On Mon, Jul 16, 2007 at 04:01:15PM +1000, Nick Piggin wrote:
> Matt Mackall wrote:
> >The version of SLOB in -mm always scans its free list from the
> >beginning, which results in small allocations and free segments
> >clustering at the beginning of the list over time. This causes the
> >average se
Pekka Enberg wrote:
On 7/16/07, Nick Piggin <[EMAIL PROTECTED]> wrote:
Actually SLOB potentially has some fundamental CPU cache hotness
advantages over the other allocators, for the same reasons as
its space advantages.
Because consecutive allocations hit the same cache-hot page regardless
o
On 7/16/07, Nick Piggin <[EMAIL PROTECTED]> wrote:
Actually SLOB potentially has some fundamental CPU cache hotness
advantages over the other allocators, for the same reasons as
its space advantages.
Because consecutive allocations hit the same cache-hot page regardless
of requested size where
Matt Mackall wrote:
The version of SLOB in -mm always scans its free list from the
beginning, which results in small allocations and free segments
clustering at the beginning of the list over time. This causes the
average search to scan over a large stretch at the beginning on each
allocation.
B
4 matches
Mail list logo
