linux-kernel - Fix for 10-second delay bug for -stable

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [day] [month] [year] [list]

Date:   Wed, 31 Oct 2018 15:45:15 -0700
From:   "Paul E. McKenney" <paulmck@...ux.ibm.com>
To:     mhillenb@...zon.de, dwmw2@...radead.org
Cc:     peterz@...radead.org, linux-kernel@...r.kernel.org
Subject: Fix for 10-second delay bug for -stable

Hello!

I have lightly tested the following backport of 92aa39e9dc77 ("rcu:
Make need_resched() respond to urgent RCU-QS needs") on v4.12-v4.19.
Does it look reasonable from your viewpoint?

If I don't hear otherwise, I will send it along to -stable at the end
of this coming weekend, Pacific Time.

This patch does not apply to v4.11 and earlier.  I will therefore
ignore those releases unless someone reproduces the problem on them.

							Thanx, Paul

------------------------------------------------------------------------

commit 0a555b6d58f80dcd93a116c5b197208aeaf9f7ef
Author: Paul E. McKenney <paulmck@...ux.vnet.ibm.com>
Date:   Mon Jul 9 13:47:30 2018 -0700

    rcu: Make need_resched() respond to urgent RCU-QS needs
    
    commit 92aa39e9dc77 upstream.
    
    The per-CPU rcu_dynticks.rcu_urgent_qs variable communicates an urgent
    need for an RCU quiescent state from the force-quiescent-state processing
    within the grace-period kthread to context switches and to cond_resched().
    Unfortunately, such urgent needs are not communicated to need_resched(),
    which is sometimes used to decide when to invoke cond_resched(), for
    but one example, within the KVM vcpu_run() function.  As of v4.15, this
    can result in synchronize_sched() being delayed by up to ten seconds,
    which can be problematic, to say nothing of annoying.
    
    This commit therefore checks rcu_dynticks.rcu_urgent_qs from within
    rcu_check_callbacks(), which is invoked from the scheduling-clock
    interrupt handler.  If the current task is not an idle task and is
    not executing in usermode, a context switch is forced, and either way,
    the rcu_dynticks.rcu_urgent_qs variable is set to false.  If the current
    task is an idle task, then RCU's dyntick-idle code will detect the
    quiescent state, so no further action is required.  Similarly, if the
    task is executing in usermode, other code in rcu_check_callbacks() and
    its called functions will report the corresponding quiescent state.
    
    Reported-by: Marius Hillenbrand <mhillenb@...zon.de>
    Reported-by: David Woodhouse <dwmw2@...radead.org>
    Suggested-by: Peter Zijlstra <peterz@...radead.org>
    Signed-off-by: Paul E. McKenney <paulmck@...ux.vnet.ibm.com>
    [ paulmck: Backported to make patch apply cleanly on older versions. ]
    Tested-by: Marius Hillenbrand <mhillenb@...zon.de>
    Cc: <stable@...r.kernel.org> # 4.12.x - 4.19.x

diff --git a/kernel/rcu/tree.c b/kernel/rcu/tree.c
index 0b760c1369f7..15301ed19da6 100644
--- a/kernel/rcu/tree.c
+++ b/kernel/rcu/tree.c
@@ -2662,6 +2662,15 @@ void rcu_check_callbacks(int user)
 		rcu_bh_qs();
 	}
 	rcu_preempt_check_callbacks();
+	/* The load-acquire pairs with the store-release setting to true. */
+	if (smp_load_acquire(this_cpu_ptr(&rcu_dynticks.rcu_urgent_qs))) {
+		/* Idle and userspace execution already are quiescent states. */
+		if (!rcu_is_cpu_rrupt_from_idle() && !user) {
+			set_tsk_need_resched(current);
+			set_preempt_need_resched();
+		}
+		__this_cpu_write(rcu_dynticks.rcu_urgent_qs, false);
+	}
 	if (rcu_pending())
 		invoke_rcu_core();