#asahi-dev on 2024-09-24 — irc logs at oftc.irclog.whitequark.org

00:02 pb17 has quit [Ping timeout: 480 seconds]

00:32 pb17 has joined #asahi-dev

01:00 KxCORP5894000345 has quit [Quit: Bye!]

01:00 ddxtanx has joined #asahi-dev

01:00 KxCORP5894000345 has joined #asahi-dev

01:13 opticron has quit [Ping timeout: 480 seconds]

01:14 ipatch has quit [Ping timeout: 480 seconds]

01:35 ipatch has joined #asahi-dev

01:46 ipatch has quit [Ping timeout: 480 seconds]

01:48 opticron has joined #asahi-dev

02:18 pb17 has quit [Ping timeout: 480 seconds]

02:23 alih has quit []

02:28 tobhe_ has joined #asahi-dev

02:34 tobhe has quit [Ping timeout: 480 seconds]

02:48 pb17 has joined #asahi-dev

03:54 nora_ has joined #asahi-dev

03:56 nora has quit [Ping timeout: 480 seconds]

04:20 john-cabaj has quit [Ping timeout: 480 seconds]

04:23 ipatch has joined #asahi-dev

04:33 pb17 has quit [Ping timeout: 480 seconds]

04:49 pb17 has joined #asahi-dev

05:29 ddxtanx has quit [Remote host closed the connection]

06:46 pb17 has quit [Ping timeout: 480 seconds]

07:03 sand4rt has joined #asahi-dev

07:08 pb17 has joined #asahi-dev

07:22 sand4rt has quit [Quit: WeeChat 4.4.2]

07:26 ddxtanx has joined #asahi-dev

08:00 sand4rt has joined #asahi-dev

08:12 opticron has quit [Read error: Connection reset by peer]

08:12 opticron has joined #asahi-dev

08:18 weitcis has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]

08:27 weitcis has joined #asahi-dev

08:29 nela has quit [Ping timeout: 480 seconds]

08:30 JayBeeFOSS has quit [Ping timeout: 480 seconds]

08:31 JayBeeFOSS has joined #asahi-dev

08:36 pb17 has quit [Ping timeout: 480 seconds]

08:55 pb17 has joined #asahi-dev

09:12 midou has joined #asahi-dev

09:46 allmazz has joined #asahi-dev

09:58 Tramtris1 has joined #asahi-dev

10:00 tramtrist has quit [Ping timeout: 480 seconds]

10:05 Tramtris1 is now known as Tramtrist

10:06 pb17 has quit [Ping timeout: 480 seconds]

10:30 ___nick___ has joined #asahi-dev

10:53 pb17 has joined #asahi-dev

10:59 qdot has joined #asahi-dev

11:02 rafsan has joined #asahi-dev

11:07 sand4rt has quit [Quit: WeeChat 4.4.2]

11:19 rafsan has quit [Quit: rafsan]

11:20 rafsan has joined #asahi-dev

11:43 rafsan has quit [Quit: rafsan]

12:07 pb17 has quit [Ping timeout: 480 seconds]

12:08 Tramtrist has quit [Remote host closed the connection]

12:19 sand4rt has joined #asahi-dev

12:48 pb17 has joined #asahi-dev

13:08 alih has joined #asahi-dev

13:42 john-cabaj has joined #asahi-dev

14:23 pb17 has quit [Ping timeout: 480 seconds]

14:32 malfunction54 has quit [Remote host closed the connection]

14:32 malfunction54 has joined #asahi-dev

14:47 sand4rt has quit [Ping timeout: 480 seconds]

14:50 <nickchan> chaos_princess: i think the brightness encoding is a bit wrong on a11

14:50 <nickchan> m1 /arm-io/mipi-dsim/lcddfr backlight-level-encoding = 8

14:51 <nickchan> a11 /arm-io/mipi-dsim/lcd = 11

14:51 <nickchan> a11 /arm-io/mipi-dsim/lcd backlight-level-encoding = 11

14:53 <chaos_princess> yea, i guess it has extra bits on primary display, and other devices send 2 bytes there https://elixir.bootlin.com/linux/v6.11/source/drivers/gpu/drm/drm_mipi_dsi.c#L1342

14:53 <chaos_princess> unfortunately, i do not have the pdf of the version of mipi dcs spec that has that command

14:54 <nickchan> I guess I need another compatible then

14:55 <nickchan> "apple,t8015-summit"

14:55 <chaos_princess> No need, just add a of property

14:55 <chaos_princess> https://github.com/AsahiLinux/linux/blob/bd0a1a7d465fcb60685a2360565ed424bafff354/drivers/gpu/drm/adp/panel-summit.c#L49

14:55 <chaos_princess> and fill it from there

14:55 <chaos_princess> m1 display should accept a 2 byte command too in theory

15:02 ipatch has quit [Quit: WeeChat 4.4.1]

15:03 pb17 has joined #asahi-dev

15:13 ipatch has joined #asahi-dev

15:19 <nickchan> okay now i just need to find the mmio that's being poked

15:19 <nickchan> so I can do random things in m1n1

15:29 <nickchan> ... which I have no idea how to

15:30 <nickchan> navigate layers and layers of helper functions

15:31 <chaos_princess> do you have a device tree?

15:32 <nickchan> chaos_princess: yes

15:33 <chaos_princess> in m1 device trees the relevant one is described as mipi-dsim

15:33 <chaos_princess> only one mmio address there

15:34 <chaos_princess> https://github.com/AsahiLinux/linux/blob/asahi/drivers/gpu/drm/adp/adp_drv.c#L72 - here are the relevant offsets

15:36 <chaos_princess> also, remind me later today, i might have some of my research notes saved

15:37 <nickchan> chaos_princess: figured it out

15:37 <nickchan> 2 byte write

15:37 <nickchan> brightness 0 - 2047

15:38 <nickchan> encoded as a short

15:38 <nickchan> uint16

15:38 <chaos_princess> makes sense, log2(2048) is 1

15:38 <chaos_princess> *11

15:38 <nickchan> i wonder if m1 would survive two byte writes

15:38 <nickchan> assume < 256

15:39 <chaos_princess> 99% sure it will, from poking at it in general, it implements a lot of mipi spec

15:39 <nickchan> if that is ok on m1 we can fill max brightness from dt

15:54 sand4rt has joined #asahi-dev

16:00 <nickchan> chaos_princess: https://github.com/asdfugil/linux-apple/commit/5e41e1ab0cac1fbc04c1c84ffead3b8c904d306e https://github.com/asdfugil/linux-apple/commit/f5ee205c2b8a023aaa5e637e10808a838e7968be hopefully these changes won't break the m1

16:01 <chaos_princess> sure, will test in a bit

16:05 ipatch has quit [Ping timeout: 480 seconds]

16:06 <nickchan> wait i forgot to change the width in summit_suspend

16:08 <nickchan> https://github.com/asdfugil/linux-apple/commits/apple/bits/060-adp/ okay just pick the most recent two non-dts commit

16:09 <chaos_princess> since you are touching it all anyway, it might make sense to use this helper https://elixir.bootlin.com/linux/v6.11/source/drivers/gpu/drm/drm_mipi_dsi.c#L1339

16:10 <nickchan> chaos_princess: that helper specifically doesn't work here

16:10 <chaos_princess> oh?

16:10 <nickchan> because we have more than 256 states

16:10 <nickchan> on some panels

16:10 <nickchan> anyways

16:10 <nickchan> okay wait

16:10 <nickchan> i misread the code

16:10 <chaos_princess> but it does write 16 bits, no?

16:11 <nickchan> I will try making the change

16:15 pb17 has quit [Ping timeout: 480 seconds]

16:18 <nickchan> chaos_princess: works on a11 and bushed to the branch above

16:18 <nickchan> pushed

16:19 <chaos_princess> kk

16:30 <nickchan> for the panel apple switched from samsung to synopsys in a8

16:31 <nickchan> I hope h6-display-pipe isn't too big of a difference

16:31 <nickchan> also apple,h7-display-pipe doesn't sound right...

