Commit Graph

74713 Commits

Author SHA1 Message Date
Arm Patinyasakdikul 9806f5e9dd Fix git version fetching logic. (#1981) 2025-10-17 09:17:49 -05:00
Narlo, Joseph 460cfcba1f [SWDEV-555807] TestCudaMallocAsync test power draw failing (#755)
* Clarified comments regarding power limit retrieval and its support on virtualized systems.
* Change unsupported comment to UINT32_MAX

---------

Signed-off-by: josnarlo <Joseph.Narlo@amd.com>
Signed-off-by: Arif, Maisam <Maisam.Arif@amd.com>
2025-10-17 08:57:57 -05:00
Narlo, Joseph 5ec7b213e4 [SWDEV-555807] TestCudaMallocAsync test power draw failing (#755)
* Clarified comments regarding power limit retrieval and its support on virtualized systems.
* Change unsupported comment to UINT32_MAX

---------

Signed-off-by: josnarlo <Joseph.Narlo@amd.com>
Signed-off-by: Arif, Maisam <Maisam.Arif@amd.com>

[ROCm/amdsmi commit: 460cfcba1f]
2025-10-17 08:57:57 -05:00
Sunday Clement b9b8b6110b rocrtst: Add SVM Prefetch test (#360)
this test will prefetch SVM memory, and then verify the memory is sourced
from the expected numa node.

Signed-off-by: Sunday Clement <Sunday.Clement@amd.com>
2025-10-17 09:43:46 -04:00
Milan Radosavljevic 00faa48ac2 Add flushing of perfetto buffer (#1417)
- Add flushing of perfetto buffer
- Add `ROCPROFSYS_PERFETTO_FLUSH_PERIOD_MS` config setting.
- Update CHANGELOG.sh
- Resolves SWDEV-518817

---------

Co-authored-by: David Galiffi <David.Galiffi@amd.com>
2025-10-17 09:30:29 -04:00
Edgar Gabriel feab645795 update to build system: (#277)
- make adding PMIx library to compile time based on the result of
  finding PMIx support. This is required eg if compiling rocSHMEM with ompi
  4.0/4.1, which do not have a built-in PMIx version.
- when setting USE_EXTERNAL_MPI=OFF  which ensures that we do not
  check for external MPI libraries (even if one would be available).

[ROCm/rocshmem commit: ed957302d4]
2025-10-17 07:42:11 -05:00
Edgar Gabriel ed957302d4 update to build system: (#277)
- make adding PMIx library to compile time based on the result of
  finding PMIx support. This is required eg if compiling rocSHMEM with ompi
  4.0/4.1, which do not have a built-in PMIx version.
- when setting USE_EXTERNAL_MPI=OFF  which ensures that we do not
  check for external MPI libraries (even if one would be available).
2025-10-17 07:42:11 -05:00
solaiys d348797017 Update rocm-core package to include rdhc script (#1375)
* Add rdhc script in to rocm-core package
* Create the rdhc symlink within the package itself.
* Removed hard-coding of rocm-core name, used CORE_TARGET instead.
* [RDHC] Check if the required pip pkgs are present and warn .
   rdhc checks the required pip packages are present or not.
   if not warns the user and exits gracefully.

Signed-off-by: Saravanan Solaiyappan <saravanan.solaiyappan@amd.com>
2025-10-17 10:34:00 +05:30
Ashutosh Mishra 221807b69d Fixing tool erros (#1416)
Tools running for sanity checks are
detecting buffer overrun which is
not the case. Still getting rid of
function which is causing the issue
removed and making the code more robust
 using defensive programming so that
any tool is not able to detect issues
hereafter. Fixed comments, corrected
typos and added a new return type
which is required as per refactoring

Signed-off-by: Ashutosh Mishra <ashutosh.mishra@amd.com>
2025-10-17 09:10:18 +05:30
Venkateshwar Reddy Kandula ba5b8aa16f [rocprofiler] Disable runFeatureTests and runTracerFeatureTests test to unblock psdb (#1404)
* disable feature profiler tests temporarily

* Update tracer_gtest.cpp

* Update profiler_gtest.cpp
2025-10-16 14:55:36 -07:00
Rahul Vaidya 307f787244 [Profiler plugin] Fix segfault issue with profiler plugin (#1973)
* Fix profiler plugin segfault by correctly setting p2p->func

* Look for librccl-profiler.so instead of libnccl-profiler.so

Signed-off-by: rahulvaidya20 <ravaidya@amd.com>

---------

Signed-off-by: rahulvaidya20 <ravaidya@amd.com>
Co-authored-by: Yongjie Qiu <Yongjie.Qiu@amd.com>

