We normally allow the owner of a file to override permissions checks on
IO operations, since:
- the client will take responsibility for doing an access check
on open;
- the permission checks offer no protection against malicious
clients--if they can authenticate as the file's owner then
they can always just change its permissions;
- checking permission on each IO operation breaks the usual
posix rule that permission is checked only on open.
However, we've never allowed the owner to override permissions on
readdir operations, even though the above logic would also apply to
directories. I've never heard of this causing a problem, probably
because a) simultaneously opening and creating a directory (with
restricted mode) isn't possible, and b) opening a directory, then
chmod'ing it, is rare.
Our disallowal of owner-override on directories appears to be an
accident, though--the readdir itself succeeds, and then we fail just
because lookup_one_len() calls in our filldir methods fail.
I'm not sure what the easiest fix for that would be. For now, just make
this behavior obvious by denying the override right at the start.
This also fixes some odd v4 behavior: with the rdattr_error attribute
requested, it would perform the readdir but return an ACCES error with
each entry.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
We don't need to keep openowners around in the >=4.1 case, because they
aren't needed to handle CLOSE replays any more (that's a problem for
sessions). And doing so causes unexpected failures on a subsequent
destroy_clientid to fail.
We probably also need something comparable for lock owners on last
unlock.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Actually, xfs and jfs can optionally be case insensitive; we'll handle
that case in later patches.
Cc: stable@vger.kernel.org
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Share a little common logic. And note the comments here are a little
out of date (e.g. we don't always create new state in the "new" case any
more.)
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
For the most part readers of cl_cb_state only need a value that is
"eventually" right. And the value is set only either 1) in response to
some change of state, in which case it's set to UNKNOWN and then a
callback rpc is sent to probe the real state, or b) in the handling of a
response to such a callback. UNKNOWN is therefore always a "temporary"
state, and for the other states we're happy to accept last writer wins.
So I think we're OK here.
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
According to RFC 5661, the TEST_STATEID operation is not allowed to
return NFS4ERR_STALE_STATEID. In addition, RFC 5661 says:
15.1.16.5. NFS4ERR_STALE_STATEID (Error Code 10023)
A stateid generated by an earlier server instance was used. This
error is moot in NFSv4.1 because all operations that take a stateid
MUST be preceded by the SEQUENCE operation, and the earlier server
instance is detected by the session infrastructure that supports
SEQUENCE.
I triggered NFS4ERR_STALE_STATEID while testing the Linux client's
NOGRACE recovery. Bruce suggested an additional test that could be
useful to client developers.
Lastly, RFC 5661, section 18.48.3 has this:
o Special stateids are always considered invalid (they result in the
error code NFS4ERR_BAD_STATEID).
An explicit check is made for those state IDs to avoid printk noise.
Signed-off-by: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
Initiate a CB probe when a new connection with the correct direction is added
to a session (IFF backchannel is marked as down). Without this a
BIND_CONN_TO_SESSION has no effect on the internal backchannel state, which
causes the server to reply to every SEQUENCE op with the
SEQ4_STATUS_CB_PATH_DOWN flag set until DESTROY_SESSION.
Signed-off-by: Weston Andros Adamson <dros@netapp.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
HFS+ doesn't really implement hard links - instead, hardlinks are indicated
by a magic file type which refers to an indirect node in a hidden
directory. The spec indicates that stat() should return the inode number
of the indirect node, but it turns out that this doesn't satisfy the
firmware when it's looking for a bootloader - it wants the catalog ID of
the hardlink file instead. Fix up this case.
Signed-off-by: Matthew Garrett <mjg@redhat.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The variable io_size was unsigned int, which caused the wrong sector number
to be calculated after aligning it. This then caused mount to fail with big
volumes, as backup volume header information was searched from a
wrong sector.
Signed-off-by: Janne Kalliomäki <janne@tuxera.com>
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull btrfs compile warning fixes from Chris Mason.
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs:
Btrfs: cast devid to unsigned long long for printk %llu
Btrfs: init old_generation in get_old_root
Highlights include:
- Fix a couple of mount regressions due to the recent cleanups.
- Fix an Oops in the open recovery code
- Fix an rpc_pipefs upcall hang that results from some of the
net namespace work from 3.4.x (stable kernel candidate).
- Fix a couple of write and o_direct regressions that were found
at last weeks Bakeathon testing event in Ann Arbor.
