xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-08-12 15:34:32 +08:00
Code Issues Packages Projects Releases Wiki Activity

[BugFix][DP/EP] Fix CUTLASS MLA hang under load (#26026)

Browse Source 
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Robert Shaw <robshaw@redhat.com>
Co-authored-by: rshaw@neuralmagic.com <rshaw@neuralmagic.com>
Signed-off-by: yewentao256 <zhyanwentao@126.com>
This commit is contained in:
Lucas Wilkinson 2025-10-01 15:30:00 -04:00 committed by yewentao256
parent bba7623426
commit 90529cec41
1 changed files with 32 additions and 32 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
csrc/attention/mla/cutlass_sm100_mla/kernel/sm100_fmha_mla_tma_warpspecialized.hpp
View File

@ -1900,7 +1900,7 @@ struct Sm100FmhaMlaKernelTmaWarpspecialized {
cutlass::arch::NamedBarrier( cutlass::arch::NamedBarrier(
(kNumComputeWarps + kNumLoadWarps) * NumThreadsPerWarp, (kNumComputeWarps + kNumLoadWarps) * NumThreadsPerWarp,
kNamedBarrierEpilogue kNamedBarrierEpilogue
).arrive(); ).arrive_and_wait();
return; return;
} }