On Fr, 2016-08-26 at 13:47 +0400, Marc-André Lureau wrote: > virtio-gpu does a set-scanout at each frame (it might be a driver > regression).
xorg vs wayland maybe? as far I know only wayland pageflips for each frame. > qemu_console_resize() recreate a surface even if the size > didn't change, and this shows up in profiling reports because the > surface is cleared. With this patch, I get a +15-20% glmark2 > improvement. Does anything break if you simply drop the qemu_console_resize() call from virgl_cmd_set_scanout? In theory we should not need a surface in the first place in scanout mode, but I suspect there is a reason why the call is there ... cheers, Gerd