Skip to content

GitLab

Commit 50467b06 authored by aska-0096's avatar aska-0096
Browse files

inital commit

parent 84044d78
Hide whitespace changes
Inline Side-by-side
Showing with 39 additions and 38 deletions
example/01_gemm/CMakeLists.txt
View file @ 50467b06
......@@ -25,6 +25,7 @@ add_example_dependencies(example_gemm_xdl example_gemm_xdl_fp16_v2)
add_example_executable(example_gemm_xdl_fp16_streamk_v3 gemm_xdl_fp16_streamk_v3.cpp)
add_example_dependencies(example_gemm_xdl example_gemm_xdl_fp16_streamk_v3)
add_example_executable(example_gemm_xdl_fp16_v3 gemm_xdl_fp16_v3.cpp)
target_compile_options(example_gemm_xdl_fp16_v3 PRIVATE -Wno-gnu-line-marker -save-temps=$PWD)
add_example_dependencies(example_gemm_xdl example_gemm_xdl_fp16_v3)
add_example_executable(example_gemm_xdl_fp8_v3 gemm_xdl_fp8_v3.cpp)
add_example_dependencies(example_gemm_xdl example_gemm_xdl_fp8_v3)
......
example/01_gemm/gemm_xdl_fp16_v3.cpp
View file @ 50467b06
......@@ -28,16 +28,16 @@ using DeviceGemmV2Instance =
ADataType, BDataType, CDataType, AccDataType, CShuffleDataType,
PassThrough, PassThrough, PassThrough, GemmDefault,
256,
224, 256,
64, 8, 2,
16, 16,
7, 8,
S<8, 32, 1>, S<1, 0, 2>, S<1, 0, 2>,
256, 256,
32, 8, 4,
32, 32,
4, 4,
S<4, 64, 1>, S<1, 0, 2>, S<1, 0, 2>,
2, 8, 8, 0,
S<8, 32, 1>, S<0, 2, 1>, S<0, 2, 1>,
1, 8, 2, 0,
1, 2, S<1, 32, 1, 8>, 8,
ck::BlockGemmPipelineScheduler::Intrawave,ck::BlockGemmPipelineVersion::v3>;
1, 8, 4, 0,
1, 1, S<1, 32, 1, 8>, 8,
ck::BlockGemmPipelineScheduler::Intrawave,ck::BlockGemmPipelineVersion::v1>;
// clang-format on
using ReferenceGemmInstance = ck::tensor_operation::host::
......
include/ck/tensor_operation/gpu/device/impl/device_gemm_xdl_cshuffle_v3.hpp
View file @ 50467b06
......@@ -206,24 +206,24 @@ struct DeviceGemm_Xdl_CShuffleV3 : public DeviceGemmV2<ALayout,
if constexpr(BlkGemmPipelineVer == BlockGemmPipelineVersion::v1 ||
BlkGemmPipelineVer == BlockGemmPipelineVersion::v3)
{
if(arg.KBatch > 1)
{
const auto kernel =
kernel_gemm_xdl_cshuffle_v3<GridwiseGemm,
true,
InMemoryDataOperationEnum::AtomicAdd,
minimum_occupancy>;
Run(kernel);
}
else
{
// if(arg.KBatch > 1)
// {
// const auto kernel =
// kernel_gemm_xdl_cshuffle_v3<GridwiseGemm,
// true,
// InMemoryDataOperationEnum::AtomicAdd,
// minimum_occupancy>;
// Run(kernel);
// }
// else
// {
const auto kernel =
kernel_gemm_xdl_cshuffle_v3<GridwiseGemm,
true,
InMemoryDataOperationEnum::Set,
minimum_occupancy>;
Run(kernel);
}
// }
}
// Tail number could be One to Seven
else if constexpr(BlkGemmPipelineVer == BlockGemmPipelineVersion::v2)
......@@ -558,24 +558,24 @@ struct DeviceGemm_Xdl_CShuffleV3 : public DeviceGemmV2<ALayout,
// Tail number always 1
if constexpr(BlkGemmPipelineVer == BlockGemmPipelineVersion::v1)
{
if(arg.KBatch > 1)
{
const auto kernel =
kernel_gemm_xdl_cshuffle_v3<GridwiseGemm,
false,
InMemoryDataOperationEnum::AtomicAdd,
minimum_occupancy>;
Run(kernel);
}
else
{
const auto kernel =
kernel_gemm_xdl_cshuffle_v3<GridwiseGemm,
false,
InMemoryDataOperationEnum::Set,
minimum_occupancy>;
Run(kernel);
}
// if(arg.KBatch > 1)
// {
// const auto kernel =
// kernel_gemm_xdl_cshuffle_v3<GridwiseGemm,
// false,
// InMemoryDataOperationEnum::AtomicAdd,
// minimum_occupancy>;
// Run(kernel);
// }
// else
// {
// const auto kernel =
// kernel_gemm_xdl_cshuffle_v3<GridwiseGemm,
// false,
// InMemoryDataOperationEnum::Set,
// minimum_occupancy>;
// Run(kernel);
// }
}
}
......
Markdown is supported
0% or .
You are about to add 0 people to the discussion. Proceed with caution.
Finish editing this message first!
Please register or to comment