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1.4.12 (GNU/Linux)
iQIcBAABAgAGBQJP2gmaAAoJEGcL54qWCgDyrBMP/RY/T++He8y5k3M9aEqiIv0q
D8ZVMwzID6f4Zgw4xRg96aYr02sBTw0q+0mP5x1EZmg8mK29rnBiVeKHE1iwSfXq
10/SYISlpIjhJC4I4kHXGd2KClgj7qRRCbDKFRWwoIIwYU+kJn8MRnPa9XqdL8kP
q68lrtayW8THSJDR8bk1GQn+ARxGeoY++qzHxm3vpQCbZVVb19VqKMWAWSN4VKqb
epWehOSAzB3iA7HrLRbf8Y8/sDdXewxCQpr9CC/wxuu++l5ifPphR0ToX+k9VZXI
BKFLUojCUZHTMAgCxuxjrFYehMeyClbzL2lLkz5Pgj0gQhOX6Myj+WMXoEg/uWfo
XNf51FH3yBbnfayTaOUs6Y50iuU+dQO7TUTAoWTPpW9V/iT5z/fWAKUVJhDtrPk5
DVDkR6SEgb4P1RqkehZKLq5k5GSAcTR+MZr452eDrFYXJrY8ORDE6o6kP4Rr3Nnd
n8gap0gHxzIYlhBghem6+nLN+HhpZQopWeD8mNub20VuXsChRDr9/+XWuMCSJaZF
2kleVdt2+rTDzi9bJTRYlsX397oaThL0NbRvshHAwnXIDtIQrzxx6+dUyOsEWMEu
go/EdSUUESXGNlsWTqewCBsOjPeE4L5ijI/QglfDkF+CzD5dDjrxl+5i57iMKVfc
Ydste3pQJkS7PiZu1sWA
=unbu
-----END PGP SIGNATURE-----
Merge tag 'nfs-for-3.5-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs
Pull NFS client bugfixes from Trond Myklebust:
"Highlights include:
- Fix a couple of mount regressions due to the recent cleanups.
- Fix an Oops in the open recovery code
- Fix an rpc_pipefs upcall hang that results from some of the net
namespace work from 3.4.x (stable kernel candidate).
- Fix a couple of write and o_direct regressions that were found at
last weeks Bakeathon testing event in Ann Arbor."
* tag 'nfs-for-3.5-2' of git://git.linux-nfs.org/projects/trondmy/linux-nfs:
NFS: add an endian notation for sparse
NFSv4.1: integer overflow in decode_cb_sequence_args()
rpc_pipefs: allow rpc_purge_list to take a NULL waitq pointer
NFSv4 do not send an empty SETATTR compound
NFSv2: EOF incorrectly set on short read
NFS: Use the NFS_DEFAULT_VERSION for v2 and v3 mounts
NFS: fix directio refcount bug on commit
NFSv4: Fix unnecessary delegation returns in nfs4_do_open
NFSv4.1: Convert another trivial printk into a dprintk
NFS4: Fix open bug when pnfs module blacklisted
NFS: Remove incorrect BUG_ON in nfs_found_client
NFS: Map minor mismatch error to protocol not support error.
NFS: Fix a commit bug
NFS4: Set parsed mount data version to 4
NFSv4.1: Ensure we clear session state flags after a session creation
NFSv4.1: Convert a trivial printk into a dprintk
NFSv4: Fix up decode_attr_mdsthreshold
NFSv4: Fix an Oops in the open recovery code
NFSv4.1: Fix a request leak on the back channel
Pull two nfsd bugfixes from J. Bruce Fields.
* 'for-3.5' of git://linux-nfs.org/~bfields/linux:
nfsd4: BUG_ON(!is_spin_locked()) no good on UP kernels
NFS: hard-code init_net for NFS callback transports
gcc was giving an uninit variable warning here. Strictly
speaking we don't need to init it, but this will make things
much less error prone.
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Pull btrfs update from Chris Mason:
"The dates look like I had to rebase this morning because there was a
compiler warning for a printk arg that I had missed earlier.
These are all fixes, including one to prevent using stale pointers for
device names, and lots of fixes around transaction abort cleanups
(Josef, Liu Bo).
Jan Schmidt also sent in a number of fixes for the new reference
number tracking code.
Liu Bo beat me to updating the MAINTAINERS file. Since he thought to
also fix the git url, I kept his commit."
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mason/linux-btrfs: (24 commits)
Btrfs: update MAINTAINERS info for BTRFS FILE SYSTEM
Btrfs: destroy the items of the delayed inodes in error handling routine
Btrfs: make sure that we've made everything in pinned tree clean
Btrfs: avoid memory leak of extent state in error handling routine
Btrfs: do not resize a seeding device
Btrfs: fix missing inherited flag in rename
Btrfs: fix incompat flags setting
Btrfs: fix defrag regression
Btrfs: call filemap_fdatawrite twice for compression
Btrfs: keep inode pinned when compressing writes
Btrfs: implement ->show_devname
Btrfs: use rcu to protect device->name
Btrfs: unlock everything properly in the error case for nocow
Btrfs: fix btrfs_destroy_marked_extents
Btrfs: abort the transaction if the commit fails
Btrfs: wake up transaction waiters when aborting a transaction
Btrfs: fix locking in btrfs_destroy_delayed_refs
Btrfs: pass locked_page into extent_clear_unlock_delalloc if theres an error
Btrfs: fix race in tree mod log addition
Btrfs: add btrfs_next_old_leaf
...
the items of the delayed inodes were forgotten to be freed, this patch
fixes it.
Signed-off-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Since we have two trees for recording pinned extents, we need to go through
both of them to make sure that we've done everything clean.
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
We've forgotten to clear extent states in pinned tree, which will results in
space counter mismatch and memory leak:
WARNING: at fs/btrfs/extent-tree.c:7537 btrfs_free_block_groups+0x1f3/0x2e0 [btrfs]()
...
space_info 2 has 8380416 free, is not full
space_info total=12582912, used=4096, pinned=4096, reserved=0, may_use=0, readonly=4194304
btrfs state leak: start 29364224 end 29376511 state 1 in tree ffff880075f20090 refs 1
...
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
Seeding devices are not supposed to change any more.
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
When we move a file into a directory with compression flag, we need to
inherite BTRFS_INODE_COMPRESS and clear BTRFS_INODE_NOCOMPRESS as well.
But if we move a file into a directory without compression flag, we need
to clear both of them.
It is the way how our setflags deals with compression flag, so keep
the same behaviour here.
Signed-off-by: Liu Bo <liubo2009@cn.fujitsu.com>
Signed-off-by: Chris Mason <chris.mason@fusionio.com>
If a file has 3 small extents:
| ext1 | ext2 | ext3 |
Running "btrfs fi defrag" will only defrag the last two extents, if those
extent mappings hasn't been read into memory from disk.
This bug was introduced by commit 17ce6ef8d7
("Btrfs: add a check to decide if we should defrag the range")
The cause is, that commit looked into previous and next extents using
lookup_extent_mapping() only.
While at it, remove the code that checks the previous extent, since
it's sufficient to check the next extent.
Signed-off-by: Li Zefan <lizefan@huawei.com>
I removed this in an earlier commit and I was wrong. Because compression
can return from filemap_fdatawrite() without having actually set any of it's
pages as writeback() it can make filemap_fdatawait() do essentially nothing,
and then we won't find any ordered extents because they may not have been
created yet. So not only does this make fsync() completely useless, but it
will also screw up if you truncate on a non-page aligned offset since we
zero out the end and then wait on ordered extents and then call drop caches.
We can drop the cache before the io completes and then we try to unpin the
extent we just wrote we won't find it and everything goes sideways. So fix
this by putting it back and put a giant comment there to keep me from trying
to remove it in the future. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
A user reported lots of problems using compression on the new code and it
turns out part of the problem was that igrab() was failing when we added a
new ordered extent. This is because when writing out an inode under
compression we immediately return without actually doing anything to the
pages, and then in another thread at some point down the line actually do
the ordered dance. The problem is between the point that we start writeback
and we actually add the ordered extent we could be trying to reclaim the
inode, which makes igrab() return NULL. So we need to do an igrab() when we
create the async extent and then drop it when we are done with it. This
makes sure we stay pinned in memory until the ordered extent can get a
reference on it and we are good to go. With this patch we no longer panic
in btrfs_finish_ordered_io(). Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Because btrfs can remove the device that was mounted we need to have a
->show_devname so that in this case we can print out some other device in
the file system to /proc/mount. So if there are multiple devices in a btrfs
file system we will just print the device with the lowest devid that we can
find. This will make everything consistent and deal with device removal
properly. The drawback is if you mount with a device that is higher than
the lowest devicd it won't show up as the mounted device in /proc/mounts,
but this is a small price to pay. This was inspired by Miao Xie's patch.
