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

Update internode_ll.cu (#246)

Browse Source
This commit is contained in:
fzyzcjy
2025-06-23 15:18:10 +08:00
committed by GitHub
parent c95997f8c4
commit fbcf430006
1 changed files with 1 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
csrc/kernels/internode_ll.cu
View File

@@ -498,7 +498,7 @@ combine(void* combined_x,
} }
cg::this_grid().sync(); cg::this_grid().sync();
// Reduce tokens with FP8 cast // Reduce tokens
EP_DEVICE_ASSERT(num_topk <= 32 and hidden_bf16_int4 <= num_threads); EP_DEVICE_ASSERT(num_topk <= 32 and hidden_bf16_int4 <= num_threads);
EP_STATIC_ASSERT(kHidden % (32 * kNumElemsPerInt4) == 0, "Invalid vectorization"); EP_STATIC_ASSERT(kHidden % (32 * kNumElemsPerInt4) == 0, "Invalid vectorization");
if (thread_id < hidden_bf16_int4) { if (thread_id < hidden_bf16_int4) {