Messages in this thread Patch in this message | | | Date | Fri, 20 Dec 2013 12:05:29 -0600 (CST) | From | Christoph Lameter <> | Subject | Re: [fs] inode_lru_isolate(): Move counter increment into spinlock section |
| |
On Fri, 20 Dec 2013, Dave Chinner wrote:
> If count_vm_events requires irqs to be disabled to behave correctly, > then putting __count_vm_events under a spin lock is still not irq > safe. Either way, this isn't in a performance critical path, so I'd > much prefer the simpler, safer option be used rather than leave a > landmine for other unsuspecting developers.
Subject: [fs] Use safe counter increment operations
The counter increment in inode_lru_isolate is happening with preemption on using __count_vm_events making counter increment races possible.
Use count_vm_events instead.
Signed-off-by: Christoph Lameter <cl@linux.com>
Index: linux/fs/inode.c =================================================================== --- linux.orig/fs/inode.c 2013-12-20 12:02:14.409583380 -0600 +++ linux/fs/inode.c 2013-12-20 12:02:14.409583380 -0600 @@ -722,9 +722,9 @@ inode_lru_isolate(struct list_head *item unsigned long reap; reap = invalidate_mapping_pages(&inode->i_data, 0, -1); if (current_is_kswapd()) - __count_vm_events(KSWAPD_INODESTEAL, reap); + count_vm_events(KSWAPD_INODESTEAL, reap); else - __count_vm_events(PGINODESTEAL, reap); + count_vm_events(PGINODESTEAL, reap); if (current->reclaim_state) current->reclaim_state->reclaimed_slab += reap; }
| |