Messages in this thread | | | Date | Tue, 9 Mar 2010 14:33:19 -0600 (CST) | From | Christoph Lameter <> | Subject | Re: mm: Do not iterate over NR_CPUS in __zone_pcp_update() |
| |
On Tue, 9 Mar 2010, Andrew Morton wrote:
> > __zone_pcp_update() iterates over NR_CPUS instead of limiting the > > access to the possible cpus. This might result in access to > > uninitialized areas as the per cpu allocator only populates the per > > cpu memory for possible cpus. > > > > Signed-off-by: Thomas Gleixner <tglx@linutronix.de> > > --- > > mm/page_alloc.c | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > Index: linux-2.6/mm/page_alloc.c > > =================================================================== > > --- linux-2.6.orig/mm/page_alloc.c > > +++ linux-2.6/mm/page_alloc.c > > @@ -3224,7 +3224,7 @@ static int __zone_pcp_update(void *data) > > int cpu; > > unsigned long batch = zone_batchsize(zone), flags; > > > > - for (cpu = 0; cpu < NR_CPUS; cpu++) { > > + for_each_possible_cpu(cpu) { > > struct per_cpu_pageset *pset; > > struct per_cpu_pages *pcp; > > > > I'm having trouble working out whether we want to backport this into > 2.6.33.x or earlier. Help?
Nope. This problem was created as a result of the dynamic allocation of pagesets from percpu memory that went in during the merge window. (99dcc3e5a94ed491fbef402831d8c0bbb267f995)
| |