linux-kernel.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: "Paul E. McKenney" <paulmck@linux.vnet.ibm.com>
To: Milton Miller <miltonm@bga.com>
Cc: Frederic Weisbecker <fweisbec@gmail.com>,
	Ingo Molnar <mingo@elte.hu>,
	Peter Zijlstra <a.p.zijlstra@chello.nl>,
	Thomas Gleixner <tglx@linutronix.de>,
	LKML <linux-kernel@vger.kernel.org>
Subject: Re: [1/4] rcu: Detect uses of rcu read side in extended quiescent states
Date: Mon, 6 Jun 2011 11:10:21 -0700	[thread overview]
Message-ID: <20110606181021.GL3066@linux.vnet.ibm.com> (raw)
In-Reply-To: <rcu-splat-split-printk@mdm.bga.com>

On Sun, Jun 05, 2011 at 10:44:33PM -0500, Milton Miller wrote:
> On Mon, 06 Jun 2011 about 03:10:55 -0000, Frederic Weisbecker wrote:
> > @@ -3994,8 +3995,13 @@ void lockdep_rcu_dereference(const char *file, const int line)
> >  	printk("\n===================================================\n");
> >  	printk(  "[ INFO: suspicious rcu_dereference_check() usage. ]\n");
> >  	printk(  "---------------------------------------------------\n");
> > -	printk("%s:%d invoked rcu_dereference_check() without protection!\n",
> > -			file, line);
> > +	printk("%s:%d invoked rcu_dereference_check() ", file, line);
> > +
> > +	if (type == RCU_WARN_UNPROTECTED)
> > +		printk("without protection!\n");
> > +	else if (type == RCU_WARN_EXT_QS)
> > +		printk("while in RCU extended quiescent state!\n");
> > +
> >  	printk("\nother info that might help us debug this:\n\n");
> >  	printk("\nrcu_scheduler_active = %d, debug_locks = %d\n", rcu_scheduler_active, debug_locks);
> >  	lockdep_print_held_locks(curr);
> 
> Can we keep the above in one printk?   That way the printing is
> guaranteed to come out on one line.  Probably the easiest way would
> be add char *why = "" then assign a string based on the current
> conditions.  Do all of that before the first printk which gets the
> a %s added.

I have the following queued the -rcu tree which does add the string.

Frederic, would it be possible to base on this patch?

							Thanx, Paul

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

commit c15d76f26712bd5228aa0c6af7a7e7c492a812c9
Author: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Date:   Tue May 24 08:31:09 2011 -0700

    rcu: Restore checks for blocking in RCU read-side critical sections
    
    Long ago, using TREE_RCU with PREEMPT would result in "scheduling
    while atomic" diagnostics if you blocked in an RCU read-side critical
    section.  However, PREEMPT now implies TREE_PREEMPT_RCU, which defeats
    this diagnostic.  This commit therefore adds a replacement diagnostic
    based on PROVE_RCU.
    
    Because rcu_lockdep_assert() and lockdep_rcu_dereference() are now being
    used for things that have nothing to do with rcu_dereference(), rename
    lockdep_rcu_dereference() to lockdep_rcu_suspicious() and add a third
    argument that is a string indicating what is suspicious.  This third
    argument is passed in from a new third argument to rcu_lockdep_assert().
    Update all calls to rcu_lockdep_assert() to add an informative third
    argument.
    
    Finally, add a pair of rcu_lockdep_assert() calls from within
    rcu_note_context_switch(), one complaining if a context switch occurs
    in an RCU-bh read-side critical section and another complaining if a
    context switch occurs in an RCU-sched read-side critical section.
    These are present only if the PROVE_RCU kernel parameter is enabled.
    
    Again, you must enable PROVE_RCU to see these new diagnostics.  But you
    are enabling PROVE_RCU to check out new RCU uses in any case, aren't you?
    
    Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>

diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index 4aef1dd..bfa66e7 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -545,7 +545,7 @@ do {									\
 #endif
 
 #ifdef CONFIG_PROVE_RCU
-extern void lockdep_rcu_dereference(const char *file, const int line);
+void lockdep_rcu_suspicious(const char *file, const int line, const char *s);
 #endif
 
 #endif /* __LINUX_LOCKDEP_H */
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 58b13f1..fb2933d 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -297,19 +297,20 @@ extern int rcu_my_thread_group_empty(void);
 /**
  * rcu_lockdep_assert - emit lockdep splat if specified condition not met
  * @c: condition to check
+ * @s: informative message
  */
-#define rcu_lockdep_assert(c)						\
+#define rcu_lockdep_assert(c, s)					\
 	do {								\
 		static bool __warned;					\
 		if (debug_lockdep_rcu_enabled() && !__warned && !(c)) {	\
 			__warned = true;				\
-			lockdep_rcu_dereference(__FILE__, __LINE__);	\
+			lockdep_rcu_suspicious(__FILE__, __LINE__, s);	\
 		}							\
 	} while (0)
 
 #else /* #ifdef CONFIG_PROVE_RCU */
 
-#define rcu_lockdep_assert(c) do { } while (0)
+#define rcu_lockdep_assert(c, s) do { } while (0)
 
 #endif /* #else #ifdef CONFIG_PROVE_RCU */
 
@@ -338,14 +339,16 @@ extern int rcu_my_thread_group_empty(void);
 #define __rcu_dereference_check(p, c, space) \
 	({ \
 		typeof(*p) *_________p1 = (typeof(*p)*__force )ACCESS_ONCE(p); \
-		rcu_lockdep_assert(c); \
+		rcu_lockdep_assert(c, "suspicious rcu_dereference_check()" \
+				      " usage"); \
 		rcu_dereference_sparse(p, space); \
 		smp_read_barrier_depends(); \
 		((typeof(*p) __force __kernel *)(_________p1)); \
 	})
 #define __rcu_dereference_protected(p, c, space) \
 	({ \
-		rcu_lockdep_assert(c); \
+		rcu_lockdep_assert(c, "suspicious rcu_dereference_protected()" \
+				      " usage"); \
 		rcu_dereference_sparse(p, space); \
 		((typeof(*p) __force __kernel *)(p)); \
 	})
@@ -359,7 +362,9 @@ extern int rcu_my_thread_group_empty(void);
 #define __rcu_dereference_index_check(p, c) \
 	({ \
 		typeof(p) _________p1 = ACCESS_ONCE(p); \
-		rcu_lockdep_assert(c); \
+		rcu_lockdep_assert(c, \
+				   "suspicious rcu_dereference_index_check()" \
+				   " usage"); \
 		smp_read_barrier_depends(); \
 		(_________p1); \
 	})
diff --git a/kernel/lockdep.c b/kernel/lockdep.c
index 53a6895..9789028 100644
--- a/kernel/lockdep.c
+++ b/kernel/lockdep.c
@@ -3798,7 +3798,7 @@ void lockdep_sys_exit(void)
 	}
 }
 
-void lockdep_rcu_dereference(const char *file, const int line)
+void lockdep_rcu_suspicious(const char *file, const int line, const char *s)
 {
 	struct task_struct *curr = current;
 
@@ -3807,9 +3807,10 @@ void lockdep_rcu_dereference(const char *file, const int line)
 		return;
 #endif /* #ifdef CONFIG_PROVE_RCU_REPEATEDLY */
 	/* Note: the following can be executed concurrently, so be careful. */
-	printk("\n===================================================\n");
-	printk(  "[ INFO: suspicious rcu_dereference_check() usage. ]\n");
-	printk(  "---------------------------------------------------\n");
+	printk("\n===============================\n");
+	printk(  "[ INFO: suspicious RCU usage. ]\n");
+	printk(  "-------------------------------\n");
+	printk(  "%s\n", s);
 	printk("%s:%d invoked rcu_dereference_check() without protection!\n",
 			file, line);
 	printk("\nother info that might help us debug this:\n\n");
@@ -3818,4 +3819,4 @@ void lockdep_rcu_dereference(const char *file, const int line)
 	printk("\nstack backtrace:\n");
 	dump_stack();
 }
-EXPORT_SYMBOL_GPL(lockdep_rcu_dereference);
+EXPORT_SYMBOL_GPL(lockdep_rcu_suspicious);
diff --git a/kernel/pid.c b/kernel/pid.c
index 57a8346..a7577b3 100644
--- a/kernel/pid.c
+++ b/kernel/pid.c
@@ -419,7 +419,9 @@ EXPORT_SYMBOL(pid_task);
  */
 struct task_struct *find_task_by_pid_ns(pid_t nr, struct pid_namespace *ns)
 {
-	rcu_lockdep_assert(rcu_read_lock_held());
+	rcu_lockdep_assert(rcu_read_lock_held(),
+			   "find_task_by_pid_ns() needs rcu_read_lock()"
+			   " protection");
 	return pid_task(find_pid_ns(nr, ns), PIDTYPE_PID);
 }
 
