Sorry for the noise... I mixed up two similar tests... :( I have now pushed the right skip on salsa.
In the meanwhile I have taken a deeper look to the failure and opened an issue [1] and a possible fix [2] to upstream. I'm not able to test the fix, I prefer to wait for an upstream reaction before adding a patch... is that ok? Best Chiara [1] https://github.com/joblib/joblib/issues/1353 [2] https://github.com/joblib/joblib/pull/1354 On Mon, Oct 10, 2022 at 12:30 PM Chiara Marmo <[email protected]> wrote: > > Related discussion on >> https://github.com/joblib/joblib/issues/279 >> >> Perhaps the test worth to be skipped on big endian architectures.... >> Should I add a patch? >> >> > I skipped the test for s390x and pushed to salsa: hope this will help. > > Best > Chiara > >
