All of lore.kernel.org
 help / color / mirror / Atom feed
From: Peter Zijlstra <a.p.zijlstra@chello.nl>
To: Linus Torvalds <torvalds@linux-foundation.org>,
	Ingo Molnar <mingo@elte.hu>, Thomas Gleixner <tglx@linutronix.de>
Cc: linux-kernel@vger.kernel.org, akpm@linux-foundation.org,
	efault@gmx.de, Peter Zijlstra <a.p.zijlstra@chello.nl>
Subject: [PATCH 3/4] printk, lockdep: Remove lockdep_off() usage
Date: Tue, 21 Jun 2011 17:17:28 +0200	[thread overview]
Message-ID: <20110621153806.236387567@chello.nl> (raw)
In-Reply-To: <20110621151725.705140475@chello.nl>

[-- Attachment #1: printk-remove-lockdep_off.patch --]
[-- Type: text/plain, Size: 2641 bytes --]

Remove the lockdep_off() usage from printk(). Also add a
debug_locks_off() call to zap_locks() since that'll mess up the lock
state in a royal way anyway, and zap_locks() when lockdep causes a
printk() recursion so that we can at observe the splat.

Further switch to local_irq_ ops so that the irq state is properly
tracked (raw_local_irq_* isn't tracked by lockdep, causing confusion).
Also drop superfluous preempt_disable(), disabling IRQs already avoids
scheduling.

Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
---
 kernel/printk.c |   11 ++++-------
 1 file changed, 4 insertions(+), 7 deletions(-)

Index: linux-2.6/kernel/printk.c
===================================================================
--- linux-2.6.orig/kernel/printk.c
+++ linux-2.6/kernel/printk.c
@@ -686,6 +686,7 @@ static void zap_locks(void)
 
 	oops_timestamp = jiffies;
 
+	debug_locks_off();
 	/* If a crash is occurring, make sure we can't deadlock */
 	spin_lock_init(&logbuf_lock);
 	/* And make sure that we print immediately */
@@ -838,9 +839,8 @@ asmlinkage int vprintk(const char *fmt, 
 	boot_delay_msec();
 	printk_delay();
 
-	preempt_disable();
 	/* This stops the holder of console_sem just where we want him */
-	raw_local_irq_save(flags);
+	local_irq_save(flags);
 	this_cpu = smp_processor_id();
 
 	/*
@@ -854,14 +854,13 @@ asmlinkage int vprintk(const char *fmt, 
 		 * recursion and return - but flag the recursion so that
 		 * it can be printed at the next appropriate moment:
 		 */
-		if (!oops_in_progress) {
+		if (!oops_in_progress && !lockdep_recursing(current)) {
 			recursion_bug = 1;
 			goto out_restore_irqs;
 		}
 		zap_locks();
 	}
 
-	lockdep_off();
 	spin_lock(&logbuf_lock);
 	printk_cpu = this_cpu;
 
@@ -958,11 +957,9 @@ asmlinkage int vprintk(const char *fmt, 
 	if (console_trylock_for_printk(this_cpu))
 		console_unlock();
 
-	lockdep_on();
 out_restore_irqs:
-	raw_local_irq_restore(flags);
+	local_irq_restore(flags);
 
-	preempt_enable();
 	return printed_len;
 }
 EXPORT_SYMBOL(printk);
diff --git a/include/linux/lockdep.h b/include/linux/lockdep.h
index ef820a3..132d210 100644
--- a/include/linux/lockdep.h
+++ b/include/linux/lockdep.h
@@ -343,6 +343,8 @@ extern void lockdep_trace_alloc(gfp_t mask);
 
 #define lockdep_assert_held(l)	WARN_ON(debug_locks && !lockdep_is_held(l))
 
+#define lockdep_recursing(tsk)	((tsk)->lockdep_recursion)
+
 #else /* !LOCKDEP */
 
 static inline void lockdep_off(void)
@@ -392,6 +394,8 @@ struct lock_class_key { };
 
 #define lockdep_assert_held(l)			do { } while (0)
 
+#define lockdep_recursing(tsk)			(0)
+
 #endif /* !LOCKDEP */
 
 #ifdef CONFIG_LOCK_STAT



  parent reply	other threads:[~2011-06-21 15:40 UTC|newest]

Thread overview: 10+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-06-21 15:17 [PATCH 0/4] printk: Remove lockdep_off() and wakeups -v3 Peter Zijlstra
2011-06-21 15:17 ` [PATCH 1/4] printk: Release console_sem after logbuf_lock Peter Zijlstra
2011-06-21 15:17 ` [PATCH 2/4] lockdep: Fix trace_{soft,hard}irqs_{on,off}() recursion Peter Zijlstra
2011-06-22 15:13   ` [tip:core/printk] lockdep: Fix trace_[soft,hard]irqs_[on,off]() recursion tip-bot for Peter Zijlstra
2011-06-21 15:17 ` Peter Zijlstra [this message]
2011-06-21 15:17 ` [PATCH 4/4] printk: Avoid all wakeups from printk Peter Zijlstra
2011-06-22  9:20 ` [PATCH 0/4] printk: Remove lockdep_off() and wakeups -v3 Peter Zijlstra
2011-06-22 11:50   ` Ingo Molnar
2011-06-22 13:44     ` Peter Zijlstra
2011-06-22 14:22   ` [tip:core/printk] printk: Fix console_sem vs logbuf_lock unlock race tip-bot for Peter Zijlstra

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=20110621153806.236387567@chello.nl \
    --to=a.p.zijlstra@chello.nl \
    --cc=akpm@linux-foundation.org \
    --cc=efault@gmx.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    --subject='Re: [PATCH 3/4] printk, lockdep: Remove lockdep_off() usage' \
    /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

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.