Skip to content

Pull requests: uxlfoundation/oneDNN

New pull request New
87 Open 1,649 Closed
87 Open 1,649 Closed
Author
Filter by author
Loading
Label
Filter by label
Loading
Use alt + click/return to exclude labels
or ⇧ + click/return for logical OR
Projects
Filter by project
Loading
Milestones
Filter by milestone
Loading
Reviews
Filter by reviews
No reviews Review required Approved review Changes requested
Assignee
Filter by who’s assigned
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

cpu: x64: conv: disable zp with large zp buffer on AMX platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#3116 opened Apr 17, 2025 by xuxinzen Loading…
1
build: fix hash sub for out-of-tree builds component:build
#3115 opened Apr 17, 2025 by atkassen Loading…
@atkassen
1
graph: backend: dnnl: support int8 sdpa with softmax component:graph-api Codeowner: @oneapi-src/onednn-graph
#3114 opened Apr 17, 2025 by gyhintel Loading…
@gyhintel
xe: sdpa: Safe softmax support for microkernel SDPA component:api Codeowner: @oneapi-src/onednn-arch component:examples component:graph-api Codeowner: @oneapi-src/onednn-graph component:tests Codeowner: @oneapi-src/onednn-arch documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64 platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#3113 opened Apr 17, 2025 by umar456 Loading…
4
[GPU] better perf model for FHS BMG kernel on some VLM shapes platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#3111 opened Apr 16, 2025 by skazakov1 Loading…
5
xe: jit: gemm: improve debugability platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#3110 opened Apr 16, 2025 by rjoursler Loading…
4
xe: jit: gemm: remove invalid inv platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#3109 opened Apr 16, 2025 by rjoursler Loading…
6
cpu: x64: ip: fix integer overflow in offsets platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#3107 opened Apr 16, 2025 by amakarev Loading…
2
[rls-v3.8] xe: ocl: reorder: prefer SIMD16 mode for ref backport platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#3106 opened Apr 16, 2025 by atkassen Loading…
@atkassen
1
gtest: graph: unit: limit decompose kernel test thread component:graph-api Codeowner: @oneapi-src/onednn-graph component:tests Codeowner: @oneapi-src/onednn-arch
#3103 opened Apr 16, 2025 by xiang1guo Loading…
@xiang1guo
1
[backport][rls-v3.8] fixup: gpu: conv: post-op mem access backport platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#3100 opened Apr 15, 2025 by yehudaorel Loading…
2
fixup: gpu: conv: post-op mem access platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#3099 opened Apr 15, 2025 by yehudaorel Loading…
2
[rls-v3.8] graph: interface: op: softmax to support inf_as_zero mode backport component:api Codeowner: @oneapi-src/onednn-arch component:examples component:graph-api Codeowner: @oneapi-src/onednn-graph component:tests Codeowner: @oneapi-src/onednn-arch documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64 platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#3098 opened Apr 15, 2025 by TaoLv Loading…
3
[backport][rls-v3.8] graph: api, doc, interface, utils, backend: support bottom-right implicit causal mask backport component:api Codeowner: @oneapi-src/onednn-arch component:graph-api Codeowner: @oneapi-src/onednn-graph component:tests Codeowner: @oneapi-src/onednn-arch documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc
#3096 opened Apr 15, 2025 by ElaineBao Loading…
@ElaineBao
2
Fix postop support checks during primitive descriptor creation platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#3088 opened Apr 13, 2025 by asimonov1 Loading…
6
benchdnn: utils: cfg: return fixed safe_n_acc for low precision dt (fixes MFDNN-13520) component:tests Codeowner: @oneapi-src/onednn-arch
#3082 opened Apr 12, 2025 by dzarukin Loading…
1
[wip] POC for Root Mean Square (RMS) layer normalization support component:api Codeowner: @oneapi-src/onednn-arch component:tests Codeowner: @oneapi-src/onednn-arch documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#3068 opened Apr 10, 2025 by mzhukova • Draft
@mzhukova
1
oneDNN v3.8 release notes backport documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc
#3064 opened Apr 10, 2025 by vgvozdeva Loading…
63
cpu: x64: upstream DMR support to main branch component:api Codeowner: @oneapi-src/onednn-arch component:tests Codeowner: @oneapi-src/onednn-arch documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#3063 opened Apr 10, 2025 by xuxinzen Loading…
75
[WIP] common: verbose: asynchronous verbose mode for execution time tracking component:api Codeowner: @oneapi-src/onednn-arch component:build documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc platform:gpu-generic Codeowner: @oneapi-src/onednn-gpu-generic
#3055 opened Apr 9, 2025 by avmanerikar • Draft
3 tasks
benchdnn: add zmalloc memory tracing component:tests Codeowner: @oneapi-src/onednn-arch
#3050 opened Apr 8, 2025 by rjoursler Loading…
11
benchdnn: graph: improve skip logic for cases with unsupported dt component:graph-api Codeowner: @oneapi-src/onednn-graph component:tests Codeowner: @oneapi-src/onednn-arch
#3046 opened Apr 8, 2025 by wzt1997 Loading…
@wzt1997
17
Update fp4 Documentation documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc
#3045 opened Apr 7, 2025 by kealan-barbieri Loading…
5
benchdnn: inputs: matmul: binary post-op bcast coverage for 3d cases component:tests Codeowner: @oneapi-src/onednn-arch
#3038 opened Apr 7, 2025 by pmanczak Loading…
1
cpu: aarch64: enable jit conv for 128 platform:cpu-aarch64 Codeowner: @oneapi-src/onednn-cpu-aarch64
#3022 opened Apr 4, 2025 by jondea • Draft
3 tasks
ProTip! Find all pull requests that aren't related to any open issues with -linked:issue.