#
✕ nick:marcan
#asahi-dev · 2023-04-09 10:59
marcanPCIe has definitely changed to some extent, ADT reg layout is slightly different but things do somewhat line up with the old one. will probably need some tweaking.
#asahi-dev · 2023-04-09 10:59
marcanmaybe iBoot initializes this now
#asahi-dev · 2023-04-09 10:59
marcanI mean the reg entry is there but I don't see it ever getting hit in relation to pcie except one weird sequence where it's writing ECAM stuff and the fuse read is 0 anyway?
#asahi-dev · 2023-04-09 10:56
marcaninterestingly, it seems there are no fuses for pcie
#asahi-dev · 2023-04-09 09:23
marcanif you want it faster, you can help out ;)
#asahi-dev · 2023-04-09 09:21
marcanactually someone else already did the chickens, I just need to deal with fuses I guess
#asahi-dev · 2023-04-09 09:20
marcanassuming it's what we have in the downstream kernel it works for me :p
#asahi-dev · 2023-04-09 09:19
marcanthanks!
#asahi-dev · 2023-04-09 09:05
marcanthis might end up requiring a complete rewrite of the DCP code in m1n1 to be much more complex, TBD
#asahi-dev · 2023-04-09 09:04
marcanbuying unsupported devices is on you, not us :)
#asahi-dev · 2023-04-09 09:04
marcannope.
#asahi-dev · 2023-04-09 08:14
marcanjust chickens and fuses everywhere
#asahi-dev · 2023-04-09 08:14
marcanmight not bother to stream this one since honestly it's going to be boring unless something really unexpected comes up
#asahi-dev · 2023-04-09 08:14
marcanI guess it's about time I bring up the M2 Max...
#asahi-dev · 2023-04-09 08:07
marcanChaosPrincess: can you send the t8112 PWM DT changes too? (on top of the asahi-soc/dt branch)
#asahi-dev · 2023-04-09 02:06
marcan:/
#asahi-dev · 2023-04-08 16:31
marcansven: did you have something for that on the pcie side already?
#asahi-dev · 2023-04-08 16:31
marcanwe really need to implement proper hotplug
#asahi-dev · 2023-04-08 16:31
marcanI think it's that aqantia thing that sometimes takes longer to initialize...
#asahi-dev · 2023-04-08 16:31
marcansigh.
#asahi-dev · 2023-04-08 16:31
marcansounds like possibly a timing issue?
#asahi-dev · 2023-04-08 16:30
marcanyay heisenbugs :(
#asahi-dev · 2023-04-08 15:56
marcanhm, ok
#asahi-dev · 2023-04-08 15:50
marcanwe *know* the driver can't handle power cycling anyway so that PD should get the always-on flag anyway, tbh (at least for now)
#asahi-dev · 2023-04-08 15:48
marcanso it might break some ports and not others
#asahi-dev · 2023-04-08 15:48
marcanIIRC power cycling caused unreliability
#asahi-dev · 2023-04-08 15:48
marcanbut it's shared by DARTs and such
#asahi-dev · 2023-04-08 15:48
marcanI seriously need to fix that, maybe I can come up with a patch tomorrow
#asahi-dev · 2023-04-08 15:47
marcanbecause we *know* that stuff is just horribly broken and somehow is breaking a lot more since 6.2
#asahi-dev · 2023-04-08 15:47
marcandoes it work with that bootarg to keep power domains on?
#asahi-dev · 2023-04-08 15:46
marcanand this isn't the same old power domain problem again?
#asahi-dev · 2023-04-08 15:46
marcaninteresting
#asahi-dev · 2023-04-08 15:44
marcanjannau: reproducibly fails with the latest everything?
#asahi-dev · 2023-04-08 15:44
marcanack, will test
#asahi-dev · 2023-04-08 15:31
marcanwhich machine?
#asahi-dev · 2023-04-08 15:30
marcanjannau: other people reported that in the past, I increased the timeout thinking it was that but we might have another issue
#asahi-dev · 2023-04-06 18:11
marcanso that's probably the iphone SoC it came from
#asahi-dev · 2023-04-06 18:11
marcanFWIW the macOS driver references H7 which is A8 (T7000)
#asahi-dev · 2023-04-06 18:10
marcanhah, cute
#asahi-dev · 2023-04-06 18:09
marcanbut I guess they kept most of the init then?
#asahi-dev · 2023-04-06 18:09
marcanyeah, that I know
#asahi-dev · 2023-04-06 18:08
marcan(since I remember someone got linux running on it in dumb fb mode, definitely no driver for that in sandcastle)
#asahi-dev · 2023-04-06 18:08
marcanalso AIUI on T2 Macs that's also initialized by iBoot, so it probably came from there
#asahi-dev · 2023-04-06 18:07
marcanI guess it's just... older pre-DCP code at that point, unrelated to much else
#asahi-dev · 2023-04-06 18:07
marcanbut it's interesting that they'd keep the iboot init
#asahi-dev · 2023-04-06 18:07
marcanfrom what I saw of the driver it *is* a copypasted primary iphone display hardware-wise (a few generations back)
#asahi-dev · 2023-04-06 18:04
marcanI wasn't expecting that
#asahi-dev · 2023-04-06 18:04
marcankind of surprising iboot sets up the touch bar... huh.
#asahi-dev · 2023-04-06 18:03
marcanbut Apple seems to like their "registerstream" concept from various strings I've seen, so no wonder it's a fifo
#asahi-dev · 2023-04-06 18:03
marcaneither that or outright shadowing

plus ›