You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Would it make sense to skip the tests in tests/functional/hw-counters in the make check process?
I believe these tests would be useful in the "production" system but are irrelevant in the "building" one. For example, one could do a build in MN5 without the need for the perfparanoid constraint in the salloc (and thus these tests would fail) but then use this very same installation without any problem once the constraint is included. So maybe removing these tests from the make check step would be a good idea (and maybe installing them in the make install process to have simple tests available in the "production" machine).
The problem comes because we are trying to add an Extrae easyconfig to the easybuild repository (see easybuilders/easybuild-easyconfigs#20690), and failing tests in the make check step are preventing the PR from being accepted.
I have already committed 3d8295c
for an XML test that was failing for the same reason, but I think adding the same hack for these hw-counters tests would make no sense as they are precisely testing that.
The text was updated successfully, but these errors were encountered:
Would it make sense to skip the tests in tests/functional/hw-counters in the
make check
process?I believe these tests would be useful in the "production" system but are irrelevant in the "building" one. For example, one could do a build in MN5 without the need for the
perfparanoid
constraint in thesalloc
(and thus these tests would fail) but then use this very same installation without any problem once the constraint is included. So maybe removing these tests from the make check step would be a good idea (and maybe installing them in themake install
process to have simple tests available in the "production" machine).The problem comes because we are trying to add an Extrae easyconfig to the easybuild repository (see easybuilders/easybuild-easyconfigs#20690), and failing tests in the
make check
step are preventing the PR from being accepted.I have already committed
3d8295c
for an XML test that was failing for the same reason, but I think adding the same hack for these hw-counters tests would make no sense as they are precisely testing that.
The text was updated successfully, but these errors were encountered: