[MPlayer-dev-eng] Removal of backing store

[Stephane Marchesin]

On Fri, Aug 22, 2008 at 2:28 AM, Vladimir Mosgalin
<mosgalin at vm10124.spb.edu> wrote:
> Hi Stephane Marchesin!
>
>  On 2008.08.22 at 02:01:20 +0200, Stephane Marchesin wrote next:
>
>> some recent drivers (notably the open source ones using EXA). There
>
> Have you checked it for intel, ati and radeonhd drivers?
>

I don't have the hardware, but it should make no difference.

>> - most importantly, backing store prevents from using a real hardware
>> overlay (as opposed to blitter/texture adapters) because this forces
>
> In what case? For example I can choose any xv port (real overlay or
> textured video) on Intel 945 integrated chip. Both always worked fine.
> AFAIR both work even when compiz is enabled (too lazy to check it out
> now).

In nouveau, we have a port that is the hardware overlay, but it will
transparently fallback to the blitter when run in composite
conditions. Since composite can be enabled/disabled on the fly, we
have to do that. I suppose that's the only solution for drivers that
wish to retain hardware overlay performance.

>
>> Now, here are some benchmarks obtained when running mplayer -nosound
>
> Well, I tried this on radeon X1900 card with fglrx 8.8, using
> -vo gl:yuv=2:force-pbo, -vo xv and -vo x11 output drivers and
> performance was identical in each case, with or without patch. Is this
> patch nvidia-specific?
>

OpenGL is not affected by backing store, so your benchmark is meaningless.

However, this change is beneficial to:
1. drivers that use EXA and implement a hardware overlay which is able
to fallback to a blitter/texture
2. drivers that have a real hw overlay where you waste vram

Stephane