Re: [PATCH 2.6.17-rc6 7/9] Remove some of the kmemleak false positives

[Peter Zijlstra <a.p.zijlstra@chello.nl>]

On Mon, 2006-06-12 at 09:11 +0100, Catalin Marinas wrote:
> On 12/06/06, Pekka Enberg <penberg@cs.helsinki.fi> wrote:
> > Hi Catalin,
> >
> > On 6/11/06, Catalin Marinas <catalin.marinas@gmail.com> wrote:
> > > There are allocations for which the main pointer cannot be found but they
> > > are not memory leaks. This patch fixes some of them.
> >
> > Can we fix this by looking for pointers to anywhere in the allocated
> > memory block instead of just looking for the start?
> 
> I thought about this as well (I think that's how Valgrind works) but
> it would increase the chances of missing real leaks. It currently
> looks for the start of the block and a few locations inside the block
> (those from which the main pointer is computed using the
> container_of() macro).
> 
> I need to do some tests to see how it works but I won't be able to use
> the radix_tree (as storing each location in the block would lead to a
> huge tree).

A radix-priority-search-tree would allow to store intervals and query
addresses.

Peter