From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([209.51.188.92]:34722) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1gt5lt-0004P1-Tr for qemu-devel@nongnu.org; Mon, 11 Feb 2019 02:12:54 -0500 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1gt5ls-0007zO-AJ for qemu-devel@nongnu.org; Mon, 11 Feb 2019 02:12:53 -0500 Date: Mon, 11 Feb 2019 15:12:35 +0800 From: Peter Xu Message-ID: <20190211071235.GE1011@xz-x1> References: <20190206174328.9736-1-marcandre.lureau@redhat.com> <20190206174328.9736-4-marcandre.lureau@redhat.com> <20190207181415.07d0d7ae.cohuck@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Disposition: inline In-Reply-To: <20190207181415.07d0d7ae.cohuck@redhat.com> Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v2 3/6] terminal3270: do not use backend timer sources List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: Cornelia Huck Cc: =?utf-8?Q?Marc-Andr=C3=A9?= Lureau , qemu-devel@nongnu.org, Paolo Bonzini , Halil Pasic , Christian Borntraeger , "open list:S390" On Thu, Feb 07, 2019 at 06:14:15PM +0100, Cornelia Huck wrote: > On Wed, 6 Feb 2019 18:43:25 +0100 > Marc-Andr=C3=A9 Lureau wrote: >=20 > > terminal3270 uses the front-end side of the chardev. It shouldn't > > create sources from backend side context (with backend > > functions). > >=20 > > send_timing_mark_cb calls qemu_chr_fe_write_all() which should be > > thread safe. > >=20 > > This partially reverts changes from commit > > 2c716ba1506769c9be2caa02f0f6d6e7c00f4304. > >=20 > > CC: Peter Xu > > Signed-off-by: Marc-Andr=C3=A9 Lureau > > --- > > hw/char/terminal3270.c | 15 ++++++--------- > > 1 file changed, 6 insertions(+), 9 deletions(-) >=20 > I've verified that 3270 continues to work as before. >=20 > Acked-by: Cornelia Huck A pure question: is it broken before this patch? Asked since I don't understand this patch and what it tries to fix... E.g., in terminal_init(), qemu_chr_fe_set_handlers() is always passing NULL as chardev context so AFAICT this patch changes nothing from functional-wise for now. Meanwhile, if we pass in some non-NULL into it in the future, IMHO this patch would break it instead of fixing anything... anything I've missed? Thanks, --=20 Peter Xu