[ROCm/rccl commit: 624f68b2b2]
2025-10-16 16:33:18 -05:00
Rahul Vaidya 624f68b2b2 [Profiler plugin] Fix segfault issue with profiler plugin (#1973)
* Fix profiler plugin segfault by correctly setting p2p->func

* Look for librccl-profiler.so instead of libnccl-profiler.so

Signed-off-by: rahulvaidya20 <ravaidya@amd.com>

---------

Signed-off-by: rahulvaidya20 <ravaidya@amd.com>
Co-authored-by: Yongjie Qiu <Yongjie.Qiu@amd.com>
2025-10-16 16:33:18 -05:00
Bindhiya Kanangot Balakrishnan 97b6e806da SWDEV-560768 - SMI test return if no devices available (#1369)
Return from Setup if no monitor devices are available.

Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
2025-10-16 15:35:18 -05:00
Bindhiya Kanangot Balakrishnan b4288fd8d4 SWDEV-554099 - Update rsmi tests expected output (#1364)
Updated rsmitsts expected outputs to accomodate
returned status.

Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
2025-10-16 15:34:07 -05:00
Bindhiya Kanangot Balakrishnan 4dd1c1042a Move build_xcp_dict def to helpers.
Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>
2025-10-16 15:25:10 -05:00
Bindhiya Kanangot Balakrishnan baa8fa3042 Move build_xcp_dict def to helpers.
Signed-off-by: Bindhiya Kanangot Balakrishnan <Bindhiya.KanangotBalakrishnan@amd.com>


[ROCm/amdsmi commit: 4dd1c1042a]
2025-10-16 15:25:10 -05:00
Aurelien Bouteiller bb8406b013 Runtime selection of IONIC (#272)
* Split ionic code to a subdirectory; dyld libionicl; move the fntable to provider_gda_xxx.hpp
pass the pattr to ionic_setup_pd, include endian.hpp
Enable building IONIC conduit for runtime selection

* Uniform style for the fntable between ionic and the rest

* Move mlx5 gda conduit to a subdir; resolve conflict with backend_can_run
function

* Don't forget to init qp for ionic, move mlx5 specialized init qp code to
the mlx5 subdir

* Don't add cmakecaches...

Typo: GDA_BXNT

* Add gda-ionic to all_backends build scripts

* Apply suggestion from reviews

Co-authored-by: Omri Mor <omri50@gmail.com>
Co-authored-by: Edgar Gabriel <edgar.gabriel@amd.com>

* Remove duplicate definitiion of DLSYM macros

---------

Co-authored-by: Omri Mor <omri50@gmail.com>
Co-authored-by: Edgar Gabriel <edgar.gabriel@amd.com>

[ROCm/rocshmem commit: 3cfe76522e]
2025-10-16 15:53:01 -04:00
Aurelien Bouteiller 3cfe76522e Runtime selection of IONIC (#272)
* Split ionic code to a subdirectory; dyld libionicl; move the fntable to provider_gda_xxx.hpp
pass the pattr to ionic_setup_pd, include endian.hpp
Enable building IONIC conduit for runtime selection

* Uniform style for the fntable between ionic and the rest

* Move mlx5 gda conduit to a subdir; resolve conflict with backend_can_run
function

* Don't forget to init qp for ionic, move mlx5 specialized init qp code to
the mlx5 subdir

* Don't add cmakecaches...

Typo: GDA_BXNT

* Add gda-ionic to all_backends build scripts

* Apply suggestion from reviews

Co-authored-by: Omri Mor <omri50@gmail.com>
Co-authored-by: Edgar Gabriel <edgar.gabriel@amd.com>

* Remove duplicate definitiion of DLSYM macros

---------

Co-authored-by: Omri Mor <omri50@gmail.com>
Co-authored-by: Edgar Gabriel <edgar.gabriel@amd.com>
2025-10-16 15:53:01 -04:00
Pryor, Adam ee1445e2cc [SWDEV-558993] Fix list() groups printout (#772)
* Updated groups printing
* added parameters to check_required_groups
	* two device groups since kfd and render require the same group

---------

Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Co-authored-by: Maisam Arif <Maisam.Arif@amd.com>
2025-10-16 11:23:49 -05:00
Pryor, Adam 1c6147ead5 [SWDEV-558993] Fix list() groups printout (#772)
* Updated groups printing
* added parameters to check_required_groups
	* two device groups since kfd and render require the same group

---------

Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Co-authored-by: Maisam Arif <Maisam.Arif@amd.com>

