linux.git/fs/file_table.c, branch v2.6.37

fs: allow for more than 2^31 files

2010-10-26T23:52:15+00:00

Robin Holt tried to boot a 16TB system and found af_unix was overflowing
a 32bit value :



We were seeing a failure which prevented boot.  The kernel was incapable
of creating either a named pipe or unix domain socket.  This comes down
to a common kernel function called unix_create1() which does:

        atomic_inc(&unix_nr_socks);
        if (atomic_read(&unix_nr_socks) > 2 * get_max_files())
                goto out;

The function get_max_files() is a simple return of files_stat.max_files.
files_stat.max_files is a signed integer and is computed in
fs/file_table.c's files_init().

        n = (mempages * (PAGE_SIZE / 1024)) / 10;
        files_stat.max_files = n;

In our case, mempages (total_ram_pages) is approx 3,758,096,384
(0xe0000000).  That leaves max_files at approximately 1,503,238,553.
This causes 2 * get_max_files() to integer overflow.



Fix is to let /proc/sys/fs/file-nr & /proc/sys/fs/file-max use long
integers, and change af_unix to use an atomic_long_t instead of atomic_t.

get_max_files() is changed to return an unsigned long.  get_nr_files() is
changed to return a long.

unix_nr_socks is changed from atomic_t to atomic_long_t, while not
strictly needed to address Robin problem.

Before patch (on a 64bit kernel) :
# echo 2147483648 >/proc/sys/fs/file-max
# cat /proc/sys/fs/file-max
-18446744071562067968

After patch:
# echo 2147483648 >/proc/sys/fs/file-max
# cat /proc/sys/fs/file-max
2147483648
# cat /proc/sys/fs/file-nr
704     0       2147483648

Reported-by: Robin Holt 
Signed-off-by: Eric Dumazet 
Acked-by: David Miller 
Reviewed-by: Robin Holt 
Tested-by: Robin Holt 
Cc: Al Viro 
Cc: Christoph Hellwig 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

fs: scale files_lock

2010-08-18T12:35:48+00:00

fs: scale files_lock

Improve scalability of files_lock by adding per-cpu, per-sb files lists,
protected with an lglock. The lglock provides fast access to the per-cpu lists
to add and remove files. It also provides a snapshot of all the per-cpu lists
(although this is very slow).

One difficulty with this approach is that a file can be removed from the list
by another CPU. We must track which per-cpu list the file is on with a new
variale in the file struct (packed into a hole on 64-bit archs). Scalability
could suffer if files are frequently removed from different cpu's list.

However loads with frequent removal of files imply short interval between
adding and removing the files, and the scheduler attempts to avoid moving
processes too far away. Also, even in the case of cross-CPU removal, the
hardware has much more opportunity to parallelise cacheline transfers with N
cachelines than with 1.

A worst-case test of 1 CPU allocating files subsequently being freed by N CPUs
degenerates to contending on a single lock, which is no worse than before. When
more than one CPU are allocating files, even if they are always freed by
different CPUs, there will be more parallelism than the single-lock case.

Testing results:

On a 2 socket, 8 core opteron, I measure the number of times the lock is taken
to remove the file, the number of times it is removed by the same CPU that
added it, and the number of times it is removed by the same node that added it.

Booting:    locks=  25049 cpu-hits=  23174 (92.5%) node-hits=  23945 (95.6%)
kbuild -j16 locks=2281913 cpu-hits=2208126 (96.8%) node-hits=2252674 (98.7%)
dbench 64   locks=4306582 cpu-hits=4287247 (99.6%) node-hits=4299527 (99.8%)

So a file is removed from the same CPU it was added by over 90% of the time.
It remains within the same node 95% of the time.

Tim Chen ran some numbers for a 64 thread Nehalem system performing a compile.

                throughput
2.6.34-rc2      24.5
+patch          24.9

                us      sys     idle    IO wait (in %)
2.6.34-rc2      51.25   28.25   17.25   3.25
+patch          53.75   18.5    19      8.75

So significantly less CPU time spent in kernel code, higher idle time and
slightly higher throughput.

Single threaded performance difference was within the noise of microbenchmarks.
That is not to say penalty does not exist, the code is larger and more memory
accesses required so it will be slightly slower.

