Skip to content

Pull requests: uxlfoundation/oneDNN

New pull request New
142 Open 4,035 Closed
142 Open 4,035 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

x64: injectors: eliminate implicit narrowing conversions platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5646 opened Jul 21, 2026 by tczeszun Contributor • Draft
cpu: x64: matmul: add MXFP4 (f4_e2m1 + e8m0) weights decompression for f32 matmul component:common component:tests Codeowner: @oneapi-src/onednn-arch platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5645 opened Jul 21, 2026 by amakarev Contributor Loading…
3
avoid out-of-bounds write in deserializer_t::pop_array component:common component:tests Codeowner: @oneapi-src/onednn-arch
#5644 opened Jul 21, 2026 by Keerthana-64 Loading…
3 of 4 tasks
rfcs : Added Embedding Bag RFC documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc RFC A design document
#5643 opened Jul 21, 2026 by sshinde-amd Loading…
2 of 10 tasks
GEMV IR: Support for Bias, Post-ops, Scales and Beta=1 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
#5642 opened Jul 20, 2026 by densamoilov Contributor Loading…
2
xe: gemm: jit: generator: avoid implicit kChain reduction platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5641 opened Jul 20, 2026 by rjoursler Contributor Loading…
1
x64: brgemm: other: eliminate implicit narrowing conversions platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5640 opened Jul 20, 2026 by tczeszun Contributor • Draft
2
x64: inner product: eliminate implicit narrowing conversions platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5639 opened Jul 20, 2026 by tczeszun Contributor • Draft
cpu: rnn: eliminate implicit narrowing conversions platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5638 opened Jul 20, 2026 by tczeszun Contributor • Draft
x64: matmul: eliminate implicit narrowing conversions platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5637 opened Jul 20, 2026 by tczeszun Contributor • Draft
gpu: intel: sdpa: fix fwd micro kernel miscompilation with split head sizes platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5636 opened Jul 20, 2026 by h-sadia Contributor Loading…
2
x64: lrn + prelu: fix implicit narrowing conversions platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5635 opened Jul 20, 2026 by inteldimitrius Contributor • Draft
[GPU] Gemm fixes platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel third_party
#5634 opened Jul 20, 2026 by dyoussif Contributor Loading…
5
Fix assignment operator of optional_t for non-trivial types component:common
#5633 opened Jul 20, 2026 by asimonov1 Contributor • Draft
3
Optimize the oneDNN matmul to close the gap with AMX intrinsics for DeepSeek R1 shapes MoE on GNR platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5632 opened Jul 20, 2026 by asimonov1 Contributor • Draft
xe: sdpa: Update heuristic kernel gen failure with 2d Dequan slm loads platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5627 opened Jul 18, 2026 by umar456 Contributor Loading…
2
fp8: saturate the output value after dynamic scaling instead of overflowing component:common component:tests Codeowner: @oneapi-src/onednn-arch platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5626 opened Jul 18, 2026 by dzarukin Contributor Loading…
7
[GPU] Fixup GEMM tail handling platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5625 opened Jul 18, 2026 by kealan-barbieri Contributor Loading…
4 tasks
1
[GPU] Properly handle atomic load cache policy platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5624 opened Jul 17, 2026 by kealan-barbieri Contributor Loading…
3 tasks
1
xe: sdpa: fix ineffective xe_hpg guard in bwd primitive init platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#5621 opened Jul 17, 2026 by h-sadia Contributor Loading…
3
[backport][rls-v3.13] cpu: x64: matmul: fix extendable_k with padded src leading dimension backport platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5620 opened Jul 17, 2026 by tprimak Contributor Loading…
x64: normalization: fix implicit narrowing conversions platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#5619 opened Jul 17, 2026 by inteldimitrius Contributor • Draft
cpu: rv64: adjust the eltwise code style platform:cpu-rv64 RISC-V
#5617 opened Jul 17, 2026 by zhangfeiv0 Contributor Loading…
graph: backend: dnnl: kernels: rename g_engine and g_stream component:graph-api Codeowner: @oneapi-src/onednn-graph component:tests Codeowner: @oneapi-src/onednn-arch
#5616 opened Jul 17, 2026 by TaoLv Contributor Loading…
2
cpu: rv64: add f16 and backward batch normalization component:common platform:cpu-rv64 RISC-V
#5615 opened Jul 17, 2026 by xiazhuozhao Member Loading…
ProTip! Follow long discussions with comments:>50.