diff options
author | Linus Torvalds <torvalds@linux-foundation.org> | 2014-01-23 19:11:50 -0800 |
---|---|---|
committer | Linus Torvalds <torvalds@linux-foundation.org> | 2014-01-23 19:11:50 -0800 |
commit | 3aacd625f20129f5a41ea3ff3b5353b0e4dabd01 (patch) | |
tree | 7cf4ea65397f80098b30494df31cfc8f5fa26d63 /mm/slab.h | |
parent | 7e21774db5cc9cf8fe93a64a2f0c6cf47db8ab24 (diff) | |
parent | 2a1d689c9ba42a6066540fb221b6ecbd6298b728 (diff) | |
download | blackbird-op-linux-3aacd625f20129f5a41ea3ff3b5353b0e4dabd01.tar.gz blackbird-op-linux-3aacd625f20129f5a41ea3ff3b5353b0e4dabd01.zip |
Merge branch 'akpm' (incoming from Andrew)
Merge second patch-bomb from Andrew Morton:
- various misc bits
- the rest of MM
- add generic fixmap.h, use it
- backlight updates
- dynamic_debug updates
- printk() updates
- checkpatch updates
- binfmt_elf
- ramfs
- init/
- autofs4
- drivers/rtc
- nilfs
- hfsplus
- Documentation/
- coredump
- procfs
- fork
- exec
- kexec
- kdump
- partitions
- rapidio
- rbtree
- userns
- memstick
- w1
- decompressors
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (197 commits)
lib/decompress_unlz4.c: always set an error return code on failures
romfs: fix returm err while getting inode in fill_super
drivers/w1/masters/w1-gpio.c: add strong pullup emulation
drivers/memstick/host/rtsx_pci_ms.c: fix ms card data transfer bug
userns: relax the posix_acl_valid() checks
arch/sh/kernel/dwarf.c: use rbtree postorder iteration helper instead of solution using repeated rb_erase()
fs-ext3-use-rbtree-postorder-iteration-helper-instead-of-opencoding-fix
fs/ext3: use rbtree postorder iteration helper instead of opencoding
fs/jffs2: use rbtree postorder iteration helper instead of opencoding
fs/ext4: use rbtree postorder iteration helper instead of opencoding
fs/ubifs: use rbtree postorder iteration helper instead of opencoding
net/netfilter/ipset/ip_set_hash_netiface.c: use rbtree postorder iteration instead of opencoding
rbtree/test: test rbtree_postorder_for_each_entry_safe()
rbtree/test: move rb_node to the middle of the test struct
rapidio: add modular rapidio core build into powerpc and mips branches
partitions/efi: complete documentation of gpt kernel param purpose
kdump: add /sys/kernel/vmcoreinfo ABI documentation
kdump: fix exported size of vmcoreinfo note
kexec: add sysctl to disable kexec_load
fs/exec.c: call arch_pick_mmap_layout() only once
...
Diffstat (limited to 'mm/slab.h')
-rw-r--r-- | mm/slab.h | 26 |
1 files changed, 25 insertions, 1 deletions
diff --git a/mm/slab.h b/mm/slab.h index 0859c4241ba1..8184a7cde272 100644 --- a/mm/slab.h +++ b/mm/slab.h @@ -160,12 +160,36 @@ static inline const char *cache_name(struct kmem_cache *s) return s->name; } +/* + * Note, we protect with RCU only the memcg_caches array, not per-memcg caches. + * That said the caller must assure the memcg's cache won't go away. Since once + * created a memcg's cache is destroyed only along with the root cache, it is + * true if we are going to allocate from the cache or hold a reference to the + * root cache by other means. Otherwise, we should hold either the slab_mutex + * or the memcg's slab_caches_mutex while calling this function and accessing + * the returned value. + */ static inline struct kmem_cache * cache_from_memcg_idx(struct kmem_cache *s, int idx) { + struct kmem_cache *cachep; + struct memcg_cache_params *params; + if (!s->memcg_params) return NULL; - return s->memcg_params->memcg_caches[idx]; + + rcu_read_lock(); + params = rcu_dereference(s->memcg_params); + cachep = params->memcg_caches[idx]; + rcu_read_unlock(); + + /* + * Make sure we will access the up-to-date value. The code updating + * memcg_caches issues a write barrier to match this (see + * memcg_register_cache()). + */ + smp_read_barrier_depends(); + return cachep; } static inline struct kmem_cache *memcg_root_cache(struct kmem_cache *s) |