[ROCm/amdsmi commit: ee1445e2cc]
2025-10-16 11:23:49 -05:00
Sunday Clement c23c320b4d rocr: Make IPC Handles Unique (#795)
Query IPC handles on shared memory export/import for any metadata as a
means to uniquely identify handles that happen to be backed by buffers
that point to the same memory.
2025-10-16 14:37:02 +05:30
marantic-amd f2ccc96cfd Add missing counter events handling for ROCPD (#1305)
* Add missing counter events handling for ROCPD

* Update projects/rocprofiler-systems/source/lib/rocprof-sys/library/rocprofiler-sdk/counters.cpp

* Update projects/rocprofiler-systems/source/lib/rocprof-sys/library/rocprofiler-sdk/counters.cpp

* Fixed formatting

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

---------

Signed-off-by: Marjan Antic <Marjan.Antic@amd.com>
Co-authored-by: David Galiffi <David.Galiffi@amd.com>
2025-10-15 23:15:19 -04:00
David Galiffi 38a81ac4e3 Update VERSION to 1.3.0 (#1368) 2025-10-15 23:12:10 -04:00
David Galiffi b75423b173 Update installation and ROCPD documentation (#1300)
* Updating install doc page

* Removing the Quick Start page

* Add documentation for rocpd output

* Update links to reference rocm-systems repo

* Update README.md

Installation instructions references ROCm Docs link.

* Updated git clone instructions

Back to using https to clone the repository

* Fix formatting

* Update projects/rocprofiler-systems/docs/how-to/understanding-rocprof-sys-output.rst

* Add reference to "rocpd" section to the "Profiling Python" section

* Update CONTRIBUTING.md

* For ROCPD, document minimum version of SDK.

* Update CHANGELOGS

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

* Update CHANGELOG.md

Updated based on feedback from docs team

* Update CONTRIBUTING.md

* Update CONTRIBUTING.md.

Simplify and remove setup information overlapping with the "rocm-systems" contributing documentation.

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* Update CHANGELOG.md

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

* Apply suggestion from @prbasyal-amd

Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

---------

Signed-off-by: David Galiffi <David.Galiffi@amd.com>
Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>
2025-10-15 23:11:46 -04:00
Dimple Prajapati 6c4325d131 Add host API for enqueuing barrier on given stream (#274)
* add host API for enqueuing barrier on given stream

[ROCm/rocshmem commit: a44b581997]
2025-10-15 14:29:07 -07:00
Dimple Prajapati a44b581997 Add host API for enqueuing barrier on given stream (#274)
* add host API for enqueuing barrier on given stream
2025-10-15 14:29:07 -07:00
Young Hui - AMD 02bf0a8492 [rocprofiler-compute] Source files updated to reference super-repo URL (#1330)
* source files updated to reference super-repo URL
2025-10-15 15:35:11 -04:00
Young Hui - AMD 161e44c425 [rocprof-compute] Documentation changes for move to super-repo for 7.1 (#1329)
- also remove json output mention in docs
2025-10-15 15:32:54 -04:00
vedithal-amd 454e935448 Fix docker compose (#1323)
Co-authored-by: Yanyao Wang <yanywang@amd.com>
2025-10-15 14:26:56 -05:00
vedithal-amd ecf0d32644 Update CHANGELOG.md for ROCm 7.1.0 release (#1362) 2025-10-15 14:25:34 -05:00
Alysa Liu 4342579645 libhsakmt: Fix memory leak for events_page metadata (#807) 2025-10-15 14:52:40 -04:00
Alysa Liu d5cbdc104d rocrtst: Add Memory_Async_Copy_On_Engine Test (#885)
Increase test coverage involving:
hsa_amd_memory_get_preferred_copy_engine()
hsa_amd_memory_copy_engine_status()
hsa_amd_memory_async_copy_on_engine()
2025-10-15 14:51:54 -04:00
alex-breslow-amd a5256e6219 MSCCL: Unland PR1788 + Fix for MSCCL Data Corruption (#1960)
- Earlier fix PR1788 is no longer necessary after ROCr fix and pre-ROCr fix workaround
- Inserts an s_waitcnt vmcnt(0), which fixes a data corruption issue in MSCCL

[ROCm/rccl commit: 154350baaf]
2025-10-15 10:32:25 -07:00
alex-breslow-amd 154350baaf MSCCL: Unland PR1788 + Fix for MSCCL Data Corruption (#1960)
- Earlier fix PR1788 is no longer necessary after ROCr fix and pre-ROCr fix workaround
- Inserts an s_waitcnt vmcnt(0), which fixes a data corruption issue in MSCCL
2025-10-15 10:32:25 -07:00
Saurabh Verma 31a7f3d5dd Update gfx9_primitives.h and gfx9_block_table.h to use gc_9_4_2_offset.h (#859)
* Initial commit

