SWDEV-2 - Change OpenCL version number from 2129 to 2130.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1876 edit
[ROCm/clr commit: 63118b2740]
SWDEV-2 - Change OpenCL version number from 2129 to 2130.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1876 edit
SWDEV-2 - Change OpenCL version number from 2128 to 2129.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1875 edit
[ROCm/clr commit: 9456725e52]
SWDEV-2 - Change OpenCL version number from 2128 to 2129.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1875 edit
SWDEV-2 - Change OpenCL version number from 2127 to 2128.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1874 edit
[ROCm/clr commit: acee9c2e2c]
SWDEV-2 - Change OpenCL version number from 2127 to 2128.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1874 edit
SWDEV-95095 - Bristol Ridge parts report their OpenCL device ID name as "Carrizo"
- Detect the ASIC revision for the name swap with Bristo Ridgel instead of Carrizo
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpucompiler.cpp#154 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#548 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#168 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/backend.h#13 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palcompiler.cpp#2 edit
[ROCm/clr commit: 9cbc946a1d]
SWDEV-95095 - Bristol Ridge parts report their OpenCL device ID name as "Carrizo"
- Detect the ASIC revision for the name swap with Bristo Ridgel instead of Carrizo
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpucompiler.cpp#154 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudevice.cpp#548 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#168 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/backend.h#13 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palcompiler.cpp#2 edit
SWDEV-79445 - Fix a TC regression after CL#1273440
- Add the both conditions to the restrictions: sorce and destination alignment
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#167 edit
[ROCm/clr commit: 23ec683890]
SWDEV-79445 - Fix a TC regression after CL#1273440
- Add the both conditions to the restrictions: sorce and destination alignment
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#167 edit
SWDEV-2 - Change OpenCL version number from 2126 to 2127.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1873 edit
[ROCm/clr commit: d9cea8af16]
SWDEV-2 - Change OpenCL version number from 2126 to 2127.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1873 edit
SWDEV-94881 - [OpenCL]increasing the size limit of image buffer to 2^27 which is for the moment at 65536
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudefs.hpp#136 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldefs.hpp#3 edit
[ROCm/clr commit: 624f331317]
SWDEV-94881 - [OpenCL]increasing the size limit of image buffer to 2^27 which is for the moment at 65536
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpudefs.hpp#136 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/paldefs.hpp#3 edit
SWDEV-93545 - HSA HLC: target option interface between complib and BE and denorm control refactoring
Global variables and associated options to control target GPU for optimizations and fp32 denorm support are removed.
Instead standard llvm -mcpu=<cpu> is used to pass chip family name and fp32 denorm is turned into a subtarget feature.
Subtarget feature can be set for llc as standard -mattr=+fp32-denormals and corresponding code to pass feature string
to the BE is added to the compiler lib, mimicing what we used to have for AMDIL.
Device name HSAIL metadata will now reflect an actual GPU family passed to the HSAIL BE instead of "generic".
Denorm support can be switched on as a feature bit in the target mapping. It is on starting from VI. However, just
switching this bit for a family will not produce denorm supporting code. The option -cl-denorms-are-zero can be used
to override this and runtime passes it for configs where CL_FP_DENORM is not reported.
Currently CL_FP_DENORM is not reported for any device, however it can be changed with AMD_GPU_FORCE_SINGLE_FP_DENORM
environment variable. If set it will be honored only starting from VI as set in the target mapping.
Implemented isFMAFasterThanFMulAndFAdd to handle use of v_fma_f32 on GFX9 instead of a direct chip family check.
Testing: smoke, precheckin
Reviewed by Evgeny Mankov
Affected files ...
