Commits · 16a45b3a281805ea4d4ff3908cef512fdf6d9f84 · OpenDAS / vllm_cscc · GitLab

28 Aug, 2025 3 commits

[NVIDIA] Support SiluMul + NVFP4 quant fusion (#23671) · 16a45b3a

elvischenv authored Aug 29, 2025


Signed-off-by: jindih <jindih@nvidia.com>
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
Co-authored-by: jindih <jindih@nvidia.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Luka Govedic <lgovedic@redhat.com>

16a45b3a

[Bugfix] Add fake mode around passes (#23349) · db74d604
Angela Yi authored Aug 28, 2025
```
Signed-off-by: angelayi <yiangela7@gmail.com>
```
db74d604
[Doc]: fix typos in Python scripts (#23828) · d3da2eea
Didier Durand authored Aug 28, 2025
```
Signed-off-by: Didier Durand <durand.didier@gmail.com>
```
d3da2eea

27 Aug, 2025 1 commit
- [XPU] Add xpu torch.compile support (#22609) · fce10dbe
  Kunshang Ji authored Aug 27, 2025
```
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
```
  fce10dbe
26 Aug, 2025 3 commits

[Misc] Add override for allreduce fusion thresholds (#23639) · 7ea22e42
nvjullin authored Aug 26, 2025
```
Signed-off-by: Julien Lin <jullin@nvidia.com>
```
7ea22e42

Remove graph_pool as member of VllmBackend and argument to CUDAGraphWrapper (#23385) · 6fad29b1

Copilot authored Aug 25, 2025

Signed-off-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: copilot-swe-agent[bot] <198982749+Copilot@users.noreply.github.com>
Co-authored-by: ProExpertProg <11367180+ProExpertProg@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>

6fad29b1

Update Flashinfer to 0.2.14.post1 (#23537) · ae067888

weiliang authored Aug 26, 2025


Signed-off-by: Siyuan Fu <siyuanf@nvidia.com>
Signed-off-by: siyuanf <siyuanf@nvidia.com>
Signed-off-by: Weiliang Liu <weiliangl@nvidia.com>
Signed-off-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Siyuan Fu <siyuanf@nvidia.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>

ae067888

24 Aug, 2025 1 commit
- fix incompatibililty with non cuda platform for nvfp4 (#23478) · c7fc6b13
  Lucia Fang authored Aug 24, 2025
```
Signed-off-by: Lu Fang <fanglu@fb.com>
Co-authored-by: Lucia (Lu) Fang <fanglu@meta.com>
```
  c7fc6b13
22 Aug, 2025 2 commits
- [NVIDIA][torch.compile] Support Flashinfer TRTLLM FP8-q/kv NVFP4-out Attention Kernel (#22703) · 24d0c9e6
  elvischenv authored Aug 23, 2025
```
Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
```
  24d0c9e6
- [Doc]: fix various typos in multiple files (#23179) · 22cf679a
  Didier Durand authored Aug 22, 2025
```
Signed-off-by: Didier Durand <durand.didier@gmail.com>
```
  22cf679a
20 Aug, 2025 1 commit
- [torch.compile] Support conditional torch.compile per module (#22269) · dfd23820
  Yong Hoon Shin authored Aug 20, 2025
```
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
```
  dfd23820
19 Aug, 2025 2 commits

[NVIDIA] Support Flashinfer TRTLLM FP8-q/kv/out Attention Kernel (#21716) · 03752dba

elvischenv authored Aug 19, 2025

Signed-off-by: elvischenv <219235043+elvischenv@users.noreply.github.com>
Co-authored-by: Michael Goin <mgoin64@gmail.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>

03752dba

chore: disable enable_cpp_symbolic_shape_guards (#23048) · a4454e94
Xiao authored Aug 18, 2025
```
Signed-off-by: Xiao Liu <xiszishu@gmail.com>
```
a4454e94

