/srv/irclogs.ubuntu.com/2022/06/16/#mir-server.txt

sophie-wRAOF (he/his): if mangled symbols can be different for different architectures, and our libmircore2.symbols file is the same for every architecture (because it's checked into git), does that mean putting mangled symbols into it is definitely wrong?00:58
RAOFsophie-w: You can put mangled symbols into it, just as long as they're gated on the relevant architecture (like the `arch-bits=32` bits)01:00
RAOFYou can also just have a different `.symbols` file or architecture.01:01
RAOFs/or/per/01:02
RAOFOh, wow! I think this gbm-kms platform test might actually have caught a bug in my code!06:35
SaviqGood mo'in06:52
alan_g[m]<RAOF> "You can put mangled symbols into..." <- That's how we've done it for miroil and now mircore07:06
RAOFUrgh. Putting a breakpoint on `mmap` is less than helpful 😬08:04
RAOF<alan_g[m]> "That's how we've done it for..." <- We do that because the *unmangled* symbol is different, but the same principle applies, yeah.08:05
alan_g[m]RAOF: Why would we use mangled symbols?!08:07
* alan_g[m] didin't read carefully08:08
RAOF<alan_g[m]> "Why would we use mangled symbols..." <- Because then we could list the *actual* symbols, rather than requiring a bunch of wildcards.08:17
* RAOF almost let autocomplete spell that “wildcats”08:17
RAOFThis bit is not actually important, though. I'll raise the interesting bit in m-sync this evening.08:18
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** Saviq drafted [pull request #70](https://github.com/MirServer/ubuntu-frame/pull/70): auth: add `ubuntu-frame-vnc` authorizations08:52
-GitHub[m]:#mir-server- **[MirServer/mir]** bors[bot] merged [pull request #2299](https://github.com/MirServer/mir/pull/2299): Filter 2 distinct bogus touch event scenarios09:36
-GitHub[m]:#mir-server- **[MirServer/mir]** Saviq opened [issue #2476](https://github.com/MirServer/mir/issues/2476): Average FPS exceeds that of screen refresh on a Pi... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/54ff01158f9c9ba019e0aae06bf61c962f17ec89)09:44
-GitHub[m]:#mir-server- **[MirServer/mir]** Saviq edited [issue #2476](https://github.com/MirServer/mir/issues/2476): Average FPS exceeds that of screen refresh on a Pi09:45
-GitHub[m]:#mir-server- **[MirServer/mir]** Saviq edited [issue #2476](https://github.com/MirServer/mir/issues/2476): Compositor average FPS reported exceeds that of screen refresh on a Pi09:45
alan_g[m]Frame dropping?09:46
SaviqLatency reported well below the 16.6ms threshold… also resolution is very low, so it should be enough oomph?09:52
Saviq> Maybe will try to backport xkbcommon09:56
SaviqThis seems to have worked \o/09:56
SaviqGot a core22-based snap with GPU acceleration (or at least no complaints about _no_ acceleration).09:56
SaviqHad to disable h264 for now, libav too old in focal and didn't bother to build it yet.09:56
SaviqIt should start building here in a moment https://launchpad.net/~mir-team/+snap/ubuntu-frame-vnc-edge09:56
alan_g[m]Saviq: This seems to have worked \o/09:56
Saviqs/core22/core20/!09:56
alan_g[m]Got a core22-based snap with GPU acceleration (or at least no complaints about _no_ acceleration).09:56
alan_g[m]Had to disable h264 for now, libav too old in focal and didn't bother to build it yet.09:56
alan_g[m]backporting and core22?09:56
SaviqMuscle memory, sorry :) Meant core20.09:57
alan_g[m]Just checking09:57
Saviqhttps://github.com/MirServer/ubuntu-frame-vnc/commit/544554834199151bdcb2a94a9285c173d676739c09:58
Saviqbbl09:58
-GitHub[m]:#mir-server- **[MirServer/mir]** AlanGriffiths opened [pull request #2477](https://github.com/MirServer/mir/pull/2477): Tidy up eglstream probe10:10
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** AlanGriffiths closed [pull request #69](https://github.com/MirServer/ubuntu-frame/pull/69): Workaround gbm-kms/Nvidia weirdness11:29
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** Saviq drafted [pull request #71](https://github.com/MirServer/ubuntu-frame/pull/71): github: migrate to new Snapcraft login method11:42
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** Saviq requested a review from AlanGriffiths for [pull request #71](https://github.com/MirServer/ubuntu-frame/pull/71): github: migrate to new Snapcraft login method11:49
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** Saviq marked [pull request #71](https://github.com/MirServer/ubuntu-frame/pull/71): github: migrate to new Snapcraft login method as ready for review11:49
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** bors[bot] edited [pull request #70](https://github.com/MirServer/ubuntu-frame/pull/70): [Merged by Bors] - auth: add `ubuntu-frame-vnc` authorizations11:49
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** bors[bot] closed [pull request #70](https://github.com/MirServer/ubuntu-frame/pull/70): [Merged by Bors] - auth: add `ubuntu-frame-vnc` authorizations11:49
