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

Fully remove forwarders' and NVL receivers' code

Browse Source
This commit is contained in:
Chenggang Zhao
2025-06-19 13:48:07 +08:00
parent 3a3398f686
commit a0a6e22eff
3 changed files with 78 additions and 309 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
csrc/kernels/utils.cuh
View File

@@ -366,7 +366,7 @@ __device__ __forceinline__ void unpack2(const dtype_b_t& packed, dtype_a_t& x, d
}
template <typename dtype_t>
__device__ __forceinline__ dtype_t broadcast(dtype_t& ptr, int src_lane_idx) {
__device__ __forceinline__ dtype_t broadcast(dtype_t ptr, int src_lane_idx) {
EP_STATIC_ASSERT(sizeof(dtype_t) % sizeof(int) == 0, "");
auto send_int_values = reinterpret_cast<int*>(&ptr);
int recv_int_values[sizeof(dtype_t) / sizeof(int)];