00:11 <alyssa> it is concerning how good i'm getting at supertuxkart from all this "testing"

00:13 <cr1901> nullptr::live tourney when?

00:19 <alyssa> i'm down

00:23 <alyssa> i should not be finding the expert challenges "about a medium" ...

00:45 <alyssa> oh and I need tests for compressed border formats

00:45 <alyssa> of course ... I couldn't get those to work ..

00:55 <alyssa> but I seem to be able to guess them

00:56 <alyssa> but I seem to be able to guess them

00:56 <alyssa> well some of them

00:56 <alyssa> ETC2 RGB8 == RGB8

00:56 <alyssa> ETC2 RGBA8 == RGBA8

00:56 <alyssa> ETC2 RGB8A1 == RGBA8

00:56 <alyssa> sRGB " == sRGB "

00:57 <alyssa> not sure what R11 maps to yet

00:58 <alyssa> unless it's 11-bit unorm internally ... nn

01:02 <alyssa> oh, yeah. it is. What a delight.

01:19 <alyssa> i hate border colours so much

01:36 <alyssa> ok. have a branch passing the relevant deqp's

01:36 <alyssa> but certainly not the relevant piglits (-:

01:44 <alyssa> i feel like these piglit results get worse every time i run them

01:44 <alyssa> i guess that's the problem with enabling more exts

01:45 <alyssa> including known broken stuff

01:45 <alyssa> (this is agx/next, going to have a lot less fails/crashes on the release obviously)

01:46 <alyssa> fewer fails than Dec 22, there's that :p

01:46 <alyssa> Dec 22 was peak broken

01:46 <alyssa> at 2716 fails+crash

01:46 <alyssa> down to 2484, yeet

01:47 <alyssa> 1153 of those are clipflat

01:47 <alyssa> of the remaining 1331..

01:48 <alyssa> 199 transform feedback

01:49 <alyssa> pile of texwrap fails for compression, yep that was expected given the state of this code

01:49 <alyssa> and 1d

12:05 <eric_engestrom> lina: your "kmsro: Fix renderonly_scanout BO aliasing" fixed other issues as well, so thanks for that, but unfortunately it also breaks the build on systems where valgrind.h isn't directly in /usr/include/ (eg. on my system it's at /usr/include/valgrind/valgrind.h)

12:05 <eric_engestrom> lina: could you take a look at the fix (https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/20530) and r-b/ack it? it simply adds the dependency in meson where needed (some drivers already had it)

12:24 <eric_engestrom> (where "my system" = arch, which I believe you're also on, so you can test this locally with `pacman -S valgrind` and see that mesa doesn't build anymore without that MR)

13:59 <lina> eric_engestrom: I've actually had that problem with the Asahi driver all along and always wondered why it only broke for me... never got around to looking at it...

14:02 <eric_engestrom> haha

14:02 <eric_engestrom> feel free to ping me next time you have mesa build issues :)

14:03 <lina> Thank you! ^^

14:04 <lina> I just tested it and it fixes the asahi/main build with valgrind enabled!

14:13 <eric_engestrom> thanks! merging ✨

18:54 <zzywysm> alyssa and lina: disregard me if you're already on top of this, but you probably don't want to let the commits between these two branches get too out of sync:

18:54 <zzywysm> https://gitlab.freedesktop.org/mesa/mesa/-/commits/main/src/asahi

18:54 <zzywysm> https://gitlab.freedesktop.org/asahi/mesa/-/commits/main/src/asahi