#asahi-dev · 2025-07-16 10:06
mazsven: is SMC on track to be picked up this time around? I don't see it in -next just yet.#asahi-dev · 2025-07-16 09:48
svenyeah, there’s a bunch of stuff I have to pick up #asahi-dev · 2025-07-16 09:38
nickchansven: bothering again the t8012-j132 framebuffer fix is still there https://lore.kernel.org/asahi/20250620-j132-fb-v2-1-65f100182085@gmail.com/#asahi-dev · 2025-07-15 17:42
flokli^ DannyB on the wifi stuff #asahi-dev · 2025-07-15 16:30
flokliHmmh nope, the wifi thing is not stable, even with disabled power management. I wonder if there's a chance to try with other firmware.#asahi-dev · 2025-07-15 14:27
mazgot basic 6.16 running on M2-pro mini with just dart and SMC added. Ugly conflicts in io-pgtables-dart.c, but it has booted -- must be perfect! ;-)#asahi-dev · 2025-07-15 14:11
jannauI've pushed asahi-6.15.6-1 to github#asahi-dev · 2025-07-15 13:09
flokliI added this to my iwd config, as per https://man.archlinux.org/man/iwd.config.5#DriverQuirks 's PowerSaveDisable= option. I'll see how stable it is.#asahi-dev · 2025-07-15 13:06
flokliGoogling around a bit seems to hint at power saving to be part of the issue, and a `iw wlan0 set power_save off` and some more toggling seems to have made it associate for now.#asahi-dev · 2025-07-15 13:00
floklirebooted with that kernel, still the same traces in dmesg: https://paste.linuxlounge.net/BU#asahi-dev · 2025-07-15 12:59
flokliok#asahi-dev · 2025-07-15 12:57
chaos_princessyes, they are now apple_spmi_nvmem#asahi-dev · 2025-07-15 12:54
floklisame for the simple-mfd-spmi kernel module#asahi-dev · 2025-07-15 12:49
floklijannau: I can verify it builds now. Though now some later tooling cannot find the nvmem_spmi_mfd kernel module. Was that removed?#asahi-dev · 2025-07-15 12:26
jannauflokli: I've pushed a fix and verified it compiles with 1.87#asahi-dev · 2025-07-15 12:24
flokliI'll try 1.88#asahi-dev · 2025-07-15 12:23
floklichadmed: it does not build with 1.86, same error message#asahi-dev · 2025-07-15 12:02
jannauI'm compiling with 1.88#asahi-dev · 2025-07-15 12:01
chadmeddowngrade to 1.86 and see #asahi-dev · 2025-07-15 12:00
floklihttps://dpaste.com/AKQAZBFDM is the full build log#asahi-dev · 2025-07-15 11:59
floklichaos_princess: 1.87.0#asahi-dev · 2025-07-15 11:17
chaos_princesswhich rust version are you on?#asahi-dev · 2025-07-15 11:14
floklichaos_princess: fails to buils with https://paste.linuxlounge.net/W4, "the trait bound `*mut kernel::bindings::fwnode_handle: Default` is not satisfied"#asahi-dev · 2025-07-15 10:54
flokliok, will build 5c2a16cd2eafc8485aefc601d5727b40203d3d3b from 28mins ago#asahi-dev · 2025-07-15 10:53
chaos_princessyes, it is based on 6.15#asahi-dev · 2025-07-15 10:53
floklichaos_princess: asahi-wip branch, you mean?#asahi-dev · 2025-07-15 10:52
chaos_princessmind trying out 6.15?#asahi-dev · 2025-07-15 10:51
flokli6.14.8-asahi, M2 Air#asahi-dev · 2025-07-15 10:48
floklihttps://paste.linuxlounge.net/NU#asahi-dev · 2025-07-15 10:47
flokliI'm at a camping event with wifi, and the wifi driver seems really really unhappy about something happening there, I guess OWE. Getting kernel traces every second or so#asahi-dev · 2025-07-15 10:24
jannauerr, yes. I forgot to do that#asahi-dev · 2025-07-15 10:21
chaos_princesssure. you can also delete the power state toggling between idle and off in SndSocAopDriver::new then#asahi-dev · 2025-07-15 10:18
jannausee asahi-wip#asahi-dev · 2025-07-15 10:18
jannauchaos_princess: I was still seeing probe order issues and have moved the snd_pcm_set_managed_buffer call to aop_pcm_open. This is what we did for DCP audio as well to avoid the same issue#asahi-dev · 2025-07-14 18:59
Lightswordchaos_princess, some build failures from right before it went offline that show disk corruption, errors started appearing with null bytes in files I guess https://autobuild.buildroot.net/results/4e0/4e0a85a0a1eefe86b5dd9afbea36e1f6c0d25ed6/build-end.log#asahi-dev · 2025-07-14 18:08
chaos_princess*prevent or fix#asahi-dev · 2025-07-14 18:08
chaos_princessit probably will be able to catch it, but obviously won't prevent or it. and no additional debug flags that i know of#asahi-dev · 2025-07-14 18:05
Lightswordchaos_princess, I'm now using btrfs instead of ext4 so I wonder if the checksumming there will catch filesystem corruption before it causes catastrophic filesystem failure, there aren't any flags for adding additional debugging/validation for the nvme driver right?#asahi-dev · 2025-07-14 18:01
chaos_princessmaybe, maybe not. probably not. Anyway, the dmesg message above is normal and can be ignored#asahi-dev · 2025-07-14 18:01
Lightswordchaos_princess, there anything I should do so that if it happens again I can get useful diagnostics?#asahi-dev · 2025-07-14 18:00
Lightswordchaos_princess, I'm suspecting it's potentially a nvme driver issue that only appears under high iops or over a long period of time if anything since I'm using the mac mini as a server for buildroot autobuilders(basically continuously cross-compiling full linux systems in parallel), it usually goes a few months before I see disk corruption related failures however so it's not exactly easy to reproduce#asahi-dev · 2025-07-14 17:55
chaos_princesssame ext4 driver as all other linux installs, and if it is the nvme driver at fault, we'd get more corruption reports on fedora too#asahi-dev · 2025-07-14 17:55
Lightswordonly known issue I was finding online were related to apple's tools showing issues(which isn't something I've ever run into)#asahi-dev · 2025-07-14 17:53
Lightswordlike this last time I was able to ssh in and see a ton of ext4 errors, fsck failed to restore it#asahi-dev · 2025-07-14 17:53
Lightswordchaos_princess, so no reports of filesystem corruption on arch? I've seen ext4 filesystem corruption a few times in the past and had to reinstall(also had a logic board replacement after the mac mini m1 completely died)#asahi-dev · 2025-07-14 17:50
chaos_princessand no, it should not randomly corrupt filesystems#asahi-dev · 2025-07-14 17:50
chaos_princessno, it is probably unrelated, i also have a bunch of those in dmesg#asahi-dev · 2025-07-14 17:50
Lightswordis filesystem corruption a known issue in general?#asahi-dev · 2025-07-14 17:49
Lightswordchaos_princess, ah, wonder if it's related to the high iops workload I'm running, I just recently migrated from the old arch based distro to fedora after the main ext4 filesystem decided to corrupt itself#asahi-dev · 2025-07-14 17:47
chaos_princessapple knows, but they will not tell. (it is a log message from the nvme controller itself)