SGI-PV: 955302
SGI-Modid: xfs-linux-melb:xfs-kern:26746a
Signed-off-by: NEric Sandeen <sandeen@sandeen.net>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

SGI-PV: 953687
SGI-Modid: xfs-linux-melb:xfs-kern:26639a
Signed-off-by: NVlad Apostolov <vapo@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

SGI-PV: 953819
SGI-Modid: xfs-linux-melb:xfs-kern:26628a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

ramdisks.

SGI-PV: 954802
SGI-Modid: xfs-linux-melb:xfs-kern:26627a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

list, to increase our potential readahead window and in turn improve
bulkstat performance.

SGI-PV: 944409
SGI-Modid: xfs-linux-melb:xfs-kern:26607a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

batches of inode cluster buffers at once, before any blocking reads are
issued.

SGI-PV: 944409
SGI-Modid: xfs-linux-melb:xfs-kern:26606a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

SGI-PV: 944409
SGI-Modid: xfs-linux-melb:xfs-kern:26603a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

extract inline attributes out of the bulkstat buffer (for that case),
rather than using an (extremely expensive for large icount filesystems)
iget for fetching attrs.

SGI-PV: 944409
SGI-Modid: xfs-linux-melb:xfs-kern:26602a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

code.

SGI-PV: 954372
SGI-Modid: xfs-linux-melb:xfs-kern:26583a
Signed-off-by: NTim Shimmin <tes@sgi.com>

SGI-PV: 954366
SGI-Modid: xfs-linux-melb:xfs-kern:26565a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

current kernels

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26564a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

semantics.

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26563a
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26562a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

incore users.

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26561a
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

__be64 and let the callers use the proper macros.

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26560a
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26559a
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

conversion

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26558a
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

handling with sparse now, no need for comments.

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26557a
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26556a
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

used for ondisk values.

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26553a
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

overloaded gfp_t param.

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26552a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

SGI-PV: 954580
SGI-Modid: xfs-linux-melb:xfs-kern:26551a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

Avoids doing an unnecessary inode to vnode conversion and avoids a memory
allocation.

SGI-PV: 904196
SGI-Modid: xfs-linux-melb:xfs-kern:26492a
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

SGI-PV: 954365
SGI-Modid: xfs-linux-melb:xfs-kern:26406a
Signed-off-by: NTim Shimmin <tes@sgi.com>

analysis.

Under a sequential create+allocate workload, blktrace reported backward
writes being issued by xfsbufd, and frequent inappropriate queue unplugs.
We now insert at the tail when moving from the delwri lists to the temp
lists, which maintains correct ordering, and we avoid unplugging queues
deep in the submit paths when we'd shortly do it at a higher level anyway.
blktrace now reports much healthier write patterns from xfsbufd for this
workload (and likely many others).

SGI-PV: 954310
SGI-Modid: xfs-linux-melb:xfs-kern:26396a
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

"inode => vnode => inode" conversion, but only flags and mode of final
inode are looked at. Pass original inode instead.

SGI-PV: 904196
SGI-Modid: xfs-linux-melb:xfs-kern:26395a
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NNathan Scott <nathans@sgi.com>
Signed-off-by: NTim Shimmin <tes@sgi.com>

Ingo Oeser pointed out that because current expands to an inline function
it is more space efficient and somewhat faster to simply keep a cached copy
of current in another variable.  This patch implements that for the
de_thread function.

(akpm: saves nearly 100 bytes of text on x86)
Signed-off-by: NEric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Acked-by: NTrond Myklebust <Trond.Myklebust@netapp.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Add missing \n to dprintk
Signed-off-by: NMartin Bligh <mbligh@google.com>
Acked-by: NTrond Myklebust <trond.myklebust@fys.uio.no>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

In de_thread we move pids from one process to another, a rather ugly case.
The function transfer_pid makes it clear what we are doing, and makes the
action atomic.  This is useful we ever want to atomically traverse the
process group and session lists, in a rcu safe manner.

Even if the atomic properties this change should be a win as transfer_pid
should be less code to execute than executing both attach_pid and
detach_pid, and this should make de_thread slightly smaller as only a
single function call needs to be emitted.  The only downside is that the
code might be slower to execute as the odds are against transfer_pid being
in cache.
Signed-off-by: NEric W. Biederman <ebiederm@xmission.com>
Cc: Oleg Nesterov <oleg@tv-sign.ru>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Since sys_sysctl is deprecated start allow it to be compiled out.  This
should catch any remaining user space code that cares, and paves the way
for further sysctl cleanups.

[akpm@osdl.org: If sys_sysctl() is not compiled-in, emit a warning]
Signed-off-by: NEric W. Biederman <ebiederm@xmission.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

free_fdset(NULL, ...) is legal.
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Since the nolargeio option no longer has any effect, print a warning
instead of setting a write-only variable.
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Chris Mason <mason@suse.com>
Cc: Hans Reiser <reiser@namesys.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

This eliminates the i_blksize field from struct inode.  Filesystems that want
to provide a per-inode st_blksize can do so by providing their own getattr
routine instead of using the generic_fillattr() function.

Note that some filesystems were providing pretty much random (and incorrect)
values for i_blksize.

[bunk@stusta.de: cleanup]
[akpm@osdl.org: generic_fillattr() fix]
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: NAdrian Bunk <bunk@stusta.de>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Move the i_cdev pointer in struct inode into a union.
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Move the i_bdev pointer in struct inode into a union.
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

The following patches reduce the size of the VFS inode structure by 28 bytes
on a UP x86.  (It would be more on an x86_64 system).  This is a 10% reduction
in the inode size on a UP kernel that is configured in a production mode
(i.e., with no spinlock or other debugging functions enabled; if you want to
save memory taken up by in-core inodes, the first thing you should do is
disable the debugging options; they are responsible for a huge amount of bloat
in the VFS inode structure).

This patch:

The filesystem or device-specific pointer in the inode is inside a union,
which is pretty pointless given that all 30+ users of this field have been
using the void pointer.  Get rid of the union and rename it to i_private, with
a comment to explain who is allowed to use the void pointer.  This is just a
cleanup, but it allows us to reuse the union 'u' for something something where
the union will actually be used.

[judith@osdl.org: powerpc build fix]
Signed-off-by: N"Theodore Ts'o" <tytso@mit.edu>
Signed-off-by: NJudith Lebzelter <judith@osdl.org>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

get_blocks() was removed.  So, this removes it on fat, and will take
advantage of the multi block mapping.
Signed-off-by: NOGAWA Hirofumi <hirofumi@mail.parknet.co.jp>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

For a long time now I have had a problem with not being able to return a
lookup failure on an existsing directory.  In autofs this corresponds to a
mount failure on a autofs managed mount entry that is browsable (and so the
mount point directory exists).

While this problem has been present for a long time I've avoided resolving
it because it was not very visible.  But now that autofs v5 has "mount and
expire on demand" of nested multiple mounts, such as is found when mounting
an export list from a server, solving the problem cannot be avoided any
longer.

I've tried very hard to find a way to do this entirely within the autofs4
module but have not been able to find a satisfactory way to achieve it.

So, I need to propose a change to the VFS.
Signed-off-by: NIan Kent <raven@themaw.net>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>

Move the fallback arch_vma_name() to a sensible place (kernel/signal.c).

Currently it's in fs/proc/task_mmu.c, a file that is dependent on both
CONFIG_PROC_FS and CONFIG_MMU being enabled, but it's used from
kernel/signal.c from where it is called unconditionally.

[akpm@osdl.org: build fix]
Signed-off-by: NDavid Howells <dhowells@redhat.com>
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>