Skip to content

[ck] refine wmma kernel build time #3517

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Draft
linqun wants to merge 1 commit into
base: develop
Choose a base branch
from
+368 −273
Draft

[ck] refine wmma kernel build time #3517

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 2 additions & 2 deletions example/35_splitK_gemm/gemm_wmma_splitk_reduce_multi_d_fp16.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -33,10 +33,10 @@ using DeviceGemmV2Instance =
ADataType, BDataType, DsDataType, CDataType, AccDataType, CShuffleDataType,
AElementOp, BElementOp, CDEElementOp, GemmDefault,
256,
128, 256, 64,
128, 128, 64,
8, 8,
16, 16,
4, 4,
4, 2,
S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>,
1, 1, 8, true,
S<4, 64, 1>, S<0, 2, 1>, S<0, 2, 1>,
Expand Down
6 changes: 3 additions & 3 deletions example/60_gemm_multi_ABD/gemm_multi_ABD_wmma_bias_fastgelu_bf16_i8.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -81,13 +81,13 @@ using DeviceOpInstance = ck::tensor_operation::device::DeviceGemmMultipleABD_Wmm
GemmSpec,
256,
128,
128,
64,
64,
8,
8,
16,
16,
4,
2,
2,
S<8, 32, 1>,
S<1, 0, 2>,
Expand All @@ -104,7 +104,7 @@ using DeviceOpInstance = ck::tensor_operation::device::DeviceGemmMultipleABD_Wmm
8,
0,
1,
1,
2,
S<1, 32, 1, 8>,
S<8, 8, 8>,
ck::BlockGemmPipelineScheduler::Intrawave,
Expand Down
2 changes: 1 addition & 1 deletion example/60_gemm_multi_ABD/gemm_multi_ABD_wmma_fastgelu_bf16_i8.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -106,7 +106,7 @@ using DeviceOpInstance = ck::tensor_operation::device::DeviceGemmMultipleABD_Wmm
S<1, 32, 1, 8>,
S<8, 8, 8>,
ck::BlockGemmPipelineScheduler::Intrawave,
ck::BlockGemmPipelineVersion::v3>;
ck::BlockGemmPipelineVersion::v1>;

int main(int argc, char* argv[])
{
Expand Down
2 changes: 1 addition & 1 deletion example/60_gemm_multi_ABD/gemm_multi_ABD_wmma_multiply_bias_fastgelu_bf16_i8.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -107,7 +107,7 @@ using DeviceOpInstance = ck::tensor_operation::device::DeviceGemmMultipleABD_Wmm
S<1, 32, 1, 8>,
S<8, 8, 8>,
ck::BlockGemmPipelineScheduler::Intrawave,
ck::BlockGemmPipelineVersion::v3>;
ck::BlockGemmPipelineVersion::v1>;

int main(int argc, char* argv[])
{
Expand Down
6 changes: 3 additions & 3 deletions example/65_gemm_multiply_multiply/gemm_multiply_multiply_wmma_fp16_bpreshuffle.cpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -65,12 +65,12 @@ using DeviceOpInstance =
A0DataType, B0DataType, DsDataType, EDataType, AccDataType, CShuffleDataType,
AElementOp, BElementOp, CDEElementOp, GemmSpec,
128,
32, 128, 128,
32, 128, 64,
8, 8,
16, 16,
2, 2,
S<16, 8, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0,
S<16, 8, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0,
S<8, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0,
S<8, 16, 1>, S<1, 0, 2>, S<1, 0, 2>, 2, 8, 8, 0,
1, 1, S<1, 16, 1, 8>, S<4, 4, 1>,
ck::BlockGemmPipelineScheduler::Intrawave,
ck::BlockGemmPipelineVersion::v1,
Expand Down
Loading