I'm running a shmem pagefault test case (see attached file) under a 64 CPU
system.  Profile shows shmem_inode_info->lock is heavily contented and
100% CPUs time are trying to get the lock.  In the pagefault (no swap)
case, shmem_getpage gets the lock twice, the last one is avoidable if we
prealloc a page so we could reduce one time of locking.  This is what
below patch does.

The result of the test case:
2.6.35-rc3: ~20s
2.6.35-rc3 + patch: ~12s
so this is 40% improvement.

One might argue if we could have better locking for shmem.  But even shmem
is lockless, the pagefault will soon have pagecache lock heavily contented
because shmem must add new page to pagecache.  So before we have better
locking for pagecache, improving shmem locking doesn't have too much
improvement.  I did a similar pagefault test against a ramfs file, the
test result is ~10.5s.

[akpm@linux-foundation.org: fix comment, clean up code layout, elimintate code duplication]
Signed-off-by: NShaohua Li <shaohua.li@intel.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: "Zhang, Yanmin" <yanmin.zhang@intel.com>
Cc: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

The current implementation of tmpfs is not scalable.  We found that
stat_lock is contended by multiple threads when we need to get a new page,
leading to useless spinning inside this spin lock.

This patch makes use of the percpu_counter library to maintain local count
of used blocks to speed up getting and returning of pages.  So the
acquisition of stat_lock is unnecessary for getting and returning blocks,
improving the performance of tmpfs on system with large number of cpus.
On a 4 socket 32 core NHM-EX system, we saw improvement of 270%.

The implementation below has a slight chance of race between threads
causing a slight overshoot of the maximum configured blocks.  However, any
overshoot is small, and is bounded by the number of cpus.  This happens
when the number of used blocks is slightly below the maximum configured
blocks when a thread checks the used block count, and another thread
allocates the last block before the current thread does.  This should not
be a problem for tmpfs, as the overshoot is most likely to be a few blocks
and bounded.  If a strict limit is really desired, then configured the max
blocks to be the limit less the number of cpus in system.
Signed-off-by: NTim Chen <tim.c.chen@linux.intel.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

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

Make sure we check the truncate constraints early on in ->setattr by adding
those checks to inode_change_ok.  Also clean up and document inode_change_ok
to make this obvious.

As a fallout we don't have to call inode_newsize_ok from simple_setsize and
simplify it down to a truncate_setsize which doesn't return an error.  This
simplifies a lot of setattr implementations and means we use truncate_setsize
almost everywhere.  Get rid of fat_setsize now that it's trivial and mark
ext2_setsize static to make the calling convention obvious.

Keep the inode_newsize_ok in vmtruncate for now as all callers need an
audit for its removal anyway.

Note: setattr code in ecryptfs doesn't call inode_change_ok at all and
needs a deeper audit, but that is left for later.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Make sure we call inode_change_ok before doing any changes in ->setattr,
and make sure to call it even if our fs wants to ignore normal UNIX
permissions, but use the ATTR_FORCE to skip those.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Despite its name it's now a generic implementation of ->setattr, but
rather a helper to copy attributes from a struct iattr to the inode.
Rename it to setattr_copy to reflect this fact.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

mtime and ctime should be changed only if the file size has actually
changed. Patches changing ext2 and tmpfs from vmtruncate to new truncate
sequence has caused regressions where they always update timestamps.

There is some strange cases in POSIX where truncate(2) must not update
times unless the size has acutally changed, see 6e656be8.

This area is all still rather buggy in different ways in a lot of
filesystems and needs a cleanup and audit (ideally the vfs will provide
a simple attribute or call to direct all filesystems exactly which
attributes to change). But coming up with the best solution will take a
while and is not appropriate for rc anyway.

So fix recent regression for now.
Signed-off-by: NNick Piggin <npiggin@suse.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Cc: Christoph Hellwig <hch@lst.de>
Acked-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NNick Piggin <npiggin@suse.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

We don't name our generic fsync implementations very well currently.
The no-op implementation for in-memory filesystems currently is called
simple_sync_file which doesn't make too much sense to start with,
the the generic one for simple filesystems is called simple_fsync
which can lead to some confusion.

