Commits · 580864d81eb03d9fb1383e1782636ff6a9425fa2 · OpenDAS / vllm_cscc

"vscode:/vscode.git/clone" did not exist on "456941cfe4313f5b89390e55fa4ed0b9771731ce"

09 Mar, 2026 7 commits
- [Attention][Perf][Kernel] Replace torch.cat with vectorized CUDA kernel MLA... · 580864d8
  Roberto L. Castro authored Mar 09, 2026
```
[Attention][Perf][Kernel] Replace torch.cat with vectorized CUDA kernel MLA query concat - DeepSeek-V3.2 (#34917)
Signed-off-by: LopezCastroRoberto <rocastro@redhat.com>
Signed-off-by: Roberto L. Castro <38211239+LopezCastroRoberto@users.noreply.github.com>
```
  580864d8
- [Bugfix] Clear stale CG keys after memory profiling (#36416) · 00c4cb56
  Matthew Bonanni authored Mar 09, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  00c4cb56
- Reapply [Attention] Refactor `check_and_update_config` (#35122) · 77a73458
  Matthew Bonanni authored Mar 09, 2026
```
Signed-off-by: Matthew Bonanni <mbonanni@redhat.com>
```
  77a73458
- [Deprecation][1/2] Remove items deprecated in v0.18 (#36470) · f96c3ab0
  Cyrus Leung authored Mar 09, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  f96c3ab0
- [Bugfix] Avoid to replace non-tensor members in cpu model runner (#36430) · 217f2759
  Li, Jiang authored Mar 09, 2026
```
Signed-off-by: jiang1.li <jiang1.li@intel.com>
```
  217f2759
- feat(attention): extract KV-cache update from FlexAttention backend (#36263) · 74743104
  cong-or authored Mar 09, 2026
```
Signed-off-by: cong-or <conchubhar.gannon@gmail.com>
```
  74743104
- [Bugfix] Fix CPU OMP autobind assertion to use local_world_size (#35815) · 43aa3892
  Weiguang Li authored Mar 09, 2026
```
Signed-off-by: liweiguang <codingpunk@gmail.com>
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
Co-authored-by: Li, Jiang <jiang1.li@intel.com>
```
  43aa3892
08 Mar, 2026 1 commit
- [Frontend] Add GPU-less render serving path (`vllm launch render`) (#36166) · 4497431d
  Sage authored Mar 08, 2026
  
  4497431d
07 Mar, 2026 6 commits
- [Perf] Support FP8 KV cache for Flashinfer MLA Sparse (#35891) · 379689d5
  Wei Zhao authored Mar 07, 2026
  
  379689d5
- [Core] NGram GPU Implementation compatible with Async Scheduler (#29184) · a6be75db
  PatchyTIS authored Mar 08, 2026
  
  a6be75db
- [UX][Startup] Account for CUDA graphs during memory profiling (#30515) · ebb9cc5f
  Matthew Bonanni authored Mar 07, 2026
  
  ebb9cc5f
- Fix CUDA graph decode capture crash in AITER FlashAttention (#36042) · 1a971808
  Mengtao (Martin) Yuan authored Mar 06, 2026
```
Signed-off-by: Martin Yuan <myuan@meta.com>
Co-authored-by: Martin Yuan <myuan@meta.com>
```
  1a971808
- [Model Runner V2] Fix warmup for pipeline parallel (#36280) · b3546865
  Nick Hill authored Mar 06, 2026
```
Signed-off-by: Nick Hill <nickhill123@gmail.com>
```
  b3546865
- [Core] Fix benign error log during normal shutdown (#36270) · 6a18d878
  Nick Hill authored Mar 06, 2026
```
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Mark McLoughlin <markmc@redhat.com>
```
  6a18d878
06 Mar, 2026 15 commits
- [ROCm] Support MLA with nhead<16 and FP8 KV cache for TP=8 (Kimi K2.5/Linear) (#35850) · c188749b
  Chuan (Richard) Li authored Mar 06, 2026
```
Signed-off-by: Li <chuali@amd.com>
```
  c188749b
- Revert "[BugFix] Fix engine hanging after KV cache initialization fai… (#36262) · 26bd43b5
  Nick Hill authored Mar 06, 2026
  
