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

[Bugfix] Fix some narrowing conversion warnings (#20141)

Browse Source 
Signed-off-by: Tyler Michael Smith <tyler@neuralmagic.com>
This commit is contained in:
Tyler Michael Smith
2025-06-27 12:01:28 -04:00
committed by GitHub
parent c6c983053d
commit e8c3bd2cd1
6 changed files with 8 additions and 14 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
csrc/quantization/fp4/nvfp4_quant_kernels.cu
View File

@@ -347,7 +347,7 @@ void scaled_fp4_quant_sm100a(torch::Tensor const& output,
auto input_sf_ptr = static_cast<float const*>(input_sf.data_ptr());
auto sf_out = static_cast<int32_t*>(output_sf.data_ptr());
auto output_ptr = static_cast<int64_t*>(output.data_ptr());
at::cuda::CUDAGuard device_guard{(char)input.get_device()};
const at::cuda::OptionalCUDAGuard device_guard(device_of(input));
auto stream = at::cuda::getCurrentCUDAStream(input.get_device());
// We don't support e8m0 scales at this moment.