Re: [PATCH 2.6.18-rc4 00/10] Kernel memory leak detector 0.9

[Catalin Marinas]

On 18/08/06, Michal Piotrowski <michal.k.k.piotrowski@xxxxxxxxx> wrote:
> On 18/08/06, Catalin Marinas <catalin.marinas@xxxxxxxxx> wrote:
> > On 18/08/06, Michal Piotrowski <michal.k.k.piotrowski@xxxxxxxxx> wrote:
> > > Lockdep still detects this bug
> >
> > Do you have the complete dmesg output? There should be a different
> > path which I missed (I didn't get this yestarday but I haven't tried
> > with the latest patch I sent to you today).
>
> Here is dmesg output
> http://www.stardust.webpages.pl/files/o_bugs/kmemleak-0.9/kml-dmesg2

Thanks. It is strange - I was under the impression that calling
radix_tree_preload() outside the memleak_lock holding would actually
prevent radix_tree_insert() from allocating a node (and therefore call
kmem_cache_alloc and acquire the list_lock). I'll look at the
radix_tree code this weekend. I'm looking at implementing some kind of
RCU mechanism in kmemleak to avoid future problems (might need changed
to radix_tree as well).

--
Catalin
-
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/