biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[NVIDIA] Support Cutlass w8a8 FP8 for Blackwell Geforce GPUs (sm120) (#17280)

Browse Source 
Signed-off-by: kaln27 <liaojuncheng123@foxmail.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
This commit is contained in:
Joonchen Liau
2025-07-02 20:47:19 +08:00
committed by GitHub
parent 0c600b9ab6
commit 9e5552aa13
7 changed files with 238 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

17
csrc/quantization/cutlass_w8a8/scaled_mm_entry.cu
View File

@@ -41,6 +41,14 @@ void cutlass_moe_mm_sm90(
#endif
#if defined ENABLE_SCALED_MM_SM120 && ENABLE_SCALED_MM_SM120
void cutlass_scaled_mm_sm120(torch::Tensor& c, torch::Tensor const& a,
torch::Tensor const& b,
torch::Tensor const& a_scales,
torch::Tensor const& b_scales,
std::optional<torch::Tensor> const& bias);
#endif
#if defined ENABLE_SCALED_MM_SM100 && ENABLE_SCALED_MM_SM100
void cutlass_scaled_mm_sm100(torch::Tensor& c, torch::Tensor const& a,
torch::Tensor const& b,
@@ -168,8 +176,15 @@ void cutlass_scaled_mm(torch::Tensor& c, torch::Tensor const& a,
at::cuda::OptionalCUDAGuard const device_guard(device_of(a));
int32_t version_num = get_sm_version_num();
#if defined ENABLE_SCALED_MM_SM120 && ENABLE_SCALED_MM_SM120
if (version_num >= 120) {
cutlass_scaled_mm_sm120(c, a, b, a_scales, b_scales, bias);
return;
}
#endif
#if defined ENABLE_SCALED_MM_SM100 && ENABLE_SCALED_MM_SM100
if (version_num >= 100) {
if (version_num >= 100 && version_num < 120) {
cutlass_scaled_mm_sm100(c, a, b, a_scales, b_scales, bias);
return;
}