diff --git a/kernel/rcutree.c b/kernel/rcutree.c
index 88547c8..8b4b3da 100644
--- a/kernel/rcutree.c
+++ b/kernel/rcutree.c
@@ -153,6 +153,12 @@ void rcu_bh_qs(int cpu)
  */
 void rcu_note_context_switch(int cpu)
 {
+	rcu_lockdep_assert(!lock_is_held(&rcu_bh_lock_map),
+			   "Illegal context switch in RCU-bh"
+			   " read-side critical section");
+	rcu_lockdep_assert(!lock_is_held(&rcu_sched_lock_map),
+			   "Illegal context switch in RCU-sched"
+			   " read-side critical section");
 	rcu_sched_qs(cpu);
 	rcu_preempt_note_context_switch(cpu);
 }

  reply	other threads:[~2011-06-06 18:10 UTC|newest]

Thread overview: 29+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-06  3:10 [PATCH 0/4] rcu: Detect rcu uses under extended quiescent state, and fix some Frederic Weisbecker
2011-06-06  3:10 ` Frederic Weisbecker
2011-06-06  3:10 ` [PATCH 1/4] rcu: Detect uses of rcu read side in extended quiescent states Frederic Weisbecker
2011-06-06  3:44   ` [1/4] " Milton Miller
2011-06-06 18:10     ` Paul E. McKenney [this message]
2011-06-06 18:20       ` Frederic Weisbecker
2011-06-06 18:37         ` Paul E. McKenney
2011-06-07  0:19       ` Frederic Weisbecker
2011-06-07  0:42         ` Paul E. McKenney
2011-06-07  1:36           ` Frederic Weisbecker
2011-06-07  4:40             ` Paul E. McKenney
2011-06-07 12:58               ` Frederic Weisbecker
2011-06-07 18:34                 ` Paul E. McKenney
2011-06-07 18:49                   ` Frederic Weisbecker
2011-06-07 19:22                     ` Paul E. McKenney
2011-06-10  8:58             ` Michel Lespinasse
2011-06-06  3:10 ` [PATCH 2/4] nohz: Split extended quiescent state handling from nohz switch Frederic Weisbecker
2011-06-06  3:20   ` [PATCH 2/4 v2] " Frederic Weisbecker
2011-06-06  3:20     ` [PATCH 2/4] " Frederic Weisbecker
2011-06-08  1:15       ` Guan Xuetao
2011-06-06 15:16     ` [PATCH 2/4 v2] " Hans-Christian Egtvedt
2011-06-06 15:24     ` Ralf Baechle
2011-06-06 18:43     ` Mike Frysinger
2011-06-06 20:30     ` Chris Metcalf
2011-06-06  3:58   ` [PATCH 2/4] " David Miller
2011-06-09 23:08   ` Frederic Weisbecker
2011-06-06  3:10 ` [PATCH 3/4] x86: Don't call idle notifier inside rcu extended QS Frederic Weisbecker
2011-06-06  3:10 ` [PATCH 4/4] x86: Call idle_exit() after irq_enter() Frederic Weisbecker
2011-06-06 18:12 ` [PATCH 0/4] rcu: Detect rcu uses under extended quiescent state, and fix some Paul E. McKenney

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20110606181021.GL3066@linux.vnet.ibm.com \
    --to=paulmck@linux.vnet.ibm.com \
    --cc=a.p.zijlstra@chello.nl \
    --cc=fweisbec@gmail.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=miltonm@bga.com \
    --cc=mingo@elte.hu \
    --cc=tglx@linutronix.de \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).