Thanks,
Reviewed-by: Miao Xie <miaox@cn.fujitsu.com>
Signed-off-by: Josef Bacik <josef@redhat.com>
Al pointed out that we can just toss out the old name on a device and add a
new one arbitrarily, so anybody who uses device->name in printk could
possibly use free'd memory. Instead of adding locking around all of this he
suggested doing it with RCU, so I've introduced a struct rcu_string that
does just that and have gone through and protected all accesses to
device->name that aren't under the uuid_mutex with rcu_read_lock(). This
protects us and I will use it for dealing with removing the device that we
used to mount the file system in a later patch. Thanks,
Reviewed-by: David Sterba <dsterba@suse.cz>
Signed-off-by: Josef Bacik <josef@redhat.com>
I was getting hung on umount when a transaction was aborted because a range
of one of the free space inodes was still locked. This is because the nocow
stuff doesn't unlock anything on error. This fixed the problem and I
verified that is what was happening. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
So we're forcing the eb's to have their ref count set to 1 so invalidatepage
works but this breaks lots of things, for example root nodes, and is just
plain wrong, we don't need to just evict all of this stuff. Also drop the
invalidatepage altogether and add a page_cache_release(). With this patch
we no longer hang when trying to access the root nodes after an aborted
transaction and we no longer leak memory. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
If a transaction commit fails we don't abort it so we don't set an error on
the file system. This patch fixes that by actually calling the abort stuff
and then adding a check for a fs error in the transaction start stuff to
make sure it is caught properly. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
I was getting lots of hung tasks and a NULL pointer dereference because we
are not cleaning up the transaction properly when it aborts. First we need
to reset the running_transaction to NULL so we don't get a bad dereference
for any start_transaction callers after this. Also we cannot rely on
waitqueue_active() since it's just a list_empty(), so just call wake_up()
directly since that will do the barrier for us and such. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
The transaction abort stuff was throwing warnings from the list debugging
code because we do a list_del_init outside of the delayed_refs spin lock.
The delayed refs locking makes baby Jesus cry so it's not hard to get wrong,
but we need to take the ref head mutex to make sure it's not being processed
currently, and so if it is we need to drop the spin lock and then take and
drop the mutex and do the search again. If we can take the mutex then we
can safely remove the head from the list and carry on. Now when the
transaction aborts I don't get the list debugging warnings. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
While doing my enospc work I got a transaction abortion that resulted in a
panic when we tried to unlock_page() an already unlocked page. This is
because we aren't calling extent_clear_unlock_delalloc with the locked page
so it was unlocking all the pages in the range. This is wrong since
__extent_writepage expects to have the page locked still unless we return
*page_started as 1. This should keep us from panicing. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Most frequent symptom was a BUG triggering in expire_client, with the
server locking up shortly thereafter.
Introduced by 508dc6e110 "nfsd41:
free_session/free_client must be called under the client_lock".
Cc: stable@kernel.org
Cc: Benny Halevy <bhalevy@tonian.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
In case of destroying mount namespace on child reaper exit, nsproxy is zeroed
to the point already. So, dereferencing of it is invalid.
This patch hard-code "init_net" for all network namespace references for NFS
callback services. This will be fixed with proper NFS callback
containerization.
Signed-off-by: Stanislav Kinsbursky <skinsbursky@parallels.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
When adding to the tree modification log, we grab two locks at different
stages. We must not drop the outer lock until we're done with section
protected by the inner lock. This moves the unlock call for the outer lock
to the appropriate position.
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
To make sense of the tree mod log, the backref walker not only needs
btrfs_search_old_slot, but it also called btrfs_next_leaf, which in turn was
calling btrfs_search_slot. This obviously didn't give the correct result.
This commit adds btrfs_next_old_leaf, a drop-in replacement for
btrfs_next_leaf with a time_seq parameter. If it is zero, it behaves exactly
like btrfs_next_leaf. If it is non-zero, it will use btrfs_search_old_slot
with this time_seq parameter.
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
In __tree_mod_log_oldest_root() we must return the found operation even if
it's not a ROOT_REPLACE operation. Otherwise, the caller assumes that there
are no operations to be rewinded and returns immediately.
