- Updates:
- Adding the following metrics to allow new calculations for violation status:
- Per XCP metrics gfx_below_host_limit_ppt_acc
- Per XCP metrics gfx_below_host_limit_thm_acc
- Per XCP metrics gfx_low_utilization_acc
- Per XCP metrics gfx_below_host_limit_total_acc
- Increasing available JPEG engines to 40. Current ASICs may not support all 40. These will be indicated as UINT16_MAX or N/A in CLI.
Signed-off-by: Juan Castillo <juan.castillo@amd.com>
Co-authored-by: Charis Poag <Charis.Poag@amd.com>
- Updates:
- Adding the following metrics to allow new calculations for violation status:
- Per XCP metrics gfx_below_host_limit_ppt_acc
- Per XCP metrics gfx_below_host_limit_thm_acc
- Per XCP metrics gfx_low_utilization_acc
- Per XCP metrics gfx_below_host_limit_total_acc
- Increasing available JPEG engines to 40. Current ASICs may not support all 40. These will be indicated as UINT16_MAX or N/A in CLI.
Signed-off-by: Juan Castillo <juan.castillo@amd.com>
Co-authored-by: Charis Poag <Charis.Poag@amd.com>
[ROCm/amdsmi commit: 7c882b2f69]
- PR runs targeting develop will require approval from at least one person in a predefined list.
- Nightly runs are separated and will not require approval.
- Sample pytest script is provided for expanding test coverage.
[ROCm/rccl commit: 6403afff4a]
- PR runs targeting develop will require approval from at least one person in a predefined list.
- Nightly runs are separated and will not require approval.
- Sample pytest script is provided for expanding test coverage.
* solve the error that makes name passed by -n not used in multi-node applications
* isort and black formatted
[ROCm/rocprofiler-compute commit: 23b42e90c9]
The over arching goal it so provide an API that pre-silicon models can latch into for software bring up.# Please enter the commit message for your changes. Lines starting
The over arching goal it so provide an API that pre-silicon models can latch into for software bring up.# Please enter the commit message for your changes. Lines starting
[ROCm/ROCR-Runtime commit: d4b85b6bf5]
* Update primitive tests for multi-workgroup support
* Update workgroup primitive tests for multi-workgroup support
* Update workfront primitive tests for multi-workgroup support
* Update team based primitive tests for multi-workgroup support
* Update RMA functional tests to capture timing after quiet call
- Modified RMA functional tests to record the time after a `quiet` call in thread, wavefront, and workgroup RMA calls.
* Improve error handling and memory management
- Replaced `cout` with `cerr` for improved error reporting.
- Ensured all allocated memory is freed when `rocshmem_malloc` fails.
* Update start time in primitive tests and latency calculations
- Modified primitive tests to capture the earliest start time.
- Updated latency calculations in functional tests.
* Remove `GetSwarmTester`
* Update start time in team primitive tests
* Invoke quiet call from a single thread within a block on a rocshmem context
[ROCm/rocshmem commit: aa3121a967]
* Update primitive tests for multi-workgroup support
* Update workgroup primitive tests for multi-workgroup support
* Update workfront primitive tests for multi-workgroup support
* Update team based primitive tests for multi-workgroup support
* Update RMA functional tests to capture timing after quiet call
- Modified RMA functional tests to record the time after a `quiet` call in thread, wavefront, and workgroup RMA calls.
* Improve error handling and memory management
- Replaced `cout` with `cerr` for improved error reporting.
- Ensured all allocated memory is freed when `rocshmem_malloc` fails.
* Update start time in primitive tests and latency calculations
- Modified primitive tests to capture the earliest start time.
- Updated latency calculations in functional tests.
* Remove `GetSwarmTester`
* Update start time in team primitive tests
* Invoke quiet call from a single thread within a block on a rocshmem context
* * rocDecode/Picture file sample: Fixed a bug that can result in incorrect output frame count when reconfiguration occurs and with more than one iterations.
* * rocDecode: Updated change log.
[ROCm/rocdecode commit: 34cdd7f709]
* * rocDecode/Picture file sample: Fixed a bug that can result in incorrect output frame count when reconfiguration occurs and with more than one iterations.
* * rocDecode: Updated change log.
When using rocprof v3:
* Use --kernel-include-regex for kernel name filtering
* Use --kernel-iteration-range for kernel dispatch filtering
Update changelog
[ROCm/rocprofiler-compute commit: 45b8937d5d]
When using rocprof v3:
* Use --kernel-include-regex for kernel name filtering
* Use --kernel-iteration-range for kernel dispatch filtering
Update changelog
* rocprofv3 might not collect any counters for MI 100, handle this case gracefully to prevent test failures
[ROCm/rocprofiler-compute commit: 64ccd588de]
Poll the dependent signals twice on all gfx9.0 GPUs except gfx90a.
This is needed as a work-around for a rare issue where SDMA_POLL_REGMEM
may return before the memory is actually cleared.
Poll the dependent signals twice on all gfx9.0 GPUs except gfx90a.
This is needed as a work-around for a rare issue where SDMA_POLL_REGMEM
may return before the memory is actually cleared.
[ROCm/ROCR-Runtime commit: 6903a41b1d]
Set locale to C.utf8 instead of en_US.UTF-8
Avoid forcing the user to use en_US.UTF-8. Most Linux systems have C.utf8.
[ROCm/rocprofiler-compute commit: 96a25e8cbc]