#
✕ nick:marcan
#asahi-dev · 2022-10-24 16:51
marcannull: yeah, I read it correctly :)
#asahi-dev · 2022-10-24 16:51
marcanhey, it's pretty low traffic (so far) :p
#asahi-dev · 2022-10-24 16:50
marcankettenis, j`ey, sven, jannau, anyone else? I assume everyone's subscribed to asahi@ anyway, but let me know if you're explicitly interested
#asahi-dev · 2022-10-24 16:47
marcanthanks! you have a good day :)
#asahi-dev · 2022-10-24 16:47
marcangot it, I'll see if I can type something up tomorrow
#asahi-dev · 2022-10-24 16:46
marcancool. I assume maz might also be interested, and the usual suspects from around here
#asahi-dev · 2022-10-24 16:46
marcan(l-a-k, lkml, asahi, anything else?)
#asahi-dev · 2022-10-24 16:45
marcanany particular lists I should CC if I start this?
#asahi-dev · 2022-10-24 16:45
marcanyeah, I need to get some sleep too
#asahi-dev · 2022-10-24 16:44
marcansure, and I'd be perfectly fine with "PSCI-inspired but simpler UEFI runtime protocol"
#asahi-dev · 2022-10-24 16:42
marcanyeah, you'd need to play tricks to unmap it sometimes
#asahi-dev · 2022-10-24 16:41
marcanthen you wouldn't have to mess with TTBRs
#asahi-dev · 2022-10-24 16:40
marcan(some inspiration from coprocessor mapping stuff there)
#asahi-dev · 2022-10-24 16:40
marcanrequire that the entire runtime thing is position-independent, and provide a descriptor of what MMIO/RAM regions it wants mapped. kernel makes sure those are mapped, fills in the pointers, maps it in kernel space, then just calls it
#asahi-dev · 2022-10-24 16:39
marcanI wonder if it would make sense to just have it "integrate" into the kernel
#asahi-dev · 2022-10-24 16:38
marcanyeah
#asahi-dev · 2022-10-24 16:36
marcanbut I imagine the overhead is probably measurable
#asahi-dev · 2022-10-24 16:36
marcanmight be okay for just a cpuidle trampoline though, that doesn't have to touch RAM other than for the GPR save/restore...
#asahi-dev · 2022-10-24 16:36
marcanalso re MMU off, that *works* but performance of these things with caches off is pretty abysmal, FWIW
#asahi-dev · 2022-10-24 16:35
marcanyeah
#asahi-dev · 2022-10-24 16:35
marcanif you think there's room for something other than spin-table and psci that isn't just a version of either, then that changes the situation
#asahi-dev · 2022-10-24 16:34
marcanbut my impression was that people were really dead set on PSCI as the One True Way, hence the idea of tunneling it via UEFI
#asahi-dev · 2022-10-24 16:34
marcanI think if it's not going to be PSCI, and it's going to be via runtime services, then we might as well drop the layer and just call it a bespoke UEFI protocol
#asahi-dev · 2022-10-24 16:33
marcanwhat are the major differences?
#asahi-dev · 2022-10-24 16:32
marcannot that I'm opposed to doing "PSCI, but simpler" if you think it's the way to go, but it's going to have to be some kind of call-based interface at some point
#asahi-dev · 2022-10-24 16:31
marcanI think regardless of whether UEFI is in the mix or not, we're going to end up reinventing PSCI poorly...
#asahi-dev · 2022-10-24 16:30
marcanyeah, we'd need something for idle, but also that doesn't really work power-wise unless we come up with a signaling mechanism more refined than "just spin on a wfe"
#asahi-dev · 2022-10-24 16:29
marcanbut what's your take on it?
#asahi-dev · 2022-10-24 16:29
marcanand it's also been long enough to be worth having a proper discussion anew anyway
#asahi-dev · 2022-10-24 16:28
marcanyeah, I agree
#asahi-dev · 2022-10-24 16:28
marcanso I'm very interested in what you think a first approximation take should be like and I'd be happy to run with that and see how it goes
#asahi-dev · 2022-10-24 16:27
marcan(and yes, this is for the ML at some point, but it also doesn't hurt to have a PoC and it's not like we haven't shipped dozens of those to our users already, and changed them 3 times each, much to kettenis' chagrin :p)
#asahi-dev · 2022-10-24 16:26
marcanISTR that back in the day that was considered and the conclusion was we'd be reinventing UEFI, but it's been a while
#asahi-dev · 2022-10-24 16:26
marcannull: I'm certainly fine with non-UEFI if you think we can specify exactly what the call semantics should be cleanly (especially re the MMU)
#asahi-dev · 2022-10-24 16:05
marcaninstalls macOS 12.3 on the MBA
#asahi-dev · 2022-10-24 16:04
marcanso then maybe it's just something dumb
#asahi-dev · 2022-10-24 16:04
marcancould be (since iboot has to set the backlight already)
#asahi-dev · 2022-10-24 16:01
marcan_jannau_: it sounds like we probably need to upload some/all that data and then just set the backlight property, probably?
#asahi-dev · 2022-10-24 15:58
marcanwhat the hell is a marketing table...
#asahi-dev · 2022-10-24 15:58
marcanbacklight-marketing-table = 0000000000000400e1fa050066a60800703d0c007a1411005c8f1700c2352000c2b52b0014ee3a008f024f006666690000008c003333b900291cf400a4b040010000a401
#asahi-dev · 2022-10-24 15:58
marcannot sure if you saw that
#asahi-dev · 2022-10-24 15:58
marcanthere's a pile of backlight calibration data in the ADT
#asahi-dev · 2022-10-24 15:57
marcan_jannau_: we can't ship DCP until this is in, so...
#asahi-dev · 2022-10-24 15:40
marcan(otherwise it's going to be the PSCI thing next I guess?)
#asahi-dev · 2022-10-24 15:39
marcan_jannau_: so should I schedule a stream to work out brightness control, or do you want to look at that yourself? :)
#asahi-dev · 2022-10-24 15:34
marcanI guess that means it's not :)
#asahi-dev · 2022-10-24 15:31
marcanis that supposed to be ""?
#asahi-dev · 2022-10-24 15:31
marcan {"region-id-74", "", true, false, false},
#asahi-dev · 2022-10-24 15:31
marcanalso
#asahi-dev · 2022-10-24 15:30
marcanbut I guess the Max has all the Ultra ones for whatever reason

plus ›