From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759919Ab1LPPX7 (ORCPT ); Fri, 16 Dec 2011 10:23:59 -0500 Received: from cam-admin0.cambridge.arm.com ([217.140.96.50]:50253 "EHLO cam-admin0.cambridge.arm.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1755763Ab1LPPXz (ORCPT ); Fri, 16 Dec 2011 10:23:55 -0500 Date: Fri, 16 Dec 2011 15:23:34 +0000 From: Catalin Marinas To: Peter Zijlstra Cc: "frank.rowand@am.sony.com" , "tglx@linutronix.de" , "linux-kernel@vger.kernel.org" , "rostedt@goodmis.org" , Linus Torvalds Subject: Re: [PATCH] PREEMPT_RT_FULL: ARM context switch needs IRQs enabled Message-ID: <20111216152334.GD6342@arm.com> References: <4EEAB90D.3050504@am.sony.com> <1324029272.18942.73.camel@twins> <20111216111319.GB6342@arm.com> <20111216114320.GC6342@arm.com> <1324039793.18942.80.camel@twins> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1324039793.18942.80.camel@twins> User-Agent: Mutt/1.5.20 (2009-06-14) Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org On Fri, Dec 16, 2011 at 12:49:53PM +0000, Peter Zijlstra wrote: > On Fri, 2011-12-16 at 11:43 +0000, Catalin Marinas wrote: > > From 26d87e955f089fd246bee29bb388f22da1297e0c Mon Sep 17 00:00:00 2001 > > From: Catalin Marinas > > Date: Fri, 16 Dec 2011 11:32:26 +0000 > > Subject: [PATCH] sched, mm: Use activate_mm() instead of switch_mm() > > > > The ARM port tries to remove __ARCH_WANT_INTERRUPTS_ON_CTXSW. Since the > > actual pgd switching requires interrupts to be enabled on ARM (for > > latency on ARMv5 and earlier and IPIs on ARMv6+), the solution is to > > defer the pgd switching to a post context switch hook that is run with > > interrupts enabled. There are however two additional direct calls to > > switch_mm() without the additional post-switch hook and ARM would fail > > to set the new pgd. > > > > This patch changes the switch_mm() call with activate_mm() which ensures > > that the required pgd has been set. The activate_mm() function must be > > called with interrupts enabled. > > > > Signed-off-by: Catalin Marinas > > Cc: Ingo Molnar > > Cc: Peter Zijlstra > > --- > > kernel/sched.c | 2 +- > > mm/mmu_context.c | 2 +- > > 2 files changed, 2 insertions(+), 2 deletions(-) > > > > diff --git a/kernel/sched.c b/kernel/sched.c > > index 7b46a39..3976157 100644 > > --- a/kernel/sched.c > > +++ b/kernel/sched.c > > @@ -6310,7 +6310,7 @@ void idle_task_exit(void) > > BUG_ON(cpu_online(smp_processor_id())); > > > > if (mm != &init_mm) > > - switch_mm(mm, &init_mm, current); > > + activate_mm(mm, &init_mm); > > mmdrop(mm); > > } > > > > diff --git a/mm/mmu_context.c b/mm/mmu_context.c > > index cf332bc..4e44ac4 100644 > > --- a/mm/mmu_context.c > > +++ b/mm/mmu_context.c > > @@ -32,7 +32,7 @@ void use_mm(struct mm_struct *mm) > > tsk->active_mm = mm; > > } > > tsk->mm = mm; > > - switch_mm(active_mm, mm, tsk); > > + activate_mm(active_mm, mm); > > task_unlock(tsk); > > > > if (active_mm != mm) > > > > alpha, cris, microblaze, powerpc appear to actually use that task > argument of switch_mm(), so I'm not convinced this is actually correct. We use it on ARM as well but in both cases above the task was current. However, I can see that the activate_mm() implementation on other architectures doesn't just call switch_mm(current), so the two may not be functionally equivalent. An alternative would be to only defer the pgd switching on ARM if irqs_disabled() and leave the generic code untouched (well, apart from the post-switch hook). -- Catalin