#asahi-dev · 2026-10-25 15:01
svenit also still seems to hang very early#asahi-dev · 2026-10-25 15:00
svensnooze6214[m]: you want to dump the state of all cores, it's also helpful if you have a development kernel with symbols#asahi-dev · 2026-10-25 14:52
svennvram is the only stuff in the adt that contains private stuff like wlan ssids, etc. isn't it?#asahi-dev · 2026-10-25 14:50
svencould be, i've also switched cables from c-to-c to a-to-c via a hub#asahi-dev · 2026-10-25 14:46
svenokay, works now#asahi-dev · 2026-10-25 14:39
svenodd. let me try to reboot everything then#asahi-dev · 2026-10-25 14:38
svenyuka: should usb device mode work on m5 inside m1n1?#asahi-dev · 2026-10-25 14:17
svensomeone contributed some python tests at least iirc#asahi-dev · 2026-10-25 13:43
svenit's already in the guest there fwiw#asahi-dev · 2026-10-25 13:41
svenmaybe also add a panic to mirror_entry when i >= 0x800#asahi-dev · 2026-10-25 13:39
svenok, time to add debug prints then#asahi-dev · 2026-10-25 13:38
svenit's just gonna reserve 2GiB + 0x1000#asahi-dev · 2026-10-25 13:38
svenhrm.. looking at it again it shouldn't actually cause an issue#asahi-dev · 2026-10-25 13:36
svenmaybe#asahi-dev · 2026-10-25 13:36
svensize = ALIGN_UP(size + 0x1000, SPRR_CHUNK_TABLES * SPRR_TABLE_SIZE); that's your issue#asahi-dev · 2026-10-25 13:36
svenyeah well#asahi-dev · 2026-10-25 13:25
svenare there some hv_sprr.c changes after all?#asahi-dev · 2026-10-25 13:25
sven"diff --git a/src/hv/hv_sprr.c b/src/hv/hv_s" <--#asahi-dev · 2026-10-25 13:25
svenuhh... why is that thing you pasted cut off?#asahi-dev · 2026-10-25 13:16
svenif the i passed in to mirror_entry is too large that would cause it to write past its bounds i think. #asahi-dev · 2026-10-25 13:14
sveni wouldn't be surprised if there's some subtle error in there#asahi-dev · 2026-10-25 13:14
svenand probably also in mirror_entry and maybe mirror_table since that's where it probably faults#asahi-dev · 2026-10-25 13:12
svene.g. https://github.com/AsahiLinux/m1n1/blob/51b747dced07a58eacf5dabc31ed6f00bcfa20fd/src/hv/hv_sprr.c#L226 to confirm the bounds and then in https://github.com/AsahiLinux/m1n1/blob/51b747dced07a58eacf5dabc31ed6f00bcfa20fd/src/hv/hv_sprr.c#L234 for each allocation#asahi-dev · 2026-10-25 13:10
svensnooze6214[m]: dunno how that can happen, i'd add some debug prints around the sprr allocation stuff#asahi-dev · 2026-10-25 13:10
svenhttps://github.com/torvalds/linux/commit/f918e380449880824c415a39214b553aa7f64d5d#asahi-dev · 2026-10-25 13:07
svenmyria: i think you're confused, https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=be72934da8f5f1e432b98eff76c4d4df0d7b531e contains it, doesn't it?#asahi-dev · 2026-10-25 13:06
svenit looks like it's trying to access past the end of the sprr reservation: PT[10166e28000:101e6e2c000] -> RESERVED SPRR POOL#asahi-dev · 2026-10-25 13:05
svenany additional local changes?#asahi-dev · 2026-10-25 13:04
sventhis is m1n1's main branch, right?#asahi-dev · 2026-10-25 13:03
svensnooze6214[m]: thanks. SErrors are async so it likely fauled a few instructions earlier#asahi-dev · 2026-10-25 13:02
sventhat shared pin should be part of the drivers soc PR unless I messed up#asahi-dev · 2026-10-25 12:52
svencan you upload your m1n1-raw.elf and/or figure out in which function that happened?#asahi-dev · 2026-10-25 12:31
sventhe cpu and cluster power stuff #asahi-dev · 2026-10-25 12:30
svenI think I saw that on m4 pro as well and fixed it with some of the bootargs I inject #asahi-dev · 2026-10-25 10:06
sven*all#asahi-dev · 2026-10-25 10:05
sventhe fault address is usually al over ram#asahi-dev · 2026-10-25 10:05
svenI'd also like to know wtf that interrupt is and why it happens but it made no sense to me#asahi-dev · 2026-10-25 10:04
svenstill kinda surprised amcc is such as mess. it looked so nice in the device tree :/#asahi-dev · 2026-10-25 09:52
svenI’ll dump the adt later today #asahi-dev · 2026-10-24 16:00
svenI've only tested it on my m4 pro#asahi-dev · 2026-10-24 16:00
svensweet, let me know if it also works on the a18 pro#asahi-dev · 2026-10-24 13:44
svenyeah, true. might as well try if it just boots without loading all the secure kernel / exclave stuff#asahi-dev · 2026-10-24 13:43
svendepends on if m5 works without secure kernel :D#asahi-dev · 2026-10-24 13:41
svenhrm, see if the hv works on m5 or m4 pro atcphy next? :>#asahi-dev · 2026-10-24 10:41
svenI don't think my m4 pro has been up for more than 10 minutes at a time so far :D#asahi-dev · 2026-10-24 10:36
svenI have mine upside down on my desk :D#asahi-dev · 2026-10-24 10:26
svenunfortunately Germany will probably take another 10 years before that can be done here :D#asahi-dev · 2026-10-24 10:25
svenyeah, that worked quite well when I was in France