SWDEV-79445 - OCL generic changes and code clean-up
- Improve image fill performance with multiple writes in a single thread. The current split has 3 regions
Affected files ...
... //depot/stg/opencl/drivers/opencl/library/common.hsa/src/blitKernels.cl#4 edit
... //depot/stg/opencl/drivers/opencl/library/common/src/blitKernels.cl#4 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpublit.cpp#123 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpublit.hpp#40 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palblit.cpp#8 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palblit.hpp#4 edit
SWDEV-94610 - Restore the amdgpu_metadata.[ch]pp files. We need to share these files between different projects, and should avoid branching them. Ideally, they would be part of a metadata utility library.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/amdgpu_metadata.cpp#1 add
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/amdgpu_metadata.hpp#1 branch
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.cpp#9 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmetadata.cpp#3 delete
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmetadata.hpp#4 delete
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.hpp#11 edit
[ROCm/clr commit: ee978692f5]
SWDEV-94610 - Restore the amdgpu_metadata.[ch]pp files. We need to share these files between different projects, and should avoid branching them. Ideally, they would be part of a metadata utility library.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/amdgpu_metadata.cpp#1 add
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/amdgpu_metadata.hpp#1 branch
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rockernel.cpp#9 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmetadata.cpp#3 delete
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocmetadata.hpp#4 delete
... //depot/stg/opencl/drivers/opencl/runtime/device/rocm/rocprogram.hpp#11 edit
enum CUresult was merged with enum cudaError_t into single hipError_t.
Thus a majority of HIP error codes has a reflection to Driver's and RT's corresponding error code at the same time. For instance:
cuda2hipRename["CUDA_SUCCESS"] = {"hipSuccess", CONV_ERR, API_DRIVER};
cuda2hipRename["cudaSuccess"] = {"hipSuccess", CONV_ERR, API_RUNTIME};
There are a few CUDA return error codes which are RT or Driver specific. For instance:
cuda2hipRename["CUDA_ERROR_INVALID_CONTEXT"] = {"hipErrorInvalidContext", CONV_ERR, API_DRIVER};
cuda2hipRename["cudaErrorInvalidMemcpyDirection"] = {"hipErrorInvalidMemcpyDirection", CONV_ERR, API_RUNTIME};
Matchers were changed from "cuda.* | cublas.*" to "cu.*" as CUDA API functions/types starts with 'cu'.
[ROCm/clr commit: 22dca6794d]
enum CUresult was merged with enum cudaError_t into single hipError_t.
Thus a majority of HIP error codes has a reflection to Driver's and RT's corresponding error code at the same time. For instance:
cuda2hipRename["CUDA_SUCCESS"] = {"hipSuccess", CONV_ERR, API_DRIVER};
cuda2hipRename["cudaSuccess"] = {"hipSuccess", CONV_ERR, API_RUNTIME};
There are a few CUDA return error codes which are RT or Driver specific. For instance:
cuda2hipRename["CUDA_ERROR_INVALID_CONTEXT"] = {"hipErrorInvalidContext", CONV_ERR, API_DRIVER};
cuda2hipRename["cudaErrorInvalidMemcpyDirection"] = {"hipErrorInvalidMemcpyDirection", CONV_ERR, API_RUNTIME};
Matchers were changed from "cuda.* | cublas.*" to "cu.*" as CUDA API functions/types starts with 'cu'.
enum CUresult was merged with enum cudaError_t into single hipError_t.
Thus a majority of HIP error codes has a reflection to Driver's and RT's corresponding error code at the same time. For instance:
cuda2hipRename["CUDA_SUCCESS"] = {"hipSuccess", CONV_ERR, API_DRIVER};
cuda2hipRename["cudaSuccess"] = {"hipSuccess", CONV_ERR, API_RUNTIME};
There are a few CUDA return error codes which are RT or Driver specific. For instance:
cuda2hipRename["CUDA_ERROR_INVALID_CONTEXT"] = {"hipErrorInvalidContext", CONV_ERR, API_DRIVER};
cuda2hipRename["cudaErrorInvalidMemcpyDirection"] = {"hipErrorInvalidMemcpyDirection", CONV_ERR, API_RUNTIME};
Matchers were changed from "cuda.* | cublas.*" to "cu.*" as CUDA API functions/types starts with 'cu'.
[ROCm/hip commit: 43998387aa]
enum CUresult was merged with enum cudaError_t into single hipError_t.
