76333 Commits

Autor SHA1 Mensaje Fecha
Gollamandala, Srinivasarao f5398c7dde SWDEV-504658-[catch2][dtest]-Reduce the lock scope of the kernel object look-up (#73)
[ROCm/hip-tests commit: e540c3b94a]
2025-05-05 14:24:11 +05:30
Gollamandala, Srinivasarao e540c3b94a SWDEV-504658-[catch2][dtest]-Reduce the lock scope of the kernel object look-up (#73) 2025-05-05 14:24:11 +05:30
Poag, Charis b5a43b7744 [SWDEV-528647/SWDEV-528450] Reduce API load times and libdrm/libdrm_amdgpu dynamic loading (#333)
Changes:
- Removed libdrm/libdrm_amdgpu dependencies
- Added/updated new internal libdrm/libdrm_amdgpu/xf86drm APIs
  to allow our APIs to reference before dynamic loading
  the libdrm/libdrm_amdgpu libraries:
  1. amdgpu_drm.h to what's seen in mainline
  2. Added xf86drm.h to whats seen in mainline
- Modified internal DRM capabilities:
  1. Require each API to independently connect to libdrm/libdrm_amdgpu
     + validate API handles reponses accordingly
  2. Initialization of AMD SMI no longer has as strong of a tie to
     libdrm
- Updated internal implementations of several APIs which have
connections to libdrm/libdrm_amdgpu or APIs which have conflicts
with open libdrm/libdrm_amdgpu connections:
  1. amdsmi_init()
  2. amdsmi_get_gpu_vram_usage()
  3. amdsmi_get_gpu_asic_info()
  4. amdsmi_get_gpu_vram_info()
  5. amdsmi_get_gpu_vbios_info()
  6. amdsmi_get_gpu_driver_info()
  7. amdsmi_get_gpu_virtualization_mode()
  8. amdsmi_set_gpu_memory_partition()
  9. amdsmi_set_gpu_memory_partition_mode()
- Cleaned up effected tests/APIs

Change-Id: I96e2cf1b06b0cfee1b01a5e991ccc6116c4245a8
2025-05-02 21:58:53 -05:00
Poag, Charis fb3d7b9fb9 [SWDEV-528647/SWDEV-528450] Reduce API load times and libdrm/libdrm_amdgpu dynamic loading (#333)
Changes:
- Removed libdrm/libdrm_amdgpu dependencies
- Added/updated new internal libdrm/libdrm_amdgpu/xf86drm APIs
  to allow our APIs to reference before dynamic loading
  the libdrm/libdrm_amdgpu libraries:
  1. amdgpu_drm.h to what's seen in mainline
  2. Added xf86drm.h to whats seen in mainline
- Modified internal DRM capabilities:
  1. Require each API to independently connect to libdrm/libdrm_amdgpu
     + validate API handles reponses accordingly
  2. Initialization of AMD SMI no longer has as strong of a tie to
     libdrm
- Updated internal implementations of several APIs which have
connections to libdrm/libdrm_amdgpu or APIs which have conflicts
with open libdrm/libdrm_amdgpu connections:
  1. amdsmi_init()
  2. amdsmi_get_gpu_vram_usage()
  3. amdsmi_get_gpu_asic_info()
  4. amdsmi_get_gpu_vram_info()
  5. amdsmi_get_gpu_vbios_info()
  6. amdsmi_get_gpu_driver_info()
  7. amdsmi_get_gpu_virtualization_mode()
  8. amdsmi_set_gpu_memory_partition()
  9. amdsmi_set_gpu_memory_partition_mode()
- Cleaned up effected tests/APIs

Change-Id: I96e2cf1b06b0cfee1b01a5e991ccc6116c4245a8

[ROCm/amdsmi commit: b5a43b7744]
2025-05-02 21:58:53 -05:00
Siu Chi Chan be0761502d rccl-UnitTests - link to dl library (#1673)
[ROCm/rccl commit: 9525c5b2ef]
2025-05-02 21:20:22 -05:00
Siu Chi Chan 9525c5b2ef rccl-UnitTests - link to dl library (#1673) 2025-05-02 21:20:22 -05:00
David Galiffi 80423ff010 Updating release build dockers (#195)
- Bringing in recent changes from rocm-6.4 branch (https://github.com/ROCm/rocprofiler-systems/pull/171)
- Add libdrm-devel to rhel and suse files
- Update ROCm installation method in Ubuntu file
- Add additional output to `test-release.sh` to catch failures due to a Python version not included
- Add Python 3.13 to Dockers

[ROCm/rocprofiler-systems commit: 83a9eb3d7c]
2025-05-02 19:29:15 -04:00
David Galiffi 83a9eb3d7c Updating release build dockers (#195)
- Bringing in recent changes from rocm-6.4 branch (https://github.com/ROCm/rocprofiler-systems/pull/171)
- Add libdrm-devel to rhel and suse files
- Update ROCm installation method in Ubuntu file
- Add additional output to `test-release.sh` to catch failures due to a Python version not included
- Add Python 3.13 to Dockers
2025-05-02 19:29:15 -04:00
Searles, Mark e480220c6a Fix typos in warning msgs (#231)
[ROCm/clr commit: cd9bc61559]
2025-05-02 14:31:42 -07:00
Searles, Mark cd9bc61559 Fix typos in warning msgs (#231) 2025-05-02 14:31:42 -07:00
David Galiffi 6fe19b681a Fix path to post-processing merge script (#187)
- Path to merge script not found unless user explicitly sources "share/rocprofiler-systems/setup-env.sh" to setup PATHs.
- Instead, let's derive the path when the application loads and use it when executing the helper script
- Rename script to rocprof-sys-merge-output.sh.
- Change install folder to <prefix>/libexec/rocprofiler-systems based on dev-ops feedback.
- Updated PATH variable in the modulefile and source scrtipt.
- For SWDEV-528101

[ROCm/rocprofiler-systems commit: adc66956b0]
2025-05-02 16:52:54 -04:00
David Galiffi adc66956b0 Fix path to post-processing merge script (#187)
- Path to merge script not found unless user explicitly sources "share/rocprofiler-systems/setup-env.sh" to setup PATHs.
- Instead, let's derive the path when the application loads and use it when executing the helper script
- Rename script to rocprof-sys-merge-output.sh.
- Change install folder to <prefix>/libexec/rocprofiler-systems based on dev-ops feedback.
- Updated PATH variable in the modulefile and source scrtipt.
- For SWDEV-528101
2025-05-02 16:52:54 -04:00
anujshuk-amd 31dc2414e3 Reverting PR-154 Changes since VCN data not seen on Perfetto file (#191)
[ROCm/rocprofiler-systems commit: ff109912c2]
2025-05-02 16:19:43 -04:00
anujshuk-amd ff109912c2 Reverting PR-154 Changes since VCN data not seen on Perfetto file (#191) 2025-05-02 16:19:43 -04:00
Welton, Benjamin 4b414f1ee5 Remove unnecessary log line (#387)
Co-authored-by: Benjamin Welton <bewelton@amd.com>

[ROCm/rocprofiler-sdk commit: 65f60bbb96]
2025-05-02 12:25:32 -07:00
Welton, Benjamin 65f60bbb96 Remove unnecessary log line (#387)
Co-authored-by: Benjamin Welton <bewelton@amd.com>
2025-05-02 12:25:32 -07:00
David Galiffi 490ff33d25 Conditionally include ROCPROFILER_BUFFER_TRACING_PAGE_MIGRATION (#193)
- Include only if ROCPROFILER_SDK_VERSION < 1.0.0, as it is being removed
- For SWDEV-530639

