#asahi-dev · 2025-07-28 22:06
jannauany guess whether that's specific to my system, mac pros in general or certain refurbished devices?#asahi-dev · 2025-07-28 22:00
chaos_princessshould be a bunch of img4s, idk which is wifi, but the long string at the end is the device serial so try matching that way#asahi-dev · 2025-07-28 21:59
chaos_princessfdr partition (assuming "factory data restore") Should be at /System/Volumes/Hardware/FactoryData/System/Library/Caches/com.apple.factorydata/#asahi-dev · 2025-07-28 21:59
jannauthe template ADT from ipsw has those properties though#asahi-dev · 2025-07-28 21:57
jannauany idea what '"wlan.msf.source" = "fdr"' (from ioreg) could mean? wlan/bt misses calibration data in the ADT on my (refurbished) mac pro. ioreg has wifi-calibration-fdr instead of wifi-calibration-msf#asahi-dev · 2025-07-28 20:08
sventhere's also irq support for the gpio driver but i'm not sure we even use that for anything#asahi-dev · 2025-07-28 20:06
svenso if anyone's looking to help and wants a simple driver: rtc and power-supply both should be reasonably easy to upstream after -rc1 is out now#asahi-dev · 2025-07-28 20:05
sveni guess at least rtc will need dt bindings as well#asahi-dev · 2025-07-28 20:04
chaos_princessnp, just making sure#asahi-dev · 2025-07-28 20:04
sveni usually consider fixing stuff like that to be part of upstreaming ;)#asahi-dev · 2025-07-28 20:04
chaos_princessmacsmc-battery still does a bunch of dmesg spam, so should probably be fixed first#asahi-dev · 2025-07-28 20:03
sveni'll do hid next (https://git.kernel.org/pub/scm/linux/kernel/git/sven/linux.git/log/?h=wip/smc-hid) and I guess chadmed_ is going to work on hwmon#asahi-dev · 2025-07-28 20:02
svenso.. what are the smc sub-device we still have to upstream? hid, rtc, hwmon, and power-supply?#asahi-dev · 2025-07-27 02:49
dramforevernevermind i had APPLE_WATCHDOG as m and forgot to put that in my initrd#asahi-dev · 2025-07-27 02:38
dramforeverearly as in like, in the initrd early#asahi-dev · 2025-07-27 02:36
dramforeverhow do i / can i disable the watchdog manually to give me time to debug early boot failures?#asahi-dev · 2025-07-26 15:47
jannauI've pushed asahi-6.15.8-1 to github with a reshuffled bits/020-dart#asahi-dev · 2025-07-26 00:03
chadmed_i have to write bindings too...#asahi-dev · 2025-07-26 00:02
chadmed_would anyone be opposed to me submitting the hwmon driver as-is (with some cleanup of course) as an RFC#asahi-dev · 2025-07-25 09:47
ruiconhi#asahi-dev · 2025-07-24 12:33
sveni don't necessarily want to bypass dt review, might just be easier to hardcode the keys inside the driver and match on the top level board compatible to select the right set#asahi-dev · 2025-07-24 11:48
chadmedi guess the benefit is bypassing dt review but the core design issue is still there#asahi-dev · 2025-07-24 11:48
chadmedthe annoying thing is that its not really composable. even keys youd think should be common like core temps are not common across machines with the same soc#asahi-dev · 2025-07-24 11:43
svenbut yeah, submitting the other SMC sub devices sounds like a good plan #asahi-dev · 2025-07-24 11:43
svenor maybe just hardcore all that in the driver instead #asahi-dev · 2025-07-24 11:26
chadmednamely the DT bindings which have an almost 100% chance of needing changes and the runtime code that walks the DT and builds all the hwmon structs. as far as i could see in the tree no other driver does anything like this#asahi-dev · 2025-07-24 11:26
chadmedthere are a couple of controversial things in it#asahi-dev · 2025-07-24 11:26
chadmednow that smc is in should i clean up and submit hwmon in the next merge window? #asahi-dev · 2025-07-24 10:41
jannauI think it's only annoying in one place. reording shouldn't be a mayor issue. I simply did not think of it#asahi-dev · 2025-07-24 10:38
svent602x support should be separate, but reordering the commits is quite annoying but required to send that one first #asahi-dev · 2025-07-24 10:33
chadmeddoes it make sense to hold that stuff back and submit it together?#asahi-dev · 2025-07-24 09:25
jannauideally the display darts shouldn't be reset until the first device attaches#asahi-dev · 2025-07-24 09:24
jannaubut iommu groups have to share the same domain ops#asahi-dev · 2025-07-24 09:23
jannauwhich is annoying since locked darts ideally would use different iommu domain ops#asahi-dev · 2025-07-24 09:23
jannaubut it's annoying for the full display support as it would mean to support combinations of locked and unlocked darts#asahi-dev · 2025-07-24 09:21
jannaum1n1 locks currently the dart of the boot display controller. this ensures the boot framebuffer works#asahi-dev · 2025-07-24 09:20
jannaudisplay makes upstreaming dart rather annoying, I probably should reorder the t602x dart with the locked dart support#asahi-dev · 2025-07-24 09:17
jannaudart is on my list#asahi-dev · 2025-07-24 09:17
sveni think jannau wanted to take those#asahi-dev · 2025-07-24 09:16
mazdart and iopgtables would be good as well -- they tend to badly clash on each rebase...#asahi-dev · 2025-07-24 09:15
svennext up is usb i guess. that's gonna be even more 'fun'#asahi-dev · 2025-07-24 09:11
mazsven: party!#asahi-dev · 2025-07-24 09:08
svenlooks like SMC made it just-in-time for the merge window :)#asahi-dev · 2025-07-22 16:40
svenalright, very late pull requests for 6.17 sent. i'll start picking up patches again after the merge window#asahi-dev · 2025-07-22 14:56
chaos_princessin short - no. you can try figuring out how gtk and qt do dialogs, but it is inherently a wild west there and you will not be able to make it 100% reliable#asahi-dev · 2025-07-22 14:48
dev-muhammad-adelis theres way to detect info about dialogs like buttons(ok,cancel,confirm, shutdown ,etc) to draw in touchbar?#asahi-dev · 2025-07-22 14:46
dev-muhammad-adelhello #asahi-dev · 2025-07-20 10:54
JesusChrist[Chat] Disabled auto-message sending, I've officially moved the bot to #llm#asahi-dev · 2025-07-20 08:51
LLM[LLM] Training Neural Network...#asahi-dev · 2025-07-20 08:51
LLM[LLM] Corpus Loaded With 867 Messages