* SWDEV-533237 Add initial support for hipOccupancyAvailableDynamicSMemPerBlock API
* SWDEV-533237 Add hipOccupancyAvailableDynamicSMemPerBlock wrapper for nvidia
* SWDEV-533237 Add implementation of hipOccupancyAvailableDynamicSMemPerBlock API
* SWDEV-533237 Add LDSAlignment field in Isa table
---------
Co-authored-by: Rahul Manocha <rmanocha@amd.com>
* Add examples to tools folder
* Correct P2P memory access section
* Sync poriting guide
* Add HIP Graph tutorial
* Add hint about using amdgpu-dkms for IPC API
* Add a few more env variables
* Added amd-smi set --pcie command
* Removed current pcie level due to it not being static
* Added pcie information to static --bus
---------
Signed-off-by: Pham, Gabriel <Gabriel.Pham@amd.com>
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
* Added amd-smi set --pcie command
* Removed current pcie level due to it not being static
* Added pcie information to static --bus
---------
Signed-off-by: Pham, Gabriel <Gabriel.Pham@amd.com>
Signed-off-by: Maisam Arif <Maisam.Arif@amd.com>
[ROCm/amdsmi commit: 9e3537d778]
- **Added evicted_time metric for kfd processes**.
- Time that queues are evicted on a GPU in milliseconds
- Added to CLI in `amd-smi monitor -q` and `amd-smi process`
- Added to C API and Python API:
- amdsmi_get_gpu_process_list()
- amdsmi_get_gpu_compute_process_info()
- amdsmi_get_gpu_compute_process_info_by_pid()
---------
Signed-off-by: Pryor, Adam <Adam.Pryor@amd.com>
- **Added evicted_time metric for kfd processes**.
- Time that queues are evicted on a GPU in milliseconds
- Added to CLI in `amd-smi monitor -q` and `amd-smi process`
- Added to C API and Python API:
- amdsmi_get_gpu_process_list()
- amdsmi_get_gpu_compute_process_info()
- amdsmi_get_gpu_compute_process_info_by_pid()
---------
Signed-off-by: Pryor, Adam <Adam.Pryor@amd.com>
[ROCm/amdsmi commit: 2144cfbba4]
somehow the test whether we requested MPI support or not stopped
working, although no obvious code change can be located.
Make the if-statement more stringent by explicitely testing whether
USE_MPI_SUPPORT is "ON".
[ROCm/rocshmem commit: c0285ac0ce]
somehow the test whether we requested MPI support or not stopped
working, although no obvious code change can be located.
Make the if-statement more stringent by explicitely testing whether
USE_MPI_SUPPORT is "ON".
* Implements casting key loads and stores to address_space(1) so that vector global load and store instructions are emitted by the compiler instead of more costly flat loads and stores
* Removes nontemporality from some key stores for gfx950.
[ROCm/rccl commit: e69b11eba5]
* Implements casting key loads and stores to address_space(1) so that vector global load and store instructions are emitted by the compiler instead of more costly flat loads and stores
* Removes nontemporality from some key stores for gfx950.
* Updated CODEOWNERS to instead use RCCL-Reviewers team
* Apply suggestion from @nileshnegi
Co-authored-by: Nilesh M Negi <Nilesh.Negi@amd.com>
---------
Co-authored-by: Nilesh M Negi <Nilesh.Negi@amd.com>
[ROCm/rccl commit: f290e302d3]
* Updated CODEOWNERS to instead use RCCL-Reviewers team
* Apply suggestion from @nileshnegi
Co-authored-by: Nilesh M Negi <Nilesh.Negi@amd.com>
---------
Co-authored-by: Nilesh M Negi <Nilesh.Negi@amd.com>
- `amdgpu-install` is no longer recommended. Link to separate driver
installation docs.
- add verify step
- update readme
- add package info
Signed-off-by: Park, Peter <Peter.Park@amd.com>
- `amdgpu-install` is no longer recommended. Link to separate driver
installation docs.
- add verify step
- update readme
- add package info
Signed-off-by: Park, Peter <Peter.Park@amd.com>
[ROCm/amdsmi commit: 12fb58c30b]
* Updates:
- [ASAN] GCC does not support `-shared-libsan flags`, so removed this one
- [Clang] Fixed refernces to local binding errors (name collision)
& other strict scope/structure/lamda binding errors
- [Clang] Fix rsmi_wrapper error: \"error: missing default argument on parameter \'args\'\"
- [ASAN] Fixed stack-buffer-overflow found in
`amdsmi_get_gpu_accelerator_partition_profile()`
Change-Id: I854007efb75d828dbb8088c0d56dbc125081f0f2
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
* Updates:
- [ASAN] GCC does not support `-shared-libsan flags`, so removed this one
- [Clang] Fixed refernces to local binding errors (name collision)
& other strict scope/structure/lamda binding errors
- [Clang] Fix rsmi_wrapper error: \"error: missing default argument on parameter \'args\'\"
- [ASAN] Fixed stack-buffer-overflow found in
`amdsmi_get_gpu_accelerator_partition_profile()`
Change-Id: I854007efb75d828dbb8088c0d56dbc125081f0f2
Signed-off-by: Charis Poag <Charis.Poag@amd.com>
[ROCm/amdsmi commit: 00a04f5810]
* SWDEV-546311 - implement hipKernelGetLibrary & hipLibraryEnumerateKernels API
* Fix for LibraryEnumerateKernel and KernelGetName
* Update Enumerate Kernels to handle 0 numKernels
* Minor fixes to function names
* fix error checking in internal function
* Update changelog for new apis
---------
Co-authored-by: Rahul Manocha <rmanocha@amd.com>
* Solved error: [Errno 2] No such file or directory: '/usr/local/lib/python3.10/dist-packages/argcomplete/bash_completion.d/python-argcomplete.sh'
[ROCm/amdsmi commit: 013d6cb511]
xuchen-amd
dsicarov-amd
Istvan Kiss
Geo Min
Pham, Gabriel
Pryor, Adam
corey-derochie-amd
Ajay GunaShekar
Atul Kulkarni
Edgar Gabriel
alex-breslow-amd
David Galiffi
Venkateshwar Reddy Kandula
Gopesh Bhardwaj
Swati Rawat
akolliasAMD
ywang103-amd
dependabot[bot]
Park, Peter
Charis Poag
systems-assistant[bot]
Rahul Manocha
Yiltan
Shadi Dashmiz
Narlo, Joseph
David Yat Sin
Saeed, Oosman
AL Musaffar, Yazen
Billakanti, Koushik