Thus a majority of HIP error codes has a reflection to Driver's and RT's corresponding error code at the same time. For instance:
cuda2hipRename["CUDA_SUCCESS"] = {"hipSuccess", CONV_ERR, API_DRIVER};
cuda2hipRename["cudaSuccess"] = {"hipSuccess", CONV_ERR, API_RUNTIME};
There are a few CUDA return error codes which are RT or Driver specific. For instance:
cuda2hipRename["CUDA_ERROR_INVALID_CONTEXT"] = {"hipErrorInvalidContext", CONV_ERR, API_DRIVER};
cuda2hipRename["cudaErrorInvalidMemcpyDirection"] = {"hipErrorInvalidMemcpyDirection", CONV_ERR, API_RUNTIME};
Matchers were changed from "cuda.* | cublas.*" to "cu.*" as CUDA API functions/types starts with 'cu'.
1. Removed vcpy_isa.ptx as it should be generated during make
2. Made argument padding specific to hcc path
3. Renamed --gencodeobject to --genco
4. Changed Makefile to work on both nvcc and hcc path
Change-Id: Ifd053d541085d9ce4fd37bc21b07674786c7163e
[ROCm/hip-tests commit: ebc17b0d6e]
1. Removed vcpy_isa.ptx as it should be generated during make
2. Made argument padding specific to hcc path
3. Renamed --gencodeobject to --genco
4. Changed Makefile to work on both nvcc and hcc path
Change-Id: Ifd053d541085d9ce4fd37bc21b07674786c7163e
1. Removed vcpy_isa.ptx as it should be generated during make
2. Made argument padding specific to hcc path
3. Renamed --gencodeobject to --genco
4. Changed Makefile to work on both nvcc and hcc path
Change-Id: Ifd053d541085d9ce4fd37bc21b07674786c7163e
[ROCm/clr commit: f22fda1291]
1. Removed vcpy_isa.ptx as it should be generated during make
2. Made argument padding specific to hcc path
3. Renamed --gencodeobject to --genco
4. Changed Makefile to work on both nvcc and hcc path
Change-Id: Ifd053d541085d9ce4fd37bc21b07674786c7163e
1. Removed vcpy_isa.ptx as it should be generated during make
2. Made argument padding specific to hcc path
3. Renamed --gencodeobject to --genco
4. Changed Makefile to work on both nvcc and hcc path
Change-Id: Ifd053d541085d9ce4fd37bc21b07674786c7163e
[ROCm/hip commit: cc8ce4d58c]
1. Removed vcpy_isa.ptx as it should be generated during make
2. Made argument padding specific to hcc path
3. Renamed --gencodeobject to --genco
4. Changed Makefile to work on both nvcc and hcc path
Change-Id: Ifd053d541085d9ce4fd37bc21b07674786c7163e
- hipcc honors HIPCC_VERBOSE for gencodeobject too
- fixed gencodeobject generation logic for nvcc path
- hipgenisa.sh is no longer passed the --gencodeobject flag
Change-Id: Iab83c18a6decef445fda8468ba5de10cd7935046
[ROCm/clr commit: 1446ad3b92]
- hipcc honors HIPCC_VERBOSE for gencodeobject too
- fixed gencodeobject generation logic for nvcc path
- hipgenisa.sh is no longer passed the --gencodeobject flag
Change-Id: Iab83c18a6decef445fda8468ba5de10cd7935046
- hipcc honors HIPCC_VERBOSE for gencodeobject too
- fixed gencodeobject generation logic for nvcc path
- hipgenisa.sh is no longer passed the --gencodeobject flag
Change-Id: Iab83c18a6decef445fda8468ba5de10cd7935046
[ROCm/hip commit: 14b0fc80c7]
- hipcc honors HIPCC_VERBOSE for gencodeobject too
- fixed gencodeobject generation logic for nvcc path
- hipgenisa.sh is no longer passed the --gencodeobject flag
Change-Id: Iab83c18a6decef445fda8468ba5de10cd7935046
SWDEV-2 - Change OpenCL version number from 2213 to 2214.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1960 edit
[ROCm/clr commit: 51b56820d8]
SWDEV-2 - Change OpenCL version number from 2213 to 2214.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1960 edit
foreman
Evgeny Mankov
Aditya Atluri
Maneesh Gupta
Rahul Garg