openpower-isa.git
2 years agopower_enum: tune SVExtra representation
Dmitry Selyutin [Mon, 5 Sep 2022 17:58:13 +0000 (20:58 +0300)]
power_enum: tune SVExtra representation

2 years agopower_insn: support extra_reg routine
Dmitry Selyutin [Mon, 5 Sep 2022 17:39:17 +0000 (20:39 +0300)]
power_insn: support extra_reg routine

2 years agopower_insn: move extras to SVP64Record
Dmitry Selyutin [Mon, 5 Sep 2022 11:19:52 +0000 (14:19 +0300)]
power_insn: move extras to SVP64Record

2 years agoremoving two unused fields (E) which somehow
Luke Kenneth Casson Leighton [Tue, 6 Sep 2022 11:42:30 +0000 (12:42 +0100)]
removing two unused fields (E) which somehow
made it into the 1.6.2 section

2 years agoadd dummy fixedsync.mdwn pseudocode for lwarx/stbcx. LR/SC operations
Luke Kenneth Casson Leighton [Tue, 6 Sep 2022 11:01:29 +0000 (12:01 +0100)]
add dummy fixedsync.mdwn pseudocode for lwarx/stbcx. LR/SC operations

2 years agosort out demo of remap_preduce_yield.py
Luke Kenneth Casson Leighton [Tue, 6 Sep 2022 01:22:02 +0000 (02:22 +0100)]
sort out demo of remap_preduce_yield.py
originally based on wiki preduce.py, split into two separate SVSHAPEs
one for left operand the other for right

2 years agoadd first version of parallel reduction yield
Luke Kenneth Casson Leighton [Tue, 6 Sep 2022 00:39:00 +0000 (01:39 +0100)]
add first version of parallel reduction yield

2 years agoadd a new log option to pysvp64dis and make it a command-line
Luke Kenneth Casson Leighton [Mon, 5 Sep 2022 23:12:10 +0000 (00:12 +0100)]
add a new log option to pysvp64dis and make it a command-line
console script using setup.py entry_points

2 years agouse log function for warnings about .mdwn files in pagereader.py
Luke Kenneth Casson Leighton [Mon, 5 Sep 2022 23:09:06 +0000 (00:09 +0100)]
use log function for warnings about .mdwn files in pagereader.py

2 years agomove reading of os.environ out of a global which prohibits
Luke Kenneth Casson Leighton [Mon, 5 Sep 2022 23:07:46 +0000 (00:07 +0100)]
move reading of os.environ out of a global which prohibits
setting of os.environ at runtime

2 years agoremove parallel-reduction mode from decoder and sv/trans/svp64.py
Luke Kenneth Casson Leighton [Mon, 5 Sep 2022 16:23:12 +0000 (17:23 +0100)]
remove parallel-reduction mode from decoder and sv/trans/svp64.py
parallel reduction has to be done through REMAP due to two critical factors:
1) the amount of gates in joining REMAP with PREDUCE as a "Mode"
2) the differing Vector Length (similar to Matrix) from the number of
   operations needed to be performed
the complexity arising is too great which means it has to be done as REMAP

2 years agorename remap_debug to remap_set_steps
Luke Kenneth Casson Leighton [Mon, 5 Sep 2022 11:40:04 +0000 (12:40 +0100)]
rename remap_debug to remap_set_steps

2 years agoplease do not use format-specifiers
Luke Kenneth Casson Leighton [Mon, 5 Sep 2022 11:25:26 +0000 (12:25 +0100)]
please do not use format-specifiers

2 years agoremove yet more f"" specifiers
Luke Kenneth Casson Leighton [Mon, 5 Sep 2022 11:20:26 +0000 (12:20 +0100)]
remove yet more f"" specifiers

2 years agoremoving use of format. please do not use format
Luke Kenneth Casson Leighton [Mon, 5 Sep 2022 11:14:28 +0000 (12:14 +0100)]
removing use of format. please do not use format

2 years agorename "PARALLEL" enums to "PTREDUCE" - parallel tree reduce
Luke Kenneth Casson Leighton [Mon, 5 Sep 2022 09:21:20 +0000 (10:21 +0100)]
rename "PARALLEL" enums to "PTREDUCE" - parallel tree reduce