This patch renames the generic file fsync method to generic_file_fsync
to match the other generic_file_* routines it is supposed to be used
with, and the no-op implementation to noop_fsync to make it obvious
what to expect.  In addition add some documentation for both methods.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

This patch adds support for moving charge of file pages, which include
normal file, tmpfs file and swaps of tmpfs file.  It's enabled by setting
bit 1 of <target cgroup>/memory.move_charge_at_immigrate.

Unlike the case of anonymous pages, file pages(and swaps) in the range
mmapped by the task will be moved even if the task hasn't done page fault,
i.e.  they might not be the task's "RSS", but other task's "RSS" that maps
the same file.  And mapcount of the page is ignored(the page can be moved
even if page_mapcount(page) > 1).  So, conditions that the page/swap
should be met to be moved is that it must be in the range mmapped by the
target task and it must be charged to the old cgroup.

[akpm@linux-foundation.org: coding-style fixes]
[akpm@linux-foundation.org: fix warning]
Signed-off-by: NDaisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Acked-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Balbir Singh <balbir@in.ibm.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

prep_new_page() will call set_page_private(page, 0) to initialise the
page, so the code is redundant.
Signed-off-by: NHuang Shijie <shijie8@gmail.com>
Reviewed-by: NMinchan Kim <minchan.kim@gmail.com>
Acked-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

- seems what ramfs_get_inode is only locally, make it static.
[AV: the hell it is; it's used by shmem, so shmem needed conversion too
and no, that function can't be made static]
Signed-off-by: NDmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

The entries in xattr handler table should be immutable (ie const)
like other operation tables.

Later patches convert common filesystems. Uncoverted filesystems
will still work, but will generate a compiler warning.
Signed-off-by: NStephen Hemminger <shemminger@vyatta.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Replacing
	error = 0;
	if (error)
		op
with nothing is not quite an equivalent transformation ;-)
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Now that we cache the ACL pointers in the generic inode all the generic_acl
cruft can go away and generic_acl.c can directly implement xattr handlers
dealing with the full Posix ACL semantics for in-memory filesystems.
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

Add a flags argument to struct xattr_handler and pass it to all xattr
handler methods.  This allows using the same methods for multiple
handlers, e.g. for the ACL methods which perform exactly the same action
for the access and default ACLs, just using a different underlying
attribute.  With a little more groundwork it'll also allow sharing the
methods for the regular user/trusted/secure handlers in extN, ocfs2 and
jffs2 like it's already done for xfs in this patch.

Also change the inode argument to the handlers to a dentry to allow
using the handlers mechnism for filesystems that require it later,
e.g. cifs.

[with GFS2 bits updated by Steven Whitehouse <swhiteho@redhat.com>]
Signed-off-by: NChristoph Hellwig <hch@lst.de>
Reviewed-by: NJames Morris <jmorris@namei.org>
Acked-by: NJoel Becker <joel.becker@oracle.com>
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

There are 2 groups of alloc_file() callers:
	* ones that are followed by ima_counts_get
	* ones giving non-regular files
So let's pull that ima_counts_get() into alloc_file();
it's a no-op in case of non-regular files.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

... and have the caller grab both mnt and dentry; kill
leak in infiniband, while we are at it.
Signed-off-by: NAl Viro <viro@zeniv.linux.org.uk>

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

While we're fiddling with the swap_map values, let's assign a particular
value to shmem/tmpfs swap pages: their swap counts are never incremented,
and it helps swapoff's try_to_unuse() a little if it can immediately
distinguish those pages from process pages.

