Fix a memory leak due to allocation in the XDR layer. In cases where the
RPC call needs to be retransmitted, we end up allocating new pages without
clearing the old ones. Fix this by moving the allocation into
nfs3_proc_setacls().

Also fix an issue discovered by Kevin Rudd, whereby the amount of memory
reserved for the acls in the xdr_buf->head was miscalculated, and causing
corruption.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

It appears that 'jiffies' timestamps do not have high enough resolution for
nfs_inode_attrs_need_update(). One problem is that a GETATTR can be
launched within < 1 jiffy of the last operation that updated the attribute.
Another problem is that RPC calls can take < 1 jiffy to execute.

We can fix this by switching the variables to use a simple global counter
that gets incremented every time we start another GETATTR call.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Peter Staubach suggested reducing NFS4_SETCLIENTID_NAMELEN by one byte so
as to avoid 7 bytes of unnecessary padding.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

The sc_name field is currently 56 bytes long.  This is not large enough
to hold a pair of IPv6 addresses, the authentication type, the protocol
name, and a uniquifier number.  The maximum possible size of the name
string using IPv6 addresses is just under 110 bytes, so I increased the
size of the sc_name field to accomodate this maximum.

In addition, the strings in the nfs4_setclientid structure are
constructed with scnprintf(), which wants to terminate its output with
'\0'.  The sc_netid field was large enough only for a three byte netid
string and a '\0' so inet6 netids were being truncated.  Perhaps we
don't need the overhead of scnprintf() to do a simple string copy, but
I fixed this by increasing the size of the buffer by one byte.

Since all three of the string buffers in nfs4_setclientid are
constructed with scnprintf(), I increased the size of all three by one
byte to document the requirement, although I don't think either the
universal address field or the name field will be so small that these
strings get truncated in this way.

The size of the Linux client's client ID on the wire will be larger
than before.  RFC 3530 suggests the size limit for client IDs is 1024,
and we are still well below that.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

remove 8 bytes of padding from struct nfs_fattr on 64 bit builds

This also removes padding from several nfs structures, including
16 bytes from  nfs4_opendata, nfs4_createdata,nfs3_createdata
& 8 bytes from nfs_read_data,nfs_write_data,nfs_removeres,nfs4_closedata

This also reduces the reported stack usage of many nfs functions (30+).
Signed-off-by: NRichard Kennedy <richard@rsk.demon.co.uk>
----

This patch is against the latest git 2.6.27-rc4.
I've built & run this on my AMD64 desktop, & successfully run _simple_
tests with a  64 bit client => 32 bit server & 32 bit client to 64 bit
server.

On fedora with gcc (GCC) 4.3.0 20080428 (Red Hat 4.3.0-8) checkpatch
reports 33 functions with reduced stack usage.
e.g.
__nfs_revalidate_inode [nfs] 216 => 200
_nfs4_proc_access [nfs] 304 => 288
_nfs4_proc_link [nfs] 536 => 504
_nfs4_proc_remove [nfs] 304 => 288
_nfs4_proc_rename [nfs] 584 => 552
nfs3_proc_access [nfs] 272 => 256
nfs3_proc_getacl [nfs] 384 => 368
nfs3_proc_link [nfs] 496 => 464
etc
I can supply the complete list if anyone is interested.

regards
Richard
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

All instances are set to nfs_open(), so we should just remove the redundant
indirection. Ditto for the file_release op
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

NFSv2 file locking currently fails the Connectathon tests, because the
calls to the VFS locking code do not return an EINVAL error if the
struct file_lock overflows the 32-bit boundaries.

The problem is due to the fact that we occasionally call helpers from
fs/locks.c in order to avoid RPC calls to the server when we know that a
local process holds the lock. These helpers are, of course, always
64-bit enabled, so EINVAL is not returned in cases when it would if
the call had gone to the NLM code.

