From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([209.51.188.92]:46580) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1h3NGP-0006QS-NW for qemu-devel@nongnu.org; Mon, 11 Mar 2019 11:54:54 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1h3NDw-0006vw-HM for qemu-devel@nongnu.org; Mon, 11 Mar 2019 11:52:21 -0400 Received: from mx1.redhat.com ([209.132.183.28]:60558) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1h3NDw-0006vY-9h for qemu-devel@nongnu.org; Mon, 11 Mar 2019 11:52:20 -0400 From: Markus Armbruster References: <20180103214925.16677-1-f4bug@amsat.org> <20180103214925.16677-2-f4bug@amsat.org> <07100472-e75f-88c7-6740-21a56e225f80@redhat.com> <20190311134306.GJ8899@habkost.net> Date: Mon, 11 Mar 2019 16:52:13 +0100 In-Reply-To: (Peter Maydell's message of "Mon, 11 Mar 2019 13:48:00 +0000") Message-ID: <87imwpl87m.fsf@dusky.pond.sub.org> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v2 1/4] sdbus: add a QMP command to access a SDBus List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Peter Maydell Cc: Eduardo Habkost , Kevin Wolf , Edgar Iglesias , Thomas Huth , Alistair Francis , Philippe =?utf-8?Q?Mathieu-Daud=C3=A9?= , QEMU Developers , Stefan Hajnoczi , =?utf-8?Q?Marc-Andr=C3=A9?= Lureau , Paolo Bonzini Peter Maydell writes: > On Mon, 11 Mar 2019 at 13:43, Eduardo Habkost wrote: >> >> On Mon, Mar 11, 2019 at 12:49:50PM +0100, Thomas Huth wrote: >> > On 08/03/2019 17.11, Philippe Mathieu-Daud=C3=A9 wrote: >> > > Hi Markus, >> > > >> > > [Asking again from the correct series thread] >> > > >> > > On 1/3/18 10:49 PM, Philippe Mathieu-Daud=C3=A9 wrote: >> > >> Use Base64 to serialize the binary blobs in JSON. >> > >> So far at most 512 bytes will be transfered, which result >> > >> in a 684 bytes payload. >> > >> Since this command is intented for qtesting, it is acceptable. >> > > >> > > Any comment regarding QMP for this patch? >> > >> > Is this useful for anybody else than qtest? If not, I think this should >> > rather go into the qtest protocol instead, since QMP is our "public" >> > protocol. >> >> Extending qtest requires writing parsers by hand. Do we really >> want to go that route and start extending the qtest protocol more >> often? > > Perhaps we could have qtest-only QMP commands that only get > recognized if qtest_enabled() ? QMP commands are defined at compile time. We just got rid of the hack to "unrecognize" selected commands dynamically at run-time (commit 0b69f6f72ce), which we had because our compile-time facilities were lacking. I'd hate to bring this hack back. What's easy is to have QMP commands that fail unless qtest_enabled() :)