From mboxrd@z Thu Jan  1 00:00:00 1970
From: Avi Kivity <avi@redhat.com>
Subject: Re: [PATCH v4 2/9] KVM: Switch IRQ subsystem to SRCU
Date: Tue, 09 Nov 2010 12:49:06 +0200
Message-ID: <4CD92722.2060908@redhat.com>
References: <cover.1289215310.git.jan.kiszka@siemens.com> <8cee16e41fc414c185e1d7c170ce44da19797fc1.1289215310.git.jan.kiszka@siemens.com>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Content-Transfer-Encoding: 7bit
Cc: Marcelo Tosatti <mtosatti@redhat.com>, kvm <kvm@vger.kernel.org>,
	Alex Williamson <alex.williamson@redhat.com>,
	"Michael S. Tsirkin" <mst@redhat.com>
To: Jan Kiszka <jan.kiszka@siemens.com>
Return-path: <kvm-owner@vger.kernel.org>
Received: from mx1.redhat.com ([209.132.183.28]:44314 "EHLO mx1.redhat.com"
	rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP
	id S1754830Ab0KIKtM (ORCPT <rfc822;kvm@vger.kernel.org>);
	Tue, 9 Nov 2010 05:49:12 -0500
In-Reply-To: <8cee16e41fc414c185e1d7c170ce44da19797fc1.1289215310.git.jan.kiszka@siemens.com>
Sender: kvm-owner@vger.kernel.org
List-ID: <kvm.vger.kernel.org>

On 11/08/2010 01:21 PM, Jan Kiszka wrote:
> Ack and mask notifiers typically call back into kvm_set_irq, thus may
> iterate over all VCPUs of a VM. Better keep this path preemptible to
> prevent that user-space can massivle influence scheduling latencies. Use
> sleepable RCU for the protection of irq_routing and the notfier lists.
>

What about preemptible RCU, now in mainline?

/*
  * Tree-preemptable RCU implementation for rcu_read_lock().
  * Just increment ->rcu_read_lock_nesting, shared state will be updated
  * if we block.
  */
void __rcu_read_lock(void)
{
     current->rcu_read_lock_nesting++;
     barrier();  /* needed if we ever invoke rcu_read_lock in rcutree.c */
}
EXPORT_SYMBOL_GPL(__rcu_read_lock);

(wow, quite clever in its simplicity)

-- 
error compiling committee.c: too many arguments to function