Commit Graph

74713 Commits

Author SHA1 Message Date
Yiannis Papadopoulos 2d2c47bdef rocr/aie: Increment write pointer upon packet submission 2025-04-08 15:36:40 -05:00
Yiannis Papadopoulos 96b7e42776 rocr/aie: Increment write pointer upon packet submission
[ROCm/ROCR-Runtime commit: 2d2c47bdef]
2025-04-08 15:36:40 -05:00
Yiltan 4d6fd799ef Enable RO CI (#65)
[ROCm/rocshmem commit: 25e7109b64]
2025-04-08 16:12:22 -04:00
Yiltan 25e7109b64 Enable RO CI (#65) 2025-04-08 16:12:22 -04:00
Nikhil-Nunna ee9da06c80 Fetching RCCL version from shared object file. (#1569)
* added rccl version using rccl-tests

* Added function to get rccl version from rccl-tests

* removed whitespace

* Added rccl version

* Updated readme and fixed formatting

* removed debug prints

[ROCm/rccl commit: 3dc0478722]
2025-04-08 14:09:39 -05:00
Nikhil-Nunna 3dc0478722 Fetching RCCL version from shared object file. (#1569)
* added rccl version using rccl-tests

* Added function to get rccl version from rccl-tests

* removed whitespace

* Added rccl version

* Updated readme and fixed formatting

* removed debug prints
2025-04-08 14:09:39 -05:00
Choudhary, Rahul 08ef6d6023 Update rocm_ci_caller.yml enabling !verify and re-test trigger (#6)
[ROCm/rocm-core commit: b793b7a765]
2025-04-08 10:50:47 -07:00
Choudhary, Rahul b793b7a765 Update rocm_ci_caller.yml enabling !verify and re-test trigger (#6) 2025-04-08 10:50:47 -07:00
Alex Breslow 9da345dadf Add instructions to README regarding benchmarking on pre ROCm 6.4.x versions with HSA_NO_SCRATCH_RECLAIM=1 (#114)
[ROCm/rccl-tests commit: 284ff2ac84]
2025-04-08 09:59:57 -07:00
Alex Breslow 284ff2ac84 Add instructions to README regarding benchmarking on pre ROCm 6.4.x versions with HSA_NO_SCRATCH_RECLAIM=1 (#114) 2025-04-08 09:59:57 -07:00
Eric Huang df6048429c kfdtest: fix max queues on multi-gpu mode
The max queues per process is 1024 in KFD,
KFDQMTest.OverSubscribeCpQueues fails with multi-gpu mode
on more than 15 gpus, because 65x16=1040 exceeds 1024, so
changing MAX_CP_QUEUES to adapt it will fix the issue.

Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
2025-04-08 12:57:00 -04:00
Eric Huang 13cdca7fb3 kfdtest: fix max queues on multi-gpu mode
The max queues per process is 1024 in KFD,
KFDQMTest.OverSubscribeCpQueues fails with multi-gpu mode
on more than 15 gpus, because 65x16=1040 exceeds 1024, so
changing MAX_CP_QUEUES to adapt it will fix the issue.

Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>


[ROCm/ROCR-Runtime commit: df6048429c]
2025-04-08 12:57:00 -04:00
Brzak, Branislav d4275741ba SWDEV-525653 - Make hipGetDeviceProperties and hipChooseDevice use the new API (#159)
[ROCm/clr commit: b006380ff6]
2025-04-08 18:54:05 +02:00
Brzak, Branislav b006380ff6 SWDEV-525653 - Make hipGetDeviceProperties and hipChooseDevice use the new API (#159) 2025-04-08 18:54:05 +02:00
Pratik Basyal 7115d36ceb Broken link and reference text updated (#664)
[ROCm/rocprofiler-compute commit: 3b67a00bc9]
2025-04-08 12:44:09 -04:00
Pratik Basyal 3b67a00bc9 Broken link and reference text updated (#664) 2025-04-08 12:44:09 -04:00
Avinash Kethineedi 9bd2b04899 Update Barrier and Sync APIs (#73)
* Add thread, wavefront, and workgroup-level `barrier` APIs in IPC and RO conduits; remove collectives on default context
 - Implemented `barrier` APIs for thread, wavefront, and workgroup scopes
 - Added support into both IPC and RO conduits
 - Added functional tests to cover all `barrier` APIs
 - Removed collective operations on default context

* Add thread, wavefront, and workgroup-level `sync` APIs in IPC and RO conduits.
  - Implemented `sync` APIs for thread, wavefront, and workgroup scopes
  - Added support into both IPC and RO conduits
  - Added functional tests to cover all `sync` APIs

