Search results

Performance difference for Quantized Matmul between v2.6.3 and v3.x

platform:cpu-x64Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64

question

Status: Open.

#4663 In uxlfoundation/oneDNN;

· SriAlavandar opened on Feb 9, 2026

Feasibility of Adding a New Backend Library to oneDNN

platform:cpu-x64Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64

question

Status: Open.

#4446 In uxlfoundation/oneDNN;

· avinashcpandey opened on Dec 10, 2025

[Graph API] Is compiled_partition thread-safe for concurrent calls on same instance?

question

Status: Open.

#4381 In uxlfoundation/oneDNN;

· lxq2t opened on Nov 25, 2025

dnnl_lstm_backward_primitive_desc_create() can be called with parameter prop_kind=dnnl_forward_training

bugA confirmed library bugA confirmed library bug

Status: Open.

#4306 In uxlfoundation/oneDNN;

· lacak-sk opened on Nov 10, 2025

Support for long integers (aka int64, s64, long)?

enhancementA feature or an optimization requestA feature or an optimization request

help wanted

Status: Open.

#4299 In uxlfoundation/oneDNN;

· blueberry opened on Nov 7, 2025

[Doc] Update documentation on "flags" parameter in dnnl_lstm_backward_primitive_desc_create()

documentationA request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-docA request to change/fix/improve the documentation. Codeowner: @oneapi-src/onednn-doc

Status: Open.

#4216 In uxlfoundation/oneDNN;

· lacak-sk opened on Oct 27, 2025

How to query for memory descriptors for weights_layer_md and weights_iter_md in RNN LSTM from primitive descriptor

question

Status: Open.

#4215 In uxlfoundation/oneDNN;

· lacak-sk opened on Oct 27, 2025

BWDW JIT 256 error reproducible through gtests, but not through benchdnn

component:testsCodeowner: @oneapi-src/onednn-archCodeowner: @oneapi-src/onednn-arch

help wanted

platform:cpu-aarch64Codeowner: @oneapi-src/onednn-cpu-aarch64Codeowner: @oneapi-src/onednn-cpu-aarch64

sightingSuspicious library behavior. Should be promoted to a bug when confirmedSuspicious library behavior. Should be promoted to a bug when confirmed

Status: Open.

#4124 In uxlfoundation/oneDNN;

· gassan-arm opened on Oct 13, 2025

Large kernels can generate xbyak_aarch64 exceptions

help wanted

platform:cpu-aarch64Codeowner: @oneapi-src/onednn-cpu-aarch64Codeowner: @oneapi-src/onednn-cpu-aarch64

sightingSuspicious library behavior. Should be promoted to a bug when confirmedSuspicious library behavior. Should be promoted to a bug when confirmed

Status: Open.

#4089 In uxlfoundation/oneDNN;

· Sqvid opened on Oct 8, 2025

Temporary fix nightly crash for large l_pad failing test

bugA confirmed library bugA confirmed library bug

help wanted

platform:cpu-aarch64Codeowner: @oneapi-src/onednn-cpu-aarch64Codeowner: @oneapi-src/onednn-cpu-aarch64

Status: Open.

#3981 In uxlfoundation/oneDNN;

· kasturedeeksha opened on Sep 22, 2025

AArch64: graph fusion and graph ci cases failed in AArch64 platform

bugA confirmed library bugA confirmed library bug

platform:cpu-aarch64Codeowner: @oneapi-src/onednn-cpu-aarch64Codeowner: @oneapi-src/onednn-cpu-aarch64

Status: Open.

Bug

#3825 In uxlfoundation/oneDNN;

· xiang1guo opened on Aug 26, 2025

Please privide more 4bit matmul docs and example?

platform:cpu-x64Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64Intel64/AMD64 processors. Codeowner: @oneapi-src/onednn-cpu-x64

question

Status: Open.

#3641 In uxlfoundation/oneDNN;

· alanzhai219 opened on Jul 21, 2025