JanLJL
fb7f1a289d
flake8 formatting
2025-03-05 10:19:10 +01:00
pleroy
b4d342266d
Add support for the Intel syntax supported by MSVC and ICC
2025-02-02 14:02:16 +01:00
stefandesouza
62c21a7f31
Port pressure returned in tuple with Memory Operand
2024-03-04 20:00:43 +01:00
stefandesouza
9cd841cd08
Added updated files
2024-02-27 14:47:55 +01:00
stefandesouza
33ad20dc3a
More formatting
2024-01-10 13:26:50 +01:00
stefandesouza
1885ce6ddb
flake8 standards
2023-12-03 21:04:58 +01:00
stefandesouza
23d10d10cb
Black formatting
2023-12-03 17:22:11 +01:00
stefandesouza
62d575714a
Fixed semantic and marker tests. Now only dump needs to be adjusted
2023-12-03 16:49:33 +01:00
stefandesouza
37ca6670c7
pre/post-indexed to pre/post_indexed. Now have use ImmediateOperand type for mem offset. Changed some parser tests also
2023-12-02 16:56:43 +01:00
stefandesouza
e77dfb4eb3
Fixed issue with throughput assignment
2023-10-30 19:32:05 +01:00
stefandesouza
78ca6fe855
Added condition operand, adjusted tests to parse it & a few changes to get the kernelDG tests working
2023-10-29 16:36:00 +01:00
stefandesouza
cce05e44cb
Changed style to conform to PEP-8 conventions; Added source and destination attributes to parent Operand class
2023-10-29 13:52:49 +01:00
stefandesouza
17cd1a70c7
Updated tests to use the now class style iforms in isa_data
2023-10-23 16:25:31 +02:00
stefandesouza
2f8c2f56cf
Black formatting
2023-09-12 12:45:28 +02:00
stefandesouza
a761ee43f1
Converting operand types read in from YAML files
2023-09-12 00:23:59 +02:00
stefandesouza
14ecefd677
Changes for operand matching, instruction loading
2023-09-11 18:23:57 +02:00
stefandesouza
0f182551f2
Changes to accomodate the new OO style
2023-08-28 15:19:46 +02:00
stefandesouza
5ca37a2a3f
Added 2 operand types and made changes for attribute usage
2023-08-20 21:01:44 +02:00
JanLJL
7b83ef7b50
fixed UnboundLocalError if tp assignment loop is not executed
2023-07-17 14:52:22 +02:00
JanLJL
d4f6314928
fixed read out of store TP from DB
2023-06-20 21:20:41 +02:00
JanLJL
72f69fa707
added fallback search in arch/ISA model for ARM instructions with shape/cc suffixes
2023-03-03 15:11:40 +01:00
JanLJL
405f390e03
bugfix in store throughput calculation
2022-09-28 14:21:46 +02:00
JanLJL
d5aaceb9c4
added Zen3 support
2022-09-27 18:39:14 +02:00
JanLJL
54da7568d9
small bugfixes
2022-03-17 16:38:28 +01:00
JanLJL
bc7761007c
fixed formatting with correct line length
2021-10-04 15:00:17 +02:00
JanLJL
314feb4104
black formatting
2021-10-04 14:33:28 +02:00
JanLJL
1b40c10a1f
applied flake8 and black rules
2021-08-26 16:58:19 +02:00
Julian
04836cf3f9
Validation ( #71 )
...
Validating of OSACA predictions for IVB, SKX, ZEN1, ZEN2, A64FX and TX2 with different kernels.
build_and_run.py contains the configuration used at RRZE's testcluster and UR's qpace4, Analysis.ipynb contains the analysis script and results. Raw data from measurements (122MB) will be attached to next OSACA release.
For now, find the raw data here: https://hawo.net/~sijuhamm/d/UPIhBOtz/validation-data.tar.gz
The analysis report can be viewed at https://nbviewer.jupyter.org/github/RRZE-HPC/OSACA/blob/validation/validation/Analysis.ipynb
Quite a few changes on OSACA included:
Feature: register change tracking via semantic understanding of operations
Feature: recording LCD latency along path and exposing this to frontend
Feature: support for memory reference aliases
Feature: store throughput scaling (similar to load throughput scaling)
Fix: model importer works with latest uops.info export
Fix: immediate type tracking on ARM now preserves type in internal representaion
Removed unused KerncraftAPI
2021-04-15 14:42:37 +02:00
Julian Hammer
0f8e53e918
making flake8 happy
2021-03-11 12:29:14 +01:00
Julian Hammer
afa607e546
migrate code style to Black
2021-03-11 12:02:45 +01:00
Julian Hammer
7f8b596fc9
improved performance of arch_semantics and reg dependency matching
2020-11-09 19:27:47 +01:00
Julian Hammer
8d7a61aed6
added comment
2020-11-09 12:35:13 +01:00
Julian Hammer
9d59864b51
performance improvement of throughput summation
2020-11-09 12:01:00 +01:00
JanLJL
e5dce9acb4
bugfix in OoO scheduling
2020-07-28 14:52:30 +02:00
JanLJL
7bc39c1343
prepared for aarch64 8.2 support
2020-07-23 15:54:54 +02:00
JanLJL
68964bb2a1
enhanced TP scheduling
2020-07-06 18:49:46 +02:00
JanLJL
9318181902
added reg-only fallback for mem-instructions not found in ISA DB
2020-03-10 17:15:57 +01:00
JanLJL
458b5954b9
more documentation
2020-03-05 18:39:38 +01:00
JanLJL
17018ac6ad
made detection of flag dependencies as opt_in for now
2020-01-29 13:03:43 +01:00
JanLJL
953ec251fd
added new instructions and fixed false positive assignment of stores by dynamic TP/LT combination for aarch64
2020-01-22 21:40:11 +01:00
JanLJL
a3cc742a87
enhanced for dynamic ST throughput combination
2020-01-14 10:49:47 +01:00
JanLJL
e250f69821
added --ignore-unknown flag and major updates in x86 parser
2020-01-09 17:57:08 +01:00
JanLJL
f477255e99
check for non-GAS-suffix mnemonics for instruction forms with MEM ops
2020-01-09 13:21:11 +01:00
JanLJL
5d55f25e4f
tiny bugfix for src_dst operands
2020-01-07 19:42:13 +01:00
JanLJL
7c1dc83658
bugfix
2019-12-19 18:54:47 +01:00
JanLJL
35d2762468
enhanced dynamic combine of LD and arithmetic instr
2019-12-19 18:50:48 +01:00
JanLJL
11f91fe9e1
added default load TP and relocation in identifier
2019-12-18 16:56:20 +01:00
Julian Hammer
63ca1e66bf
FIX #46 untangled semantic and non-semantic operand info
2019-11-14 16:43:33 +01:00
Julian Hammer
51a4b989d0
Merge branch 'master' of github.com:RRZE-HPC/OSACA
2019-11-13 12:52:34 +01:00
Julian Hammer
455fd1a972
fixed semantics testcase to join lists
2019-11-13 12:39:16 +01:00
