
maddhd[u]/maddld are official ops

2 years agoformat code
Jacob Lifshay [Fri, 23 Sep 2022 00:37:33 +0000 (17:37 -0700)]
format code

2 years agoadd first (correctly-working) ctr-mode sv.bc test
Luke Kenneth Casson Leighton [Thu, 22 Sep 2022 23:47:26 +0000 (00:47 +0100)]
add first (correctly-working) ctr-mode sv.bc test

2 years agocomment need for waiting on binutils update
Luke Kenneth Casson Leighton [Thu, 22 Sep 2022 12:41:23 +0000 (13:41 +0100)]
comment need for waiting on binutils update

2 years agofix no of iterations in comment, harmless but wrong
Konstantinos Margaritis [Thu, 22 Sep 2022 11:05:34 +0000 (11:05 +0000)]
fix no of iterations in comment, harmless but wrong

2 years agodump memory
Konstantinos Margaritis [Thu, 22 Sep 2022 08:43:46 +0000 (08:43 +0000)]
dump memory

2 years agobetter handling of memory copies, fix vpx_get4x4sse_cs_svp64
Konstantinos Margaritis [Thu, 22 Sep 2022 08:43:26 +0000 (08:43 +0000)]
better handling of memory copies, fix vpx_get4x4sse_cs_svp64

2 years agoremove extra setvl instruction
Konstantinos Margaritis [Thu, 22 Sep 2022 08:42:05 +0000 (08:42 +0000)]
remove extra setvl instruction

2 years agoadd series of double-stride options to test_caller_svp64_dct.py
Luke Kenneth Casson Leighton [Wed, 21 Sep 2022 19:49:44 +0000 (20:49 +0100)]
add series of double-stride options to test_caller_svp64_dct.py

2 years agodo not set striding on costables, keep them contiguous.
Luke Kenneth Casson Leighton [Wed, 21 Sep 2022 19:17:41 +0000 (20:17 +0100)]
do not set striding on costables, keep them contiguous.
not totally sure this is a good idea, but hey

2 years agogetting better, get rid of the ctr, group src/ref loads
Konstantinos Margaritis [Wed, 21 Sep 2022 18:33:24 +0000 (18:33 +0000)]
getting better, get rid of the ctr, group src/ref loads

2 years agoscale-up svshape pseudo-code for striding in DCT/FFT
Luke Kenneth Casson Leighton [Wed, 21 Sep 2022 17:18:14 +0000 (18:18 +0100)]
scale-up svshape pseudo-code for striding in DCT/FFT

2 years agofix dct/fft test-functions with new "scaling" parameter
Luke Kenneth Casson Leighton [Wed, 21 Sep 2022 16:56:57 +0000 (17:56 +0100)]
fix dct/fft test-functions with new "scaling" parameter
https://bugs.libre-soc.org/show_bug.cgi?id=930

2 years agomissed setting zdim in svshape on DCT modes
Luke Kenneth Casson Leighton [Wed, 21 Sep 2022 14:53:53 +0000 (15:53 +0100)]
missed setting zdim in svshape on DCT modes

2 years agouse sv.subf
Konstantinos Margaritis [Wed, 21 Sep 2022 15:33:49 +0000 (15:33 +0000)]
use sv.subf

2 years agofix braces
Konstantinos Margaritis [Wed, 21 Sep 2022 15:30:45 +0000 (15:30 +0000)]
fix braces

2 years agowhoops stride already has +1 from SVSTATE class
Luke Kenneth Casson Leighton [Wed, 21 Sep 2022 14:46:46 +0000 (15:46 +0100)]
whoops stride already has +1 from SVSTATE class

2 years agoadd SVzd to REMAP (svshape) "stride"
Luke Kenneth Casson Leighton [Wed, 21 Sep 2022 14:40:16 +0000 (15:40 +0100)]
add SVzd to REMAP (svshape) "stride"
https://bugs.libre-soc.org/show_bug.cgi?id=930

2 years agoadd stride-multiplier for 2D DCT/FFT "in-place" offsets
Luke Kenneth Casson Leighton [Wed, 21 Sep 2022 14:36:29 +0000 (15:36 +0100)]
add stride-multiplier for 2D DCT/FFT "in-place" offsets
https://bugs.libre-soc.org/show_bug.cgi?id=930

2 years agoInitial SVP64 attempt to vpx_get4x4sse_cs_svp64_real()
Konstantinos Margaritis [Wed, 21 Sep 2022 14:28:49 +0000 (14:28 +0000)]
Initial SVP64 attempt to vpx_get4x4sse_cs_svp64_real()

2 years agouse mr instead of li/addi pair
Konstantinos Margaritis [Wed, 21 Sep 2022 14:28:14 +0000 (14:28 +0000)]
use mr instead of li/addi pair

2 years agofix comments
Konstantinos Margaritis [Wed, 21 Sep 2022 13:29:02 +0000 (13:29 +0000)]
fix comments

2 years agoadd vpx_get4x4sse_cs_svp64_real() and wrapper
Konstantinos Margaritis [Wed, 21 Sep 2022 13:07:03 +0000 (13:07 +0000)]
add vpx_get4x4sse_cs_svp64_real() and wrapper