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** bors[bot] edited [pull request #71](https://github.com/MirServer/ubuntu-frame/pull/71): [Merged by Bors] - github: migrate to new Snapcraft login method11:55
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** bors[bot] closed [pull request #71](https://github.com/MirServer/ubuntu-frame/pull/71): [Merged by Bors] - github: migrate to new Snapcraft login method11:55
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** AlanGriffiths opened [issue #72](https://github.com/MirServer/ubuntu-frame/issues/72): Does not pick up "mir-quirks" from a graphics provider until `configure` runs... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/b26f0d7a5a62b0af84a5f47e3bb81eae7a81f74e)13:32
-GitHub[m]:#mir-server- **[MirServer/mir]** Saviq drafted [pull request #2478](https://github.com/MirServer/mir/pull/2478): github: use canonical/actions/build-snap13:37
-GitHub[m]:#mir-server- **[MirServer/mir]** Saviq marked [pull request #2478](https://github.com/MirServer/mir/pull/2478): github: use canonical/actions/build-snap as ready for review13:37
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** Saviq opened [pull request #73](https://github.com/MirServer/ubuntu-frame/pull/73): Fix snapcraft auth13:41
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** Saviq edited [pull request #73](https://github.com/MirServer/ubuntu-frame/pull/73): github: use build/close-snap actions13:46
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** AlanGriffiths opened [pull request #74](https://github.com/MirServer/ubuntu-frame/pull/74): Workaround gbm-kms/Nvidia weirdness... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/e4b2122c0cbbf915896890c3d296fd6d01f19ca1)13:53
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** bors[bot] edited [pull request #74](https://github.com/MirServer/ubuntu-frame/pull/74): [Merged by Bors] - Workaround gbm-kms/Nvidia weirdness14:33
-GitHub[m]:#mir-server- **[MirServer/ubuntu-frame]** bors[bot] closed [pull request #74](https://github.com/MirServer/ubuntu-frame/pull/74): [Merged by Bors] - Workaround gbm-kms/Nvidia weirdness14:33
-GitHub[m]:#mir-server- **[MirServer/mir]** frank-dspeed closed [issue #2467](https://github.com/MirServer/mir/issues/2467): Auto-selection of X11 platforms fails with Nvidia drivers (470 and 515) using LXDE14:38
Saviqalan_g made a mess, can you please push 5b4d3b8 as mir-kiosk master?15:04
alan_g[m]Huh?! OK, will se if that makes sense...15:05
SaviqI forced a temp branch to master, not pulled beforehand so don't have the commit referenced anywhere…15:06
alan_g[m]There15:09
Saviq:thankyou:15:10
SaviqOK that's all for now15:20
SaviqTalk tmrw o/15:20
alan_g[m]Have a good one!15:20
-GitHub[m]:#mir-server- **[MirServer/mir]** graysonguarino drafted [pull request #2479](https://github.com/MirServer/mir/pull/2479): Refactor event matchers20:59
-GitHub[m]:#mir-server- 20:59
-GitHub[m]:#mir-server- > Mir's event matchers are not descriptive with why tests are failing, they simply report a failure. This PR refactors the event matchers and allows them to be more descriptive.20:59
RAOFsophie: I'm not wild about `ProofOfMutexLock`, and I'm not sure we should be promoting it to mircore?23:46
RAOF(I've never really been wild about it, but it was a pragmatic low-effort partial solution to the problem of ensuring things are called under lock)23:46
RAOFBut particularly in BasicSurface there is just so much state that is guarded by the lock (or is it?)23:47
RAOFAnd now that we're adding a “trust me, this doesn't *need* locking” escape hatch to the `ProofOfMutexLock`, maybe we could instead just (work out what state needs to be locked and then) shuffle all the relevant state into a `struct State {}; mir::Mutex<State> state`?23:49
RAOFs/*need*/_need_/, s/`?/;`?/23:50
RAOF * And now that we're adding a “trust me, this doesn't _need_ locking” escape hatch to the `ProofOfMutexLock`, maybe we could instead just (work out what state needs to be locked and then) shuffle all the relevant state into a `struct State {…}; mir::Mutex<State> state;`?23:50
-GitHub[m]:#mir-server- **[MirServer/mir]** wmww opened [issue #2480](https://github.com/MirServer/mir/issues/2480): Surface damage notifications do not cover all cases... (full message at https://libera.ems.host/_matrix/media/r0/download/libera.chat/c51eb298300a7d7297e908381ebf1411f46626b1)23:51
sophie-w@[RAOF (he/his)] all the mutable state in BasicSurface is guarded by the mutex. I don't see all that much value in such a refactor, but I don't object to it23:55
RAOFReally? Huh. That's a lot of state guarded by the one mutex :)23:57

Generated by irclog2html.py 2.7 by Marius Gedminas - find it at mg.pov.lt!