Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Test wino_conv_test_t.TestSmallPadding of 3.5.3 fails on aarch64 with Arm Compute Library 24.08 #2109

Closed
ggardet opened this issue Sep 23, 2024 · 0 comments · Fixed by #2128
Labels
platform:cpu-aarch64 Codeowner: @oneapi-src/onednn-cpu-aarch64 sighting Suspicious library behavior. Should be promoted to a bug when confirmed

Comments

@ggardet
Copy link

ggardet commented Sep 23, 2024

Summary

Test wino_conv_test_t.TestSmallPadding of 3.5.3 fails on aarch64 with Arm Compute Library 24.08 on openSUSE Tumbleweed aarch64.

Version

3.5.3

Environment

oneDNN includes hardware-specific optimizations and may behave
differently on depending on the compiler and build environment. Include
the following information to help reproduce the issue:

  • CPU make and model (try lscpu; if your lscpu does not list CPU flags,
    try running cat /proc/cpuinfo | grep flags | sort -u)
  • OS version (uname -a)
  • Compiler version (gcc --version) gcc (SUSE Linux) 14.2.0
  • CMake version (cmake --version) cmake version 3.30.3
  • CMake output log
  • git hash (git log -1 --format=%H)

Steps to reproduce

Build oneDNN 3.5.3 with Arm Compute Library 24.08

Observed behavior

Fails in Unit Test:

[ 1156s] 62/99 Test #64: test_iface_wino_convolution .............................***Failed    0.04 sec
[ 1156s] Note: Google Test filter = *:-*_GPU*
[ 1156s] [==========] Running 3 tests from 1 test suite.
[ 1156s] [----------] Global test environment set-up.
[ 1156s] [----------] 3 tests from wino_conv_test_t
[ 1156s] [ RUN      ] wino_conv_test_t.TestSmallPadding
[ 1156s] /home/abuild/rpmbuild/BUILD/oneDNN-3.5.3/tests/gtests/test_iface_wino_convolution.cpp:78: Failure
[ 1156s] Expected: fwd_hint = convolution_forward::primitive_desc(eng, prop_kind::forward, algorithm::convolution_winograd, src_md, wei_md, dst_md, {1, 1}, {1, 1}, {1, 1}) doesn't throw an exception.
[ 1156s]   Actual: it throws dnnl::error with description "could not create a primitive descriptor for a convolution forward propagation primitive".
[ 1156s] [  FAILED  ] wino_conv_test_t.TestSmallPadding (0 ms)

Expected behavior

All tests should pass.

@ggardet ggardet added the sighting Suspicious library behavior. Should be promoted to a bug when confirmed label Sep 23, 2024
@vpirogov vpirogov added the platform:cpu-aarch64 Codeowner: @oneapi-src/onednn-cpu-aarch64 label Sep 24, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
platform:cpu-aarch64 Codeowner: @oneapi-src/onednn-cpu-aarch64 sighting Suspicious library behavior. Should be promoted to a bug when confirmed
Projects
None yet
Development

Successfully merging a pull request may close this issue.

2 participants