Cc: linux-kernel@vger.kernel.org
Cc: Tim Chen 
Cc: Andi Kleen 
Signed-off-by: Nick Piggin 
Signed-off-by: Al Viro

fs: cleanup files_lock locking

2010-08-18T12:35:47+00:00

fs: cleanup files_lock locking

Lock tty_files with a new spinlock, tty_files_lock; provide helpers to
manipulate the per-sb files list; unexport the files_lock spinlock.

Cc: linux-kernel@vger.kernel.org
Cc: Christoph Hellwig 
Cc: Alan Cox 
Acked-by: Andi Kleen 
Acked-by: Greg Kroah-Hartman 
Signed-off-by: Nick Piggin 
Signed-off-by: Al Viro

Revert "fsnotify: store struct file not struct path"

2010-08-12T21:23:04+00:00

This reverts commit 3bcf3860a4ff9bbc522820b4b765e65e4deceb3e (and the
accompanying commit c1e5c954020e "vfs/fsnotify: fsnotify_close can delay
the final work in fput" that was a horribly ugly hack to make it work at
all).

The 'struct file' approach not only causes that disgusting hack, it
somehow breaks pulseaudio, probably due to some other subtlety with
f_count handling.

Fix up various conflicts due to later fsnotify work.

Signed-off-by: Linus Torvalds

vfs: improve comment describing fget_light()

2010-08-11T15:59:02+00:00

Improve the description of fget_light(), which is currently incorrect
about needing a prior refcnt (judging by the way it is actually used).

Signed-off-by: Tony Battersby 
Cc: Al Viro 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

vfs/fsnotify: fsnotify_close can delay the final work in fput

2010-07-28T14:18:51+00:00

fanotify almost works like so:

user context calls fsnotify_* function with a struct file.
   fsnotify takes a reference on the struct path
user context goes about it's buissiness

at some later point in time the fsnotify listener gets the struct path
   fanotify listener calls dentry_open() to create a file which userspace can deal with
      listener drops the reference on the struct path
at some later point the listener calls close() on it's new file

With the switch from struct path to struct file this presents a problem for
fput() and fsnotify_close().  fsnotify_close() is called when the filp has
already reached 0 and __fput() wants to do it's cleanup.

The solution presented here is a bit odd.  If an event is created from a
struct file we take a reference on the file.  We check however if the f_count
was already 0 and if so we take an EXTRA reference EVEN THOUGH IT WAS ZERO.
In __fput() (where we know the f_count hit 0 once) we check if the f_count is
non-zero and if so we drop that 'extra' ref and return without destroying the
file.

Signed-off-by: Eric Paris

get rid of the magic around f_count in aio

2010-05-28T02:03:07+00:00

__aio_put_req() plays sick games with file refcount.  What
it wants is fput() from atomic context; it's almost always
done with f_count > 1, so they only have to deal with delayed
work in rare cases when their reference happens to be the
last one.  Current code decrements f_count and if it hasn't
hit 0, everything is fine.  Otherwise it keeps a pointer
to struct file (with zero f_count!) around and has delayed
work do __fput() on it.

Better way to do it: use atomic_long_add_unless( , -1, 1)
instead of !atomic_long_dec_and_test().  IOW, decrement it
only if it's not the last reference, leave refcount alone
if it was.  And use normal fput() in delayed work.

I've made that atomic_long_add_unless call a new helper -
fput_atomic().  Drops a reference to file if it's safe to
do in atomic (i.e. if that's not the last one), tells if
it had been able to do that.  aio.c converted to it, __fput()
use is gone.  req->ki_file *always* contributes to refcount
now.  And __fput() became static.

Signed-off-by: Al Viro

vfs: take f_lock on modifying f_mode after open time

2010-03-06T19:26:25+00:00

We'll introduce FMODE_RANDOM which will be runtime modified.  So protect
all runtime modification to f_mode with f_lock to avoid races.

Signed-off-by: Wu Fengguang 
Cc: Al Viro 
Cc: Christoph Hellwig 
Cc: Trond Myklebust 
Cc: Chuck Lever 
Cc: 			[2.6.33.x]
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

Take ima_file_free() to proper place.

2010-02-07T08:07:29+00:00

Hooks: Just Say No.

Signed-off-by: Al Viro

alloc_file(): simplify handling of mnt_clone_write() errors

2009-12-22T17:27:33+00:00

When alloc_file() and init_file() were combined, the error handling of
mnt_clone_write() was taken into alloc_file() in a somewhat obfuscated
way.  Since we don't use the error code for anything except warning,
we might as well warn directly without an extra variable.

Signed-off-by: Roland Dreier 
Signed-off-by: Al Viro