[Regression] `intel-opencl-rt` 2022.2 breaks PyOpenCL CI

[inducer]

Solution to issue cannot be found in the documentation.

• [X] I checked the documentation.

Issue

See for example this CI job:

https://github.com/inducer/pyopencl/actions/runs/3172205054/jobs/5166430208

2022.1 worked fine; the breakage appears to have started with 2022.2 about a day ago: https://anaconda.org/conda-forge/intel-opencl-rt/files

Installed packages

# Name                    Version                   Build  Channel
_libgcc_mutex             0.1                 conda_forge    conda-forge
_openmp_mutex             4.5                       2_gnu    conda-forge
bzip2                     1.0.8                h7f98852_4    conda-forge
c-ares                    1.18.1               h7f98852_0    conda-forge
ca-certificates           2022.9.24            ha878542_0    conda-forge
curl                      7.83.1               h2283fc2_0    conda-forge
expat                     2.4.9                h27087fc_0    conda-forge
gettext                   0.19.8.1          h27087fc_1009    conda-forge
git                       2.37.3          pl5321h04cb727_0    conda-forge
icu                       70.1                 h27087fc_0    conda-forge
importlib-metadata        4.11.4          py310hff52083_0    conda-forge
intel-cmplr-lib-rt        2022.2.0          h6239696_8734    conda-forge
intel-cmplr-lic-rt        2022.2.0          ha770c72_8734    conda-forge
intel-opencl-rt           2022.2.0          hce74451_8734    conda-forge
keyutils                  1.6.1                h166bdaf_0    conda-forge
krb5                      1.19.3               h08a2579_0    conda-forge
ld_impl_linux-64          2.36.1               hea4e1c9_2    conda-forge
libblas                   3.9.0           16_linux64_openblas    conda-forge
libcblas                  3.9.0           16_linux64_openblas    conda-forge
libcurl                   7.83.1               h2283fc2_0    conda-forge
libedit                   3.1.20191231         he28a2e2_2    conda-forge
libev                     4.33                 h516909a_1    conda-forge
libffi                    3.4.2                h7f98852_5    conda-forge
libgcc-ng                 12.1.0              h8d9b700_16    conda-forge
libgfortran-ng            12.1.0              h69a702a_16    conda-forge
libgfortran5              12.1.0              hdcd56e2_16    conda-forge
libgomp                   12.1.0              h8d9b700_16    conda-forge
libiconv                  1.17                 h166bdaf_0    conda-forge
liblapack                 3.9.0           16_linux64_openblas    conda-forge
libnghttp2                1.47.0               hff17c54_1    conda-forge
libnsl                    2.0.0                h7f98852_0    conda-forge
libopenblas               0.3.21          pthreads_h78a6416_3    conda-forge
libsqlite                 3.39.3               h753d276_0    conda-forge
libssh2                   1.10.0               hf14f497_3    conda-forge
libstdcxx-ng              12.1.0              ha89aaad_16    conda-forge
libuuid                   2.32.1            h7f98852_1000    conda-forge
libxml2                   2.9.14               h22db469_4    conda-forge
libzlib                   1.2.12               h166bdaf_3    conda-forge
mako                      1.2.3              pyhd8ed1ab_0    conda-forge
markupsafe                2.1.1           py310h5764c6d_1    conda-forge
ncurses                   6.3                  h27087fc_1    conda-forge
numpy                     1.23.3          py310h53a5b5f_0    conda-forge
ocl-icd                   2.3.1                h7f98852_0    conda-forge
openssl                   3.0.5                h166bdaf_2    conda-forge
pcre2                     10.37                hc3806b6_1    conda-forge
perl                      5.32.1          2_h7f98852_perl5    conda-forge
pip                       22.2.2             pyhd8ed1ab_0    conda-forge
pybind11                  2.10.0          py310hbf28c38_0    conda-forge
pybind11-global           2.10.0          py310hbf28c38_0    conda-forge
python                    3.10.6          ha86cf86_0_cpython    conda-forge
python_abi                3.10                    2_cp310    conda-forge
readline                  8.1.2                h0f457ee_0    conda-forge
setuptools                65.4.0             pyhd8ed1ab_0    conda-forge
tbb                       2021.6.0             h924138e_0    conda-forge
tk                        8.6.12               h27826a3_0    conda-forge
tzdata                    2022d                h191b570_0    conda-forge
wheel                     0.37.1             pyhd8ed1ab_0    conda-forge
xz                        5.2.6                h166bdaf_0    conda-forge
zipp                      3.8.1              pyhd8ed1ab_0    conda-forge

Environment info

(see CI log)

[inducer]

cc @zhaomaosu (based on our conversation in https://github.com/intel/llvm/issues/2038) @isuruf (for interest)

[isuruf]

Looks like this happens only for slightly older processors. Cascade Lake processors don't have this issue, but Haswell processors do.

[zhaomaosu]

@inducer, thanks for the information. The regression will be fixed in the next release of oneAPI.

[napetrov]

@inducer - i would assume this one have been fixed?