#
…

2022-11-24

00:30alyssalina: "ail: Do not unconditionally page-align sizes" is breaking t-rex for me
00:32alyssaunclear why
00:33alyssaif I functionally revert it, works again
00:34alyssaI am unsure where you got the logic from
00:36alyssaif I align for uncompressed but not for twiddled, it seems happy
00:36alyssaalthough my unti test failing no
00:37alyssaok did the obvious fix and it's happy
00:38alyssaWorth noting that t-rex uses (IIRC) ETC2 or ASTC cube maps
00:38alyssawhich are uncompressed twiddled and the alignment matters
00:41alyssasplat for you
00:42alyssait should not be possible to provoke an rtkit crash from userspace so this makes me profoundly uncomfortable
00:46alyssaalso just hit the oom killer yikes
00:47alyssaleaking staging resources, just fixed that locally
00:49* alyssafixes stuff locally
00:50Tramtristthe only way to fix things really
00:53alyssaTramtrist: i've fixed plenty of stuff remotely :-p
00:57alyssacube maps + compression is botched because we can't do linear staging 2D Array resources
01:00alyssai've just disabled non-2D compression for now because that's nontrivial to workaround (though certainly doable)
01:05alyssasimilarly, compression to non-renderable formats doesn't work
01:05alyssain principle all formats are renderable if we want them to be, but..
01:06alyssait's worth seeing if Apple uses this compression for funny stuff that shouldn't be renderable :p
01:10le0n has joined #asahi-gpu
01:11ahubcap has joined #asahi-gpu
01:12alyssaok, bunch of fixes and we're back to no gles2 regressions
01:13alyssagles3 is still going but seems about as broken as usual
01:14millenialhacker has joined #asahi-gpu
01:15le0n_ has quit [Ping timeout: 480 seconds]
01:16ahubcap_ has joined #asahi-gpu
01:22millenialhacker has quit [Ping timeout: 480 seconds]
01:22ahubcap has quit [Ping timeout: 480 seconds]
01:30ahubcap_ has quit [Quit: Leaving...]
01:51millenialhacker has joined #asahi-gpu
01:59millenialhacker has quit [Ping timeout: 480 seconds]
03:31millenialhacker has joined #asahi-gpu
03:39millenialhacker has quit [Ping timeout: 480 seconds]
03:59alyssathe latter should help performance noticeably, and it's pretty well tested now
04:00alyssaI tested my current set of { GNOME, neverball, supertuxkart, quake, xonotic }
04:00alyssabut maybe other people will find other issues
04:00alyssajannau: yuka: You may wish to upgrade ^^
04:00alyssare performance, I have a LOT more patches where that came from but letting them trickle out to give time to test everything
04:02alyssa[My unstable Linux integration branch has asahi/main + NIR UBO/VBO lowering + jmp_exec_none + framebuffer compression + DISCARD_WHOLE_RESOURCE]
04:03alyssa(Seeing as 2 of those were written today and 1 of those I found big issues with as recently as this afternoon, they'll be a bit longer. But they're coming)
05:14CatDentures has quit [Quit: Konversation terminated!]
05:56trsohmers has quit [Ping timeout: 480 seconds]
05:57trsohmers has joined #asahi-gpu
06:03SSJ_GZ has joined #asahi-gpu
06:34millenialhacker has joined #asahi-gpu
06:42millenialhacker has quit [Ping timeout: 480 seconds]
06:54millenialhacker has joined #asahi-gpu
06:59jannaualyssa: works and is faster, thanks
07:02millenialhacker has quit [Ping timeout: 480 seconds]
07:06zimsneexh has joined #asahi-gpu
07:11zimsneexh has quit []
07:14zimsneexh has joined #asahi-gpu
07:28zimsneexh has quit [Quit: zimsneexh]
07:28zimsneexh_ has joined #asahi-gpu
07:28zimsneexh_ is now known as zimsneexh
07:30millenialhacker has joined #asahi-gpu
07:30capta1nt0ad has joined #asahi-gpu
07:38millenialhacker has quit [Ping timeout: 480 seconds]
07:42millenialhacker has joined #asahi-gpu
07:50millenialhacker has quit [Ping timeout: 480 seconds]
07:51zimsneexh has quit [Ping timeout: 480 seconds]
07:52millenialhacker has joined #asahi-gpu
08:12millenialhacker has quit [Ping timeout: 480 seconds]
08:34millenialhacker has joined #asahi-gpu
08:43millenialhacker has quit [Ping timeout: 480 seconds]
08:45millenialhacker has joined #asahi-gpu
08:46millenia_ has joined #asahi-gpu
08:53millenialhacker has quit [Ping timeout: 480 seconds]
08:54millenia_ has quit []
09:03MajorBiscuit has joined #asahi-gpu
09:07Major_Biscuit has joined #asahi-gpu
09:14MajorBiscuit has quit [Ping timeout: 480 seconds]
09:16a_plastic_bag has joined #asahi-gpu
09:29a_plastic_bag has quit [Quit: WeeChat 3.7.1]
11:02Major_Biscuit has quit [Ping timeout: 480 seconds]
11:14mpshi. newbie question about building kernel asahi gpu with rust. build failed. is it ok to post paste link with build log here?
