From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([208.118.235.92]:43869)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <ronniesahlberg@gmail.com>) id 1SWlwc-0007uD-8S
	for qemu-devel@nongnu.org; Tue, 22 May 2012 06:04:04 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <ronniesahlberg@gmail.com>) id 1SWlwV-0006DR-TE
	for qemu-devel@nongnu.org; Tue, 22 May 2012 06:03:57 -0400
Received: from mail-ob0-f173.google.com ([209.85.214.173]:39194)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <ronniesahlberg@gmail.com>) id 1SWlwV-0006Cr-JZ
	for qemu-devel@nongnu.org; Tue, 22 May 2012 06:03:51 -0400
Received: by obbwd20 with SMTP id wd20so11775220obb.4
	for <qemu-devel@nongnu.org>; Tue, 22 May 2012 03:03:49 -0700 (PDT)
MIME-Version: 1.0
In-Reply-To: <CAN05THSfrtm1NGQRANiUgj-sihzuOwudu3CPduKAevhf1Cndng@mail.gmail.com>
References: <CAN05THQg76tu66DXJhtCu-DQ-6iVfLdzdGcbh0y+=oD0FQ+_4w@mail.gmail.com>
	<4FBA1CC4.7020009@redhat.com>
	<CAN05THQRmdJvNt8XyOYDMu1gqAbWGFE9Z2d6pKEy-s-iS3Q98g@mail.gmail.com>
	<CAJSP0QXOEuYEvXwdTRw0yu-KyDmc6VUP20depv-n+K3BwifjOg@mail.gmail.com>
	<CAN05THQEhaeD8_-5mk_8Hm_-xGRtQorUK_HbUt-+Ozc-s+Px-w@mail.gmail.com>
	<CAN05THTh4oQeo0cRZ8bjY4i7L8TNNsY0mNJuDyjJzW7NBZ_V9g@mail.gmail.com>
	<4FBB5C91.4060504@redhat.com>
	<CAN05THSfrtm1NGQRANiUgj-sihzuOwudu3CPduKAevhf1Cndng@mail.gmail.com>
Date: Tue, 22 May 2012 20:03:49 +1000
Message-ID: <CAN05THR5Wt5RPkyNfddQ+8HM8dVYAXk39nDvqk0_qr==4r-1wg@mail.gmail.com>
From: ronnie sahlberg <ronniesahlberg@gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: quoted-printable
Subject: Re: [Qemu-devel] Weird iscsi/fd-event issue since recent merge of
 event system changes
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: Paolo Bonzini <pbonzini@redhat.com>
Cc: Kevin Wolf <kwolf@redhat.com>, Stefan Hajnoczi <stefanha@gmail.com>, qemu-devel <qemu-devel@nongnu.org>, Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>

None of the other drivers in block/*.c call qemu_notify_event()

Do you you think those should be audited and have this call added to
where required ?


regards
ronnie sahlberg

On Tue, May 22, 2012 at 7:48 PM, ronnie sahlberg
<ronniesahlberg@gmail.com> wrote:
> On Tue, May 22, 2012 at 7:29 PM, Paolo Bonzini <pbonzini@redhat.com> wrot=
e:
>> Il 22/05/2012 11:15, ronnie sahlberg ha scritto:
>>> Hi,
>>>
>>> Now that I see what happens, I can easily workaround this in block/iscs=
i.c
>>> by the patch below, but I dont know if this is the right thing to do.
>>>
>>> It does appear that here, =A0when I use qemu_set_fd_handler() and add a
>>> handler for "writeble" it takes 55ms before the event system notices
>>> this and reacts.
>>>
>>>
>>>
>>> diff --git a/block/iscsi.c b/block/iscsi.c
>>> index d37c4ee..1ebff0f 100644
>>> --- a/block/iscsi.c
>>> +++ b/block/iscsi.c
>>> @@ -105,6 +105,10 @@ iscsi_set_events(IscsiLun *iscsilun)
>>> =A0{
>>> =A0 =A0 =A0struct iscsi_context *iscsi =3D iscsilun->iscsi;
>>>
>>> + =A0 =A0if (iscsi_which_events(iscsi) & POLLOUT) {
>>> + =A0 =A0 =A0 =A0iscsi_process_write(iscsilun);
>>> + =A0 =A0}
>>> +
>>> =A0 =A0 =A0qemu_aio_set_fd_handler(iscsi_get_fd(iscsi), iscsi_process_r=
ead,
>>> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 (iscsi_which_ev=
ents(iscsi) & POLLOUT)
>>> =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 =A0 ? iscsi_process=
_write : NULL,
>>
>> Doh, now I remember. =A0Whenever you change the aio handlers you need to
>> call qemu_notify_event() afterwards, if the handler may fire right away.
>
> Thanks. =A0I just confirmed that qemu_notify_event() fixes the issue.
> Ill send a patch that uses qemu_notify_event() and a comment why this is =
needed.
>
>
> regards
> ronnie sahlberg