xfs_ialloc_log_di is only used to log the full inode core + di_next_unlinked.
That means all the offset magic is not nessecary and we can simply use
xfs_trans_log_buf directly.  Also add a comment describing what we should do
here instead.

(First sent on October 7th)
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

Move all fields from xlog_iclog_fields_t into xlog_in_core_t instead of having
them in a substructure and the using #defines to make it look like they were
directly in xlog_in_core_t.  Also document that xlog_in_core_2_t is grossly
misnamed, and make all references to it typesafe.

(First sent on Semptember 15th)
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

Add a helper to read the AGF header and perform basic verification.
Based on hunks from a larger patch from Dave Chinner.

(First sent on Juli 23rd)
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

Add a helper to read the AGI header and perform basic verification.
Based on hunks from a larger patch from Dave Chinner.

(First sent on Juli 23rd)
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

i_gen is incremented in directory operations when the
directory is changed. It is never read or otherwise used
so it should be removed to help reduce the size of the
struct xfs_inode.

The patch also removes a duplicate logging of the directory
inode core. We only need to do this once per transaction
so kill the one associated with the i_gen increment.
Signed-off-by: NDave Chinner <david@fromorbit.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

The only thing left is xfs_do_force_shutdown which already has a defintion
in xfs_mount.h.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

The only thing left are the forced shutdown flags and freeze macros which
fit into xfs_mount.h much better.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

This adds the fiemap inode_operation, which for us converts the
fiemap values & flags into a getbmapx structure which can be sent
to xfs_getbmap.  The formatter then copies the bmv array back into
the user's fiemap buffer via the fiemap helpers.

If we wanted to be more clever, we could also return mapping data
for in-inode attributes, but I'm not terribly motivated to do that
just yet.
Signed-off-by: NEric Sandeen <sandeen@sandeen.net>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

This adds a new output flag, BMV_OF_LAST to indicate if we've hit
the last extent in the inode.  This potentially saves an extra call
from userspace to see when the whole mapping is done.

It also adds BMV_IF_DELALLOC and BMV_OF_DELALLOC to request, and
indicate, delayed-allocation extents.  In this case bmv_block
is set to -2 (-1 was already taken for HOLESTARTBLOCK; unfortunately
these are the reverse of the in-kernel constants.)

These new flags facilitate addition of the new fiemap interface.

Rather than adding sh_delalloc, remove sh_unwritten & just test
the flags directly.
Signed-off-by: NEric Sandeen <sandeen@sandeen.net>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

Preliminary work to hook up fiemap, this allows us to pass in an
arbitrary formatter to copy extent data back to userspace.

The formatter takes info for 1 extent, a pointer to the user "thing*"
and a pointer to a "filled" variable to indicate whether a userspace
buffer did get filled in (for fiemap, hole "extents" are skipped).

I'm just using the getbmapx struct as a "common denominator" because
as far as I can see, it holds all info that any formatters will care
about.

("*thing" because fiemap doesn't pass the user pointer around, but rather
has a pointer to a fiemap info structure, and helpers associated with it)
Signed-off-by: NEric Sandeen <sandeen@sandeen.net>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

gcc is warning about an uninitialised variable in xfs_growfs_rt().
This is a false positive. Fix it by changing the scope of the
transaction pointer to wholly within the internal loop inside
the function.

While there, preemptively change xfs_growfs_rt_alloc() in the
same way as it has exactly the same structure as xfs_growfs_rt()
but gcc is not warning about it. Yet.
Signed-off-by: NDave Chinner <david@fromorbit.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

XFS gets the sign of the error wrong in several places when
gathering the error from generic linux functions. These functions
return negative error values, while the core XFS code returns
positive error values. Hence when XFS inverts the error to be
returned to the VFS, it can incorrectly invert a negative
error and this error will be ignored by the syscall return.

Fix all the problems related to calling filemap_* functions.

Problem initially identified by Nick Piggin in xfs_fsync().
Signed-off-by: NDave Chinner <david@fromorbit.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

Some recent gcc warnings don't like passing string variables to
printf-like functions without using at least a "%s" format string.
Change the two occurances of that in xfs to please gcc.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NEric Sandeen <sandeen@sandeen.net>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

Now that we've stopped using the Linux inode cache when can trivally
support the inode64 mount option on 32bit architectures.  As far as the
kernel and most userspace is concerned this works perfectly, but
applications still using really old stat and readdir interfaces will get
an EOVERFLOW error when hitting an inode number not fitting into 32
bits (that problem of course also exists when using these applications
on a 64bit kernel).

