> On Wed, 3 Oct 2007, Yasunori Goto wrote:
>
> > >
> > > That would work. But it would be better to shrink the cache first. The
> > > first 2 slabs on a node may be empty and the shrinking will remove those.
> > > If you do not shrink then the code may falsely assume that there are
> > > objec
On Wed, 3 Oct 2007, Yasunori Goto wrote:
> >
> > That would work. But it would be better to shrink the cache first. The
> > first 2 slabs on a node may be empty and the shrinking will remove those.
> > If you do not shrink then the code may falsely assume that there are
> > objects on the node
> On Tue, 2 Oct 2007, Yasunori Goto wrote:
>
> > Do you mean that just nr_slabs should be checked like followings?
> > I'm not sure this is enough.
> >
> > :
> > if (s->node[nid]) {
> > n = get_node(s, nid);
> > if (!atomic_read(&n->nr_slabs)) {
> > s->node[nid] = NULL;
>
On Tue, 2 Oct 2007, Yasunori Goto wrote:
> Do you mean that just nr_slabs should be checked like followings?
> I'm not sure this is enough.
>
> :
> if (s->node[nid]) {
> n = get_node(s, nid);
> if (!atomic_read(&n->nr_slabs)) {
> s->node[nid] = NULL;
>
> On Mon, 1 Oct 2007, Yasunori Goto wrote:
>
> > +#ifdef CONFIG_MEMORY_HOTPLUG
> > +static void __slab_callback_offline(int nid)
> > +{
> > + struct kmem_cache_node *n;
> > + struct kmem_cache *s;
> > +
> > + list_for_each_entry(s, &slab_caches, list) {
> > + if (s->node[nid]) {
>
On Mon, 1 Oct 2007, Yasunori Goto wrote:
> +#ifdef CONFIG_MEMORY_HOTPLUG
> +static void __slab_callback_offline(int nid)
> +{
> + struct kmem_cache_node *n;
> + struct kmem_cache *s;
> +
> + list_for_each_entry(s, &slab_caches, list) {
> + if (s->node[nid]) {
> +
6 matches
Mail list logo
