DeepSeek/DeepEP
1
0
Fork 0
mirror of https://github.com/deepseek-ai/DeepEP synced 2025-06-26 18:28:11 +00:00
Code Issues Actions Packages Projects Releases Wiki Activity

Use TMA instead of LD/ST for intra-node normal kernels (#191)

Browse Source 
* Update CMake files

* Use TMA instead of LD/ST for intranode dispatch

* Use TMA instead of LD/ST for intranode combine

* Adjust configs

* Test default configs as well

* More warps for combine

* Add inter-thread fence

* Enable more warps

* Do not use TMA for senders

* Update configs

* Remove useless wait
This commit is contained in:
Chenggang Zhao
2025-06-06 15:40:17 +08:00
committed by GitHub
parent df4debe30c
commit c8dceba110
6 changed files with 230 additions and 87 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
csrc/kernels/configs.cuh
View File

@@ -19,8 +19,6 @@
#ifdef __CLION_IDE__
#define __CUDA_ARCH__ 900 // NOLINT(*-reserved-identifier)
#define __CUDACC_RDC__ // NOLINT(*-reserved-identifier)
__host__ __device__ __forceinline__ void host_device_printf(const char* format, ...) { asm volatile("trap;"); }
#define printf host_device_printf
#endif
// Remove Torch restrictions