#
✕ nick:marcan
#asahi-dev · 2023-04-18 16:22
marcan>>>
#asahi-dev · 2023-04-18 16:22
marcanHypervisor exited. Entering shell.
#asahi-dev · 2023-04-18 16:22
marcanTTY> HV: All CPUs exited
#asahi-dev · 2023-04-18 16:22
marcanTTY> HV: Exiting from CPU 7
#asahi-dev · 2023-04-18 16:22
marcanTTY> HV: Requesting exit of CPU#7 from the guest
#asahi-dev · 2023-04-18 16:22
marcan[cpu7] Guest is shutting down CPU
#asahi-dev · 2023-04-18 16:22
marcanthis happens:
#asahi-dev · 2023-04-18 16:21
marcananyway, I implemented CPU shutdown enough to survive a suspend cycle
#asahi-dev · 2023-04-18 16:12
marcanmaybe it's due to the dodgy power readings?
#asahi-dev · 2023-04-18 16:12
marcanweird, so clpc or something is shutting down the Pcore clusters on this t602x and then they never come back, processes just stay stuck on the ecores
#asahi-dev · 2023-04-18 15:16
marcanand yes you can do what DCP used to do and drop it in the driver (it's a hack but it works), though eventually that won't be necessary
#asahi-dev · 2023-04-18 15:15
marcanit is a known problem, don't design around it
#asahi-dev · 2023-04-18 15:15
marcanthe always-on thing will be fixed in core code eventually
#asahi-dev · 2023-04-18 15:15
marcanwe don't want to be passing around unnecessary blobs
#asahi-dev · 2023-04-18 15:15
marcanChaosPrincess: no, read the regs in the kernel
#asahi-dev · 2023-04-18 14:57
marcanif it's just a save/restore it's not worth using the existing driver
#asahi-dev · 2023-04-18 14:57
marcannew driver
#asahi-dev · 2023-04-18 12:08
marcanthat should do it
#asahi-dev · 2023-04-18 12:08
marcanioreg -l -p IODeviceTree | grep voltage-states
#asahi-dev · 2023-04-18 12:07
marcanyes, you can see it in the pmgr stuff
#asahi-dev · 2023-04-18 12:06
marcanI'm guessing the console isn't quite right then
#asahi-dev · 2023-04-18 12:06
marcanthat kind of makes sense because macsmc tries to do a printk when you press the button
#asahi-dev · 2023-04-18 12:06
marcanwell
#asahi-dev · 2023-04-18 12:06
marcanoh, heh
#asahi-dev · 2023-04-18 11:47
marcanneeds a m1n1 stage1 update to fix "properly" but I can probably just restore the bits in stage2 as a one-off
#asahi-dev · 2023-04-18 11:47
marcanI think that made us lose at least one bit on pcores and I don't know what it does
#asahi-dev · 2023-04-18 11:47
marcanwe were clearing all bits of the pstate registers because I wrote & FOO instead of & ~FOO
#asahi-dev · 2023-04-18 11:47
marcannp, let me confirm that I can work around the snafu
#asahi-dev · 2023-04-18 11:45
marcanwhat machine are you testing on?
#asahi-dev · 2023-04-18 11:45
marcanthere's a couple m1n1 changes that might affect that
#asahi-dev · 2023-04-18 11:45
marcanchadmed: might want to hold off on measurements
#asahi-dev · 2023-04-18 11:45
marcanoh yeah, there's something else...
#asahi-dev · 2023-04-18 10:31
marcanbbl
#asahi-dev · 2023-04-18 10:31
marcanI should get dinner, I clearly am not thinking right :p
#asahi-dev · 2023-04-18 10:31
marcanmax we have on t8103 is 15/3204, t600x is 15/3228, t8112 17/3504, t6021/j416 19/3696
#asahi-dev · 2023-04-18 10:30
marcaner another typo, let me redo that
#asahi-dev · 2023-04-18 10:30
marcanso this sounds like it should be exclusive to that
#asahi-dev · 2023-04-18 10:30
marcan*t6001/j416
#asahi-dev · 2023-04-18 10:30
marcanextra pstates on t600x seem unlikely given they ran out of bits in the register
#asahi-dev · 2023-04-18 10:29
marcanmax we have on t8103 is 15/3104, t600x is 15/3228, t8112 17/3504 t6002/j416 19/3696
#asahi-dev · 2023-04-18 10:27
marcanI'm interested in what pstates the devicetree has
#asahi-dev · 2023-04-18 10:09
marcanalso, have we confirmed that the "extra" pstates on t802x are only for j416c? if so we need to adjust the DTs to that end.
#asahi-dev · 2023-04-18 10:03
marcanI expect the boost states to be woefully power-inefficient, so it really shouldn't switch to them until it really knows it needs to
#asahi-dev · 2023-04-18 09:53
marcanso I think it got added in 13.0
#asahi-dev · 2023-04-18 09:53
marcan(would be really nice to get newer keys...)
#asahi-dev · 2023-04-18 09:52
marcanand the firmware I'm reverse engineering is some early 13.0 beta (last one I have a key for on t8103)
#asahi-dev · 2023-04-18 09:52
marcanit is in 13.3 in j416
#asahi-dev · 2023-04-18 09:52
marcannot in 12.4 j314 SMC, no
#asahi-dev · 2023-04-18 09:51
marcanisn't there on my j314 with 12.4
#asahi-dev · 2023-04-18 09:50
marcanohh and this feature was introduced in some newer SMC firmware, 13.0 maybe

plus ›