#
✕ nick:mixi
#asahi · 2025-08-20 20:39
mixidoes the volume go to a sensible value if you set the volume, reboot, start to record from the device?
#asahi · 2025-08-20 20:37
mixino clue about the input, but it would make sense to me if it has a symmetrical bug
#asahi · 2025-08-20 20:35
mixiat least the random times of playback with 100% volume did not occur for me anymore after I started automatically playing 1s of silence on every boot
#asahi · 2025-08-20 20:33
mixinote on that bug: while normally it goes back to the saved volume as soon as you start playing audio, it seems to save the volume as 100% if you boot and shut down without playing any audio in between, which means it'll stay on full blast on the next boot
#asahi · 2024-10-28 06:54
mixinicolas17: that should work. I never really used the launcher script myself. you can probably modify a copy of it somewhere if you don't want to fight the package manager on each update
#asahi · 2024-10-26 08:18
mixinicolas17: for the keyboard I found setkxbmap in the VM before starting your actual application warns that you're not supposed to run it against xwayland, but works nonetheless
#asahi · 2024-10-26 08:17
mixinicolas17: adding "-e SOMMELIER_SCALE=0.5" (or similar, but note that it is the inverse of the usual scaling factor) to your muvm invocation should work for the scaling
#asahi · 2024-05-16 16:47
miximaz: that's what I get for trying out new mail clients. I thought I already added them...
#asahi · 2024-05-16 16:44
mixijannau: right. that was a mistype. it was since 6.8
#asahi · 2024-05-16 16:23
mixiand the bisect was with the upstream kernel
#asahi · 2024-05-16 16:23
mixidepending on how well it works on this train I might have sent the mail
#asahi · 2024-05-16 15:56
mixi(found by git bisect)
#asahi · 2024-05-16 15:55
mixias a follow-up to the problems with dm-crypt flokli mentioned yesterday: they seem to be gone with 2632e2521769 ("arm64: fpsimd: Implement lazy restore for kernel mode FPSIMD") reverted
#asahi · 2024-05-15 15:31
mixiah, the bug report already says it is
#asahi · 2024-05-15 15:30
mixiflokli: are your filesystems on top of dm-crypt? because I observed the same for ext4 and xfs
#asahi · 2023-12-17 10:15
mixinot sure whether that is the right node for hdmi, but what does "tr '\0' '\n' < /proc/device-tree/$(tr '\0' '\n' < /proc/device-tree/aliases/dcpext0)/status" say for you?
#asahi · 2023-03-06 18:41
mixiI put it as a comment on the proposed patch github gist, as I do not have any mastodon account, but feel free to relay it there as well
#asahi · 2023-03-06 18:29
mixi(see the NLMSG_GOODSIZE macro in include/linux/netlink.h)
#asahi · 2023-03-06 18:29
mixi*=16k
#asahi · 2023-03-06 18:29
mixi"it's 16k on Asahi because the kernel gives you page-sized buffers" <-- it's not 4k for >16k page size, but the kernel limits it at min(PAGE_SIZE, 8k) + overhead
#asahi · 2022-09-02 08:04
mixiand if you forgot something in the initramfs, but already set up the encrypted root you have to start from square one, as you already deleted the unencrypted one and can't go back there to regenerate a bigger initramfs
#asahi · 2022-09-02 08:03
mixiyou have to be especially sure you can boot from files on the esp alone, as grub might load grub modules, kernels or initramfs from the linux /
#asahi · 2022-09-02 07:59
mixibut that is quite fiddly and easy to break
#asahi · 2022-09-02 07:58
mixibestouff[m]: the last time I installed, I set up the unencrypted /, booted it once adding everything required for luks (cryptsetup), creating the new fs (mkfs.*) and network (to get the contents back into the encrypted rootfs) to the initramfs, rebooted into the initramfs rescue shell and set it up from there
#asahi · 2022-07-07 13:32
mixithe qemu efi support was referring to also passing -bios QEMU_EFI.fd with an appropriate QEMU_EFI.fd stolen from some distribution package
#asahi · 2022-07-07 13:31
mixithe qemu efi support was of course not part of the qcow2-file, I just rewrote that sentence wrongly when trying to add extra information to it
#asahi · 2022-07-07 13:29
miximps: I created my own esp + kernel partition with and efi bootloader and qemu efi support as a qcow2-file on macos, and passed that and the root partition as separate hard drives to qemu; but I think it would have been a lot easier to just pass the root partition and have qemu load the kernel and maybe initramfs you want to boot directly
#asahi · 2022-07-05 16:35
miximps: I can confirm it works with raw partitions, as I had the same setup (booting the same root via qemu on macos or asahi)
#asahi · 2022-03-10 13:24
mixiand from there the aarch64-unknown-none-softfp target libraries are only one command away
#asahi · 2022-03-10 13:23
miximps: why not? the rust project provides aarch64-unknown-linux-musl binaries at least, so for a m1-hosted alpine you should be able to install rustup
#asahi · 2022-02-27 17:49
mixibut that's just guessing
#asahi · 2022-02-27 17:49
mixiI'd attribute it to using the software rendering path that's probably not well tested on other platforms
#asahi · 2022-02-27 17:48
mixibut it's rare enough to not be a problem
#asahi · 2022-02-27 17:48
mixithere is an issue where rarely some clients crash with the pixman renderer I was too lazy to figure out
#asahi · 2022-02-27 17:47
mixis/fine/well enough for me/
#asahi · 2022-02-27 17:44
mixiyes, it works fine on my machine
#asahi · 2022-02-27 17:10
mixipitust[m]: is your ~/code/initramfs-linux-fallback.img compressed with gzip or xz?
#asahi · 2022-02-21 07:54
mixiGlanzmann: the command you're looking for should be "dtc -I dtb -O dts /sys/firmware/fdt"
#asahi · 2022-02-20 22:07
mixiit should also have created /sys/firmware/devicetree/base/backlight entry from the device tree node alone though, so my first guess would be that you are loading the wrong device tree again
#asahi · 2022-02-20 22:07
mixiGlanzmann: https://tg.st/u/backlight_notes.txt looks right to me, and should make a /sys/class/backlight/backlight appear (the "backlight:" is only necessary if you want to refer to it from another node though)
#asahi · 2022-02-06 22:47
mixiyes, the default for fields left out in designated initializers is 0 ("the same as objects that have static storage duration") -> fast_io is 0 / false if it is left out of the initializer
#asahi · 2022-02-06 21:43
mixij`ey: the regmap lock will be a mutex instead, which means we will not end up in an atomic context and will be allowed to sleep
#asahi · 2022-02-06 21:25
mixij`ey, jeffmiw: while it is allowed to sleep in *_probe, regmap_raw_read takes the regmap lock (a spin lock because apple-spmi-sera-pmu sets regmap_config.fast_io = true) -> we are in an atomic context and may not sleep anymore
#asahi · 2022-01-28 17:35
mixiGlanzmann: you have both CONFIG_SND_SIMPLE_CARD=y and CONFIG_SND_SIMPLE_CARD=m in https://github.com/AsahiLinux/docs/wiki/DesktopKernel