#
✕ nick:DmitrySharshakov[m]
#asahi-dev · 2022-08-28 08:24
DmitrySharshakov[m]Have you figured out the DP thing you asked recently?
#asahi-dev · 2022-08-28 08:23
DmitrySharshakov[m]Something like PD-out and DP tunneling?
#asahi-dev · 2022-08-28 08:22
DmitrySharshakov[m]And what's left/which questions are to be answered?
#asahi-dev · 2022-08-28 08:21
DmitrySharshakov[m]Ah, no kernel driver yet?
#asahi-dev · 2022-08-28 08:20
DmitrySharshakov[m]sven: btw how are your ATCPHY/TB-NHI/USB3 works? Are you testing locally without a branch or there's a branch to watch the news?
#asahi-dev · 2022-08-28 08:18
DmitrySharshakov[m]does arm even have deep sleep? Is it like x86 where EC wakes the system up? Probably ARM has a different concept
#asahi-dev · 2022-08-28 08:15
DmitrySharshakov[m]sven: I meant cannot you watch macOS suspending and resuming wireless with m1n1 hypervizor?
#asahi-dev · 2022-08-28 08:15
DmitrySharshakov[m]sven: Yea. Why would it break kernel concepts?
#asahi-dev · 2022-08-28 08:13
DmitrySharshakov[m]I guess suspend protocol can be reversed by m1n1 introspection, or doesn't macOS suspend work while booted under HV?
#asahi-dev · 2022-08-28 08:12
DmitrySharshakov[m]sven: So you just suspended to idle, without bringing CPUs down?
#asahi-dev · 2022-08-28 08:12
DmitrySharshakov[m]But yes, I guess that'd duplicate what is done by default PSCI
#asahi-dev · 2022-08-28 08:11
DmitrySharshakov[m]x86 does its s2idle (S0ix) like this: 1. bring down non-boot cpus, 2. suspend pcie peripherals, 3. drive the CPU 0 into low-power idle
#asahi-dev · 2022-08-28 08:09
DmitrySharshakov[m]Well, not all ARM SoCs use PSCI for PM. Should M1 do so? Does it use PSCI in macOS? Can it be avoided by simply doing all by register writes (SMP bringup reversed)?
#asahi-dev · 2022-08-28 08:08
DmitrySharshakov[m]sven: Hasn't it yet been proposed?
#asahi-dev · 2022-08-28 08:07
DmitrySharshakov[m]What's wrong exactly? Might it be a generic brcmfmac issue or platform one?
#asahi-dev · 2022-08-28 08:07
DmitrySharshakov[m]sven: WiFi on its own or because APCIe breaks in some way?
#asahi-dev · 2022-08-28 07:43
DmitrySharshakov[m]j`ey: yes it is, thanks
#asahi-dev · 2022-08-28 07:42
DmitrySharshakov[m]Probably better to wait for marcan, as he certainly has more fresh info about PM than I do
#asahi-dev · 2022-08-28 07:41
DmitrySharshakov[m]PSCI is power system coordination interface iirc
#asahi-dev · 2022-08-28 07:41
DmitrySharshakov[m]Well, PCIe controller driver should definitely be updated for suspend/resume xD
#asahi-dev · 2022-08-28 07:39
DmitrySharshakov[m]Quite a bit of testing should be needed to ensure the best efficiency and stability
#asahi-dev · 2022-08-28 07:39
DmitrySharshakov[m]I'm not sure (not a maintainer, I'm here mostly for news + I do some work for userspace audio in PipeWire), but 99% you can do this
#asahi-dev · 2022-08-28 07:37
DmitrySharshakov[m]Typically for PM ops we (the kernel living in EL2 while being capable of virtualization) must ask someone higher (EL3) to manage power interfaces. This is the PSCI calling ABI. But Apple SoC has no TrustZone extension, so PSCI need to be handled by the kernel itself. I guess marcan has already invented the idea of this fix
#asahi-dev · 2022-08-28 07:34
DmitrySharshakov[m]Presumably the main task is to figure out PSCI without the ARM Trusted Firmware. Correct me if my info is terribly outdated
#asahi-dev · 2022-08-28 07:34
DmitrySharshakov[m]marcan: btw when is the next kernel stream planned?
#asahi-dev · 2022-08-28 07:33
DmitrySharshakov[m]Oh nice, welcoome! marcan is planning to work on s2idle the next stream or so (?).
