* rocprofv3: do not abort if counter does not have dimensions
* Relax error handling further in rocprofv3 metadata
---------
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
[ROCm/rocprofiler-sdk commit: 3071199386]
* rocprofv3: do not abort if counter does not have dimensions
* Relax error handling further in rocprofv3 metadata
---------
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
- VA API tracing using Timemory gotcha wrappers.
- rocDecode API tracing integration using callback to ROCPROFILER_CALLBACK_TRACING_ROCDECODE_API
- Updated videodecode ctest to validate rocDecode APIs in perfetto trace.
[ROCm/rocprofiler-systems commit: 697d1ac02f]
- VA API tracing using Timemory gotcha wrappers.
- rocDecode API tracing integration using callback to ROCPROFILER_CALLBACK_TRACING_ROCDECODE_API
- Updated videodecode ctest to validate rocDecode APIs in perfetto trace.
* Accept an EXPLICIT_ROCM_VERSION and use that vs inspecting the environment if provided.
* Use CMake's built in file reading support vs execute_process (without error checking) to avoid silent but deadly later failures.
* Properly quote some comparisons to avoid syntax errors if they happen to have an empty string.
* Guard against ROCM_PATH being an empty string, avoiding stray path extensions to root directories, etc.
Co-authored-by: Stella Laurenzo <stellaraccident@gmail.com>
[ROCm/rccl commit: 92ac136db5]
* Accept an EXPLICIT_ROCM_VERSION and use that vs inspecting the environment if provided.
* Use CMake's built in file reading support vs execute_process (without error checking) to avoid silent but deadly later failures.
* Properly quote some comparisons to avoid syntax errors if they happen to have an empty string.
* Guard against ROCM_PATH being an empty string, avoiding stray path extensions to root directories, etc.
Co-authored-by: Stella Laurenzo <stellaraccident@gmail.com>
* Switched cmake_host_system_information feature to a manual parse to remain cmake 3.5 compliant.
* Updating minimum cmake to 3.16 to conform with the rest of ROCm. This change still applies.
[ROCm/rccl commit: 42ab425037]
* Switched cmake_host_system_information feature to a manual parse to remain cmake 3.5 compliant.
* Updating minimum cmake to 3.16 to conform with the rest of ROCm. This change still applies.
Most sysfs reads do not require cross-process level mutex, and writes to sysfs
should be protected by the kernel already.
Users can still switch to the old behavior by setting the environment variable AMDSMI_MUTEX_CROSS_PROCESS=1.
Most sysfs reads do not require cross-process level mutex, and writes to sysfs
should be protected by the kernel already.
Users can still switch to the old behavior by setting the environment variable AMDSMI_MUTEX_CROSS_PROCESS=1.
[ROCm/amdsmi commit: 8e74024b11]
- Fix regression for D2H pinned copies which adds systemscope release.
- Skip cpu wait for D2H unpinned copies as we can pass the signal of the
barrier to rocr copy.
- Fix an old bug in sdmaEngineRetainCount_ logic
- Improve logging
Change-Id: If074bddb05564b15949b0d5f9bf12acd3692174e
[ROCm/clr commit: 4c95ee5e1e]
- Fix regression for D2H pinned copies which adds systemscope release.
- Skip cpu wait for D2H unpinned copies as we can pass the signal of the
barrier to rocr copy.
- Fix an old bug in sdmaEngineRetainCount_ logic
- Improve logging
Change-Id: If074bddb05564b15949b0d5f9bf12acd3692174e
* Adding pc sampling how to guide
* doc update
* Fixing indentation
* updating index
* udpating doc
* updating doc
* Added field information
* Fixing Formatting
* fix formatting error
* Added json format for pc sampling
* feedback resolved
* formatting for text
* PC Sampling API doc
* Reformatted
* Note for shared systems
* update docs
* correcting relative path for cross-referencing
---------
Co-authored-by: vlaindic_amdeng <vladimir.indic@amd.com>
[ROCm/rocprofiler-sdk commit: cdf22eba7d]
* Adding pc sampling how to guide
* doc update
* Fixing indentation
* updating index
* udpating doc
* updating doc
* Added field information
* Fixing Formatting
* fix formatting error
* Added json format for pc sampling
* feedback resolved
* formatting for text
* PC Sampling API doc
* Reformatted
* Note for shared systems
* update docs
* correcting relative path for cross-referencing
---------
Co-authored-by: vlaindic_amdeng <vladimir.indic@amd.com>
* initial hack to fix for v3 stucking on mi100 becasue of -m parameter and missing counter csv file
* proper formating
* refactored profiler option function to take soc arch
* resolve missing step that casued error for profiler option
* fix typo of arch name
* change method of putting soc info into profiler option
* isort and black format
* add comment for the part that handles missing counter csv file
* remove unncecessary import
---------
Co-authored-by: YANG WANG <ywang@ywang-ubuntu.amd.com>
[ROCm/rocprofiler-compute commit: 5ee37b3353]
* initial hack to fix for v3 stucking on mi100 becasue of -m parameter and missing counter csv file
* proper formating
* refactored profiler option function to take soc arch
* resolve missing step that casued error for profiler option
* fix typo of arch name
* change method of putting soc info into profiler option
* isort and black format
* add comment for the part that handles missing counter csv file
* remove unncecessary import
---------
Co-authored-by: YANG WANG <ywang@ywang-ubuntu.amd.com>
Environment variable HSA_HIGH_PRECISION_MODE can be used to control MFMA
precision
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: Ib78dd9dd8867025e090a3cca96ab6db4f65dea12
Environment variable HSA_HIGH_PRECISION_MODE can be used to control MFMA
precision
Signed-off-by: Harish Kasiviswanathan <Harish.Kasiviswanathan@amd.com>
Change-Id: Ib78dd9dd8867025e090a3cca96ab6db4f65dea12
[ROCm/ROCR-Runtime commit: 2a64fa5e06]
* Force HSA_AMD_MEMORY_POOL_EXECUTABLE_FLAG value to be used with HSA calls
Fix for CI
* More tweaks
* Increase reproducible-runtime kernel sleep granularity
* Fix data race in synchronous device counter collection sample
* Update device counting service
- add get_active_context function
---------
Co-authored-by: Benjamin Welton <bewelton@amd.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
[ROCm/rocprofiler-sdk commit: 080b2ba451]
* Force HSA_AMD_MEMORY_POOL_EXECUTABLE_FLAG value to be used with HSA calls
Fix for CI
* More tweaks
* Increase reproducible-runtime kernel sleep granularity
* Fix data race in synchronous device counter collection sample
* Update device counting service
- add get_active_context function
---------
Co-authored-by: Benjamin Welton <bewelton@amd.com>
Co-authored-by: Jonathan R. Madsen <jonathanrmadsen@gmail.com>
Madsen, Jonathan
Wenkai Du
Kandula, Venkateshwar reddy
Rahul Manocha
Sajina PK
rahulc1984
cfallows-amd
Lakshmi Kumar
corey-derochie-amd
Nilesh M Negi
Liu, Shuzhou (Bill)
kjayapra-amd
Arif, Maisam
Saleel Kudchadker
Bhardwaj, Gopesh
Elwazir, Ammar
ywang103-amd
xuchen-amd
Harish Kasiviswanathan
Yiltan
Welton, Benjamin