public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
* [PATCH] Fix of stack dump in {SOFT|HARD}IRQs
@ 2004-10-01 10:51 Kirill Korotaev
  2004-10-01 22:08 ` Andrew Morton
  0 siblings, 1 reply; 2+ messages in thread
From: Kirill Korotaev @ 2004-10-01 10:51 UTC (permalink / raw)
  To: linux-kernel, Linus Torvalds, Andrew Morton

[-- Attachment #1: Type: text/plain, Size: 270 bytes --]

This patch fixes incorrect check for stack ptr in 
show_trace()->valid_stack_ptr(). When called from hardirq/softirq 
show_trace() prints "Stack pointer is garbage, not printing trace" 
message instead of call traces.

Signed-Off-By: Kirill Korotaev <dev@sw.ru>

Kirill

[-- Attachment #2: diff-dumpstack --]
[-- Type: text/plain, Size: 1472 bytes --]

--- ./arch/i386/kernel/traps.c.dumpstack	2004-10-01 14:21:40.000000000 +0400
+++ ./arch/i386/kernel/traps.c	2004-10-01 14:30:10.109733400 +0400
@@ -95,11 +95,16 @@ static int kstack_depth_to_print = 24;
 
 static int valid_stack_ptr(struct task_struct *task, void *p)
 {
-	if (p <= (void *)task->thread_info)
-		return 0;
-	if (kstack_end(p))
-		return 0;
-	return 1;
+	extern int is_irq_stack_ptr(struct task_struct *, void *);
+
+	if (is_irq_stack_ptr(task, p))
+		return 1;
+	if (p >= (void *)task->thread_info &&
+	    p < (void *)task->thread_info + THREAD_SIZE &&
+	    !kstack_end(p))
+		return 1;
+
+	return 0;
 }
 
 #ifdef CONFIG_FRAME_POINTER
--- ./arch/i386/kernel/irq.c.dumpstack	2004-09-20 14:14:58.000000000 +0400
+++ ./arch/i386/kernel/irq.c	2004-10-01 14:28:15.806110192 +0400
@@ -1126,6 +1126,21 @@ void init_irq_proc (void)
 static char softirq_stack[NR_CPUS * THREAD_SIZE]  __attribute__((__aligned__(THREAD_SIZE)));
 static char hardirq_stack[NR_CPUS * THREAD_SIZE]  __attribute__((__aligned__(THREAD_SIZE)));
 
+int is_irq_stack_ptr(struct task_struct *task, void *p)
+{
+	unsigned long off;
+
+	off = task->thread_info->cpu * THREAD_SIZE;
+	if (p >= (void *)hardirq_stack + off &&
+	    p < (void *)hardirq_stack + off + THREAD_SIZE)
+		return 1;
+	if (p >= (void *)softirq_stack + off &&
+	    p < (void *)softirq_stack + off + THREAD_SIZE)
+		return 1;
+
+	return 0;
+}
+
 /*
  * allocate per-cpu stacks for hardirq and for softirq processing
  */

^ permalink raw reply	[flat|nested] 2+ messages in thread

end of thread, other threads:[~2004-10-01 22:09 UTC | newest]

Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2004-10-01 10:51 [PATCH] Fix of stack dump in {SOFT|HARD}IRQs Kirill Korotaev
2004-10-01 22:08 ` Andrew Morton

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox