#
✕ nick:marcan
#asahi-dev · 2023-04-17 07:47
marcan*14"
#asahi-dev · 2023-04-17 07:47
marcanI assume you mean 24"
#asahi-dev · 2023-04-17 07:46
marcankaazoo: there is no 13" M2 Pro
#asahi-dev · 2023-04-16 16:38
marcanthe dva is >> 12 in the message
#asahi-dev · 2023-04-16 16:37
marcanneeds the left shift
#asahi-dev · 2023-04-16 16:29
marcanI wonder what that 0xff8 thing is about though...
#asahi-dev · 2023-04-16 16:29
marcanOK, so that's what the old logic was for, pre-assigned addresses. so we need to check for dva != 0 and reintroduce the old code.
#asahi-dev · 2023-04-16 15:26
marcanyeah, that would make sense
#asahi-dev · 2023-04-16 15:16
marcanbut on the other hand, I sure would expect a mute to happen before that anyway
#asahi-dev · 2023-04-16 15:15
marcanpovik: cs43130.c has a hw_free callback that switches to internal clock, so maybe that's the right way
#asahi-dev · 2023-04-16 15:03
marcanI won't claim to know what the "right" thing is here but that's what's happening right now
#asahi-dev · 2023-04-16 15:03
marcanso clearly hw_free alone isn't going to work unless we add an explicit mute path in there
#asahi-dev · 2023-04-16 15:02
marcanhw_free is getting called from the ioctl path, while the mute and later shutdown are getting called from the device close path
#asahi-dev · 2023-04-16 15:01
marcanwell it is calling it after prepare :p
#asahi-dev · 2023-04-16 15:01
marcanso if you want to do it that way the codec drivers would have to mute on hw_free too (I think some do)
#asahi-dev · 2023-04-16 15:01
marcanhw_free is an ioctl that userspace can send at any time AIUI, and which does not trigger a mute
#asahi-dev · 2023-04-16 15:00
marcanhttps://mrcn.st/p/qAy7Uwdw one of the logs
#asahi-dev · 2023-04-16 14:55
marcanhw_free is symmetrical to hw_params AIUI
#asahi-dev · 2023-04-16 14:55
marcanshutdown is called when the device is closed
#asahi-dev · 2023-04-16 14:55
marcanpovik: I added WARN_ON(1) to get stack traces and printks to log what is happening
#asahi-dev · 2023-04-16 13:05
marcanspi though, I think that would get matched already?
#asahi-dev · 2023-04-16 13:04
marcanso yeah, I guess this doesn't block dockchannel
#asahi-dev · 2023-04-16 13:04
marcanif it's just the default keyboard it's fine, it won't match the apple one nor do Fn emu
#asahi-dev · 2023-04-16 13:00
marcanOTOH I could just upstream the keyboard stuff with the no Fn emulation idea to begin with, even if we still carry downstream patches to add emulation until the xkb stuff is ready
#asahi-dev · 2023-04-16 12:59
marcanif we add this driver, I think the keyboard will work to some extent even without other patches, and then we might end up in trouble
#asahi-dev · 2023-04-16 12:59
marcanbecause right now the keyboards don't work *at all*, which means doing whatever I want with Fn is not a regression
#asahi-dev · 2023-04-16 12:59
marcanand yeah, I need to clean up the shutdown path and submit this to linux, *however*, I'm partially slacking on that on purpose so we can figure out the Fn key story
#asahi-dev · 2023-04-16 12:58
marcannah, we're still good for the size, as long as they don't double it
#asahi-dev · 2023-04-16 12:56
marcanthey did increase the init message lengths though, I hope we aren't going to overflow the fifo with our trick for that...
#asahi-dev · 2023-04-16 12:55
marcanwe don't even parse that stuff in u-boot
#asahi-dev · 2023-04-16 12:55
marcanoh actually, don't need to
#asahi-dev · 2023-04-16 12:54
marcanI can do that one
#asahi-dev · 2023-04-16 12:54
marcanalso I guess I should fix mtp there too
#asahi-dev · 2023-04-16 12:53
marcan(the GPU sign-extends so it needs the high bit)
#asahi-dev · 2023-04-16 12:53
marcanoh yeah, APPLE_RTKIT_BUFFER_REQUEST_IOVA should be 43, 0 (you don't care, only the GPU cares about those high bits, but just out of technicality)
#asahi-dev · 2023-04-16 12:52
marcanthey already had 44 bits, which is enough for current machines with up to 42-bit address space everywhere
#asahi-dev · 2023-04-16 12:52
marcanyeah, it's weird...
#asahi-dev · 2023-04-16 12:49
marcanit's like the other buffer requests except the field defs are different and the address is in 4K pages
#asahi-dev · 2023-04-16 12:49
marcankettenis: if you feel like implementing oslog in u-boot, https://github.com/AsahiLinux/linux/commit/ea40aa13292f59a90688955bee581dbc3037fdd0
#asahi-dev · 2023-04-16 12:47
marcanthat's quite the crash
#asahi-dev · 2023-04-16 12:47
marcanhuh, wow.
#asahi-dev · 2023-04-16 12:45
marcanhaven't tested it as stage1 though
#asahi-dev · 2023-04-16 12:44
marcanjannau: seems to work for me with what's in m1n1:main (drop the logic, keep the ep start)
#asahi-dev · 2023-04-16 12:42
marcanno idea
#asahi-dev · 2023-04-16 12:39
marcanjannau: how does it break?
#asahi-dev · 2023-04-16 12:24
marcanthe change was backwards-compatible, I had just misinterpreted the protocol
#asahi-dev · 2023-04-16 12:23
marcanjannau: pushed m1n1 and the bringup branch, mtp should work now
#asahi-dev · 2023-04-16 12:15
marcanm1n1 uses it for what looks like a bug workaround which *may* be needed in linux eventually but I'm not sure, we can always add it later
#asahi-dev · 2023-04-16 12:15
marcandidn't I drop that?
#asahi-dev · 2023-04-16 12:10
marcanyeah, so we can drop it then

plus ›