e81be86a31
rocr: use driver interface for scratch memory deallocation
...
Replace direct hsaKmtFreeMemory call with driver's FreeMemory interface
in GpuAgent::ReleaseResources(). This change improves code abstraction
by handling memory deallocation through the unified driver interface.
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com >
[ROCm/ROCR-Runtime commit: 8216787d4c
2025-07-16 13:15:45 +08:00
da8dd9e1e3
rocr/driver: add scratch memory allocation into driver interface
...
Add AllocateScratchMemory interface to Driver base class and implement it
in both KFD and XDNA drivers. This change encapsulates the low-level
scratch memory allocation details within driver implementations, making
the code more maintainable and the interface cleaner.
The main changes include:
- Add AllocateScratchMemory virtual method to Driver interface
- Implement the interface in KfdDriver with existing allocation logic
- Add stub implementation in XdnaDriver
- Update GpuAgent to use the new interface instead of direct KMT calls
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com >
2025-07-16 13:15:45 +08:00
1ad79f04d0
rocr/driver: add scratch memory allocation into driver interface
...
Add AllocateScratchMemory interface to Driver base class and implement it
in both KFD and XDNA drivers. This change encapsulates the low-level
scratch memory allocation details within driver implementations, making
the code more maintainable and the interface cleaner.
The main changes include:
- Add AllocateScratchMemory virtual method to Driver interface
- Implement the interface in KfdDriver with existing allocation logic
- Add stub implementation in XdnaDriver
- Update GpuAgent to use the new interface instead of direct KMT calls
Signed-off-by: Honglei Huang <Honglei1.Huang@amd.com >
[ROCm/ROCR-Runtime commit: da8dd9e1e3
2025-07-16 13:15:45 +08:00
5a767cf272
Remove an undefined submodule from the GOTCHA submodule ( #285 )
...
A recursive submodule update, `git submodule uppdate --recursive
--init`,
would fail due to an improperly defined and empty submodule in GOTCHA.
Signed-off-by: David Galiffi <David.Galiffi@amd.com >
[ROCm/rocprofiler-systems commit: e9b09b6ade
2025-07-15 21:32:43 -04:00
e9b09b6ade
Remove an undefined submodule from the GOTCHA submodule ( #285 )
...
A recursive submodule update, `git submodule uppdate --recursive
--init`,
would fail due to an improperly defined and empty submodule in GOTCHA.
Signed-off-by: David Galiffi <David.Galiffi@amd.com >
2025-07-15 21:32:43 -04:00
3b1957e674
[SWDEV-531904] Added test_get_gpu_revision ( #533 )
...
* [SWDEV-531904] Added test_get_gpu_revision
New:
- amdsmi_get_gpu_revision() previously not implemented in amdsmi_interface.py
- test_get_gpu_revision() missing integration test.
Updated:
-changelog.md added new doc fields for ROCm 7.1
-amdsmi-py-api.md added field|description doc fields
Signed-off-by: Juan Castillo <juan.castillo@amd.com >
2025-07-15 19:35:54 -05:00
801dbaedec
[SWDEV-531904] Added test_get_gpu_revision ( #533 )
...
* [SWDEV-531904] Added test_get_gpu_revision
New:
- amdsmi_get_gpu_revision() previously not implemented in amdsmi_interface.py
- test_get_gpu_revision() missing integration test.
Updated:
-changelog.md added new doc fields for ROCm 7.1
-amdsmi-py-api.md added field|description doc fields
Signed-off-by: Juan Castillo <juan.castillo@amd.com >
[ROCm/amdsmi commit: 3b1957e674
2025-07-15 19:35:54 -05:00
9a442748ea
Fix docker testing instructions ( #813 )
...
[ROCm/rocprofiler-compute commit: ccb34e8a91
2025-07-15 17:06:22 -04:00
ccb34e8a91
Fix docker testing instructions ( #813 )
2025-07-15 17:06:22 -04:00
816f7a850f
[SWDEV-379269] Add all gpus as default to dmon ( #211 )
...
Change-Id: Idb17e9018c39479830a4366f2002d02725d66873
Signed-off-by: adapryor <Adam.pryor@amd.com >
2025-07-15 16:03:28 -05:00
010ac416b1
[SWDEV-379269] Add all gpus as default to dmon ( #211 )
...
Change-Id: Idb17e9018c39479830a4366f2002d02725d66873
Signed-off-by: adapryor <Adam.pryor@amd.com >
[ROCm/rdc commit: 816f7a850f
2025-07-15 16:03:28 -05:00
03414e20ee
SWDEV-539482: Different sizes of mem leaks observed in amdsmitst ( #538 )
...
