IRC channel logs
2019-12-11.log
back to list of logs
***Server sets mode: +nt
<youpi>gnu_srs1: by "packages at -ports", you mean "packages in unreleased" ? <youpi>the sid packages in the ports archive do appear on buildd.Debian.org <youpi>the unreleased packages don't, I don't think stats are done for them <gnu_srs1>So the build stats are only referring to packages in sid. Is there any method to include the unreleased ones? <gnu_srs1>Additionally, when building packages in unreleased you close bugs, but they are not officially closed in sid. <youpi>17:53:23) jrtc27: unreleased isn't tracked by the BTS <youpi>(19:36:12) Samuel Thibault: gnu_srs1: yes, because they're not done: the main archive doesn't contain the fixes <gnu_srs1>I just scrolled back and did not find your reply. I must have been offline then. <gnu_srs1>(17:55:11) srs: So what's the idea of flagging bugs as closed then? <gnu_srs1>(17:56:20) srs: It would be better to make an NMU of the package in question, or?? <youpi>gnu_srs1: could you build-test and submit a patch to backport the semaphore timing fix to gcc-9? <youpi>perhaps golang packages will not actually need that fix to build fine, but probably better have it <gnu_srs1>building gcc-9 with the sem_timedwait fix (and the sigaltstack fix in glibc-2.29) <gnu_srs1>I see that you are now building golang packages, exciting :D <gnu_srs1>of course, but you enabled the builds right? <youpi>I gave back what uses sigaltstack and failed, yes <youpi>that's not only golang packages <youpi>gnu_srs1: golang-goprotobuf seems to need to investigation <gnu_srs1>youpi: Sorry to disappoint you: About the same number of libgo tests fails. The sem_timedwait did not seem to change anything. <gnu_srs1>However, two SIGSEGV failures have now disappeared, but still fails, probably due to the gcc fix of sigaltstack. <gnu_srs1>Most tests fails due to timeouts (I have increased the value several times, no change) <youpi>gnu_srs1: that the test don't sigsegvs any more means progress :)