[ROCm/rocprofiler-systems commit: 0f16d45445]
2025-05-02 15:05:27 -04:00
David Galiffi 0f16d45445 Conditionally include ROCPROFILER_BUFFER_TRACING_PAGE_MIGRATION (#193)
- Include only if ROCPROFILER_SDK_VERSION < 1.0.0, as it is being removed
- For SWDEV-530639
2025-05-02 15:05:27 -04:00
Aurelien Bouteiller 394b52c0ed func-tests: Don't rely on asserts to catch invalid argv/env params (#96)
[ROCm/rocshmem commit: 2bbe21db56]
2025-05-02 12:00:35 -04:00
Aurelien Bouteiller 2bbe21db56 func-tests: Don't rely on asserts to catch invalid argv/env params (#96) 2025-05-02 12:00:35 -04:00
Aurelien Bouteiller 19e98852af cleanup leftovers from SOS testers removal (#97)
Followup to pr#85

[ROCm/rocshmem commit: f0501550f7]
2025-05-02 11:59:52 -04:00
Aurelien Bouteiller f0501550f7 cleanup leftovers from SOS testers removal (#97)
Followup to pr#85
2025-05-02 11:59:52 -04:00
Bertan Dogancay b435c75068 [Graph] Try using P2P by default (#1670)
[ROCm/rccl commit: acfac55516]
2025-05-02 11:54:30 -04:00
Bertan Dogancay acfac55516 [Graph] Try using P2P by default (#1670) 2025-05-02 11:54:30 -04:00
Roy, Rakesh c986600da6 SWDEV-530762 - Disable failing test in Linux PSDB (#162)
Change-Id: If4ab846d74932384fd847c69f76d932f31d4523c

[ROCm/hip-tests commit: 4794a250ba]
2025-05-02 18:31:24 +05:30
Roy, Rakesh 4794a250ba SWDEV-530762 - Disable failing test in Linux PSDB (#162)
Change-Id: If4ab846d74932384fd847c69f76d932f31d4523c
2025-05-02 18:31:24 +05:30
David Galiffi 091ae42b98 Install rocprof-sys-causal-print to /opt/rocm/libexec/rocprofiler-systems (#175)
For SWDEV-528225

Signed-off-by: David Galiffi <David.Galiffi@amd.com>

[ROCm/rocprofiler-systems commit: 7d74dc7d5d]
2025-05-01 19:49:11 -04:00
David Galiffi 7d74dc7d5d Install rocprof-sys-causal-print to /opt/rocm/libexec/rocprofiler-systems (#175)
For SWDEV-528225

Signed-off-by: David Galiffi <David.Galiffi@amd.com>
2025-05-01 19:49:11 -04:00
David Galiffi 2b61118a14 Re-enable RPM_PACKAGE_AUTOREQ (#174)
Removes workaround for SWDEV-525268

Signed-off-by: David Galiffi <David.Galiffi@amd.com>

[ROCm/rocprofiler-systems commit: e9e194a3ca]
2025-05-01 19:46:24 -04:00
David Galiffi e9e194a3ca Re-enable RPM_PACKAGE_AUTOREQ (#174)
Removes workaround for SWDEV-525268

Signed-off-by: David Galiffi <David.Galiffi@amd.com>
2025-05-01 19:46:24 -04:00
Sajina PK 8c424f2074 Fix to overlapping VCN and JPEG tracks in perfetto (#192)
- Fix overlapping VCN and JPEG activity values in Perfetto output.
- Modify the storage of the activity values to be more efficient.

[ROCm/rocprofiler-systems commit: 99a411fe52]
2025-05-01 19:40:49 -04:00
Sajina PK 99a411fe52 Fix to overlapping VCN and JPEG tracks in perfetto (#192)
- Fix overlapping VCN and JPEG activity values in Perfetto output.
- Modify the storage of the activity values to be more efficient.
2025-05-01 19:40:49 -04:00
Aryan Salmanpour d64d6326f8 VideoDecodePerf sample: use a display delay of 1 by default (#578)
[ROCm/rocdecode commit: cf85e6c46c]
2025-05-01 16:03:38 -04:00
Aryan Salmanpour cf85e6c46c VideoDecodePerf sample: use a display delay of 1 by default (#578) 2025-05-01 16:03:38 -04:00
jeffqjiangNew a8f665e2be * Roc decoder utility: Do not throw exceptions on picture decode or decoder reconfig errors. (#577)
- This allows the decoder to be able to handle more AV1 error streams.