  26bd43b5
- [Bugfix] Quickfix followups to busy loop removal in #28053 (#36068) · 6b625a88
  Travis Johnson authored Mar 06, 2026
```
Signed-off-by: Travis Johnson <tsjohnso@us.ibm.com>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
```
  6b625a88
- [Bugfix] Fix `cudagraph_mode:FULL` dispatch (This does not impact... · 39f9ea0d
  Raphaël Rialland authored Mar 06, 2026
```
[Bugfix] Fix `cudagraph_mode:FULL` dispatch (This does not impact `FULL_AND_PIECEWISE` (default)) (#36165)
```
  39f9ea0d
- [Core][KVConnector] Support HMA+NixlConnector (#35758) · 5b3ba94a
  Nicolò Lucchesi authored Mar 06, 2026
```
Signed-off-by: NickLucche <nlucches@redhat.com>
```
  5b3ba94a
- [Spec Decode][KV Connector] Fix KV transfer in PD + speculative decoding (#35158) · 90f3c01f
  zhanqiuhu authored Mar 06, 2026
```
Signed-off-by: Claude <noreply@anthropic.com>
Signed-off-by: Zhanqiu Hu <zh338@cornell.edu>
Co-authored-by: Nicolò Lucchesi <nlucches@redhat.com>
```
  90f3c01f
- [ROCm][CI] Fix tool use test stability - disable skinny GEMM, prefix caching,... · 807d6803
  Andreas Karatzas authored Mar 06, 2026
```
[ROCm][CI] Fix tool use test stability - disable skinny GEMM, prefix caching, eliminate batch variance (#35553)
Signed-off-by: Andreas Karatzas <akaratza@amd.com>
```
  807d6803
- [Bugfix] Fix inner_dp_world initialization order for multi-node TP (#35892) · 86e1060b
  Yongye Zhu authored Mar 06, 2026
```
Signed-off-by: Yongye Zhu <zyy1102000@gmail.com>
Signed-off-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Claude Opus 4.5 <noreply@anthropic.com>
Co-authored-by: Nick Hill <nickhill123@gmail.com>
Co-authored-by: Nick Hill <nhill@redhat.com>
```
  86e1060b
- [Frontend][Core] Add shutdown timeout - allowing in-flight requests to finish (#34730) · 27066d1b
  Mark McLoughlin authored Mar 06, 2026
```
Signed-off-by: Mark McLoughlin <markmc@redhat.com>
Co-authored-by: Claude Sonnet 4.5 <noreply@anthropic.com>
```
  27066d1b
- perf: add __slots__ to KVCacheBlock (#36164) · 57c84ff1
  cong-or authored Mar 06, 2026
```
Signed-off-by: cong-or <conchubhar.gannon@gmail.com>
```
  57c84ff1
- [BugFix] Fix engine hanging after KV cache initialization failure (#35478) · 0a208d1f
  Shiyan Deng authored Mar 05, 2026
```
Signed-off-by: Shiyan Deng <dsy842974287@meta.com>
Co-authored-by: Lu Fang <30275821+houseroad@users.noreply.github.com>
```
  0a208d1f
- [Feature] Add --distributed-timeout-seconds CLI option (#36047) · 03a49bb8
  Shiyan Deng authored Mar 05, 2026
```
Signed-off-by: Shiyan Deng <dsy842974287@meta.com>
Co-authored-by: Lu Fang <30275821+houseroad@users.noreply.github.com>
```
  03a49bb8
- [Misc] Rename `group_mm_kwargs_by_modality -> group_and_batch_mm_kwargs` (#36158) · 6dd30265
  Cyrus Leung authored Mar 06, 2026
```
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
```
  6dd30265
- Reenable features for ROCm attention backends (#36185) · c5362c73
  Rohan Potdar authored Mar 05, 2026
```
Signed-off-by: Rohan138 <rohanpotdar138@gmail.com>
```
  c5362c73
- [Performance] Extract KV-cache update from TreeAttention backend (#35384) · ebed80a7
  Dor Huri authored Mar 06, 2026
```
Signed-off-by: dorhuri123 <dor.huri1@live.biu.ac.il>
```
  ebed80a7
05 Mar, 2026 9 commits
- [Model Runner V2] Fix warmup for very small kvcache and/or blocksizes (#36176) · a73af584
  Nick Hill authored Mar 05, 2026
```
Signed-off-by: Nick Hill <nickhill123@gmail.com>
```
  a73af584
- [BugFix] Fallback from FA4->FA2 for Batch Invariance (#36059) · a57c877f
  Frank Wang authored Mar 05, 2026
```
Signed-off-by: frankwang28 <frank.wbb@hotmail.com>
```
  a57c877f
- [XPU] Enable ModelRunnerV2 on XPU (#36078) · d8839ef7
  Xinyu Chen authored Mar 06, 2026
```
Signed-off-by: Xinyu Chen <xinyu1.chen@intel.com>
```
  d8839ef7
- [Bugfix][CI] fix typos (#34934) · 6a895197
  Jiayi Yan authored Mar 06, 2026
```
Signed-off-by: 1195343015 <1195343015@qq.com>
Signed-off-by: Jiayi Yan <66017932+1195343015@users.noreply.github.com>
Co-authored-by: Harry Mellor <19981378+hmellor@users.noreply.github.com>
```
  6a895197
- [ROCm] Refactor ROCm attention backend selection logic (#35246) · 8c760b6a
  Sage Moore authored Mar 05, 2026
```
Signed-off-by: Sage Moore <sage@neuralmagic.com>
```
  8c760b6a
- [KVConnector] Scheduler: Fix num_computed_tokens after async KV load (#34616) · 612e7729
  Or Ozeri authored Mar 05, 2026
```
Signed-off-by: Or Ozeri <oro@il.ibm.com>
```
  612e7729
- [Hardware] Replace `torch.cuda.synchronize()` api with `torch.accelerator.synchronize` (#36085) · 66a22096
  Kunshang Ji authored Mar 05, 2026
```
Signed-off-by: Kunshang Ji <kunshang.ji@intel.com>
```
  66a22096
- [Bugfix] Fix block_size for hybrid model MTP (#36036) · 57c629e9
  Benjamin Chislett authored Mar 05, 2026
```
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
```
  57c629e9
- [Core] Move ray-specific WorkerWrapperBase methods to RayWorkerWrapper (#35328) · 16c472ab
  Nick Hill authored Mar 04, 2026
```
Signed-off-by: Nick Hill <nickhill123@gmail.com>
```
  16c472ab
04 Mar, 2026 2 commits
- [Model Runner V2] Misc code simplification (#35941) · a3299c3d
  Nick Hill authored Mar 04, 2026
```
Signed-off-by: Nick Hill <nickhill123@gmail.com>
```
  a3299c3d
- [Bugfix] Fix race in non-blocking num_accepted_tokens GPU->CPU copy (#36013) · be0a3f75
  Thomas Parnell authored Mar 04, 2026
```
Co-authored-by: Claude Opus 4.6 <noreply@anthropic.com>
```
  be0a3f75