Skip to content

[CUBLAS] Set fp32 compute and scale dtypes in fp16 matmul #16892

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

Merged
vinx13 merged 1 commit into from
Apr 16, 2024
Merged

[CUBLAS] Set fp32 compute and scale dtypes in fp16 matmul #16892

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
6 changes: 0 additions & 6 deletions src/runtime/contrib/cublas/cublas.cc
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -150,8 +150,6 @@ void CallCublasLt(cublasLtHandle_t hdl, cudaStream_t stream,
cudaDataType_t c_type = CUDA_R_32F;
float one_fp32 = 1.0;
float zero_fp32 = 0.0;
auto one_fp16 = __truncXfYf2__<float, uint32_t, 23, uint16_t, uint16_t, 10>(1.0);
auto zero_fp16 = __truncXfYf2__<float, uint32_t, 23, uint16_t, uint16_t, 10>(0.0);
int32_t one_i32 = 1;
int32_t zero_i32 = 0;
void* alpha = &one_fp32;
Expand All @@ -165,10 +163,6 @@ void CallCublasLt(cublasLtHandle_t hdl, cudaStream_t stream,

if (TypeMatch(C->dtype, kDLFloat, 16)) {
c_type = CUDA_R_16F;
compute_type = CUBLAS_COMPUTE_16F;
scale_type = CUDA_R_16F;
alpha = &one_fp16;
beta = &zero_fp16;
} else if (TypeMatch(C->dtype, kDLInt, 32)) {
c_type = CUDA_R_32I;
compute_type = CUBLAS_COMPUTE_32I;
Expand Down