| Saviq | > <@grayson-g:matrix.org> Frame Diagnostic documentation is [now up](https://mir-server.io/docs/ubuntu-frame-diagnostic-documentation).... (full message at <https://libera.ems.host/_matrix/media/r0/download/libera.chat/1a1880c99efd86f25ff4094e98f7968b9343d513>) | 07:17 |
|---|---|---|
| Saviq | Hmm wonder if some of my troubles are just different manifestations of https://bugs.launchpad.net/ubuntu/+source/mutter/+bug/1985869 | 08:41 |
| Saviq | Oh oh!... (full message at <https://libera.ems.host/_matrix/media/r0/download/libera.chat/ba871bf0c44c7f04b5594c0710b979fdaebbc8e8>) | 09:02 |
| Saviq | I didn't manage to catch more, will put it under gdb again, but this happened when waking the screens up | 09:03 |
| Saviq | I only have the trace after this converted to SIGABRT alraedy | 09:03 |
| Saviq | There's likely too many instances of `mir::geometry::Width` to track this without a trace, are there not… | 09:08 |
| alan_g[m] | Even that may not be enough: looks like a user after free, and the trace is only about the use (not the free) | 09:10 |
| -GitHub[m]:#mir-server- **[MirServer/mir]** bors[bot] merged [pull request #2684](https://github.com/MirServer/mir/pull/2684): Deal with confused state of KMS outputs better | 11:18 | |
| -GitHub[m]:#mir-server- **[MirServer/mir]** bors[bot] closed [issue #2661](https://github.com/MirServer/mir/issues/2661): Mir sometimes shuts down when moving to a different TTY when logind-driven | 11:18 | |
| -GitHub[m]:#mir-server- **[MirServer/mir]** bors[bot] closed [issue #2674](https://github.com/MirServer/mir/issues/2674): Waking screens up after `idle-timeout` often stops or locks Mir up | 11:18 | |
| Saviq | I now wonder… do we even want a bugfix release, or should we just go for a full one… https://github.com/MirServer/mir/compare/v2.9.0...main | 11:33 |
| alan_g[m] | I thought maintaining ABI and features would make SRU easier | 11:34 |
| Saviq | Not SRU, FWIW (that's Stable Release Update), just a freeze exception: | 11:35 |
| Saviq | https://wiki.ubuntu.com/FreezeExceptionProcess#FeatureFreeze_Exceptions | 11:35 |
| Saviq | And according to this, yeah, it would be easier | 11:36 |
| Saviq | Until Oct 13 we can simply upload a bugfix | 11:36 |
| Saviq | OK I'll get a list of merges that we can walk through to try and cherry-pick | 11:37 |
| alan_g[m] | It's pretty easy to pull the list together and chrerry-pick it in CLion. I've done it a few times | 11:44 |
| alan_g[m] | Anyway, the rule: . If it changes include/... o9r debian/... then it isn't a "bugfix" | 11:45 |
| alan_g[m] | *or | 11:45 |
| Saviq | https://paste.ubuntu.com/p/SBCX3QtpRx/ | 11:45 |
| alan_g[m] | That's not very useful: no way to expand the changes | 11:47 |
| * alan_g[m] uploaded an image: (518KiB) < https://libera.ems.host/_matrix/media/r0/download/matrix.org/bTDlpQUTUboEqzHUKhCPwEjv/1665056946014.png > | 11:49 | |
| alan_g[m] | For example, #2623 changes the mircommon and platform APIs | 11:54 |
| Saviq | I have a feeling it would be easier to get the FFe than to release a huge bugfix release now and then the actual release soon after… | 12:00 |
| Saviq | We have zero dependents in the archive | 12:01 |
| alan_g[m] | Is it that huge? (I've not counted, but didn't think it that bad) | 12:02 |
| Saviq | There's 61 merges, most of them will be bugfixes, no? | 12:03 |
| alan_g[m] | OK, I see what you mean | 13:16 |
| Saviq | <Saviq> "I didn't manage to catch more..." <- So this happened when plugging / unplugging a display:... (full message at <https://libera.ems.host/_matrix/media/r0/download/libera.chat/8a30e7bfb09d0fd92599272f61178eae886aee91>) | 14:02 |
| Saviq | Again, it continued to a SIGABRT on its own. Grr. How do I catch it at the SIGSEGV? | 14:02 |
| alan_g[m] | Saviq: Maybe https://discourse.ubuntu.com/t/debugging-using-rr/17505 | 14:04 |
| Saviq | Will try that. | 14:05 |
| alan_g[m] | That's EOD folks! Interact with you tomorrow... o/ | 17:01 |
Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!