[ROCm/rocdecode commit: d751b24ed5]
2025-05-01 12:50:28 -04:00
jeffqjiangNew d751b24ed5 * Roc decoder utility: Do not throw exceptions on picture decode or decoder reconfig errors. (#577)
- This allows the decoder to be able to handle more AV1 error streams.
2025-05-01 12:50:28 -04:00
Aurelien Bouteiller 27d1189ff3 Substitute pow2bin allocator with a dlmalloc based allocator (#71)
* Add dlmalloc_strat allocator strategy
 - Use mspace variant to ease encapsulation
 - Make pow2bins and dlmalloc cmake selectable
* Add unit tester for dlmalloc, rework single_heap, pow2bins unit testers
accordingly
 - add dlmalloc get_used/get_avail, and have all strats allocators also have a get_used
 - Rework memallocator unit tests: bin size is per strat, alignment is verified in singleheap
* bugfix: dlmalloc exposed that the pingpong test would write past end of
allocation with -w 32
* iostream leakage/mixed usage of cerr and fprintf(stderr

---------

Signed-off-by: Aurelien Bouteiller <aurelien.bouteiller@amd.com>

[ROCm/rocshmem commit: b835de6cd5]
2025-05-01 11:55:23 -04:00
Aurelien Bouteiller b835de6cd5 Substitute pow2bin allocator with a dlmalloc based allocator (#71)
* Add dlmalloc_strat allocator strategy
 - Use mspace variant to ease encapsulation
 - Make pow2bins and dlmalloc cmake selectable
* Add unit tester for dlmalloc, rework single_heap, pow2bins unit testers
accordingly
 - add dlmalloc get_used/get_avail, and have all strats allocators also have a get_used
 - Rework memallocator unit tests: bin size is per strat, alignment is verified in singleheap
* bugfix: dlmalloc exposed that the pingpong test would write past end of
allocation with -w 32
* iostream leakage/mixed usage of cerr and fprintf(stderr

---------

Signed-off-by: Aurelien Bouteiller <aurelien.bouteiller@amd.com>
2025-05-01 11:55:23 -04:00
Rawat, Swati 4a230f0180 fix broken link (#169)
Update job_stats_sample.rst

Co-authored-by: srawat <120587655+SwRaw@users.noreply.github.com>
2025-05-01 10:49:31 -05:00
Rawat, Swati f9ceb0e6b9 fix broken link (#169)
Update job_stats_sample.rst

Co-authored-by: srawat <120587655+SwRaw@users.noreply.github.com>

[ROCm/rdc commit: 4a230f0180]
2025-05-01 10:49:31 -05:00
Apurv Mishra aa0a32a166 kfdtest: Update ROCr homepage in CMakeLists.txt
Signed-off-by: Apurv Mishra <Apurv.Mishra@amd.com>
2025-05-01 11:22:49 -04:00
Apurv Mishra aa896090f8 kfdtest: Update ROCr homepage in CMakeLists.txt
Signed-off-by: Apurv Mishra <Apurv.Mishra@amd.com>


[ROCm/ROCR-Runtime commit: aa0a32a166]
2025-05-01 11:22:49 -04:00
Chaudhary, Jatin Jaikishan f619372ae6 SWDEV-514560 - add fp6 header implementation (#54)
Co-authored-by: rahul manocha <rmanocha_amdeng>

[ROCm/clr commit: 12febe6782]
2025-05-01 15:17:38 +01:00
Chaudhary, Jatin Jaikishan 12febe6782 SWDEV-514560 - add fp6 header implementation (#54)
Co-authored-by: rahul manocha <rmanocha_amdeng>
2025-05-01 15:17:38 +01:00
Kandula, Venkateshwar reddy d9bd0903f8 [PSDB] [CI] SWDEV-528922: Modify summary MEMORY_ALLOCATION for summary.txt (#375)
* Update validate.py

* increase memalloc to 4 to account new vmem.

