marcan changed the topic of #asahi-dev to: Asahi Linux: porting Linux to Apple Silicon macs | General development | GitHub: https://alx.sh/g | Wiki: https://alx.sh/w | Logs: https://alx.sh/l/asahi-dev
dougall_ is now known as dougall
zkrx has quit [Ping timeout: 256 seconds]
choozy has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
really2 has joined #asahi-dev
really2 has quit [Remote host closed the connection]
zkrx has joined #asahi-dev
Stary is now known as Stary2001
Stary2001 is now known as Stary
<marcan> sven: I want it to support both, in parallel. I was already thinking about that yesterday :-)
<marcan> but get your USB backend merged; I'll refactor all of that to make it work like that
<marcan> it just means making the uartproxy stuff into a state machine, which really only has to handle the command header (I can probably assume that once I get the magic command header in via an input, that I can block on the rest rom there)
<marcan> same with the console, which will also cover the FB. I'll add a dispatcher or something and a function to let you turn console outputs on and off for performance
jeffmiw has joined #asahi-dev
jeffmiw has quit [Ping timeout: 256 seconds]
<marcan> *from
jeffmiw has joined #asahi-dev
jeffmiw has quit [Ping timeout: 265 seconds]
phiologe has quit [Quit: https://quassel-irc.org - Chat comfortably. Anywhere.]
phiologe has joined #asahi-dev
phiologe has quit [Ping timeout: 260 seconds]
odmir has quit [Remote host closed the connection]
phiologe has joined #asahi-dev
KindOne has quit [Ping timeout: 272 seconds]
KindOne has joined #asahi-dev
KindOne has quit [Ping timeout: 260 seconds]
KindTwo has joined #asahi-dev
KindTwo is now known as KindOne
mxw39 has quit [Quit: Konversation terminated!]
VinDuv has joined #asahi-dev
<sven> marcan: sounds good. regarding the console: we'll also need a way to disable it for certain prints. (e.g. printing to usb from within the usb code might be bad a idea)
<marcan> I bet I can handle that entirely within the console code
<marcan> add some reentrance check or something
<marcan> per-target
jeffmiw has joined #asahi-dev
jeffmiw has quit [Remote host closed the connection]
jeffmiw has joined #asahi-dev
jeffmiw has quit [Remote host closed the connection]
jeffmiw has joined #asahi-dev
jeffmiw has quit [Remote host closed the connection]
jeffmiw has joined #asahi-dev
jeffmiw_ has joined #asahi-dev
jeffmiw has quit [Read error: Connection reset by peer]
jeffmiw_ has quit [Remote host closed the connection]
jeffmiw has joined #asahi-dev
jeffmiw has quit [Ping timeout: 260 seconds]
<sven> https://github.com/AsahiLinux/m1n1/pull/36 so that's essentially the usb code without the PHY bringup (because I can't really sign that off) and the uartproxy/console integration
<sven> and https://github.com/svenpeter42/m1n1/tree/usb-dwc3-serial-wip is with phy bringup (based on pongoos/kext disassembly :() and a quick uartproxy hack
<marcan> sven: this problem will go away once we have mmiotrace, but for the time being since this is kind of a chicken-and-egg there, if you can just add a printf() to all the read32/write32 calls and run it and paste a dump somewhere, we can call that clean
<marcan> I'll write a version based on that
<marcan> (will look into this later, have some unrelated last-minute stuff going on today)
<sven> marcan: https://f.svpe.de/749afd07925249e765a83887ad2281d363ba7e81a764dba9b75ecf3742c450de_phy_bringup.txt and that one is with the reg ranges extracted from the device tree
<marcan> thanks!
