Re: [PATCH] em28xx: fix usb alternate setting for analog and digital video endpoints > 0

["Frank Schäfer" <fschaefer.oss@googlemail.com>]

Am 06.02.2013 17:03, schrieb Mauro Carvalho Chehab:
> Em Wed, 06 Feb 2013 16:31:41 +0100
> Frank Schäfer <fschaefer.oss@googlemail.com> escreveu:
>
>> Am 05.02.2013 23:06, schrieb Mauro Carvalho Chehab:
>>> Em Tue, 05 Feb 2013 22:37:50 +0100
>>> Frank Schäfer <fschaefer.oss@googlemail.com> escreveu:
>>>
>>>> Am 05.02.2013 21:57, schrieb Mauro Carvalho Chehab:
>>>>> Em Fri, 18 Jan 2013 18:25:48 +0100
>>>>> Frank Schäfer <fschaefer.oss@googlemail.com> escreveu:
>>>>>
>>>>>> While the current code handles sound interfaces with a number > 0 correctly, it
>>>>>> assumes that the interface number for analog + digital video is always 0 when
>>>>>> changing the alternate setting.
>>>>>>
>>>>>> (NOTE: the "SpeedLink VAD Laplace webcam" (EM2765) uses interface number 3 for video)
>>>>>>
>>>>>> Signed-off-by: Frank Schäfer <fschaefer.oss@googlemail.com>
>>>>>> ---
>>>>>>  drivers/media/usb/em28xx/em28xx-audio.c |   10 +++++-----
>>>>>>  drivers/media/usb/em28xx/em28xx-cards.c |    2 +-
>>>>>>  drivers/media/usb/em28xx/em28xx-core.c  |    2 +-
>>>>>>  drivers/media/usb/em28xx/em28xx-dvb.c   |    2 +-
>>>>>>  drivers/media/usb/em28xx/em28xx.h       |    3 +--
>>>>>>  5 Dateien geändert, 9 Zeilen hinzugefügt(+), 10 Zeilen entfernt(-)
>>>>>>
>>>>>> diff --git a/drivers/media/usb/em28xx/em28xx-audio.c b/drivers/media/usb/em28xx/em28xx-audio.c
>>>>>> index 2fdb66e..cdbfe0a 100644
>>>>>> --- a/drivers/media/usb/em28xx/em28xx-audio.c
>>>>>> +++ b/drivers/media/usb/em28xx/em28xx-audio.c
>>>>>> @@ -283,15 +283,15 @@ static int snd_em28xx_capture_open(struct snd_pcm_substream *substream)
>>>>>>  	}
>>>>>>  
>>>>>>  	runtime->hw = snd_em28xx_hw_capture;
>>>>>> -	if ((dev->alt == 0 || dev->audio_ifnum) && dev->adev.users == 0) {
>>>>>> -		if (dev->audio_ifnum)
>>>>>> +	if ((dev->alt == 0 || dev->ifnum) && dev->adev.users == 0) {
>>>>>> +		if (dev->ifnum)
>>>>> Please don't merge a non-fix change (variable rename) with a fix.
>>>> Ok, sorry, it seems to be trivial...
>>>>
>>>>> Btw, audio_ifnum is a better name, as it avoids it to be miss-interpreted.
>>>> Did you read the complete patch ? ;)
>>>> Or do you really want the video interface number to be called audio_ifnum ?
>>> There are two types of em28xx audio vendor class. In one type, the audio IF
>>> is the same as the video one, but on the other one, it is different.
>> Sure, but if I'm not misunderstanding the code, we have two device
>> instances with separate device structs when audio is on a separate
>> interface.
>>
>> Hence we don't need two fields for the interface number in the struct
>> and that's why renamed it.
> No. The way the extension mechanism was written, the same data instance
> is used by all modules. This is what is there at em28xx core:
>
> void em28xx_init_extension(struct em28xx *dev)
> {
> 	const struct em28xx_ops *ops = NULL;
>
> 	mutex_lock(&em28xx_devlist_mutex);
> 	list_add_tail(&dev->devlist, &em28xx_devlist);
> 	list_for_each_entry(ops, &em28xx_extension_devlist, next) {
> 		if (ops->init)
> 			ops->init(dev);
> 	}
> 	mutex_unlock(&em28xx_devlist_mutex);
> }
>
> The ops->init() receives the data from the main driver. It doesn't matter
> if the USB interface is the same or not.