Signed-off-by:Oosman Saeed <oossaeed@amd.com >
2025-07-15 14:33:27 -05:00
9092f915fe
SWDEV-539482: Different sizes of mem leaks observed in amdsmitst ( #538 )
...
Signed-off-by:Oosman Saeed <oossaeed@amd.com >
[ROCm/amdsmi commit: 03414e20ee
2025-07-15 14:33:27 -05:00
bfe76cf94e
rocr: Using MemoryRegion::GetInfo(HSA_REGION_INFO_ALLOC_MAX_SIZE) for HSA_AMD_MEMORY_POOL_INFO_ALLOC_MAX_SIZE
2025-07-15 14:22:56 -05:00
a0ef6f6473
rocr: Using MemoryRegion::GetInfo(HSA_REGION_INFO_ALLOC_MAX_SIZE) for HSA_AMD_MEMORY_POOL_INFO_ALLOC_MAX_SIZE
...
[ROCm/ROCR-Runtime commit: bfe76cf94e
2025-07-15 14:22:56 -05:00
7af4bd7ddf
Add roccv to scope
2025-07-15 15:20:42 -04:00
10ead63735
Update workflows for building ( #812 )
...
* Update workflows for building testing in pipeline: rhel8 now tests rhel8 and 9, and for both rhel and ubuntu files we are using the same image as rocprofiler-systems which is consistently updated on docker hub.
---------
Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com >
[ROCm/rocprofiler-compute commit: 10c696521d
2025-07-15 15:12:21 -04:00
10c696521d
Update workflows for building ( #812 )
...
* Update workflows for building testing in pipeline: rhel8 now tests rhel8 and 9, and for both rhel and ubuntu files we are using the same image as rocprofiler-systems which is consistently updated on docker hub.
---------
Signed-off-by: Carrie Fallows <Carrie.Fallows@amd.com >
2025-07-15 15:12:21 -04:00
1b18684171
Add aqlprofile to readme
2025-07-15 14:53:47 -04:00
f96cafaa60
[rocprofv3] Sanitize string of command in rocpd ( #467 )
...
Sanitize string of command in rocpd
Co-authored-by: Huanran Wang <huanrwan@amd.com >
[ROCm/rocprofiler-sdk commit: 5c45c77ec7
2025-07-15 13:01:14 -04:00
5c45c77ec7
[rocprofv3] Sanitize string of command in rocpd ( #467 )
...
Sanitize string of command in rocpd
Co-authored-by: Huanran Wang <huanrwan@amd.com >
2025-07-15 13:01:14 -04:00
930b32e2cb
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 >
[ROCm/rocprofiler-compute commit: 1d834f4190
2025-07-15 12:42:27 -04:00
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
c3f8f57c80
Reset state of repo with updated README
2025-07-15 12:29:35 -04:00
b439b4e719
[rocprofv3] Fix incorrect position of Allocation_Size header in memory_allocation csv ( #510 )
...
Fix incorrect position of Allocation_Size header in memory_allocation csv
[ROCm/rocprofiler-sdk commit: 4af73d078f
2025-07-15 18:15:22 +02:00
4af73d078f
[rocprofv3] Fix incorrect position of Allocation_Size header in memory_allocation csv ( #510 )
...
Fix incorrect position of Allocation_Size header in memory_allocation csv
2025-07-15 18:15:22 +02:00
52363f797e
SWDEV-540629 - Update the current device ID ( #699 )
...
[ROCm/clr commit: 0313e20706
2025-07-15 21:40:26 +05:30
0313e20706
SWDEV-540629 - Update the current device ID ( #699 )
2025-07-15 21:40:26 +05:30
680c7fbf64
SWDEV-483895 - Use ROCr to reserve memory for HMM ( #590 )
...
[ROCm/clr commit: a1f056bd11
2025-07-15 21:40:13 +05:30
a1f056bd11
SWDEV-483895 - Use ROCr to reserve memory for HMM ( #590 )
2025-07-15 21:40:13 +05:30
b6d69d162e
SWDEV-542832 - Refactor vector types alignment ( #713 )
...
[ROCm/clr commit: 918b0bdb7a
2025-07-15 21:09:31 +05:30
918b0bdb7a
SWDEV-542832 - Refactor vector types alignment ( #713 )
2025-07-15 21:09:31 +05:30
670966f86b
Fix inline compilation issue with LL ( #1806 )
...
