Messages in this thread Patch in this message | | | Date | Mon, 19 May 2008 23:24:48 -0400 | From | Dave Jones <> | Subject | [PATCH] Make |
| |
Arjan noted that the list_head debugging is BUG'ing when it detects corruption. By causing the box to panic immediately, we're possibly losing some bug reports. Changing this to a WARN_ON should mean we at the least start seeing reports collected at kerneloops.org
[ I chose to BUG() when I first added that code, because I was chasing a bug which caused a lockup anyway, so it made little difference to me. ]
Signed-off-by: Dave Jones <davej@redhat.com>
diff --git a/lib/list_debug.c b/lib/list_debug.c index 4350ba9..09f28bf 100644 --- a/lib/list_debug.c +++ b/lib/list_debug.c @@ -24,13 +24,13 @@ void __list_add(struct list_head *new, printk(KERN_ERR "list_add corruption. next->prev should be " "prev (%p), but was %p. (next=%p).\n", prev, next->prev, next); - BUG(); + WARN_ON(1); } if (unlikely(prev->next != next)) { printk(KERN_ERR "list_add corruption. prev->next should be " "next (%p), but was %p. (prev=%p).\n", next, prev->next, prev); - BUG(); + WARN_ON(1); } next->prev = new; new->next = next; @@ -64,12 +64,12 @@ void list_del(struct list_head *entry) if (unlikely(entry->prev->next != entry)) { printk(KERN_ERR "list_del corruption. prev->next should be %p, " "but was %p\n", entry, entry->prev->next); - BUG(); + WARN_ON(1); } if (unlikely(entry->next->prev != entry)) { printk(KERN_ERR "list_del corruption. next->prev should be %p, " "but was %p\n", entry, entry->next->prev); - BUG(); + WARN_ON(1); } __list_del(entry->prev, entry->next); entry->next = LIST_POISON1; -- http://www.codemonkey.org.uk
| |