Skip to content

Pull requests: uxlfoundation/oneDNN

New pull request New
146 Open 4,367 Closed
146 Open 4,367 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

Flatten BRGEMM kernel hierarchy platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5995 opened Sep 12, 2026 by densamoilov Contributor • 2/2 Loading…
1
[backport] gpu: rnn: fix backward propagation for dhc == 1 backport platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5994 opened Sep 11, 2026 by karturov Contributor Loading…
1
Introduce IR-based brgemm platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5993 opened Sep 11, 2026 by densamoilov Contributor • 1/2 Loading…
4
reorder: enable primitive_desc_iterator support component:common component:tests Codeowner: @oneapi-src/onednn-arch platform:cpu-aarch64 Codeowner: @oneapi-src/onednn-cpu-aarch64 platform:cpu-ppc64 PowerPC platform:cpu-rv64 RISC-V platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64 platform:gpu-amd Codeowner: @oneapi-src/onednn-gpu-amd platform:gpu-generic Codeowner: @oneapi-src/onednn-gpu-generic platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel platform:gpu-nvidia Codeowner: @oneapi-src/onednn-gpu-nvidia
#5992 opened Sep 11, 2026 by dzarukin Contributor Loading…
6
xe: ggemm: Add support for u3 to grouped gemm(rebased dyoussif/gemm_u3_v2 on mzhukova/u3) component:common component:tests Codeowner: @oneapi-src/onednn-arch platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5988 opened Sep 10, 2026 by umar456 Contributor • 2/2 • Draft
src: gpu: intel: add optimized u2 kernels component:api Codeowner: @oneapi-src/onednn-arch component:common component:tests Codeowner: @oneapi-src/onednn-arch documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5987 opened Sep 10, 2026 by mdvoretc-intel • Draft
4
cpu: x64: zen: add f16 matmul and weight prepack support component:tests Codeowner: @oneapi-src/onednn-arch platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5986 opened Sep 10, 2026 by AyushSinghBaiswar Contributor Loading…
3 of 6 tasks
6
benchdnn: matmul: introduce weekly set component:build component:tests Codeowner: @oneapi-src/onednn-arch
#5985 opened Sep 9, 2026 by dzarukin Contributor Loading…
benchdnn: enhance check-ref-impl to distinguish backends it applies to component:tests Codeowner: @oneapi-src/onednn-arch documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc
#5984 opened Sep 9, 2026 by dzarukin Contributor Loading…
3
cpu: rv64: select wider LMUL for pure-copy reorder on VLEN>=256 platform:cpu-rv64 RISC-V
#5982 opened Sep 9, 2026 by ww8191201-coder Contributor • Draft
cpu: rv64: widen vector groups in the eltwise JIT kernel platform:cpu-rv64 RISC-V
#5981 opened Sep 9, 2026 by ww8191201-coder Contributor • Draft
cpu: rv64: reuse xf16 softmax strided JIT kernels platform:cpu-rv64 RISC-V
#5980 opened Sep 9, 2026 by ww8191201-coder Contributor Loading…
gpu: intel: sdpa: add configs for small head sizes with a thin q component:tests Codeowner: @oneapi-src/onednn-arch platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5979 opened Sep 8, 2026 by karturov Contributor Loading…
2 tasks done
2
[GPU] Extend scale support in matmul component:common component:tests Codeowner: @oneapi-src/onednn-arch platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5978 opened Sep 8, 2026 by echeresh Contributor Loading…
8
cpu: aarch64: add ASIMD 16/32-bit transpose platform:cpu-aarch64 Codeowner: @oneapi-src/onednn-cpu-aarch64
#5975 opened Sep 8, 2026 by jondea Contributor • Draft
3 tasks done
x64: align clang narrowing warning flags to MSVC component:common platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5974 opened Sep 7, 2026 by inteldimitrius Contributor Loading…
2
common: attr: include binary select src2 in post-op identity component:common component:tests Codeowner: @oneapi-src/onednn-arch
#5970 opened Sep 6, 2026 by MaxFreedomPollard Loading…
3 tasks done
benchdnn: matmul: paneled (periodic) fast native reference (--mode-modifier=A) component:tests Codeowner: @oneapi-src/onednn-arch documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc
#5969 opened Sep 6, 2026 by karturov Contributor • Draft
xe: gemm: jit: disable k-streaming for grouped quant platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5968 opened Sep 4, 2026 by kealan-barbieri Contributor Loading…
4 tasks
2
api, src: common: add u2 data type component:api Codeowner: @oneapi-src/onednn-arch component:common component:tests Codeowner: @oneapi-src/onednn-arch documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5965 opened Sep 4, 2026 by mdvoretc-intel Loading…
8
[GPU] Use SYCL profiling tags for async verbose profiling in SYCL streams component:build component:common documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5964 opened Sep 3, 2026 by avmanerikar Contributor Loading…
31
benchdnn: reorder: enhancements (prepare for generated coverage) 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
#5962 opened Sep 3, 2026 by dzarukin Contributor Loading…
9
xe: gemm: jit: fixup host scalar sign platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5960 opened Sep 3, 2026 by dyoussif Contributor Loading…
5
[WIP] Improve performance of fused SDPA/GQA component:graph-api Codeowner: @oneapi-src/onednn-graph component:tests Codeowner: @oneapi-src/onednn-arch platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5959 opened Sep 3, 2026 by asimonov1 Contributor • Draft
3
graph: backend: dnnl: kernels: eliminate reorders in sdp_decomp component:graph-api Codeowner: @oneapi-src/onednn-graph component:tests Codeowner: @oneapi-src/onednn-arch
#5958 opened Sep 3, 2026 by TaoLv Contributor Loading…
7
ProTip! Type g i on any issue or pull request to go back to the issue listing page.