* update naming convention for context-based `barrier` APIs

[ROCm/rocshmem commit: dc61bca066]
2025-04-08 11:25:31 -05:00
Avinash Kethineedi dc61bca066 Update Barrier and Sync APIs (#73)
* Add thread, wavefront, and workgroup-level `barrier` APIs in IPC and RO conduits; remove collectives on default context
 - Implemented `barrier` APIs for thread, wavefront, and workgroup scopes
 - Added support into both IPC and RO conduits
 - Added functional tests to cover all `barrier` APIs
 - Removed collective operations on default context

* Add thread, wavefront, and workgroup-level `sync` APIs in IPC and RO conduits.
  - Implemented `sync` APIs for thread, wavefront, and workgroup scopes
  - Added support into both IPC and RO conduits
  - Added functional tests to cover all `sync` APIs

* update naming convention for context-based `barrier` APIs
2025-04-08 11:25:31 -05:00
Galantsev, Dmitrii 62c10bfe3c CMAKE - Force INSTALL_LIBDIR to be lib
On some systems it defaults to lib64, on others to lib.

Change-Id: I973b488253d106ded518ee590a0edb370927f9a4
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
2025-04-08 16:02:43 +00:00
Galantsev, Dmitrii 7bbe33c94d CMAKE - Force INSTALL_LIBDIR to be lib
On some systems it defaults to lib64, on others to lib.

Change-Id: I973b488253d106ded518ee590a0edb370927f9a4
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>


[ROCm/amdsmi commit: 62c10bfe3c]
2025-04-08 16:02:43 +00:00
Mallya, Ameya Keshava 4d8e9cfa1d Added !verify features
[ROCm/rocm_smi_lib commit: d18e40f204]
2025-04-08 07:59:30 -07:00
Mallya, Ameya Keshava d18e40f204 Added !verify features 2025-04-08 07:59:30 -07:00
Eric Huang d3265234e9 kfdtest: fix ptrace error on multi-gpu mode
The parent process can only be ptraced by 1 process
once, to avoid the error we have to add mutex to
synchronize the ptrace call.

Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>
2025-04-08 09:58:28 -04:00
Eric Huang 9055cf8092 kfdtest: fix ptrace error on multi-gpu mode
The parent process can only be ptraced by 1 process
once, to avoid the error we have to add mutex to
synchronize the ptrace call.

Signed-off-by: Eric Huang <jinhuieric.huang@amd.com>


[ROCm/ROCR-Runtime commit: d3265234e9]
2025-04-08 09:58:28 -04:00
Patel, Jaydeepkumar 2f3bc7f01c SWDEV-521011 - Allow max stack size as per ISA. (#73)
[ROCm/clr commit: 9e7248aa36]
2025-04-08 10:15:38 +05:30
Patel, Jaydeepkumar 9e7248aa36 SWDEV-521011 - Allow max stack size as per ISA. (#73) 2025-04-08 10:15:38 +05:30
Galantsev, Dmitrii 24024f0e4f Revert "Implement CPU discovery support"
This reverts commit f967f8a17d15e148464393fcd145af01dc0e1525.
2025-04-07 20:45:19 -05:00
Galantsev, Dmitrii 5276903800 Revert "Implement CPU discovery support"
This reverts commit f967f8a17d15e148464393fcd145af01dc0e1525.


[ROCm/rdc commit: 24024f0e4f]
2025-04-07 20:45:19 -05:00
Galantsev, Dmitrii c96f5db52c Revert "Fix breaking changes introduced with CPU support"
This reverts commit e9ac9e4626e3e45ebdfafb39e251d073091429f1.
2025-04-07 20:45:19 -05:00
Galantsev, Dmitrii 8afcedfc96 Revert "Fix breaking changes introduced with CPU support"
This reverts commit e9ac9e4626e3e45ebdfafb39e251d073091429f1.


[ROCm/rdc commit: c96f5db52c]
2025-04-07 20:45:19 -05:00
xuchen-amd 76e7d942f6 Add mi325 specs. (#663)
[ROCm/rocprofiler-compute commit: e7a7af539a]
2025-04-07 17:03:40 -04:00
xuchen-amd e7a7af539a Add mi325 specs. (#663) 2025-04-07 17:03:40 -04:00
Williams, Justin e3ab8cf71b [SWDEV-500518] Updated CI Structure (#244)
Signed-off-by: Justin Williams <Justin.Williams@amd.com>
2025-04-07 15:02:57 -05:00
Williams, Justin 52195d6505 [SWDEV-500518] Updated CI Structure (#244)
Signed-off-by: Justin Williams <Justin.Williams@amd.com>

[ROCm/amdsmi commit: e3ab8cf71b]
2025-04-07 15:02:57 -05:00
Galantsev, Dmitrii 0aeceefcb3 Fix breaking changes introduced with CPU support
Changes introduced in 3bdca8b8b6
broke RDC if it was compiled without ESMI support, or if esmi driver is
not loaded when RDC is being used.

