120b8fba7ce730308b5a07faf530dcdc8705f541
SWDEV-77682 - Wave Limiter: Add __attribute__((amd_hint_waves_per_simd(n))) to FE for setting waves/simd hint per kernel. 0 for adaptive wave limitting. 1~10 for fixed waves/simd. Add query of WavesPerSimdHint metadata to compiler lib. HSAI path shouls use aclQueryInfo RT_WAVES_PER_SIMD_HINT to get the waves/simd hint value. This should provide a flexible workaround for bug SWDEV-56552 [CQE OCL][PERF][LNX][QR][G] 19% Perf drop is observed with chunkBp kernel in Carrizo. Affected files ... ... //depot/stg/opencl/drivers/opencl/compiler/clang/include/clang/Basic/Attr.td#9 edit ... //depot/stg/opencl/drivers/opencl/compiler/clang/lib/CodeGen/CodeGenFunction.cpp#16 edit ... //depot/stg/opencl/drivers/opencl/compiler/clang/lib/Sema/SemaDeclAttr.cpp#8 edit ... //depot/stg/opencl/drivers/opencl/compiler/clc/src/e2lModule.cpp#95 edit ... //depot/stg/opencl/drivers/opencl/compiler/clc/src/e2lModule.h#44 edit ... //depot/stg/opencl/drivers/opencl/compiler/clc/src/oclEntry.cpp#49 edit ... //depot/stg/opencl/drivers/opencl/compiler/clc/src/oclMeta.cpp#32 edit ... //depot/stg/opencl/drivers/opencl/compiler/clc/src/oclMeta.h#32 edit ... //depot/stg/opencl/drivers/opencl/compiler/edg/src/amd_ocl_attribute.c#26 edit ... //depot/stg/opencl/drivers/opencl/compiler/edg/src/attribute.c#29 edit ... //depot/stg/opencl/drivers/opencl/compiler/edg/src/err_codes.h#42 edit ... //depot/stg/opencl/drivers/opencl/compiler/edg/src/err_data.h#56 edit ... //depot/stg/opencl/drivers/opencl/compiler/edg/src/il_def.h#57 edit ... //depot/stg/opencl/drivers/opencl/compiler/edg/src/il_display.c#25 edit ... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/common/v0_8/if_acl.cpp#77 edit ... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/MDParser/AMDILMDInterface.h#3 edit ... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/MDParser/AMDILMDParser.l#4 edit ... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/MDParser/AMDILMDTypes.cpp#3 edit ... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/MDParser/lex.yy.cpp#5 edit ... //depot/stg/opencl/drivers/opencl/compiler/lib/backends/gpu/metadata.cpp#9 edit ... //depot/stg/opencl/drivers/opencl/compiler/lib/include/v0_8/aclEnums.h#23 edit ... //depot/stg/opencl/drivers/opencl/compiler/lib/include/v0_8/aclStructs.h#18 edit ... //depot/stg/opencl/drivers/opencl/compiler/llvm/include/llvm/Transforms/IPO/AMDKernelPerfHint.h#4 edit ... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Target/HSAIL/HSAILKernelManager.cpp#37 edit ... //depot/stg/opencl/drivers/opencl/compiler/llvm/lib/Transforms/IPO/AMDKernelPerfHint.cpp#6 edit ... //depot/stg/opencl/drivers/opencl/compiler/llvm32/include/llvm/Transforms/IPO/AMDKernelPerfHint.h#3 edit ... //depot/stg/opencl/drivers/opencl/compiler/llvm32/lib/Target/AMDIL/AMDILKernelManager.cpp#4 edit ... //depot/stg/opencl/drivers/opencl/compiler/llvm32/lib/Transforms/IPO/AMDKernelPerfHint.cpp#6 edit ... //depot/stg/opencl/drivers/opencl/tests/hsa/src/llc/opt/wave_limiter/explicit.cl#1 add ... //depot/stg/opencl/drivers/opencl/tests/hsa/tlst/llc_opt.tlst#57 edit ... //depot/stg/opencl/drivers/opencl/tests/ocltst/module/complib/CLAssumptionCheck.cpp#44 edit ... //depot/stg/opencl/drivers/opencl/tests/ocltst/module/complib/CLEnumCheck.cpp#48 edit
Popis
Nebyl uveden žádný popis
Jazyky
C++
67.5%
C
20.6%
Python
6.6%
CMake
3.4%
Shell
0.6%
Jiný
1.1%