From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:41572) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1ffIh3-000267-TI for qemu-devel@nongnu.org; Tue, 17 Jul 2018 01:38:39 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1ffIh0-0006LR-Q4 for qemu-devel@nongnu.org; Tue, 17 Jul 2018 01:38:37 -0400 Received: from mx3-rdu2.redhat.com ([66.187.233.73]:53708 helo=mx1.redhat.com) by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32) (Exim 4.71) (envelope-from ) id 1ffIh0-0006LF-Ju for qemu-devel@nongnu.org; Tue, 17 Jul 2018 01:38:34 -0400 Received: from smtp.corp.redhat.com (int-mx04.intmail.prod.int.rdu2.redhat.com [10.11.54.4]) (using TLSv1.2 with cipher AECDH-AES256-SHA (256/256 bits)) (No client certificate requested) by mx1.redhat.com (Postfix) with ESMTPS id E88B9401EF0B for ; Tue, 17 Jul 2018 05:38:33 +0000 (UTC) From: Markus Armbruster References: <20180706121354.2021-1-marcandre.lureau@redhat.com> <20180706121354.2021-6-marcandre.lureau@redhat.com> Date: Tue, 17 Jul 2018 07:38:32 +0200 In-Reply-To: <20180706121354.2021-6-marcandre.lureau@redhat.com> (=?utf-8?Q?=22Marc-Andr=C3=A9?= Lureau"'s message of "Fri, 6 Jul 2018 14:13:47 +0200") Message-ID: <87va9eh15z.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 05/12] monitor: no need to save need_resume List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: =?utf-8?Q?Marc-Andr=C3=A9?= Lureau Cc: qemu-devel@nongnu.org, armbru@redhat.com, peterx@redhat.com Marc-Andr=C3=A9 Lureau writes: > There is no need for per-command need_resume granularity, it should > resume after running an non-oob command on oob-disabled monitor. > > Signed-off-by: Marc-Andr=C3=A9 Lureau > --- > monitor.c | 12 +++--------- > 1 file changed, 3 insertions(+), 9 deletions(-) > > diff --git a/monitor.c b/monitor.c > index 462cf96f7b..ec40a80d81 100644 > --- a/monitor.c > +++ b/monitor.c > @@ -257,12 +257,6 @@ struct QMPRequest { > */ > QObject *req; > Error *err; > - /* > - * Whether we need to resume the monitor afterward. This flag is > - * used to emulate the old QMP server behavior that the current > - * command must be completed before execution of the next one. > - */ > - bool need_resume; Note for later: this comment goes away without a replacement. > }; > typedef struct QMPRequest QMPRequest; >=20=20 > @@ -4079,11 +4073,13 @@ static void monitor_qmp_bh_dispatcher(void *data) > { > QMPRequest *req_obj =3D monitor_qmp_requests_pop_any(); > QDict *rsp; > + bool need_resume; >=20=20 > if (!req_obj) { > return; > } >=20=20 > + need_resume =3D !qmp_oob_enabled(req_obj->mon); > if (req_obj->req) { > trace_monitor_qmp_cmd_in_band(qobject_get_try_str(req_obj->id) ?= : ""); > monitor_qmp_dispatch(req_obj->mon, req_obj->req, req_obj->id); > @@ -4094,7 +4090,7 @@ static void monitor_qmp_bh_dispatcher(void *data) > qobject_unref(rsp); > } >=20=20 > - if (req_obj->need_resume) { > + if (need_resume) { > /* Pairs with the monitor_suspend() in handle_qmp_command() */ Why not if (!qmp_oob_enabled(req_obj->mon)) { and drop the variable? Perhaps you keep the variable to have its name provide a hint on why we resume. But even with that hint and the "pairs with" comment, it's still less than clear. What about... > monitor_resume(req_obj->mon); > } > @@ -4146,7 +4142,6 @@ static void handle_qmp_command(JSONMessageParser *p= arser, GQueue *tokens) > req_obj->id =3D id; > req_obj->req =3D req; > req_obj->err =3D err; > - req_obj->need_resume =3D false; >=20=20 > /* Protect qmp_requests and fetching its length. */ > qemu_mutex_lock(&mon->qmp.qmp_queue_lock); > @@ -4159,7 +4154,6 @@ static void handle_qmp_command(JSONMessageParser *p= arser, GQueue *tokens) > */ > if (!qmp_oob_enabled(mon)) { ... adding a replacement for the deleted comment here, say /* * Emulate traditional QMP server behavior: read next command * only after current command completed. Pairs with * monitor_resume() in monitor_qmp_bh_dispatcher(). */ Would you then be okay with cutting out the variable? > monitor_suspend(mon); > - req_obj->need_resume =3D true; > } else { > /* Drop the request if queue is full. */ > if (mon->qmp.qmp_requests->length >=3D QMP_REQ_QUEUE_LEN_MAX) { Your patch works fine as far as I can tell, so Reviewed-by: Markus Armbruster