From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:52115)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1c9ElC-0002xs-AW
	for qemu-devel@nongnu.org; Tue, 22 Nov 2016 12:21:35 -0500
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1c9El9-0007Pr-3V
	for qemu-devel@nongnu.org; Tue, 22 Nov 2016 12:21:34 -0500
Received: from mx1.redhat.com ([209.132.183.28]:47448)
	by eggs.gnu.org with esmtps (TLS1.0:DHE_RSA_AES_256_CBC_SHA1:32)
	(Exim 4.71) (envelope-from <pbonzini@redhat.com>) id 1c9El8-0007PW-So
	for qemu-devel@nongnu.org; Tue, 22 Nov 2016 12:21:31 -0500
References: <1479832306-26440-1-git-send-email-stefanha@redhat.com>
	<1479832306-26440-10-git-send-email-stefanha@redhat.com>
From: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <e2085f33-351d-f413-b8c7-577e8faaf474@redhat.com>
Date: Tue, 22 Nov 2016 18:21:24 +0100
MIME-Version: 1.0
In-Reply-To: <1479832306-26440-10-git-send-email-stefanha@redhat.com>
Content-Type: text/plain; charset=utf-8
Content-Transfer-Encoding: quoted-printable
Subject: Re: [Qemu-devel] [PATCH v3 09/10] aio: add .io_poll_begin/end()
 callbacks
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel/>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Stefan Hajnoczi <stefanha@redhat.com>, qemu-devel@nongnu.org
Cc: borntraeger@de.ibm.com, Karl Rister <krister@redhat.com>, Fam Zheng <famz@redhat.com>



On 22/11/2016 17:31, Stefan Hajnoczi wrote:
> +static bool try_poll_mode(AioContext *ctx, bool enable)
> +{
> +    if (enable && aio_poll_max_ns && ctx->poll_disable_cnt =3D=3D 0) {
> +        /* See qemu_soonest_timeout() uint64_t hack */
> +        int64_t max_ns =3D MIN((uint64_t)aio_compute_timeout(ctx),
> +                             (uint64_t)aio_poll_max_ns);
> +
> +        if (max_ns) {
> +            poll_set_started(ctx, true);
> +
> +            if (run_poll_handlers(ctx, max_ns)) {
> +                return true;
> +            }
> +        }
> +    }
> +
> +    poll_set_started(ctx, false);

You could do a single iteration even if enable =3D=3D false (which I'd
rename to blocking, BTW, because poll_start can be false on exit even if
enable =3D=3D true).

In fact, since (like virtio_queue_host_notifier_aio_poll_end) all
.io_poll_end() callbacks are going to poll once more, what about adding
here:

    return run_poll_handlers(ctx, 0);

or just an instance of the loop, without qemu_clock_get_ns and the
tracepoints:

    return run_poll_handlers_once(ctx);

and removing from patch 10 the

    /* Handle any buffers that snuck in after we finished polling */
    virtio_queue_host_notifier_aio_poll(n);

?

Thanks,

Paolo