From mboxrd@z Thu Jan 1 00:00:00 1970 From: "Paul E. McKenney" Subject: Re: [RFC] Make need_resched() return true when rcu_urgent_qs requested Date: Thu, 12 Jul 2018 05:53:51 -0700 Message-ID: <20180712125351.GP3593@linux.vnet.ibm.com> References: <20180709220823.GA18045@linux.vnet.ibm.com> <1531319025.8759.57.camel@infradead.org> <20180711144303.GQ3593@linux.vnet.ibm.com> <20180711164952.GA29994@linux.vnet.ibm.com> <9ee4d6fd-02e2-2c73-36a7-36ef4f6413b0@de.ibm.com> <20180711201759.GB3593@linux.vnet.ibm.com> <1531340384.8759.86.camel@infradead.org> <20180711210828.GD3593@linux.vnet.ibm.com> <1531396842.8759.125.camel@infradead.org> Reply-To: paulmck@linux.vnet.ibm.com Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: 8bit Cc: Christian Borntraeger , Peter Zijlstra , mhillenb@amazon.de, linux-kernel , kvm To: David Woodhouse Return-path: Content-Disposition: inline In-Reply-To: <1531396842.8759.125.camel@infradead.org> Sender: linux-kernel-owner@vger.kernel.org List-Id: kvm.vger.kernel.org On Thu, Jul 12, 2018 at 01:00:42PM +0100, David Woodhouse wrote: > > > On Wed, 2018-07-11 at 14:08 -0700, Paul E. McKenney wrote: > > > > > Also... why in $DEITY's name was the existing > > > rcu_virt_note_context_switch() not actually sufficient? If we had that > > > there, why did we need an additional explicit calls to rcu_all_qs() in > > > the KVM loop, or the more complex fixes to need_resched() which > > > ultimately had the same effect, to avoid ten-second latencies? > > > > My guess is that this was because control passed through the > > rcu_virt_note_context_switch() only once, and then subsequent > > scheduling-clock interrupts bypassed this code. Gah! My guess was instead that the code did a rcu_kvm_enter() going in, but somehow managed to miss the rcu_kvm_exit() going out. But that makes absolutely no sense -- had that happened, rcutorture would likely have screamed bloody murder, loudly and often. No mere near misses! And besides, thus far, -ENOREPRODUCE. :-/ Which indicates that I have an opportunity to improve rcutorture and that this patch was with high probability an innocent bystander. > >   But that is just a guess. > > I need to defer to someone who understands the KVM code better than I do. > > I think it's more likely that we just never happened at all. It's > conditional. From the latest patch iteration (see it being removed): > > @@ -118,12 +118,12 @@ static inline void guest_enter_irqoff(void) >          * one time slice). Lets treat guest mode as quiescent state, just like >          * we do with user-mode execution. >          */ > -       if (!context_tracking_cpu_is_enabled()) > -               rcu_virt_note_context_switch(smp_processor_id()); > +       rcu_kvm_enter(); >  } > > > Given the vmexit overhead, I don't think we can do the currently- > proposed rcu_kvm_enter() thing except for CONFIG_NO_HZ_FULL where it's > really necessary. I'll make that conditional, but probably on the RCU > side. > > Without CONFIG_NO_HZ_FULL, rcu_kvm_exit() can do nothing, and > rcu_kvm_enter() can do rcu_virt_note_context_switch(). > > OK? Makes sense to me! And a big "thank you!" to Christian for testing and analyzing this in a timely fashion!!! Thanx, Paul