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/codegen.cpp#70 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/linker.cpp#146 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/opt_level.cpp#30 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/libUtils.cpp#18 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/libUtils.h#27 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings.h#42 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_hsail.h#32 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_hsail64.h#27 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/BRIGAsmPrinter.cpp#155 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/BRIGAsmPrinter.h#69 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAIL.td#11 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILISelDAGToDAG.cpp#71 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILISelLowering.cpp#116 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILISelLowering.h#29 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILInstrInfo.cpp#42 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILInstrInfo.h#18 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILInstructions.td#22 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILKernelManager.cpp#55 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILSubtarget.cpp#14 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILSubtarget.h#15 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILTargetMachine.cpp#57 edit
... //depot/stg/opencl/drivers/opencl/tests/hsa/bin/test_driver.pl#29 edit
... //depot/stg/opencl/drivers/opencl/tests/hsa/tlst/llc_opt.tlst#97 edit
... //depot/stg/opencl/drivers/opencl/tests/hsa/tlst/ocl_features.tlst#58 edit
[ROCm/clr commit: 712c0c4009]
SWDEV-93545 - HSA HLC: target option interface between complib and BE and denorm control refactoring
Global variables and associated options to control target GPU for optimizations and fp32 denorm support are removed.
Instead standard llvm -mcpu=<cpu> is used to pass chip family name and fp32 denorm is turned into a subtarget feature.
Subtarget feature can be set for llc as standard -mattr=+fp32-denormals and corresponding code to pass feature string
to the BE is added to the compiler lib, mimicing what we used to have for AMDIL.
Device name HSAIL metadata will now reflect an actual GPU family passed to the HSAIL BE instead of "generic".
Denorm support can be switched on as a feature bit in the target mapping. It is on starting from VI. However, just
switching this bit for a family will not produce denorm supporting code. The option -cl-denorms-are-zero can be used
to override this and runtime passes it for configs where CL_FP_DENORM is not reported.
Currently CL_FP_DENORM is not reported for any device, however it can be changed with AMD_GPU_FORCE_SINGLE_FP_DENORM
environment variable. If set it will be honored only starting from VI as set in the target mapping.
Implemented isFMAFasterThanFMulAndFAdd to handle use of v_fma_f32 on GFX9 instead of a direct chip family check.
Testing: smoke, precheckin
Reviewed by Evgeny Mankov
Affected files ...
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/codegen.cpp#70 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/linker.cpp#146 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/opt_level.cpp#30 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/libUtils.cpp#18 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/libUtils.h#27 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings.h#42 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_hsail.h#32 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/utils/v0_8/target_mappings_hsail64.h#27 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/BRIGAsmPrinter.cpp#155 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/BRIGAsmPrinter.h#69 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAIL.td#11 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILISelDAGToDAG.cpp#71 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILISelLowering.cpp#116 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILISelLowering.h#29 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILInstrInfo.cpp#42 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILInstrInfo.h#18 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILInstructions.td#22 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILKernelManager.cpp#55 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILSubtarget.cpp#14 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILSubtarget.h#15 edit
... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILTargetMachine.cpp#57 edit
... //depot/stg/opencl/drivers/opencl/tests/hsa/bin/test_driver.pl#29 edit
... //depot/stg/opencl/drivers/opencl/tests/hsa/tlst/llc_opt.tlst#97 edit
... //depot/stg/opencl/drivers/opencl/tests/hsa/tlst/ocl_features.tlst#58 edit
SWDEV-94161 - [Adobe Photoshop] TDR is observed while applying Blur effect on the selected portion of an Image
- Restore back the original condition for SDMA restriction (CL#1195487). The root cause wasn't identified and probably somewhere in GSL, since PAL backend works fine.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#166 edit
[ROCm/clr commit: 24b23e3f72]
SWDEV-94161 - [Adobe Photoshop] TDR is observed while applying Blur effect on the selected portion of an Image
- Restore back the original condition for SDMA restriction (CL#1195487). The root cause wasn't identified and probably somewhere in GSL, since PAL backend works fine.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gslbe/src/rt/GSLDevice.cpp#166 edit
SWDEV-85602 - HSA HLC: serialize linking
Since we do not have build referenced symbol map now during linking we do not have to keep all library modules
in memory during the whole link process. This change serializes linking to save some memory.
