Re: Debug: sleeping function called from invalid context

[Andrew Morton <akpm@osdl.org>]

Dave Jones <davej@redhat.com> wrote:
>
> How spooky. now I got one too, (minus the noise).
> 
> Call Trace:
>  [<c0120022>] __might_sleep+0x5b/0x5f

It would be useful to know whether this was triggered by in_atomic() or by
irqs_disabled().  We're suspecting the latter.


diff -puN kernel/sched.c~might_sleep-diags kernel/sched.c
--- 25/kernel/sched.c~might_sleep-diags	Mon Aug 18 14:09:41 2003
+++ 25-akpm/kernel/sched.c	Mon Aug 18 14:11:55 2003
@@ -2795,13 +2795,19 @@ void __might_sleep(char *file, int line)
 {
 #if defined(in_atomic)
 	static unsigned long prev_jiffy;	/* ratelimiting */
+	char *msg = NULL;
 
-	if (in_atomic() || irqs_disabled()) {
+	if (in_atomic())
+		msg = "in atomic section";
+	else if (irqs_disabled())
+		msg = "with interrupts disabled";
+
+	if (msg) {
 		if (time_before(jiffies, prev_jiffy + HZ))
 			return;
 		prev_jiffy = jiffies;
-		printk(KERN_ERR "Debug: sleeping function called from invalid"
-				" context at %s:%d\n", file, line);
+		printk(KERN_ERR "Debug: sleeping function "
+				"called %s at %s:%d\n", msg, file, line);
 		dump_stack();
 	}
 #endif

_