ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-20 14:59:17 +00:00
Code Issues Packages Projects Releases Wiki Activity

Enable gfx941 and gfx942 architectures. (#752)

Browse Source 
* enable gfx941/942 targets

* fix clang format

* fix the cmake logic for multiple targets

* fix cmake syntax for looping over targets

* add gfx941/942 support for gemm_xdl instances
This commit is contained in:
Illia Silin
2023-06-15 08:20:59 -07:00
committed by GitHub
parent 309b1c6461
commit 027e46ee82
76 changed files with 346 additions and 150 deletions
Download Patch File Download Diff File Expand all files Collapse all files

21
example/31_batched_gemm_gemm/CMakeLists.txt
View File

@@ -1,12 +1,23 @@
if(GPU_TARGETS MATCHES "gfx908" OR GPU_TARGETS MATCHES "gfx90a" OR GPU_TARGETS MATCHES "gfx940")
list(APPEND gpu_list1 gfx908 gfx90a gfx940 gfx941 gfx942)
list(APPEND gpu_list2 gfx908 gfx90a)
set(target 0)
foreach(gpu IN LISTS GPU_TARGETS)
if(gpu IN_LIST gpu_list1 AND target EQUAL 0)
add_example_executable(example_batched_gemm_gemm_xdl_fp32 batched_gemm_gemm_xdl_fp32.cpp)
add_example_executable(example_batched_gemm_gemm_xdl_fp16 batched_gemm_gemm_xdl_fp16.cpp)
add_example_executable(example_batched_gemm_gemm_xdl_bf16 batched_gemm_gemm_xdl_bf16.cpp)
if(NOT GPU_TARGETS MATCHES "gfx940")
add_example_executable(example_batched_gemm_gemm_xdl_int8 batched_gemm_gemm_xdl_int8.cpp)
endif()
if(USE_BITINT_EXTENSION_INT4)
add_example_executable(example_batched_gemm_gemm_xdl_int4 batched_gemm_gemm_xdl_int4.cpp)
endif(USE_BITINT_EXTENSION_INT4)
endif()
set(target 1)
endif()
endforeach()
set(target 0)
foreach(gpu IN LISTS GPU_TARGETS)
if(gpu IN_LIST gpu_list2 AND target EQUAL 0)
add_example_executable(example_batched_gemm_gemm_xdl_int8 batched_gemm_gemm_xdl_int8.cpp)
set(target 1)
endif()
endforeach()