On Sat 14-12-13 12:15:34, Vladimir Davydov wrote:
> The vmalloc was introduced by patch 333279 ("memcgroup: use vmalloc for
> mem_cgroup allocation"), because at that time MAX_NUMNODES was used for
> defining the per-node array in the mem_cgroup structure so that the
> structure could be huge even if the system had the only NUMA node.
>
> The situation was significantly improved by patch 45cf7e ("memcg: reduce
> the size of struct memcg 244-fold"), which made the size of the
> mem_cgroup structure calculated dynamically depending on the real number
> of NUMA nodes installed on the system (nr_node_ids), so now there is no
> point in using vmalloc here: the structure is allocated rarely and on
> most systems its size is about 1K.
>
> Personally I'd like to remove this vmalloc, because I'm considering
> using wait_on_bit() on mem_cgroup::kmem_account_flags in the kmemcg
> shrinkers implementation, which is impossible on vmalloc'd areas.
>
> Signed-off-by: Vladimir Davydov <vdavy...@parallels.com>
> Cc: Michal Hocko <mho...@suse.cz>
> Cc: Glauber Costa <glom...@openvz.org>
> Cc: Johannes Weiner <han...@cmpxchg.org>
> Cc: Balbir Singh <bsinghar...@gmail.com>
> Cc: KAMEZAWA Hiroyuki <kamezawa.hir...@jp.fujitsu.com>
yes this makes sense to me even without wait_on_bit part.
Maybe I would just merge it with 1/2.
Acked-by: Michal Hocko <mho...@suse.cz>
Thanks!
> ---
> mm/memcontrol.c | 28 ++++++----------------------
> 1 file changed, 6 insertions(+), 22 deletions(-)
>
> diff --git a/mm/memcontrol.c b/mm/memcontrol.c
> index 7f1a356..205eb7b 100644
> --- a/mm/memcontrol.c
> +++ b/mm/memcontrol.c
> @@ -48,7 +48,6 @@
> #include <linux/sort.h>
> #include <linux/fs.h>
> #include <linux/seq_file.h>
> -#include <linux/vmalloc.h>
> #include <linux/vmpressure.h>
> #include <linux/mm_inline.h>
> #include <linux/page_cgroup.h>
> @@ -335,12 +334,6 @@ struct mem_cgroup {
> /* WARNING: nodeinfo must be the last member here */
> };
>
> -static size_t memcg_size(void)
> -{
> - return sizeof(struct mem_cgroup) +
> - nr_node_ids * sizeof(struct mem_cgroup_per_node *);
> -}
> -
> /* internal only representation about the status of kmem accounting. */
> enum {
> KMEM_ACCOUNTED_ACTIVE = 0, /* accounted by this cgroup itself */
> @@ -6139,14 +6132,12 @@ static void free_mem_cgroup_per_zone_info(struct
> mem_cgroup *memcg, int node)
> static struct mem_cgroup *mem_cgroup_alloc(void)
> {
> struct mem_cgroup *memcg;
> - size_t size = memcg_size();
> + size_t size;
>
> - /* Can be very big if nr_node_ids is very big */
> - if (size < PAGE_SIZE)
> - memcg = kzalloc(size, GFP_KERNEL);
> - else
> - memcg = vzalloc(size);
> + size = sizeof(struct mem_cgroup);
> + size += nr_node_ids * sizeof(struct mem_cgroup_per_node *);
>
> + memcg = kzalloc(size, GFP_KERNEL);
> if (!memcg)
> return NULL;
>
> @@ -6157,10 +6148,7 @@ static struct mem_cgroup *mem_cgroup_alloc(void)
> return memcg;
>
> out_free:
> - if (size < PAGE_SIZE)
> - kfree(memcg);
> - else
> - vfree(memcg);
> + kfree(memcg);
> return NULL;
> }
>
> @@ -6178,7 +6166,6 @@ out_free:
> static void __mem_cgroup_free(struct mem_cgroup *memcg)
> {
> int node;
> - size_t size = memcg_size();
>
> mem_cgroup_remove_from_trees(memcg);
>
> @@ -6199,10 +6186,7 @@ static void __mem_cgroup_free(struct mem_cgroup *memcg)
> * the cgroup_lock.
> */
> disarm_static_keys(memcg);
> - if (size < PAGE_SIZE)
> - kfree(memcg);
> - else
> - vfree(memcg);
> + kfree(memcg);
> }
>
> /*
> --
> 1.7.10.4
>
--
Michal Hocko
SUSE Labs
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
