#
✕ nick:nickchan
#asahi-dev · 2025-05-17 16:37
nickchansomething tells me that upstream wouldn't exactly like "apple,t8103-atcphy-capcode-workaround" but not sure
#asahi-dev · 2025-05-17 16:29
nickchani think "usb: typec: tipd: WIP" could use some work on the commit message
#asahi-dev · 2025-05-17 10:02
nickchanwill submit spmi nodes for t2 and a11 later too those also appears to be compatible
#asahi-dev · 2025-05-17 09:56
nickchanand said submission would only have a7-a11 not t2 since there's issues accessing the i2c and the associated pmgr ps regs on t2
#asahi-dev · 2025-05-17 09:54
nickchanright now i have the i2c3 in the dts without the pinctrl properties of course
#asahi-dev · 2025-05-17 09:51
nickchanmaybe i should remove that i2c3 entirely from the dts file before i submit
#asahi-dev · 2025-05-17 09:51
nickchanand nothing is connected to that i2c3
#asahi-dev · 2025-05-10 15:12
nickchanit also only happens in gcc
#asahi-dev · 2025-05-10 14:59
nickchaneven though the original driver already has like 10 W=1 warnings already
#asahi-dev · 2025-05-10 14:58
nickchani will send the pmu support series again i guess to fix that one W=1 compile warning that syzbot complained about
#asahi-dev · 2025-05-10 14:57
nickchanand the deep sleep serror issues above is only a11 since a10 and below only has one cluster and obviously cpu0 is not in sleep
#asahi-dev · 2025-05-10 14:55
nickchansven: they were not cleared before
#asahi-dev · 2025-05-10 14:54
nickchansven: clear the wfi retention disable bit on startup on cores that supports arch retention and it should be good
#asahi-dev · 2025-05-10 14:54
nickchansven: In any case at least the chickens PR does not have any *new* regressions on A7, A9, A10, A11
#asahi-dev · 2025-05-10 14:40
nickchani will make sure it works for booting and secondary start i guess
#asahi-dev · 2025-05-10 14:40
nickchanuntil i want to bother with idle in linux of course
#asahi-dev · 2025-05-10 14:40
nickchanugh in any case i don't think i really care about what happens on that code path
#asahi-dev · 2025-05-10 14:36
nickchana non deep sleep but i am quite sure that it was okay at one point
#asahi-dev · 2025-05-10 14:36
nickchanokay sleeping also doesn't work without that pr right now
#asahi-dev · 2025-05-10 14:29
nickchanhttps://github.com/apple-oss-distributions/xnu/blob/bb611c8fecc755a0d8e56e2fa51513527c5b7a0e/osfmk/arm64/cpu.c#L999
#asahi-dev · 2025-05-10 14:29
nickchansven: the xnu sources stated that the global sleep register must not be modified when there are CPUs in wfi
#asahi-dev · 2025-05-10 14:01
nickchanand there's some workarounds in those sources
#asahi-dev · 2025-05-10 13:57
nickchanuh how about mindlessly changing the sequence to match xnu sources' sleep
#asahi-dev · 2025-05-10 13:42
nickchana11 is HAS_CLUSTER false in xnu
#asahi-dev · 2025-05-10 13:34
nickchanokay so the register accesses are getting killed in what's supposedly a not-deep-sleep
#asahi-dev · 2025-05-10 13:28
nickchanlooks like "Turning off the last core in a cluster kills our register accesses"
#asahi-dev · 2025-05-10 13:27
nickchanTTY> L2C_ERR_ADR: 0x110000208150100
#asahi-dev · 2025-05-10 13:25
nickchanOk the chickens pr is absolutely not fine on A11
#asahi-dev · 2025-05-10 11:35
nickchanwow that must be talking about T2
#asahi-dev · 2025-05-10 11:35
nickchanhttps://github.com/AsahiLinux/linux/blob/8dc6d00309e5feeca4f3f802976c0b7c9c406a49/drivers/nvme/host/pci.c#L2720
#asahi-dev · 2025-05-10 11:33
nickchanNVME_QUIRK_128_BYTES_SQES
#asahi-dev · 2025-05-10 11:17
nickchansven: after cleaning up I see it is just the other mode which is already mentioned in the driver https://github.com/AsahiLinux/linux/blob/8dc6d00309e5feeca4f3f802976c0b7c9c406a49/drivers/nvme/host/apple.c#L1110
#asahi-dev · 2025-05-10 11:17
nickchanokay cleaned up a11's nvme code
#asahi-dev · 2025-05-10 08:09
nickchancurrently cleaning up some apple nvme code after that i should check out the m1n1 chicken bits pr
#asahi-dev · 2025-05-07 13:41
nickchanit probably doesn't like genmask inside a function prototype or something
#asahi-dev · 2025-05-07 13:39
nickchanokay it is causing the warning but it's a bit complicated
#asahi-dev · 2025-05-05 09:22
nickchanuh i don't think i could review the m1n1 chickens changes today either i will review later
#asahi-dev · 2025-05-04 18:16
nickchansven: i do think the CYC_OVRD_DISABLE_WFI_RET bit should be cleared on startup on cores that supports arch state retention
#asahi-dev · 2025-05-04 18:14
nickchantalking about midr comparisons
#asahi-dev · 2025-05-04 18:13
nickchanalso do
#asahi-dev · 2025-05-04 18:13
nickchanwhich this refractor als odo
#asahi-dev · 2025-05-04 18:13
nickchanthink it's a good idea to open code everything too since there's always some Apple Watch cores or ASC cores in the middle breaking the flfow
#asahi-dev · 2025-05-04 18:09
nickchanand that horrdenously *broken* given that it's an arm64 core fabbed in 2012
#asahi-dev · 2025-05-04 18:09
nickchanat least no one should run m1n1 on Alcatraz (A7) A0
#asahi-dev · 2025-05-04 18:08
nickchansven: and a7-a11 cores definitely has more rev specific chickens it is just quite unlikely the chickens for previous revs will ever be used
#asahi-dev · 2025-05-04 18:03
nickchani will review chickens tomorrow
#asahi-dev · 2025-05-04 17:24
nickchanerm I mean mailboxes for the cortex-a7 cores
#asahi-dev · 2025-05-04 17:24
nickchancurrently calling the mailboxes "apple,t8015-marin-mailbox", "apple,marin-mailbox" since the mbox for the kingfisher cores is most definitely different
#asahi-dev · 2025-05-04 17:20
nickchanand see which bits can be set
#asahi-dev · 2025-05-04 17:20
nickchansart is something that could be figured out just by reading out what iboot configured the register as and then try writing 0xffffffff to it

plus ›