JanLJL
607d459569
keep dependency paths as generators instead of lists
2021-04-17 12:46:44 +02:00
JanLJL
b033b3b7aa
allow different base with prefix for offset values
2021-04-17 11:06:39 +02:00
Julian Hammer
0c295dc847
version bump
2021-04-15 14:47:09 +02:00
Julian
08440ed5e1
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
25a0e0607d
added missing instructions to all DBs
2021-04-05 16:47:52 +02:00
JanLJL
b0e35316f0
changed consideration of masking for database back to NO
2021-03-25 11:50:17 +01:00
JanLJL
94313ec772
added new instructions and bugfixed the wrong import from uops.info for masks for A FEW instructions
2021-03-25 11:47:15 +01:00
Julian Hammer
63563ecabc
flake8 to ignore some errors and small style improvements
2021-03-11 12:52:34 +01:00
Julian Hammer
b7625a4a25
making flake8 happy
2021-03-11 12:29:14 +01:00
Julian Hammer
6204c90934
migrate code style to Black
2021-03-11 12:02:45 +01:00
Julian Hammer
1ebe5ecfbd
sanity check validity of operand entries
2021-03-11 11:38:25 +01:00
Julian Hammer
0c5ac26f3f
fixed typos and added missing default load/store info
2021-03-11 10:10:09 +01:00
JanLJL
9a13e5cbc5
guarantee 0 latency for None values in DB
2021-03-11 01:55:57 +01:00
Julian Hammer
dcf3e38612
fix #66 typo in icl.yml
...
cmp failed with immediates
2021-03-10 11:57:21 +01:00
Julian Hammer
d7a687909e
passing parsing errors to the outside
2021-03-05 18:07:36 +01:00
JanLJL
74a479fb95
fixed AArch64 parser for register shifts and new instructions for A64FX
2021-02-25 07:43:42 +01:00
Julian Hammer
4fdf312622
add "sub gpr, gpr" to skx
2021-02-15 11:41:21 +01:00
Julian Hammer
803da767f2
add "vmovapd zmm, zmm" to skx
2021-02-15 11:41:20 +01:00
JanLJL
0e69db9de9
new instructions
2021-01-31 18:11:06 +01:00
JanLJL
9f87606ce8
minor model fixes
2021-01-26 12:56:19 +01:00
JanLJL
768a90de10
cover kernel with only unknown instructions
2021-01-07 12:40:16 +01:00
Jan
8c9557760e
Merge pull request #62 from jdomke/attfix
...
att parser: workaround for crash with "jg,pt" mnemonic
For now we will ignore the branch taken/not-taken indication and will only keep the condition in the mnemonic.
2021-01-07 10:23:39 +01:00
Jan
4dbcfbda5d
suppress branch indication in mnemonic for now
2021-01-07 10:20:57 +01:00
JanLJL
ed13cde61b
fixed DV port bug
2021-01-07 09:59:18 +01:00
jdomke
4f8ed13309
att parser: support for more mnemonic
...
found some 'jg,pt' in icc/mkl generated binaries which crashed the
parser, here an example:
dd8ccd: 3e 7f 90 jg,pt dd8c60 <mkl_blas_avx2_dtrsm_kernel...
2021-01-05 15:26:22 +09:00
JanLJL
3c7971b347
fixes #61
2020-12-28 10:33:54 +01:00
JanLJL
60f792c4b2
new instructions
2020-12-17 12:38:58 +01:00
JanLJL
449932d75b
version bump
2020-12-11 01:00:30 +01:00
JanLJL
8e3d613843
new instructions
2020-12-09 11:52:10 +01:00
JanLJL
2093610bbf
add support for :: delimiter in x86 labels and identifiers
2020-12-07 02:45:49 +01:00
JanLJL
f9f1120da6
added support for absolute address operands without brackets
2020-12-07 01:57:06 +01:00
JanLJL
e87ab5d6ca
new instruction
2020-12-07 01:18:32 +01:00
JanLJL
82b35e7649
new instruction
2020-12-07 01:18:32 +01:00
JanLJL
23623ca18a
enhancements for lookup and parsing AArch64 instrs
2020-12-07 01:18:32 +01:00
JanLJL
b9e434d124
new instructions
2020-12-07 01:18:32 +01:00
JanLJL
81ce395115
added the possibility of a 5th operand
2020-12-06 18:05:59 +01:00
JanLJL
f41854a0a6
better port order and a few new instructions
2020-12-06 18:05:36 +01:00
JanLJL
d7e5e12961
version bump
2020-11-23 10:43:00 +01:00
JanLJL
6bc6349c25
fetch version from __init__ file and write uarch in upper case
2020-11-21 21:33:33 +01:00
JanLJL
f69b5f88f0
removed false entries
2020-11-21 21:02:44 +01:00
JanLJL
596a323dfb
bugfixes
2020-11-21 21:00:58 +01:00
JanLJL
08b4586b71
replaced b4799d1d with smarter solution
2020-11-20 15:44:49 +01:00
JanLJL
ffb263e20f
more instructions
2020-11-20 15:37:32 +01:00
JanLJL
b4799d1d45
bugfix for immediate shifts with hex-base
2020-11-20 15:22:05 +01:00
Julian Hammer
4ff8fdc4ab
version bump
2020-11-11 15:14:27 +01:00
JanLJL
92c162daa2
new instructions
2020-11-11 13:54:23 +01:00
JanLJL
87ea8f0f0a
new instructions
2020-11-11 12:27:49 +01:00
Julian Hammer
cb04efc384
fixed typo
2020-11-10 13:33:24 +01:00
JanLJL
14c0ea6180
bugfixes
2020-11-09 23:29:42 +01:00
Julian Hammer
314ff4cf9d
improved performance of arch_semantics and reg dependency matching
2020-11-09 19:27:47 +01:00
