Re: [alsa-devel] USB Audio Interface / Denon MC7000 and MC8000 controller

[Tobias <toszlanyi@yahoo.de>]

Thank you so much Alexander!
I used latest Kernel and patched as you suggested. The Device is working 
now giving sound on all 4 channels, even though dmesg still shows the 
error message as you can see here:

uname -a:
Linux tobias-V130 5.5.2 #1 SMP Thu Feb 6 09:41:57 CET 2020 x86_64 x86_64 
x86_64 GNU/Linux

dmesg:
[   62.918777] usb 1-1.3: new high-speed USB device number 6 using xhci_hcd
[   62.939293] usb 1-1.3: New USB device found, idVendor=15e4, 
idProduct=8004, bcdDevice=11.10
[   62.939295] usb 1-1.3: New USB device strings: Mfr=1, Product=2, 
SerialNumber=3
[   62.939297] usb 1-1.3: Product: DENON DJ MC7000
[   62.939298] usb 1-1.3: Manufacturer: DENON DJ
[   62.939299] usb 1-1.3: SerialNumber: 201603
[   62.942232] usb 1-1.3: clock source 65 is not valid, cannot use
[   62.943998] usb 1-1.3: clock source 65 is not valid, cannot use
[   63.013306] usb 1-1.3: clock source 65 is not valid, cannot use
[   63.028912] usb 1-1.3: clock source 65 is not valid, cannot use
[   63.029675] usb 1-1.3: clock source 65 is not valid, cannot use
[   63.037813] usb 1-1.3: clock source 65 is not valid, cannot use
[   63.063865] usb 1-1.3: clock source 65 is not valid, cannot use

I checked in file /sound/usb/clock.c that within functions

static int __uac_clock_find_source
static int __uac3_clock_find_source

there is another check that possibly gives the warning.

Maybe the warning "cannot use" should not be displayed when a Denon 
Audio device is attached as it is misleading.

Thanks a lot for your next patch that I would like to test.
Tobias



Am 05.02.20 um 20:07 schrieb Alexander Tsoy:
> В Пн, 03/02/2020 в 11:23 +0100, Tobias пишет:
>> Dear Alexander - unfortunately I did hot hear back from you. I guess
>> this may not be your highest priority but still - do you have any
>> other
>> idea to make the MC7000 sound device working? Thanks a lot.
> I think it should be safe to ignore clock validity check result if:
> - only one single sample rate is supported;
> - the clock source is internal,
> - there is no clock selector.
>
> Could you try the following patch?
>
>
> diff --git a/sound/usb/clock.c b/sound/usb/clock.c
> index 018b1ecb5404..636c340d4f9f 100644
> --- a/sound/usb/clock.c
> +++ b/sound/usb/clock.c
> @@ -197,6 +197,38 @@ static bool uac_clock_source_is_valid(struct snd_usb_audio *chip,
>   	return data ? true :  false;
>   }
>   
> +/*
> + * Assume the clock is valid if clock source supports only one single sample
> + * rate, its type is not external and there is no clock selector. This is needed
> + * for some Denon DJ controllers, that always report that clock is invalid.
> + */
> +static bool uac_clock_source_is_valid_quirk(struct snd_usb_audio *chip,
> +					    struct audioformat *fmt,
> +					    int clock)
> +{
> +	if (fmt->protocol == UAC_VERSION_3) {
> +		struct uac3_clock_source_descriptor *cs_desc =
> +			snd_usb_find_clock_source_v3(chip->ctrl_intf, clock);
> +
> +		if (!cs_desc)
> +			return false;
> +
> +		return (fmt->nr_rates == 1 &&
> +			(fmt->clock & 0xff) == cs_desc->bClockID &&
> +			(cs_desc->bmAttributes & 0x1) != UAC3_CLOCK_SOURCE_TYPE_EXT);
> +	} else { /* UAC_VERSION_2 */
> +		struct uac_clock_source_descriptor *cs_desc =
> +			snd_usb_find_clock_source(chip->ctrl_intf, clock);
> +
> +		if (!cs_desc)
> +			return false;
> +
> +		return (fmt->nr_rates == 1 &&
> +			(fmt->clock & 0xff) == cs_desc->bClockID &&
> +			(cs_desc->bmAttributes & 0x3) != UAC_CLOCK_SOURCE_TYPE_EXT);
> +	}
> +}
> +
>   static int __uac_clock_find_source(struct snd_usb_audio *chip, int entity_id,
>   				   unsigned long *visited, bool validate)
>   {
> @@ -577,7 +609,8 @@ static int set_sample_rate_v2v3(struct snd_usb_audio *chip, int iface,
>   
>   validation:
>   	/* validate clock after rate change */
> -	if (!uac_clock_source_is_valid(chip, fmt->protocol, clock))
> +	if (!uac_clock_source_is_valid(chip, fmt->protocol, clock) &&
> +	    !uac_clock_source_is_valid_quirk(chip, fmt, clock))
>   		return -ENXIO;
>   	return 0;
>   }
>

_______________________________________________
Alsa-devel mailing list
Alsa-devel@alsa-project.org
https://mailman.alsa-project.org/mailman/listinfo/alsa-devel