#asahi-dev · 2022-05-31 11:32
marcanbut I think in practice it will ~definitely work for a single-modeset; what I'm not so sure about is the stage1->2 latency. but stage2 will ~always have to do a re-check anyway because stage1 might be old or broken or we might need to change the resolution anyway#asahi-dev · 2022-05-31 11:31
marcanVinDuv: you can't check if it's connected without DCP, so you'd just have to re-modeset and do it all over again yes#asahi-dev · 2022-05-30 20:05
marcananyway, will figure out the details tomorrow#asahi-dev · 2022-05-30 20:04
marcanfor users who want to set non-default modes, that'd have to include the trip into stage2#asahi-dev · 2022-05-30 20:04
marcanthere's still a race, but as long as we can detect hpd -> modeset -> shut down before the monitor goes into the disconnect cycle, we're good#asahi-dev · 2022-05-30 20:02
marcanI believe so#asahi-dev · 2022-05-30 20:02
marcanyes#asahi-dev · 2022-05-30 19:51
marcanI'll implement it in the m1n1 C side tomorrow, need sleep#asahi-dev · 2022-05-30 19:51
marcanthat means no more hotplug nonsense, and HDMI stays up even if the monitor glitches :-)#asahi-dev · 2022-05-30 19:51
marcanand doesn't modeset!#asahi-dev · 2022-05-30 19:50
marcanand it comes back up if you PMGR reset it and do a proper boot, and it even does so quickly!#asahi-dev · 2022-05-30 19:50
marcanlike all the way :-)#asahi-dev · 2022-05-30 19:50
marcanjannau: so... turns out you can shut down DCP.#asahi-dev · 2022-05-30 16:21
marcaneither way the fix is what I said, we need to block, not fail, when the queue is full#asahi-dev · 2022-05-30 16:21
marcanthat would explain it, yes#asahi-dev · 2022-05-30 16:21
marcanoh, so we ack 8 messages all at once (presumably for 8 syslog pings coming in) then try to send the EPIC out before DCP clears out any of them?#asahi-dev · 2022-05-30 16:07
marcanare there 8 messages in the FIFO? that seems unlikely#asahi-dev · 2022-05-30 16:06
marcan_jannau_: right, but why does it return false?#asahi-dev · 2022-05-30 14:30
marcanand there's an iterator so you can do stuff like `for i in cpus(): context(); bt()`#asahi-dev · 2022-05-30 14:30
marcanalso the hypervisor now lets you cpu() to switch CPU without exiting the shell#asahi-dev · 2022-05-30 14:30
marcanmerged a bunch of m1n1 PR stuff that was in the queue#asahi-dev · 2022-05-26 03:29
marcanbut honestly almost nobody needs it at this point#asahi-dev · 2022-05-26 03:28
marcanalso the supply chain is still a mess#asahi-dev · 2022-05-25 04:35
marcanthanks :)#asahi-dev · 2022-05-24 15:16
marcanI give up#asahi-dev · 2022-05-24 15:16
marcansven: so apparently that bizarro peek thing was intended?#asahi-dev · 2022-05-24 14:52
marcanI'm building a new kernel and I'm going to throw some of the m1n1 display stuff in too (without config tooling for now)#asahi-dev · 2022-05-24 14:51
marcanI found it in the previous stream, just hadn't actually looked into it#asahi-dev · 2022-05-24 12:55
marcanI give up, I'm sending just the apple patches and that peek_data can keep having a terribad name#asahi-dev · 2022-05-24 12:54
marcan1 (1) person acked one of my mailbox patches in that cleanup series#asahi-dev · 2022-05-24 12:54
marcan< sven> i feel like i say this for every kernel version but i hope 5.20 will finally have useable upstream support for the m1 macbooks as well :D <- speaking of that (which needs smc...)#asahi-dev · 2022-05-24 02:55
marcangreat! means no touching SMC after all#asahi-dev · 2022-05-23 16:35
marcansomething to test from m1n1#asahi-dev · 2022-05-23 16:35
marcanalternatively maybe those devices are kind of broken and need an actual reset to come back up after a power gate#asahi-dev · 2022-05-23 16:34
marcanit could be a domain parenting issue#asahi-dev · 2022-05-23 09:53
marcanfor 4-byte alignment#asahi-dev · 2022-05-23 09:53
marcan0x..fffc#asahi-dev · 2022-05-23 09:52
marcanlet me see if I can give it a shot tomorrow#asahi-dev · 2022-05-23 09:52
marcanI think so#asahi-dev · 2022-05-21 17:24
marcanthat hardcoded addr will only work on t8103 PMUs#asahi-dev · 2022-05-21 17:24
marcanneeds to properly read the ADT node to support t600x though#asahi-dev · 2022-05-21 09:42
marcanalso apparently lina pushed m1n1/hw/spmi.py last night which probably also overlaps#asahi-dev · 2022-05-21 09:40
marcanmillenialhacker: the i2c stuff looks like the usual, I assume you've seen m1n1/hw/i2c.py#asahi-dev · 2022-05-21 09:04
marcanthe bit defs for those PS regs are in the Linux driver (and m1n1)#asahi-dev · 2022-05-21 09:02
marcan # 27 ISPSENS3 id: 28 psreg: 11:19 flags: C unk1_0: 0 unk1_1: 0 ctl_reg: 1:0x35 unk3: 0 0 0 0 @ 0x23b700198 = 0x04000244 #asahi-dev · 2022-05-21 09:02
marcan # 26 ISPSENS2 id: 27 psreg: 11:18 flags: C unk1_0: 0 unk1_1: 0 ctl_reg: 1:0x34 unk3: 0 0 0 0 @ 0x23b700190 = 0x04000244 #asahi-dev · 2022-05-21 09:02
marcan # 25 ISPSENS1 id: 26 psreg: 11:17 flags: C unk1_0: 0 unk1_1: 0 ctl_reg: 1:0x33 unk3: 0 0 0 0 @ 0x23b700188 = 0x04000244 #asahi-dev · 2022-05-21 09:02
marcan # 24 ISPSENS0 id: 25 psreg: 11:16 flags: C unk1_0: 0 unk1_1: 0 ctl_reg: 1:0x32 unk3: 0 0 0 0 @ 0x23b700180 = 0x04000244 #asahi-dev · 2022-05-21 09:02
marcanalso these might be relevant:#asahi-dev · 2022-05-21 09:02
marcanbut likely everything outside of that you don't