Graf commitů

703 Commity

Autor SHA1 Zpráva Datum
xuchen-amd 9ddee8de2e TUI: Kernel Selection (#769) 2025-08-01 11:31:43 -04:00
Fei Zheng 6cbda3f1d8 Format code (#842) 2025-07-31 08:20:49 -06:00
Fei Zheng 3a40e4b58c quick fix how to call v3 with pc sampling 2025-07-31 08:17:06 -06:00
vedithal-amd ea4769f604 Fix rocprof avail usage (#840) 2025-07-30 13:52:55 -04:00
fxmarty-amd 56d040156e bugfix to make amd-smi usage backward compatible (#836)
* Update soc_base.py

Fixes https://github.com/ROCm/rocprofiler-compute/issues/835

Signed-off-by: fxmarty-amd <felmarty@amd.com>

* address comments

---------

Signed-off-by: fxmarty-amd <felmarty@amd.com>
2025-07-30 09:40:04 -04:00
vedithal-amd 80ea339217 Fix test cases (#837)
* Fix formatting
2025-07-28 16:19:45 -04:00
vedithal-amd 03d27c0ba0 Enable rocpd output format with rocprofiler sdk (#790)
* Add `rocpd` choice for `--format-rocprof-output` option
* Add rocpd_data.py which defines SQL queries to extract data from rocpd database
* Use sqlite3 package to read the database
* Add `--retain-rocpd-output` option in profile mode to retain raw
  rocpd database
* Add warning notice to say `--format-rocprof-output rocpd` will be
  default in future release

For rocpd output:
* Use only `pmc_perf.csv` instead of reading individual coll_level results csv files
* Post process csv files using pandas in analysis mode instead of profile mode
* Use ACCUM counters instead of SQ_ACCUM_PREV_HIRES

* Add test cases for rocpd output format
* Fix code formatting issues
* Update CHANGELOG
2025-07-28 11:02:28 -04:00
vedithal-amd 6885cb068d add description for MI100 counters (#834) 2025-07-26 15:33:23 -04:00
vedithal-amd bb44e90b2d Unified configuration for metrics (#726)
* Show description of metrics during analysis
    * Use --include-cols Description show the Description column in analyze mode (this is hidden by default)
    * Remove tips field from analysis config

* Align metric names in analysis config and documentation

* Add unified config utils/unified_config.yaml

* Add python script utils/split_config.py to auto generate analysis configuration and documentation metrics description
   * Add test case to ensure unified config is older than auto-generated config
   * Auto generate analysis config and documentation metrics description

* Update CONTRIBUTING.md to add instructions to build documentation assets
    * Add docker image and compose file to build documentation

* Update CHANGELOG and Documentation

* Use jinja template instead of hardcoding metric tables in documentation
2025-07-25 14:01:34 -04:00
xuchen-amd 99a6e67bcc Improve --time-unit arg (#807) 2025-07-24 12:15:52 -04:00
vedithal-amd dbcaccb9de Fix rocprofv3 supported counters not being detected (#832)
* Fix rocprofv3 supported counters not being detected

* Fix rocprof interface deprecation warning appearing twice
2025-07-24 11:50:07 -04:00
vedithal-amd d4c316a730 Improve baseline comparison (#817)
* Do not force unsupported metrics to be specified in older gpu
  architectures as None

* Remove metrics which are explicitly set to None

* Update CHANGELOG

* Fix analysis configuration to fix baseline comparisons across all gpu
  architectures
    * Add missing 1812 section for gfx908
    * Add missing 1812 section for gfx90a

* Baseline comparision will only show common metrics
   * First workload will be used to set Metric ID index column
2025-07-24 11:49:02 -04:00
Fei Zheng 137f35e700 Fix L2 read/write/atomic bandwidths on MI350 (#831) 2025-07-23 15:46:19 -06:00
vedithal-amd a70ae40ddc Improve block filtering to accept metric ids (#821)
* Fix tests
* Update CHANGELOG and documentation
2025-07-23 16:16:29 -04:00
vedithal-amd 068e5aaef4 Fix tests and formatting (#826) 2025-07-22 17:49:16 -04:00
vedithal-amd e132b047cd Standalone GUI bugfix (#825)
* Fix barchart elements table ids

* Add HBM bandwidth section to L2 cache report for gfx950

* bugfix for standlone GUI

Co-authored-by: Felix Marty <Felix.Marty@amd.com>
2025-07-22 15:31:54 -04:00
vedithal-amd 0fdd186dd5 Fix memory clock detection with amd-smi (#824) 2025-07-22 13:46:06 -04:00
vedithal-amd 98bb0f4237 Remove hardware IP block based filtering (#820)
* Analysis report block based filtering is the default now

* Update documentation

* Update CHANGELOG

* Fix tests
    * Replace hardware block based filtering tests with report block
      based filtering tests
2025-07-21 09:37:35 -04:00
vedithal-amd 537a269e95 Remove MI50/MI60 gfx906 support per documentation (#819) 2025-07-18 09:57:33 -04:00
cfallows-amd 1d834f4190 Fix roofline block print during CLI output (#811)
Fix roofline panel in CLI analyze stage when using --block option. Improve roofline CLI output checks and logs.