Note that because inode64 is simply a mount option we can currently
mount a filesystem having > 32 bit inode numbers and cause a variety of
problems, all this is solved but this patch which enables XFS_BIG_INUMS,
even when inode64 is not used.

(First sent on October 18th)
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

Currently there's no ->open method set for directories on XFS.  That
means we don't perform any check for opening too large directories
without O_LARGEFILE, we don't check for shut down filesystems, and we
don't actually do the readahead for the first block in the directory.

Instead of just setting the directories open routine to xfs_file_open
we merge the shutdown check directly into xfs_file_open and create
a new xfs_dir_open that first calls xfs_file_open and then performs
the readahead for block 0.

(First sent on September 29th)
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NDave Chinner <david@fromorbit.com>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

xfs_log_force_umount may be called very early during log recovery where

If we fail a buffer read in xlog_recover_do_inode_trans we abort the mount.
But at that point log recovery has started delayed writeback of inode
buffers.   As part of the aborted mount we try to flush out all delwri
buffers, but at that point we have already freed the superblock, and set
mp->m_sb_bp to NULL, and xfs_log_force_umount which gets called after
the inode buffer writeback trips over it.

Make xfs_log_force_umount a little more careful when accessing mp->m_sb_bp
to avoid this.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NEric Sandeen <sandeen@sandeen.net>
Signed-off-by: NNiv Sardi <xaiki@sgi.com>

If a connection with open file handles has gone down
and come back up and reconnected without reopening
the file handle yet, do not attempt to send an SMB close
request for this handle in cifs_close.  We were
checking for the connection being invalid in cifs_close
but since the connection may have been reconnected
we also need to check whether the file handle
was marked invalid (otherwise we could close the
wrong file handle by accident).
Acked-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

fs/hostfs/hostfs_user.c defines do_readlink() as non-static, and so does
fs/xfs/linux-2.6/xfs_ioctl.c when CONFIG_XFS_DEBUG=y.  So rename
do_readlink() in hostfs to hostfs_do_readlink().

I think it's better if XFS guys will also rename their do_readlink(),
it's not necessary to use such a general name.
Signed-off-by: NWANG Cong <wangcong@zeuux.org>
Cc: Jeff Dike <jdike@addtoit.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Peter Cordes is sorry that he rm'ed his swapfiles while they were in use,
he then had no pathname to swapoff.  It's a curious little oversight, but
not one worth a lot of hackery.  Kudos to Willy Tarreau for turning this
around from a discussion of synthetic pathnames to how to prevent unlink.
Mimic immutable: prohibit unlinking an active swapfile in may_delete()
(and don't worry my little head over the tiny race window).
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Cc: Willy Tarreau <w@1wt.eu>
Acked-by: NChristoph Hellwig <hch@infradead.org>
Cc: Peter Cordes <peter@cordes.ca>
Cc: Bodo Eggert <7eggert@gmx.de>
Cc: David Newall <davidn@davidnewall.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

I have received some reports of out-of-memory errors on some older AMD
architectures.  These errors are what I would expect to see if
crypt_stat->key were split between two separate pages.  eCryptfs should
not assume that any of the memory sent through virt_to_scatterlist() is
all contained in a single page, and so this patch allocates two
scatterlist structs instead of one when processing keys.  I have received
confirmation from one person affected by this bug that this patch resolves
the issue for him, and so I am submitting it for inclusion in a future
stable release.

Note that virt_to_scatterlist() runs sg_init_table() on the scatterlist
structs passed to it, so the calls to sg_init_table() in
decrypt_passphrase_encrypted_session_key() are redundant.
Signed-off-by: NMichael Halcrow <mhalcrow@us.ibm.com>
Reported-by: NPaulo J. S. Silva <pjssilva@ime.usp.br>
Cc: "Leon Woestenberg" <leon.woestenberg@gmail.com>
Cc: Tim Gardner <tim.gardner@canonical.com>
Cc: <stable@kernel.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

This was recently changed to check for need_reconnect, but should
actually be a check for a tidStatus of CifsExiting.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Block ext devt conversion missed md_autodetect_dev() call in
rescan_partitions() leaving md autodetect unable to see partitions.
Fix it.
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Make add_partition() return pointer to the new hd_struct on success
and ERR_PTR() value on failure.  This change will be used to fix md
autodetection bug.
Signed-off-by: NTejun Heo <tj@kernel.org>
Cc: Neil Brown <neilb@suse.de>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Partition stats structure was not freed on devt allocation failure
path.  Fix it.
Signed-off-by: NTejun Heo <tj@kernel.org>
Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Fixes a data corruption under heavy stress in which pages could be left
dirty after all open instances of a inode have been closed.

