#
✕ nick:sven
#asahi-dev · 2025-05-02 13:35
sven(minus that one commit description ofc)
#asahi-dev · 2025-05-02 13:35
svenhttps://github.com/AsahiLinux/linux/tree/sven/smc-v4 okay, happy with this now I think. will send this out on sunday probably
#asahi-dev · 2025-05-02 11:12
sventhere are some comments in the linux driver iirc
#asahi-dev · 2025-05-02 10:44
svenokay, looks similar with smc-v4
#asahi-dev · 2025-05-02 10:44
sventhanks
#asahi-dev · 2025-05-02 10:40
svenyes, that would be great
#asahi-dev · 2025-05-02 10:34
sven(current == the one we ship)
#asahi-dev · 2025-05-02 10:33
svenjannau: can you run the current smc driver with trace_smc.py (just a regular boot is fine) and paste the results? i wanna compare if everything looks the same
#asahi-dev · 2025-05-02 10:28
sventhough maybe we return the wrong errno and we're supposed to just set the timeout to the max value and userspace can directly query it again afterwards?
#asahi-dev · 2025-05-02 10:28
sveni don't see any get max timeout function in the uapi either
#asahi-dev · 2025-05-02 10:17
svenoh, and add the reboot nodes for t600x and t8112
#asahi-dev · 2025-05-02 10:16
svenhttps://github.com/AsahiLinux/linux/tree/sven/smc-v4 need to fix that one commit description/etc. and write a cover letter but this should be mostly ready for submission otherwise
#asahi-dev · 2025-05-02 10:11
svenno idea why it failed from m1n1 as well though, I probably just messed up there
#asahi-dev · 2025-05-02 10:10
svenworks now
#asahi-dev · 2025-05-02 10:10
svenlol, okay, dumb bug because the smc protocol is a bit dumb
#asahi-dev · 2025-05-02 10:01
sven[cpu7] [SMCTracer@/arm-io/smc] [smceptracer] [2] >W: <MBSE> = (0) <-- well, that looks wrong :D
#asahi-dev · 2025-05-02 09:52
svennope, all manual
#asahi-dev · 2025-05-02 09:47
svener, *on this mac mini
#asahi-dev · 2025-05-02 09:47
svenlet me update the firmware on this m1n1, it's still running whatever was current when i bought this one a few years ago, lol
#asahi-dev · 2025-05-02 09:45
sven" * 'offw' - shutdown warning"
#asahi-dev · 2025-05-02 09:45
svenno, shutdown works just fine even without that
#asahi-dev · 2025-05-02 09:43
svenseems like i can't even write that key from m1n1 here
#asahi-dev · 2025-05-02 09:43
svendoes anyone else see "[ 13.012798] macsmc-reboot macsmc-reboot: Failed to issue MBSE = offw (0x6f666677) (reboot_prepare)" when running shutdown? (probably needs to be running under hv since this is one of the very last messages before power off)
#asahi-dev · 2025-05-01 18:46
svenlet's see how i broke that
#asahi-dev · 2025-05-01 18:46
svenhttps://github.com/AsahiLinux/linux/tree/sven/smc-v4 shutdown works again now as well except for a strange macsmc-reboot macsmc-reboot: Failed to issue MBSE = offw (0x6f666677) (reboot_prepare) message
#asahi-dev · 2025-04-24 11:40
svenit might help to get some tested-by or reviewed-by tags by people who are using it to show that there's an active user base
#asahi-dev · 2025-04-22 19:28
svenhttps://github.com/AsahiLinux/linux/commits/b4/pasemi-fixes/ pasemi fixes, v3. will send that out tomorrow probably
#asahi-dev · 2025-04-16 07:08
svenand right now I can kinda reproduce it rather easily
#asahi-dev · 2025-04-16 07:07
svenIt’s probably not bad for end users since m1n1 either boots Linux or fails but this could get really annoying with the hv
#asahi-dev · 2025-04-16 07:06
svenwhich is why I’d really like to track this down
#asahi-dev · 2025-04-16 07:06
svenyeah, same
#asahi-dev · 2025-04-16 06:27
svenI’m not sure this is a compiler issue tbh, but I can try an older clang this evening
#asahi-dev · 2025-04-15 20:20
svenassembly looks very different so this might also just be similar to adding printfs somewhere
#asahi-dev · 2025-04-15 20:18
svenjannau: ugh.. looks like building with gcc on either a macos or a linux host doesn't run into the same issue
#asahi-dev · 2025-04-15 18:52
svenpower also still needs dt bindings i guess
#asahi-dev · 2025-04-15 18:52
svenhttps://github.com/AsahiLinux/linux/commits/sven/smc-v4/ current smc code, especially the power part is untested because that has a bunch of dependencies
#asahi-dev · 2025-04-15 18:14
svenah
#asahi-dev · 2025-04-15 18:12
svenbroken resume?
#asahi-dev · 2025-04-15 16:59
svenand adding e.g. debug prints or too many lol = __line__; hacks makes it go away
#asahi-dev · 2025-04-15 16:58
svenbut i think it has to be some issue in exception or irq handling or something like that because it doesn't always happen
#asahi-dev · 2025-04-15 16:58
svenbut at least the assembly code for smp_secondary_entry and mmu_secondary_setup looked okay to me
#asahi-dev · 2025-04-15 16:57
svennot yet, i was going to try a linux host next
#asahi-dev · 2025-04-15 16:09
svenbut it still works fine for my very outdated stage1 m1n1
#asahi-dev · 2025-04-15 16:09
sveni first thought i broke something with https://github.com/AsahiLinux/m1n1/pull/460 but this happens with the current m1n1 main for me
#asahi-dev · 2025-04-15 16:07
svenbut when i shared my m1n1.bin with jannau it also crashed for him
#asahi-dev · 2025-04-15 16:07
svencouldn't bisect it to a specific commit and no one else seems to have run into that
#asahi-dev · 2025-04-15 16:07
svensee logs from a few days ago, _something_ corrupts LR I think in the smp_call4 -> [other core] -> smp_secondary_entry -> mmu_secondary_setup path
#asahi-dev · 2025-04-15 15:50
svenhttps://lore.kernel.org/asahi/20250415-pasemi-fixes-v2-0-c543bf53151a@svenpeter.dev/ pasemi fixes v2, time to get back to smc now (or get nerd sniped by that weird m1n1 bug again...)
#asahi-dev · 2025-04-13 20:24
svennope, just crashed when trying to start cpu 1
#asahi-dev · 2025-04-13 20:23
svenlet me try a few time

plus ›