[04:42] <duflu> RAOF: Reconsider? https://code.launchpad.net/~vanvugt/mir/relax-libmirplatform2/+merge/233682
[04:42] <duflu> I mean... "Fixed"
[04:51] <RAOF> duflu: Enjoy
[05:15] <duflu> camako: Are you sure the fixes we want to release quickly are ABI breakers? I don't think they are. We could just do a 0.7.1
[05:16] <camako> duflu, probably not, but I've requested a silo for the 0.8.0, so we might as well have them go out with 0.8..
[05:20] <duflu> camako: Well, you have the option of not doing a major release. Any resulting pain was your choice :)
[05:22] <camako> Well, I'll be spreading the pain around this time :-)
[05:26] <camako> duflu, yeay privatize-underused is landing... if you wanna land the server-abi-26 branch by hand... be my guest! Anyways, it's way too late for me to be here.. good night!
[05:28] <duflu> camako: Unfortunately landing that one by hand is likely to break CI for *all* other branches
[05:28] <duflu> So I won't yet
[05:28] <duflu> camako: Night
[05:42] <RAOF> There we go; first cut at symbol versioning HOWTO.
[06:07] <duflu> (the crowd goes wild)
[06:07] <duflu> (or something)
[06:10]  * RAOF wonders what would fix LTO for clang
[06:17] <RAOF> duflu: Oh, did I argue enough for dont-kill-the-poor-clients?
[06:17] <duflu> RAOF: I haven't checked it today, but doubt it :)
[06:17] <RAOF> Heh
[06:19] <duflu> RAOF: Happy to be your tester if you can't reproduce the issue
[06:20] <RAOF> The MP doesn't actually touch any of the code around that codepath anyway.
[06:20] <RAOF> Well, once you remove the eglapp changes.
[06:20] <duflu> Extra weird then
[06:20] <RAOF> It's an existing race, I tells ya!@
[08:51] <duflu> Does anyone know why we have deep library directories? Seems at least one level is superfluous: /usr/lib/x86_64-linux-gnu/mir/platformgraphics/mesa/libmirplatformgraphics.so as "platformgraphics" is mentioned twice