In order to write contiguous pages whenever possible, cifs_writepages()
asks pagevec_lookup_tag() for more pages than it may write at one time.
Normally, it then resets index just past the last page written before calling
pagevec_lookup_tag() again.

If cifs_writepages() can't write the first page returned, it wasn't resetting
index, and the next call to pagevec_lookup_tag() resulted in skipping all of
the pages it previously returned, even though cifs_writepages() did nothing
with them.  This can result in data loss when the file descriptor is about
to be closed.

This patch ensures that index gets set back to the next returned page so
that none get skipped.
Signed-off-by: NDave Kleikamp <shaggy@linux.vnet.ibm.com>
Acked-by: NJeff Layton <jlayton@redhat.com>
Cc: Shirish S Pargaonkar <shirishp@us.ibm.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Since these hit the same routines, and are relatively small, it is easier to review
them as one patch.

Fixed incorrect handling of the last option in some cases
Fixed prefixpath handling convert path_consumed into host depended string length (in bytes)
Use non default separator if it is provided in the original mount options
Acked-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NIgor Mammedov <niallain@gmail.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

set tcon->ses earlier

If the inital tree connect fails, we'll end up calling cifs_put_smb_ses
with a NULL pointer. Fix it by setting the tcon->ses earlier.
Acked-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

When an I/O error occurs during an intermediate commit on a rolling
transaction, xfs_trans_commit() will free the transaction structure
and the related ticket. However, the duplicate transaction that
gets used as the transaction continues still contains a pointer
to the ticket. Hence when the duplicate transaction is cancelled
and freed, we free the ticket a second time.

Add reference counting to the ticket so that we hold an extra
reference to the ticket over the transaction commit. We drop the
extra reference once we have checked that the transaction commit
did not return an error, thus avoiding a double free on commit
error.

Credit to Nick Piggin for tripping over the problem.

SGI-PV: 989741
Signed-off-by: NDave Chinner <david@fromorbit.com>
Reviewed-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NLachlan McIlroy <lachlan@sgi.com>

Signed-off-by: NSteve French <sfrench@us.ibm.com>

Use a similar approach to the SMB session sharing. Add a list of tcons
attached to each SMB session. Move the refcount to non-atomic. Protect
all of the above with the cifs_tcp_ses_lock. Add functions to
properly find and put references to the tcons.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

D'oh...
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Reported-and-tested-by: NPeter Palfrader <peter@palfrader.org>
Cc: stable@kernel.org
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Inotify watch removals suck violently.

To kick the watch out we need (in this order) inode->inotify_mutex and
ih->mutex.  That's fine if we have a hold on inode; however, for all
other cases we need to make damn sure we don't race with umount.  We can
*NOT* just grab a reference to a watch - inotify_unmount_inodes() will
happily sail past it and we'll end with reference to inode potentially
outliving its superblock.

Ideally we just want to grab an active reference to superblock if we
can; that will make sure we won't go into inotify_umount_inodes() until
we are done.  Cleanup is just deactivate_super().

However, that leaves a messy case - what if we *are* racing with
umount() and active references to superblock can't be acquired anymore?
We can bump ->s_count, grab ->s_umount, which will almost certainly wait
until the superblock is shut down and the watch in question is pining
for fjords.  That's fine, but there is a problem - we might have hit the
window between ->s_active getting to 0 / ->s_count - below S_BIAS (i.e.
the moment when superblock is past the point of no return and is heading
for shutdown) and the moment when deactivate_super() acquires
->s_umount.

We could just do drop_super() yield() and retry, but that's rather
antisocial and this stuff is luser-triggerable.  OTOH, having grabbed
->s_umount and having found that we'd got there first (i.e.  that
->s_root is non-NULL) we know that we won't race with
inotify_umount_inodes().

