NAK: [Utopic PATCH 0/2] fix pop and white noise problem for dell xps13 2015 model

[Hui Wang]

On 07/17/2015 05:53 PM, Luis Henriques wrote:
> On Thu, Jul 16, 2015 at 03:02:38PM +0800, Hui Wang wrote:
>> On 07/16/2015 02:34 PM, Stefan Bader wrote:
>>> On 16.07.2015 08:29, Stefan Bader wrote:
>>>> On 16.07.2015 03:31, Hui Wang wrote:
>>>>> In the upstream kernel and vivid kernel, these two patches introduce a
>>>>> regression, after applied these two patches, the headset jack will not
>>>>> detect plug and unplug anymore. So we need one more patch (831bfdf9)
>>>>> to fix this regression in the upstream kernel or in the vivid kernel.
>>>>>
>>>>> But in the utopic kernel, these two patches will not introduce the
>>>>> regression since headset jack detection fixing patch has not been
>>>>> in the utopic yet and it is not easy to backport headset jack
>>>>> detection fixing patches to utopic since the patches depend on
>>>>> the macro UPDATE_COEF(), to introduce UPDATE_COEF() in the utopic
>>>>> kernel, the patch_realtek.c will be greatly changed.
>>>>>
>>>>> As a result, I just send these two patches to fix noise problem.
>>>>>
>>>>> Also please refer to upstream bug for these 2 patches:
>>>>> Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=99851
>>>>>
>>>>> Takashi Iwai (2):
>>>>>    ALSA: hda - Fix audio crackles on Dell Latitude E7x40
>>>>>    ALSA: hda - Fix noisy outputs on Dell XPS13 (2015 model)
>>>> This (and the Trusty series, too) seems to miss
>>>>
>>>> ALSA: hda - restore the MIC FIXUP for some Dell machines
>>>>
>>>> which you included in the Vivid series.
>> Because that patch is for fixing a regression, it is really a regression for
>> vivid since vivid merged these two patches to fix the headset problem:
>> f3b703326541d0c1ce85f5e570f6d2b6bd4296ec
>> e1e62b98ebddc3234f3259019d3236f66fc667f8
>>
>>
>> But for trusty and utopic, they have not merged above two commits, so there
>> is no regression for trusty and utopic after applied the patches I sent out,
>> as a result, we don't need to merge this patch "ALSA: hda - restore the MIC
>> FIXUP for some Dell machines" .
>>> Hm, I realized that you heavily modified the first patch. Personally I am not
>>> really happy with that. Because now there is a patch for the Latitude which in
>>> no way does do anything for that model.
>> Yes, I can understand it.  The reason i introduced this patch is the machine
>> xps 13 also needs this method to fix noise problem.
> I understand this and it all makes sense... but I also understand
> Stefan's point: it's really confusing to have a patch with subject:
>
>   ALSA: hda - Fix audio crackles on Dell Latitude E7x40
>
> where most of the bits related to the Dell Latitude are actually
> dropped.  And there are some questions, such as: do we really need
> 'ALC292_FIXUP_DISABLE_AAMIX'  in the alc269_fixups?  If not, why don't
> we drop it since we're already modifying this patch?
'ALC292_FIXUP_DISABLE_AAMIX' is necessary to fix this problem. The patch 
added two FIXUP macros, the first one 'ALC292_FIXUP_DELL_E7X' is for 
fixing the pop noise, the second one 'ALC292_FIXUP_DISABLE_AAMIX' is for 
fixing the white noise, and the second one is chained to the first one 
in the quirk table, the machine xps13 will apply DELL_E7X first, then 
will apply DISABLE_AAMIX.
> Anyway, after a chat with Stefan, it's clear that this patch will
> confuse everyone (it already did :-) ).  Two possible solutions:
>
> 1) Include the changes in the first patch in a 'UBUNTU: SAUCE: (no-up)...'
>     patch, that would include *only* the required changes from
>     8b99aba70c5f ("ALSA: hda - Fix audio crackles on Dell Latitude E7x40")
>     and that would document exactly what it is doing, where did it came
>     from (the origin SHA1), and why it is needed.
>
> 2) Merge only the required changes from the 1st patch into the 2nd
>     patch, documenting those changes (and it's provenience).  This
>     solution is particularly appealing if the
>     'ALC292_FIXUP_DISABLE_AAMIX' changes aren't really necessary.
I will prepare a new patch according to solution 2, the #2 looks less 
confusing. :-)

Thanks for your review,  I will send the V2 soon.

Regards,
Hui.
>
> Cheers,
> --
> Luís
>