Since we've no use for SWAP_MAP_BAD | COUNT_CONTINUED,
we might as well use that 0xbf value for SWAP_MAP_SHMEM.
Signed-off-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Reviewed-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

* mark struct vm_area_struct::vm_ops as const
* mark vm_ops in AGP code

But leave TTM code alone, something is fishy there with global vm_ops
being used.
Signed-off-by: NAlexey Dobriyan <adobriyan@gmail.com>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Signed-off-by: NJens Axboe <jens.axboe@oracle.com>

Fixes the following kmemcheck false positive (the compiler is using
a 32-bit mov to load the 16-bit sbinfo->mode in shmem_fill_super):

[    0.337000] Total of 1 processors activated (3088.38 BogoMIPS).
[    0.352000] CPU0 attaching NULL sched-domain.
[    0.360000] WARNING: kmemcheck: Caught 32-bit read from uninitialized
memory (9f8020fc)
[    0.361000]
a44240820000000041f6998100000000000000000000000000000000ff030000
[    0.368000]  i i i i i i i i i i i i i i i i u u u u i i i i i i i i i i u
u
[    0.375000]                                                          ^
[    0.376000]
[    0.377000] Pid: 9, comm: khelper Not tainted (2.6.31-tip #206) P4DC6
[    0.378000] EIP: 0060:[<810a3a95>] EFLAGS: 00010246 CPU: 0
[    0.379000] EIP is at shmem_fill_super+0xb5/0x120
[    0.380000] EAX: 00000000 EBX: 9f845400 ECX: 824042a4 EDX: 8199f641
[    0.381000] ESI: 9f8020c0 EDI: 9f845400 EBP: 9f81af68 ESP: 81cd6eec
[    0.382000]  DS: 007b ES: 007b FS: 00d8 GS: 0000 SS: 0068
[    0.383000] CR0: 8005003b CR2: 9f806200 CR3: 01ccd000 CR4: 000006d0
[    0.384000] DR0: 00000000 DR1: 00000000 DR2: 00000000 DR3: 00000000
[    0.385000] DR6: ffff4ff0 DR7: 00000400
[    0.386000]  [<810c25fc>] get_sb_nodev+0x3c/0x80
[    0.388000]  [<810a3514>] shmem_get_sb+0x14/0x20
[    0.390000]  [<810c207f>] vfs_kern_mount+0x4f/0x120
[    0.392000]  [<81b2849e>] init_tmpfs+0x7e/0xb0
[    0.394000]  [<81b11597>] do_basic_setup+0x17/0x30
[    0.396000]  [<81b11907>] kernel_init+0x57/0xa0
[    0.398000]  [<810039b7>] kernel_thread_helper+0x7/0x10
[    0.400000]  [<ffffffff>] 0xffffffff
[    0.402000] khelper used greatest stack depth: 2820 bytes left
[    0.407000] calling  init_mmap_min_addr+0x0/0x10 @ 1
[    0.408000] initcall init_mmap_min_addr+0x0/0x10 returned 0 after 0 usecs
Reported-by: NIngo Molnar <mingo@elte.hu>
Analysed-by: NVegard Nossum <vegard.nossum@gmail.com>
Signed-off-by: NPekka Enberg <penberg@cs.helsinki.fi>
Acked-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

CONFIG_SHMEM off gives you (ramfs masquerading as) tmpfs, even when
CONFIG_TMPFS is off: that's a little anomalous, and I'd intended to make
more sense of it by removing CONFIG_TMPFS altogether, always enabling its
code when CONFIG_SHMEM; but so many defconfigs have CONFIG_SHMEM on
CONFIG_TMPFS off that we'd better leave that as is.

But there is no point in asking for CONFIG_TMPFS if CONFIG_SHMEM is off:
make TMPFS depend on SHMEM, which also prevents TMPFS_POSIX_ACL
shmem_acl.o being pointlessly built into the kernel when SHMEM is off.

And a selfish change, to prevent the world from being rebuilt when I
switch between CONFIG_SHMEM on and off: the only CONFIG_SHMEM in the
header files is mm.h shmem_lock() - give that a shmem.c stub instead.
Signed-off-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Acked-by: NMatt Mackall <mpm@selenic.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix the following 'make includecheck' warning:

  mm/shmem.c: linux/vfs.h is included more than once.
Signed-off-by: NJaswinder Singh Rajput <jaswinderrajput@gmail.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

After commit 355cfa73 ("mm: modify swap_map and add SWAP_HAS_CACHE flag"),
only the context which have set SWAP_HAS_CACHE flag by swapcache_prepare()
or get_swap_page() would call add_to_swap_cache().  So add_to_swap_cache()
doesn't return -EEXIST any more.

