On Tue, Oct 07, 2008 at 03:03:07PM -0500, Adam Litke wrote:
> On Fri, 2008-10-03 at 18:37 +0100, Andy Whitcroft wrote:
> > @@ -446,6 +447,91 @@ void __lh_setup_mounts(void)
> > debug_show_page_sizes();
> > }
> >
> > +#define MAX_POOLS 32
> > +int __lh_hpool_sizes(struct hpage_pool *pools, int pcnt)
> > +{
> > + long default_size;
> > + int which = 0;
> > + DIR *dir;
> > + struct dirent *entry;
> > +
> > + long nr_static;
> > + long nr_over;
> > + long nr_used;
> > + long nr_resv;
> > + long nr_surp;
> > +
> > + default_size = size_to_smaller_unit(file_read_ulong(MEMINFO,
> > + "Hugepagesize:"));
> > + if (default_size >= 0) {
> > + nr_used = get_huge_page_counter(default_size, HUGEPAGES_TOTAL);
> > + nr_surp = get_huge_page_counter(default_size, HUGEPAGES_SURP);
> > + nr_over = get_huge_page_counter(default_size, HUGEPAGES_OC);
> > + nr_resv = get_huge_page_counter(default_size, HUGEPAGES_RSVD);
> > +
> > + /*
> > + * XXX: this is racy with respect to ongoing
> > + * allocations. We need to a nr_static_hugepages.
> > + */
> > + nr_static = nr_used - nr_surp;
>
> Should we do some race detection and either print a warning message or
> retry when a race is observed? We could recheck nr_used and nr_surp and
> complain/retry if either value has changed.
Yep, we should just retry until they stablise.
-apw
-------------------------------------------------------------------------
This SF.Net email is sponsored by the Moblin Your Move Developer's challenge
Build the coolest Linux based applications with Moblin SDK & win great prizes
Grand prize is a trip for two to an Open Source event anywhere in the world
http://moblin-contest.org/redirect.php?banner_id=100&url=/
_______________________________________________
Libhugetlbfs-devel mailing list
Libhugetlbfs-devel@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/libhugetlbfs-devel