The code in the caller is modified to improve readability.
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
get_old_root could race with root node updates because we weren't locking
the node early enough. Use btrfs_read_lock_root_node to grab the root locked
in the very beginning and release the lock as soon as possible (just like
btrfs_search_slot does).
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
When resolving indirect refs, we used to call btrfs_next_leaf in case we
didn't find an exact match. While we should find exact matches most of the
time, in case we don't, we must continue searching. Treating those matches
differently depending on the level we're searching doesn't make sense.
Even worse, we might end up searching for a key larger than the largest, in
which case there is no next_leaf and subsequent jobs would fail. This commit
drops the bogous lines.
Signed-off-by: Jan Schmidt <list.btrfs@jan-o-sch.net>
This is supposed to be a __be32 value. Sparse complains a lot:
fs/nfs/callback_xdr.c:699:30: warning: incorrect type in initializer (different base types)
fs/nfs/callback_xdr.c:699:30: expected unsigned int [unsigned] status
fs/nfs/callback_xdr.c:699:30: got restricted __be32 const [usertype] csr_status
fs/nfs/callback_xdr.c:715:9: warning: cast to restricted __be32
fs/nfs/callback_xdr.c:716:16: warning: incorrect type in return expression (different base types)
fs/nfs/callback_xdr.c:716:16: expected restricted __be32
fs/nfs/callback_xdr.c:716:16: got unsigned int [unsigned] status
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
This seems like it could overflow on 32 bits. Use kmalloc_array() which
has overflow protection built in.
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Fix sparse non-ANSI function warning:
fs/exofs/sys.c:112:28: warning: non-ANSI function declaration of function 'exofs_sysfs_dbg_print'
Signed-off-by: Randy Dunlap <rdunlap@xenotime.net>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Commit 536e43d12b ATTR_OPEN check can result in
an ia_valid with only ATTR_FILE set, and no NFS_VALID_ATTRS attributes to
request from the server.
Signed-off-by: Andy Adamson <andros@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
In cases where the server returns fewer bytes then those requested, we
can incorrectly set the eof flag for the file. Fixing this allows the
request to be retried with updated offset and count arguments.
Signed-off-by: Sachin Prabhu <sprabhu@redhat.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Older versions of nfs utils don't always pass a "vers=" mount option for
NFS. This chould lead to attempts at using NFS v0 due to a zeroed out
nfs_parsed_mount_data struct. I solve this by setting the default NFS
version to NFS_DEFAULT_VERSION in the v2 and v3 cases (v4 has already been
taken care of by a similar patch).
Reported-by: Joerg Roedel <joro@&bytes.org>
Signed-off-by: Bryan Schumaker <bjschuma@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
This reverts a hunk from commit 0427708657
"NFS: Clean up - Simplify reference counting in fs/nfs/direct.c"
The cleanups in that patch affect the write path, but by the time
processing hits commit the removed reference has been added back by
nfs_scan_commit_list(). Without this reversion, any page that is
sent to commit holds on to an unbalanced reference that is never
freed. The immediate effect is an imbalance over the wire between
OPENs and CLOSEs.
Signed-off-by: Fred Isaman <iisaman@netapp.com>
Signed-off-by: Trond Myklebust <Trond.Myklebust@netapp.com>
Fix bug introduced by 169ebd90. We have to have wb_list_lock locked when
restarting writeback loop after having waited for inode writeback.
Bug description by Ted Tso:
I can reproduce this fairly easily by using ext4 w/o a journal, running
under KVM with 1024megs memory, with fsstress (xfstests #13):
[ 45.153294] =====================================
[ 45.154784] [ BUG: bad unlock balance detected! ]
[ 45.155591] 3.5.0-rc1-00002-gb22b1f1 #124 Not tainted
[ 45.155591] -------------------------------------
[ 45.155591] flush-254:16/2499 is trying to release lock (&(&wb->list_lock)->rlock) at:
[ 45.155591] [<c022c3da>] writeback_sb_inodes+0x160/0x327
[ 45.155591] but there are no more locks to release!
Reported-by: Theodore Ts'o <tytso@mit.edu>
Tested-by: Theodore Ts'o <tytso@mit.edu>
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: Fengguang Wu <fengguang.wu@intel.com>
J. Bruce Fields