[PATCH] inotify: lock avoidance with parent watch status in dentry

Previous inotify work avoidance is good when inotify is completely unused,
but it breaks down if even a single watch is in place anywhere in the
system.  Robin Holt notices that udev is one such culprit - it slows down a
512-thread application on a 512 CPU system from 6 seconds to 22 minutes.

Solve this by adding a flag in the dentry that tells inotify whether or not
its parent inode has a watch on it.  Event queueing to parent will skip
taking locks if this flag is cleared.  Setting and clearing of this flag on
all child dentries versus event delivery: this is no in terms of race
cases, and that was shown to be equivalent to always performing the check.

The essential behaviour is that activity occuring _after_ a watch has been
added and _before_ it has been removed, will generate events.

Signed-off-by: Nick Piggin <npiggin@suse.de>
Cc: Robert Love <rml@novell.com>
Cc: John McCutchan <ttb@tentacle.dhs.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This commit is contained in:
Nick Piggin 2006-03-25 03:07:09 -08:00 committed by Linus Torvalds
parent bf36b9011e
commit c32ccd87bf
5 changed files with 117 additions and 10 deletions

View file

@ -16,6 +16,25 @@
#include <linux/dnotify.h>
#include <linux/inotify.h>
/*
* fsnotify_d_instantiate - instantiate a dentry for inode
* Called with dcache_lock held.
*/
static inline void fsnotify_d_instantiate(struct dentry *entry,
struct inode *inode)
{
inotify_d_instantiate(entry, inode);
}
/*
* fsnotify_d_move - entry has been moved
* Called with dcache_lock and entry->d_lock held.
*/
static inline void fsnotify_d_move(struct dentry *entry)
{
inotify_d_move(entry);
}
/*
* fsnotify_move - file old_name at old_dir was moved to new_name at new_dir
*/