Define (global) data structures to store ids, uids and gids, to which a
SID maps.  There are two separate trees, one for SID/uid and another one
for SID/gid.

A new type of key, cifs_idmap_key_type, is used.

Keys are instantiated and searched using credential of the root by
overriding and restoring the credentials of the caller requesting the key.

Id mapping functions are invoked under config option of cifs acl.
Signed-off-by: NShirish Pargaonkar <shirishpargaonkar@gmail.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Remove config flag CIFS_EXPERIMENTAL.
Do export operations under new config flag CIFS_NFSD_EXPORT
Signed-off-by: NShirish Pargaonkar <shirishpargaonkar@gmail.com>
Reviewed-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

The CIFSSMBNotify worker is unused, pending changes to allow it to be called
via inotify, so move it into its own experimental config option so it does
not get built in, until the necessary VFS support is fixed.  It used to
be used in dnotify, but according to Jeff, inotify needs minor changes
before we can reenable this.

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

ino is unused in function cifs_root_iget().
Signed-off-by: NShirish Pargaonkar <shirishpargaonkar@gmail.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

This is more or less the same patch as before, but with some merge
conflicts fixed up.

If a process has a dirty page mapped into its page tables, then it has
the ability to change it while the client is trying to write the data
out to the server. If that happens after the signature has been
calculated then that signature will then be wrong, and the server will
likely reset the TCP connection.

This patch adds a page_mkwrite handler for CIFS that simply takes the
page lock. Because the page lock is held over the life of writepage and
writepages, this prevents the page from becoming writeable until
the write call has completed.

With this, we can also remove the "sign_zero_copy" module option and
always inline the pages when writing.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Commit 522440ed made cifs set backing_dev_info on the mapping attached
to new inodes. This change caused a fairly significant read performance
regression, as cifs started doing page-sized reads exclusively.

By virtue of the fact that they're allocated as part of cifs_sb_info by
kzalloc, the ra_pages on cifs BDIs get set to 0, which prevents any
readahead. This forces the normal read codepaths to use readpage instead
of readpages causing a four-fold increase in the number of read calls
with the default rsize.

Fix it by setting ra_pages in the BDI to the same value as that in the
default_backing_dev_info.

Fixes https://bugzilla.kernel.org/show_bug.cgi?id=31662

Cc: stable@kernel.org
Reported-and-Tested-by: NTill <till2.schaefer@uni-dortmund.de>
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

We artificially limited the user name to 32 bytes, but modern servers handle
larger.  Set the maximum length to a reasonable 256, and make the user name
string dynamically allocated rather than a fixed size in session structure.
Also clean up old checkpatch warning.
Signed-off-by: NSteve French <sfrench@us.ibm.com>

This flag currently only affects whether we allow "zero-copy" writes
with signing enabled. Typically we map pages in the pagecache directly
into the write request. If signing is enabled however and the contents
of the page change after the signature is calculated but before the
write is sent then the signature will be wrong. Servers typically
respond to this by closing down the socket.

Still, this can provide a performance benefit so the "Experimental" flag
was overloaded to allow this. That's really not a good place for this
option however since it's not clear what that flag does.

Move that flag instead to a new module parameter that better describes
its purpose. That's also better since it can be set at module insertion
time by configuring modprobe.d.
Reviewed-by: NSuresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

If we don't have Exclusive oplock we write a data to the server.
Also set invalidate_mapping flag on the inode if we wrote something
to the server. Add cifs_iovec_write to let the client write iovec
buffers through CIFSSMBWrite2.
Signed-off-by: NPavel Shilovsky <piastryyy@gmail.com>
Reviewed-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Read from the cache if we have at least Level II oplock - otherwise
read from the server. Add cifs_user_readv to let the client read into
iovec buffers.
Reviewed-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NPavel Shilovsky <piastryyy@gmail.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Invalidate inode mapping if we don't have at least Level II oplock.
Reviewed-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NPavel Shilovsky <piastryyy@gmail.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Invalidate inode mapping if we don't have at least Level II oplock in
cifs_strict_fsync. Also remove filemap_write_and_wait call from cifs_fsync
because it is previously called from vfs_fsync_range. Add file operations'
structures for strict cache mode.
Reviewed-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NPavel Shilovsky <piastryyy@gmail.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

If the server isn't responding to echoes, we don't want to leave tasks
hung waiting for it to reply. At that point, we'll want to reconnect
so that soft mounts can return an error to userspace quickly.

If the client hasn't received a reply after a specified number of echo
intervals, assume that the transport is down and attempt to reconnect
the socket.

The number of echo_intervals to wait before attempting to reconnect is
tunable via a module parameter. Setting it to 0, means that the client
will never attempt to reconnect. The default is 5.
Signed-off-by: NJeff Layton <jlayton@redhat.com>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Reduce false inode collisions by using the CreationTime like an
i_generation field. This way, even if the server ends up reusing
a uniqueid after a delete/create cycle, we can avoid matching
the inode incorrectly.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Signed-off-by: NNick Piggin <npiggin@kernel.dk>

RCU free the struct inode. This will allow:

- Subsequent store-free path walking patch. The inode must be consulted for
  permissions when walking, so an RCU inode reference is a must.
- sb_inode_list_lock to be moved inside i_lock because sb list walkers who want
  to take i_lock no longer need to take sb_inode_list_lock to walk the list in
  the first place. This will simplify and optimize locking.
- Could remove some nested trylock loops in dcache code
- Could potentially simplify things a bit in VM land. Do not need to take the
  page lock to follow page->mapping.

