[PATCH] dri2: Don't short-circuit to blitting for vblank_mode=0.

[Ben Skeggs]

On Tue, Nov 13, 2012 at 10:22:54AM -0800, Eric Anholt wrote:
> When someone asks for no vblank syncing, they're also going for high
> performance, which forcing a blit is not.  Leave this up to the
> driver's normal SwapBuffers path, which can choose something
> appropriate to do based off of the target/divisor/remainder.
> 
> Note that this currently reduces performance on the Intel driver
> because it will still sync to the next vblank even though you asked
> for a swap target that's passed.
Any news on if/when this will get applied?

> ---
>  hw/xfree86/dri2/dri2.c |    4 ++--
>  1 file changed, 2 insertions(+), 2 deletions(-)
> 
> diff --git a/hw/xfree86/dri2/dri2.c b/hw/xfree86/dri2/dri2.c
> index d3b3c73..b17f45c 100644
> --- a/hw/xfree86/dri2/dri2.c
> +++ b/hw/xfree86/dri2/dri2.c
> @@ -878,8 +878,8 @@ DRI2SwapBuffers(ClientPtr client, DrawablePtr pDraw, CARD64 target_msc,
>          return BadDrawable;
>      }
>  
> -    /* Old DDX or no swap interval, just blit */
> -    if (!ds->ScheduleSwap || !pPriv->swap_interval) {
> +    /* Old DDX can only blit */
> +    if (!ds->ScheduleSwap) {
>          BoxRec box;
>          RegionRec region;
>  
> -- 
> 1.7.10.4
> 
> _______________________________________________
> xorg-devel at lists.x.org: X.Org development
> Archives: http://lists.x.org/archives/xorg-devel
> Info: http://lists.x.org/mailman/listinfo/xorg-devel