Dmitry Zarukin

rfc: support implicit concat

2

This RFC covers oneDNN vision on implicit concat support on user side, provides an example how it may be implemented and potential challenges to implement it. [Link to rendered document](https://github.com/dzarukin/oneDNN/blob/dzarukin/rfcs/support_implicit_concat/rfcs/20211118-support-implicit-concat/README.md).

[Link to a rendered document](https://github.com/dzarukin/oneDNN/tree/dzarukin/rfcs/v3_cleanup/rfcs/20220815-v3.0-API-cleanup). This document mentions all (or at least most) changes coming in v3.0 when it comes to names changed or removed. P.S. Though RFC shows 20k...

[nvidia] batch normalization primitive fails correctness check

2

# Summary oneDNN validation for Nvidia backend hits a correctness issue on backward batch normalization with shift and scale under benchdnn. # Version Latest master. # Environment ## Hardware: NVIDIA...

# Summary oneDNN validation for Nvidia backend hits a correctness issue under benchdnn on int8 convolution problems with dst scale set. # Steps to reproduce ## Build ``` mkdir -p...

# Summary oneDNN validation for Nvidia backend hits a correctness issue on forward linear resampling for specific shapes under benchdnn. # Version Latest master. # Environment ## Hardware: NVIDIA A100...

[DO NOT REVIEW][TEST] common: on-purpose check removal to trigger new coverity issues

1

This PR is to validate external repo Coverity integration.

cpu: some improvements

1

This is some groundwork for future asynchronous runtime support: * There are overflow issues caught by `-ftrapv` option when running test_conv_all for conv and ip. * There's a rename for...

rfcs: ukernel API updates to support exception-free mode and rt lds

1

[Link to a rendered document](https://github.com/uxlfoundation/oneDNN/tree/dzarukin/rfcs/ukernel_api_updates/rfcs/20250312-ukernel-api-updates)

[WIP] benchdnn: enable parallel creation on Windows

1

Use rls-v3.8 branch as more stable to identify side effects.

all: introduce dynamic_fp quantization mode

1

[MFDNN-14462](https://jira.devtools.intel.com/browse/MFDNN-14462)