[Kernel] Adding bias epilogue support for cutlass_scaled_mm (#5560)

Co-authored-by: Chih-Chieh-Yang <7364402+cyang49@users.noreply.github.com>
Co-authored-by: Lucas Wilkinson <lwilkinson@neuralmagic.com>
This commit is contained in:
Luka Govedič
2024-06-26 15:16:00 +00:00
committed by GitHub
co-authored by Chih-Chieh-Yang Lucas Wilkinson
parent 6984c02a27
commit 5bfd1bbc98
8 changed files with 385 additions and 136 deletions
+2 -1
View File
@@ -96,7 +96,8 @@ bool cutlass_scaled_mm_supports_fp8(int64_t cuda_device_capability);
void cutlass_scaled_mm(torch::Tensor& out, torch::Tensor const& a,
torch::Tensor const& b, torch::Tensor const& a_scales,
torch::Tensor const& b_scales);
torch::Tensor const& b_scales,
c10::optional<torch::Tensor> const& bias);
#endif