Skip to content

Pull requests: uxlfoundation/oneDNN

New pull request New
116 Open 3,213 Closed
116 Open 3,213 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
Assigned to nobody Loading
Sort
Sort by
Newest Oldest Most commented Least commented Recently updated Least recently updated Best match
Most reactions
👍 👎 😄 🎉 😕 ❤️ 🚀 👀

Pull requests list

benchdnn: graph: remove obsolete verbose print component:graph-api Codeowner: @oneapi-src/onednn-graph component:tests Codeowner: @oneapi-src/onednn-arch
#4771 opened Mar 5, 2026 by TaoLv Loading…
cpu: rv64: conv: add brgemm conv kernel for rv64 component:common platform:cpu-rv64 RISC-V
#4770 opened Mar 5, 2026 by zhangjian29 Loading…
3 tasks done
2
[Backport v3.11] xe: gemm: reintroduce kernels to spotfix regressions backport performance platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4768 opened Mar 5, 2026 by Simonsays095 Loading…
[Backport v3.12-pc] xe: gemm: reintroduce kernels to spotfix regressions backport performance platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4767 opened Mar 5, 2026 by Simonsays095 Loading…
xe: gemm: reintroduce kernels to spotfix regressions platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4766 opened Mar 5, 2026 by Simonsays095 Loading…
1
[FOR TESTING] GEMM zero pool evaluation platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4765 opened Mar 4, 2026 by echeresh Loading…
1
tests: benchdnn: inputs: add more matmul runtime dims coverage component:tests Codeowner: @oneapi-src/onednn-arch
#4764 opened Mar 4, 2026 by asimonov1 • Draft
1
test ov jenkins documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc
#4763 opened Mar 4, 2026 by pmanczak • Draft
1
src: gpu: intel: jit: enable tiled grf allocation for buffers platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4762 opened Mar 4, 2026 by hidefromkgb Loading…
20
rfc: graph: revisit sdpa training backward documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc RFC A design document
#4761 opened Mar 4, 2026 by ElaineBao Loading…
@ElaineBao
[GPU][rls-v3.12-pc] Fixup mx dst swizzle for xe3p+ backport platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4760 opened Mar 4, 2026 by kealan-barbieri Loading…
[GPU] Fixup mx dst swizzle for xe3p+ platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4759 opened Mar 4, 2026 by kealan-barbieri Loading…
4 tasks done
11
[GPU][Rls-v3.12-pc] Fix mx dst kernel selection requirement backport platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4755 opened Mar 3, 2026 by kealan-barbieri Loading…
cpu: aarch64: add f16 support to JIT softmax platform:cpu-aarch64 Codeowner: @oneapi-src/onednn-cpu-aarch64
#4753 opened Mar 3, 2026 by Anndrey24 Loading…
5
cpu: x64: update APX feature detection platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#4752 opened Mar 3, 2026 by xuxinzen Loading…
2
Kealanba/backport xe3p fixes platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4748 opened Mar 2, 2026 by kealan-barbieri Loading…
4 tasks done
4
WIP: xe: use double with post-ops with f64 to preserve accuracy component:tests Codeowner: @oneapi-src/onednn-arch platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4746 opened Mar 2, 2026 by echeresh Loading…
6
fix: common: memory_desc: add upper bounds check for md tensor dimensions component:common
#4745 opened Mar 2, 2026 by avmanerikar Loading…
11
cpu: x64: xbyak fix ModRM encoding for prefetchrst2 third_party
#4737 opened Feb 27, 2026 by georgen117 Loading…
6
cpu: rv64: conv: add rvv winograd kernel component:common platform:cpu-rv64 RISC-V
#4735 opened Feb 27, 2026 by zhangjian29 Loading…
11
cpu: rv64: add JIT affine kernel optimization for softmax f32 with runtime ISA check platform:cpu-rv64 RISC-V
#4734 opened Feb 27, 2026 by xiazhuozhao Loading…
3
[WIP] src: gpu: intel: gated_mlp: ukernel-based horizontal fusion component:common platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4727 opened Feb 25, 2026 by hidefromkgb Loading…
benchdnn: respect accumulation mode in safe digits calculation component:tests Codeowner: @oneapi-src/onednn-arch
#4723 opened Feb 24, 2026 by echeresh Loading…
2
Do not call div_up(a,b) with a<0 component:common platform:cpu-aarch64 Codeowner: @oneapi-src/onednn-cpu-aarch64 platform:cpu-rv64 RISC-V platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64 platform:gpu-generic Codeowner: @oneapi-src/onednn-gpu-generic platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#4712 opened Feb 20, 2026 by asimonov1 • Draft
5
cpu: x64: matmul: added checks for invalid blocking options platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#4701 opened Feb 18, 2026 by yair-obodovsky Loading…
1
ProTip! What’s not been updated in a month: updated:<2026-02-05.