Графік комітів

674 Коміти

Автор SHA1 Повідомлення Дата
cfallows-amd e085f3cc54 Fix method return for python version compatibility (#784)
Support for python3.8 and above means union defined in one of the methods was throwing errors for anything less than python3.10. Swapping out | operand for Optional[] resolves errors on systems using <3.10. No functional changes.

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: a59b1ea6e6]
2025-07-02 10:36:37 -04:00
vedithal-amd 740fe492c2 Fix rocprofiler sdk list counters (#765)
[ROCm/rocprofiler-compute commit: 894046f283]
2025-07-02 09:41:51 -04:00
xuchen-amd bfe24beec6 Remove cell backgrouund colors. (#778)
[ROCm/rocprofiler-compute commit: 222dbdc6bd]
2025-06-28 10:38:28 -04:00
jamessiddeley-amd 60ace87e1a added 0400 analysis config yaml files (#779)
[ROCm/rocprofiler-compute commit: 1bb7bf4dd4]
2025-06-27 16:21:19 -04:00
vedithal-amd 86e243201e Fix SPI block counter limit (#780)
* Fix SPI counter collection limit

* Update counter collection limits from aqlprofile

* Remove special handling for MI 350 SPI pipe counters

[ROCm/rocprofiler-compute commit: 6e1bbb5efb]
2025-06-27 14:07:56 -04:00
vedithal-amd 1d59cbb06d Add support for MI 100 with rocprofiler-sdk (#768)
* Add custom rocprofiler-sdk counter definitions file for MI 100

* Update CHANGELOG to mention that accumulation counters will not be
  collected when profiling on MI 100 using rocprofiler-sdk/rocprofv3

* Migrate accum_counters.yaml to code

[ROCm/rocprofiler-compute commit: a95a45d69a]
2025-06-26 09:03:18 -04:00
cfallows-amd 429b17a1e0 Add roofline PDF output to general profiling runs (#774)
Change when Roofline PDFs are generated- during general profiling and --roof-only profiling (skip only when --no-roof option is present)

---------

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: 630bc149ff]
2025-06-25 01:19:28 -04:00
xuchen-amd b7c6c4eaac Fix system sol units, remove units if None value. (#773)
* Fix system sol units.

[ROCm/rocprofiler-compute commit: 74e9853f8e]
2025-06-24 16:37:55 -04:00
xuchen-amd 7ab4a3f03f GUI: prevent big number overflow (#772)
* Prevent number overflow by formatting values.

* Improve value formatting logic based on comments.

[ROCm/rocprofiler-compute commit: 92ce4732bc]
2025-06-24 16:27:06 -04:00
vedithal-amd 9873460265 Fix counter collection for MI 350 (#764)
* Fix TCC counter collection bug
* Fix accumulate counter collection
* Fix formatting

[ROCm/rocprofiler-compute commit: 5950a02ff6]
2025-06-21 08:48:25 -04:00
jamessiddeley-amd 979c59fadc added progress printout in rocprof-compute profile (#727)
* added progress printout for profiler

* added comments and fixed readability

* removed redundant newlines

* moved format_time helper function to utils

* removed tqdm and redundant time calc

[ROCm/rocprofiler-compute commit: 036866fc12]
2025-06-18 14:09:09 -04:00
xuchen-amd 9d50115bd7 Integrate roofline in tui (#762)
* Update changelog, remove unused code.

* enable roofline in TUI.

* Remove roofline section when data not available.

* Fix workload dir path.

[ROCm/rocprofiler-compute commit: 5d8fd55e36]
2025-06-18 13:58:42 -04:00
xuchen-amd 0b6e03adb0 CLI: enable roofline plot (#760)
* enable roofline plot in cli.

* Add roofline to analysis config.

* Unify global variables.

* Disable roofline for baseline comparison and gfx908.

* Add check for roofline.csv

[ROCm/rocprofiler-compute commit: 80159ee117]
2025-06-18 13:19:58 -04:00
Vladimir Indic f49b70142f [PC Sampling] Separate counts for issued and stalled samples (#756)
[ROCm/rocprofiler-compute commit: 451be8a600]
2025-06-16 15:09:21 -06:00
vedithal-amd fccb9195ff Print counter list in DEBUG level (#758)
[ROCm/rocprofiler-compute commit: 7c0ffd14a9]
2025-06-16 16:53:00 -04:00
ywang103-amd 72e8576a7e format the code after mem chart's fix of test (#753)
[ROCm/rocprofiler-compute commit: b8dd6d049d]
2025-06-13 12:03:42 -04:00
ywang103-amd dab2c49342 fix broken test with --cols options (#750)
[ROCm/rocprofiler-compute commit: bb0c417871]
2025-06-12 19:45:24 -04:00
Vignesh Edithal 696c1407af Bugfix for PR #744
[ROCm/rocprofiler-compute commit: 6054b3b7fd]
2025-06-12 16:09:15 -04:00
vedithal-amd 89bc1fc812 Change default rocprof to rocprofv3 (#748)
* Revert of https://github.com/ROCm/rocprofiler-compute/pull/738

