From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:32940)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1XqjQZ-0004lX-GP
	for qemu-devel@nongnu.org; Tue, 18 Nov 2014 09:06:50 -0500
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1XqjQS-0001ZX-QO
	for qemu-devel@nongnu.org; Tue, 18 Nov 2014 09:06:43 -0500
Received: from mx1.redhat.com ([209.132.183.28]:55703)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1XqjQS-0001ZN-Ir
	for qemu-devel@nongnu.org; Tue, 18 Nov 2014 09:06:36 -0500
Message-ID: <546B525F.3070001@redhat.com>
Date: Tue, 18 Nov 2014 15:06:23 +0100
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
References: <1415286601-30715-1-git-send-email-ming.lei@canonical.com>
	<1415286601-30715-3-git-send-email-ming.lei@canonical.com>
In-Reply-To: <1415286601-30715-3-git-send-email-ming.lei@canonical.com>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH v3 2/3] linux-aio: handling -EAGAIN for
 !s->io_q.plugged case
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: Ming Lei <ming.lei@canonical.com>, qemu-devel@nongnu.org, Peter Maydell <peter.maydell@linaro.org>, Stefan Hajnoczi <stefanha@redhat.com>, Kevin Wolf <kwolf@redhat.com>
Cc: =?windows-1252?Q?Beno=EEt_Canet?= <benoit.canet@irqsave.net>



On 06/11/2014 16:10, Ming Lei wrote:
> +    /* don't submit until next completion for -EAGAIN of non plug case */
> +    if (unlikely(!s->io_q.plugged)) {
> +        return 0;
> +    }
> +

Is this an optimization or a fix for something?

> +        /*
> +         * Switch to queue mode until -EAGAIN is handled, we suppose
> +         * there is always uncompleted I/O, so try to enqueue it first,
> +         * and will be submitted again in following aio completion cb.
> +         */
> +        if (ret == -EAGAIN) {
> +            goto enqueue;
> +        } else if (ret < 0) {
>              goto out_free_aiocb;
>          }

Better:

     if (!s->io_q.plugged && !s->io_q.idx) {
        ret = io_submit(s->ctx, 1, &iocbs);
        if (ret >= 0) {
            return &laiocb->common;
        }
        if (ret != -EAGAIN) {
            goto out_free_aiocb;
        }
    }

    /* code for queue mode.  */

Paolo