Sure, but the device struct pointer "dev" passed to the module is different.

> If the changes at the probing code changed that, then it broke audio
> support for several devices.

If I'm not misunderstanding the code completely, em28xx_usb_probe() is
called for each USB interface.
If an analog video, audio or DVB endpoint ist detected, it creates a new
device instance (which has it's own device struct).

I don't know if this has ever been working differently, but at least it
didn't change recently. I also think this is how it should be.

Unfortunately I don't have one of these device for testing, so I don't
know if audio is currently broken for them.
What I can say for sure is, that devices with the isoc video/DVB
endpoints at an interface > 0 are currently broken, which is what I'm
trying to fix with this patch.

Regards,
Frank

>
>> Regards,
>> Frank
>>
>>> That's why audio_ifnum were added in the first place.
>>>
>>> See this commit:
>>>
>>> commit 4f83e7b3ef938eb9a01eadf81a0f3b2c67d3afb6
>>> Author: Mauro Carvalho Chehab <mchehab@redhat.com>
>>> Date:   Fri Jun 17 15:15:12 2011 -0300
>>>
>>>     [media] em28xx: Add support for devices with a separate audio interface
>>>     
>>>     Some devices use a separate interface for the vendor audio class.
>>>     
>>>     Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
>>>
>>>>>>  			dev->alt = 1;
>>>>>>  		else
>>>>>>  			dev->alt = 7;
>>>>>>  
>>>>>>  		dprintk("changing alternate number on interface %d to %d\n",
>>>>>> -			dev->audio_ifnum, dev->alt);
>>>>>> -		usb_set_interface(dev->udev, dev->audio_ifnum, dev->alt);
>>>>>> +			dev->ifnum, dev->alt);
>>>>>> +		usb_set_interface(dev->udev, dev->ifnum, dev->alt);
>>>>>>  
>>>>>>  		/* Sets volume, mute, etc */
>>>>>>  		dev->mute = 0;
>>>>>> @@ -642,7 +642,7 @@ static int em28xx_audio_init(struct em28xx *dev)
>>>>>>  	static int          devnr;
>>>>>>  	int                 err;
>>>>>>  
>>>>>> -	if (!dev->has_alsa_audio || dev->audio_ifnum < 0) {
>>>>>> +	if (!dev->has_alsa_audio) {
>>>>>>  		/* This device does not support the extension (in this case
>>>>>>  		   the device is expecting the snd-usb-audio module or
>>>>>>  		   doesn't have analog audio support at all) */
>>>>>> diff --git a/drivers/media/usb/em28xx/em28xx-cards.c b/drivers/media/usb/em28xx/em28xx-cards.c
>>>>>> index 0a5aa62..553db17 100644
>>>>>> --- a/drivers/media/usb/em28xx/em28xx-cards.c
>>>>>> +++ b/drivers/media/usb/em28xx/em28xx-cards.c
>>>>>> @@ -3376,7 +3376,7 @@ static int em28xx_usb_probe(struct usb_interface *interface,
>>>>>>  	dev->alt   = -1;
>>>>>>  	dev->is_audio_only = has_audio && !(has_video || has_dvb);
>>>>>>  	dev->has_alsa_audio = has_audio;
>>>>>> -	dev->audio_ifnum = ifnum;
