Re: [BUG]i2c_hid_acpi broken with 4.17.2 on Framework Laptop 13 AMD

[Sébastien Chaumat <euidzero@xxxxxxxxx>]

Le mer. 6 mars 2024 à 19:08, Mario Limonciello <mario.limonciello@xxxxxxx> a écrit :

On 3/6/2024 12:05, Sébastien Chaumat wrote:
>
>
> Le mer. 6 mars 2024 à 18:33, Mario Limonciello
> <mario.limonciello@xxxxxxx <mailto:mario.limonciello@xxxxxxx>> a écrit :
>
>     On 3/6/2024 11:28, Sébastien Chaumat wrote:
>      >
>      >
>      >
>      >
>      >     Reasoning backward  (using a  kernel without the pinctrl_amd
>     driver
>      >     to ensure xen only is at stake) :
>      >       checking the diff in IOAPIC  between bare metal and xen
>     (IRQ7 is
>      >     on pin07 on APIC )
>      >
>      >     using kernel argument : apic=debug
>      >
>      >     bare metal :
>      >     [    0.715330] fedora kernel: ... APIC VERSION: 81050010
>      >     ...
>      >     [    0.715433] fedora kernel:  pin07, disabled, edge , high,
>     V(00),
>      >     IRR(0), S(0), physical, D(0000), M(0)
>      >
>      >     xen :
>      >     [    2.249582] fedora kernel: ... APIC VERSION: 00000014
>      >     ...
>      >     [    2.249730] fedora kernel:  pin07, disabled, level, low ,
>     V(60),
>      >     IRR(0), S(0), physical, D(0000), M(0)
>      >
>      >     So the APIC table is not the same.
>      >
>      >     As strange as it looks the  (IOAPIC 0) pin07 is correctly
>     described
>      >     by the APIC in xen but yet differently than in baremetal.
>      >     But the APIC message comes long after the
>      >     [    1.833145] fedora kernel: xen: registering gsi 7 triggering 0
>      >     polarity 1
>      >
>      >     so I wonder if the APIC pin07 info had any influence.
>      >
>      > Finally found the fix : adding ioapic_ack=new to xen boot parameters.
>      > Not only the trackpad is now working but also the ACPI Embedded
>      > Controller which is completely disabled.
>      >
>      > Sébastien
>      >
>     That's great news!  I'm personally totally unfamiliar with
>     ioapic_ack=new, so I did a quick search and found out it's a Xen
>     parameter (I came across
>     https://xenbits.xen.org/docs/4.5-testing/misc/xen-command-line.html
>     <https://xenbits.xen.org/docs/4.5-testing/misc/xen-command-line.html>).
>
>     This mentions that "new" should be the default, so why isn't it the
>     case?
>
>
>   "This is the the default unless directed-EOI is supported"
> xl dmesg without forcing the parameters shows :
>
> (XEN) Enabled directed EOI with ioapic_ack_old on!

Got it; it sounds to me like a logic mismatch in Xen then.

>
>     Also; I'd be really interested to hear what happens with s2idle with
>     Xen
>     now (if it works).
>
>
> suspend to RAM now works :)

Do you see /sys/power/suspend_stats/last_hw_sleep increasing with your
suspend cycle?

No,  it does not increase (0).