From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:58284)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1VlyIK-0005mH-Tm
	for qemu-devel@nongnu.org; Thu, 28 Nov 2013 04:54:06 -0500
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1VlyIE-00016k-VP
	for qemu-devel@nongnu.org; Thu, 28 Nov 2013 04:54:00 -0500
Received: from mx1.redhat.com ([209.132.183.28]:55165)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1VlyIE-00016g-Lh
	for qemu-devel@nongnu.org; Thu, 28 Nov 2013 04:53:54 -0500
Message-ID: <529712A1.8090207@redhat.com>
Date: Thu, 28 Nov 2013 10:53:37 +0100
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
References: <52949847.6020908@redhat.com>
	<CAEbWaipAnmoa=gMbB1aNb=btU6LgYXhcnmQWHJ_89m4yvw6Dug@mail.gmail.com>
	<5294A68F.6060301@redhat.com>
	<CAF950W+-UiX6xv4vYmnxji9aWDU8ds2rnx7JugnbHQWJdCCD-Q@mail.gmail.com>
	<5294B461.5000405@redhat.com>
	<5294B634.4050801@cloudius-systems.com>
	<20131126150357.GA20352@redhat.com> <5294BC3B.6070902@redhat.com>
	<D3E216785288A145B7BC975F83A2ED10448BBAB3@SZXEMA510-MBS.china.huawei.com>
	<5297050E.6000700@redhat.com> <20131128091903.GA4609@kernel.org>
	<5297118C.3050104@cloudius-systems.com>
In-Reply-To: <5297118C.3050104@cloudius-systems.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [RFC] create a single workqueue for each vm to
 update vm irq routing table
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: Avi Kivity <avi@cloudius-systems.com>
Cc: "Huangweidong (C)" <weidong.huang@huawei.com>, Gleb Natapov <gleb@redhat.com>, KVM <kvm@vger.kernel.org>, "Michael S. Tsirkin" <mst@redhat.com>, "Zhanghaoyu (A)" <haoyu.zhang@huawei.com>, Luonengjun <luonengjun@huawei.com>, "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>, Zanghongyong <zanghongyong@huawei.com>, Avi Kivity <avi.kivity@gmail.com>, Gleb Natapov <gleb@minantech.com>, "Jinxin (F)" <jinxin712@huawei.com>

Il 28/11/2013 10:49, Avi Kivity ha scritto:
> Linux is safe, it does interrupt migration from within the interrupt
> handler.  If you do that before the device-specific EOI, you won't get
> another interrupt until programming the MSI is complete.
> 
> Is virtio safe? IIRC it can post multiple interrupts without guest acks.
> 
> Using call_rcu() is a better solution than srcu IMO.  Less code changes,
> consistently faster.

call_rcu() has the problem of rate limiting, too.  It wasn't such a
great idea, I think.

The QRCU I linked would work great latency-wise (it has roughly the same
latency of an rwsem but readers are lock-free).  However, the locked
operations in the read path would hurt because of cache misses, so it's
not good either.

Paolo