Hi Aurelien! On 03/08/2016 07:40 PM, Aurelien Jarno wrote: > The glibc testsuite in version 2.22 which has been uploaded to sid 2 > days ago fails to pass on both ravirin and landau build daemon, while > it passes fine on my system. It also fails on andi which just built glibc with the latest 4.4 kernel release we have in Debian. > The ravirin build was done with glibc 2.22-0experimental3. There was a > single failure: > > FAIL: nptl/tst-dlsym1 > > This test hasn't changed for year, and it is likely due to the fact the > experimental chroot contains a mix of gcc 5 and gcc 6. Ah, yes. This is probably because of an oversight on my side, sorry. I assume, I should set the default target release for APT to "unstable" and just have "experimental" added to the sources.list? I'm not absolutely sure, to be honest, what the proper setup for an experimental chroot is. > The landau buildd tried to build twice glibc version 2.22-1 with both > time exactly the same failures related to threads: > > FAIL: nptl/tst-cond10 > FAIL: nptl/tst-cond16 > FAIL: nptl/tst-cond17 > FAIL: nptl/tst-cond18 > FAIL: nptl/tst-cond25 > FAIL: rt/tst-mqueue5 > > The error returned by some of these tests is "Resource temporarily > unavailable". Given they are new issues that don't happen on ravirin > nor on my system, it could be a kernel issue. Hmm, ok. So raverin was on 4.3.0 at the time of the test while landau was on 4.4.0. So there might be the possibility that this is a problem with the newer kernel, indeed. > Given I am unable to reproduce the issues, can someone please > investigate them? I can try a test build with an older kernel next week. But we should also notify Oracle's userland and kernel developers about the issue. Adrian -- .''`. John Paul Adrian Glaubitz : :' : Debian Developer - glaubitz@debian.org `. `' Freie Universitaet Berlin - glaubitz@physik.fu-berlin.de `- GPG: 62FF 8A75 84E0 2956 9546 0006 7426 3B37 F5B5 F913
Attachment:
signature.asc
Description: OpenPGP digital signature