Change-Id: Id54e1e9002d2e3cf09240081149eed84178700af
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>
2025-04-07 14:41:46 -05:00
Galantsev, Dmitrii 3e8f56c430 Fix breaking changes introduced with CPU support
Changes introduced in f0f44d977f
broke RDC if it was compiled without ESMI support, or if esmi driver is
not loaded when RDC is being used.

Change-Id: Id54e1e9002d2e3cf09240081149eed84178700af
Signed-off-by: Galantsev, Dmitrii <dmitrii.galantsev@amd.com>


[ROCm/rdc commit: 0aeceefcb3]
2025-04-07 14:41:46 -05:00
cfallows-amd 04919c13e0 Fixes for roofline datatype plot outputs (#659)
Profile mode:
Fix roofline plots for datatypes that have peakVALU only. Check for highest roofline to plot the bandwidth lines to proper height, don't rely on existence of peakMFMA for every datatype.
Analyze mode:
Add roofline-data-type option for viewing pdfs in standalone gui. Default is same as profile mode, FP32.

---------

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

[ROCm/rocprofiler-compute commit: c45e20f325]
2025-04-07 12:10:37 -04:00
cfallows-amd c45e20f325 Fixes for roofline datatype plot outputs (#659)
Profile mode:
Fix roofline plots for datatypes that have peakVALU only. Check for highest roofline to plot the bandwidth lines to proper height, don't rely on existence of peakMFMA for every datatype.
Analyze mode:
Add roofline-data-type option for viewing pdfs in standalone gui. Default is same as profile mode, FP32.

---------

Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com>
2025-04-07 12:10:37 -04:00
Choudhary, Rahul 5b4c717208 Update workflow to use mainline branch 2025-04-07 09:36:52 -04:00
Choudhary, Rahul 480ca4d5e7 Update workflow to use mainline branch
[ROCm/ROCR-Runtime commit: 5b4c717208]
2025-04-07 09:36:52 -04:00
Choudhary, Rahul a0b80c825c Update rocm_ci_caller.yml updating push trigger to amd-mainline
Signed-off-by: Choudhary, Rahul <Rahul.Choudhary@amd.com>
2025-04-07 09:36:52 -04:00
Choudhary, Rahul d605e9256c Update rocm_ci_caller.yml updating push trigger to amd-mainline
Signed-off-by: Choudhary, Rahul <Rahul.Choudhary@amd.com>

[ROCm/ROCR-Runtime commit: a0b80c825c]
2025-04-07 09:36:52 -04:00
Betigeri, Sourabh e3e31f1486 SWDEV-523281 - Adds hipLaunchKernelEx C++ wrapper (#44)
[ROCm/hip commit: 3998c11287]
2025-04-07 09:24:01 +05:30
Betigeri, Sourabh 3998c11287 SWDEV-523281 - Adds hipLaunchKernelEx C++ wrapper (#44) 2025-04-07 09:24:01 +05:30
Pham, Gabriel e2c371ece4 [SWDEV-524288] Fixed duplication of GPU id in events. (#233)
* Fixed duplication of GPU id in events.

---------

Signed-off-by: gabrpham_amdeng <Gabriel.Pham@amd.com>
2025-04-04 18:31:08 -05:00
Pham, Gabriel b485d4ba70 [SWDEV-524288] Fixed duplication of GPU id in events. (#233)
* Fixed duplication of GPU id in events.

---------

Signed-off-by: gabrpham_amdeng <Gabriel.Pham@amd.com>

[ROCm/amdsmi commit: e2c371ece4]
2025-04-04 18:31:08 -05:00
Arif, Maisam 50d7d5287f Removed unnecessary rocm-smi files (#217)
* Removed unnecessary rocm-smi files
* Moved the update wrapper script into the tools folder

---------

Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
2025-04-04 18:26:43 -05:00
Arif, Maisam cb3c979dfe Removed unnecessary rocm-smi files (#217)
* Removed unnecessary rocm-smi files
* Moved the update wrapper script into the tools folder

---------

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

[ROCm/amdsmi commit: 50d7d5287f]
2025-04-04 18:26:43 -05:00
Maisam Arif 0da6613b99 Updated imports on amdsmi_quick_start
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Change-Id: Ideb0f2addf61fb6bdb728e549a8b0f133682d7d6
2025-04-04 18:25:17 -05:00
Maisam Arif dab3e0657b Updated imports on amdsmi_quick_start
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
Change-Id: Ideb0f2addf61fb6bdb728e549a8b0f133682d7d6


[ROCm/amdsmi commit: 0da6613b99]
2025-04-04 18:25:17 -05:00