Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1030820AbdIZUpg (ORCPT ); Tue, 26 Sep 2017 16:45:36 -0400 Received: from mail.efficios.com ([167.114.142.141]:48807 "EHLO mail.efficios.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1030665AbdIZUpf (ORCPT ); Tue, 26 Sep 2017 16:45:35 -0400 Date: Tue, 26 Sep 2017 20:46:30 +0000 (UTC) From: Mathieu Desnoyers To: "Paul E. McKenney" , Peter Zijlstra , Ingo Molnar , Thomas Gleixner , "H. Peter Anvin" Cc: linux-kernel , Boqun Feng , Andrew Hunter , maged michael , gromer , Avi Kivity , Benjamin Herrenschmidt , Paul Mackerras , Michael Ellerman , Dave Watson , x86@kernel.org Message-ID: <1068963628.19295.1506458790412.JavaMail.zimbra@efficios.com> In-Reply-To: <20170926175151.14264-3-mathieu.desnoyers@efficios.com> References: <20170926175151.14264-1-mathieu.desnoyers@efficios.com> <20170926175151.14264-3-mathieu.desnoyers@efficios.com> Subject: Re: [PATCH v4 for 4.14 3/3] membarrier: Document scheduler barrier requirements MIME-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 7bit X-Originating-IP: [167.114.142.141] X-Mailer: Zimbra 8.7.11_GA_1854 (ZimbraWebClient - FF52 (Linux)/8.7.11_GA_1854) Thread-Topic: membarrier: Document scheduler barrier requirements Thread-Index: lQ9SqjUhkXRoMN33GQFfFkjZZT37CA== Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 6137 Lines: 159 ----- On Sep 26, 2017, at 1:51 PM, Mathieu Desnoyers mathieu.desnoyers@efficios.com wrote: > Document the membarrier requirement on having a full memory barrier in > __schedule() after coming from user-space, before storing to rq->curr. > It is provided by smp_mb__after_spinlock() in __schedule(). Missed a few maintainers that should have been CC'd. Adding them now. This patch is aimed to go through Paul E. McKenney's tree. Thanks, Mathieu > > Document that membarrier requires a full barrier on transition from > kernel thread to userspace thread. We currently have an implicit barrier > from atomic_dec_and_test() in mmdrop() that ensures this. > > The x86 switch_mm_irqs_off() full barrier is currently provided by many > cpumask update operations as well as write_cr3(). Document that > write_cr3() provides this barrier. > > Changes since v1: > - Update comments to match reality for code paths which are after > storing to rq->curr, before returning to user-space. > Changes since v2: > - Update changelog (smp_mb__before_spinlock -> smp_mb__after_spinlock). > Changes since v3: > - Clarify comments following feeback from Peter Zijlstra. > > Signed-off-by: Mathieu Desnoyers > CC: Peter Zijlstra > CC: Paul E. McKenney > CC: Boqun Feng > CC: Andrew Hunter > CC: Maged Michael > CC: gromer@google.com > CC: Avi Kivity > CC: Benjamin Herrenschmidt > CC: Paul Mackerras > CC: Michael Ellerman > CC: Dave Watson > --- > arch/x86/mm/tlb.c | 5 +++++ > include/linux/sched/mm.h | 5 +++++ > kernel/sched/core.c | 38 +++++++++++++++++++++++++++----------- > 3 files changed, 37 insertions(+), 11 deletions(-) > > diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c > index 93fe97cce581..5ba86b85953b 100644 > --- a/arch/x86/mm/tlb.c > +++ b/arch/x86/mm/tlb.c > @@ -143,6 +143,11 @@ void switch_mm_irqs_off(struct mm_struct *prev, struct > mm_struct *next, > } > #endif > > + /* > + * The membarrier system call requires a full memory barrier > + * before returning to user-space, after storing to rq->curr. > + * Writing to CR3 provides that full memory barrier. > + */ > if (real_prev == next) { > VM_BUG_ON(this_cpu_read(cpu_tlbstate.ctxs[prev_asid].ctx_id) != > next->context.ctx_id); > diff --git a/include/linux/sched/mm.h b/include/linux/sched/mm.h > index d3b81e48784d..1bd10c2c0893 100644 > --- a/include/linux/sched/mm.h > +++ b/include/linux/sched/mm.h > @@ -38,6 +38,11 @@ static inline void mmgrab(struct mm_struct *mm) > extern void __mmdrop(struct mm_struct *); > static inline void mmdrop(struct mm_struct *mm) > { > + /* > + * The implicit full barrier implied by atomic_dec_and_test is > + * required by the membarrier system call before returning to > + * user-space, after storing to rq->curr. > + */ > if (unlikely(atomic_dec_and_test(&mm->mm_count))) > __mmdrop(mm); > } > diff --git a/kernel/sched/core.c b/kernel/sched/core.c > index b9d731283946..6254f87645de 100644 > --- a/kernel/sched/core.c > +++ b/kernel/sched/core.c > @@ -2649,6 +2649,12 @@ static struct rq *finish_task_switch(struct task_struct > *prev) > finish_arch_post_lock_switch(); > > fire_sched_in_preempt_notifiers(current); > + /* > + * When transitioning from a kernel thread to a userspace > + * thread, mmdrop()'s implicit full barrier is required by the > + * membarrier system call, because the current active_mm can > + * become the current mm without going through switch_mm(). > + */ > if (mm) > mmdrop(mm); > if (unlikely(prev_state == TASK_DEAD)) { > @@ -2754,6 +2760,13 @@ context_switch(struct rq *rq, struct task_struct *prev, > */ > arch_start_context_switch(prev); > > + /* > + * If mm is non-NULL, we pass through switch_mm(). If mm is > + * NULL, we will pass through mmdrop() in finish_task_switch(). > + * Both of these contain the full memory barrier required by > + * membarrier after storing to rq->curr, before returning to > + * user-space. > + */ > if (!mm) { > next->active_mm = oldmm; > mmgrab(oldmm); > @@ -3290,6 +3303,9 @@ static void __sched notrace __schedule(bool preempt) > * Make sure that signal_pending_state()->signal_pending() below > * can't be reordered with __set_current_state(TASK_INTERRUPTIBLE) > * done by the caller to avoid the race with signal_wake_up(). > + * > + * The membarrier system call requires a full memory barrier > + * after coming from user-space, before storing to rq->curr. > */ > rq_lock(rq, &rf); > smp_mb__after_spinlock(); > @@ -3337,17 +3353,17 @@ static void __sched notrace __schedule(bool preempt) > /* > * The membarrier system call requires each architecture > * to have a full memory barrier after updating > - * rq->curr, before returning to user-space. For TSO > - * (e.g. x86), the architecture must provide its own > - * barrier in switch_mm(). For weakly ordered machines > - * for which spin_unlock() acts as a full memory > - * barrier, finish_lock_switch() in common code takes > - * care of this barrier. For weakly ordered machines for > - * which spin_unlock() acts as a RELEASE barrier (only > - * arm64 and PowerPC), arm64 has a full barrier in > - * switch_to(), and PowerPC has > - * smp_mb__after_unlock_lock() before > - * finish_lock_switch(). > + * rq->curr, before returning to user-space. > + * > + * Here are the schemes providing that barrier on the > + * various architectures: > + * - mm ? switch_mm() : mmdrop() for x86, s390, sparc, > + * - finish_lock_switch() for weakly-ordered > + * architectures where spin_unlock is a full barrier, > + * - switch_to() for arm64 (weakly-ordered, spin_unlock > + * is a RELEASE barrier), > + * - membarrier_arch_sched_in() for PowerPC, > + * (weakly-ordered, spin_unlock is a RELEASE barrier). > */ > ++*switch_count; > > -- > 2.11.0 -- Mathieu Desnoyers EfficiOS Inc. http://www.efficios.com