Re: Wrong piix4_smbus address / slow trackpoint on Thinkpad P14s gen 2 (AMD)

From: Andrea Ippolito <andrea.ippo@gmail.com>
To: "Miroslav Bendík" <miroslav.bendik@gmail.com>
Cc: linux-i2c@vger.kernel.org, linux-input@vger.kernel.org
Subject: Re: Wrong piix4_smbus address / slow trackpoint on Thinkpad P14s gen 2 (AMD)
Date: Thu, 6 Jan 2022 19:27:05 +0100	[thread overview]
Message-ID: <CAGhUXvCMyVD3aP7Mi+i2hz0A0FePJvPSceubXm=u+4AR+wgesA@mail.gmail.com> (raw)
In-Reply-To: <f8c13907-d296-baa6-7637-c5f8aa96b7ff@gmail.com>

Hello Miroslav,

forgive me if it's a dumb question, but would this also apply to Intel
platforms, or this kind of stuff chip maker specific?

I got an Intel Tiger Lake.

Thanks!
Andrea IPPOLITO

Il giorno gio 6 gen 2022 alle ore 16:03 Miroslav Bendík
<miroslav.bendik@gmail.com> ha scritto:
>
> Address detection does not work because cd6h/cd7h port io can be
> disabled, but
> it's accessible using mmio. This patch:
> https://lore.kernel.org/all/20210715221828.244536-1-Terry.Bowman@amd.com/
> with
> modified AMD_PCI_SMBUS_REVISION_MMIO fixed base address detection.
>
> Problem with RMI4 touchpad / trackpoint remains, because rmi4-smbus
> needs host
> notify feature. I have tried implement this feature, without success.
> Interrupts
> on IRQ 7 (SMBus) are generated only for block data transfers, but not for
> trackpoint / touchpad move actions. I have looked at pinctrl_amd and it
> looks,
> that activity is signaled using GPIO. This looks promising:
>
> cat /sys/kernel/debug/pinctrl/AMDI0030:00/pingroups:
>
> group: i2c0
> pin 145 (GPIO_145)
> pin 146 (GPIO_146)
>
> group: i2c1
> pin 147 (GPIO_147)
> pin 148 (GPIO_148)
>
> group: i2c2
> pin 113 (GPIO_113)
> pin 114 (GPIO_114)
>
> group: i2c3
> pin 19 (GPIO_19)
> pin 20 (GPIO_20)
>
> After loading psmouse with forced enabled intertouch pin 19/20 started
> generating events.
>