[lkml]   [2015]   [Jan]   [29]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: [PATCH -mm v2 1/3] slub: never fail to shrink cache
On Thu, Jan 29, 2015 at 3:27 AM, Andrew Morton
<> wrote:
> On Wed, 28 Jan 2015 19:22:49 +0300 Vladimir Davydov <> wrote:
>> SLUB's version of __kmem_cache_shrink() not only removes empty slabs,
>> but also tries to rearrange the partial lists to place slabs filled up
>> most to the head to cope with fragmentation. To achieve that, it
>> allocates a temporary array of lists used to sort slabs by the number of
>> objects in use. If the allocation fails, the whole procedure is aborted.
>> This is unacceptable for the kernel memory accounting extension of the
>> memory cgroup, where we want to make sure that kmem_cache_shrink()
>> successfully discarded empty slabs. Although the allocation failure is
>> utterly unlikely with the current page allocator implementation, which
>> retries GFP_KERNEL allocations of order <= 2 infinitely, it is better
>> not to rely on that.
>> This patch therefore makes __kmem_cache_shrink() allocate the array on
>> stack instead of calling kmalloc, which may fail. The array size is
>> chosen to be equal to 32, because most SLUB caches store not more than
>> 32 objects per slab page. Slab pages with <= 32 free objects are sorted
>> using the array by the number of objects in use and promoted to the head
>> of the partial list, while slab pages with > 32 free objects are left in
>> the end of the list without any ordering imposed on them.
>> ...
>> @@ -3375,51 +3376,56 @@ int __kmem_cache_shrink(struct kmem_cache *s)
>> struct kmem_cache_node *n;
>> struct page *page;
>> struct page *t;
>> - int objects = oo_objects(s->max);
>> - struct list_head *slabs_by_inuse =
>> - kmalloc(sizeof(struct list_head) * objects, GFP_KERNEL);
>> + LIST_HEAD(discard);
>> + struct list_head promote[SHRINK_PROMOTE_MAX];
> 512 bytes of stack. The call paths leading to __kmem_cache_shrink()
> are many and twisty. How do we know this isn't a problem?
> The logic behind choosing "32" sounds rather rubbery. What goes wrong
> if we use, say, "4"?

This much space in the stack may be fertile grounds for kernel stack
overflow code execution :) Another perspective could be that there
should be allocations that are not penalized to a particular cgroup
(from an accounting perspective), should come from the reserved pool.

Balbir Singh.

 \ /
  Last update: 2015-01-29 09:41    [W:0.160 / U:7.148 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site