15 Aug, 2025 1 commit

[Core] Allow full cudagraph with separate attention routines and orthogonal to... · 74f441f4

fhl2000 authored Aug 15, 2025


[Core] Allow full cudagraph with separate attention routines and orthogonal to compilation, add support for FA2 and FlashInfer (#20059)
Signed-off-by: fhl <2410591650@qq.com>
Signed-off-by: fhl2000 <63384265+fhl2000@users.noreply.github.com>
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
Signed-off-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>
Co-authored-by: Luka Govedič <ProExpertProg@users.noreply.github.com>
Co-authored-by: Lucas Wilkinson <lwilkins@redhat.com>
Co-authored-by: Lucas Wilkinson <LucasWilkinson@users.noreply.github.com>

74f441f4

05 Aug, 2025 1 commit
- [ROCm][Bugfix] Compilation passes fix (#22202) · 031ca762
  Gregory Shtrasberg authored Aug 04, 2025
```
Signed-off-by: Gregory Shtrasberg <Gregory.Shtrasberg@amd.com>
```
  031ca762
02 Aug, 2025 1 commit
- Revert "[compile][startup] Disable C++ compilation of symbolic shapes" (#22122) · 554df8a6
  Xiao authored Aug 02, 2025
```
Signed-off-by: Xiao Liu <xiszishu@gmail.com>
```
  554df8a6
01 Aug, 2025 3 commits
- [compile][startup] Disable C++ compilation of symbolic shapes (#20836) · 9659bc7f
  Animesh Jain authored Aug 01, 2025
```
Signed-off-by: Animesh Jain <anijain@umich.edu>
```
  9659bc7f
- [BugFix] Update AttnFusionPass cache key (#21947) · 8026a335
  Richard Zou authored Aug 01, 2025
```
Signed-off-by: Richard Zou <zou3519@gmail.com>
```
  8026a335
- [BUG] [ROCm] Fix import bug on ROCm (#22083) · 26b5f7bd
  TJian authored Aug 01, 2025
```
Signed-off-by: tjtanaa <tunjian.tan@embeddedllm.com>
```
  26b5f7bd
31 Jul, 2025 2 commits
- Add FlashInfer allreduce RMSNorm Quant fusion (#21069) · 6e672daf
  Ilya Markov authored Jul 31, 2025
```
Signed-off-by: ilmarkov <imarkov@redhat.com>
Signed-off-by: ilmarkov <markovilya197@gmail.com>
Co-authored-by: ilmarkov <imarkov@redhat.com>
```
  6e672daf
- [ez] Remove a trailing space from compilation/decorators.py (#22028) · 7349d526
  Zhengxu Chen authored Jul 31, 2025
  
  7349d526
30 Jul, 2025 1 commit
- [Feature] Add async tensor parallelism for scaled mm (#20155) · 287f527f
  cascade authored Jul 30, 2025
```
Signed-off-by: cascade812 <cascade812@outlook.com>
```
  287f527f
29 Jul, 2025 1 commit
- [Bugfix] VLLM_V1 supports passing other compilation levels (#19340) · 04e38500
  Richard Zou authored Jul 29, 2025
```
Signed-off-by: Richard Zou <zou3519@gmail.com>
```
  04e38500
24 Jul, 2025 1 commit
- [XPU] Conditionally import CUDA-specific passes to avoid import errors on xpu platform (#21036) · d9f9a3fd
  Chaojun Zhang authored Jul 24, 2025
```
Signed-off-by: chzhang <chaojun.zhang@intel.com>
```
  d9f9a3fd
23 Jul, 2025 1 commit
- Add test case for compiling multiple graphs (#21044) · 4ac7713e
  Yong Hoon Shin authored Jul 23, 2025
```
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
```
  4ac7713e
22 Jul, 2025 1 commit

Fix Flashinfer Allreduce+Norm enable disable calculation based on... · ae268b63

Xin Li authored Jul 22, 2025


Fix Flashinfer Allreduce+Norm enable disable calculation based on `fi_allreduce_fusion_max_token_num` (#21325)
Signed-off-by: XIn Li <xinli@nvidia.com>

ae268b63

15 Jul, 2025 3 commits
- [Misc] Refactor AllReduceFusionPass. Remove parameter (#20918) · 37a7d5d7
  Ilya Markov authored Jul 15, 2025
```
Signed-off-by: ilmarkov <imarkov@redhat.com>
Co-authored-by: ilmarkov <imarkov@redhat.com>
```
  37a7d5d7
- [cold start] replace VLLM_COMPILE_DEPYF with debug_dump_dir (#20940) · 91b3d190
  Boyuan Feng authored Jul 14, 2025
```
Signed-off-by: Boyuan Feng <boyuan@meta.com>
```
  91b3d190
- [BugFix] VLLM_DISABLE_COMPILE_CACHE=1 should disable all reads and writes from the cache (#20942) · ba8c3000
  Richard Zou authored Jul 14, 2025
```
Signed-off-by: Richard Zou <zou3519@gmail.com>
```
  ba8c3000
14 Jul, 2025 1 commit
- Fall back if flashinfer comm module not found (#20936) · 61e20828
  Yong Hoon Shin authored Jul 14, 2025
```
Signed-off-by: Yong Hoon Shin <yhshin@meta.com>
```
  61e20828
12 Jul, 2025 2 commits
- [cold start time] add envs.VLLM_COMPILE_DEPYF to guard decompile (#20790) · c1c8ca57
  Boyuan Feng authored Jul 11, 2025
```
Signed-off-by: Boyuan Feng <boyuan@meta.com>
```
  c1c8ca57
- Integration SM100 FlashInfer fused allreduce RMSNorm (#20691) · fc0f41d1
  Ilya Markov authored Jul 12, 2025
```
Signed-off-by: ilmarkov <imarkov@redhat.com>
Co-authored-by: ilmarkov <imarkov@redhat.com>
```
  fc0f41d1
11 Jul, 2025 2 commits
- [Bugfix] Upgrade depyf to 0.19 and streamline custom pass logging (#20777) · 762be26a
  Luka Govedič authored Jul 11, 2025
```
Signed-off-by: Luka Govedic <lgovedic@redhat.com>
Signed-off-by: luka <lgovedic@redhat.com>
```
  762be26a
- [Perf][fp8] Use CustomOp abstraction for fp8 quant for better perf (#19830) · 31d5c179
  Luka Govedič authored Jul 11, 2025
```
Signed-off-by: Luka Govedic <lgovedic@redhat.com>
Co-authored-by: mgoin <mgoin64@gmail.com>
```
  31d5c179
08 Jul, 2025 1 commit
- [Misc] Improve logging for dynamic shape cache compilation (#20573) · d2e841a1
  Kyle Yu authored Jul 07, 2025
```
Signed-off-by: kyolebu <kyu@redhat.com>
```
  d2e841a1
04 Jul, 2025 1 commit
- [Misc] Add SPDX-FileCopyrightText (#20428) · 1caca5a5
  Jee Jee Li authored Jul 04, 2025
```
Signed-off-by: Jee Jee Li <pandaleefree@gmail.com>
```
  1caca5a5
24 Jun, 2025 1 commit
- use .dev for version comparison with pytorch nightly release (#20031) · c01d1c5a
  Boyuan Feng authored Jun 24, 2025
```
Signed-off-by: Boyuan Feng <boyuan@meta.com>
```
  c01d1c5a
23 Jun, 2025 1 commit
- [Feature] Support sequence parallelism for static fp8 quantization (#19181) · e6327c9b
  cascade authored Jun 23, 2025
```
Signed-off-by: cascade812 <cascade812@outlook.com>
```
  e6327c9b
19 Jun, 2025 1 commit
- [BugFix] Fix use_cudagraph=False (#19612) · ed333497
  Richard Zou authored Jun 18, 2025
```
Signed-off-by: Richard Zou <zou3519@gmail.com>
```
  ed333497
13 Jun, 2025 1 commit
- [CUDA] Enable full cudagraph for FlashMLA (#18581) · 3597b06a
  Luka Govedič authored Jun 13, 2025
```
Signed-off-by: luka <luka@neuralmagic.com>
```
  3597b06a