Merge first patch-bomb from Andrew Morton:
- A few hotfixes which missed 4.4 becasue I was asleep. cc'ed to
-stable
- A few misc fixes
- OCFS2 updates
- Part of MM. Including pretty large changes to page-flags handling
and to thp management which have been buffered up for 2-3 cycles now.
I have a lot of MM material this time.
[ It turns out the THP part wasn't quite ready, so that got dropped from
this series - Linus ]
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (117 commits)
zsmalloc: reorganize struct size_class to pack 4 bytes hole
mm/zbud.c: use list_last_entry() instead of list_tail_entry()
zram/zcomp: do not zero out zcomp private pages
zram: pass gfp from zcomp frontend to backend
zram: try vmalloc() after kmalloc()
zram/zcomp: use GFP_NOIO to allocate streams
mm: add tracepoint for scanning pages
drivers/base/memory.c: fix kernel warning during memory hotplug on ppc64
mm/page_isolation: use macro to judge the alignment
mm: fix noisy sparse warning in LIBCFS_ALLOC_PRE()
mm: rework virtual memory accounting
include/linux/memblock.h: fix ordering of 'flags' argument in comments
mm: move lru_to_page to mm_inline.h
Documentation/filesystems: describe the shared memory usage/accounting
memory-hotplug: don't BUG() in register_memory_resource()
hugetlb: make mm and fs code explicitly non-modular
mm/swapfile.c: use list_for_each_entry_safe in free_swap_count_continuations
mm: /proc/pid/clear_refs: no need to clear VM_SOFTDIRTY in clear_soft_dirty_pmd()
mm: make sure isolate_lru_page() is never called for tail page
vmstat: make vmstat_updater deferrable again and shut down on idle
...
Following the previous patch, further reduction of /proc/pid/smaps cost
is possible for private writable shmem mappings with unpopulated areas
where the page walk invokes the .pte_hole function. We can use radix
tree iterator for each such area instead of calling find_get_entry() in
a loop. This is possible at the extra maintenance cost of introducing
another shmem function shmem_partial_swap_usage().
To demonstrate the diference, I have measured this on a process that
creates a private writable 2GB mapping of a partially swapped out
/dev/shm/file (which cannot employ the optimizations from the prvious
patch) and doesn't populate it at all. I time how long does it take to
cat /proc/pid/smaps of this process 100 times.
Before this patch:
real 0m3.831s
user 0m0.180s
sys 0m3.212s
After this patch:
real 0m1.176s
user 0m0.180s
sys 0m0.684s
The time is similar to the case where a radix tree iterator is employed
on the whole mapping.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The previous patch has improved swap accounting for shmem mapping, which
however made /proc/pid/smaps more expensive for shmem mappings, as we
consult the radix tree for each pte_none entry, so the overal complexity
is O(n*log(n)).
We can reduce this significantly for mappings that cannot contain COWed
pages, because then we can either use the statistics tha shmem object
itself tracks (if the mapping contains the whole object, or the swap
usage of the whole object is zero), or use the radix tree iterator,
which is much more effective than repeated find_get_entry() calls.
This patch therefore introduces a function shmem_swap_usage(vma) and
makes /proc/pid/smaps use it when possible. Only for writable private
mappings of shmem objects (i.e. tmpfs files) with the shmem object
itself (partially) swapped outwe have to resort to the find_get_entry()
approach.
Hopefully such mappings are relatively uncommon.
To demonstrate the diference, I have measured this on a process that
creates a 2GB mapping and dirties single pages with a stride of 2MB, and
time how long does it take to cat /proc/pid/smaps of this process 100
times.
Private writable mapping of a /dev/shm/file (the most complex case):
real 0m3.831s
user 0m0.180s
sys 0m3.212s
Shared mapping of an almost full mapping of a partially swapped /dev/shm/file
(which needs to employ the radix tree iterator).
real 0m1.351s
user 0m0.096s
sys 0m0.768s
Same, but with /dev/shm/file not swapped (so no radix tree walk needed)
real 0m0.935s
user 0m0.128s
sys 0m0.344s
Private anonymous mapping:
real 0m0.949s
user 0m0.116s
sys 0m0.348s
The cost is now much closer to the private anonymous mapping case, unless
the shmem mapping is private and writable.