2 years agoFirst form of fully working SVP64 version
Konstantinos Margaritis [Wed, 21 Sep 2022 13:06:13 +0000 (13:06 +0000)]
First form of fully working SVP64 version

2 years agoreduce number of iterations in test, as it takes too long
Konstantinos Margaritis [Wed, 21 Sep 2022 13:05:13 +0000 (13:05 +0000)]
reduce number of iterations in test, as it takes too long

2 years agonecessary changes for run_a_simulation to work with pypowersim_wrapper
Konstantinos Margaritis [Wed, 21 Sep 2022 08:50:13 +0000 (08:50 +0000)]
necessary changes for run_a_simulation to work with pypowersim_wrapper

2 years agoInitial attempt for SVP64 asm version of vpx_get_mb_ss_svp64_real()
Konstantinos Margaritis [Wed, 21 Sep 2022 08:49:11 +0000 (08:49 +0000)]
Initial attempt for SVP64 asm version of vpx_get_mb_ss_svp64_real()

2 years agoadd sv.madd* to sv_analysis
Luke Kenneth Casson Leighton [Wed, 21 Sep 2022 00:15:46 +0000 (01:15 +0100)]
add sv.madd* to sv_analysis

2 years agoadd sv.maddld test case
Jacob Lifshay [Wed, 21 Sep 2022 00:00:48 +0000 (17:00 -0700)]
add sv.maddld test case

2 years agominor codemorph, whitespace
Luke Kenneth Casson Leighton [Tue, 20 Sep 2022 23:46:18 +0000 (00:46 +0100)]
minor codemorph, whitespace

2 years agosv.bc reclassified as RM-2P-1S by eliminating SPRs.
Luke Kenneth Casson Leighton [Tue, 20 Sep 2022 20:27:33 +0000 (21:27 +0100)]
sv.bc reclassified as RM-2P-1S by eliminating SPRs.
strictly it should be RM-1P-1S but there is a bug. needs investigation.
sv_analysis temporarily classifies as twin predication for now

2 years agoPoC simplified and isolated unit test for libvpx (VP8 & VP9) that uses pypowersim_wrapper
Konstantinos Margaritis [Tue, 20 Sep 2022 20:16:25 +0000 (20:16 +0000)]
PoC simplified and isolated unit test for libvpx (VP8 & VP9) that uses pypowersim_wrapper

2 years agoInitial PoC for calling pypowersim from within C code
Konstantinos Margaritis [Tue, 20 Sep 2022 20:14:46 +0000 (20:14 +0000)]
Initial PoC for calling pypowersim from within C code

2 years agoremove messy string identification, use RM Mode from database
Luke Kenneth Casson Leighton [Tue, 20 Sep 2022 19:44:51 +0000 (20:44 +0100)]
remove messy string identification, use RM Mode from database
in sv/trans/svp64.py

2 years agoadd quick test and loooong test of pysvp64dis - branches split out
Luke Kenneth Casson Leighton [Tue, 20 Sep 2022 16:43:32 +0000 (17:43 +0100)]
add quick test and loooong test of pysvp64dis - branches split out

2 years agopysvp64asm: fix sz handling
Dmitry Selyutin [Tue, 20 Sep 2022 14:14:03 +0000 (17:14 +0300)]
pysvp64asm: fix sz handling

2 years agopower_insn: unify predicates
Dmitry Selyutin [Tue, 20 Sep 2022 13:39:56 +0000 (16:39 +0300)]
power_insn: unify predicates

2 years agotest_pysvp64dis: test vli specifier
Dmitry Selyutin [Tue, 20 Sep 2022 12:55:33 +0000 (15:55 +0300)]
test_pysvp64dis: test vli specifier

2 years agopysvp64asm: support vli specifier
Dmitry Selyutin [Tue, 20 Sep 2022 12:55:22 +0000 (15:55 +0300)]
pysvp64asm: support vli specifier

2 years agopower_insn: support vli specifier
Dmitry Selyutin [Tue, 20 Sep 2022 12:55:11 +0000 (15:55 +0300)]
power_insn: support vli specifier

2 years agopower_insn: simplify specifiers sorting
Dmitry Selyutin [Tue, 20 Sep 2022 11:36:18 +0000 (14:36 +0300)]
power_insn: simplify specifiers sorting

2 years agomissed one sorting order in test_pysvp64dis.py
Luke Kenneth Casson Leighton [Tue, 20 Sep 2022 11:13:43 +0000 (12:13 +0100)]
missed one sorting order in test_pysvp64dis.py

2 years agosort specifiers in pysvp64dis in lexicographical order
Luke Kenneth Casson Leighton [Tue, 20 Sep 2022 11:11:57 +0000 (12:11 +0100)]
sort specifiers in pysvp64dis in lexicographical order

2 years agoadd two extra tests, sv.bc/m=r3/sz
Luke Kenneth Casson Leighton [Tue, 20 Sep 2022 11:06:00 +0000 (12:06 +0100)]
add two extra tests, sv.bc/m=r3/sz

2 years agopower_insn: custom sz handling for branches
Dmitry Selyutin [Tue, 20 Sep 2022 10:23:44 +0000 (13:23 +0300)]
power_insn: custom sz handling for branches