dhewg has quit [Ping timeout: 265 seconds]
dhewg has joined #asahi-dev
<sven> loading kernels with this is fun :>
<sven> the bottleneck now actually is the reboot + building the kernel itself instead of waiting for the upload to finish
<marcan> yup
<marcan> parallelize those :D
<marcan> also ccache helps
_whitelogger has joined #asahi-dev
<kettenis> regarding my earlier message: the secondary CPUs defenitely spin up with the MMU (and caches) disabled
jeffmiw has joined #asahi-dev
jeffmiw has quit [Client Quit]
dottedmag has quit [Quit: QUIT]
dottedmag has joined #asahi-dev
dottedmag has joined #asahi-dev
taziden has quit [Ping timeout: 264 seconds]
taziden has joined #asahi-dev
jeffmiw has joined #asahi-dev
zkrx has quit [Ping timeout: 245 seconds]
KindOne has quit [Ping timeout: 272 seconds]
<jeffmiw> sven: thanks a lot, playing with it right now, it is enumerating, got a /dev/ttyACM0, can't get anything else yet, proxy shell or proxy linux are getting some proxy.UartTimeout checking what is wrong with my setup
KindOne has joined #asahi-dev
<jeffmiw> Looking at how to get the fb-console to debug also
<sven> jeffmiw: does lsusb show the device with the string descriptors?
<sven> i've only tested this with a mac os host so far but it should also work with linux
<sven> hm, strange. i had to run shell.py twice before it work on a linux host
<sven> or maybe not
<sven> seems to work fine even with a linux host here
<kettenis> marcan: seems the memory where the secondar CPUs are spinning isn't reserved?
jeffmiw has quit [Ping timeout: 265 seconds]
jeffmiw has joined #asahi-dev
<jeffmiw> yes lsusb shows the device correctly
<jeffmiw> probably something with my setup
<sven> nothing strange in dmesg either?
zkrx has joined #asahi-dev
bastilian has quit [Quit: Idle for 30+ days]
really2 has joined #asahi-dev
odmir has joined #asahi-dev
<jeffmiw> sven: no nothing strange in dmesg
really2 has quit [Ping timeout: 256 seconds]
<marcan> kettenis: whoops, I could've sworn I reserved all of m1n1 but I guess I didn't
<kettenis> by the looks of it you reserve none of it
<kettenis> and when my kernel lands on top of it, the secondary cores don't spin up
mxw39 has joined #asahi-dev
<marcan> kettenis: should be fixed now, sorry about that
<kettenis> hmm, that doesn't quite work
<kettenis> u-boot ends up marking that memory as EfiBootServicesData
<kettenis> and kernel thinks it can re-use this block of memory
<kettenis> need to read some docs to see what is supposed to happen
<marcan> that seems... wrong
<kettenis> I don't necessarily disagree with you, but it seems that u-boot only considers entries under /reserved-memory that have the no-map property to be permanently off-limits
<marcan> linux definitely treats them as reserved forever, so this seems inconsistent
<marcan> is there any known rationale for that behavior in u-boot?
<kettenis> this was changed not too long ago and the change references the EBBR
<kettenis> and the clarification it claims to be made in the EBBR is completely absent as far as I can tell
<kettenis> Which points back at the device tree spec ;)
<kettenis> And the "Memory Reservation Block and UEFI" sections is crystal clear
<kettenis> u-boot is buggy
<kettenis> I'll send them a patch
<marcan> which admittely is... strange
<marcan> I would only expect EfiBootServicesData for stuff marked `reusable`
<kettenis> well, I suppose they're mostly concerned about linux,cma-default and linux,dma-default nodes
Chainsaw has joined #asahi-dev
<kettenis> patch sent; let's see what they say
<kettenis> also pushed that patch to the apple-m1-m1n1 branch of my u-boot fork
VinDuv has quit [Read error: Connection reset by peer]
VinDuv has joined #asahi-dev
<kettenis> thanks for the quick fix
KindTwo has joined #asahi-dev
KindOne has quit [Ping timeout: 256 seconds]
KindTwo is now known as KindOne
jeffmiw has quit [Ping timeout: 245 seconds]
jeffmiw has joined #asahi-dev
VinDuv has quit [Quit: Leaving.]
jeffmiw has quit [Ping timeout: 256 seconds]
vlixa has quit [Remote host closed the connection]
yrlf has quit [Quit: The Lounge - https://thelounge.chat]
yrlf has joined #asahi-dev
<sven> marcan: huh, looks like your -interline-spacing -1 produces a rather strange font.b file on mac os. the individual chars don't seem to line up nicely anymore
<sven> Version: ImageMagick 7.0.11-3 Q16 arm 2021-03-06 https://imagemagick.org fwiw
<sven> anyway, bedtime for me. i'll update the fb console MR once the usb stuff is merged