Merge branch 'develop' into lwpck-929

test/elementwise_normalization/CMakeLists.txt

-if(DTYPES MATCHES "fp16" OR NOT DEFINED DTYPES)
-  add_custom_target(test_elementwise_normalization)
-  add_gtest_executable(test_elementwise_layernorm_fp16 test_elementwise_layernorm_fp16.cpp)
+add_custom_target(test_elementwise_normalization)
+add_gtest_executable(test_elementwise_layernorm_fp16 test_elementwise_layernorm_fp16.cpp)
+if(result EQUAL 0)
   target_link_libraries(test_elementwise_layernorm_fp16 PRIVATE utility device_elementwise_normalization_instance)
   add_dependencies(test_elementwise_normalization test_elementwise_layernorm_fp16)
 endif()
\ No newline at end of file

test/gemm/CMakeLists.txt

-if(DTYPES MATCHES "fp32" OR NOT DEFINED DTYPES)
 add_test_executable(test_gemm_fp32 gemm_fp32.cpp)
-target_link_libraries(test_gemm_fp32 PRIVATE utility)
-target_link_libraries(test_gemm_fp32 PRIVATE device_gemm_instance)
+if(result EQUAL 0)
+    target_link_libraries(test_gemm_fp32 PRIVATE utility device_gemm_instance)
 endif()
-if(DTYPES MATCHES "fp16" OR NOT DEFINED DTYPES)
 add_test_executable(test_gemm_fp16 gemm_fp16.cpp)
-target_link_libraries(test_gemm_fp16 PRIVATE utility)
-target_link_libraries(test_gemm_fp16 PRIVATE device_gemm_instance)
-add_library(gemm_standalone_xdl_fp16_instances STATIC
+if(result EQUAL 0)
+    target_link_libraries(test_gemm_fp16 PRIVATE utility device_gemm_instance)
+    add_library(gemm_standalone_xdl_fp16_instances STATIC
     instance/gemm_f16_nn_instance.cpp
     instance/gemm_f16_nt_instance.cpp
     instance/gemm_f16_tn_instance.cpp
     instance/gemm_wavelet_f16_tn_instance.cpp
     instance/gemm_f16_tt_instance.cpp
-)
+    )
+endif()
 add_test_executable(test_gemm_standalone_xdl_fp16 gemm_standalone_xdl_fp16.cpp)
-target_link_libraries(test_gemm_standalone_xdl_fp16 PRIVATE gemm_standalone_xdl_fp16_instances utility)
-target_include_directories(test_gemm_standalone_xdl_fp16 PRIVATE instance/)
+if(result EQUAL 0)
+    target_link_libraries(test_gemm_standalone_xdl_fp16 PRIVATE gemm_standalone_xdl_fp16_instances utility)
+    target_include_directories(test_gemm_standalone_xdl_fp16 PRIVATE instance/)
 endif()
-if(DTYPES MATCHES "bf16" OR NOT DEFINED DTYPES)
 add_test_executable(test_gemm_bf16 gemm_bf16.cpp)
-target_link_libraries(test_gemm_bf16 PRIVATE utility)
-target_link_libraries(test_gemm_bf16 PRIVATE device_gemm_instance)
+if(result EQUAL 0)
+    target_link_libraries(test_gemm_bf16 PRIVATE utility device_gemm_instance)
 endif()
-if(DTYPES MATCHES "int8" OR NOT DEFINED DTYPES)
 add_test_executable(test_gemm_int8 gemm_int8.cpp)
-target_link_libraries(test_gemm_int8 PRIVATE utility)
-target_link_libraries(test_gemm_int8 PRIVATE device_gemm_instance)
+if(result EQUAL 0)
+    target_link_libraries(test_gemm_int8 PRIVATE utility device_gemm_instance)
 endif()
\ No newline at end of file

test/gemm_layernorm/CMakeLists.txt

@@ -2,12 +2,12 @@ list(APPEND gpu_list gfx908 gfx90a gfx940 gfx941 gfx942)
 set(target 0)
 foreach(gpu IN LISTS GPU_TARGETS)
  if(gpu IN_LIST gpu_list AND target EQUAL 0)
-  if(DTYPES MATCHES "fp16" OR NOT DEFINED DTYPES)
    add_custom_target(test_gemm_layernorm)
    add_gtest_executable(test_gemm_add_relu_add_layernorm_fp16 test_gemm_add_relu_add_layernorm_fp16.cpp)
-   target_link_libraries(test_gemm_add_relu_add_layernorm_fp16 PRIVATE utility device_gemm_add_relu_add_layernorm_instance)
-   add_dependencies(test_gemm_layernorm test_gemm_add_relu_add_layernorm_fp16)
-   set(target 1)
-  endif()
+   if(result EQUAL 0)
+     target_link_libraries(test_gemm_add_relu_add_layernorm_fp16 PRIVATE utility device_gemm_add_relu_add_layernorm_instance)
+     add_dependencies(test_gemm_layernorm test_gemm_add_relu_add_layernorm_fp16)
+     set(target 1)
+   endif()
  endif()
 endforeach()

