#asahi-dev · 2026-06-08 09:01
StanfordAppliedCyber[m]if i can just turn off bluetooth that would indeed be nice#asahi-dev · 2026-06-08 09:01
StanfordAppliedCyber[m]thanks for the help!#asahi-dev · 2026-06-08 09:01
StanfordAppliedCyber[m]unfortunate#asahi-dev · 2026-06-08 09:00
StanfordAppliedCyber[m]you have any idea how i can debug this though? like ultimately the sep is opaque, i cant inspect its internals so if it just doesn't respond i cant say why#asahi-dev · 2026-06-08 08:59
StanfordAppliedCyber[m]okay i suppose that is the brute force solution#asahi-dev · 2026-06-08 08:58
StanfordAppliedCyber[m]i also see both in my msgbuf#asahi-dev · 2026-06-08 08:58
StanfordAppliedCyber[m]msg=0x424c51580000c013 ep=0x13 tag=0xc0 opcode=0x0 data=0x424c5158#asahi-dev · 2026-06-08 08:58
StanfordAppliedCyber[m]msg=0x170610 ep=0x10 tag=0x6 opcode=0x17#asahi-dev · 2026-06-08 08:58
StanfordAppliedCyber[m]chaos_princess: #asahi-dev · 2026-06-08 08:58
StanfordAppliedCyber[m]xart and xarm are up and we see traffic on both#asahi-dev · 2026-06-08 08:54
StanfordAppliedCyber[m]it's being called from TrustedAccessory#asahi-dev · 2026-06-08 08:54
StanfordAppliedCyber[m]chaos_princess: stac#asahi-dev · 2026-06-08 08:53
StanfordAppliedCyber[m]they have names?#asahi-dev · 2026-06-08 08:53
StanfordAppliedCyber[m]ep=0x18 / 24... (full message at <https://matrix.org/oftc/media/v1/media/download/AXKd9WfnzvQtpPbs7Nhy6nBxpsJePWyHo-n9hOP48l3DizuAS1qrnrtQi3PGo0nR08DXc8myR0YCRcR60CSywgBCee7ojOhAAG1hdHJpeC5vcmcvTUl6TGxMUmlqSkRVS05UVkdoVFB6R0ph>)#asahi-dev · 2026-06-08 08:52
StanfordAppliedCyber[m]endpoint 24, cmd 0xb#asahi-dev · 2026-06-08 08:52
StanfordAppliedCyber[m]i'm trying to figure out why this endpoint won't respond#asahi-dev · 2026-06-08 08:52
StanfordAppliedCyber[m]i presume it's basically the same everywhere and hasn't changed much#asahi-dev · 2026-06-08 08:51
StanfordAppliedCyber[m]what's the most recent sep fw version we can decrypt?#asahi-dev · 2026-06-08 08:46
StanfordAppliedCyber[m]yeah allowing comical amounts of fragmentation isn't great#asahi-dev · 2026-06-08 08:41
StanfordAppliedCyber[m]and yes i'm aware of the economics/licensing/bullshit behind all of this#asahi-dev · 2026-06-08 08:41
StanfordAppliedCyber[m]true#asahi-dev · 2026-06-08 08:39
StanfordAppliedCyber[m]seriously? they had to build new hardware just to piss us off and make our lives harder?#asahi-dev · 2026-06-08 08:29
StanfordAppliedCyber[m]oh well#asahi-dev · 2026-06-08 08:29
StanfordAppliedCyber[m]well unfortunately sven it does not appear as if i owe you a beer-- turns out my handling was not correct and was likely not the issue even if it was wrong#asahi-dev · 2026-06-08 07:38
StanfordAppliedCyber[m].....well i don't know if this is the problem but i am definitely not typing the txm sep channel correctly#asahi-dev · 2026-06-08 07:35
StanfordAppliedCyber[m]also does anyone have a copy of the full xnu msgbuf lying around? i just want to see what teh full one from boot to login screen looks like and my google-fu isn't working apparently#asahi-dev · 2026-06-08 07:30
StanfordAppliedCyber[m]or a $REASONABLY_PRICED_BEVERAGE_HERE#asahi-dev · 2026-06-08 07:30
StanfordAppliedCyber[m]i'll buy you a beer if this solves my problem#asahi-dev · 2026-06-08 07:29
StanfordAppliedCyber[m]oh no, wait, that does remind me there is a txm thing that is responsible for messing with SEP#asahi-dev · 2026-06-08 07:29
StanfordAppliedCyber[m]cmd 0xb no reply#asahi-dev · 2026-06-08 07:29
StanfordAppliedCyber[m]if i send cmd 0xf then it works#asahi-dev · 2026-06-08 07:28
StanfordAppliedCyber[m]i can clearly see the SEP is on, the endpoint it's trying to talk to is 24#asahi-dev · 2026-06-08 07:28
StanfordAppliedCyber[m]in that order#asahi-dev · 2026-06-08 07:28
StanfordAppliedCyber[m]my current issue is trustedaccessory (or something like that) from userspace is trying to do some SEP thing, does a syscall, xnu tries to talk to SEP, timeouts, panics#asahi-dev · 2026-06-08 07:27
StanfordAppliedCyber[m]yeah, panics#asahi-dev · 2026-06-08 07:27
StanfordAppliedCyber[m]i can get a response from some endpoints#asahi-dev · 2026-06-08 07:26
StanfordAppliedCyber[m]I can definitely verify SEP is loaded and works#asahi-dev · 2026-06-08 07:26
StanfordAppliedCyber[m]weird#asahi-dev · 2026-06-08 07:03
StanfordAppliedCyber[m]i'm fairly deep in userspace right now, not at "it works" yet but we get a lot of messages from userspace in the msgbuf#asahi-dev · 2026-06-08 06:58
StanfordAppliedCyber[m]i probed it a bit and i wasn't able to get a reply from SEP#asahi-dev · 2026-06-08 06:57
StanfordAppliedCyber[m]i'm dealing with some problem where xnu submits a message and then just never gets a reply in the outbox#asahi-dev · 2026-06-08 06:57
StanfordAppliedCyber[m]do we know if raw boot modifies SEP at all? like are certain SEP endpoints disabled in raw boot mode?#asahi-dev · 2026-06-07 22:11
StanfordAppliedCyber[m]ack, will submit a PR shortly then, writing these docs really made me remember how annoying figuring all this stuff out in the first place was#asahi-dev · 2026-06-07 22:11
StanfordAppliedCyber[m]this is the first out of 6 documents, xnu bootstrap (all the page table mapping stuff), txm, then dart/uat/sart/nvme#asahi-dev · 2026-06-07 22:10
StanfordAppliedCyber[m]fw is the place that stuck out to me but figured i'd just ask#asahi-dev · 2026-06-07 22:08
StanfordAppliedCyber[m]where should i put the file under docs?#asahi-dev · 2026-06-07 22:07
StanfordAppliedCyber[m]i'm preparing a PR for the xnu bootstrap SPTM table document (this is the biggest one and, aside from txm, the only one that isn't an iommu whose core functionality we already understand)#asahi-dev · 2026-06-07 21:21
StanfordAppliedCyber[m]yeah i knew about ppl on m1, i read your blog post on that ages ago#asahi-dev · 2026-06-07 21:21
StanfordAppliedCyber[m]huh#asahi-dev · 2026-06-07 21:17
StanfordAppliedCyber[m]i suppose they definitely rolled it out on iphone chips first