Even though it doesn't return -EEXIST, it's not good behavior conceptually
to call swapcache_prepare() in the -EEXIST case, because it means clearing
SWAP_HAS_CACHE flag while the entry is on swap cache.

This patch removes redundant codes and comments from callers of it, and
adds VM_BUG_ON() in error path of add_to_swap_cache() and some comments.
Signed-off-by: NDaisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Reviewed-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Balbir Singh <balbir@linux.vnet.ibm.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Enable removing of corrupted pages through truncation
for a bunch of file systems: ext*, xfs, gfs2, ocfs2, ntfs
These should cover most server needs.

I chose the set of migration aware file systems for this
for now, assuming they have been especially audited.
But in general it should be safe for all file systems
on the data area that support read/write and truncate.

Caveat: the hardware error handler does not take i_mutex
for now before calling the truncate function. Is that ok?

Cc: tytso@mit.edu
Cc: hch@infradead.org
Cc: mfasheh@suse.com
Cc: aia21@cantab.net
Cc: hugh.dickins@tiscali.co.uk
Cc: swhiteho@redhat.com
Signed-off-by: NAndi Kleen <ak@linux.intel.com>

The dirtying of page and set_page_dirty() can be moved into the page lock.

- In shmem_write_end(), the page was dirtied while the page lock was held,
  but it's being marked dirty just after dropping the page lock.
- In shmem_symlink(), both dirtying and marking can be moved into page lock.

It's valuable for the hwpoison code to know whether one bad page can be dropped
without losing data. It mainly judges by testing the PG_dirty bit after taking
the page lock. So it becomes important that the dirtying of page and the
marking of dirtiness are both done inside the page lock. Which is a common
practice, but sadly not a rule.

The noticeable exceptions are
- mapped pages
- pages with buffer_heads
The above pages could go dirty at any time. Fortunately the hwpoison will
unmap the page and release the buffer_heads beforehand anyway.

Many other types of pages (eg. metadata pages) can also be dirtied at will by
their owners, the hwpoison code cannot do meaningful things to them anyway.
Only the dirtiness of pagecache pages owned by regular files are interested.

v2: AK: Add comment about set_page_dirty rules (suggested by Peter Zijlstra)
Acked-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Reviewed-by: NWANG Cong <xiyou.wangcong@gmail.com>
Signed-off-by: NWu Fengguang <fengguang.wu@intel.com>
Signed-off-by: NAndi Kleen <ak@linux.intel.com>

Devtmpfs lets the kernel create a tmpfs instance called devtmpfs
very early at kernel initialization, before any driver-core device
is registered. Every device with a major/minor will provide a
device node in devtmpfs.

Devtmpfs can be changed and altered by userspace at any time,
and in any way needed - just like today's udev-mounted tmpfs.
Unmodified udev versions will run just fine on top of it, and will
recognize an already existing kernel-created device node and use it.
The default node permissions are root:root 0600. Proper permissions
and user/group ownership, meaningful symlinks, all other policy still
needs to be applied by userspace.

If a node is created by devtmps, devtmpfs will remove the device node
when the device goes away. If the device node was created by
userspace, or the devtmpfs created node was replaced by userspace, it
will no longer be removed by devtmpfs.

If it is requested to auto-mount it, it makes init=/bin/sh work
without any further userspace support. /dev will be fully populated
and dynamic, and always reflect the current device state of the kernel.
With the commonly used dynamic device numbers, it solves the problem
where static devices nodes may point to the wrong devices.

It is intended to make the initial bootup logic simpler and more robust,
by de-coupling the creation of the inital environment, to reliably run
userspace processes, from a complex userspace bootstrap logic to provide
a working /dev.
Signed-off-by: NKay Sievers <kay.sievers@vrfy.org>
Signed-off-by: NJan Blunck <jblunck@suse.de>
Tested-By: NHarald Hoyer <harald@redhat.com>
Tested-By: NScott James Remnant <scott@ubuntu.com>
Signed-off-by: NGreg Kroah-Hartman <gregkh@suse.de>

shmfs wants purely standard POSIX ACL semantics, so we can use the new
generic VFS layer POSIX ACL checking rather than cooking our own
'permission()' function.
Reviewed-by: NJames Morris <jmorris@namei.org>
Acked-by: NSerge Hallyn <serue@us.ibm.com>
Acked-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

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

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

As function shmem_file_setup does not modify/allocate/free/pass given
filename - mark it as const.
Signed-off-by: NSergei Trofimovich <slyfox@inbox.ru>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

In a following patch, the usage of swap cache is recorded into swap_map.
This patch is for necessary interface changes to do that.