#asahi-dev · 2022-08-28 05:39
DmitrySharshakov[m]I have a bit of concern about Fedora aarch64. It appears that it's hard/impossible to setup multiarch (x86_64 libraries on aarch64 system), so that might cause issues with emulation (box86/box64), which is an integral part of Apple Silicon experience and is generally needed for full compatibility
#asahi-dev · 2022-08-26 07:00
DmitrySharshakov[m]recovery is necessary, boot with option held down
#asahi-dev · 2022-08-26 07:00
DmitrySharshakov[m]So it'd also be easier for distributions. Say we have Fedora. We first install m1n1, allow it to boot with Fedora keys (once secureboot is there), add a recovery partition (ideally in the future simply dd Fedora ISO into it). Then user boots into there, bootloader cannot find the main system thus gets them into recovery and the installation could be done from there as ordinary (well, ideally). Then main partition is used. When
#asahi-dev · 2022-08-26 06:57
DmitrySharshakov[m]I guess going the (2) way would match the desired platform UX, since macOS gets installed from its recovery
#asahi-dev · 2022-08-26 06:53
DmitrySharshakov[m]IIRC it's only for LUKS volumes, so we cannot use it on a non-LUKS volume
#asahi-dev · 2022-08-26 06:53
DmitrySharshakov[m]Hm, we can do it as well. But what about UX of this? Should it be autostarted on the first boot?
#asahi-dev · 2022-08-26 06:47
DmitrySharshakov[m]This allows to use essentially the same FS drivers (cryptsetup and kernel ext4/btrfs), so no compatibility issues ever
#asahi-dev · 2022-08-26 06:46
DmitrySharshakov[m]We got a partition from diskutil, we got a config file, then a VM is started with our partition as a virtual disk, another virtual disk with an automatic headless installer. It just spins up, reads the config, sets everything up, writes a log file and shuts the VM down. After that, installer shows that a reboot into the recoveryOS can now be performed to authenticate the bootloader
#asahi-dev · 2022-08-26 06:43
DmitrySharshakov[m]3. Crazy, but still an idea. We can investigate thin VMM on hypervisor framework. Yes, just spin up a thin Linux VM with the installer after we collect preferences from the user.
#asahi-dev · 2022-08-26 06:42
DmitrySharshakov[m]Bonus points here: we want to have a recovery system image together with Asahi to fix things users can break
#asahi-dev · 2022-08-26 06:42
DmitrySharshakov[m]2. Do like system76. They ship Pop OS with a recovery image, then the initial setup program asks whether you want FDE. If you do, the installer welcomes you into the recovery, where you do a reinstall with proper partitions.
#asahi-dev · 2022-08-26 06:40
DmitrySharshakov[m]1. Use some userspace solutions based on FUSE (there's FUSE btrfs and some WIP userspace LUKS). This is not really suitable as will likely result in broken volumes later as those userspace FS drivers are rather alpha-stage and might have incompatibility
#asahi-dev · 2022-08-26 06:39
DmitrySharshakov[m]0. Not using FDE but homedir encryption, has no relation to the Asahi installer and has flaws
#asahi-dev · 2022-08-26 06:38
DmitrySharshakov[m]I've previously seen a topic about LUKS on Asahi and have been investigating into it now. Well, I've found a couple solutions that could be relied upon
#asahi-dev · 2022-08-23 07:57
DmitrySharshakov[m]The advantage should be used when updated audio bits (povik sent those for review for Asahi branch yesterday) land, so would be great to have in less time than WP 0.5
#asahi-dev · 2022-08-23 07:55
DmitrySharshakov[m]Just need some feedback on how do my patches fit the use case + I'll fix some pieces and et voila we have it in a released WP version. Then 0.5 comes implementing a better config discovery mechanism
#asahi-dev · 2022-08-23 07:53
DmitrySharshakov[m]<chadmed> "plus the supposedly massive..." <- Yes, so it'll be going into 0.4 for now and 0.5 just going it's way as an unstable branch
#asahi-dev · 2022-08-23 07:53
DmitrySharshakov[m]<chadmed> "and besides, the current..." <- Well, no DT editing should be needed after updated audio bits go in (volume and HPF caps are in place in that series)
#asahi-dev · 2022-08-23 05:21
DmitrySharshakov[m]Notice: it's worth testing at least something before continuing. Did the config I wrote work for somebody?
#asahi-dev · 2022-08-23 04:42
DmitrySharshakov[m]discovery mechanism as well as optimizing things like hardware volume
#asahi-dev · 2022-08-23 04:42
DmitrySharshakov[m]chadmed: yes, although Julian wrote that it should be okay for daily use, but still Asahi is a distro which is most likely to have to deal with Pro Audio production workloads etc. So let's stay with 0.4 which is familiar to people. Then I'll first fix correctness and merge into 0.4 (maybe without hardware volume if that'd be hard). This 0.4.x can then be shipped and I'll continue my work on WP 0.5, improving configuration
#asahi-dev · 2022-08-22 19:53
DmitrySharshakov[m]Do we want to ship WirePlumber 0.5? https://gitlab.freedesktop.org/pipewire/wireplumber/-/issues/336#note_1517863
#asahi-dev · 2022-08-22 14:45
DmitrySharshakov[m]Will probably ping somebody of WirePlumber maintainers to review parts necessary
#asahi-dev · 2022-08-22 14:44
DmitrySharshakov[m]I guess when somebody sets those for MBA then audio is ready to be shipped by default and for MBP will have to either ship asahi-audio PipeWire configs or wait for WirePlumber patches

plus ›