huggingface/candle
1
0
Fork 0
mirror of https://github.com/huggingface/candle.git synced 2025-06-21 04:10:46 +00:00
Code Issues Packages Projects Releases Wiki Activity

Update the flash attn kernels. (#2333)

Browse Source
This commit is contained in:
Laurent Mazare
2024-07-15 20:37:36 +02:00
committed by GitHub
parent d74fbed334
commit 30cdd769f9
51 changed files with 2279 additions and 904 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
candle-flash-attn/kernels/error.h Normal file
View File

@ -0,0 +1,8 @@
#pragma once
#define C10_CUDA_CHECK(EXPR) \
do { \
const cudaError_t __err = EXPR; \
} while (0)
#define C10_CUDA_KERNEL_LAUNCH_CHECK() C10_CUDA_CHECK(cudaGetLastError())