frozenleaves/pytorch
1
0
Fork 0
mirror of https://github.com/pytorch/pytorch.git synced 2025-10-20 21:14:14 +08:00
Code Issues Packages Projects Releases Wiki Activity

CUDA 13 -- sm_120 -- Nvidia 5090 -- ptxas warning : Value of threads … (#161380)

Browse Source 
bug fix:

i have opened a issue ( https://github.com/pytorch/pytorch/issues/161376 ) and i suggest this bug fix.

In this metod compile fine.

Fixes #161376

Pull Request resolved: https://github.com/pytorch/pytorch/pull/161380
Approved by: https://github.com/eqy, https://github.com/malfet

Co-authored-by: Aidyn-A <31858918+Aidyn-A@users.noreply.github.com>
This commit is contained in:
DrStone71
2025-09-02 13:27:54 +00:00
committed by PyTorch MergeBot
parent e304ea4e69
commit bc4db2c27f
1 changed files with 2 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
torch/headeronly/macros/Macros.h
View File

@ -259,7 +259,8 @@ using namespace c10::xpu;
// to resolve potential warnings.
#if __CUDA_ARCH__ == 750
constexpr uint32_t CUDA_MAX_THREADS_PER_SM = 1024;
#elif __CUDA_ARCH__ == 860 || __CUDA_ARCH__ == 870 || __CUDA_ARCH__ == 890
#elif __CUDA_ARCH__ == 860 || __CUDA_ARCH__ == 870 || __CUDA_ARCH__ == 890 || \
__CUDA_ARCH__ == 1200
constexpr uint32_t CUDA_MAX_THREADS_PER_SM = 1536;
#else
constexpr uint32_t CUDA_MAX_THREADS_PER_SM = 2048;