[10:15] <oSoMoN> hey kernel team! is bug #1699772 supposed to be fully fixed with 4.4.0-83.106 ?
[10:16] <oSoMoN> I’m asking because I’m still seeing build failures for libreoffice 5.3.3 and 5.3.4 on i386 (bug #1700692)
[10:55] <smb> oSoMoN, That kernel version replaced the fix with the current upstream approach. That should fix the currently known real use-cases but showed some failures with some test cases. However upstream decided that they want to see real use-cases. So behavior is probably not back to how it was before but might need convincing arguments for upstream that something breaking is a legit use.
[11:41] <oSoMoN> smb, ack, so where should I forward that use-case for upstream to consider it?
[11:57] <apw> oSoMoN, the first step is to look at what is it doing in that unit test and confirm it is actually sensible, and was not getting lucky beforehand
[11:57] <apw> oSoMoN, do you have a reference to the test ?
[12:05] <oSoMoN> apw, there's a full stack trace here: https://bugs.launchpad.net/ubuntu/+source/libreoffice/+bug/1700692/comments/3