#
✕ nick:marcan
#asahi · 2021-10-14 06:31
marcanI think we have a bunch of kernel newbies around, so that's no big deal :)
#asahi · 2021-10-14 06:30
marcanit's pretty much pick something that piques your interest and roll with it (e.g. how povik just picked up audio)
#asahi · 2021-10-14 06:29
marcannobodynada: are you interested in any particular aspect?
#asahi · 2021-10-13 19:20
marcanso there's not much reason to be concerned about backdoors other than in the SEP services themselves
#asahi · 2021-10-13 19:20
marcanit still has a DART etc
#asahi · 2021-10-13 19:20
marcanit's basically two-way isolation
#asahi · 2021-10-13 19:20
marcanthough AIUI the SEP is designed not to have power over the AP either
#asahi · 2021-10-13 19:19
marcanif you don't, don't
#asahi · 2021-10-13 19:19
marcanif you want to use its services and trust the secret blob, boot it
#asahi · 2021-10-13 19:19
marcanthat's one reason it makes sense for it to be off when the OS boots
#asahi · 2021-10-13 19:18
marcanwhich none of the others are
#asahi · 2021-10-13 19:18
marcanSEP firmware is also encrypted
#asahi · 2021-10-13 19:17
marcanif you compromise iBoot you compromise all the other ASCs, but not SEP
#asahi · 2021-10-13 19:16
marcanand signing
#asahi · 2021-10-13 19:16
marcanand memory encryption
#asahi · 2021-10-13 19:16
marcanit also has way, way fancier memory protection
#asahi · 2021-10-13 19:16
marcanwhich the other ASCs don't
#asahi · 2021-10-13 19:16
marcanit has its own secureboot chain
#asahi · 2021-10-13 19:16
marcanyeah
#asahi · 2021-10-13 19:16
marcanit's off when we boot though, right?
#asahi · 2021-10-13 19:15
marcan?
#asahi · 2021-10-13 19:15
marcanit has to
#asahi · 2021-10-13 19:14
marcaniBoot has a driver for it
#asahi · 2021-10-13 19:08
marcanthe memory controller has a lot of carveout/protection features
#asahi · 2021-10-13 18:56
marcanbut maybe the DCP isn't one of those
#asahi · 2021-10-13 18:56
marcansven: ISTR being able to read various firmwares from the AP, way back when I was dumping memory
#asahi · 2021-10-13 07:08
marcanfair point...
#asahi · 2021-10-13 07:07
marcan(but at least we'll do keyboard emu)
#asahi · 2021-10-13 07:07
marcan(which is unfortunate for M1 MBP owners since... there probably will never be enough momentum for a good ecosystem for those)
#asahi · 2021-10-13 07:06
marcantouchbar is dead
#asahi · 2021-10-13 07:06
marcanthat's pretty much confirmed
#asahi · 2021-10-13 05:28
marcanhttps://www.macrumors.com/2021/10/12/apple-event-unleashed-october-18/
#asahi · 2021-10-13 05:28
marcanJTL: M1X/M2
#asahi · 2021-10-13 05:19
marcanwell, sounds like this weekend is going to be interesting
#asahi · 2021-10-12 13:45
marcanjust create an empty one, write all the data into it, seek to 0 and pass it on :)
#asahi · 2021-10-12 13:44
marcanI think we can afford the transient memory usage and just use BytesIO there
#asahi · 2021-10-12 13:42
marcanpovik: the idea yes, but I find the Concat class a bit... overkill :)
#asahi · 2021-10-12 05:42
marcankettenis: btw, feel free to CC me on the u-boot stuff (I also just subscribed to the ML)
#asahi · 2021-10-12 01:18
marcanit could be a separate repo too though
#asahi · 2021-10-12 01:18
marcanjust need to make it all sync better with the linux version
#asahi · 2021-10-12 01:17
marcankettenis: I was thinking of keeping a copy in the m1n1 source (there's one already, horribly outdated)
#asahi · 2021-10-11 17:34
marcanthough I think the spinlock should guarantee that is not needed if we wrote it properly, but...
#asahi · 2021-10-11 17:34
marcan(probably overkill but)
#asahi · 2021-10-11 17:34
marcanj_ey: try adding a sysop("dsb sy") after those queued flag sets
#asahi · 2021-10-11 17:32
marcanit could be a memory barrier problem too
#asahi · 2021-10-11 17:30
marcanmight want to add a break to those loops when it finds the mpidr, and a check for not found
#asahi · 2021-10-11 17:30
marcanmaybe some mpidr value is wrong?
#asahi · 2021-10-11 17:28
marcanbut I'm not sure what's broken there :)
#asahi · 2021-10-11 17:27
marcanthought so
#asahi · 2021-10-11 17:00
marcan:o

plus ›