2 interfaces:

  - swapcache_prepare()
  - swapcache_free()

are added for allocating/freeing refcnt from swap-cache to existing swap
entries.  But implementation itself is not changed under this patch.  At
adding swapcache_free(), memcg's hook code is moved under
swapcache_free().  This is better than using scattered hooks.
Signed-off-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Reviewed-by: NDaisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Acked-by: NBalbir Singh <balbir@in.ibm.com>
Cc: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Dhaval Giani <dhaval@linux.vnet.ibm.com>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Based on discussion on lkml (Andrew Morton and Eric Paris),
move ima_counts_get down a layer into shmem/hugetlb__file_setup().
Resolves drm shmem_file_setup() usage case as well.

HD comment:
  I still think you're doing this at the wrong level, but recognize
  that you probably won't be persuaded until a few more users of
  alloc_file() emerge, all wanting your ima_counts_get().

  Resolving GEM's shmem_file_setup() is an improvement, so I'll say
Acked-by: NHugh Dickins <hugh.dickins@tiscali.co.uk>
Signed-off-by: NMimi Zohar <zohar@us.ibm.com>
Signed-off-by: NJames Morris <jmorris@namei.org>

- Add support in ima_path_check() for integrity checking without
incrementing the counts. (Required for nfsd.)
- rename and export opencount_get to ima_counts_get
- replace ima_shm_check calls with ima_counts_get
- export ima_path_check
Signed-off-by: NMimi Zohar <zohar@us.ibm.com>
Signed-off-by: NJames Morris <jmorris@namei.org>

Current mem_cgroup_shrink_usage() has two problems.

1. It doesn't call mem_cgroup_out_of_memory and doesn't update
   last_oom_jiffies, so pagefault_out_of_memory invokes global OOM.

2. Considering hierarchy, shrinking has to be done from the
   mem_over_limit, not from the memcg which the page would be charged to.

mem_cgroup_try_charge_swapin() does all of these things properly, so we
use it and call cancel_charge_swapin when it succeeded.

The name of "shrink_usage" is not appropriate for this behavior, so we
change it too.
Signed-off-by: NDaisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Acked-by: NKAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Li Zefan <lizf@cn.fujitsu.cn>
Cc: Paul Menage <menage@google.com>
Cc: Dhaval Giani <dhaval@linux.vnet.ibm.com>
Cc: Daisuke Nishimura <nishimura@mxp.nes.nec.co.jp>
Cc: YAMAMOTO Takashi <yamamoto@valinux.co.jp>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: David Rientjes <rientjes@google.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

SHMEM_MAX_BYTES was derived from the maximum size of its triple-indirect
swap vector, forgetting to take the MAX_LFS_FILESIZE limit into account.
Never mind 256kB pages, even 8kB pages on 32-bit kernels allowed files to
grow slightly bigger than that supposed maximum.

Fix this by using the min of both (at build time not run time).  And it
happens that this calculation is good as far as 8MB pages on 32-bit or
16MB pages on 64-bit: though SHMSWP_MAX_INDEX gets truncated before that,
it's truncated to such large numbers that we don't need to care.

[akpm@linux-foundation.org: it needs pagemap.h]
[akpm@linux-foundation.org: fix sparc64 min() warnings]
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Cc: Yuri Tikhonov <yur@emcraft.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Randy Dunlap <randy.dunlap@oracle.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Fix a division by zero which we have in shmem_truncate_range() and
shmem_unuse_inode() when using big PAGE_SIZE values (e.g.  256kB on
ppc44x).