For consistency, we therefore add support for a bounds-checking helper.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

It is quite possible that the OPEN, CLOSE, LOCK, LOCKU,... compounds fail
before the actual stateful operation has been executed (for instance in the
PUTFH call). There is no way to tell from the overall status result which
operations were executed from the COMPOUND.

The fix is to move incrementing of the sequence id into the XDR layer,
so that we do it as we process the results from the stateful operation.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

RPC protocol version numbers are unsigned.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Ensure that the RPC buffer size specified for NFSv4 SETCLIENTID procedures
matches what we are encoding into the buffer.  See the definition of
struct nfs4_setclientid {} and the encode_setclientid() function.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Move the common code for setting up the nfs_write_data and nfs_read_data
structures into fs/nfs/read.c, fs/nfs/write.c and fs/nfs/direct.c.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

NFSv2 and v4 don't offer weak cache consistency attributes on WRITE calls.
In NFSv3, returning wcc data is optional. In all cases, we want to prevent
the client from invalidating our cached data whenever ->write_done()
attempts to update the inode attributes.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

It doesn't really make sense to cache an access call without also
revalidating the attributes.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Fix a couple of bugs:
 - Don't rely on the parent dentry still being valid when the call completes.
   Fixes a race with shrink_dcache_for_umount_subtree()

 - Don't remove the file if the filehandle has been labelled as stale.

Fix a couple of inefficiencies
 - Remove the global list of sillyrenamed files. Instead we can cache the
   sillyrename information in the dentry->d_fsdata
 - Move common code from unlink_setup/unlink_done into fs/nfs/unlink.c
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

We need a common structure for setting up an unlink() rpc call in order to
fix the asynchronous unlink code.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Currently we just use a 32-bit counter.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

The Linux NFS4 client simply skips over the bitmask in an O_EXCL open
call and so it doesn't bother to reset any fields that may be holding
the verifier. This patch has us save the first two words of the bitmask
(which is all the current client has #defines for). The client then
later checks this bitmask and turns on the appropriate flags in the
sattr->ia_verify field for the following SETATTR call.

This patch only currently checks to see if the server used the atime
and mtime slots for the verifier (which is what the Linux server uses
for this). I'm not sure of what other fields the server could
reasonably use, but adding checks for others should be trivial.
Signed-off-by: NJeff Layton <jlayton@redhat.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Many struct inode_operations in the kernel can be "const".  Marking them const
moves these to the .rodata section, which avoids false sharing with potential
dirty data.  In addition it'll catch accidental writes at compile time to
these shared resources.
Signed-off-by: NArjan van de Ven <arjan@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

It makes no sense to maintain 2 parallel systems for reading in pages.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Maintaining two parallel ways of doing synchronous writes is rather
pointless. This patch gets rid of the legacy nfs_writepage_sync(), and
replaces it with the faster asynchronous writes.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Acked-by: NTrond Myklebust <trond.myklebust@fys.uio.no>
Acked-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

on-the-wire data is big-endian

[in large part pulled from Alexey's patch]
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>
Acked-by: NTrond Myklebust <trond.myklebust@fys.uio.no>
Acked-by: NNeil Brown <neilb@suse.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Now that we have a copy of the symlink path in the page cache, we can pass
a struct page down to the XDR routines instead of a string buffer.

Test plan:
Connectathon, all NFS versions.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

If the LOOKUP or GETATTR in nfs_instantiate fail, nfs_instantiate will do a
d_drop before returning.  But some callers already do a d_drop in the case
of an error return.  Make certain we do only one d_drop in all error paths.

This issue was introduced because over time, the symlink proc API diverged
slightly from the create/mkdir/mknod proc API.  To prevent other coding
mistakes of this type, change the symlink proc API to be more like
create/mkdir/mknod and move the nfs_instantiate call into the symlink proc
routines so it is used in exactly the same way for create, mkdir, mknod,
and symlink.