16:31 <nickchan> the compatible

16:32 <nickchan> really should have been apple,t7000-display-pipe

16:32 <nickchan> especially if t7000 is going to be supported by the kernel

16:34 <nickchan> not going to propose any changes to the compaitble until I have the whole driver working though

16:34 <nickchan> currently it is just backlight

16:41 <chaos_princess> now im kinda sad that my iphone 10 went spicy pillow and shattered the screen some time ago :P

16:45 <nickchan> the amount of stuff that works or almost works is amazing

16:46 <nickchan> the amount of blocks that's still seemingly compatible going back 10 generation to A7 is very sigificant

16:47 <nickchan> DART isn't like that though so that's something to fiddle around with

16:48 <nickchan> well I think it still isn't *too* different

16:48 billak has joined #asahi-dev

16:50 <nickchan> the otg role-switch controller though, is likely completely different

16:50 <nickchan> and it is not some IP block either

17:01 billak has quit [Quit: Konversation terminated!]

17:10 ipatch has joined #asahi-dev

17:21 pb17 has joined #asahi-dev

17:40 ipatch has quit [Ping timeout: 480 seconds]

17:57 ipatch has joined #asahi-dev

18:32 balrog_ has joined #asahi-dev

18:32 balrog has quit [Ping timeout: 480 seconds]

18:38 sand4rt has quit [Ping timeout: 480 seconds]

18:54 ___nick___ has quit []

18:56 ___nick___ has joined #asahi-dev

18:57 ___nick___ has quit []

18:58 <jannau> tobhe_: do you have usb proxy access to m1n1 on the black screen sequoia/m1n1-1.415 macbook? m1n1 log would be helpful

18:59 ___nick___ has joined #asahi-dev

19:03 ipatch has quit [Ping timeout: 480 seconds]

19:06 pb17 has quit [Ping timeout: 480 seconds]

19:24 ipatch has joined #asahi-dev

19:51 pb17 has joined #asahi-dev

20:04 ___nick___ has quit [Remote host closed the connection]

20:09 <tobhe_> have never set that up but I could try

20:09 tobhe_ is now known as tobhe

20:23 sand4rt has joined #asahi-dev

20:27 <jannau> tobhe: if you're kernel has CONFIG_MTD_PHRAM you could patch your 2nd stage m1n1 with https://github.com/AsahiLinux/m1n1/pull/399 if you have ssh access

20:28 <jannau> kernel logs form a boot with black screen would might be helpful

20:28 <tobhe> I can easily get those

20:28 <jannau> probably should create an issue in the m1n1 repository to collect information

20:29 <tobhe> yes, i was going to ask

20:29 <tobhe> if you create one I can add all my info or I can create one if you prefer that

20:29 <tobhe> I'll try the m1n1 stage 2 patch

20:29 <jannau> yes, currently in process

20:44 <jannau> tobhe: https://github.com/AsahiLinux/m1n1/issues/406

20:44 <tobhe> thx

21:01 sand4rt has quit [Ping timeout: 480 seconds]

21:03 cow123 has joined #asahi-dev

21:23 chrisl has quit [Remote host closed the connection]

21:23 chrisl has joined #asahi-dev

21:26 sand4rt has joined #asahi-dev

21:34 ipatch has quit [Ping timeout: 480 seconds]

21:44 pb17 has quit [Ping timeout: 480 seconds]

21:50 sand4rt has quit [Ping timeout: 480 seconds]

22:08 <tobhe> your m1n1 patches work. interestingly I can't trigger the problem now, instead of being entirely dark the backlight is off during boot and comes back with my graphical session starting.

22:09 cow123 has quit [Ping timeout: 480 seconds]

22:14 john-cabaj has quit [Ping timeout: 480 seconds]

22:18 cow123 has joined #asahi-dev

22:25 pb17 has joined #asahi-dev

22:25 ddxtanx has quit [Remote host closed the connection]

22:49 <jannau> tobhe: maybe that's the issue?

22:51 <tobhe> added some more context in a comment

22:52 <tobhe> the backlight thing certainly didn't happen before

22:54 <jannau> it turning on either on dcp linux driver init or Displaymanager start?

22:55 <jannau> if you can still find apple-dcp log messages for the weird boots please attach them in the issue

23:03 sting has joined #asahi-dev

23:03 <tobhe> I'm pretty sure gnome is what resets the backlight

23:03 <tobhe> I'll look for logs

23:04 ipatch has joined #asahi-dev

23:05 kettenis has joined #asahi-dev

23:27 <tobhe> added some. hope that helps

23:49 pb17 has quit [Ping timeout: 480 seconds]