2 years agoadd detection of Parallel-Reduction Mode into SVP64RMModeDecode
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 19:47:40 +0000 (20:47 +0100)]
add detection of Parallel-Reduction Mode into SVP64RMModeDecode

2 years agoadd parallel-reduction (subvl=0) in sv/trans/svp64.py
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 19:35:53 +0000 (20:35 +0100)]
add parallel-reduction (subvl=0) in sv/trans/svp64.py

2 years agopower_insn: support mode description
Dmitry Selyutin [Sun, 4 Sep 2022 18:21:49 +0000 (21:21 +0300)]
power_insn: support mode description

2 years agopower_insn: decouple mode function
Dmitry Selyutin [Sun, 4 Sep 2022 17:36:16 +0000 (20:36 +0300)]
power_insn: decouple mode function

2 years agopower_insn: pass database instance everywhere
Dmitry Selyutin [Sun, 4 Sep 2022 17:27:41 +0000 (20:27 +0300)]
power_insn: pass database instance everywhere

2 years agopower_insn: refactor operation order
Dmitry Selyutin [Sun, 4 Sep 2022 16:17:48 +0000 (19:17 +0300)]
power_insn: refactor operation order

2 years agowhoops forgot to write maxvl in PowerDecoder2
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 15:06:50 +0000 (16:06 +0100)]
whoops forgot to write maxvl in PowerDecoder2

2 years agopower_insn: drop argument in PPCDatabase
Dmitry Selyutin [Sun, 4 Sep 2022 13:28:29 +0000 (16:28 +0300)]
power_insn: drop argument in PPCDatabase

2 years agopower_table: replace prints with logging
Dmitry Selyutin [Sun, 4 Sep 2022 12:33:23 +0000 (15:33 +0300)]
power_table: replace prints with logging

2 years agopower_table: simplify the code
Dmitry Selyutin [Sun, 4 Sep 2022 12:31:35 +0000 (15:31 +0300)]
power_table: simplify the code

2 years agopower_insn: inherit PPCMultiRecord from frozenset
Dmitry Selyutin [Sun, 4 Sep 2022 12:12:01 +0000 (15:12 +0300)]
power_insn: inherit PPCMultiRecord from frozenset

2 years agopower_insn: try exact name matching first
Dmitry Selyutin [Sun, 4 Sep 2022 12:10:56 +0000 (15:10 +0300)]
power_insn: try exact name matching first

2 years agobrainmelt moment, added a comma into a comment in a csv file
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 11:35:34 +0000 (12:35 +0100)]
brainmelt moment, added a comma into a comment in a csv file

2 years agouse maxvl not vl in impicit-RS
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 11:05:33 +0000 (12:05 +0100)]
use maxvl not vl in impicit-RS

2 years agopower_insn: remove the whitespaces properly
Dmitry Selyutin [Sun, 4 Sep 2022 11:23:45 +0000 (14:23 +0300)]
power_insn: remove the whitespaces properly

2 years agocomments for 3-in 2-out ops
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 10:45:32 +0000 (11:45 +0100)]
comments for 3-in 2-out ops

2 years agowhitespace cleanup
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 10:32:08 +0000 (11:32 +0100)]
whitespace cleanup

2 years agocomments on ffmadds fft 3-in 2-out
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 10:29:05 +0000 (11:29 +0100)]
comments on ffmadds fft 3-in 2-out

2 years agoassert prints out XO
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 10:14:49 +0000 (11:14 +0100)]
assert prints out XO

2 years agomore comments
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 10:10:48 +0000 (11:10 +0100)]
more comments

2 years agopower_insn: refactor immediate operands
Dmitry Selyutin [Sun, 4 Sep 2022 10:06:24 +0000 (13:06 +0300)]
power_insn: refactor immediate operands

2 years agopower_insn: inherit Operands from tuple
Dmitry Selyutin [Sun, 4 Sep 2022 09:49:42 +0000 (12:49 +0300)]
power_insn: inherit Operands from tuple

2 years agocode-cleanup (comments) and rename "i" to "XO"
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 10:08:19 +0000 (11:08 +0100)]
code-cleanup (comments) and rename "i" to "XO"

2 years agopower_insn: support immediate operands
Dmitry Selyutin [Sun, 4 Sep 2022 09:26:48 +0000 (12:26 +0300)]
power_insn: support immediate operands

2 years agopower_insn: make operand classes public
Dmitry Selyutin [Sun, 4 Sep 2022 09:04:29 +0000 (12:04 +0300)]
power_insn: make operand classes public

2 years agoRevert "target_addr in b and bc pseudo-code has no corresponding"
Dmitry Selyutin [Sun, 4 Sep 2022 08:58:08 +0000 (11:58 +0300)]
Revert "target_addr in b and bc pseudo-code has no corresponding"

This reverts commit 7656b425de39cb28e6c52b9ee32c0f12517bae69.

