Hi Dmitriy,

On Tuesday 24 March 2009 23:13:01 Dmitriy Geels wrote:
> 2009/3/24 Laurent Pinchart <[email protected]>:
> > The goal was, if I remember well, to have rotation implemented in a
> > completely transparent way for userspace applications. The applications
> > will, of course, have to use libv4l(2).
>
> Mmm... sounds confusing. Does all applications use libv4l?

Not yet, but most are either ported or are getting ported.

> > libv4l(2) should detect that the device requires rotation (based on the
> > USB VID:PID) and rotate the image in software if hardware rotation isn't
> > supported.
> >
> >> I will try to make some reverse engineering on original drivers, may
> >> be there is something, that control hardware flipping.
> >
> > What camera are you using ? Please post the USB descriptors (retrieved
> > with 'lsusb -v' using usbutils 0.72 or newer - 0.73+ preferred).
>
> Well known one:
> Bus 002 Device 003: ID 04f2:b012 Chicony Electronics Co., Ltd 1.3
> MPixel UVC webcam
>
> On packard bell easynote bg46 laptop (seems to be designed and
> manufactured by asus).
>
> lsusb output: http://paste.org.ru/index.pl?ht6viv

While your camera probably supports image rotation in hardware (most sensor 
can be read out upside down) the function is not exported to the host :-S

Best regards,

Laurent Pinchart

_______________________________________________
Linux-uvc-devel mailing list
[email protected]
https://lists.berlios.de/mailman/listinfo/linux-uvc-devel

Reply via email to