Re: repeatable slab corruption with LTP msgctl08

[Andrew Morton]

On Thu, 12 Jun 2008 10:35:55 +0200 Andi Kleen <andi@xxxxxxxxxxxxxx> wrote:

> BTW a great way to debug slab corruptions with LTP faster is to run with
> a slab thrasher stress module like http://firstfloor.org/~andi/crasher-26.diff

Well I tried that.  It didn't actually seem to do much (no CPU time
consumed) so I revved it up a bit: 

--- a/drivers/char/crasher.c~crasher-26-speedup
+++ a/drivers/char/crasher.c
@@ -59,7 +59,7 @@ struct mem_buf {
 static unsigned long crasher_random(void)
 {
         rand_seed = rand_seed*69069L+1;
-        return rand_seed^jiffies;
+        return (rand_seed^jiffies) & 3;
 }
 
 void crasher_srandom(unsigned long entropy)
_


But it hasn't crashed after 57 minutes.

I don't think that is how we should fix this bug ;)

I'm pretty much out of time on this one.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/