The downsides of this is the performance cost of using RCU. In a simple
creat/unlink microbenchmark, performance drops by about 10% due to inability to
reuse cache-hot slab objects. As iterations increase and RCU freeing starts
kicking over, this increases to about 20%.

In cases where inode lifetimes are longer (ie. many inodes may be allocated
during the average life span of a single inode), a lot of this cache reuse is
not applicable, so the regression caused by this patch is smaller.

The cache-hot regression could largely be avoided by using SLAB_DESTROY_BY_RCU,
however this adds some complexity to list walking and store-free path walking,
so I prefer to implement this at a later date, if it is shown to be a win in
real situations. I haven't found a regression in any non-micro benchmark so I
doubt it will be a problem.
Signed-off-by: NNick Piggin <npiggin@kernel.dk>

Make connect logic more ip-protocol independent and move RFC1001 stuff into
a separate function. Also replace union addr in TCP_Server_Info structure
with sockaddr_storage.
Signed-off-by: NPavel Shilovsky <piastryyy@gmail.com>
Reviewed-and-Tested-by: NJeff Layton <jlayton@samba.org>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

...this string is zeroed out and nothing ever changes it.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Currently, the attribute cache timeout for CIFS is hardcoded to 1 second. This
means that the client might have to issue a QPATHINFO/QFILEINFO call every 1
second to verify if something has changes, which seems too expensive. On the
other hand, if the timeout is hardcoded to a higher value, workloads that
expect strict cache coherency might see unexpected results.

Making attribute cache timeout as a tunable will allow us to make a tradeoff
between performance and cache metadata correctness depending on the
application/workload needs.

Add 'actimeo' tunable that can be used to tune the attribute cache timeout.
The default timeout is set to 1 second. Also, display actimeo option value in
/proc/mounts.

It appears to me that 'actimeo' and the proposed (but not yet merged)
'strictcache' option cannot coexist, so care must be taken that we reset the
other option if one of them is set.

Changes since last post:
   - fix option parsing and handle possible values correcly
Reviewed-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSuresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Reviewed-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSuresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

All the callers already have a pointer to struct cifsInodeInfo. Use it.
Signed-off-by: NSuresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: NPavel Shilovsky <piastryyy@gmail.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Radix trees are ideal when you want to track a bunch of pointers and
can't embed a tracking structure within the target of those pointers.
The tradeoff is an increase in memory, particularly if the tree is
sparse.

In CIFS, we use the tlink_tree to track tcon_link structs. A tcon_link
can never be in more than one tlink_tree, so there's no impediment to
using a rb_tree here instead of a radix tree.

Convert the new multiuser mount code to use a rb_tree instead. This
should reduce the memory required to manage the tlink_tree.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

The caller allocated it, the caller should free it.

The only issue so far is that we could change the flp pointer even on an
error return if the fl_change callback failed.  But we can simply move
the flp assignment after the fl_change invocation, as the callers don't
care about the flp return value if the setlease call failed.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

We modified setlease to require the caller to allocate the new lease in
the case of creating a new lease, but forgot to fix up the filesystem
methods.

Cc: Steven Whitehouse <swhiteho@redhat.com>
Cc: Steve French <sfrench@samba.org>
Cc: Trond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: NJ. Bruce Fields <bfields@redhat.com>
Acked-by: NArnd Bergmann <arnd@arndb.de>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

write_behind_rc is redundant and just adds complexity to the code. What
we really want to do instead is to use mapping_set_error to reset the
flags on the mapping when we find a writeback error and can't report it
to userspace yet.

For cifs_flush and cifs_fsync, we shouldn't reset the flags since errors
returned there do get reported to userspace.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Reviewed-by: NSuresh Jayaraman <sjayaraman@suse.de>
Reviewed-by: NDave Kleikamp <shaggy@linux.vnet.ibm.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

cifs_tcp_ses_lock is a rwlock with protects the cifs_tcp_ses_list,
server->smb_ses_list and the ses->tcon_list. It also protects a few
ref counters in server, ses and tcon. In most cases the critical section
doesn't seem to be large, in a few cases where it is slightly large, there
seem to be really no benefit from concurrent access. I briefly considered RCU
mechanism but it appears to me that there is no real need.

Replace it with a spinlock and get rid of the last rwlock in the cifs code.
Signed-off-by: NSuresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Convert this lock to a regular spinlock

A rwlock_t offers little value here. It's more expensive than a regular
spinlock unless you have a fairly large section of code that runs under
the read lock and can benefit from the concurrency.

Additionally, we need to ensure that the refcounting for files isn't
racy and to do that we need to lock areas that can increment it for
write. That means that the areas that can actually use a read_lock are
very few and relatively infrequently used.

