Search results

[Compile BUG] Can't compile on Intel <code>LLVM</code> Project Compiler

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

Status: Open.

#3711 In uxlfoundation/oneDNN;

· izelnakri opened on Aug 2, 2025

testcase fails: test_benchdnn_modeC_graph_ci_cpu

bugA confirmed library bugA confirmed library bug

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

Status: Open.

#3676 In uxlfoundation/oneDNN;

· ilotasw opened on Jul 28, 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

Static Link Time Optimization is Broken on MSVC2022

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

Status: Open.

#3383 In uxlfoundation/oneDNN;

· Skylion007 opened on Jun 1, 2025

Example of dnnl::vanilla_rnn_backward

question

Status: Open.

#3257 In uxlfoundation/oneDNN;

· dtecta opened on May 13, 2025

How to switch different doc version?

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

enhancementA feature or an optimization requestA feature or an optimization request

help wanted

Status: Open.

#2843 In uxlfoundation/oneDNN;

· alanzhai219 opened on Mar 10, 2025

AArch64: convolution diff issue with gemm_s8u8s32 kernel

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.

#2675 In uxlfoundation/oneDNN;

· xiang1guo opened on Feb 12, 2025

post-op shift operations

enhancementA feature or an optimization requestA feature or an optimization request

question

Status: Open.

#2600 In uxlfoundation/oneDNN;

· ArielPom opened on Feb 5, 2025

Test test_bfloat16_plus_float fails on s390x

help wanted

platform:cpu-s390xIBM Z systems based on s390x architectureIBM Z systems based on s390x architecture

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

Status: Open.

#2228 In uxlfoundation/oneDNN;

· skriesch opened on Nov 26, 2024

[ARM] Support fp16 data type in JIT Reorder kernel

enhancementA feature or an optimization requestA feature or an optimization request

help wanted

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

Status: Open.

#2185 In uxlfoundation/oneDNN;

· dmitry-gorokhov opened on Oct 28, 2024

Extend support for JIT Backward Convolution Operators with ARM SVE 128bit

enhancementA feature or an optimization requestA feature or an optimization request

help wanted

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

Status: Open.

#2165 In uxlfoundation/oneDNN;

· snadampal opened on Oct 14, 2024

[ARM] Support 8bit/4bit weights decompression for Matmul primitive

enhancementA feature or an optimization requestA feature or an optimization request

help wanted

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

Status: Open.

#2081 In uxlfoundation/oneDNN;

· dmitry-gorokhov opened on Sep 4, 2024