76333 Υποβολές

Συγγραφέας SHA1 Μήνυμα Ημερομηνία
Chaudhary, Jatin Jaikishan 07c7d3e860 SWDEV-520627 - include sync_warp header instead of warp function header (#18)
Change-Id: Ic3f54b0f5bfee8565a8bbb6218fb0ccdb900c9ea

[ROCm/clr commit: 5c030840d6]
2025-04-10 21:50:25 +05:30
Chaudhary, Jatin Jaikishan 5c030840d6 SWDEV-520627 - include sync_warp header instead of warp function header (#18)
Change-Id: Ic3f54b0f5bfee8565a8bbb6218fb0ccdb900c9ea
2025-04-10 21:50:25 +05:30
Dingming Wu 63c6180130 Adding #include <dlfcn.h> in profiler.cc to pass build (#1632)
w/o the header, experience following errors:
```
error: use of undeclared identifier 'RTLD_NOW'
error: use of undeclared identifier 'RTLD_LOCAL'
error: use of undeclared identifier 'dlerror'
error: use of undeclared identifier 'dlsym'
error: use of undeclared identifier 'dlclose'
```

[ROCm/rccl commit: 1786c0268b]
2025-04-10 08:48:18 -07:00
Dingming Wu 1786c0268b Adding #include <dlfcn.h> in profiler.cc to pass build (#1632)
w/o the header, experience following errors:
```
error: use of undeclared identifier 'RTLD_NOW'
error: use of undeclared identifier 'RTLD_LOCAL'
error: use of undeclared identifier 'dlerror'
error: use of undeclared identifier 'dlsym'
error: use of undeclared identifier 'dlclose'
```
2025-04-10 08:48:18 -07:00
Arm Patinyasakdikul f29d59aa00 Add device synchronization before destroying proxy thread. (#1631)
This commit ensures that GPU finishes all kernel before destroying
communicator thread.

[ROCm/rccl commit: 52654e2301]
2025-04-10 10:44:16 -05:00
Arm Patinyasakdikul 52654e2301 Add device synchronization before destroying proxy thread. (#1631)
This commit ensures that GPU finishes all kernel before destroying
communicator thread.
2025-04-10 10:44:16 -05:00
Pedram Alizadeh 93ac2ea61e all_reduce LL/LL128 and Ring/Tree multi-node tuning for MI300 (#1627)
* Enabling LL128 by default on MI300

* Add missing CUDACHECK

* Adjust BW correction factors to fix the Tree->Ring switching point

* Refactor and add ll128 AR logarithmic factor to tuning models

* Move RCCL tuning changes to a separate file 

* Use enum for tunable indexing

* Use explicit indexing in tuning models to avoid mismatch issues

* Place rcclGetSizePerRank in a function

* Remove HIP ifdef for rccl-only call

---------

Co-authored-by: Mustafa Abduljabbar <mustafa.abduljabbar@amd.com>

[ROCm/rccl commit: e40ff4f84a]
2025-04-10 11:43:54 -04:00
Pedram Alizadeh e40ff4f84a all_reduce LL/LL128 and Ring/Tree multi-node tuning for MI300 (#1627)
* Enabling LL128 by default on MI300

* Add missing CUDACHECK

* Adjust BW correction factors to fix the Tree->Ring switching point

* Refactor and add ll128 AR logarithmic factor to tuning models

* Move RCCL tuning changes to a separate file 

* Use enum for tunable indexing

* Use explicit indexing in tuning models to avoid mismatch issues

* Place rcclGetSizePerRank in a function

* Remove HIP ifdef for rccl-only call

---------

Co-authored-by: Mustafa Abduljabbar <mustafa.abduljabbar@amd.com>
2025-04-10 11:43:54 -04:00
Andryeyev, German 90e3d2619a SWDEV-525725 - Enable resource cache for SVM (#156)
- Make sure reserved_va_ updated before svmPtr overwrite

[ROCm/clr commit: 94cd9bc4f7]
2025-04-10 10:54:28 -04:00
Andryeyev, German 94cd9bc4f7 SWDEV-525725 - Enable resource cache for SVM (#156)
- Make sure reserved_va_ updated before svmPtr overwrite
2025-04-10 10:54:28 -04:00
Pedram Alizadeh b225281747 single-node AR msccl algorithm tuning for MI300 (#1629)
[ROCm/rccl commit: 5b36b68d06]
2025-04-10 10:42:28 -04:00
Pedram Alizadeh 5b36b68d06 single-node AR msccl algorithm tuning for MI300 (#1629) 2025-04-10 10:42:28 -04:00
Edgar Gabriel 5b22ddd1ff add new flag to build instructions (#78)
This flag is required to link a pytorch use-case correctly.
It doesn't seem to impact the rocSHMEM code.

[ROCm/rocshmem commit: 5e49567b6c]
2025-04-10 08:39:54 -05:00
Edgar Gabriel 5e49567b6c add new flag to build instructions (#78)
This flag is required to link a pytorch use-case correctly.
It doesn't seem to impact the rocSHMEM code.
2025-04-10 08:39:54 -05:00
Patel, Jaydeepkumar c2ed585737 SWDEV-521262 - Adding MSVC compiler options to fix the conflict with SC module while building hip in debug. (#24)
[ROCm/clr commit: 997519fa94]
2025-04-10 15:25:58 +05:30
Patel, Jaydeepkumar 997519fa94 SWDEV-521262 - Adding MSVC compiler options to fix the conflict with SC module while building hip in debug. (#24) 2025-04-10 15:25:58 +05:30
Patel, Jaydeepkumar dea880f9da SWDEV-508973 - If total # of threads/block is more than HW capacity, it's invalid config issue and should return invalid config error. (#25)
[ROCm/clr commit: 8531cd3bbe]
2025-04-10 15:16:16 +05:30
Patel, Jaydeepkumar 8531cd3bbe SWDEV-508973 - If total # of threads/block is more than HW capacity, it's invalid config issue and should return invalid config error. (#25) 2025-04-10 15:16:16 +05:30
Stojiljkovic, Vladana 81a566e397 SWDEV-505795 - Return the same ptr from hipIpcOpenMemHandle if it is called multiple times (#93)
* SWDEV-505795 - Return the same ptr from hipIpcOpenMemHandle if it is called multiple times

* Move initialization outside of if statement

[ROCm/clr commit: e91cb4f320]
2025-04-10 11:20:36 +02:00
Stojiljkovic, Vladana e91cb4f320 SWDEV-505795 - Return the same ptr from hipIpcOpenMemHandle if it is called multiple times (#93)
* SWDEV-505795 - Return the same ptr from hipIpcOpenMemHandle if it is called multiple times

* Move initialization outside of if statement
2025-04-10 11:20:36 +02:00
Galantsev, Dmitrii 91be467cad Profiler - Fix eval fields
The 'value' pointer was being written to a lot and then used for reading
within the same function. This likely caused issues all over RDC when
reading the metrics.

This commit changes it so *value is written to only once.

Change-Id: I83c158c1e46c6ce46ff87d8a2e769f26ffa8c0da
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
2025-04-09 20:06:21 -05:00
Galantsev, Dmitrii d87fe5bada Profiler - Fix eval fields
The 'value' pointer was being written to a lot and then used for reading
within the same function. This likely caused issues all over RDC when
reading the metrics.

This commit changes it so *value is written to only once.

Change-Id: I83c158c1e46c6ce46ff87d8a2e769f26ffa8c0da
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>


[ROCm/rdc commit: 91be467cad]
2025-04-09 20:06:21 -05:00
jeffqjiangNew a0dd3fc6b3 AVC: Added support for in stream DPB buffer size change. (#553)
* * AVC: Added support for in stream DPB buffer size change.

* * AVC: Updated change log.

---------

Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com>

[ROCm/rocdecode commit: 3d1a1b638e]
2025-04-09 20:57:52 -04:00
jeffqjiangNew 3d1a1b638e AVC: Added support for in stream DPB buffer size change. (#553)
* * AVC: Added support for in stream DPB buffer size change.

* * AVC: Updated change log.

