#
✕ nick:sven
#asahi-dev · 2022-11-23 17:03
svenwon't work with dcpext yet though because the HPD OOB notification won't work yet
#asahi-dev · 2022-11-23 17:03
svenit also contains the "proper" code to recognize and switch to dp-altmode
#asahi-dev · 2022-11-23 17:02
svenjannau: https://github.com/AsahiLinux/linux/commits/atcphy-WIP may or may not fix that m1n1 WARN. didn't test it very well yet and still need to rebase it on something more recent though.
#asahi-dev · 2022-11-23 09:27
svenand running something equivalent to "1+2" is already a big step when reverse engineering :)
#asahi-dev · 2022-11-23 09:26
svenyeah, these things go a bit different when reverse engineering hardware. it makes sense to start these discussion once it comes to implementing a linux driver but before that it's just guesswork because it's unclear what the hardware can even do
#asahi-dev · 2022-11-23 09:03
sven<sven> please stop skipping to step 100 when step 1 isn't even done :D
#asahi-dev · 2022-11-23 08:58
svenlina's gpu UAPI also started with essentially a hack and now she's only slowly converging towards something upstreamable
#asahi-dev · 2022-11-23 08:55
svenplease stop skipping to step 100 when step 1 isn't even done :D
#asahi-dev · 2022-11-23 07:39
svenand again, just make it ping the kernel and the kernel shutdown on timeout
#asahi-dev · 2022-11-23 07:38
svenI don’t think anyone proposed an api yet and it doesn’t make much sense to discuss it before that
#asahi-dev · 2022-11-23 07:35
sven*some
#asahi-dev · 2022-11-23 07:35
svenbecause you generally want to avoid floating point in the kernel. you can also just make the watchdog ping it and if it hasn’t seen any feedback in aome time just kill the output
#asahi-dev · 2022-11-22 21:32
svenpovik: soc for stuff that will only ever run on apple Silicon SoCs (e.g. rtkit), platform for things that are also useful for older apple machines (x86 macs etc.) imho
#asahi-dev · 2022-11-22 20:49
svenokay, wtf, I just had access to the tbt pcie and everything else but now I rebooted and now the same kernel results in SErrors again?!
#asahi-dev · 2022-11-22 18:51
svenguess I’ll have to convince macOS to go into device mode and see what it does differently
#asahi-dev · 2022-11-22 18:51
svenrqou_: ah, pity. it’s probably still broken then
#asahi-dev · 2022-11-22 17:40
svenwhenever modules are enable which kinda makes sense maybe
#asahi-dev · 2022-11-22 17:40
sventhere’s a timeout that’s set to 10s
#asahi-dev · 2022-11-22 17:39
svendrivers/base/dd.c maybe
#asahi-dev · 2022-11-22 17:37
svenso it just gives up then?
#asahi-dev · 2022-11-22 17:37
sventhere’s a 10s deferral delay? hah
#asahi-dev · 2022-11-22 14:52
svenbut the kernel currently hardcodes the f2fs blocksize to 4k
#asahi-dev · 2022-11-22 14:52
svenuhh...
#asahi-dev · 2022-11-22 14:50
svenbut I guess they have another patch for the kernel to allow it?
#asahi-dev · 2022-11-22 14:50
svenalso, does that even touch the kernel? it looks like it just changes something in the userspace tools
#asahi-dev · 2022-11-22 14:49
svenyes, on the upstream kernel where that patch wouldn't be accepted anyway
#asahi-dev · 2022-11-22 14:47
svenokay, maybe let me rephrase that: it won't even allow you to mount pages created on a 4k kernel when you're running a 16k kernel
#asahi-dev · 2022-11-22 14:45
sventhat's not a proper fix though, it just makes f2fs only work on 16k PAGE_SIZE machines
#asahi-dev · 2022-11-22 14:22
svenpretty sure that f2fs already complains if its block size is != PAGE_SIZE
#asahi-dev · 2022-11-22 13:19
svenyeah, that's fine
#asahi-dev · 2022-11-22 13:19
sveni've tweeted/tooted it already but these progress reports usually get much more attention
#asahi-dev · 2022-11-22 13:18
sventhat's fine
#asahi-dev · 2022-11-22 13:17
sven:D
#asahi-dev · 2022-11-22 13:15
svenplease just don't mention external displays very prominently yet. i've already had at least three very enthusiastic people asking me in /msg how to set that up who didn't quite realize how broken it is right now
#asahi-dev · 2022-11-22 13:14
svenprobably won't have time but the atcphy part sounded good to me
#asahi-dev · 2022-11-22 11:24
svenlol
#asahi-dev · 2022-11-22 11:22
svenohhh… that target disk thing is what I remember in tbt mode probably
#asahi-dev · 2022-11-22 11:21
svenI’ve seen it on my Linux laptop early on when I was implementing dwc3 in m1n1
#asahi-dev · 2022-11-22 11:21
svenI think it was that
#asahi-dev · 2022-11-22 11:21
svenyes
#asahi-dev · 2022-11-22 11:20
sven*some
#asahi-dev · 2022-11-22 11:20
svenI thought tb was sone different mode but maybe I misremember
#asahi-dev · 2022-11-22 11:20
svenpretty sure it’s usb
#asahi-dev · 2022-11-22 11:15
sveni don’t think I ever traced macOS to see if it configures the phy differently
#asahi-dev · 2022-11-22 11:15
svenhm, did anyone test what happens with usb3 in device mode btw? does that just work?
#asahi-dev · 2022-11-22 10:52
svenI think some of them are hacks and/or legacy code because they didn't want to implement a phy
#asahi-dev · 2022-11-22 10:52
svennone of them
#asahi-dev · 2022-11-22 10:52
svenoh.. those
#asahi-dev · 2022-11-22 10:51
svendwc3 is a single driver
#asahi-dev · 2022-11-22 10:51
svenhmm?

plus ›