From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:44749) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XcHXf-0008N9-6Y for qemu-devel@nongnu.org; Thu, 09 Oct 2014 13:30:25 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1XcHXW-0004sn-VW for qemu-devel@nongnu.org; Thu, 09 Oct 2014 13:30:19 -0400 Received: from mx1.redhat.com ([209.132.183.28]:5797) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1XcHXW-0004sU-OL for qemu-devel@nongnu.org; Thu, 09 Oct 2014 13:30:10 -0400 Message-ID: <5436C618.90703@redhat.com> Date: Thu, 09 Oct 2014 13:30:00 -0400 From: Cole Robinson MIME-Version: 1.0 References: <20141004081842.GC2401@slate> <1412582801.5003.14.camel@nilsson.home.kraxel.org> <1412591302.5003.22.camel@nilsson.home.kraxel.org> In-Reply-To: Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] Fullscreen Bug with GTK List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Maydell , Gerd Hoffmann Cc: Cedric Sodhi , QEMU Developers On 10/06/2014 07:24 AM, Peter Maydell wrote: > On 6 October 2014 11:28, Gerd Hoffmann wrote: >> On Mo, 2014-10-06 at 12:14 +0200, Cedric Sodhi wrote: >>> The reporter explicitly mentioned that removing the Menu disables the >>> Accels and therefore he has provided a patch which attaches the Accels >>> to the Window instead (which, in fact, is how it should be done). >> >> Oh, missed that detail. /me recommends submitting the patches to >> qemu-devel then (and cc me so I don't miss them). > > ...with the appropriate signed-off-by lines, which are > missing from the patch in the bug. > The patches in the original report were not meant to be submitted as is, and instead I hoped it would start a discussion, because they depend on APIs only available with gtk3.8, and the fallback loses some usefulness. Also the example patch isn't complete, since I think we would want to install all the accelerators on the parent window, rather than just the fullscreen one. Then in fullscreen mode, every accelerator would continue to work, which matches SDL behavior IIRC. I think if someone fleshed out my patch examples it should be fine for them to use their own Signed-off-by. I don't have any plans to submit a patch myself in the short term. Also there's other possible approaches. Like disabling accelerators entirely when in fullscreen mode, and instead use a hiding toolbar like virt-manager and virt-viewer do. danpb pointed out that gtk 3.10 has a GtkRevealer widget that could probably be used for that without having to import the autodrawer code that virt-viewer uses. - Cole