From mboxrd@z Thu Jan 1 00:00:00 1970 From: Avi Kivity Subject: Re: [PATCH 4/6] kvm tools: Add rwlock wrapper Date: Sun, 29 May 2011 09:41:23 +0300 Message-ID: <4DE1EA93.6040401@redhat.com> References: <1306426743.3065.34.camel@lappy> <20110526180518.GA3572@elte.hu> <4DDE97CE.4000302@redhat.com> <20110526202531.GA2765@elte.hu> <20110526230508.GA15983@Krystal> <20110527102533.GA24608@elte.hu> <20110527110729.GA26920@elte.hu> <4DE13AF0.2080001@redhat.com> <20110528183259.GA15019@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=ISO-8859-1; format=flowed Content-Transfer-Encoding: 7bit Cc: Mathieu Desnoyers , Pekka Enberg , Sasha Levin , john@jfloren.net, kvm@vger.kernel.org, asias.hejun@gmail.com, gorcunov@gmail.com, prasadjoshi124@gmail.com, "Paul E. McKenney" To: Ingo Molnar Return-path: Received: from mx1.redhat.com ([209.132.183.28]:63453 "EHLO mx1.redhat.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750916Ab1E2Gl7 (ORCPT ); Sun, 29 May 2011 02:41:59 -0400 In-Reply-To: <20110528183259.GA15019@elte.hu> Sender: kvm-owner@vger.kernel.org List-ID: On 05/28/2011 09:32 PM, Ingo Molnar wrote: > * Avi Kivity wrote: > > > > So if you set a notification signal via fcntl(F_SETOWN) on the > > > scheduler context switch event fd, the user-space RCU code will > > > get a signal on every context switch. > > > > Context switches are completely uninteresting for userspace rcu: > > > > rcu_read_lock(); > > ---> context switch > > > > have we learned anything from that? no. User code is always > > preemptible and migratable. If rcu_read_lock() prevented migration > > somehow, then we'd know that a context switch means we've started a > > grace period for this thread. But it doesn't, so we don't. > > Well, in the next mail i mentioned that we can do migration events as > well, which would be useful: instead of having to keep track of > nr_tasks RCU grace periods we could simplify it down to nr_cpus. I don't see how a migration event helps. It is completely transparent from the task's point of view. > But if we indexed by the TID then we wouldnt need any scheduler > bindings at all - this is the simpler approach. Yes, and it maps 1:1 to the kernel implementation (cpu = task). > > What's needed are explicit notifications about grace periods. For > > the vcpu threads, calling KVM_VCPU_RUN seems like a good point. > > For I/O threads, completion of processing of an event is also a > > good point. > > Grace period notifications are needed too, obviously. I'd think they're sufficient, no? Is something else needed? -- I have a truly marvellous patch that fixes the bug which this signature is too narrow to contain.