>>>>>> +	dev->ifnum = ifnum;
>>>>>>  
>>>>>>  	/* Checks if audio is provided by some interface */
>>>>>>  	for (i = 0; i < udev->config->desc.bNumInterfaces; i++) {
>>>>>> diff --git a/drivers/media/usb/em28xx/em28xx-core.c b/drivers/media/usb/em28xx/em28xx-core.c
>>>>>> index ce4f252..210859a 100644
>>>>>> --- a/drivers/media/usb/em28xx/em28xx-core.c
>>>>>> +++ b/drivers/media/usb/em28xx/em28xx-core.c
>>>>>> @@ -862,7 +862,7 @@ set_alt:
>>>>>>  	}
>>>>>>  	em28xx_coredbg("setting alternate %d with wMaxPacketSize=%u\n",
>>>>>>  		       dev->alt, dev->max_pkt_size);
>>>>>> -	errCode = usb_set_interface(dev->udev, 0, dev->alt);
>>>>>> +	errCode = usb_set_interface(dev->udev, dev->ifnum, dev->alt);
>>>>>>  	if (errCode < 0) {
>>>>>>  		em28xx_errdev("cannot change alternate number to %d (error=%i)\n",
>>>>>>  				dev->alt, errCode);
>>>>> This hunk doesn't apply upstream:
>>>>>
>>>>> patching file drivers/media/usb/em28xx/em28xx-core.c
>>>>> Hunk #1 FAILED at 862.
>>>>> 1 out of 1 hunk FAILED -- rejects in file drivers/media/usb/em28xx/em28xx-core.c
>>>> It applies after
>>>>
>>>> http://patchwork.linuxtv.org/patch/16197/
>>>>
>>>> has been applied.
>>>>
>>>> Regards,
>>>> Frank
>>>>
>>>>>> diff --git a/drivers/media/usb/em28xx/em28xx-dvb.c b/drivers/media/usb/em28xx/em28xx-dvb.c
>>>>>> index a81ec2e..dbeed6c 100644
>>>>>> --- a/drivers/media/usb/em28xx/em28xx-dvb.c
>>>>>> +++ b/drivers/media/usb/em28xx/em28xx-dvb.c
>>>>>> @@ -196,7 +196,7 @@ static int em28xx_start_streaming(struct em28xx_dvb *dvb)
>>>>>>  		dvb_alt = dev->dvb_alt_isoc;
>>>>>>  	}
>>>>>>  
>>>>>> -	usb_set_interface(dev->udev, 0, dvb_alt);
>>>>>> +	usb_set_interface(dev->udev, dev->ifnum, dvb_alt);
>>>>>>  	rc = em28xx_set_mode(dev, EM28XX_DIGITAL_MODE);
>>>>>>  	if (rc < 0)
>>>>>>  		return rc;
>>>>>> diff --git a/drivers/media/usb/em28xx/em28xx.h b/drivers/media/usb/em28xx/em28xx.h
>>>>>> index 5f0b2c5..0dc5b73 100644
>>>>>> --- a/drivers/media/usb/em28xx/em28xx.h
>>>>>> +++ b/drivers/media/usb/em28xx/em28xx.h
>>>>>> @@ -487,8 +487,6 @@ struct em28xx {
>>>>>>  
>>>>>>  	unsigned char disconnected:1;	/* device has been diconnected */
>>>>>>  
>>>>>> -	int audio_ifnum;
>>>>>> -
>>>>>>  	struct v4l2_device v4l2_dev;
>>>>>>  	struct v4l2_ctrl_handler ctrl_handler;
>>>>>>  	/* provides ac97 mute and volume overrides */
>>>>>> @@ -597,6 +595,7 @@ struct em28xx {
>>>>>>  
>>>>>>  	/* usb transfer */
>>>>>>  	struct usb_device *udev;	/* the usb device */
>>>>>> +	int ifnum;			/* usb interface number */
>>>>>>  	u8 analog_ep_isoc;	/* address of isoc endpoint for analog */
>>>>>>  	u8 analog_ep_bulk;	/* address of bulk endpoint for analog */
>>>>>>  	u8 dvb_ep_isoc;		/* address of isoc endpoint for DVB */