leavelet/cutlass
1
0
Fork 0
Code Issues Pull Requests Actions 29 Packages Projects Releases Wiki Activity

Adding more Threadblock Tiles for Mixed-input TensorOp (BF16 * S8) in cutlass_library (#1132)

Browse Source 
* Adding more tiles in the cutlass_library for mixed-input support.

* fix rebase issue

* more tiles to upcast a
This commit is contained in:
Manish Gupta
2023-10-13 11:33:15 -04:00
committed by GitHub
parent fa8dfe631f
commit 757275f279
14 changed files with 830 additions and 20 deletions
Download Patch File Download Diff File Expand all files Collapse all files
tools/library/src/reduction/init_reduction_operations.cu
+2
View File
@@ -44,6 +44,7 @@ namespace library {
///////////////////////////////////////////////////////////////////////////////////////////////
void initialize_reduce_add_linear_combination_f16_f16_f16(Manifest &manifest);
void initialize_reduce_add_linear_combination_f32_f32_f16(Manifest &manifest);
void initialize_reduce_add_linear_combination_f32_f32_bf16(Manifest &manifest);
void initialize_reduce_add_linear_combination_f32_f32_f32(Manifest &manifest);
void initialize_reduce_add_linear_combination_f64_f64_f64(Manifest &manifest);
void initialize_reduce_add_linear_combination_cf32_cf32_cf32(Manifest &manifest);
@@ -55,6 +56,7 @@ void initialize_all_reduction_op(Manifest &manifest) {
initialize_reduce_add_linear_combination_f16_f16_f16(manifest);
initialize_reduce_add_linear_combination_f32_f32_f16(manifest);
initialize_reduce_add_linear_combination_f32_f32_bf16(manifest);
initialize_reduce_add_linear_combination_f32_f32_f32(manifest);
initialize_reduce_add_linear_combination_f64_f64_f64(manifest);
initialize_reduce_add_linear_combination_cf32_cf32_cf32(manifest);