Re: [Qemu-devel] [PATCH v5 04/14] audio: -audiodev command line option basic implementation

[Zoltán Kővágó]

On 2019-03-08 08:21, Markus Armbruster wrote:
> "Zoltán Kővágó" <address@hidden> writes:
> 
>> On 2019-03-07 16:56, Gerd Hoffmann wrote:
>>> On Tue, Feb 26, 2019 at 02:39:38AM +0100, Zoltán Kővágó wrote:
>>>> On 2019-02-20 22:37, Kővágó, Zoltán wrote:
>>>> [...]
>>>>> diff --git a/audio/audio.c b/audio/audio.c
>>>>> index ce8e6ea8c2..8ad8cbe559 100644
>>>>> --- a/audio/audio.c
>>>>> +++ b/audio/audio.c
>>>> [...]
>>>>> @@ -2129,3 +1866,170 @@ void AUD_set_volume_in (SWVoiceIn *sw, int mute, 
>>>>> uint8_t lvol, uint8_t rvol)
>>>>>          }
>>>>>      }
>>>>>  }
>>>>> +
>>>>> +void audio_create_pdos(Audiodev *dev)
>>>>> +{
>>>>> +    switch (dev->driver) {
>>>>> +#define CASE(DRIVER, driver, pdo_name)                              \
>>>>> +    case AUDIODEV_DRIVER_##DRIVER:                                  \
>>>>> +        dev->u.driver.in = g_malloc0(                               \
>>>>> +            sizeof(Audiodev##pdo_name##PerDirectionOptions));       \
>>>> This should check has_in before overwriting. It'll work correctly when
>>>> called from audio_legacy.c, but when using -audiodev it will overwrite
>>>> the options passed by user (and leak memory) when called from
>>>> audio_validate_opts. I'll fix it in the next update.
>>>
>>> Ping.  4.0 freeze is next tuesday.  Any chance for a v6 early enough
>>> that we have a chance to get the first chunk into 4.0?  Monday latest,
>>> preferably earlier ...
>>
>> I'll try to do something this weekend, but I can't promise anything. I
>> still haven't got to reading through Markus' comments...
> 
> Quoting myself: "We're down to minor stylistic issues.  Good work!"
> 
> Addressing these should not be hard.
> 

I blame it on my horrible time management.  Also the recent sdl/audio
patches generated some conflicts, I had to solve them first.  And magic
errors like "'vfio-pci' is not a valid device model name" after update.

Regards,
Zoltan