weston_surface_create() already inits the opaque region, so the second
init in create_black_surface() is logically wrong. Whether this was a
memory leak or not, depends on Pixman internals.
Fini before initing again.
Signed-off-by: Pekka Paalanen <ppaalanen@gmail.com>
surface->private = fs_surface;
weston_surface_configure(surface, x, y, w, h);
weston_surface_set_color(surface, 0.0, 0.0, 0.0, 1);
+ pixman_region32_fini(&surface->opaque);
pixman_region32_init_rect(&surface->opaque, 0, 0, w, h);
return surface;