Re: [Regression] Changes to i2c-piix4.c initialisation prevent loading of sp5100_tco watchdog driver on AMD SB800 chipset

[Guenter Roeck <linux@roeck-us.net>]

Hi Paul,

On 03/31/2017 12:17 AM, Paul Menzel wrote:
> Dear Wolfram,
>
>
> Thank you for the reply, which we talked about briefly at the
> Chemnitzer LinuxTage.
>
>
> Am Freitag, den 03.03.2017, 11:17 +0100 schrieb Wolfram Sang:
>>> Unfortunately, commit 2fee61d22e (i2c: piix4: Add support for
>>> multiplexed main adapter in SB800) [1] caused a regression. Tim
>>> reported that to the Linux Kernel Bugtracker as bug #170741 last
>>> September [2], but it looks like the affected subsystems don’t use it.
>>
>> Jean Delvare pointed out this issue amongst others[1] last year already.
>> Let me quote:
>>
>> ===
>>
>> 5* The I/O ports used for SMBus configuration and port switching are
>> also needed by a watchdog driver, sp5100_tco. Both drivers request the
>> region, so the first one wins, and the other driver can't be loaded.
>> sp5100_tco was there first, so the changes done to the i2c-piix4 driver
>> recently will cause a regression for some users by preventing them
>> from using the sp5100_tco and i2c-piix4 drivers at the same time. In
>> the long run I guess we will need a helper module to handle this shared
>> resource. Unless IORESOURCE_MUXED can be used for that. Either way,
>> that's more work than I can put into this before kernel v4.5 is
>> released. For the time being, I think we should simply make it
>> non-fatal if the I/O ports can't be requested, and continue without
>> multiplexing (as before.)
>>
>> ===
>>
>> Seems nobody had the resources, so far.
>
> I still don’t understand, why Jean then not immediately reverted the
> commit to adhere to the Linux Kernel’s no-regression-policy.
>
>> I don't have the HW and not much experience with non-embedded
>> platforms. I wonder, though, if we really need to convert the drivers
>> to MFD ones, or if we could use the simpler MFD_SYSCON mechanism
>> which helps in exactly such cases for embedded platforms. But I am
>> really lacking details here and am afraid this is probably all the
>> input I can give currently.
>
> Zoltan stepped up, and uploaded a patch for review to the Kernel.org
> Bugzilla [2], also attached to this message.
>

Please don't send patches as attachments.

request_muxed_region() can fail, and literally every other driver
using it checks for that failure. Please do the same.

The sp5100_tco_dev_name change in the watchdog driver is unnecessary.
There are some unnecessary { } in the watchdog driver after the patch
is applied.

Please split the patch into two patches so they can be reviewed and
applied separately.

Thanks,
Guenter