[PATCH] glx/dri2: Disable AIGLX if indirect GLX is disabled

Chris Wilson chris at chris-wilson.co.uk
Wed Mar 4 23:47:06 PST 2015


On Thu, Mar 05, 2015 at 01:18:51PM +0900, Michel Dänzer wrote:
> On 04.03.2015 21:16, Chris Wilson wrote:
> > There is no point in setting up the acceleration for indirect GLX if
> > indirect GLX is itself disabled.
> > 
> > Signed-off-by: Chris Wilson <chris at chris-wilson.co.uk>
> > ---
> >  glx/glxdri2.c | 3 +++
> >  1 file changed, 3 insertions(+)
> > 
> > diff --git a/glx/glxdri2.c b/glx/glxdri2.c
> > index ec86a73..7cb0f28 100644
> > --- a/glx/glxdri2.c
> > +++ b/glx/glxdri2.c
> > @@ -945,6 +945,9 @@ __glXDRIscreenProbe(ScreenPtr pScreen)
> >      size_t buffer_size;
> >      ScrnInfoPtr pScrn = xf86ScreenToScrn(pScreen);
> >  
> > +    if (!enableIndirectGLX)
> > +	    return NULL;
> > +
> >      screen = calloc(1, sizeof *screen);
> >      if (screen == NULL)
> >          return NULL;
> > 
> 
> Can the GLX visual information still be probed from the DRI driver with
> this patch?

No. This patch kills the AIGLX messages and also DRI. Sadly as probing
mesa/i965 is the single most CPU expensive step in starting X.
-Chris

-- 
Chris Wilson, Intel Open Source Technology Centre


More information about the xorg-devel mailing list