test/gemm_reduce/CMakeLists.txt

-if(DTYPES MATCHES "fp16" OR NOT DEFINED DTYPES)
-  add_test_executable(test_gemm_reduce_fp16 gemm_reduce_fp16.cpp)
-  target_link_libraries(test_gemm_reduce_fp16 PRIVATE utility)
-  target_link_libraries(test_gemm_reduce_fp16 PRIVATE device_gemm_reduce_instance)
+add_test_executable(test_gemm_reduce_fp16 gemm_reduce_fp16.cpp)
+if(result EQUAL 0)
+  target_link_libraries(test_gemm_reduce_fp16 PRIVATE utility device_gemm_reduce_instance)
 endif()
\ No newline at end of file

test/grouped_convnd_bwd_weight/test_grouped_convnd_bwd_weight.cpp

@@ -33,8 +33,8 @@ class TestGroupedConvndBwdWeight : public ::testing::Test
 
     bool skip_case(const ck::utils::conv::ConvParam& params, const ck::index_t split_k)
     {
-        // K or C are odd is supported only by DL kernel (only applies to fp16)
-        // DL kernel is only supported for split_k=1
+        // Odd K or C values are supported only by DL kernel (only applies to fp16)
+        // DL kernel currently supports only `split_k=1`
         if constexpr(std::is_same_v<InDataType, ck::half_t>)
         {
             if(split_k != 1 && (params.K_ % 2 != 0 || params.C_ % 2 != 0))

test/grouped_gemm/CMakeLists.txt

-if(DTYPES MATCHES "fp16" OR NOT DEFINED DTYPES)
 list(APPEND gpu_list gfx908 gfx90a gfx940 gfx941 gfx942)
 set(target 0)
 foreach(gpu IN LISTS GPU_TARGETS)
@@ -13,4 +12,3 @@ foreach(gpu IN LISTS GPU_TARGETS)
    set(target 1)
  endif()
 endforeach()
-endif()

test/normalization/CMakeLists.txt

-if(DTYPES MATCHES "fp16" OR DTYPES MATCHES "fp32" OR NOT DEFINED DTYPES)
-  add_custom_target(test_normalization)
-endif()
-if(DTYPES MATCHES "fp32" OR NOT DEFINED DTYPES)
-  add_gtest_executable(test_layernorm2d_fp32 test_layernorm2d_fp32.cpp)
-  add_gtest_executable(test_groupnorm_fp32 test_groupnorm_fp32.cpp)
+add_custom_target(test_normalization)
+add_gtest_executable(test_layernorm2d_fp32 test_layernorm2d_fp32.cpp)
+if(result EQUAL 0)
   target_link_libraries(test_layernorm2d_fp32 PRIVATE utility device_normalization_instance)
-  target_link_libraries(test_groupnorm_fp32 PRIVATE utility device_normalization_instance)
   add_dependencies(test_normalization test_layernorm2d_fp32)
+endif()
+add_gtest_executable(test_groupnorm_fp32 test_groupnorm_fp32.cpp)
+if(result EQUAL 0)
+  target_link_libraries(test_groupnorm_fp32 PRIVATE utility device_normalization_instance)
   add_dependencies(test_normalization test_groupnorm_fp32)
 endif()
-if(DTYPES MATCHES "fp16" OR NOT DEFINED DTYPES)
-  add_gtest_executable(test_layernorm2d_fp16 test_layernorm2d_fp16.cpp)
-  add_gtest_executable(test_groupnorm_fp16 test_groupnorm_fp16.cpp)
+add_gtest_executable(test_layernorm2d_fp16 test_layernorm2d_fp16.cpp)
+if(result EQUAL 0)
   target_link_libraries(test_layernorm2d_fp16 PRIVATE utility device_normalization_instance)
-  target_link_libraries(test_groupnorm_fp16 PRIVATE utility device_normalization_instance)
   add_dependencies(test_normalization test_layernorm2d_fp16)
+endif()
+add_gtest_executable(test_groupnorm_fp16 test_groupnorm_fp16.cpp)
+if(result EQUAL 0)
+  target_link_libraries(test_groupnorm_fp16 PRIVATE utility device_normalization_instance)
   add_dependencies(test_normalization test_groupnorm_fp16)
 endif()

test/reduce/CMakeLists.txt

 add_test_executable(test_reduce_no_index reduce_no_index.cpp)
 add_test_executable(test_reduce_with_index reduce_with_index.cpp)
-target_link_libraries(test_reduce_no_index PRIVATE utility)
-target_link_libraries(test_reduce_no_index PRIVATE device_reduce_instance)
-target_link_libraries(test_reduce_with_index PRIVATE utility)
-target_link_libraries(test_reduce_with_index PRIVATE device_reduce_instance)
+target_link_libraries(test_reduce_no_index PRIVATE utility device_reduce_instance)
+target_link_libraries(test_reduce_with_index PRIVATE utility device_reduce_instance)