#asahi-gpu · 2022-03-03 17:38
svenno, it just doesn’t come back up#asahi-gpu · 2022-03-03 17:37
sveni know that the ans asc can’t be restarted after it’s been crashed e.g.#asahi-gpu · 2022-03-03 17:21
svenhttps://github.com/AsahiLinux/linux/blob/asahi/drivers/soc/apple/rtkit-crashlog.c#asahi-gpu · 2022-03-03 17:21
svenit’s also not as pretty as macOS though #asahi-gpu · 2022-03-03 17:20
sveni wrote the one for the kernel #asahi-gpu · 2022-03-03 17:20
svenpretty sure marcan wrote the Python decoder #asahi-gpu · 2022-02-21 12:50
svenhttps://i.kym-cdn.com/photos/images/newsfeed/000/572/078/d6d.jpg#asahi-gpu · 2022-02-14 13:52
svenand i guess that GPTBAT pointer is not inside the CTRR firmware region either?#asahi-gpu · 2022-02-14 12:20
svenhrm... unless they also limit the physical addresses that are allowed to contain code like they do for the XNU kernel#asahi-gpu · 2022-02-14 12:19
sven(including the code running in GL2)#asahi-gpu · 2022-02-14 12:18
svenyeah. if we can modify the full pagetable we can just map the firmware (or parts of it) somewhere else#asahi-gpu · 2022-02-14 12:18
sveni didn't follow the pagetable discussion, but I think that if you control whatever TTBR0/1 points to you can even hijack the GL2 code#asahi-gpu · 2022-02-14 12:13
svenwe can only disable the first one#asahi-gpu · 2022-02-14 12:13
svenone for the kernel and one for the co-processor firmware#asahi-gpu · 2022-02-14 12:13
sventhey have two CTRR regions#asahi-gpu · 2022-02-14 12:13
svenalmost!#asahi-gpu · 2022-02-14 12:13
sventhey do the same for XNU but that protection can be disabled in 1TR#asahi-gpu · 2022-02-14 12:12
sveni think apple calls that CTRR. iBoot loads the firmware of all co-procs into a memory region and setups CTRR to write protect it#asahi-gpu · 2022-02-14 12:11
svenit's in a region of RAM that's write protected#asahi-gpu · 2022-02-14 12:11
sven*replace#asahi-gpu · 2022-02-14 12:11
svenso we can't easily replaced the firmware#asahi-gpu · 2022-02-14 12:09
svenGXF is what apple called "guarded exception level". it's a lateral level to EL2 called GL2 with a different interpretation of pagetable permissions and usually the only code that gets to modify pagetables#asahi-gpu · 2022-02-14 12:09
sveni'm not sure i understand that question#asahi-gpu · 2022-02-14 12:04
svenand then there was ofc that one kernel with llvm bitcode released under the open source license. that should contain full symbols and then some for GXF as well :D#asahi-gpu · 2022-02-14 12:03
sven*release#asahi-gpu · 2022-02-14 12:03
svenbut then they were removed the next released again#asahi-gpu · 2022-02-14 12:02
sveniirc GXF also has (had?) some symbols in some XNU kernel#asahi-gpu · 2022-02-09 08:48
svenweird that gfx-asc can essentially get access to every page#asahi-gpu · 2022-02-09 08:47
svenhttps://github.com/AsahiLinux/m1n1/blob/41f0874f9999d2ac95ef5418dfae4fdd42cdc507/proxyclient/m1n1/fw/asc/crash.py#L137#asahi-gpu · 2022-02-09 08:46
svenjust send 0x40 or 0x22 to it and it should crash #asahi-gpu · 2022-02-09 08:28
svenendpoint 0x1 is probably the normal rtkit crashlog#asahi-gpu · 2022-02-07 19:18
sven*that#asahi-gpu · 2022-02-07 19:18
svensounds like it maps those with thus UAT iommu as well, which also makes sense I guess#asahi-gpu · 2022-02-07 08:08
svenjust be glad you didn't decide to look into DCP instead. that thing's full of magic structs from what i can tell#asahi-gpu · 2022-02-07 07:16
svenyes#asahi-gpu · 2022-02-07 07:15
svenerm, let me check. but iirc it was just a list of u64s (paddr, size, iova) or something similar#asahi-gpu · 2022-02-07 07:15
svenusually the CPU needs to map in the firmware for most of these co-processors. it's usually just already done by iBoot fwiw#asahi-gpu · 2022-02-07 07:13
sventhere's a segment-ranges property in the ADT that usually contains the mapping for the firmware fwiw#asahi-gpu · 2022-02-07 07:12
svenyeah, sounds like it#asahi-gpu · 2022-02-07 07:11
svenphire> Thats RTKSTACK repeated <-- sounds like that's also how the firmware and RAM of the co-processor is mapped then#asahi-gpu · 2021-10-18 19:43
svenhuh, interesting#asahi-gpu · 2021-07-10 17:43
svenfair enough :-)#asahi-gpu · 2021-07-10 17:42
svenfwiw, so far all dma transactions i've seen between the cpu and any peripheral have been cache coherent#asahi-gpu · 2021-05-30 19:14
svenoh... more streams. that means i have more excuses to procrastinate :> #asahi-gpu · 2021-05-06 18:53
sventhat line was probably still stuck in my head when i wrote that post yesterday night#asahi-gpu · 2021-05-06 18:52
sven(and yet fools seldom differ)#asahi-gpu · 2021-05-06 18:52
sven<@bloom> Hey that's my line :-p <-- great minds think alike ;)