linux-kernel - Re: [PATCH 0/8] rcu: Fix expedited GP deadlock (and cleanup some nocb stuff)

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

Message-ID: <ZXdrNJFCaXAFMITp@localhost.localdomain>
Date:   Mon, 11 Dec 2023 21:04:04 +0100
From:   Frederic Weisbecker <frederic@...nel.org>
To:     "Paul E. McKenney" <paulmck@...nel.org>
Cc:     LKML <linux-kernel@...r.kernel.org>,
        Boqun Feng <boqun.feng@...il.com>,
        Joel Fernandes <joel@...lfernandes.org>,
        Neeraj Upadhyay <neeraj.upadhyay@....com>,
        Uladzislau Rezki <urezki@...il.com>,
        Zqiang <qiang.zhang1211@...il.com>, rcu <rcu@...r.kernel.org>
Subject: Re: [PATCH 0/8] rcu: Fix expedited GP deadlock (and cleanup some
 nocb stuff)

Le Mon, Dec 11, 2023 at 08:38:59AM -0800, Paul E. McKenney a écrit :
> On Fri, Dec 08, 2023 at 11:05:37PM +0100, Frederic Weisbecker wrote:
> > TREE04 can trigger a writer stall if run with memory pressure. This
> > is due to a circular dependency between waiting for expedited grace
> > period and polling on expedited grace period when workqueues go back
> > to mayday serialization.
> > 
> > Here is a proposal fix.
> 
> The torture.sh "acceptance test" with KCSAN and --duration 30 ran
> fine except for this in TREE09:
> 
> kernel/rcu/tree_nocb.h:1785:13: error: unused function '__call_rcu_nocb_wake' [-Werror,-Wunused-function]
> 
> My guess is that the declaration of __call_rcu_nocb_wake() in
> kernel/rcu/tree.h needs an "#ifdef CONFIG_SMP", but you might have a
> better fix.

Could be because if CONFIG_RCU_NO_CB_CPU=n, the function is only called
(though as dead code) from rcutree_migrate_callbacks() which in turn only
exists if CONFIG_HOTPLUG_CPU=y.

Something like that then:

diff --git a/kernel/rcu/tree.h b/kernel/rcu/tree.h
index 35f7af331e6c..e1ff53d5084c 100644
--- a/kernel/rcu/tree.h
+++ b/kernel/rcu/tree.h
@@ -445,6 +445,8 @@ static void rcu_qs(void);
 static int rcu_preempt_blocked_readers_cgp(struct rcu_node *rnp);
 #ifdef CONFIG_HOTPLUG_CPU
 static bool rcu_preempt_has_tasks(struct rcu_node *rnp);
+static void __call_rcu_nocb_wake(struct rcu_data *rdp, bool was_empty,
+				 unsigned long flags);
 #endif /* #ifdef CONFIG_HOTPLUG_CPU */
 static int rcu_print_task_exp_stall(struct rcu_node *rnp);
 static void rcu_preempt_check_blocked_tasks(struct rcu_node *rnp);
@@ -466,8 +468,6 @@ static bool rcu_nocb_flush_bypass(struct rcu_data *rdp, struct rcu_head *rhp,
 				  unsigned long j, bool lazy);
 static void call_rcu_nocb(struct rcu_data *rdp, struct rcu_head *head,
 			  rcu_callback_t func, unsigned long flags, bool lazy);
-static void __call_rcu_nocb_wake(struct rcu_data *rdp, bool was_empty,
-				 unsigned long flags);
 static int rcu_nocb_need_deferred_wakeup(struct rcu_data *rdp, int level);
 static bool do_nocb_deferred_wakeup(struct rcu_data *rdp);
 static void rcu_boot_init_nocb_percpu_data(struct rcu_data *rdp);