[libvirt] [PULL 04/14] audio: -audiodev command line option basic implementation

[Zoltán Kővágó]

On 2019-03-14 10:46, Peter Maydell wrote:
> On Tue, 12 Mar 2019 at 07:13, Gerd Hoffmann <kraxel at redhat.com> wrote:
>>
>> From: Kővágó, Zoltán <dirty.ice.hu at gmail.com>
>>
>> Audio drivers now get an Audiodev * as config paramters, instead of the
>> global audio_option structs.  There is some code in audio/audio_legacy.c
>> that converts the old environment variables to audiodev options (this
>> way backends do not have to worry about legacy options).  It also
>> contains a replacement of -audio-help, which prints out the equivalent
>> -audiodev based config of the currently specified environment variables.
> 
> Hi; Coverity complains (CID 1399706) about this, which isn't
> a change in this patch as such, but the code change has
> probably caused it to reanalyze:
> 
>>
>>      if (!done) {
>>          driver = audio_driver_lookup("none");
>> -        done = !audio_driver_init(s, driver, false);
>> +        done = !audio_driver_init(s, driver, false, dev);
> 
> Everywhere else we call audio_driver_lookup() we check
> whether the return value is NULL before using it,
> but here we don't. I guess this is a false positive
> because the "none" driver must always exist ?
> If so, I can just silence the warning in the coverity UI.

Yes, "none" (implemented in noaudio.c) is currently unconditionally
compiled in along with "wav".  "none" is used as a fallback when nothing
else works, so I think it's a bug somewhere else if it doesn't exist.

PS. I think I managed to break something, Thunderbird complained that
non-ascii characters in email addresses are not supported.  Somehow
Kővágó@redhat.com ended up on the recipient list.

Regards,
Zoltan