* Replaced gc_9_2_1_sh_mask.h with gc_9_4_2_sh_mask.h

* properly replace gc_9_2_1_sh_mask.h and gc_9_2_1_offset.h for all gfx9 asics
2025-10-15 12:13:35 -05:00
gilbertlee-amd bb85692891 Enabling gdrcopy option for gfx950 (#1955)
[ROCm/rccl commit: fedddb452c]
2025-10-15 10:55:25 -06:00
gilbertlee-amd fedddb452c Enabling gdrcopy option for gfx950 (#1955) 2025-10-15 10:55:25 -06:00
Venkateshwar Reddy Kandula 9404178ea5 [rocprofiler-sdk][CI] rhel sles workflow fix (#1373)
* bug fix.

* add backslash

* add export for path, bug
2025-10-15 11:48:59 -05:00
alex-breslow-amd 455d516dc4 [gfx950] Make bypassing __threadfence the default for multinode. (#1947)
* Gate based on ROCM version, safe for ROCm 7.0.2 and beyond.
* Updates naming to gfx9CheapFenceOff since we use this for gfx942 and gfx950.  Thanks Nilesh.
* Add info logging statement to NCCL_INIT to print whether enabled when INFO logging is enabled.

[ROCm/rccl commit: c70f5b4621]
2025-10-15 09:15:36 -07:00
alex-breslow-amd c70f5b4621 [gfx950] Make bypassing __threadfence the default for multinode. (#1947)
* Gate based on ROCM version, safe for ROCm 7.0.2 and beyond.
* Updates naming to gfx9CheapFenceOff since we use this for gfx942 and gfx950.  Thanks Nilesh.
* Add info logging statement to NCCL_INIT to print whether enabled when INFO logging is enabled.
2025-10-15 09:15:36 -07:00
adapryor a64e9b4ac4 [SWDEV-560778] Update gpu metrics factory to return a new pointer every time 2025-10-15 11:00:44 -05:00
adapryor cda730140f [SWDEV-560778] Update gpu metrics factory to return a new pointer every time
[ROCm/amdsmi commit: a64e9b4ac4]
2025-10-15 11:00:44 -05:00
Mythreya Kuricheti ac8adbacff [CI][rocprofiler-sdk] Fix codeql jobs (#1366) 2025-10-15 10:34:29 -05:00
Saurabh Verma 946385d0ff Reverts #1379 and properly migrates the docs (#1381)
Reverts #1379 and properly migrates the docs

---------

Co-authored-by: Matt Williams <matt.williams@amd.com>
2025-10-15 10:48:27 -04:00
Saurabh Verma b6a187aed1 migrate aqlprofile docs 7.0.1 from standalone repo (#1379)
This PR migrates the aqlprofile/docs folder from standalone repo to monorepo
Link to the docs branch:
https://github.com/ROCm/aqlprofile/commits/docs/7.0.1

---------

Co-authored-by: Matt Williams <matt.williams@amd.com>
Co-authored-by: pbhandar-amd <138039281+pbhandar-amd@users.noreply.github.com>
2025-10-15 10:01:36 -04:00
Gerardo Hernandez fc5551a724 SWDEV-536360 - fix another bullet point in reduce sync operations section not being displayed on its own line (#1374) 2025-10-15 14:51:43 +01:00
Danylo Lytovchenko 59a30bb117 Add ignore revs file (#1126)
* Add ignore revs file

* Fix rev file name
2025-10-15 13:57:56 +02:00
ajanicijamd 259ef6348b Fixed issues with nic-performance test (#1168)
- On some hosts the wget can finish too soon and PAPI doesn't catch even a single network event.
- On some hosts, there are multiple default NICs and the scripts didn't work in that case.
- The test script was writing the output of wget to /tmp directory, which causes a problem if another user tries to run the same test. Because the output file with the same name already exists in the same directory, but with a different owner, the test fails

---------

Co-authored-by: David Galiffi <David.Galiffi@amd.com>
2025-10-14 23:45:08 -04:00
Gerardo Hernandez bfbc48bb0e SWDEV-536360 - fix bullet points in reduce sync operations section not being displayed on different lines in the browser (#1346) 2025-10-14 22:02:34 +01:00
axie_amdeng dde482d224 rocr: unitialized size variable caused huge memory/space allocation (#1232)
Signed-off-by: Alex Xie <AlexBin.Xie@amd.com>
2025-10-14 16:57:10 -04:00