Received-SPF: pass (google.com: domain of linux-kernel+bounces-11806-linux.lists.archive=gmail.com@vger.kernel.org designates 2604:1380:45e3:2400::1 as permitted sender) client-ip=2604:1380:45e3:2400::1;
Precedence: bulk
MIME-Version: 1.0
References: <20231220051350.392350-1-kai.heng.feng@canonical.com>
 <5a81eda0-5044-4088-9aba-cf725e6fca9a@roeck-us.net> <CAAd53p4NNX9c5pUJgv12V=s_1YJxM3=G+OCYTgcRPqSJVbdi8w@mail.gmail.com>
 <3d777aee-9ec1-4236-98d2-f56b8026b96a@gmx.de>
In-Reply-To: <3d777aee-9ec1-4236-98d2-f56b8026b96a@gmx.de>
From: Kai-Heng Feng <kai.heng.feng@canonical.com>
Date: Wed, 27 Dec 2023 11:39:46 +0800
Message-ID: <CAAd53p6POFCNEP4J8U_jTexBkxOn-15ULRqQj+r2GA3q2+1raQ@mail.gmail.com>
Subject: Re: [PATCH] hwmon: (acpi_power_meter) Install IPMI handler for Dell systems
To: Armin Wolf <W_Armin@gmx.de>
Cc: Guenter Roeck <linux@roeck-us.net>, jdelvare@suse.com, linux-hwmon@vger.kernel.org, 
	linux-kernel@vger.kernel.org
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Hi Armin,

On Fri, Dec 22, 2023 at 8:08=E2=80=AFPM Armin Wolf <W_Armin@gmx.de> wrote:
>
> Am 22.12.23 um 04:57 schrieb Kai-Heng Feng:
>
> > Hi Guenter,
> >
> > On Thu, Dec 21, 2023 at 7:49=E2=80=AFAM Guenter Roeck <linux@roeck-us.n=
et> wrote:
> >> On Wed, Dec 20, 2023 at 01:13:50PM +0800, Kai-Heng Feng wrote:
> >>> The following error can be observed at boot:
> >>> [    3.717920] ACPI Error: No handler for Region [SYSI] (00000000ab9e=
62c5) [IPMI] (20230628/evregion-130)
> >>> [    3.717928] ACPI Error: Region IPMI (ID=3D7) has no handler (20230=
628/exfldio-261)
> >>>
> >>> [    3.717936] No Local Variables are initialized for Method [_GHL]
> >>>
> >>> [    3.717938] No Arguments are initialized for method [_GHL]
> >>>
> >>> [    3.717940] ACPI Error: Aborting method \_SB.PMI0._GHL due to prev=
ious error (AE_NOT_EXIST) (20230628/psparse-529)
> >>> [    3.717949] ACPI Error: Aborting method \_SB.PMI0._PMC due to prev=
ious error (AE_NOT_EXIST) (20230628/psparse-529)
> >>> [    3.717957] ACPI: \_SB_.PMI0: _PMC evaluation failed: AE_NOT_EXIST
> >>>
> >>> On Dell systems several methods of acpi_power_meter access variables =
in
> >>> IPMI region [0], so request module 'ipmi_si' which will load 'acpi_ip=
mi'
> >>> and install the region handler accordingly.
> >>>
> >>> [0] https://www.dell.com/support/manuals/en-us/redhat-enterprise-linu=
x-v8.0/rhel8_rn_pub/advanced-configuration-and-power-interface-acpi-error-m=
essages-displayed-in-dmesg?guid=3Dguid-0d5ae482-1977-42cf-b417-3ed5c3f5ee62
> >>>
> >>> Signed-off-by: Kai-Heng Feng <kai.heng.feng@canonical.com>
> >>> ---
> >>>   drivers/hwmon/acpi_power_meter.c | 2 ++
> >>>   1 file changed, 2 insertions(+)
> >>>
> >>> diff --git a/drivers/hwmon/acpi_power_meter.c b/drivers/hwmon/acpi_po=
wer_meter.c
> >>> index 703666b95bf4..b9db53166bc9 100644
> >>> --- a/drivers/hwmon/acpi_power_meter.c
> >>> +++ b/drivers/hwmon/acpi_power_meter.c
> >>> @@ -882,6 +882,8 @@ static int acpi_power_meter_add(struct acpi_devic=
e *device)
> >>>        strcpy(acpi_device_name(device), ACPI_POWER_METER_DEVICE_NAME)=
;
> >>>        strcpy(acpi_device_class(device), ACPI_POWER_METER_CLASS);
> >>>        device->driver_data =3D resource;
> >>> +     if (dmi_match(DMI_SYS_VENDOR, "Dell Inc."))
> >>> +             request_module("ipmi_si");
> >>>
> >> This looks like a terrible hack to me. Is there precedent of similar h=
acks
> >> elsewhere showing that this is the "way to go" ?
> > Yes it's ugly.
> >
> > The error happens in the ACPI ASL code, so it's not possible to know
> > if any method of apci_power_meter requires IPMI region.
> > I really can't think of any better solution for it.
> >
> > Kai-Heng
>
> Maybe we could use an ACPI scan handler for that?
> Basically we would call request_module() upon discovering an ACPI IPMI de=
vice,
> so that the necessary module is available for handling the IPMI opregion.
> This would also prevent any issues should other devices also require IPMI=
 access.

I think the ACPI core already does this. But since the modules are
loading and probing asynchronously, the order needs to be enforced.
I'll send another revision that doesn't use request_module().

Kai-Heng

>
> Armin Wolf
>
> >> Guenter
> >>
> >>>        res =3D read_capabilities(resource);
> >>>        if (res)
> >>> --
> >>> 2.34.1
> >>>