Skip to content

Pull requests: oneapi-src/oneDNN

New pull request New
88 Open 1,116 Closed
88 Open 1,116 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

(v3.7 backport) src: gpu: intel: jit: conv: fix reorder zp precomp page faults on xe3 backport platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#2560 opened Jan 29, 2025 by hidefromkgb Loading…
src: gpu: intel: jit: conv: fix reorder zp precomp page faults on xe3 platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#2559 opened Jan 29, 2025 by hidefromkgb Loading…
3
cpu: x64: brgemm: update conditions for zp with big ow platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2558 opened Jan 29, 2025 by xuxinzen Loading…
1
cpu: x64: rnn: multiplier is not required for gru backport platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2557 opened Jan 29, 2025 by densamoilov Loading…
cpu: x64: rnn: multiplier is not required for gru backport platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2556 opened Jan 29, 2025 by densamoilov Loading…
x64: reorder: change dispatching for xf16 and blocks_size < 8 backport platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2555 opened Jan 29, 2025 by tczeszun Loading…
x64: reorder: change dispatching for xf16 and blocks_size < 8 (fixes MFDNN-13118) platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2554 opened Jan 29, 2025 by tczeszun Loading…
xe: conv: add heuristic to perform reorder for plain layouts backport platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#2553 opened Jan 29, 2025 by echeresh Loading…
2
Fix GRU brgemm implementation platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2552 opened Jan 29, 2025 by densamoilov Loading…
3
[backport] x64: updates for brgemm kernel and avx2 brgemm convolutions backport platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2551 opened Jan 29, 2025 by ankalinin Loading…
[backport] x64: updates for brgemm kernel and avx2 brgemm convolutions backport component:tests Codeowner: @oneapi-src/onednn-arch platform:cpu-aarch64 Codeowner: @oneapi-src/onednn-cpu-aarch64 platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2550 opened Jan 29, 2025 by ankalinin Loading…
xe: Add internal tests for SDPA component:tests Codeowner: @oneapi-src/onednn-arch
#2549 opened Jan 29, 2025 by umar456 Loading…
27
gpu: nvidia: fix relu type cast for dimension (backport) (#2519) backport platform:gpu-nvidia Codeowner: @oneapi-src/onednn-gpu-nvidia
#2547 opened Jan 29, 2025 by s-Nick Loading…
[GPU] Shapeless convolution: generate block 2D requirements on the fly platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#2545 opened Jan 29, 2025 by echeresh Loading…
benchdnn: introduce --summary knob for better usability component:tests Codeowner: @oneapi-src/onednn-arch documentation A request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc
#2544 opened Jan 29, 2025 by dzarukin Loading…
1
x64: brgemm kernel: update Vmm usage platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2540 opened Jan 28, 2025 by ankalinin Loading…
benchdnn, ngen, xe: fix memory use miss-estimation triggering out-of-memory issues component:tests Codeowner: @oneapi-src/onednn-arch
#2535 opened Jan 28, 2025 by rjoursler Loading…
9
cpu: x64: matmul: fix strides for weights in ba when K == 1 platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2533 opened Jan 28, 2025 by densamoilov Loading…
1
api: ukernel: make get_B_pack_type static for better usability component:api Codeowner: @oneapi-src/onednn-arch component:examples component:tests Codeowner: @oneapi-src/onednn-arch platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2531 opened Jan 28, 2025 by dzarukin Loading…
1
[backport 3.7] xe: fix sdpa out-of-bounds accesses. extend --dt support specifying tensor id backport 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:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#2530 opened Jan 27, 2025 by umar456 Loading…
1
cpu: fixes some coverity issues(overflow, dereference null pointer issues) platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2528 opened Jan 27, 2025 by xuxinzen Loading…
2
xe: conv: add heuristic to perform reorder for plain layouts platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#2527 opened Jan 27, 2025 by echeresh Loading…
2
gpu: ocl: softmax reusable enable half-precision support platform:gpu-intel Codeowner: @oneapi-src/onednn-gpu-intel
#2525 opened Jan 27, 2025 by pv-pterab-s Loading…
6
x64: conv: addressing coverity issues (backport into rls-v3.7) backport platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2524 opened Jan 27, 2025 by inteldimitrius Loading…
x64: conv: addressing coverity issues platform:cpu-x64 Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64
#2523 opened Jan 27, 2025 by inteldimitrius Loading…
2 tasks done
13
ProTip! no:milestone will show everything without a milestone.