[Libva] Error in Coded Offset value in encode sample app.

[Sambhav]

Hi,

I was able to get the proper encoded content after setting the offset to 16.



When a Surface is created with VA_RT_FORMAT_YUV420 and vaDeriveImage API is
called on the this surface, it returns fourcc format as VA_FOURCC_NV12. Is
this the only format supported ?
What all YUV formats does encoding acclerators in Moorestown platform
support ?

 Driver version : GMA500: psb_drv - 5.1.0.32L.0124
OS               : Moblin 2.1

Regards,
Sambhav

On Tue, Mar 2, 2010 at 10:27 AM, Yuan, Shengquan
<shengquan.yuan at gmail.com>wrote:

> On Mon, Mar 1, 2010 at 4:55 PM, Sambhav <isambhav at gmail.com> wrote:
> > Hi,
> >
> > I was trying the libva encode sample app on the Moorestown platform.
> > Encoding happens properly.
> > But when writing the encoded data to a file inside
> > the function "save_coded_buf" the coded_offset value returns junk.
> > The comment says "second DWord is byte offset". Is this correct ?
> Yes. It is correct.
>
> I am not sure which driver version you are using. The old version
> doesn't program the offset correctly.
> For Moorestown platform, the offset is always 16.
>
> >
> > Was anyone able to successfully run the encode sampe app ?
> >
> > Regards,
> > Sambhav
> >
> > _______________________________________________
> > Libva mailing list
> > Libva at lists.freedesktop.org
> > http://lists.freedesktop.org/mailman/listinfo/libva
> >
> >
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.freedesktop.org/archives/libva/attachments/20100302/b55423b6/attachment.html>