Testing: smoke, precheckin
Reviewed by Evgeny Mankov
Affected files ...
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/linker.cpp#145 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/linker/include/AMDResolveLinker.h#5 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/linker/lib/AMDResolveLinker.cpp#5 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/linker/tools/opencl-link/opencl-link.cpp#7 edit
[ROCm/clr commit: 1161118bc3]
SWDEV-85602 - HSA HLC: serialize linking
Since we do not have build referenced symbol map now during linking we do not have to keep all library modules
in memory during the whole link process. This change serializes linking to save some memory.
Testing: smoke, precheckin
Reviewed by Evgeny Mankov
Affected files ...
... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/linker.cpp#145 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/linker/include/AMDResolveLinker.h#5 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/linker/lib/AMDResolveLinker.cpp#5 edit
... //depot/stg/opencl/drivers/opencl/compiler/lib/linker/tools/opencl-link/opencl-link.cpp#7 edit
SWDEV-2 - Change OpenCL version number from 2125 to 2126.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1872 edit
[ROCm/clr commit: 06ef7ffe6e]
SWDEV-2 - Change OpenCL version number from 2125 to 2126.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1872 edit
Squashed commit of the following:
commit bc44bcee461e46b0cf5cb9fe09213dca450b081a
Author: Daniil Fukalov <daniil.fukalov@amd.com>
Date: Mon May 16 20:50:05 2016 +0300
added note about errors without CUDA sdk installed
commit 5fd73ba90c0940bdc977737894362a99f4232b56
Author: Daniil Fukalov <daniil.fukalov@amd.com>
Date: Mon May 16 20:31:47 2016 +0300
move clang-hipify info to its own README
commit 21d81a6d5acd3f093d77ac4d584e6f5bbe48f8cc
Author: Daniil Fukalov <daniil.fukalov@amd.com>
Date: Mon May 16 20:30:00 2016 +0300
initial version
Change-Id: I157294699a7be3d0bb38b2ee4a137a94280529c9
Squashed commit of the following:
commit bc44bcee461e46b0cf5cb9fe09213dca450b081a
Author: Daniil Fukalov <daniil.fukalov@amd.com>
Date: Mon May 16 20:50:05 2016 +0300
added note about errors without CUDA sdk installed
commit 5fd73ba90c0940bdc977737894362a99f4232b56
Author: Daniil Fukalov <daniil.fukalov@amd.com>
Date: Mon May 16 20:31:47 2016 +0300
move clang-hipify info to its own README
commit 21d81a6d5acd3f093d77ac4d584e6f5bbe48f8cc
Author: Daniil Fukalov <daniil.fukalov@amd.com>
Date: Mon May 16 20:30:00 2016 +0300
initial version
Change-Id: I157294699a7be3d0bb38b2ee4a137a94280529c9
SWDEV-2 - Change OpenCL version number from 2124 to 2125.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1871 edit
[ROCm/clr commit: 112e68f217]
SWDEV-2 - Change OpenCL version number from 2124 to 2125.
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/utils/versions.hpp#1871 edit
SWDEV-94524 - [OCL] Call to clBuildProgram crashes amdocl64.dll
- Detect offline device and return a fake srd handle. Compiler must reload binary for a specific device in order for the binary to have proper functionality
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuprogram.cpp#226 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#5 edit
[ROCm/clr commit: 4c47086ac4]
SWDEV-94524 - [OCL] Call to clBuildProgram crashes amdocl64.dll
- Detect offline device and return a fake srd handle. Compiler must reload binary for a specific device in order for the binary to have proper functionality
Affected files ...
... //depot/stg/opencl/drivers/opencl/runtime/device/gpu/gpuprogram.cpp#226 edit
... //depot/stg/opencl/drivers/opencl/runtime/device/pal/palprogram.cpp#5 edit
foreman
Konstantin Zhuravlyov
Besar Wicaksono
James Edwards
Maneesh Gupta
Ramesh Errabolu
Rahul Garg