[PATCH] Revert "dix: use the event mask of the grab for TryClientEvents."

Daniel Stone daniel at fooishbar.org
Fri Jun 25 01:28:50 PDT 2010


On Fri, Jun 25, 2010 at 10:51:47AM +1000, Peter Hutterer wrote:
> On Fri, Jun 25, 2010 at 01:27:10AM +0100, Daniel Stone wrote:
> > On Fri, Jun 25, 2010 at 10:16:02AM +1000, Peter Hutterer wrote:
> > > On Fri, Jun 25, 2010 at 01:09:41AM +0100, Daniel Stone wrote:
> > > > The reverted commit was meant to fix some kind of Motif bug (possibly
> > > > mouse getting confined to a widget) or something along those lines,
> > > > right? The explanation for this commit seems fine to me, but I vaguely
> > > > remember the original being some kind of bugfix ... is it no longer
> > > > necessary, or just didn't work out, or?
> > > 
> > > I think you're referring to commit 1c612acca8568fcdf9761d23f112adaf4d496f1b
> > > 
> > >     dix: if owner-events is true for passive grabs, add the window mask
> > >     (#25400)
> > > 
> > > that's lesstif bug where the pointer got stuck in a window. With this patch,
> > > the need for 1c612 is obsolete, since mask & filter is always true,
> > > regardless of owner_events. Or should be, anyway.
> > 
> > Ah OK, so the original commit was just a general desire to DTRT,
> 
> yes, AFAICT.
> 
> > 1c612acca can now almost certainly be reverted, and that may even fix
> 
> it's removed with this patch since this commit didn't revert without
> conflicts. but I'll add this to the commit message.
> 
> > #25400 if we're being optimistic?
> 
> verified as still working with the test case attached to 25400.

Cool! Now I've got your hopes up ... I don't have a great deal to
contribute here beyond my Acked-by, sorry.  But the commit message will
be a lot nicer for it. :)

Cheers,
Daniel le tease
-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/pgp-signature
Size: 198 bytes
Desc: Digital signature
URL: <http://lists.x.org/archives/xorg-devel/attachments/20100625/f7966dd6/attachment-0001.pgp>


More information about the xorg-devel mailing list