Skip to content

Commit bb64e74

Browse files
PerkzZhengPerkzZheng
authored
[None][fix] fix a bug with trtllm-gen kernels + attention sinks (#7919)
Signed-off-by: Perkz Zheng <67892460+PerkzZheng@users.noreply.github.com>
1 parent f882fb8 commit bb64e74

File tree

1,357 files changed

+7872
-4332
lines changed

Some content is hidden

Large Commits have some content hidden by default. Use the searchbox below for content that may be hidden.

1,357 files changed

+7872
-4332
lines changed

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/fmha/cubin/FmhaSm100aKernel_QE4m3KvE2m1OE4m3H128PagedKvCausalP32VarSeqQ128Kv128PersistentContext_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:874dfac674d05781b81eceba1f903a059ab327e53eab719bcb81944b8fa6570a
3-
size 631293
2+
oid sha256:b701d0a9633860f7c27858c996c0c9a5a0e1ee050550c56fd17207a961c02ee9
3+
size 671549

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/fmha/cubin/FmhaSm100aKernel_QE4m3KvE2m1OE4m3H128PagedKvCausalP32VarSeqQ128Kv128StaticContext_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:c5ed56a6219edd48875b2103ebd5eb3a717bacbe7288db7156bc051a51a13796
3-
size 597072
2+
oid sha256:b3cdd8bb13409ad08bc71c907a8c49c193b72f4f2e64a56f5be5d9b6ceb4ef7e
3+
size 634961

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/fmha/cubin/FmhaSm100aKernel_QE4m3KvE2m1OE4m3H128PagedKvCausalP64VarSeqQ128Kv128PersistentContext_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:b4502fb4cec0c4100b9113eaa10588eab37937864e2d06cb48762f18fd6be798
3-
size 610326
2+
oid sha256:4aec1713bf94b80e34e828b1191d4b4750087dad76185a67c288c2856ab30d62
3+
size 650583

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/fmha/cubin/FmhaSm100aKernel_QE4m3KvE2m1OE4m3H128PagedKvCausalP64VarSeqQ128Kv128StaticContext_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:6a331240f055a91972980b90b9d52dfc83652dfeb415e33283303473263fd33b
3-
size 581284
2+
oid sha256:b75eb8454874eb9f2818d8dd986974ecb768cb2d79690a78a68642bb33974f72
3+
size 618385

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/fmha/cubin/FmhaSm100aKernel_QE4m3KvE2m1OE4m3H128PagedKvDenseP32MultiCtasKvCgaVarSeqQ16Kv128StaticSwapsAbForGen_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:d280dcfda9c2da47d78dd7beb73d29e7325cf500fa7a795ce506932de42a4b14
3-
size 661859
2+
oid sha256:662cd1c7b3f76772bf3a7f6b7f032aa9d8d6384d8613546f3cd9b8c533ce7f24
3+
size 666595

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/fmha/cubin/FmhaSm100aKernel_QE4m3KvE2m1OE4m3H128PagedKvDenseP32MultiCtasKvCgaVarSeqQ8Kv128StaticSwapsAbForGen_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:7125bc1bee0aff133bebfeea464626ef75bec61f650a9efb8e53e0f3de452583
3-
size 621625
2+
oid sha256:4798b1f1502de45ff9ecb7adf61313756128ea75b851cd5372bea4b8376ddd1a
3+
size 623203

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/fmha/cubin/FmhaSm100aKernel_QE4m3KvE2m1OE4m3H128PagedKvDenseP32MultiCtasKvVarSeqQ16Kv128StaticSwapsAbForGen_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:3a936f0f4022b6f67978d6e8820b9a8c380ad19602d24bb7126b2729bb4cce6c
3-
size 656599
2+
oid sha256:3c9ff6f3c938733f9b96388aab58e7894c7fd0e3766ade5623098d5a10d76f7f
3+
size 660545

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/fmha/cubin/FmhaSm100aKernel_QE4m3KvE2m1OE4m3H128PagedKvDenseP32MultiCtasKvVarSeqQ8Kv128StaticSwapsAbForGen_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:30cee427aeea3aacce8c7af9c5a995348b4f1a9eddee333cf83b579c550a5ee2
3-
size 617943
2+
oid sha256:a2a0becf6f915523e1f2a577c0cc6e5b53c5c76007932297c8060a9fa1dffb74
3+
size 620311

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/fmha/cubin/FmhaSm100aKernel_QE4m3KvE2m1OE4m3H128PagedKvDenseP32VarSeqQ128Kv128PersistentContext_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:6fd7cb2f1d1346a4d58ab2f36a7fad2e12d6e07291dcb3f3f6622064a6abdcdb
3-
size 675889
2+
oid sha256:1d137005de0eb779040474369f0512f14495f6ab1376dfd90e15def29aecc5ad
3+
size 716145

‎cpp/tensorrt_llm/kernels/trtllmGenKernels/fmha/cubin/FmhaSm100aKernel_QE4m3KvE2m1OE4m3H128PagedKvDenseP32VarSeqQ128Kv128StaticContext_cubin.cpp‎

Lines changed: 2 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -1,3 +1,3 @@
11
version https://git-lfs.github.com/spec/v1
2-
oid sha256:aaca053a118aae5f46c845b564149051ea3c5da60c65eb345c22bd42abdf98df
3-
size 596378
2+
oid sha256:519bf1ba61cfdd0cba303b2f3fe46d6924d7fecdc6a629180330040ae35eb341
3+
size 634267

0 commit comments

Comments
 (0)