From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1753186AbaIHGlh (ORCPT ); Mon, 8 Sep 2014 02:41:37 -0400 Received: from terminus.zytor.com ([198.137.202.10]:51530 "EHLO terminus.zytor.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753081AbaIHGld (ORCPT ); Mon, 8 Sep 2014 02:41:33 -0400 Date: Sun, 7 Sep 2014 23:39:57 -0700 From: tip-bot for Rik van Riel Message-ID: Cc: mingo@kernel.org, torvalds@linux-foundation.org, peterz@infradead.org, guillaume@morinfr.org, mschmidt@redhat.com, riel@redhat.com, akpm@linux-foundation.org, oleg@redhat.com, tglx@linutronix.de, vdavydov@parallels.com, yangds.fnst@cn.fujitsu.com, rientjes@google.com, linux-kernel@vger.kernel.org, hpa@zytor.com, daeseok.youn@gmail.com, athorlton@sgi.com, geert@linux-m68k.org, keescook@chromium.org, ionut.m.alexa@gmail.com, mhocko@suse.cz, lizefan@huawei.com Reply-To: mingo@kernel.org, torvalds@linux-foundation.org, peterz@infradead.org, guillaume@morinfr.org, mschmidt@redhat.com, riel@redhat.com, akpm@linux-foundation.org, tglx@linutronix.de, oleg@redhat.com, rientjes@google.com, yangds.fnst@cn.fujitsu.com, vdavydov@parallels.com, linux-kernel@vger.kernel.org, hpa@zytor.com, daeseok.youn@gmail.com, keescook@chromium.org, geert@linux-m68k.org, athorlton@sgi.com, ionut.m.alexa@gmail.com, mhocko@suse.cz, lizefan@huawei.com In-Reply-To: <20140816134010.26a9b572@annuminas.surriel.com> References: <20140816134010.26a9b572@annuminas.surriel.com> To: linux-tip-commits@vger.kernel.org Subject: [tip:sched/core] time, signal: Protect resource use statistics with seqlock Git-Commit-ID: e78c3496790ee8a36522a838b59b388e8a709e65 X-Mailer: tip-git-log-daemon Robot-ID: Robot-Unsubscribe: Contact to get blacklisted from these emails MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=UTF-8 Content-Disposition: inline Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Commit-ID: e78c3496790ee8a36522a838b59b388e8a709e65 Gitweb: http://git.kernel.org/tip/e78c3496790ee8a36522a838b59b388e8a709e65 Author: Rik van Riel AuthorDate: Sat, 16 Aug 2014 13:40:10 -0400 Committer: Ingo Molnar CommitDate: Mon, 8 Sep 2014 08:17:01 +0200 time, signal: Protect resource use statistics with seqlock Both times() and clock_gettime(CLOCK_PROCESS_CPUTIME_ID) have scalability issues on large systems, due to both functions being serialized with a lock. The lock protects against reporting a wrong value, due to a thread in the task group exiting, its statistics reporting up to the signal struct, and that exited task's statistics being counted twice (or not at all). Protecting that with a lock results in times() and clock_gettime() being completely serialized on large systems. This can be fixed by using a seqlock around the events that gather and propagate statistics. As an additional benefit, the protection code can be moved into thread_group_cputime(), slightly simplifying the calling functions. In the case of posix_cpu_clock_get_task() things can be simplified a lot, because the calling function already ensures that the task sticks around, and the rest is now taken care of in thread_group_cputime(). This way the statistics reporting code can run lockless. Signed-off-by: Rik van Riel Signed-off-by: Peter Zijlstra (Intel) Cc: Alex Thorlton Cc: Andrew Morton Cc: Daeseok Youn Cc: David Rientjes Cc: Dongsheng Yang Cc: Geert Uytterhoeven Cc: Guillaume Morin Cc: Ionut Alexa Cc: Kees Cook Cc: Linus Torvalds Cc: Li Zefan Cc: Michal Hocko Cc: Michal Schmidt Cc: Oleg Nesterov Cc: Vladimir Davydov Cc: umgwanakikbuti@gmail.com Cc: fweisbec@gmail.com Cc: srao@redhat.com Cc: lwoodman@redhat.com Cc: atheurer@redhat.com Link: http://lkml.kernel.org/r/20140816134010.26a9b572@annuminas.surriel.com Signed-off-by: Ingo Molnar --- include/linux/sched.h | 1 + kernel/exit.c | 4 ++++ kernel/fork.c | 1 + kernel/sched/cputime.c | 33 ++++++++++++++++++++------------- kernel/sys.c | 2 -- kernel/time/posix-cpu-timers.c | 14 -------------- 6 files changed, 26 insertions(+), 29 deletions(-) diff --git a/include/linux/sched.h b/include/linux/sched.h index 5c2c885..dd9eb48 100644 --- a/include/linux/sched.h +++ b/include/linux/sched.h @@ -645,6 +645,7 @@ struct signal_struct { * Live threads maintain their own counters and add to these * in __exit_signal, except for the group leader. */ + seqlock_t stats_lock; cputime_t utime, stime, cutime, cstime; cputime_t gtime; cputime_t cgtime; diff --git a/kernel/exit.c b/kernel/exit.c index b93d46d..fa09b86 100644 --- a/kernel/exit.c +++ b/kernel/exit.c @@ -127,6 +127,7 @@ static void __exit_signal(struct task_struct *tsk) * the signal_struct. */ task_cputime(tsk, &utime, &stime); + write_seqlock(&sig->stats_lock); sig->utime += utime; sig->stime += stime; sig->gtime += task_gtime(tsk); @@ -140,6 +141,7 @@ static void __exit_signal(struct task_struct *tsk) sig->sum_sched_runtime += tsk->se.sum_exec_runtime; sig->nr_threads--; __unhash_process(tsk, group_dead); + write_sequnlock(&sig->stats_lock); /* * Do this under ->siglock, we can race with another thread @@ -1042,6 +1044,7 @@ static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) spin_lock_irq(&p->real_parent->sighand->siglock); psig = p->real_parent->signal; sig = p->signal; + write_seqlock(&psig->stats_lock); psig->cutime += tgutime + sig->cutime; psig->cstime += tgstime + sig->cstime; psig->cgtime += task_gtime(p) + sig->gtime + sig->cgtime; @@ -1064,6 +1067,7 @@ static int wait_task_zombie(struct wait_opts *wo, struct task_struct *p) psig->cmaxrss = maxrss; task_io_accounting_add(&psig->ioac, &p->ioac); task_io_accounting_add(&psig->ioac, &sig->ioac); + write_sequnlock(&psig->stats_lock); spin_unlock_irq(&p->real_parent->sighand->siglock); } diff --git a/kernel/fork.c b/kernel/fork.c index 0cf9cdb..9387ae8 100644 --- a/kernel/fork.c +++ b/kernel/fork.c @@ -1068,6 +1068,7 @@ static int copy_signal(unsigned long clone_flags, struct task_struct *tsk) sig->curr_target = tsk; init_sigpending(&sig->shared_pending); INIT_LIST_HEAD(&sig->posix_timers); + seqlock_init(&sig->stats_lock); hrtimer_init(&sig->real_timer, CLOCK_MONOTONIC, HRTIMER_MODE_REL); sig->real_timer.function = it_real_fn; diff --git a/kernel/sched/cputime.c b/kernel/sched/cputime.c index 3e52836..49b7cfe 100644 --- a/kernel/sched/cputime.c +++ b/kernel/sched/cputime.c @@ -288,18 +288,28 @@ void thread_group_cputime(struct task_struct *tsk, struct task_cputime *times) struct signal_struct *sig = tsk->signal; cputime_t utime, stime; struct task_struct *t; - - times->utime = sig->utime; - times->stime = sig->stime; - times->sum_exec_runtime = sig->sum_sched_runtime; + unsigned int seq, nextseq; rcu_read_lock(); - for_each_thread(tsk, t) { - task_cputime(t, &utime, &stime); - times->utime += utime; - times->stime += stime; - times->sum_exec_runtime += task_sched_runtime(t); - } + /* Attempt a lockless read on the first round. */ + nextseq = 0; + do { + seq = nextseq; + read_seqbegin_or_lock(&sig->stats_lock, &seq); + times->utime = sig->utime; + times->stime = sig->stime; + times->sum_exec_runtime = sig->sum_sched_runtime; + + for_each_thread(tsk, t) { + task_cputime(t, &utime, &stime); + times->utime += utime; + times->stime += stime; + times->sum_exec_runtime += task_sched_runtime(t); + } + /* If lockless access failed, take the lock. */ + nextseq = 1; + } while (need_seqretry(&sig->stats_lock, seq)); + done_seqretry(&sig->stats_lock, seq); rcu_read_unlock(); } @@ -611,9 +621,6 @@ void task_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st) cputime_adjust(&cputime, &p->prev_cputime, ut, st); } -/* - * Must be called with siglock held. - */ void thread_group_cputime_adjusted(struct task_struct *p, cputime_t *ut, cputime_t *st) { struct task_cputime cputime; diff --git a/kernel/sys.c b/kernel/sys.c index ce81291..b663664 100644 --- a/kernel/sys.c +++ b/kernel/sys.c @@ -862,11 +862,9 @@ void do_sys_times(struct tms *tms) { cputime_t tgutime, tgstime, cutime, cstime; - spin_lock_irq(¤t->sighand->siglock); thread_group_cputime_adjusted(current, &tgutime, &tgstime); cutime = current->signal->cutime; cstime = current->signal->cstime; - spin_unlock_irq(¤t->sighand->siglock); tms->tms_utime = cputime_to_clock_t(tgutime); tms->tms_stime = cputime_to_clock_t(tgstime); tms->tms_cutime = cputime_to_clock_t(cutime); diff --git a/kernel/time/posix-cpu-timers.c b/kernel/time/posix-cpu-timers.c index 3b89464..492b986 100644 --- a/kernel/time/posix-cpu-timers.c +++ b/kernel/time/posix-cpu-timers.c @@ -272,22 +272,8 @@ static int posix_cpu_clock_get_task(struct task_struct *tsk, if (same_thread_group(tsk, current)) err = cpu_clock_sample(which_clock, tsk, &rtn); } else { - unsigned long flags; - struct sighand_struct *sighand; - - /* - * while_each_thread() is not yet entirely RCU safe, - * keep locking the group while sampling process - * clock for now. - */ - sighand = lock_task_sighand(tsk, &flags); - if (!sighand) - return err; - if (tsk == current || thread_group_leader(tsk)) err = cpu_clock_sample_group(which_clock, tsk, &rtn); - - unlock_task_sighand(tsk, &flags); } if (!err)