Re: [PATCH 1/1] at24: Fix I²C device selection for runtime PM

[Bartosz Golaszewski <brgl@bgdev.pl>]

2017-12-01 16:35 GMT+01:00 Sakari Ailus <sakari.ailus@linux.intel.com>:
> Hi Sven,
>
> On Fri, Dec 01, 2017 at 10:20:41AM -0500, Sven Van Asbroeck wrote:
>> Thank you, it fixes the issue on the multi-address eeprom that I have access to.
>>
>> Tested-by: Sven Van Asbroeck on a 24AA16/24LC16B <svendev@arcx.com>
>>
>> One very minor remark:
>>
>> +       struct device *dev = &at24->client[0]->dev;
>>
>> It is sufficiently clear to others (and us a few months down the line)
>> why we are
>> using only client[0] for power management? Could it benefit from a separate
>> function with comments?
>>
>> struct device *dev = get_pm_device(at24);
>>
>> static struct device *get_pm_device(struct at24_data *at24)
>> {
>>     /* explain why we use client[0] and not any of the dummies */
>>     return &at24->client[0]->dev;
>> }
>
> There are no comments in assigning at24->client[0] either (or a helper
> function). I think it should be rather evident when looking at the code
> when you think about it. I certainly don't object adding a comment if you
> insist or someone else thinks it's a good idea.
>
> Thanks for testing!
>
> --
> Kind regards,
>
> Sakari Ailus
> sakari.ailus@linux.intel.com

Pushed to at24/fixes, thanks!

@Saraki: there were some conflicts with the previous fixes queued for
4.15. Could you take a look if my rebase didn't break anything? You
can find my tree at
git://git.kernel.org/pub/scm/linux/kernel/git/brgl/linux.git

Best regards,
Bartosz