On Wed, Nov 16, 2016 at 11:18:04PM +0200, Abdiel Janulgue wrote: > @@ -179,7 +111,8 @@ static void flip_to_fb(igt_display_t *dpy, int pipe, > } > igt_assert_f(nanosleep(&tv, NULL) == -1, > "flip to %s blocked waiting for busy fb", name); > - finish_fb_busy(batch, 2*TIMEOUT); > + finish_fb_busy(t->batch, 2*TIMEOUT); As you didn't adjust finish_fb_busy, this is a double munmap. finish_fb_busy() is now just a sleep to allow the worker to do its things. That could be tidied up. -Chris -- Chris Wilson, Intel Open Source Technology Centre _______________________________________________ Intel-gfx mailing list Intel-gfx@xxxxxxxxxxxxxxxxxxxxx https://lists.freedesktop.org/mailman/listinfo/intel-gfx
