#asahi-alt · 2026-08-27 08:57
elisaadohttps://github.com/nix-community/nixos-apple-silicon/issues/533 we've also seen this on NixOS, not sure if it's related#asahi-alt · 2026-08-27 01:50
JamesCalligeros[m]if it does not, then something is wrong with how youre updating m1n1#asahi-alt · 2026-08-27 01:49
JamesCalligeros[m]i think it would be helpful for them to go back to 7.1.9, then check if apple,iomfb-surfaces exists in the DCP nodes of the devicetree after rebooting (and when the failure happens)#asahi-alt · 2026-08-27 01:46
nicolohere's the github issue in case https://github.com/void-linux/void-packages/issues/62174#asahi-alt · 2026-08-27 01:46
nicoloactually, also someone with mac mini m1#asahi-alt · 2026-08-27 01:45
JamesCalligeros[m]was it everyone with a j314 that was affected or just a few users? cant rule out that theyre doing something Very Special #asahi-alt · 2026-08-27 01:45
nicolowhat am i missing?#asahi-alt · 2026-08-27 01:45
nicolowith the relevant version of dtbs (same as kernel)#asahi-alt · 2026-08-27 01:44
nicoloso assuming it works (as they confirmed), at every kernel update mini stage2 is regenerated#asahi-alt · 2026-08-27 01:44
nicoloour trigger pulls dtbs from the versioned kernel and packs them into mini stage2#asahi-alt · 2026-08-27 01:43
nicoloas for outdated dtbs: i do not understand this#asahi-alt · 2026-08-27 01:43
nicolojannau: thanks!, they confirmed the fix works#asahi-alt · 2026-08-26 09:12
jannaufixed in asahi-7.1.10-1#asahi-alt · 2026-08-26 06:16
jannauwill be fixed for 7.1.10 which I'll do later today but the root cause is likely using outdated dtbs#asahi-alt · 2026-08-26 06:10
jannauit returns -EINVAL if it is not present but I'm checking for -ENODATA#asahi-alt · 2026-08-26 06:10
JamesCalligeros[m]So still something is wrong with their dtbs not being updated#asahi-alt · 2026-08-26 06:09
JamesCalligeros[m]Ahhh rip#asahi-alt · 2026-08-26 06:09
jannauI changed it for the merge and I'm not handling the of_property_count_elems_of_size() return value correctly when the property is not present#asahi-alt · 2026-08-26 06:08
JamesCalligeros[m]But all t600x dcps should have the property#asahi-alt · 2026-08-26 06:07
JamesCalligeros[m]It should fall back to one surface unless that changed for merge#asahi-alt · 2026-08-26 06:04
jannauhmm, no I did merge the dcp plane/layer changes. Didn't I add a fallback if it is not present#asahi-alt · 2026-08-26 06:03
jannauI don't think there should be anything which requires dtb in 7.1.8. especially if the previous kernel was 7.1.6 and for dcp#asahi-alt · 2026-08-26 05:25
JamesCalligeros[m]is there anything else before the probe failure that might give a clue?#asahi-alt · 2026-08-26 05:19
LarstiQiirc m1n1 and kernel dtbs ar enot the same, though not sure which part m1n1 is responsible for setting up and which is kernel, but maybe a lead?#asahi-alt · 2026-08-26 05:18
nicoloour mini hook takes dtbs from the kernel#asahi-alt · 2026-08-26 05:17
nicolom1 pro macbook pro#asahi-alt · 2026-08-26 03:16
JamesCalligeros[m]m1 macbook pro or m1 pro macbook pro#asahi-alt · 2026-08-26 03:16
JamesCalligeros[m]nicolo: they have probably not updated their devicetrees properly#asahi-alt · 2026-08-26 02:27
nicolois anyone else experiencing this?#asahi-alt · 2026-08-26 02:27
nicolobreaks as in apple-dcp 38bc00000.dcp: probe with driver apple-dcp failed with error -22#asahi-alt · 2026-08-26 02:26
nicolothey have M1 mac book pro, while I cannot reproduce on my m1 air or m2 mini#asahi-alt · 2026-08-26 02:26
nicolosome users on Void Linux reported that the kernel 7.1.8 update breaks GPU#asahi-alt · 2026-08-25 11:25
chaos_princessthe usb stack on macs is all kind of weird, and the usb bus resets on exiting sleep, so if you have your root partition on there - kernel panic#asahi-alt · 2026-08-25 11:24
x1z53[m]chaos_princess: Understood#asahi-alt · 2026-08-25 11:24
chaos_princessnot behind some sort of an expert/yes-i-know-what-i-am-doing mode#asahi-alt · 2026-08-25 11:24
chaos_princessplease do not offer a flash install to your users#asahi-alt · 2026-08-25 11:24
x1z53[m]chaos_princess: It's Phosh#asahi-alt · 2026-08-25 11:24
chaos_princesswhy is it cut off like that#asahi-alt · 2026-08-25 11:24
x1z53[m]It's now from flash, but I guess we can rewrite installation script#asahi-alt · 2026-08-25 11:22
x1z53[m]Thanks a lot for your help!#asahi-alt · 2026-08-25 11:22
x1z53[m]It works!#asahi-alt · 2026-08-25 11:22
x1z53[m]uploaded an image: (112KiB) < https://matrix.org/oftc/media/v1/media/download/AY8gOav5kzFHm_ls5DoZYtnM6_gzF2q8lYArD37Sj8yiDXSiFe9PuF1HsA-n9Yyr2bgN5FGUMt-kay0LAApcx0NCeggMGJKAAGFsdC1nbm9tZS5ydS9LV2txTmJackJ3UkpRbVNvaFVobWxhb2U >#asahi-alt · 2026-08-25 09:42
x1z53[m]I need to shutdown and boot to Recovery OS, not reboot#asahi-alt · 2026-08-25 09:42
x1z53[m]Hm, I guess I understand now#asahi-alt · 2026-08-25 09:29
chaos_princessthey must match#asahi-alt · 2026-08-25 09:29
chaos_princesswhich os are you trying to change csrutil on, and which os is set as default boot#asahi-alt · 2026-08-25 09:27
x1z53[m]* using the `csrutil disable, * csrutil disable` command#asahi-alt · 2026-08-25 09:27
x1z53[m]By the way, here's a question:... (full message at <https://matrix.org/oftc/media/v1/media/download/AYYjuJFoyrinTzNoJVjOD3EBbpRVoEaS6tUbk_aT0mmvLOSiGEvDvnsxJ3oy3zrx_M8c1ZnqRpJHYCHBLSmdBw1CeggFe4lgAG1hdHJpeC5vcmcvQlVLYVd3dlpuelJOTnVKUWN6S3NBa2Zm>)#asahi-alt · 2026-08-25 09:19
chaos_princessmaybe it is different for non-latin keyboards then, one of mine is defaulting to a french layout since it has an azerty keyboard#asahi-alt · 2026-08-25 09:18
x1z53[m]*list of keyboards