2 years agoRevert "svbranch.mdwn: replace target_addr with BD"
Dmitry Selyutin [Sun, 4 Sep 2022 08:57:57 +0000 (11:57 +0300)]
Revert "svbranch.mdwn: replace target_addr with BD"

This reverts commit 617e5af215856d2b80f3c0d7c92d97e4a4e0b2fc.

2 years agopower_insn: switch back to target_addr
Dmitry Selyutin [Sun, 4 Sep 2022 09:02:41 +0000 (12:02 +0300)]
power_insn: switch back to target_addr

2 years agoreallocate opcodes for ffadds (converted to X-FORM) and fdmadds to make space for...
Jacob Lifshay [Sun, 4 Sep 2022 07:53:30 +0000 (00:53 -0700)]
reallocate opcodes for ffadds (converted to X-FORM) and fdmadds to make space for fptrans

2 years agoclear up assert, clean up table columns
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 01:18:32 +0000 (02:18 +0100)]
clear up assert, clean up table columns

2 years agofix power_table.py to use multi-entries
Luke Kenneth Casson Leighton [Sun, 4 Sep 2022 01:14:38 +0000 (02:14 +0100)]
fix power_table.py to use multi-entries

2 years agopower_insn: support SVP64 verbose mode
Dmitry Selyutin [Sat, 3 Sep 2022 21:15:22 +0000 (00:15 +0300)]
power_insn: support SVP64 verbose mode

2 years agopower_insn: refactor verbose output
Dmitry Selyutin [Sat, 3 Sep 2022 21:06:46 +0000 (00:06 +0300)]
power_insn: refactor verbose output

2 years agopower_insn: support verbose binary
Dmitry Selyutin [Sat, 3 Sep 2022 20:50:18 +0000 (23:50 +0300)]
power_insn: support verbose binary

2 years agopower_insn: support PPC multi-records
Dmitry Selyutin [Sat, 3 Sep 2022 20:15:54 +0000 (23:15 +0300)]
power_insn: support PPC multi-records

2 years agopower_insn: update disassembly target_addr
Dmitry Selyutin [Sat, 3 Sep 2022 17:44:33 +0000 (20:44 +0300)]
power_insn: update disassembly target_addr

2 years agopysvp64dis: support verbose mode
Dmitry Selyutin [Sat, 3 Sep 2022 17:07:46 +0000 (20:07 +0300)]
pysvp64dis: support verbose mode

2 years agopower_insn: support verbose disassembly mode
Dmitry Selyutin [Sat, 3 Sep 2022 14:18:39 +0000 (17:18 +0300)]
power_insn: support verbose disassembly mode

2 years agopower_insn: implement mode decoding
Dmitry Selyutin [Sat, 3 Sep 2022 10:52:05 +0000 (13:52 +0300)]
power_insn: implement mode decoding

2 years agopower_insn: support Rc detection
Dmitry Selyutin [Sat, 3 Sep 2022 14:51:48 +0000 (17:51 +0300)]
power_insn: support Rc detection

2 years agopower_insn: drop pack/unpack bit
Dmitry Selyutin [Sat, 3 Sep 2022 13:16:05 +0000 (16:16 +0300)]
power_insn: drop pack/unpack bit

2 years agopower_insn: rename normal field to simple
Dmitry Selyutin [Sat, 3 Sep 2022 11:16:30 +0000 (14:16 +0300)]
power_insn: rename normal field to simple

2 years agopower_insn: decouple IMM/IDX LD/ST modes
Dmitry Selyutin [Sat, 3 Sep 2022 11:08:47 +0000 (14:08 +0300)]
power_insn: decouple IMM/IDX LD/ST modes

2 years agopower_insn: support operands check
Dmitry Selyutin [Sat, 3 Sep 2022 10:09:30 +0000 (13:09 +0300)]
power_insn: support operands check