11:19bisko has quit [Quit: My MacBook has gone to sleep. ZZZzzz…]
11:29Dcowmps: pm'd you
11:35capta1nt0ad has quit [Quit: Konversation terminated!]
11:36capta1nt0ad has joined #asahi-gpu
11:54cylm has joined #asahi-gpu
12:27mpsnice, Dcow helped me to solve build problem. thank you Dcow
12:38zimsneexh has joined #asahi-gpu
12:45DcowI love procrastinate from $work sometimes
12:48zimsneexh has quit [Quit: zimsneexh]
12:53mpsnow another question, does xorg works with gpu or wayland is only option
13:05Dcowwayland is preffered
13:09_jannau_but there could be an incompatibility with xorg/modeset
13:11mpsI've got this fatal error when tried to start X '(EE) Cannot run in framebuffer mode. Please specify busIDs for all framebuffer devices'
13:15Major_Biscuit has joined #asahi-gpu
13:21chadmed_ has joined #asahi-gpu
13:21chadmed_mps: wayland and xwayland both work on my end, but same as you no "real" x11
13:25mpsthanks all for info
13:45qdot has quit [Ping timeout: 480 seconds]
14:44bisko has joined #asahi-gpu
14:46alyssajannau: +1
14:46alyssai saw real x11 working once and have since forgotten how i did it :p
14:52bisko has quit [Ping timeout: 480 seconds]
14:58qdot has joined #asahi-gpu
15:05Jondus has joined #asahi-gpu
15:05JondusHello :)
15:06JondusI got Asahi Mesa fully built. So does that mean it setups Wayland to use Mesa or do I need to do that manually?
15:09JondusI ask because my Wayland app is still not GPU accelerated. .. Maybe I need to reboot though lul
15:12_jannau_how did you build it? if you want every application using it you have install it to replace the system mesa libraries
15:13Jondus$ mkdir build $ cd build $ meson .. $ sudo ninja install
15:13JondusThat should replace system mesa right?
15:14alyssa has quit [Quit: leaving]
15:20_jannau_not sure what mesa's default install path is
15:22JondusAh.. Anyway you know of to force it to system path then?
15:23DcowJondus: you need to enable gallium asahi
15:23Dcowwhen build mesa
15:23JondusDammit I knew something was off
15:23Dcow+ I assume you kernel already with DRM_ASAHI, right?
15:23JondusI'm using bleeding-edge
15:23Dcowit's not enabled there
15:24Jondusugh. Can you tell me how to enable DRM_ASAHI then please
15:24Dcowyou need to prepare kernel yourself
15:24JondusAh so I need to compile the kernel
15:27JondusDcow: Mind if I pop you a dm?
15:28DcowI have a meeting in few minutes, so probably won't answer
15:28JondusMy apologies no problem
15:29DcowBut general advice is read this chat logs and if you still have questions about building kernel process just wait until it reach -edge kernel =)
15:30DcowBut if you have some specific questions pm is welcome, I just answer later
15:30JondusPretty sure I just need to run make LLVM=1 DRM_ASAHI=1?
15:31Dcowread about configuring kernel's
15:31Jondusyep def in the wrong repo then
15:32JondusAnyway yah, I will do some more research
15:32JondusThanks for your time
15:33JondusI was in the right repo but wrong command. I've never actually had to build a kernel like this
15:40Jondusffs.. It was make config
15:41Jondusmake menuconfig*
15:48marcanplease take build tutorials to #asahi-offtopic or PMs. this channel is for GPU development, not installation instructions.
