#
✕ nick:yuka
#asahi-dev · 2026-07-08 07:47
yuka27.0b3 on t6050: SYS_IMP_APL_VM_TMR_FIQ_ENA_EL2 is unlocked, smp is broken
#asahi-dev · 2026-07-07 23:38
yuka(This is without pmp)
#asahi-dev · 2026-07-07 23:37
yukaThe culprit was dockchannel_tty
#asahi-dev · 2026-07-07 23:36
yukaSame ballpark as cpuidle-apple
#asahi-dev · 2026-07-07 23:36
yuka5.35W
#asahi-dev · 2026-07-07 23:35
yuka*now
#asahi-dev · 2026-07-07 23:34
yukaNot I can test efi psci again
#asahi-dev · 2026-07-07 23:34
yukaFound the power usage culprit
#asahi-dev · 2026-07-07 23:26
yukafair
#asahi-dev · 2026-07-07 23:25
yukaYea
#asahi-dev · 2026-07-07 23:24
yukaI could give it a try if I don't have to debug too much, since I don't actually want to read the code
#asahi-dev · 2026-07-07 23:04
yukaAlso, I think the impact of getting your hv into an upstreamable state would be much larger than frickling some drivers for the Neo
#asahi-dev · 2026-07-07 23:02
yukaDidn't see it mentioned
#asahi-dev · 2026-07-07 23:02
yukaAlso add nvme to your list
#asahi-dev · 2026-07-07 23:01
yukanot the linux one
#asahi-dev · 2026-07-07 22:58
yukapcie is not unchanged but the changes are in my m1n1 PR
#asahi-dev · 2026-07-07 22:55
yukaDockchannel-hid needs changes because there is no stm
#asahi-dev · 2026-07-07 22:54
yukaThen, pinctrl needs a new compatible
#asahi-dev · 2026-07-07 22:54
yukaIf you could figure out the smp boot instability that would also be great
#asahi-dev · 2026-07-07 21:03
yukaok yeah the power draw is just as bad with cpuidle-apple, so it's something else in my kernel
#asahi-dev · 2026-07-07 20:51
yukaso there might be other fixes missing
#asahi-dev · 2026-07-07 20:51
yukabut that's with a mostly-mainline kernel
#asahi-dev · 2026-07-07 20:51
yuka12-14W on the M3 Max 14-core
#asahi-dev · 2026-07-07 20:51
yukamy power draw is still unexpectedly high, though I still have to compare that against cpuidle-apple (with everything else the same)
#asahi-dev · 2026-07-07 20:40
yukapowertop now shows me cpu-deep-wfi as 70-80%
#asahi-dev · 2026-07-07 20:40
yukaI definitely had to use more optimistic latencies to get it to use deep wfi at all
#asahi-dev · 2026-07-07 20:39
yukaplaying with efi psci now
#asahi-dev · 2026-07-07 16:58
yukanice
#asahi-dev · 2026-07-07 16:58
yukasdconn
#asahi-dev · 2026-07-07 16:58
yukaah
#asahi-dev · 2026-07-07 16:58
yukaAh, my device doesn't have a bit14
#asahi-dev · 2026-07-07 16:57
yukamy adt has power-gate-with-media = 1
#asahi-dev · 2026-07-07 16:53
yukaThere are tales about a gpio which can detect the presence of an sd card. Does anyone know which gpio this is or how to find it?
#asahi-dev · 2026-07-07 10:56
yukait would be a very cursed setup either way
#asahi-dev · 2026-07-07 10:28
yukabut since the thing does not even boot reliably, I would not consider it useful for any real workload
#asahi-dev · 2026-07-07 10:27
yukausb miiiight work with usb 2.0 only and no hotplugging, I haven't tried it yet
#asahi-dev · 2026-07-07 10:04
yukaYup
#asahi-dev · 2026-07-07 09:34
yukaDon't expect to get it into a useful/usable state except as development target
#asahi-dev · 2026-07-07 09:34
yukadepends on what you mean by the basics
#asahi-dev · 2026-07-07 09:12
yuka27.0 beta 3 is out, I will test later on t8132/t8140/t6050
#asahi-dev · 2026-07-07 07:41
yukaBut I was wrong to think it was the main/only issue
#asahi-dev · 2026-07-07 07:41
yukaand I think I ran into the mmu issue a few times while using the switch boot cpu code
#asahi-dev · 2026-07-07 07:41
yukaSwitching the boot cpu does involve initializing the mmu on the secondary core
#asahi-dev · 2026-07-07 07:33
yukaI will test drive it a bit
#asahi-dev · 2026-07-07 07:32
yukaOr was that unreliable for other reasons?
#asahi-dev · 2026-07-07 07:31
yukaThen we can maybe revisit switching boot cpu
#asahi-dev · 2026-07-07 07:08
yukaoops
#asahi-dev · 2026-07-07 06:32
yukabut if we had code to tear down a port and bring it back up again, I would expect to be able to unbind and bind the driver and have it repeatedly probe successfully
#asahi-dev · 2026-07-07 06:32
yukathe port setup being in init() doesn't cause any problems in practise, because we don't implement tearing down and bringing back up the pcie ports
#asahi-dev · 2026-07-07 06:30
yukaotherwise my pwrctrl code wouldn't work

plus ›