While we're at it, change the name to something easier to type, and fix
a bug in find_writable_file. cifsFileInfo_put can sleep and shouldn't be
called while holding the lock.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Reviewed-by: NSuresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Filesystems aren't really supposed to do anything with a vfsmount. It's
considered a layering violation since vfsmounts are entirely managed at
the VFS layer.

CIFS currently keeps an active reference to a vfsmount in order to
prevent the superblock vanishing before an oplock break has completed.
What we really want to do instead is to keep sb->s_active high until the
oplock break has completed. This patch borrows the scheme that NFS uses
for handling sillyrenames.

An atomic_t is added to the cifs_sb_info. When it transitions from 0 to
1, an extra reference to the superblock is taken (by bumping the
s_active value). When it transitions from 1 to 0, that reference is
dropped and a the superblock teardown may proceed if there are no more
references to it.

Also, the vfsmount pointer is removed from cifsFileInfo and from
cifs_new_fileinfo, and some bogus forward declarations are removed from
cifsfs.h.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Reviewed-by: NSuresh Jayaraman <sjayaraman@suse.de>
Acked-by: NDave Kleikamp <shaggy@linux.vnet.ibm.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

...based on CIFS_MOUNT_MULTIUSER flag.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

cifsFileInfo needs a pointer to a tcon, but it doesn't currently hold a
reference to it. Change it to keep a pointer to a tcon_link instead and
hold a reference to it.

That will keep the tcon from being freed until the file is closed.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

This prepares the removal of the big kernel lock from the
file locking code. We still use the BKL as long as fs/lockd
uses it and ceph might sleep, but we can flip the definition
to a private spinlock as soon as that's done.
All users outside of fs/lockd get converted to use
lock_flocks() instead of lock_kernel() where appropriate.

Based on an earlier patch to use a spinlock from Matthew
Wilcox, who has attempted this a few times before, the
earliest patch from over 10 years ago turned it into
a semaphore, which ended up being slower than the BKL
and was subsequently reverted.

Someone should do some serious performance testing when
this becomes a spinlock, since this has caused problems
before. Using a spinlock should be at least as good
as the BKL in theory, but who knows...
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Acked-by: NMatthew Wilcox <willy@linux.intel.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: Trond Myklebust <trond.myklebust@fys.uio.no>
Cc: "J. Bruce Fields" <bfields@fieldses.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Miklos Szeredi <mszeredi@suse.cz>
Cc: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: John Kacur <jkacur@redhat.com>
Cc: Sage Weil <sage@newdream.net>
Cc: linux-kernel@vger.kernel.org
Cc: linux-fsdevel@vger.kernel.org

The BKL is only used in put_super and fill_super that are both protected by
the superblocks s_umount rw_semaphore. Therefore it is safe to remove the
BKL entirely.
Signed-off-by: NJan Blunck <jblunck@infradead.org>
Cc: Steve French <smfrench@gmail.com>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>

This patch is a preparation necessary to remove the BKL from do_new_mount().
It explicitly adds calls to lock_kernel()/unlock_kernel() around
get_sb/fill_super operations for filesystems that still uses the BKL.

I've read through all the code formerly covered by the BKL inside
do_kern_mount() and have satisfied myself that it doesn't need the BKL
any more.

do_kern_mount() is already called without the BKL when mounting the rootfs
and in nfsctl. do_kern_mount() calls vfs_kern_mount(), which is called
from various places without BKL: simple_pin_fs(), nfs_do_clone_mount()
through nfs_follow_mountpoint(), afs_mntpt_do_automount() through
afs_mntpt_follow_link(). Both later functions are actually the filesystems
follow_link inode operation. vfs_kern_mount() is calling the specified
get_sb function and lets the filesystem do its job by calling the given
fill_super function.

Therefore I think it is safe to push down the BKL from the VFS to the
low-level filesystems get_sb/fill_super operation.

[arnd: do not add the BKL to those file systems that already
       don't use it elsewhere]
Signed-off-by: NJan Blunck <jblunck@infradead.org>
Signed-off-by: NArnd Bergmann <arnd@arndb.de>
Cc: Matthew Wilcox <matthew@wil.cx>
Cc: Christoph Hellwig <hch@infradead.org>

At mount time, we'll always need to create a tcon that will serve as a
template for others that are associated with the mount. This tcon is
known as the "master" tcon.

In some cases, we'll need to use that tcon regardless of who's accessing
the mount. Add an accessor function for the master tcon and go ahead and
switch the appropriate places to use it.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

When we convert cifs to do multiple sessions per mount, we'll need more
than one tcon per superblock. At that point "cifs_sb->tcon" will make
no sense. Add a new accessor function that gets a tcon given a cifs_sb.
For now, it just returns cifs_sb->tcon. Later it'll do more.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

...where it's available and appropriate.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NSteve French <sfrench@us.ibm.com>

If registering fs cache failed, we weren't cleaning up proc.
Acked-by: NJeff Layton <jlayton@redhat.com>
CC: Suresh Jayaraman <sjayaraman@suse.de>
Signed-off-by: NSteve French <sfrench@us.ibm.com>