* Format.

* modify test to work when vmem using and not to work.

* address comments.

---------

Co-authored-by: Venkateshwar Reddy Kandula <vkandula@amd.com>

[ROCm/rocprofiler-sdk commit: b69e12032e]
2025-05-01 00:58:00 -05:00
Kandula, Venkateshwar reddy b69e12032e [PSDB] [CI] SWDEV-528922: Modify summary MEMORY_ALLOCATION for summary.txt (#375)
* Update validate.py

* increase memalloc to 4 to account new vmem.

* Format.

* modify test to work when vmem using and not to work.

* address comments.

---------

Co-authored-by: Venkateshwar Reddy Kandula <vkandula@amd.com>
2025-05-01 00:58:00 -05:00
Madsen, Jonathan 11fdedb874 [Build System] Install FindrocDecode.cmake and FindrocJPEG.cmake (#382)
Install FindrocDecode.cmake and FindrocJPEG.cmake

Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>

[ROCm/rocprofiler-sdk commit: e9757e46ce]
2025-05-01 00:57:37 -05:00
Madsen, Jonathan e9757e46ce [Build System] Install FindrocDecode.cmake and FindrocJPEG.cmake (#382)
Install FindrocDecode.cmake and FindrocJPEG.cmake

Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
2025-05-01 00:57:37 -05:00
Trowbridge, Ian 24f054f509 Fix HIP Streams Duplication Error (#313)
* Fix stream duplication and fixed tests

* Added comments to explain stream.cpp code, change stream nullptr check to occur in update table to prevent readding null stream, simplified hip-streams bin file code, add destroyStreams to hip-streams bin file code

* Removed roctx from CMakeLists.txt

* Updated documentation

* Fix documentation

* Removed update_table for HIP compiler table and updated stream.cpp to remove support for HIP compiler table

* Added runtime initialization check for HIP

* Changed tool name, working on fixing memory management

* Added context for counter collection kernel rename combination

* Changed name from map to set and changed description

* Fix documentation description for group-by-queue

* Merged memory copy and kernel operations onto a single track when on the same stream

* Updated perfetto output to remove hardware information from track name to merge all memory copy and kernel operations on the same stream to the same track:

* Most pr comments addressed

* Added filter for counter collection and removed kernel buffer tracing hack

* Added PR comment fixes

---------

Co-authored-by: Madsen, Jonathan <Jonathan.Madsen@amd.com>

[ROCm/rocprofiler-sdk commit: e626df43eb]
2025-05-01 00:56:15 -05:00
Trowbridge, Ian e626df43eb Fix HIP Streams Duplication Error (#313)
* Fix stream duplication and fixed tests

* Added comments to explain stream.cpp code, change stream nullptr check to occur in update table to prevent readding null stream, simplified hip-streams bin file code, add destroyStreams to hip-streams bin file code

* Removed roctx from CMakeLists.txt

* Updated documentation

* Fix documentation

* Removed update_table for HIP compiler table and updated stream.cpp to remove support for HIP compiler table

* Added runtime initialization check for HIP

* Changed tool name, working on fixing memory management

* Added context for counter collection kernel rename combination

* Changed name from map to set and changed description

* Fix documentation description for group-by-queue

* Merged memory copy and kernel operations onto a single track when on the same stream

* Updated perfetto output to remove hardware information from track name to merge all memory copy and kernel operations on the same stream to the same track:

* Most pr comments addressed

* Added filter for counter collection and removed kernel buffer tracing hack

* Added PR comment fixes

---------

Co-authored-by: Madsen, Jonathan <Jonathan.Madsen@amd.com>
2025-05-01 00:56:15 -05:00