From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1755236AbbATNWQ (ORCPT ); Tue, 20 Jan 2015 08:22:16 -0500 Received: from www.linutronix.de ([62.245.132.108]:48746 "EHLO Galois.linutronix.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753965AbbATNWP (ORCPT ); Tue, 20 Jan 2015 08:22:15 -0500 Date: Tue, 20 Jan 2015 14:21:51 +0100 (CET) From: Thomas Gleixner To: Calvin Owens cc: "Paul E. McKenney" , Andrew Morton , Joe Perches , Peter Zijlstra , linux-kernel@vger.kernel.org, kernel-team@fb.com Subject: Re: [PATCH] ksoftirqd: Enable IRQs and call cond_resched() before poking RCU In-Reply-To: <20150113211618.GA21498@mail.thefacebook.com> Message-ID: References: <1420594659-16996-1-git-send-email-calvinowens@fb.com> <20150107014906.GA27996@mail.thefacebook.com> <20150107021926.GT5280@linux.vnet.ibm.com> <20150107165223.GA21555@linux.vnet.ibm.com> <20150108043329.GB27996@mail.thefacebook.com> <20150108045306.GJ5280@linux.vnet.ibm.com> <20150108214644.GC27996@mail.thefacebook.com> <20150113184301.GO9719@linux.vnet.ibm.com> <20150113211618.GA21498@mail.thefacebook.com> User-Agent: Alpine 2.11 (DEB 23 2013-08-11) MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Linutronix-Spam-Score: -1.0 X-Linutronix-Spam-Level: - X-Linutronix-Spam-Status: No , -1.0 points, 5.0 required, ALL_TRUSTED=-1,SHORTCIRCUIT=-0.0001 Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Tue, 13 Jan 2015, Calvin Owens wrote: > While debugging an issue with excessive softirq usage, I encountered the > following note in commit 3e339b5dae24a706 ("softirq: Use hotplug thread > infrastructure"): > > [ paulmck: Call rcu_note_context_switch() with interrupts enabled. ] > > ...but despite this note, the patch still calls RCU with IRQs disabled. > > This seemingly innocuous change caused a significant regression in softirq > CPU usage on the sending side of a large TCP transfer (~1 GB/s): when > introducing 0.01% packet loss, the softirq usage would jump to around 25%, > spiking as high as 50%. Before the change, the usage would never exceed 5%. > > Moving the call to rcu_note_context_switch() after the cond_sched() call, > as it was originally before the hotplug patch, completely eliminated this > problem. > > Signed-off-by: Calvin Owens > Cc: stable@vger.kernel.org > --- > This version includes the "cpu" argument to rcu_note_context_switch() in > order to apply cleanly to stable kernels. It will need to be removed to > apply to 3.18+ and 3.19 (upstream commit 38200cf2 removed the argument). > > kernel/softirq.c | 6 +++++- > 1 file changed, 5 insertions(+), 1 deletion(-) > > diff --git a/kernel/softirq.c b/kernel/softirq.c > index 501baa9..9e787d8 100644 > --- a/kernel/softirq.c > +++ b/kernel/softirq.c > @@ -656,9 +656,13 @@ static void run_ksoftirqd(unsigned int cpu) > * in the task stack here. > */ > __do_softirq(); > - rcu_note_context_switch(cpu); > local_irq_enable(); > cond_resched(); > + > + preempt_disable(); > + rcu_note_context_switch(cpu); > + preempt_enable(); > + The whole rcu_note_context_switch() in run_ksoftirqd() is silly. cond_resched() __preempt_count_add(PREEMPT_ACTIVE); __schedule(); preempt_disable(); rcu_note_context_switch(); .... __preempt_count_sub(PREEMPT_ACTIVE); Thanks, tglx