2 years agopower_insn: canonicalize Rc field name
Dmitry Selyutin [Sat, 3 Sep 2022 10:06:05 +0000 (13:06 +0300)]
power_insn: canonicalize Rc field name

2 years agopower_insn: support mode selector
Dmitry Selyutin [Fri, 2 Sep 2022 22:11:32 +0000 (01:11 +0300)]
power_insn: support mode selector

2 years agopower_insn: support normal mode
Dmitry Selyutin [Fri, 2 Sep 2022 22:04:48 +0000 (01:04 +0300)]
power_insn: support normal mode

2 years agopower_insn: support LD/ST indexed mode
Dmitry Selyutin [Fri, 2 Sep 2022 21:20:16 +0000 (00:20 +0300)]
power_insn: support LD/ST indexed mode

2 years agopower_insn: simplify fields
Dmitry Selyutin [Fri, 2 Sep 2022 21:17:39 +0000 (00:17 +0300)]
power_insn: simplify fields

2 years agopower_insn: decrease LDSTMode class nesting
Dmitry Selyutin [Fri, 2 Sep 2022 21:01:47 +0000 (00:01 +0300)]
power_insn: decrease LDSTMode class nesting

2 years agopower_insn: support LD/ST immediate mode
Dmitry Selyutin [Fri, 2 Sep 2022 20:46:10 +0000 (23:46 +0300)]
power_insn: support LD/ST immediate mode

2 years agopower_insn: make RM class public
Dmitry Selyutin [Fri, 2 Sep 2022 16:56:39 +0000 (19:56 +0300)]
power_insn: make RM class public

2 years agopower_insn: remap RM immediately
Dmitry Selyutin [Fri, 2 Sep 2022 16:34:21 +0000 (19:34 +0300)]
power_insn: remap RM immediately

2 years agopower_fields: allow slicing mappings
Dmitry Selyutin [Fri, 2 Sep 2022 21:34:29 +0000 (00:34 +0300)]
power_fields: allow slicing mappings

2 years agopower_fields: allow slicing fields
Dmitry Selyutin [Fri, 2 Sep 2022 16:34:11 +0000 (19:34 +0300)]
power_fields: allow slicing fields

2 years agopower_fields: create arrays from Array class
Dmitry Selyutin [Fri, 2 Sep 2022 15:04:32 +0000 (18:04 +0300)]
power_fields: create arrays from Array class

2 years agosv_binutils: shorten and simplify the output
Dmitry Selyutin [Fri, 2 Sep 2022 21:48:01 +0000 (00:48 +0300)]
sv_binutils: shorten and simplify the output

2 years agocreate list of opcodes by dict entry
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 18:38:53 +0000 (19:38 +0100)]
create list of opcodes by dict entry

2 years agocorrect table header
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 18:32:34 +0000 (19:32 +0100)]
correct table header

2 years agouse opcode directly
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 18:30:08 +0000 (19:30 +0100)]
use opcode directly

2 years agodivpoint 2 to match v3.1
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 18:07:39 +0000 (19:07 +0100)]
divpoint 2 to match v3.1

2 years agocreate correct divpoint to make match against v3.0 Appendix C
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 18:05:45 +0000 (19:05 +0100)]
create correct divpoint to make match against v3.0 Appendix C

2 years agocomplete markdown table
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 17:43:29 +0000 (18:43 +0100)]
complete markdown table

2 years agocorrect table-matching
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 17:00:54 +0000 (18:00 +0100)]
correct table-matching

2 years agoenumeration almost there
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 16:48:09 +0000 (17:48 +0100)]
enumeration almost there

2 years agoMSB0-order, xomask 31-start
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 16:37:47 +0000 (17:37 +0100)]
MSB0-order, xomask 31-start

2 years agoadd power_table.py start of creating markdown Appendix D tables
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 16:33:46 +0000 (17:33 +0100)]
add power_table.py start of creating markdown Appendix D tables

2 years agoadd svshape2 offset test demonstrating RA being offset by one
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 12:45:12 +0000 (13:45 +0100)]
add svshape2 offset test demonstrating RA being offset by one

2 years agoRevert "add inv option to svshape2 (only 1 bit)"
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 12:13:55 +0000 (13:13 +0100)]
Revert "add inv option to svshape2 (only 1 bit)"

This reverts commit 77a4f7104968859385e0b8117ea74041cbe6e436.

