#asahi-dev · 2022-06-21 18:50
millenialhackerThanks guys! #asahi-dev · 2022-06-21 18:46
millenialhackervery helpful! #asahi-dev · 2022-06-21 18:46
millenialhackerj`ey thanks! #asahi-dev · 2022-06-21 18:44
millenialhackerSilly question: There's a way to force reload of hypervisor scripts without actually restarting the laptop? I'm almost sure I saw marcan doing this in one of his videos long time ago, but I do not remember and rebooting more than three times makes my M1 go to 1TR/Recovery#asahi-dev · 2022-06-17 19:52
millenialhackerInteresting, makefile is not looking for toolchain at /usr/local/opt/llvm/bin/, is that new? I don't remember needing to change my toolchain path, currently it's failing due to not found error#asahi-dev · 2022-06-17 19:48
millenialhackerlet me compile it and install it again #asahi-dev · 2022-06-17 19:48
millenialhackerABI got broken? #asahi-dev · 2022-06-17 19:47
millenialhackerNo, I'm still using old version #asahi-dev · 2022-06-17 19:45
millenialhackerI just pull latest changes in m1n1, but now when I tried to run hypervisor, it fails with "OverflowError: Python int too large to convert to C ssize_t" has anyone experience this issue ? #asahi-dev · 2022-06-15 21:32
millenialhackerlol\#asahi-dev · 2022-06-15 10:14
millenialhackers/a/an/g#asahi-dev · 2022-06-15 10:14
millenialhackersven: thanks! it has been a enlightning experience :D #asahi-dev · 2022-06-14 23:12
millenialhackersven: look this beauty: https://pastebin.com/x66LdVQC#asahi-dev · 2022-06-14 23:10
millenialhackerhttps://github.com/kellermanrivero/m1n1/commit/dc514befc1ebe158c91364205493a19288c9a2ad#asahi-dev · 2022-06-14 23:10
millenialhackerI discovered that in order to read ISP logs I need to be faster than Mac OS. Mac OS reads the log buffer address, print the log and clear the buffer address from the TERMINAL channel ring buffer, that explain why when I tried to read entries the address was 0x0 (0x1). Just fixed that in latest commit: #asahi-dev · 2022-06-14 08:41
millenialhackerI think I'm gonna start cleaning up stuff to make a couple of PRs to m1n1 repo #asahi-dev · 2022-06-14 08:39
millenialhackersven: thanks #asahi-dev · 2022-06-14 08:39
millenialhackersven: at least at low level being able to see communication through is great, but I still have not idea how OS operates in the high level. :D #asahi-dev · 2022-06-13 18:28
millenialhackerSample logs (just TERMINAL channel): https://pastebin.com/Fm0SrXbF#asahi-dev · 2022-06-13 18:26
millenialhackerFor people interested in ISP, I have done some progress this week. At this point, tracer is able to read properly 4 of 7 IPC channels, including ISP internal terminal log buffer. https://github.com/kellermanrivero/m1n1/commit/6bdb879fc0cfb35667714cccfdc85f00c07745ee#asahi-dev · 2022-05-23 14:34
millenialhackerTy Yuni#asahi-dev · 2022-05-23 10:07
millenialhackerbtw, @povik and others thanks! #asahi-dev · 2022-05-23 09:56
millenialhackerI wonder how I can guess what those commands do, any suggestion beside doing static RE on Apple KEXT? #asahi-dev · 2022-05-23 09:53
millenialhackerxD #asahi-dev · 2022-05-23 09:53
millenialhacker@povik, do you mean iova & 0xfffffffc? #asahi-dev · 2022-05-23 09:52
millenialhacker@povik, do you mean iova & 0xffffff00? #asahi-dev · 2022-05-23 09:20
millenialhackerI'm learning a lot with this stuff btw#asahi-dev · 2022-05-23 09:19
millenialhackerI still have no idea what those commands do, but after two samples, command stream looks quite the same in both samples. #asahi-dev · 2022-05-23 09:18
millenialhackerIt's quite weird, maybe a bug? As long as ISP uses the address no matter if it is X or X+1, it will work, but I wonder why they have that little offset changing for some specific commands#asahi-dev · 2022-05-23 09:16
millenialhackerThe weird stuff is: Apple something uses IOVA address X and sometimes it uses X + 1#asahi-dev · 2022-05-23 09:15
millenialhackerA fun story about Apple ISP. The IO IPC channel is used to send commands to ISP, that channel is 8 bytes wide, so they put the IOVA address of the CMD struct which is apparently 8 bytes wide too, but they never change the Address, they mostly use the same iova address and replace the contents of memory. Here's a couple of sessions I traced today: https://gist.github.com/kellermanrivero/c92aec82d59d6d0a4d27b3381d1038b7#asahi-dev · 2022-05-21 09:34
millenialhackers/it/it's/g#asahi-dev · 2022-05-21 09:34
millenialhackerI need to polish my documentation, it more raw notes at this point #asahi-dev · 2022-05-21 09:31
millenialhackermarcan: Thanks! This is super helpful. Yes, the 4XXX range is ISP specific, I documented here a few weeks ago: https://github.com/AsahiLinux/docs/wiki/HW:Camera#asahi-dev · 2022-05-21 06:02
millenialhackerI need to clean it up to remove some unneeded changes, but at least some good progress tracing IPC channels#asahi-dev · 2022-05-21 06:02
millenialhackerGot some progress with ISP, here's my tracer: https://github.com/kellermanrivero/m1n1/commit/5d8c9c669ef097e541fd06649103408483a0a7bd#asahi-dev · 2022-05-17 12:36
millenialhacker@jannau: yes, it's 32 bytes per segment, each segment divided in four 8-bytes value. What do you mean by phy-add in 32-bits cells? #asahi-dev · 2022-05-17 12:05
millenialhackerif I split in 8 bytes (numbers I see doesn't look IOVA or PHY address) #asahi-dev · 2022-05-17 12:04
millenialhackere.g: 00 80 9D 00 08 00 00 00 00 00 00 00 00 00 00 00 00 80 9D 00 08 00 00 00 00 80 9B 00 03 00 00 00 00 C0 9A 01 08 00 00 00 00 80 9B 00 00 00 00 00 00 80 9B 00 0F 00 00 00 00 C0 41 00 00 00 00 00#asahi-dev · 2022-05-17 12:03
millenialhackerWhat I know so far it's that it should include IOVA / PHY / LENGHT and some REMAP (???) stuff. but when I try to read it from my ADT, it makes no sense. #asahi-dev · 2022-05-17 11:57
millenialhackerDoes anyone know how to interpret "segment-ranges" property in ADT ? #asahi-dev · 2022-04-27 08:43
millenialhackerThe funny thing: It's rotates over DARTs, sometimes you must use dart[0], sometimes dart[1] and sometimes dart[2]. The interesting stuff is that it doesn't change over frame, it just change over hardware power on. #asahi-dev · 2022-04-27 08:42
millenialhackerI got it by accident but it seems 0x3a28000 (IOVA) is the syslog of ISP. https://pastebin.com/kp64tEXL #asahi-dev · 2022-04-26 17:39
millenialhackerpoint*#asahi-dev · 2022-04-26 17:39
millenialhackerty, now I got your p\];p['#asahi-dev · 2022-04-26 17:34
millenialhackersven, regarding corellium, I just noticed they initialized iommu driver instance with some data regarding number of SIP supported by DART instance, so I wonder if we can at least see how they did it.#asahi-dev · 2022-04-26 17:32
millenialhackersven that makes sense, I noticed that independently of the DART instance, if I do IOREAD on them using a IOVA I always get same data, which made me thing they were using same page tables.#asahi-dev · 2022-04-26 17:23
millenialhackershould corellium codebase be treated specially like copyright disclaimer suggests? sven, marcan?#asahi-dev · 2022-04-26 17:11
millenialhackersame with TCR, i see the values when TCR is set written in other offsets as well#asahi-dev · 2022-04-26 17:10
millenialhackerBTW I just repurpose DART mapper in my branch to monitor six darts instances at same time, so I know those extra offsets do not belong to other DART from ISP.