---------

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>
2025-07-15 12:42:27 -04:00
Fei Zheng e1da9355c2 Update peak flops for MI350 (#810) 2025-07-15 08:46:13 -06:00
xuchen-amd b4b50502e3 Remove rocm-smi for ROCm 7.1 (#809) 2025-07-15 10:28:53 -04:00
xuchen-amd a3dc98e251 rocm-smi deprecation warning (#806) 2025-07-14 13:35:47 -04:00
xuchen-amd 23664c72f7 tui user experience improvement (#805) 2025-07-11 16:51:55 -04:00
vedithal-amd 000fd4f5b2 Fix roofline and TUI bugs (#803)
* Fix roofline rocm version bug
* Fix utils bug
* Remove unnecessary tests
* Do not check textual-fspicker package in cmake build
* Use rocprofv3 to test MI 100 and fix tests
2025-07-09 19:15:46 -04:00
xuchen-amd 8585cd8e8f Generalize config path. (#802)
* Generalize config path.

* Fix format.

* Fix typo.
2025-07-09 18:57:21 -04:00
cfallows-amd 3df0d1dded Update roofline bins for rocm7 changes (#755)
* Update current bins to have rocm6 suffix. Add new rocm7 bins, built on rocm7.0 latest due to hip updates.

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

* Re-add rocm version check for roof bins.

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

* Rebuild roofline binaries on top of latest rocm7 (#16379) after recent mainline promotions.
Adjusting version and distro combinations of bins following rocm6 vs rocm 7 supported OS.
*rhel8 not supported on rocm7, also not built anymore
*sles15 not supported on rocm7 but is still being built
*ubuntu stays as 22.04 and above for rocm7

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

* Minor fixes after testing.

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

* Fixed bad copy after finding it in testing ctest.

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

* Remove runpath from new bin

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

* Rework target_binary map return in detect_roofline- we should not be returning maps of different sizes or with different keys for the same method. Expected output should be consistent in case we run into bad position, or for testing purposes. Manually tested all possible roofline bin expected cases to comfirm functionality and expected user output.

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

* Update changelog with new roofline distro minimums

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

---------

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>
2025-07-08 16:51:50 -04:00
Fei Zheng 1b7445faa4 PC Sampling: fix potential_end_offset (#792)
* fix potential_end_offset

* format code: unrelated
2025-07-08 08:52:01 -04:00
fxmarty-amd c58ee9e73b fix normalization (#789) 2025-07-08 08:50:41 -04:00
cfallows-amd a59b1ea6e6 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>
2025-07-02 10:36:37 -04:00
vedithal-amd 894046f283 Fix rocprofiler sdk list counters (#765) 2025-07-02 09:41:51 -04:00
xuchen-amd 222dbdc6bd Remove cell backgrouund colors. (#778) 2025-06-28 10:38:28 -04:00
jamessiddeley-amd 1bb7bf4dd4 added 0400 analysis config yaml files (#779) 2025-06-27 16:21:19 -04:00
vedithal-amd 6e1bbb5efb 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
2025-06-27 14:07:56 -04:00
vedithal-amd a95a45d69a 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
2025-06-26 09:03:18 -04:00
cfallows-amd 630bc149ff 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>
2025-06-25 01:19:28 -04:00
xuchen-amd 74e9853f8e Fix system sol units, remove units if None value. (#773)
* Fix system sol units.
2025-06-24 16:37:55 -04:00
xuchen-amd 92ce4732bc GUI: prevent big number overflow (#772)
* Prevent number overflow by formatting values.

* Improve value formatting logic based on comments.
2025-06-24 16:27:06 -04:00
vedithal-amd 5950a02ff6 Fix counter collection for MI 350 (#764)
* Fix TCC counter collection bug
* Fix accumulate counter collection
* Fix formatting
2025-06-21 08:48:25 -04:00
jamessiddeley-amd 036866fc12 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
2025-06-18 14:09:09 -04:00
xuchen-amd 5d8fd55e36 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.
2025-06-18 13:58:42 -04:00
xuchen-amd 80159ee117 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
2025-06-18 13:19:58 -04:00
Vladimir Indic 451be8a600 [PC Sampling] Separate counts for issued and stalled samples (#756) 2025-06-16 15:09:21 -06:00
vedithal-amd 7c0ffd14a9 Print counter list in DEBUG level (#758) 2025-06-16 16:53:00 -04:00
ywang103-amd b8dd6d049d format the code after mem chart's fix of test (#753) 2025-06-13 12:03:42 -04:00
ywang103-amd bb0c417871 fix broken test with --cols options (#750) 2025-06-12 19:45:24 -04:00
Vignesh Edithal 6054b3b7fd Bugfix for PR #744 2025-06-12 16:09:15 -04:00
vedithal-amd d27ee69b52 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
2025-06-12 15:45:11 -04:00
vedithal-amd cdd41dee40 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
2025-06-12 14:21:24 -04:00
David Galiffi 1cd989a110 Copyright Header Compliance (#745)
- for SWDEV-537492
2025-06-12 12:02:58 -04:00