On Mon, 30 Apr 2007, Daniel Drake wrote:

Commit 93c8bf45e083b89dffe3a708363c15c1b220c723 modified the USB device matching behaviour to ignore interface class matches if the device class is vendor-specific.

This patch adds an explicit ID match for the Logitech QuickCam Messenger, which has a vendor specific device class (but standards-compliant audio interfaces).

This fixes a 2.6.20 regression where the audio component of this device was no longer usable.

http://bugs.gentoo.org/show_bug.cgi?id=175715 https://bugs.launchpad.net/ubuntu/+source/linux-source-2.6.20/+bug/93822 https://bugtrack.alsa-project.org/alsa-bug/view.php?id=3040

Based on a patch from sergiom

Signed-off-by: Daniel Drake dsd@gentoo.org

Index: linux/sound/usb/usbquirks.h

--- linux.orig/sound/usb/usbquirks.h +++ linux/sound/usb/usbquirks.h @@ -40,6 +40,21 @@ .bInterfaceClass = USB_CLASS_VENDOR_SPEC

/*

• • Logitech QuickCam: bDeviceClass is vendor-specific, so generic interface
• • class matches do not take effect without an explicit ID match.
• */

+{

• .match_flags = USB_DEVICE_ID_MATCH_DEVICE |

•        USB_DEVICE_ID_MATCH_INT_CLASS |
  

•        USB_DEVICE_ID_MATCH_INT_SUBCLASS,
  

• .idVendor = 0x046d,
• .idProduct = 0x08f0,
• .bInterfaceClass = USB_CLASS_AUDIO,
• .bInterfaceSubClass = USB_SUBCLASS_AUDIO_CONTROL

+},

+/*

• Yamaha devices

*/

This ought to fix the problem.

You might want to consider adding a macro to include/linux/usb.h to handle cases like this. Something like USB_DEVICE_AND_INTERFACE_INFO() -- it would make life easier for other people in the future when faced with the same problem.

Alan Stern