---------

Co-authored-by: Kiriti Gowda <kiritigowda@gmail.com>
2025-04-09 20:57:52 -04:00
dependabot[bot] 464d69963d Bump rocm-docs-core from 1.18.1 to 1.18.2 in /docs/sphinx (#1625)
Bumps [rocm-docs-core](https://github.com/ROCm/rocm-docs-core) from 1.18.1 to 1.18.2.
- [Release notes](https://github.com/ROCm/rocm-docs-core/releases)
- [Changelog](https://github.com/ROCm/rocm-docs-core/blob/develop/CHANGELOG.md)
- [Commits](https://github.com/ROCm/rocm-docs-core/compare/v1.18.1...v1.18.2)

---
updated-dependencies:
- dependency-name: rocm-docs-core
  dependency-version: 1.18.2
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>

[ROCm/rccl commit: b6d97a6176]
2025-04-09 17:00:31 -06:00
dependabot[bot] b6d97a6176 Bump rocm-docs-core from 1.18.1 to 1.18.2 in /docs/sphinx (#1625)
Bumps [rocm-docs-core](https://github.com/ROCm/rocm-docs-core) from 1.18.1 to 1.18.2.
- [Release notes](https://github.com/ROCm/rocm-docs-core/releases)
- [Changelog](https://github.com/ROCm/rocm-docs-core/blob/develop/CHANGELOG.md)
- [Commits](https://github.com/ROCm/rocm-docs-core/compare/v1.18.1...v1.18.2)

---
updated-dependencies:
- dependency-name: rocm-docs-core
  dependency-version: 1.18.2
  dependency-type: direct:production
  update-type: version-update:semver-patch
...

Signed-off-by: dependabot[bot] <support@github.com>
Co-authored-by: dependabot[bot] <49699333+dependabot[bot]@users.noreply.github.com>
2025-04-09 17:00:31 -06:00
GunaShekar, Ajay fc1d9a1c10 SWDEV-523281 - hipLaunchKernelExC nvidia impl (#5)
[ROCm/hipother commit: 27b12ec7cc]
2025-04-09 15:39:39 -07:00
GunaShekar, Ajay 27b12ec7cc SWDEV-523281 - hipLaunchKernelExC nvidia impl (#5) 2025-04-09 15:39:39 -07:00
Aryan Salmanpour 27a331f879 CI Update - Enhance testing for rocJPEG by using JPEG images with different chroma subsampling options (#138)
* CI Update - Enhance testing for rocJPEG by using JPEG images with different chroma subsampling options

* Add more tests

[ROCm/rocjpeg commit: 6e548b125b]
2025-04-09 15:13:44 -07:00
Aryan Salmanpour 6e548b125b CI Update - Enhance testing for rocJPEG by using JPEG images with different chroma subsampling options (#138)
* CI Update - Enhance testing for rocJPEG by using JPEG images with different chroma subsampling options

* Add more tests
2025-04-09 15:13:44 -07:00
ajanicijamd afec56b7ff Added ctest for NIC performance profiling (#143)
Added two new tests: nic-performance and validate-nic-performance-perfetto
- For ticket: SWDEV-516739

[ROCm/rocprofiler-systems commit: 11fe6a8ebb]
2025-04-09 17:38:46 -04:00
ajanicijamd 11fe6a8ebb Added ctest for NIC performance profiling (#143)
Added two new tests: nic-performance and validate-nic-performance-perfetto
- For ticket: SWDEV-516739
2025-04-09 17:38:46 -04:00
David Galiffi 07d2e46bf3 Disable AUTOREQ and manually add requirements. (#160)
- The amd-smi-lib package is not providing libamd_smi.so in its specs. We need to add a package dependency instead.
  - For ticket: SWDEV-525268
- Add support for `ROCM_DEP_ROCMCORE` in CMake

[ROCm/rocprofiler-systems commit: eaad295cfc]
2025-04-09 17:15:24 -04:00
David Galiffi eaad295cfc Disable AUTOREQ and manually add requirements. (#160)
- The amd-smi-lib package is not providing libamd_smi.so in its specs. We need to add a package dependency instead.
  - For ticket: SWDEV-525268
- Add support for `ROCM_DEP_ROCMCORE` in CMake
2025-04-09 17:15:24 -04:00
Sajina PK 04fb7e4fe7 RocJpeg cmake and document fixes (#157)
- Fix for rocjpeg sample cmake due to changes in the rocJPEG project
- Fix for rocprofiler-sdk version check - change the format
- Edits to docs for jpeg and vcn activity support - mention that these values may not be supported on all ASICs.

