#asahi-re · 2026-07-03 10:21
StanfordAppliedCyber[m]i'm also doing that tomorrow because it's 3:20 here in california and i need to sleep#asahi-re · 2026-07-03 10:21
StanfordAppliedCyber[m]if you really want i can dig through my git log, but that wouldn't be trivial for me to get#asahi-re · 2026-07-03 10:19
StanfordAppliedCyber[m]i spent a long time trying to figure this out#asahi-re · 2026-07-03 10:19
StanfordAppliedCyber[m]i tried both#asahi-re · 2026-07-03 09:56
StanfordAppliedCyber[m]definitely does not work#asahi-re · 2026-07-03 09:56
StanfordAppliedCyber[m]first thing i tried#asahi-re · 2026-07-03 09:56
StanfordAppliedCyber[m]yes#asahi-re · 2026-07-03 09:54
StanfordAppliedCyber[m]the uat continue ones i can dodge just by ensuring that the non continue endpoint completes the work all at once and never returns the continue return code#asahi-re · 2026-07-03 09:53
StanfordAppliedCyber[m]basically on retype to a page table type sptm keeps track of dirty pages and flushes them with those instructions#asahi-re · 2026-07-03 09:53
StanfordAppliedCyber[m]for context btw the weird gxf instructions occur in a few endpoints, in the uat continue endpoints, but the relevant one is the RETYPE endpoint in the xnu bootstrap table#asahi-re · 2026-07-02 21:48
StanfordAppliedCyber[m]mischa85: i'm writing cleanroom docs on my sptm emulator right now, i finished xnu-bootstrap a while ago (it's missing some stuff i learned recently, notably the cache stuff, but is overall correct), i finished sart yesterday, will work on dart + nvme + uat today, probably gonna do txm last because i may end up tossing that emulator entirely and just doing the "virtualize txm" thing separately#asahi-re · 2026-07-01 22:36
StanfordAppliedCyber[m]mischa85: what are you working on? i have a very advanced tree for a m4 hypervisor + sptm emulator (tainted) so if you're doing any virtualizatio work i'm fairly certain there's a lot of room for collaboration#asahi-re · 2026-07-01 22:35
StanfordAppliedCyber[m]aside from that there's nothing magical at all, my sptm emulator can write to it just fine#asahi-re · 2026-07-01 22:34
StanfordAppliedCyber[m]sptm just says "you shall not write to this" and enforces that in the page tables#asahi-re · 2026-07-01 22:34
StanfordAppliedCyber[m]yes i can verify everything sven + chaos said#asahi-re · 2026-06-30 23:45
StanfordAppliedCyber[m]they're writable from el2, source: currently doing so#asahi-re · 2026-06-30 08:35
StanfordAppliedCyber[m]lol#asahi-re · 2026-06-30 08:35
StanfordAppliedCyber[m]i was wondering why your messages weren't showing up#asahi-re · 2026-06-30 08:20
StanfordAppliedCyber[m]that would piss me off also#asahi-re · 2026-06-30 08:20
StanfordAppliedCyber[m]I do think there are correct ways to use them#asahi-re · 2026-06-30 08:20
StanfordAppliedCyber[m]oh yeah i'm fully with you that using llms poorly is terrible#asahi-re · 2026-06-30 08:08
StanfordAppliedCyber[m]i've found use for them in other domains (not asahi obviously)#asahi-re · 2026-06-30 08:08
StanfordAppliedCyber[m]but if it's 90% correct and 10% bs it can be useful#asahi-re · 2026-06-30 08:07
StanfordAppliedCyber[m]which they will, not denying that#asahi-re · 2026-06-30 08:06
StanfordAppliedCyber[m]not to say that the person used them well (i haven't looked at the paper)#asahi-re · 2026-06-30 08:06
StanfordAppliedCyber[m]fwiw llms are actually very good at looking through large blobs of asm and writing docs about it#asahi-re · 2026-06-18 22:56
StanfordAppliedCyber[m]literally just struct A, struct B, struct C, etc, is probably safest#asahi-re · 2026-06-18 22:56
StanfordAppliedCyber[m]which is actually better from a copyright perspective, just don't include anything tainted in whatever docs you write#asahi-re · 2026-06-18 22:56
StanfordAppliedCyber[m]i don't think we have full field info from the included symbols, so you'll have to infer based on us + surrounding function names#asahi-re · 2026-06-18 22:54
StanfordAppliedCyber[m]_oftc_jailbreakhat[m]: the most helpful thing you could do for me, imo, to get the gpu driver working, is just try to dump all the structs that comprise the firmware abi#asahi-re · 2026-06-11 21:36
StanfordAppliedCyber[m]no clue#asahi-re · 2026-06-11 21:36
StanfordAppliedCyber[m]like the iboot code?#asahi-re · 2026-06-04 08:03
StanfordAppliedCyber[m]i'm just going to include a description + offset and not include a name entirely#asahi-re · 2026-06-03 21:28
StanfordAppliedCyber[m]thanks for the assistance!#asahi-re · 2026-06-03 21:27
StanfordAppliedCyber[m]okay i'll see what i can do#asahi-re · 2026-06-03 21:27
StanfordAppliedCyber[m].........this feels like "just rename all the fields and use primitive types" except with more english around it#asahi-re · 2026-06-03 21:26
StanfordAppliedCyber[m]it would contain the exact same fields in the exact same way, just with primitive types and different names#asahi-re · 2026-06-03 21:26
StanfordAppliedCyber[m]what does "write your own struct" mean? if offset 0x20 is some u32 describing number of available pages (for example), then i'm not sure what "write your own struct" even means#asahi-re · 2026-06-03 21:25
StanfordAppliedCyber[m]sure i can do that#asahi-re · 2026-06-03 21:25
StanfordAppliedCyber[m]so basically just rename all the fields?#asahi-re · 2026-06-03 21:24
StanfordAppliedCyber[m]sk is secure kernel, whether or not exclaves are working (i set this to 0 for now, but will enable once i add exclave support, but i'll wait for yuka to do her thing first)#asahi-re · 2026-06-03 21:24
StanfordAppliedCyber[m]first available physical address#asahi-re · 2026-06-03 21:23
StanfordAppliedCyber[m]some of the fields are first_avail_phys, sk_bootstrapped and cpu_stack_papt_start#asahi-re · 2026-05-13 08:58
StanfordAppliedCyber[m]thank you for the clarification there#asahi-re · 2026-05-13 08:57
StanfordAppliedCyber[m]i'm asking because i'm not experienced with RE and want clear guidance on policy, not because i'm foaming at the mouth to disassemble something#asahi-re · 2026-05-13 08:44
StanfordAppliedCyber[m]I haven't touched them at all aside from the sptm firmware, but say the agx firmware could be nice for figuring out what the structs are exactly#asahi-re · 2026-05-13 08:43
StanfordAppliedCyber[m]question, how much are we allowed to disasm the firmware blobs?#asahi-re · 2026-05-13 05:12
StanfordAppliedCyber[m]some of it is definitely useful, it'll all need a bit of a refactor to fit, and some things will still need to be cleanroomed, but I'll discuss this further when my code works and i've cleaned it up#asahi-re · 2026-05-13 04:51
StanfordAppliedCyber[m]i'll see how it works#asahi-re · 2026-05-13 04:51
StanfordAppliedCyber[m]brain extra small