Re: [PATCH 1/2] lib: more scalable list_sort()

[Artem Bityutskiy <dedekind1@gmail.com>]

On Wed, 2010-08-04 at 17:04 +0300, Artem Bityutskiy wrote:
> On Wed, 2010-01-20 at 20:51 -0800, Don Mullis wrote:
> > The use of list_sort() by UBIFS looks like it could generate long
> > lists; this alternative implementation scales better, reaching ~3x
> > performance gain as list length approaches the L2 cache size.
> > 
> > Stand-alone program timings were run on a Core 2 duo L1=32KB L2=4MB,
> > gcc-4.4, with flags extracted from an Ubuntu kernel build.  Object
> > size is 552 bytes versus 405 for Mark J. Roberts' code.
> > 
> > Worst case for either implementation is a list length just over a POT,
> > and to roughly the same degree, so here are results for a range of
> > 2^N+1 lengths.  List elements were 16 bytes each including malloc
> > overhead; random initial order.
> 
> This patch breaks UBIFS. I did not have time to dig deeper, but the
> symptoms is that list_sort() calls the 'cmp()' function with bogus
> 'struct list_head *a' parameter, which did not exist in the original
> list.

Sorry, it appeared to be that UBIFS 'cmp()' function is broken, so your
patches revealed the issue. Sorry for noise and thanks for revealing
UBIFS problem!

-- 
Best Regards,
Artem Bityutskiy (Артём Битюцкий)