[CPU:Bugfix] Fix issue #4463 large weight offset overflow - #4741
Merged
Conversation
…r correctness+speed (incl. shader/dflash.py comment cleanup) Discussed-in: Merge-Request 28762702 , URL: https://code.alibaba-inc.com/AliNN/AliNNPrivate/codereview/28762702 GitOrigin-RevId: deccae64d33344d91759cfc99996ddae462c88cc
ORIGINAL_AUTHOR=Zeding Zhang <me@zedingzhang.com> GitOrigin-RevId: 447579e5f325290ca82f884ec4cae3d1f2e9e7c2
ORIGINAL_AUTHOR=fshuang8299 <167897122+fshuang8299@users.noreply.github.com> GitOrigin-RevId: a498397cc4b405eb25f8310965d6f1d655c646f9
Discussed-in: Merge-Request 29126354 , URL: https://code.alibaba-inc.com/AliNN/AliNNPrivate/codereview/29126354 GitOrigin-RevId: 5e8e1ab217bf5343db7a63a0476560d85b96b9e8
…4384) ORIGINAL_AUTHOR=Srikanth Patchava <srikanth.patchava@outlook.com> GitOrigin-RevId: 88c322490539a7807985e869ef085a01e256042c
ORIGINAL_AUTHOR=huluhuluu <hustzjh@hust.edu.cn> GitOrigin-RevId: fd55b7481e1bf199de614b08ed7549a80a7366f9
ORIGINAL_AUTHOR=ihb2032 <40718643+ihb2032@users.noreply.github.com> GitOrigin-RevId: 6330b07465eac2d87a9bbdb1b0a03a91678180d8
ORIGINAL_AUTHOR=xly <xlycae@126.com> GitOrigin-RevId: a04c1a6008889ce63eebc7825a5332ba754cd4f8
Discussed-in: Merge-Request 29141009 , URL: https://code.alibaba-inc.com/AliNN/AliNNPrivate/codereview/29141009 GitOrigin-RevId: 2fc5e50b74ed98447696b8c1fb4aabf538d64266
|
|
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Automated rolling sync from the internal MNN development repository.
sync/ali-to-githubCommits
e0c529b416[Metal:Bugfix] Fix DFlash SIGSEGV on Metal + fused-attention draft for correctness+speed (incl. shader/dflash.py comment cleanup) — 花熊 zhanghuaxiang.zhx@taobao.comc4ab9060b2[LLM:Bugfix] Fix embedding QNN export paths ([LLM:Bugfix] Fix embedding QNN export paths #4555) — Zeding Zhang me@zedingzhang.com0bf17e3738[Express:Bugfix] Fix loadMap input tensor losing host buffer ([Express:Bugfix] Fix loadMap input tensor losing host buffer #4733) — fshuang8299 167897122+fshuang8299@users.noreply.github.com0248961dd3[CPU:Bugfix] Fix Arm82 build without transformer fuse — wangzhaode zhaode.wzd@alibaba-inc.coma8478970b9[Core:Bugfix] Fix model-loading leaks and image-process boundaries (fix: free Content object in error paths of loadModelFile/createFromBuffer #4384) — Srikanth Patchava srikanth.patchava@outlook.com8298424b68[OpenCL:Bugfix] Fix Eagle3 KVCache reserve logic ([OpenCL:Bugfix]fix: correct KVCache reserve logic and stride alignment in Eagle3 for opencl #4738) — huluhuluu hustzjh@hust.edu.cnff26078b63[CPU:Bugfix] Fix dirty data reduction in RVV AccumulateSequence ([CPU:Bugfix] Fix dirty data reduction in RVV AccumulateSequence #4590) — ihb2032 40718643+ihb2032@users.noreply.github.com9c338656d1[LLM:Feature] Add HunyuanOCR 1.5 support ([LLM:Feature] Add HunyuanOCR 1.5 support #4625) — xly xlycae@126.com1b2466ad1e[CPU:Bugfix] Fix issue gemma-3-4b-it-q4_0-mnn: SIGSEGV in libMNN on Snapdragon 8 Gen 2 (MnnLlmChat 0.8.3 + master) #4463 large weight offset overflow — wangzhaode zhaode.wzd@alibaba-inc.com