#
…

2021-04-23

00:01odmir has joined #asahi-gpu
00:01odmir has quit [Client Quit]
00:21odmir has joined #asahi-gpu
00:27odmir has quit [Remote host closed the connection]
00:28odmir has joined #asahi-gpu
00:33odmir has quit [Ping timeout: 268 seconds]
00:56odmir has joined #asahi-gpu
01:26odmir has quit [Remote host closed the connection]
01:27odmir has joined #asahi-gpu
01:32odmir has quit [Ping timeout: 260 seconds]
01:34DarkShadow44 has quit [Quit: Free ZNC ~ Powered by LunarBNC: https://LunarBNC.net]
01:34DarkShadow44 has joined #asahi-gpu
01:38odmir has joined #asahi-gpu
01:39DarkShadow44 has quit [Client Quit]
01:40DarkShadow44 has joined #asahi-gpu
01:43odmir has quit [Ping timeout: 240 seconds]
01:50odmir has joined #asahi-gpu
01:54odmir has quit [Ping timeout: 260 seconds]
02:02odmir has joined #asahi-gpu
02:25odmir has quit [Ping timeout: 268 seconds]
02:38tbodt has quit [Quit: ZNC 1.6.6+deb1ubuntu0.2 - http://znc.in]
02:39tbodt has joined #asahi-gpu
02:49bloomok, checked in cmdstream defs for 2d textures and samplers
02:49bloomvery much incomplete but good enough to get started
02:52phiologe has quit [Ping timeout: 250 seconds]
02:53dougallawesome! looking forward to trying that out
02:54phiologe has joined #asahi-gpu
02:54bloomnot piped into the demo yet
02:54bloomthat won't be too hard but I'll wait till the ISA side is known
02:55bloomrn im more concerned about framebuffers
03:00tbodt has quit [Quit: ZNC 1.6.6+deb1ubuntu0.2 - http://znc.in]
03:01tbodt has joined #asahi-gpu
03:08tbodt has quit [Quit: ZNC 1.7.5+deb4 - https://znc.in]
03:08tbodt has joined #asahi-gpu
03:45bloommy "Render target" descriptor looks like a compact flavour of the texture descriptor
03:45bloomat least I know how formats work
04:38Graypup_ has joined #asahi-gpu
08:25tbodt has quit [Quit: ZNC 1.7.5+deb4 - https://znc.in]
08:26tbodt has joined #asahi-gpu
08:53tmlind_ is now known as tmlind
10:04eric_engestrom has quit [Ping timeout: 246 seconds]
10:06eric_engestrom has joined #asahi-gpu
15:38choozy has joined #asahi-gpu
15:40choozyI read the blog post number 3 from Alyssa Anna and it stated the GPU can allocate up to 4785MB in it's registers. Is this piece of memory used and maybe shared with the system RAM from the M1?
15:41choozyAlyssa Anne*, excuse me
15:48sveni think it's ~5MB and not ~5000MB
15:48svenand that's the total size of all register combined
16:00rockinrobstar[m] has quit [Quit: Idle for 30+ days]
17:03bloomsven: is right
17:03bloomunfortunately not a number we can *confirm* but it's a solid guess
17:04bloomand it's not in system RAM, it's a dedicated register file
18:06odmir has joined #asahi-gpu
18:21odmir has quit [Remote host closed the connection]
18:22odmir has joined #asahi-gpu
18:27odmir has quit [Ping timeout: 260 seconds]
18:53odmir has joined #asahi-gpu
19:29odmir has quit [Ping timeout: 268 seconds]
19:41odmir has joined #asahi-gpu
19:47macc24 has quit [Ping timeout: 250 seconds]
20:10macc24 has joined #asahi-gpu
20:15odmir has quit [Ping timeout: 260 seconds]
20:27odmir has joined #asahi-gpu
20:59odmir has quit [Ping timeout: 268 seconds]
21:14odmir has joined #asahi-gpu
21:47odmir has quit [Ping timeout: 240 seconds]
22:16choozy has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
22:20odmir has joined #asahi-gpu
22:50bloomTrying to wire textures into the spinning cube dmeo
22:52nimroot[m]bloom: are you able to estimate how long it might take for first MR to mesa? Months? Years?
22:52bloomI made my first MR to mesa when Mesa first moved to GitLab a few years back :-p
22:54odmir has quit [Ping timeout: 265 seconds]
22:54bloom(Actually, there was a gap. Panfrost was a mailing list hold out. https://gitlab.freedesktop.org/mesa/mesa/-/merge_requests/1120 was apparently my frist.)
22:55nimroot[m]Oh, sorry, didn't make my question precise. I meant the first MR for M1 support : D
22:58bloomUpstreaming into Mesa isn't a big milestone for me, I guess
22:58nimroot[m]Wow, missed it somehow. So, the first (docs) step is behind us, can't wait for proper hardware support.
22:58bloomIf I wanted I could push the code now and nobody would complain.
22:58nimroot[m]I'd love to help somehow, but all of my experience is python web backend dev : /
22:59bloomBut right now the extra overhead of upstream dev for an early project is more than the overhead of rebasing things.
22:59bloomI'll upstream when the curves intersect.
…