linux.git/fs/dcache.c, branch v2.6.16

[PATCH] fix file counting

2006-03-08T22:14:01+00:00

I have benchmarked this on an x86_64 NUMA system and see no significant
performance difference on kernbench.  Tested on both x86_64 and powerpc.

The way we do file struct accounting is not very suitable for batched
freeing.  For scalability reasons, file accounting was
constructor/destructor based.  This meant that nr_files was decremented
only when the object was removed from the slab cache.  This is susceptible
to slab fragmentation.  With RCU based file structure, consequent batched
freeing and a test program like Serge's, we just speed this up and end up
with a very fragmented slab -

llm22:~ # cat /proc/sys/fs/file-nr
587730  0       758844

At the same time, I see only a 2000+ objects in filp cache.  The following
patch I fixes this problem.

This patch changes the file counting by removing the filp_count_lock.
Instead we use a separate percpu counter, nr_files, for now and all
accesses to it are through get_nr_files() api.  In the sysctl handler for
nr_files, we populate files_stat.nr_files before returning to user.

Counting files as an when they are created and destroyed (as opposed to
inside slab) allows us to correctly count open files with RCU.

Signed-off-by: Dipankar Sarma 
Cc: "Paul E. McKenney" 
Cc: "David S. Miller" 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] make "struct d_cookie" depend on CONFIG_PROFILING

2006-02-03T16:32:04+00:00

Shrinks "struct dentry" from 128 bytes to 124 on x86, allowing 31 objects
per slab instead of 30.

Cc: John Levon 
Cc: Philippe Elie 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] Unlinline a bunch of other functions

2006-01-15T02:27:06+00:00

Remove the "inline" keyword from a bunch of big functions in the kernel with
the goal of shrinking it by 30kb to 40kb

Signed-off-by: Arjan van de Ven 
Signed-off-by: Ingo Molnar 
Acked-by: Jeff Garzik 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] d_instantiate_unique / NFS inode leakage

2006-01-10T16:01:41+00:00

If we have found aliased dentry that we return, inode reference is not
dropped and inode is not attached anywhere, so it seems the reference to
inode is leaked in that case.

Cc: Trond Myklebust ,
Cc: 
Cc: Christoph Hellwig 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] shrink dentry struct

2006-01-09T04:13:58+00:00

Some long time ago, dentry struct was carefully tuned so that on 32 bits
UP, sizeof(struct dentry) was exactly 128, ie a power of 2, and a multiple
of memory cache lines.

Then RCU was added and dentry struct enlarged by two pointers, with nice
results for SMP, but not so good on UP, because breaking the above tuning
(128 + 8 = 136 bytes)

This patch reverts this unwanted side effect, by using an union (d_u),
where d_rcu and d_child are placed so that these two fields can share their
memory needs.

At the time d_free() is called (and d_rcu is really used), d_child is known
to be empty and not touched by the dentry freeing.

Lockless lookups only access d_name, d_parent, d_lock, d_op, d_flags (so
the previous content of d_child is not needed if said dentry was unhashed
but still accessed by a CPU because of RCU constraints)

As dentry cache easily contains millions of entries, a size reduction is
worth the extra complexity of the ugly C union.

Signed-off-by: Eric Dumazet 
Cc: Dipankar Sarma 
Cc: Maneesh Soni 
Cc: Miklos Szeredi 
Cc: "Paul E. McKenney" 
Cc: Ian Kent 
Cc: Paul Jackson 
Cc: Al Viro 
Cc: Christoph Hellwig 
Cc: Trond Myklebust 
Cc: Neil Brown 
Cc: James Morris 
Cc: Stephen Smalley 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] Remove hlist_for_each_rcu() API, convert existing use to hlist_for_each_entry_rcu

2005-11-07T15:53:35+00:00

Remove the hlist_for_each_rcu() API, which is used only in one place, and
is trivially converted to hlist_for_each_entry_rcu(), making the code
shorter and more readable.  Any out-of-tree uses may be similarly
converted.

Signed-off-by: "Paul E. McKenney" 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] gfp_t: fs/*

2005-10-28T15:16:47+00:00

 - ->releasepage() annotated (s/int/gfp_t), instances updated
 - missing gfp_t in fs/* added
 - fixed misannotation from the original sweep caught by bitwise checks:
   XFS used __nocast both for gfp_t and for flags used by XFS allocator.
   The latter left with unsigned int __nocast; we might want to add a
   different type for those but for now let's leave them alone.  That,
   BTW, is a case when __nocast use had been actively confusing - it had
   been used in the same code for two different and similar types, with
   no way to catch misuses.  Switch of gfp_t to bitwise had caught that
   immediately...

One tricky bit is left alone to be dealt with later - mapping->flags is
a mix of gfp_t and error indications.  Left alone for now.

Signed-off-by: Al Viro 
Signed-off-by: Linus Torvalds

Make fsnotify possibly work better for the inode removal case

2005-09-20T02:54:29+00:00

Checking i_nlink is dubious, but the alternatives look even
less appetizing.

Signed-off-by: Linus Torvalds

[PATCH] janitor: fs/dcache.c: list_for_each*

2005-09-10T17:06:32+00:00

First one is list_for_each_entry (thanks maks), second 2 list_for_each_safe.

Signed-off-by: Maximilian Attems 
Signed-off-by: Domen Puncer 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

[PATCH] fsnotify_name/inoderemove

2005-08-08T18:53:47+00:00

The patch below unhooks fsnotify from vfs_unlink & vfs_rmdir.  It
introduces two new fsnotify calls, that are hooked in at the dcache
level.  This not only more closely matches how the VFS layer works, it
also avoids the problem with locking and inode lifetimes.

The two functions are

 - fsnotify_nameremove -- called when a directory entry is going away.
   It notifies the PARENT of the deletion.  This is called from
   d_delete().

 - inoderemove -- called when the files inode itself is going away.  It
   notifies the inode that is being deleted.  This is called from
   dentry_iput().

Signed-off-by: John McCutchan 
Signed-off-by: Linus Torvalds