[libdv-dev] =?ISO-8859-1?Q?Re=3A=20=5Blibdv=2Ddev=5D=20Re=3A=20Should=20dv1394=20work=20on=20PPC=3F?=

[<Ste...@ep...>]

>On Tue, 18 Jun 2002, Daniel Kobras wrote:
>
>> On Tue, Jun 18, 2002 at 09:57:49AM -0400, Bill Fink wrote:
>> > Then can you make the YUY2 part conditional since it will break DRI
>> > enabled ATI cards as well as non-ATI cards such as NVIDIA.
>> 
>> Uh.  So they want their stuff in PCI-endianness with DRI enabled?  Ugl=
y.
>> For the record, I tested the patch on the Powerbook with DRI on, so at=

>> least they want their stuff DMAed big-endian.  Bill, what does your
>> machine call mode 0x3259555 when you run xvinfo?  YUY2, or 2YUY?  It'd=

>> be rather easy to provide both little and big endian versions of the
>> conversion code, but there needs to be a reliable way to find out whic=
h
>> endianness the card does expect.
>
>Well it's actually id 0x32595559 which it says is 2YUY (the 59 at the en=
d
>of the id seems to correspond to xv port base 59 which is what is used
>with DRI enabled, while port base 55 was used with DRI disabled).
NO, thouse IDs come from X11. 
#define FOURCC_YUY2 0x32595559
is from:
xc/programs/Xserver/hw/xfree86/common/fourcc.h

playdv checks presented X11 formats against available renderer formats.
So ATI
DMA/DRI issue should be / is (??) fixed in X11 (2002-06-07 CVS).




________________________________________
Zeitschriftenabos online bestellen - jetzt neu im Infoboten! http://www.e=
post.de