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

Vulkan test failures after pinning GLIBC versions to 2.2.5 #3552

Open
qiao-bo opened this issue Nov 18, 2021 · 0 comments
Open

Vulkan test failures after pinning GLIBC versions to 2.2.5 #3552

qiao-bo opened this issue Nov 18, 2021 · 0 comments

Comments

@qiao-bo
Copy link
Collaborator

qiao-bo commented Nov 18, 2021

Taichi has a few math symbols exp, log, pow, log2 that will introduce dependencies on high GLIBC versions (such as 2.29), when building on systems such as Ubuntu 20.04. We found a way to pin these symbols back to an older version (2.2.5 GLIBC), which has been documented at #3426. Nevertheless, the method will introduce many Vulkan test failures such as:

========================================================== short test summary info ===========================================================
FAILED python/taichi/tests/test_ad_basics.py::test_poly[<lambda>0] - assert 0.0 == 0.234 ± 2.3e-05
FAILED python/taichi/tests/test_ad_basics.py::test_poly[<lambda>1] - assert 0.0 == -0.234 ± 2.3e-05
FAILED python/taichi/tests/test_ad_basics.py::test_poly[<lambda>2] - assert 0.0 == 0.054756000000000006 ± 5.5e-06
FAILED python/taichi/tests/test_ad_basics.py::test_poly[<lambda>3] - assert 0.0 == 0.054756000000000006 ± 5.5e-06
FAILED python/taichi/tests/test_ad_basics.py::test_poly[<lambda>4] - assert 0.0 == 0.012812904000000002 ± 1.3e-06
FAILED python/taichi/tests/test_ad_basics.py::test_poly[<lambda>5] - assert 0.0 == 0.00299821953...0008 ± 3.0e-07
FAILED python/taichi/tests/test_ad_basics.py::test_poly[<lambda>6] - assert 0.0 == -2.9780976 ± 3.0e-04
FAILED python/taichi/tests/test_ad_basics.py::test_poly[<lambda>7] - assert 0.0 == 2.118756 ± 2.1e-04
FAILED python/taichi/tests/test_ad_basics.py::test_poly[<lambda>8] - assert 0.0 == 2.1735119999999997 ± 2.2e-04
FAILED python/taichi/tests/test_ad_basics.py::test_frac[<lambda>0] - assert 0.0 == 4.273504273504273 ± 4.3e-04
FAILED python/taichi/tests/test_ad_basics.py::test_frac[<lambda>1] - assert 0.0 == -1.6109660574412532 ± 1.6e-04
FAILED python/taichi/tests/test_ad_basics.py::test_frac[<lambda>2] - assert 0.0 == -1.11283184054538 ± 1.1e-04
FAILED python/taichi/tests/test_ad_basics.py::test_unary[<lambda>-<lambda>0] - assert 0.0 == 0.48373546489791297 ± 4.8e-05
FAILED python/taichi/tests/test_ad_basics.py::test_unary[<lambda>-<lambda>1] - assert 0.0 == 1.2636444922077779 ± 1.3e-04
FAILED python/taichi/tests/test_ad_basics.py::test_unary[<lambda>-<lambda>2] - assert 0.0 == -1.4524341636244356 ± 1.5e-04
FAILED python/taichi/tests/test_ad_basics.py::test_minmax[<lambda>-<lambda>1] - assert 0.0 == 0.234 ± 2.3e-05
FAILED python/taichi/tests/test_ad_basics.py::test_minmax[<lambda>-<lambda>3] - assert 0.0 == 0.234 ± 2.3e-05
FAILED python/taichi/tests/test_ad_basics.py::test_minmax[<lambda>-<lambda>4] - assert 0.0 == 0.234 ± 2.3e-05
FAILED python/taichi/tests/test_ad_basics.py::test_minmax[<lambda>-<lambda>5] - assert 0.0 == 1.0 ± 1.0e-04
FAILED python/taichi/tests/test_ad_basics.py::test_minmax[<lambda>-<lambda>6] - assert 0.0 == 0.234 ± 2.3e-05
FAILED python/taichi/tests/test_ad_basics.py::test_minmax[<lambda>-<lambda>7] - assert 0.0 == 1.0 ± 1.0e-04
FAILED python/taichi/tests/test_ad_basics.py::test_atan2[<lambda>-<lambda>0] - assert 0.0 == 1.0414792570758407 ± 1.0e-04
FAILED python/taichi/tests/test_ad_basics.py::test_atan2[<lambda>-<lambda>1] - assert 0.0 == 0.529317069719056 ± 5.3e-05
FAILED python/taichi/tests/test_ad_basics.py::test_pow[<lambda>-<lambda>0] - assert 0.0 == 0.807015800773769 ± 8.1e-05
FAILED python/taichi/tests/test_ad_basics.py::test_pow[<lambda>-<lambda>1] - assert 0.0 == 0.5593534781830355 ± 5.6e-05
FAILED python/taichi/tests/test_ad_basics.py::test_obey_kernel_simplicity - assert 0.0 == 111 ± 1.1e-04
=================================== 26 failed, 19 passed, 1147 deselected, 4 warnings, 52 rerun in 10.85s ====================================
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
Status: Backlog
Development

No branches or pull requests

1 participant