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Cc: Jerome Marchand <jmarchan@redhat.com>
Cc: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Acked-by: Michal Hocko <mhocko@suse.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Mark those kmem allocations that are known to be easily triggered from
userspace as __GFP_ACCOUNT/SLAB_ACCOUNT, which makes them accounted to
memcg. For the list, see below:
- threadinfo
- task_struct
- task_delay_info
- pid
- cred
- mm_struct
- vm_area_struct and vm_region (nommu)
- anon_vma and anon_vma_chain
- signal_struct
- sighand_struct
- fs_struct
- files_struct
- fdtable and fdtable->full_fds_bits
- dentry and external_name
- inode for all filesystems. This is the most tedious part, because
most filesystems overwrite the alloc_inode method.
The list is far from complete, so feel free to add more objects.
Nevertheless, it should be close to "account everything" approach and
keep most workloads within bounds. Malevolent users will be able to
breach the limit, but this was possible even with the former "account
everything" approach (simply because it did not account everything in
fact).
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Vladimir Davydov <vdavydov@virtuozzo.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Greg Thelen <gthelen@google.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
inode_nohighmem() is sufficient to make sure that page_get_link()
won't try to allocate a highmem page. Moreover, it is sufficient
to make sure that page_symlink/__page_symlink won't do the same
thing. However, any filesystem that manually preseeds the symlink's
page cache upon symlink(2) needs to make sure that the page it
inserts there won't be a highmem one.
Fortunately, only nfs and shmem have run afoul of that...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull vfs xattr updates from Al Viro:
"Andreas' xattr cleanup series.
It's a followup to his xattr work that went in last cycle; -0.5KLoC"
* 'work.xattr' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
xattr handlers: Simplify list operation
ocfs2: Replace list xattr handler operations
nfs: Move call to security_inode_listsecurity into nfs_listxattr
xfs: Change how listxattr generates synthetic attributes
tmpfs: listxattr should include POSIX ACL xattrs
tmpfs: Use xattr handler infrastructure
btrfs: Use xattr handler infrastructure
vfs: Distinguish between full xattr names and proper prefixes
posix acls: Remove duplicate xattr name definitions
gfs2: Remove gfs2_xattr_acl_chmod
vfs: Remove vfs_xattr_cmp
Pull vfs RCU symlink updates from Al Viro:
"Replacement of ->follow_link/->put_link, allowing to stay in RCU mode
even if the symlink is not an embedded one.
No changes since the mailbomb on Jan 1"
* 'work.symlinks' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
switch ->get_link() to delayed_call, kill ->put_link()
kill free_page_put_link()
teach nfs_get_link() to work in RCU mode
teach proc_self_get_link()/proc_thread_self_get_link() to work in RCU mode
teach shmem_get_link() to work in RCU mode
teach page_get_link() to work in RCU mode
replace ->follow_link() with new method that could stay in RCU mode
don't put symlink bodies in pagecache into highmem
namei: page_getlink() and page_follow_link_light() are the same thing
ufs: get rid of ->setattr() for symlinks
udf: don't duplicate page_symlink_inode_operations
logfs: don't duplicate page_symlink_inode_operations
switch befs long symlinks to page_symlink_operations
Dmitry Vyukov provides a little program, autogenerated by syzkaller,
which races a fault on a mapping of a sparse memfd object, against
truncation of that object below the fault address: run repeatedly for a
few minutes, it reliably generates shmem_evict_inode()'s
WARN_ON(inode->i_blocks).