15:49marcanif you've never built a kernel before you probably shouldn't be trying out the GPU driver at this point.
15:51JondusI have build a kernel just this particular one. I will move to off-topic for further questions.
15:51marcanyou also need to rebuild m1n1 and update your DTs, which could leave your system unbootable.
15:51Jondusjust not this particular one*
15:51marcanunless you know what you are doing and how to recover from that, you probably shouldn't
16:03Dcowthat's a strong point^
16:09chadmed_ has quit [Remote host closed the connection]
16:25Major_Biscuit has quit [Ping timeout: 480 seconds]
16:53bisko has joined #asahi-gpu
17:14gaminglt has joined #asahi-gpu
17:15gaminglt has quit []
17:33milan has joined #asahi-gpu
17:34milan has quit []
17:34Znullptr has joined #asahi-gpu
17:51zalyx has quit [Quit: later alligator]
17:51zalyx has joined #asahi-gpu
18:50Cyrinux has quit [Remote host closed the connection]
19:06Jondus has quit [Quit: Page closed]
19:06Dcow has quit [Remote host closed the connection]
19:07Dcow has joined #asahi-gpu
19:08Dcow has quit []
19:28Cyrinux has joined #asahi-gpu
19:31Cyrinux has quit []
19:34Cyrinux has joined #asahi-gpu
19:49Major_Biscuit has joined #asahi-gpu
19:57kenzie7 has quit []
20:00kenzie7 has joined #asahi-gpu
20:08kenzie7 has quit []
20:09kenzie7 has joined #asahi-gpu
20:27Cyrinux has quit []
20:29Cyrinux has joined #asahi-gpu
20:35bisko has quit [Read error: Connection reset by peer]
20:54bisko has joined #asahi-gpu
21:22Dcow has joined #asahi-gpu
21:31erik3424 has joined #asahi-gpu
21:39Major_Biscuit has quit [Ping timeout: 480 seconds]
22:13Dcowalyssa: xonotic crash with asahi:main https://paste.gnome.org/QLosMwOdc
22:27Dcow../src/gallium/drivers/asahi/agx_state.c:1355
22:57alyssa has joined #asahi-gpu
22:57alyssadcow: Odd... is that reproducible?
22:58Dcowset ultra settings, play ~1 minute, usually earlier (second map, idk if it matter)
22:58alyssaoh I think I see the problem
22:58Dcowxonotic-sdl
22:59Cyrinux has quit [Remote host closed the connection]
22:59alyssatry replace PIPE_MAX_SAMPLERS a few lines above with ctx->stage[shader].sampler_count
22:59alyssalmk if that fixes it
22:59alyssaok yes that should fix it
23:00alyssachange PIPE_MAX_SAMPLERS to nr_samplers in line 1351
23:01alyssabind_sampler_states doesn't zero [nr_samplers, PIPE_MAX_SAMPLERS) so can get non-null garbage samplers leading to a use-after-free (segfault derefencing sampler) or a buffer overflow (writing samplers[] out of bounds, we only allocated room for nr_samplers of them) or both
23:01alyssathank you for the helpful bug report :-)
23:01* alyssareally appreciates bug reports that she can actually do something about
23:02alyssa"Reproducible segfault in this workload, here is the backtrace, here is the line number that's failing, I am using the stable tree with no patches" -- good
23:02alyssa"This workload used to work on the stable tree, but it no longer works on the stable tree. I've bisected the failure down to commit sha 123456. If I revert that commit on top of main, it works again." -- good
23:03alyssa"I just tried this random DirectX game and it doesn't work very good, here is an apitrace, please fix, cc @someone_banned_for_code_of_conduct_violations" -- not good
23:04SSJ_GZ has quit [Ping timeout: 480 seconds]
23:06Dcowyeah, that's do it. Thank you!
23:07mps has left #asahi-gpu [#asahi-gpu]
23:07Dcowsorry for delay, wanted to _test_ the whole match >_>
23:08Dcowlol xonotic have a good perf even under gdb!
23:09DcowI should go sleep or I miss start of l-i-n-a's stream again
23:10Cyrinux has joined #asahi-gpu
23:12Dcow has quit [Quit: dcow]
23:15alyssa+1, thanks!
23:16kenzie7 has quit []
23:41kenzie7 has joined #asahi-gpu
…