Re: [PATCH v2] audio: Free consumed default audio devices

[Akihiko Odaki <akihiko.odaki@daynix.com>]

On 2023/11/20 20:35, Marc-André Lureau wrote:
> Hi
> 
> On Mon, Nov 20, 2023 at 3:29 PM Akihiko Odaki <akihiko.odaki@daynix.com> wrote:
>>
>> Failed default audio devices were removed from the list but not freed,
>> and that made LeakSanitizer sad. Free default audio devices as they are
>> consumed.
>>
>> Signed-off-by: Akihiko Odaki <akihiko.odaki@daynix.com>
> 
> Acked-by: Marc-André Lureau <marcandre.lureau@redhat.com>
> 
>> ---
>> V1 -> V2: Set s->dev NULL after free (Marc-André Lureau)
>>
>>   audio/audio.c | 5 ++++-
>>   1 file changed, 4 insertions(+), 1 deletion(-)
>>
>> diff --git a/audio/audio.c b/audio/audio.c
>> index f91e05b72c..8d1e4ad922 100644
>> --- a/audio/audio.c
>> +++ b/audio/audio.c
>> @@ -1758,12 +1758,15 @@ static AudioState *audio_init(Audiodev *dev, Error **errp)
>>                   goto out;
>>               }
>>               s->dev = dev = e->dev;
>> +            QSIMPLEQ_REMOVE_HEAD(&default_audiodevs, next);
>> +            g_free(e);
>>               drvname = AudiodevDriver_str(dev->driver);
>>               driver = audio_driver_lookup(drvname);
>>               if (!audio_driver_init(s, driver, dev, NULL)) {
> 
> Maybe it would be clearer/safer to set s->dev in audio_driver_init() on success.

I considered that too, but I decided not to that because 
audio_driver_init() takes s and may expect s->dev is initialized.