Test plan:
Connectathon, all versions of NFS.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

include/linux/sunrpc/clnt.h already includes include/linux/sunrpc/xprt.h.
We can remove xprt.h from source files that already include clnt.h.
Likewise include/linux/sunrpc/timer.h.

Test plan:
Compile kernel with CONFIG_NFS enabled.
Signed-off-by: NChuck Lever <chuck.lever@oracle.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Add some extra const qualifiers into NFS.
Signed-Off-By: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Add a set_capabilities NFS RPC op so that the server capabilities can be set.
Signed-Off-By: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Add a lookup filehandle NFS RPC op so that a file handle can be looked up
without requiring dentries and inodes and other VFS stuff when doing an NFS4
pathwalk during mounting.
Signed-Off-By: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

The logic in nfs_direct_read_schedule and nfs_direct_write_schedule can
allow data->npages to be one larger than rpages.  This causes a page
pointer to be written beyond the end of the pagevec in nfs_read_data (or
nfs_write_data).

Fix this by making nfs_(read|write)_alloc() calculate the size of the
pagevec array, and initialise data->npages.

Also get rid of the redundant argument to nfs_commit_alloc().
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
Cc: Chuck Lever <chuck.lever@oracle.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Neil Brown observed that the current limit of 32 bytes isn't enough to hold two
ip addresses and the rest of the stuff we're putting in it, so it's often
truncated to the point where it's unlikely to be unique.  This can cause
spurious CLID_INUSE's from the server.
Signed-off-by: NJ. Bruce Fields <bfields@citi.umich.edu>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
(cherry picked from fc8c17ec251e984ab3df9182ed097aa5b577c915 commit)

This reverts ccf01ef7 commit.

No idea how git managed this one: when I asked it to merge the odirect
topic branch it actually generated a patch which reverted the change.

Reverting the 'merge' will once again reveal Chuck's recent NFS/O_DIRECT
work to the world.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Neil Brown observed that the kmalloc() in nfs_get_user_pages() is more
likely to fail if the I/O is large enough to require the allocation of more
than a single page to keep track of all the pinned pages in the user's
buffer.

Instead of tracking one large page array per dreq/iocb, track pages per
nfs_read/write_data, just like the cached I/O path does.  An array for
pages is already allocated for us by nfs_readdata_alloc() (and the write
and commit equivalents).

This is also required for adding support for vectored I/O to the NFS direct
I/O path.

The original reason to pin the user buffer and allocate all the NFS data
structures before trying to schedule I/O was to ensure all needed resources
are allocated on the client before starting to send requests.  This reduces
the chance that resource exhaustion on the client will cause a short read
or write.

On the other hand, for an application making very large application I/O
requests, this means that it will be nearly impossible for the application
to make forward progress on a resource-limited client.

Thus, moving the buffer pinning functionality into the I/O scheduling
loops should be good for scalability.  The next patch will do the same for
NFS data structure allocation.
Signed-off-by: NChuck Lever <cel@netapp.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Respond to a moved error on NFS lookup by setting up the referral.
Note: We don't actually follow the referral during lookup/getattr, but
later when we detect fsid mismatch in inode revalidation (similar to the
processing done for cloning submounts). Referrals will have fake attributes
until they are actually followed or traversed.
Signed-off-by: NManoj Naik <manoj@almaden.ibm.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

Use component4-style formats for decoding list of servers and pathnames in
fs_locations.
Signed-off-by: NManoj Naik <manoj@almaden.ibm.com>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

NFSv4 allows for the fact that filesystems may be replicated across
several servers or that they may be migrated to a backup server in case of
failure of the primary server.
fs_locations is an NFSv4 operation for retrieving information about the
location of migrated and/or replicated filesystems.

Based on an initial implementation by Jiaying Zhang <jiayingz@citi.umich.edu>
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>

This should enable us to detect if we are crossing a mountpoint in the
case where the server is exporting "nohide" mounts.
Signed-off-by: NTrond Myklebust <Trond.Myklebust@netapp.com>