Re: frequent slab corruption (since a long time)

[Dave Jones]

On Wed, Aug 02, 2006 at 05:37:28AM +0200, Andi Kleen wrote:
 > Dave Jones <davej@xxxxxxxxxx> writes:
 > 
 > > Every so often, I see a slab corruption bug reported against
 > > the Fedora kernels (going back as far as 2.6.11), and it's
 > > still plagueing us.
 > > 
 > > It seems to have turned up in a number of different scenarios,
 > > which makes it all the more complicated, but the footprint is
 > > always the same. We write ffffffff00000000 to freed memory.
 > 
 > DEBUG_PAGEALLOC + a small slab patch to force the 2k slab to be
 > only a single object per page (so that a kfree() immediately
 > triggers an unmap) would catch it I guess.

Problem with that approach is that DEBUG_PAGEALLOC makes things
so damned slow that it's pretty much unusable, and this bug
doesn't seem to want to repeat itself to order, so I doubt
many people would put up with the slowdown long enough to chase it down.

		Dave

-- 
http://www.codemonkey.org.uk
-
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/