* Change default rocprof backend interface to rocprofv3

* Add MI 350 support in documentation

* Added known issue that MI 100 profiling will not work unless rocprofv1
  is explicitly opted in

* Remove MI 50 soc gfx python class since MI 50 is not supported

[ROCm/rocprofiler-compute commit: d27ee69b52]
2025-06-12 15:45:11 -04:00
vedithal-amd 9e743cdff2 Remove rocscope related code and add deprecation warning for mongo db usecase (#744)
* Remove rocscope related code

* Add deprecation warning for database update mode which is used for grafana and mongodb functionality

[ROCm/rocprofiler-compute commit: cdd41dee40]
2025-06-12 14:21:24 -04:00
David Galiffi a153c96c7b Copyright Header Compliance (#745)
- for SWDEV-537492

[ROCm/rocprofiler-compute commit: 1cd989a110]
2025-06-12 12:02:58 -04:00
jamessiddeley-amd 218d77b88f fixed long kernel names cut off in --kernel-names option (#728)
* reformatted kernel roofline PDF to use table

* restored kernel symbol icons

* enhance code readability

* restored cell text wrap

---------

Signed-off-by: jamessiddeley-amd <James.Siddeley@amd.com>

[ROCm/rocprofiler-compute commit: f004aeebe9]
2025-06-12 10:23:40 -04:00
cfallows-amd 56f38e3b60 Add roofline cli_generate_plot method (#737)
Add option to print out roofline plot in terminal using plotext.
Takes in one datatype and returns the str from plot.build() which contains the visual plot of roofline analysis for said datatype.

---------

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: 0415bb9740]
2025-06-11 15:38:21 -04:00
cfallows-amd 192c63f2f6 Update roofline binaries (#741)
Update roofline binaries from rocm-amdgpu-bench
- uses hip to find number of CUs dynamically instead of hardcoded values in table

Remove duplicate AI plot points printing
- only print ai points once on plot since we are measuring using total flops and value is same
- remove datatype from legend labels

---------

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: 24d3e7eecd]
2025-06-10 15:43:56 -04:00
cfallows-amd 5de2375db1 Fix load_kernel_top arg for GUI analyze mode (#740)
--gui option for analyze mode failing due to missing arg in load_kernel_top call in pre_processing

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: ce3ef1400e]
2025-06-10 11:27:03 -04:00
vedithal-amd c7cea9c971 Revert default interface to rocprof v1 (#738)
* Add deprecation warning for rocprof v1 / v2 / v3 interfaces to use
  rocprofiler-sdk interface

[ROCm/rocprofiler-compute commit: a1ef3425c6]
2025-06-09 16:39:11 -04:00
Fei Zheng d750ea6616 TUI improvement (#732)
[ROCm/rocprofiler-compute commit: 96aa04fb13]
2025-06-09 11:29:10 -06:00
vedithal-amd dd39c649d9 Bugfix for rocprofiler sdk interface not working in MI 200 (#733)
[ROCm/rocprofiler-compute commit: 721053bd03]
2025-06-09 12:33:25 -04:00
Fei Zheng 9a392edb21 CLI: enable mem_chart for single run (#643)
[ROCm/rocprofiler-compute commit: e5b31af2a4]
2025-06-06 16:15:56 -06:00
Fei Zheng 243aa68712 Support stochastic pc sampling
[ROCm/rocprofiler-compute commit: d756aeb3fd]
2025-06-06 12:43:52 -06:00
xuchen-amd 0db02c5fd6 Introduce rocprof-compute TUI (Text User Interface) (#682)
* rocprof-compute TUI (Text User Interface) - providing users interactive analyze experience with visuals.

* Analyze results with tables, charts, plots.

* Add menu bar, terminal, directory dialog. Improve logging and ui.

* Add display config file to manipulate result categorization.

* Add support for recently opened dirs.

* Update licensing and version.

[ROCm/rocprofiler-compute commit: ca0cdaf948]
2025-06-04 17:06:08 -04:00
Fei Zheng dd2d9cddf0 Fix peak flops of F8 I8 F16 and BF16 on MI300
[ROCm/rocprofiler-compute commit: ab6665d317]
2025-06-04 12:51:46 -06:00
ywang103-amd d1feafe4db Tcc new format input yaml (#723)
[ROCm/rocprofiler-compute commit: e5c7d4795a]
2025-06-04 12:24:57 -04:00
xuchen-amd 6c40056aa9 Add chip specs (#681)
* Add perfmon config spec, enhance memory partition info.

* Add gfx950 perfmon config.

* Add High Freq variants in gfx942.

* Add backup detection methods for gpu model.

* Improve get_num_xcds logic by adding detection of 1to1 arch-to-compute_partition logic.

* Add default compute partition settings spx:8 for when gpu_model=None.

* Update gpu spec tests.

* Add backup compute partition detection.

---------

Signed-off-by: xuchen-amd <xuchen@amd.com>

[ROCm/rocprofiler-compute commit: f0fad19e8b]
2025-05-29 16:35:34 -04:00
Ben Richard 7fd93cfed3 Upgrade to Dash 3.0 (#719)
[ROCm/rocprofiler-compute commit: 45296ceb46]
2025-05-29 14:36:03 -04:00
cfallows-amd 9eaeb16acc Rename roofline bins (#717)
Rename roofline bins, remove rocm version in naming. Change method for binary search.

---------

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: bbe2e17b80]
2025-05-28 14:55:51 -04:00
cfallows-amd 689746e2cd Add F4 F6 to roofline for MI350 series (#709)
Add roofline bins with FP4 FP6 datatypes enabled for gfx950 arch

---------

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: cb2d928ecf]
2025-05-26 18:36:31 -04:00
jamessiddeley-amd ba61cc36f5 Fixed duplicate keys in analysis_configs yamls (#707)
* fixed duplicate keys in analysis_configs yamls

* Fix: removed TODO comment

Signed-off-by: James Siddeley <james.siddeley@amd.com>

---------

Signed-off-by: James Siddeley <james.siddeley@amd.com>

[ROCm/rocprofiler-compute commit: 09b6ef4508]
2025-05-20 13:12:46 -04:00
Ben Richard 4899cfa160 Update illegal character check for profile name (#703)
[ROCm/rocprofiler-compute commit: 41dd4aab90]
2025-05-16 15:45:16 -04:00
cfallows-amd 72af81c528 Check mode during soc init for roofline (#705)
Check mode before creating roofline object- skip if only printing specs

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: 43dbf38b27]
2025-05-16 12:54:53 -04:00
vedithal-amd 6dc3fad1c7 Implement interface to rocprofiler sdk (#695)
* Setting ROCPROF=rocprofiler-sdk environment variable will use rocprofiler-sdk C++ library instead of rocprofv3 python script

* Add runtime option --rocprofiler-sdk-library-path to use custom version of rocprofiler sdk library
    * Add --rocprofiler-sdk-library-path conftest option for tests

* Setup appropriate environment variables to inject rocprofiler sdk code to user command
    * Add env. vars. for counter collection and filtering
    * Add env. vars. for pc sampling

* Use python bindings to list counters supported by rocprofiler sdk

[ROCm/rocprofiler-compute commit: 5cb86e31fc]
2025-05-13 10:48:21 -04:00
cfallows-amd e025a04a60 Fix setting roofline-data-type option in both profile and analyze modes (#702)
Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: d527d77337]
2025-05-12 23:02:47 -04:00
vedithal-amd bd04d43108 Fix PC sampling analysis config issue (#697)
[ROCm/rocprofiler-compute commit: abd500593b]
2025-05-06 18:22:15 -04:00
Ben Richard ceb2df6345 Avoid crash when profiling data not generated (#694)
* Avoid crash when profiling data not generated

-Handle case where program has no kernel launches
-Improve error messages
-Avoid roofline when profiling data is missing

Signed-off-by: benrichard-amd <ben.richard@amd.com>

* Update other soc_gfx files to catch missing pmc_perf.csv

* Fix formatting

* Fix incorrectly ordered imports

---------

Signed-off-by: benrichard-amd <ben.richard@amd.com>

[ROCm/rocprofiler-compute commit: 35493f440c]
2025-05-05 16:09:48 -04:00
cfallows-amd c8d67fc97e Enable roofline for MI350 series (#677)
Rework of roofline binaries generated from rocm-amdgpu-bench
- removed arch identifier in bin name
- removed rocm5 bins altogether

Updated required distros for roofline
- updated distro checks and bin naming
- moved up ubuntu20.04->22.04 and sles15.3->15.6 per rocm support

Enabled ctests for mi350 for test_roof_*
- removed mi350 series check to skip these specific tests

---------

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>

[ROCm/rocprofiler-compute commit: 41e73650d5]
2025-04-28 16:08:23 -04:00
Daniel Su 935e2b53b8 Set rocprofV3 agent-index to absolute (#675)
Signed-off-by: Daniel Su <danielsu@amd.com>

[ROCm/rocprofiler-compute commit: b77fcf575e]
2025-04-28 15:38:07 -04:00
xuchen-amd af114a1539 Add test for gfx942 number of xcds. (#674)
* Add test for 9fx942 number of xcds.

* Improve the structure of mi gpu specs, add num_xcds_spec_class test.

* Add to ctest.

---------

Signed-off-by: xuchen-amd <xuchen@amd.com>

[ROCm/rocprofiler-compute commit: 85bfa73e2c]
2025-04-28 11:29:14 -04:00
xuchen-amd 703dc4d6ad process hip trace output. (#654)
Signed-off-by: xuchen-amd <xuchen@amd.com>

[ROCm/rocprofiler-compute commit: ee73c2a119]
2025-04-22 18:31:47 -04:00
xuchen-amd 39c53aadda Patch in new rocprofv3 metrics. (#679)
[ROCm/rocprofiler-compute commit: f145f89e30]
2025-04-22 18:30:26 -04:00
ywang103-amd 303b846a13 change default rocprof version to v3 when not setting env variable (#673)
[ROCm/rocprofiler-compute commit: 3e09f038e5]
2025-04-16 12:38:20 -04:00