So we could grab a reference to watch and do the rest as above, just
with drop_super() instead of deactivate_super(), right? Wrong.  We had
to drop ih->mutex before we could grab ->s_umount.  So the watch
could've been gone already.

That still can be dealt with - we need to save watch->wd, do idr_find()
and compare its result with our pointer.  If they match, we either have
the damn thing still alive or we'd lost not one but two races at once,
the watch had been killed and a new one got created with the same ->wd
at the same address.  That couldn't have happened in inotify_destroy(),
but inotify_rm_wd() could run into that.  Still, "new one got created"
is not a problem - we have every right to kill it or leave it alone,
whatever's more convenient.

So we can use idr_find(...) == watch && watch->inode->i_sb == sb as
"grab it and kill it" check.  If it's been our original watch, we are
fine, if it's a newcomer - nevermind, just pretend that we'd won the
race and kill the fscker anyway; we are safe since we know that its
superblock won't be going away.

And yes, this is far beyond mere "not very pretty"; so's the entire
concept of inotify to start with.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Acked-by: NGreg KH <greg@kroah.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NSteve French <sfrench@us.ibm.com>

We do this by abandoning the global list of SMB sessions and instead
moving to a per-server list. This entails adding a new list head to the
TCP_Server_Info struct. The refcounting for the cifsSesInfo is moved to
a non-atomic variable. We have to protect it by a lock anyway, so there's
no benefit to making it an atomic. The list and refcount are protected
by the global cifs_tcp_ses_lock.

The patch also adds a new routines to find and put SMB sessions and
that properly take and put references under the lock.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

The code that allows these structs to be shared is extremely racy.
Disable the sharing of SMB and tcon structs for now until we can
come up with a way to do this that's race free.

We want to continue to share TCP sessions, however since they are
required for multiuser mounts. For that, implement a new (hopefully
race-free) scheme. Add a new global list of TCP sessions, and take
care to get a reference to it whenever we're dealing with one.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

We're currently declaring both a sockaddr_in and sockaddr6_in on the
stack, but we really only need storage for one of them. Declare a
sockaddr struct and cast it to the proper type. Also, eliminate the
protocolType field in the TCP_Server_Info struct. It's redundant since
we have a sa_family field in the sockaddr anyway.

We may need to revisit this if SCTP is ever implemented, but for now
this will simplify the code.

CIFS over IPv6 also has a number of problems currently. This fixes all
of them that I found. Eventually, it would be nice to move more of the
code to be protocol independent, but this is a start.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Also adds two lines missing from the previous patch (for the need reconnect flag in the
/proc/fs/cifs/DebugData handling)

The new global_cifs_sock_list is added, and initialized in init_cifs but not used yet.
Jeff Layton will be adding code in to use that and to remove the GlobalTcon and GlobalSMBSession
lists.

CC: Jeff Layton <jlayton@redhat.com>
CC: Shirish Pargaonkar <shirishp@us.ibm.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

In preparation for Jeff's big umount/mount fixes to remove the possibility of
various races in cifs mount and linked list handling of sessions, sockets and
tree connections, this patch cleans up some repetitive code in cifs_mount,
and addresses a problem with ses->status and tcon->tidStatus in which we
were overloading the "need_reconnect" state with other status in that
field.  So the "need_reconnect" flag has been broken out from those
two state fields (need reconnect was not mutually exclusive from some of the
other possible tid and ses states).  In addition, a few exit cases in
cifs_mount were cleaned up, and a problem with a tcon flag (for lease support)
was not being set consistently for the 2nd mount of the same share

CC: Jeff Layton <jlayton@redhat.com>
CC: Shirish Pargaonkar <shirishp@us.ibm.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Fixes a regression from commit 0f8e0d9a,
"dlm: allow multiple lockspace creates".

An extraneous 'else' slipped into a code fragment being moved from
release_lockspace() to dlm_release_lockspace().  The result of the
unwanted 'else' is that dlm threads and structures are not stopped
and cleaned up when the final dlm lockspace is removed.  Trying to
create a new lockspace again afterward will fail with
"kmem_cache_create: duplicate cache dlm_conn" because the cache
was not previously destroyed.
Signed-off-by: NDavid Teigland <teigland@redhat.com>