From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:34963) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WYets-0008FJ-5D for qemu-devel@nongnu.org; Fri, 11 Apr 2014 13:06:04 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WYetm-0006zJ-SL for qemu-devel@nongnu.org; Fri, 11 Apr 2014 13:06:00 -0400 Received: from mx1.redhat.com ([209.132.183.28]:9782) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WYetm-0006z7-JQ for qemu-devel@nongnu.org; Fri, 11 Apr 2014 13:05:54 -0400 Message-ID: <534820ED.9080209@redhat.com> Date: Fri, 11 Apr 2014 13:05:49 -0400 From: Cole Robinson MIME-Version: 1.0 References: <2f8f007ce2152ac3b65f0811199662799c509225.1397155389.git.crobinso@redhat.com> <53481F15.7010902@suse.de> In-Reply-To: <53481F15.7010902@suse.de> Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH for-2.0] qom: Fix crash with qom-list and link properties List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: =?UTF-8?B?QW5kcmVhcyBGw6RyYmVy?= , qemu-devel@nongnu.org, Peter Maydell Cc: m.gibula@beyond.pl, Stefan Hajnoczi , Marcel Apfelbaum On 04/11/2014 12:57 PM, Andreas F=C3=A4rber wrote: > Am 10.04.2014 20:47, schrieb Cole Robinson: >> Commit 9561fda8d90e176bef598ba87c42a1bd6ad03ef7 changed the type of >> 'opaque' for link properties, but missed updating this call site. >> Reproducer: >> >> ./x86_64-softmmu/qemu-system-x86_64 -qmp unix:./qmp.sock,server & >> ./scripts/qmp/qmp-shell ./qmp.sock >> (QEMU) qom-list path=3D//machine/i440fx/pci.0/child[2] >=20 > I would much prefer if we could give the path as just > path=3D/machine/i440fx/pci.0/child[2] (without double slash). Peter can > you improve when applying? >=20 >> >> Reported-by: Marcin Gibu=C5=82a >> Signed-off-by: Cole Robinson >> --- >> qom/object.c | 3 ++- >> 1 file changed, 2 insertions(+), 1 deletion(-) >=20 > Reviewed-by: Andreas F=C3=A4rber >=20 > I would like to remark that specifically for regression-testing link<> > properties I had extended qom-test, and it succeeds currently. So Cole, > if you have any suggestion on how to extend it to catch this regression > that would be welcome. >=20 I haven't looked closely at tests/, I see *qmp*.c files, but is there infrastructure for running qmp commands? If so we could turn the above reproducer into a test case. - Cole