On 05/19/2010 04:39 AM, Gwenole Beauchesne wrote:
> This awful hack is not guaranteed to work. Besides, the fallback
> mechanism already takes place gracefully in gtk_canvas.cpp.
Well, it was only a one line fix to workaround the segfault. I assume
somehow I merged your original patch in wrong if this driver check
wasn't in the correct place. But at least my ugly hack let Gnash
continue so I could actually try using vaapi.
I got the other patches you posted today merged in as of revno #12191,
interestingly my cpu load went up, not down when using my mpeg4 decoding
test. I'll have to dig into it. and see what happened.
- rob -
_______________________________________________
Gnash-dev mailing list
Gnash-dev@gnu.org
http://lists.gnu.org/mailman/listinfo/gnash-dev
