From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([208.118.235.92]:43890) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1SEn8e-00054B-O9 for qemu-devel@nongnu.org; Mon, 02 Apr 2012 15:42:05 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1SEn8d-0006AV-0j for qemu-devel@nongnu.org; Mon, 02 Apr 2012 15:42:04 -0400 Received: from mail-gy0-f173.google.com ([209.85.160.173]:52184) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1SEn8c-00069z-SE for qemu-devel@nongnu.org; Mon, 02 Apr 2012 15:42:02 -0400 Received: by ghrr14 with SMTP id r14so2003531ghr.4 for ; Mon, 02 Apr 2012 12:42:01 -0700 (PDT) Message-ID: <4F7A0105.8030707@codemonkey.ws> Date: Mon, 02 Apr 2012 14:41:57 -0500 From: Anthony Liguori MIME-Version: 1.0 References: <4F79FF0E.9090701@web.de> In-Reply-To: <4F79FF0E.9090701@web.de> Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Subject: Re: [Qemu-devel] qdev property listing broken List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Jan Kiszka Cc: Anthony Liguori , qemu-devel On 04/02/2012 02:33 PM, Jan Kiszka wrote: > Hi Anthony, > > I noticed that only properties with legacy names are printed via info > qtree. Yes. Is that not what you expect? The qom properties and model is not stable for 1.1 so I was very careful in making sure they didn't leak into info qtree. > After digging through the qdev and qom property maze, it turned > out the property registration in qdev_property_add_legacy and > qdev_property_add_static is not consistent with the access in > qdev_print_props. qdev_print_props will completely die for 1.2 as will almost everything related to the human monitor in qdev-monitor.c. I just wanted to give us a full release to make sure we were happy with the various interfaces. > The latter assumes all properties are strings, the > former generate the full set of types - and add_legacy obviously an > inconsistent one, dependent on the existence of print/parse handlers. I > fail to see the right direction, ie. where to fix this. Can you provide > a hint? Is there a reason you're using info qtree instead of qom-list? qom-list gives you much more info than info qtree. Regards, Anthony Liguori > Jan > > PS: It's really no fun to understand and debug this code anymore. > Hopefully, the removal of the qdev layer can improve this again.