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

[Compile] Fix nvfp4 compile warning (#38573)

Browse Source 
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Wentao Ye
2026-04-01 14:28:57 -04:00
committed by GitHub
parent cbe7d18096
commit c9a9db0e02
1 changed files with 7 additions and 4 deletions
Download Patch File Download Diff File Expand all files Collapse all files

11
csrc/libtorch_stable/quantization/fp4/nvfp4_quant_entry.cu
View File

@@ -58,16 +58,19 @@ void silu_and_mul_scaled_fp4_experts_quant_sm1xxa(
torch::stable::Tensor const& output_scale_offset_by_experts);
#endif
#if (defined(ENABLE_NVFP4_SM100) && ENABLE_NVFP4_SM100) || \
(defined(ENABLE_NVFP4_SM120) && ENABLE_NVFP4_SM120)
static bool nvfp4_quant_sm_supported() {
const int32_t sm = get_sm_version_num();
#if defined(ENABLE_NVFP4_SM100) && ENABLE_NVFP4_SM100
#if defined(ENABLE_NVFP4_SM100) && ENABLE_NVFP4_SM100
if (sm >= 100 && sm < 120) return true;
#endif
#if defined(ENABLE_NVFP4_SM120) && ENABLE_NVFP4_SM120
#endif
#if defined(ENABLE_NVFP4_SM120) && ENABLE_NVFP4_SM120
if (sm >= 120 && sm < 130) return true;
#endif
#endif
return false;
}
#endif
void scaled_fp4_quant_out(torch::stable::Tensor const& input,
torch::stable::Tensor const& input_sf,