the reason is that the inclusion of an inv bit can only be
done by reducing the range of "offset", from 0-15 to 0-7.

as this is the *only way* to get at elements on EXTRA2-encoding
it is considered "unwise"

2 years agoadd inv option to svshape2 (only 1 bit)
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 11:59:06 +0000 (12:59 +0100)]
add inv option to svshape2 (only 1 bit)
https://bugs.libre-soc.org/show_bug.cgi?id=911

there is precious space: an sv.svshape2 can add more bits later

2 years agoupdate sv_analysis to create separate SVMode.LDST_IDX from SVMode.LDST_IMM
Luke Kenneth Casson Leighton [Sat, 3 Sep 2022 11:06:00 +0000 (12:06 +0100)]
update sv_analysis to create separate SVMode.LDST_IDX from SVMode.LDST_IMM
to help distinguish the two SVP64.RM LD/ST types

2 years agofix test_caller_svshape2.py
Jacob Lifshay [Sat, 3 Sep 2022 01:50:01 +0000 (18:50 -0700)]
fix test_caller_svshape2.py

2 years agoformat code
Jacob Lifshay [Sat, 3 Sep 2022 01:44:08 +0000 (18:44 -0700)]
format code

2 years agoadd test_caller_svshape2.py and make corrections to csv and fields.txt
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 20:46:53 +0000 (21:46 +0100)]
add test_caller_svshape2.py and make corrections to csv and fields.txt
yx needed to be SVM2-form and was in a different bitposition
offs needed to be 6:9 not 6:10
XO was off-by-one in minor_22.csv

2 years agoadd first svshape2 pseudocode, based on svindex
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 17:44:22 +0000 (18:44 +0100)]
add first svshape2 pseudocode, based on svindex
it is near-identical but is actually back to svshape in terms of
the SHAPE bits set. bits 18-20 of SVSHAPEn are set in "Matrix" mode
not "Indexed" mode though.

2 years agoadd svshape2 to ISACaller
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 17:41:52 +0000 (18:41 +0100)]
add svshape2 to ISACaller
first recognising the persistence mode bit, second as not an illegal op

2 years agoadd svshape2 to sv/trans/svp64.py
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 15:33:01 +0000 (16:33 +0100)]
add svshape2 to sv/trans/svp64.py
https://bugs.libre-soc.org/show_bug.cgi?id=911

2 years agoadd svshape2 to list of instructions in power_enums.py
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 15:29:11 +0000 (16:29 +0100)]
add svshape2 to list of instructions in power_enums.py

2 years agowhoops bit 25 is sk not vf in svshape2. matches with svindex
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 15:25:52 +0000 (16:25 +0100)]
whoops bit 25 is sk not vf in svshape2. matches with svindex

2 years agodisallow reserved SVrm values in svshape, svp64.py
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 15:16:47 +0000 (16:16 +0100)]
disallow reserved SVrm values in svshape, svp64.py

2 years agoadd svshape2 (stub pseudocode) fields, Form, and CSV file minor_22.csv
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 14:43:17 +0000 (15:43 +0100)]
add svshape2 (stub pseudocode) fields, Form, and CSV file minor_22.csv
https://bugs.libre-soc.org/show_bug.cgi?id=911

also added missing SVI-Form (svindex) SVd and rmm which are exactly
the same.  svshape2 is weird, it is a hybrid of svindex and svshapes

2 years agoadd explicit 13 patterns for svshape which make a hole for svshape2
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 14:40:59 +0000 (15:40 +0100)]
add explicit 13 patterns for svshape which make a hole for svshape2
svshape2 will take up 2 out of the 16 4-bit patterns so svshape has
to explicitly list them. this takes advantage of the new feature
added by ghostmansd: opcode merging, commit d5ec553e768

2 years agoshuffle down numbering after SVM to make room for SVM2
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 14:37:50 +0000 (15:37 +0100)]
shuffle down numbering after SVM to make room for SVM2
power_enums.py Form needs to make space for SVM2
(and svshape is no longer temporary)

2 years agoadd fix of out_sel in power_decoder.py formal proof
Luke Kenneth Casson Leighton [Fri, 2 Sep 2022 12:32:45 +0000 (13:32 +0100)]
add fix of out_sel in power_decoder.py formal proof
but there are others.  basically the list of enums needs to be
automatically listed and put into the matches