Commits · 2e32cf5ef2683a196df24ccd2cd2b4ed735ad9d3 · Kirill Smelkov / linux

25 Oct, 2013 40 commits

cifs: rcu-delay unload_nls() and freeing sbi · 2e32cf5e

Al Viro authored Oct 03, 2013

makes ->d_hash(), ->d_compare() and ->permission() safety in RCU mode
independent from vfsmount_lock.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

2e32cf5e

autofs4: make freeing sbi rcu-delayed · baa40671

Al Viro authored Oct 03, 2013

makes ->d_managed() safety in RCU mode independent from vfsmount_lock
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

baa40671

adfs: delayed freeing of sbi · 2d1d9b5b

Al Viro authored Oct 03, 2013

makes ->d_hash() and ->d_compare() safety in RCU mode independent
from vfsmount_lock.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

2d1d9b5b

hpfs: make freeing sbi and codetables rcu-delayed · 30687e0a

Al Viro authored Oct 03, 2013

makes ->d_hash() and ->d_compare() safety in RCU mode independent
from vfsmount_lock
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

30687e0a

make freeing super_block rcu-delayed · e2fec7c3
Al Viro authored Oct 04, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
e2fec7c3

vfs: introduce d_instantiate_no_diralias() · b70a80e7

Miklos Szeredi authored Oct 01, 2013

...which just returns -EBUSY if a directory alias would be created.

This is to be used by fuse mkdir to make sure that a buggy or malicious
userspace filesystem doesn't do anything nasty.  Previously fuse used a
private mutex for this purpose, which can now go away.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>

b70a80e7

move taking vfsmount_lock down into prepend_path() · 94e92a6e
Al Viro authored Oct 01, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
94e92a6e

split __lookup_mnt() in two functions · 474279dc

Al Viro authored Oct 01, 2013

Instead of passing the direction as argument (and checking it on every
step through the hash chain), just have separate __lookup_mnt() and
__lookup_mnt_last().  And use the standard iterators...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

474279dc

uninline destroy_super(), consolidate alloc_super() · 7eb5e882
Al Viro authored Oct 01, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
7eb5e882
isofs: don't pass dentry to isofs_hash{i,}_common() · 966c1f75
Al Viro authored Sep 29, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
966c1f75

new helpers: lock_mount_hash/unlock_mount_hash · 719ea2fb

Al Viro authored Sep 29, 2013

aka br_write_{lock,unlock} of vfsmount_lock.  Inlines in fs/mount.h,
vfsmount_lock extern moved over there as well.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

719ea2fb

don't bother with vfsmount_lock in mounts_poll() · aab407fc

Al Viro authored Sep 29, 2013

wake_up_interruptible/poll_wait provide sufficient barriers;
just use ACCESS_ONCE() to fetch ns->event and that's it.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

aab407fc

namespace.c: get rid of mnt_ghosts · aba809cf
Al Viro authored Sep 28, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
aba809cf
fold dup_mnt_ns() into its only surviving caller · 9559f689
Al Viro authored Sep 28, 2013
```
should've been done 6 years ago...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
9559f689

mnt_set_expiry() doesn't need vfsmount_lock · f6b742d8

Al Viro authored Sep 28, 2013

->mnt_expire is protected by namespace_sem
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

f6b742d8

finish_automount() doesn't need vfsmount_lock for removal from expiry list · 22a79192
Al Viro authored Sep 28, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
22a79192

fs/namespace.c: bury long-dead define · 085e83ff

Al Viro authored Sep 28, 2013

MNT_WRITER_UNDERFLOW_LIMIT has been missed 4 years ago when it became unused.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

085e83ff

fold mntfree() into mntput_no_expire() · 649a795a
Al Viro authored Sep 28, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
649a795a

do_remount(): pull touch_mnt_namespace() up · 6339dab8

Al Viro authored Sep 16, 2013

... and don't bother with dropping and regaining vfsmount_lock
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

6339dab8

dup_mnt_ns(): get rid of pointless grabbing of vfsmount_lock · aa7a574d

Al Viro authored Sep 16, 2013

mnt_list is protected by namespace_sem, not vfsmount_lock
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

aa7a574d

fs_is_visible only needs namespace_sem held shared · 44bb4385
Al Viro authored Sep 16, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
44bb4385
initialize namespace_sem statically · 59aa0da8
Al Viro authored Sep 16, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
59aa0da8
file->f_op is never NULL... · 72c2d531
Al Viro authored Sep 22, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
72c2d531
rtl8188eu: remove dead code · 22bd002e
Al Viro authored Sep 22, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
22bd002e

dmxdev: get rid of pointless clearing ->f_op · 79b185dd

Al Viro authored Sep 22, 2013

nobody else will see that struct file after return from ->release()
anyway; just leave ->f_op as is and let __fput() do that fops_put().
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

79b185dd

consolidate the reassignments of ->f_op in ->open() instances · e84f9e57
Al Viro authored Sep 22, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
e84f9e57
put_mnt_ns(): use drop_collected_mounts() · 7b00ed6f
Al Viro authored Sep 16, 2013
```
... rather than open-coding it
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
7b00ed6f
ncpfs: switch to %p[dD] · 84eb3532
Al Viro authored Sep 16, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
84eb3532
ubifs: switch to %pd · 4cb2a01d
Al Viro authored Sep 16, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
4cb2a01d
sunrpc: switch to %pd · 1e903eda
Al Viro authored Sep 16, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
1e903eda
nfsd: switch to %p[dD] · a6a9f18f
Al Viro authored Sep 16, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
a6a9f18f
nfs: use %p[dD] instead of open-coded (and often racy) equivalents · 6de1472f
Al Viro authored Sep 16, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
6de1472f
befs: split symlink iops in two - for short and long symlinks resp. · 48bc06e7
Al Viro authored Sep 16, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
48bc06e7

new helper: kfree_put_link() · 87dc800b

Al Viro authored Sep 16, 2013

duplicated to hell and back...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

87dc800b

libfs: get exports to definitions of objects being exported... · 12f38872
Al Viro authored Sep 15, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
12f38872
ecryptfs: ->lower_path.dentry is never NULL · cbe9c085
Al Viro authored Sep 15, 2013
```
... on anything found via ->d_fsdata
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
cbe9c085
ecryptfs: get rid of ecryptfs_set_dentry_lower{,_mnt} · 92dd1230
Al Viro authored Sep 15, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
92dd1230

ecryptfs: don't leave RCU pathwalk immediately · 2edbfbf1

Al Viro authored Sep 15, 2013

If the underlying dentry doesn't have ->d_revalidate(), there's no need to
force dropping out of RCU mode.  All we need for that is to make freeing
ecryptfs_dentry_info RCU-delayed.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>

2edbfbf1

ecryptfs: check DCACHE_OP_REVALIDATE instead of ->d_op · 3a93e17c
Al Viro authored Sep 15, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
3a93e17c
9p: make v9fs_cache_inode_{get,put,set}_cookie empty inlines for !9P_CACHEFS · ceaec15d
Al Viro authored Sep 17, 2013
```
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
```
ceaec15d