On Tue, 2011-03-29 at 08:02 -0700, Keith Packard wrote: > On Tue, 29 Mar 2011 10:44:54 -0400, Adam Jackson <[email protected]> wrote: > > > Given that, I think we have to make it so damage actions against a > > Damage that has lost its Drawable simply fizzle, and continue to leave > > cleanup to the clients. Compiled, not tested: > > Yeah, that sounds right to me. Retains the current semantics, but > removes the pointer to freed memory.
Hmph. In trying to build a testcase for this I think I've discovered
that we _do_ already gc Damages when their drawables die; it's just
harder to see because it's done as a callback from miext/damage's
wrappers around Destroy{Pixmap,Window}.
Which, I suspect, means my original patch was simply correct. But I'll
keep digging.
- ajax
signature.asc
Description: This is a digitally signed message part
_______________________________________________ [email protected]: X.Org development Archives: http://lists.x.org/archives/xorg-devel Info: http://lists.x.org/mailman/listinfo/xorg-devel