With 256kB PAGE_SIZE, the ENTRIES_PER_PAGEPAGE constant becomes too large
(0x1.0000.0000) on a 32-bit kernel, so this patch just changes its type
from 'unsigned long' to 'unsigned long long'.

Hugh: reverted its unsigned long longs in shmem_truncate_range() and
shmem_getpage(): the pagecache index cannot be more than an unsigned long,
so the divisions by zero occurred in unreached code.  It's a pity we need
any ULL arithmetic here, but I found no pretty way to avoid it.
Signed-off-by: NYuri Tikhonov <yur@emcraft.com>
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Cc: Paul Mackerras <paulus@samba.org>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>

Synopsis: if shmem_writepage calls swap_writepage directly, most shmem
swap loads benefit, and a catastrophic interaction between SLUB and some
flash storage is avoided.

shmem_writepage() has always been peculiar in making no attempt to write:
it has just transferred a shmem page from file cache to swap cache, then
let that page make its way around the LRU again before being written and
freed.

The idea was that people use tmpfs because they want those pages to stay
in RAM; so although we give it an overflow to swap, we should resist
writing too soon, giving those pages a second chance before they can be
reclaimed.

That was always questionable, and I've toyed with this patch for years;
but never had a clear justification to depart from the original design.

It became more questionable in 2.6.28, when the split LRU patches classed
shmem and tmpfs pages as SwapBacked rather than as file_cache: that in
itself gives them more resistance to reclaim than normal file pages.  I
prepared this patch for 2.6.29, but the merge window arrived before I'd
completed gathering statistics to justify sending it in.

Then while comparing SLQB against SLUB, running SLUB on a laptop I'd
habitually used with SLAB, I found SLUB to run my tmpfs kbuild swapping
tests five times slower than SLAB or SLQB - other machines slower too, but
nowhere near so bad.  Simpler "cp -a" swapping tests showed the same.

slub_max_order=0 brings sanity to all, but heavy swapping is too far from
normal to justify such a tuning.  The crucial factor on that laptop turns
out to be that I'm using an SD card for swap.  What happens is this:

By default, SLUB uses order-2 pages for shmem_inode_cache (and many other
fs inodes), so creating tmpfs files under memory pressure brings lumpy
reclaim into play.  One subpage of the order is chosen from the bottom of
the LRU as usual, then the other three picked out from their random
positions on the LRUs.

In a tmpfs load, many of these pages will be ones which already passed
through shmem_writepage, so already have swap allocated.  And though their
offsets on swap were probably allocated sequentially, now that the pages
are picked off at random, their swap offsets are scattered.

But the flash storage on the SD card is very sensitive to having its
writes merged: once swap is written at scattered offsets, performance
falls apart.  Rotating disk seeks increase too, but less disastrously.

So: stop giving shmem/tmpfs pages a second pass around the LRU, write them
out to swap as soon as their swap has been allocated.

It's surely possible to devise an artificial load which runs faster the
old way, one whose sizing is such that the tmpfs pages on their second
pass are the ones that are wanted again, and other pages not.

But I've not yet found such a load: on all machines, under the loads I've
tried, immediate swap_writepage speeds up shmem swapping: especially when
using the SLUB allocator (and more effectively than slub_max_order=0), but
also with the others; and it also reduces the variance between runs.  How
much faster varies widely: a factor of five is rare, 5% is common.

One load which might have suffered: imagine a swapping shmem load in a
limited mem_cgroup on a machine with plenty of memory.  Before 2.6.29 the
swapcache was not charged, and such a load would have run quickest with
the shmem swapcache never written to swap.  But now swapcache is charged,
so even this load benefits from shmem_writepage directly to swap.

Apologies for the #ifndef CONFIG_SWAP swap_writepage() stub in swap.h:
it's silly because that will never get called; but refactoring shmem.c
sensibly according to CONFIG_SWAP will be a separate task.
Signed-off-by: NHugh Dickins <hugh@veritas.com>
Acked-by: NPekka Enberg <penberg@cs.helsinki.fi>
Acked-by: NRik van Riel <riel@redhat.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NLinus Torvalds <torvalds@linux-foundation.org>