[ROCm/rccl commit: 106024b0db
2025-07-15 08:39:18 -07:00
106024b0db
Fix inline compilation issue with LL ( #1806 )
2025-07-15 08:39:18 -07:00
54bf215b7d
SWDEV-539145 - Fix compilation error in windows psdb ( #376 )
...
Co-authored-by: Rahul Manocha <rmanocha@amd.com >
[ROCm/hip-tests commit: 2a0bcca07b
2025-07-15 08:38:24 -07:00
2a0bcca07b
SWDEV-539145 - Fix compilation error in windows psdb ( #376 )
...
Co-authored-by: Rahul Manocha <rmanocha@amd.com >
2025-07-15 08:38:24 -07:00
a91363744c
Update TUI docs. ( #796 )
...
[ROCm/rocprofiler-compute commit: bfb2dc0795
2025-07-15 11:13:24 -04:00
bfb2dc0795
Update TUI docs. ( #796 )
2025-07-15 11:13:24 -04:00
1050e0201c
Update peak flops for MI350 ( #810 )
...
[ROCm/rocprofiler-compute commit: e1da9355c2
2025-07-15 08:46:13 -06:00
e1da9355c2
Update peak flops for MI350 ( #810 )
2025-07-15 08:46:13 -06:00
65673617e0
Remove rocm-smi for ROCm 7.1 ( #809 )
...
[ROCm/rocprofiler-compute commit: b4b50502e3
2025-07-15 10:28:53 -04:00
b4b50502e3
Remove rocm-smi for ROCm 7.1 ( #809 )
2025-07-15 10:28:53 -04:00
a20e65cfc0
gfx950 updated on LL thresholds for allreduce/allgather, update treeCorrection ( #1803 )
...
* change LL thresholds for allreduce/allgather and update treeCorrectionFactor
* update allGather LL cutoff
* adjust allgather LL/LL128 thresholds
[ROCm/rccl commit: 8d0f1a1cef
2025-07-15 09:10:19 -05:00
8d0f1a1cef
gfx950 updated on LL thresholds for allreduce/allgather, update treeCorrection ( #1803 )
...
* change LL thresholds for allreduce/allgather and update treeCorrectionFactor
* update allGather LL cutoff
* adjust allgather LL/LL128 thresholds
2025-07-15 09:10:19 -05:00
d4da72bf2d
Update gotcha submodule from timemory ( #277 )
...
* Update gotcha submodule from timemory
* Fix build failure and add copilot suggestions
* Fix formatting errors
[ROCm/rocprofiler-systems commit: d26486ad83
2025-07-14 21:12:10 -04:00
d26486ad83
Update gotcha submodule from timemory ( #277 )
...
* Update gotcha submodule from timemory
* Fix build failure and add copilot suggestions
* Fix formatting errors
2025-07-14 21:12:10 -04:00
a5ca94ab9c
Fix ROCtx event ranges in trace output ( #278 )
...
* Fix marker api traces
* Remove space
* Formatting change
* Small change
* Update Changelog
* Add period to changelog
* Update source/lib/rocprof-sys/library/rocprofiler-sdk.cpp
Co-authored-by: David Galiffi <David.Galiffi@amd.com >
* Fix roctx tests
---------
Co-authored-by: David Galiffi <David.Galiffi@amd.com >
[ROCm/rocprofiler-systems commit: c996c23a13
2025-07-14 19:31:14 -04:00
c996c23a13
Fix ROCtx event ranges in trace output ( #278 )
...
* Fix marker api traces
* Remove space
* Formatting change
* Small change
* Update Changelog
* Add period to changelog
* Update source/lib/rocprof-sys/library/rocprofiler-sdk.cpp
Co-authored-by: David Galiffi <David.Galiffi@amd.com >
* Fix roctx tests
---------
Co-authored-by: David Galiffi <David.Galiffi@amd.com >
2025-07-14 19:31:14 -04:00
1892bdef83
Make CMakeFile fixes to align with rocDecode and rocJpeg changes ( #281 )
...
[ROCm/rocprofiler-systems commit: be06384250
2025-07-14 19:23:19 -04:00
be06384250
Make CMakeFile fixes to align with rocDecode and rocJpeg changes ( #281 )
2025-07-14 19:23:19 -04:00
Honglei Huang
David Galiffi
Castillo, Juan
vedithal-amd
Pryor, Adam
Saeed, Oosman
Yiannis Papadopoulos
Joseph Macaranas
cfallows-amd
Wang, Huanran
amd-jmacaran
Tumakaev, Aleksei
Betigeri, Sourabh
Dittakavi, Satyanvesh
Wenkai Du
Manocha, Rahul
xuchen-amd
Fei Zheng
isaki001
Sajina PK
darren-amd