[ROCm/rocprofiler-systems commit: fad3a0d341]
2025-04-09 16:20:02 -04:00
Sajina PK fad3a0d341 RocJpeg cmake and document fixes (#157)
- Fix for rocjpeg sample cmake due to changes in the rocJPEG project
- Fix for rocprofiler-sdk version check - change the format
- Edits to docs for jpeg and vcn activity support - mention that these values may not be supported on all ASICs.
2025-04-09 16:20:02 -04:00
ajanicijamd 01c1cbe57f "How-To" document describing network performance profiling (#145)
---------

Co-authored-by: David Galiffi <David.Galiffi@amd.com>
Co-authored-by: Peter Park <peter.park@amd.com>
Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>

[ROCm/rocprofiler-systems commit: de84a277f2]
2025-04-09 16:11:52 -04:00
ajanicijamd de84a277f2 "How-To" document describing network performance profiling (#145)
---------

Co-authored-by: David Galiffi <David.Galiffi@amd.com>
Co-authored-by: Peter Park <peter.park@amd.com>
Co-authored-by: Pratik Basyal <pratik.basyal@amd.com>
2025-04-09 16:11:52 -04:00
Kiriti Gowda 74da3a7457 Codecov - .codecov.yaml (#554)
* Create .codecov.yaml

* Update .codecov.yaml

* Update .codecov.yaml

[ROCm/rocdecode commit: c39e320756]
2025-04-09 14:43:10 -04:00
Kiriti Gowda c39e320756 Codecov - .codecov.yaml (#554)
* Create .codecov.yaml

* Update .codecov.yaml

* Update .codecov.yaml
2025-04-09 14:43:10 -04:00
Aryan Salmanpour 20e919c6cf Code Clean up - remove unused API (#552)
* Code Clean up - remove unused API

* clean up

[ROCm/rocdecode commit: d80f807b1a]
2025-04-09 13:49:04 -04:00
Aryan Salmanpour d80f807b1a Code Clean up - remove unused API (#552)
* Code Clean up - remove unused API

* clean up
2025-04-09 13:49:04 -04:00
Justin Williams 3f75cd906f Created ABI Compliance Checker
Signed-off-by: Justin Williams <Justin.Williams@amd.com>
2025-04-09 12:40:02 -05:00
Justin Williams 574144c9a0 Created ABI Compliance Checker
Signed-off-by: Justin Williams <Justin.Williams@amd.com>


[ROCm/amdsmi commit: 3f75cd906f]
2025-04-09 12:40:02 -05:00
Chaudhary, Jatin Jaikishan b0d9ea7454 SWDEV-525969 - add gfx950 to use ocp type for fp8 (#157)
* do not use __gfx94plus_clr__ macro in fp8 header


[ROCm/clr commit: 5214d1ca07]
2025-04-09 16:21:39 +01:00
Chaudhary, Jatin Jaikishan 5214d1ca07 SWDEV-525969 - add gfx950 to use ocp type for fp8 (#157)
* do not use __gfx94plus_clr__ macro in fp8 header
2025-04-09 16:21:39 +01:00
Stojiljkovic, Vladana ef52738425 Set maxTexture2DLinear fileds in deviceProp (#89)
[ROCm/clr commit: bc474ea5af]
2025-04-09 17:13:49 +02:00
Stojiljkovic, Vladana bc474ea5af Set maxTexture2DLinear fileds in deviceProp (#89) 2025-04-09 17:13:49 +02:00
Galantsev, Dmitrii 2e429ed890 Revert "CMAKE - Force INSTALL_LIBDIR to be lib"
This reverts commit 62c10bfe3c.
2025-04-09 01:38:42 -05:00
Galantsev, Dmitrii 055f603ab9 Revert "CMAKE - Force INSTALL_LIBDIR to be lib"
This reverts commit 7bbe33c94d.


[ROCm/amdsmi commit: 2e429ed890]
2025-04-09 01:38:42 -05:00