[media] gspca core: Fix regressions gspca breaking devices with audio
authorMauro Carvalho Chehab <mchehab@redhat.com>
Tue, 7 Dec 2010 20:19:09 +0000 (17:19 -0300)
committerMauro Carvalho Chehab <mchehab@redhat.com>
Wed, 29 Dec 2010 10:16:59 +0000 (08:16 -0200)
Changeset 35680ba broke several devices:
- Sony Playstation Eye (1415:2000);
- Gigaware model 25-234 (0c45:628f);
- Logitech Messenger Plus (046d:08f6).

Probably more devices were broken by this change.

What happens is that several devices don't need to save some bandwidth
for audio.

Also, as pointed by Hans de Goede <hdegoede@redhat.com>, the logic
that implements the bandwidth reservation for audio is broken, since
it will reduce the alt number twice, on devices with audio.

So, let's just revert the broken logic, and think on a better solution
for usb 1.1 devices with audio that can't use the maximum packetsize.

Acked-by: Jean-Francois Moine <moinejf@free.fr>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
drivers/media/video/gspca/gspca.c

index 0a7af737150946f0b0903d08ca8afeb343a7bb94..5a2274b2b0e8ed962f55c33678fefb56447e7d27 100644 (file)
@@ -652,16 +652,12 @@ static struct usb_host_endpoint *get_ep(struct gspca_dev *gspca_dev)
                                   : USB_ENDPOINT_XFER_ISOC;
        i = gspca_dev->alt;                     /* previous alt setting */
        if (gspca_dev->cam.reverse_alts) {
-               if (gspca_dev->audio && i < gspca_dev->nbalt - 2)
-                       i++;
                while (++i < gspca_dev->nbalt) {
                        ep = alt_xfer(&intf->altsetting[i], xfer);
                        if (ep)
                                break;
                }
        } else {
-               if (gspca_dev->audio && i > 1)
-                       i--;
                while (--i >= 0) {
                        ep = alt_xfer(&intf->altsetting[i], xfer);
                        if (ep)