(But there's nothing specific to memfd here, nor to the fstat which it
happened to use to generate the fault: though that looked suspicious,
since a shmem_recalc_inode() had been added there recently. The same
problem can be reproduced with open+unlink in place of memfd_create, and
with fstatfs in place of fstat.)
v3.7 commit 0f3c42f522 ("tmpfs: change final i_blocks BUG to WARNING")
explains one cause of such a warning (a race with shmem_writepage to
swap), and possible solutions; but we never took it further, and this
syzkaller incident turns out to have a different cause.
shmem_getpage_gfp()'s error recovery, when a freshly allocated page is
then found to be beyond eof, looks plausible - decrementing the alloced
count that was just before incremented - but in fact can go wrong, if a
racing thread (the truncator, for example) gets its shmem_recalc_inode()
in just after our delete_from_page_cache(). delete_from_page_cache()
decrements nrpages, that shmem_recalc_inode() will balance the books by
decrementing alloced itself, then our decrement of alloced take it one
too low: leading to the WARNING when the object is finally evicted.
Once the new page has been exposed in the page cache,
shmem_getpage_gfp() must leave it to shmem_recalc_inode() itself to get
the accounting right in all cases (and not fall through from "trunc:" to
"decused:"). Adjust that error recovery block; and the reinitialization
of info and sbinfo can be removed too.
While we're here, fix shmem_writepage() to avoid the original issue: it
will be safe against a racing shmem_recalc_inode(), if it merely
increments swapped before the shmem_delete_from_page_cache() which
decrements nrpages (but it must then do its own shmem_recalc_inode()
before that, while still in balance, instead of after). (Aside: why do
we shmem_recalc_inode() here in the swap path? Because its raison d'etre
is to cope with clean sparse shmem pages being reclaimed behind our
back: so here when swapping is a good place to look for that case.) But
I've not now managed to reproduce this bug, even without the patch.
I don't see why I didn't do that earlier: perhaps inhibited by the
preference to eliminate shmem_recalc_inode() altogether. Driven by this
incident, I do now have a patch to do so at last; but still want to sit
on it for a bit, there's a couple of questions yet to be resolved.
Signed-off-by: Hugh Dickins <hughd@google.com>
Reported-by: Dmitry Vyukov <dvyukov@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
new method: ->get_link(); replacement of ->follow_link(). The differences
are:
* inode and dentry are passed separately
* might be called both in RCU and non-RCU mode;
the former is indicated by passing it a NULL dentry.
* when called that way it isn't allowed to block
and should return ERR_PTR(-ECHILD) if it needs to be called
in non-RCU mode.
It's a flagday change - the old method is gone, all in-tree instances
converted. Conversion isn't hard; said that, so far very few instances
do not immediately bail out when called in RCU mode. That'll change
in the next commits.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
kmap() in page_follow_link_light() needed to go - allowing to hold
an arbitrary number of kmaps for long is a great way to deadlocking
the system.
new helper (inode_nohighmem(inode)) needs to be used for pagecache
symlinks inodes; done for all in-tree cases. page_follow_link_light()
instrumented to yell about anything missed.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
When a file on tmpfs has an ACL or a Default ACL, listxattr should include the
corresponding xattr name.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Reviewed-by: James Morris <james.l.morris@oracle.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: linux-mm@kvack.org
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Use the VFS xattr handler infrastructure and get rid of similar code in
the filesystem. For implementing shmem_xattr_handler_set, we need a
version of simple_xattr_set which removes the attribute when value is
NULL. Use this to implement kernfs_iop_removexattr as well.
Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com>
Reviewed-by: James Morris <james.l.morris@oracle.com>
Cc: Hugh Dickins <hughd@google.com>
Cc: linux-mm@kvack.org
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Andrew stated the following
We have quite a history of remote parts of the kernel using
weird/wrong/inexplicable combinations of __GFP_ flags. I tend
to think that this is because we didn't adequately explain the
interface.
And I don't think that gfp.h really improved much in this area as
a result of this patchset. Could you go through it some time and
decide if we've adequately documented all this stuff?
This patches first moves some GFP flag combinations that are part of the MM
internals to mm/internal.h. The rest of the patch documents the __GFP_FOO
bits under various headings and then documents the flag combinations. It
will not help callers that are brain damaged but the clarity might motivate
some fixes and avoid future mistakes.
Signed-off-by: Mel Gorman <mgorman@techsingularity.net>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Rik van Riel <riel@redhat.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Vitaly Wool <vitalywool@gmail.com>
Cc: Rik van Riel <riel@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
LKP reports that v4.2 commit afa2db2fb6 ("tmpfs: truncate prealloc
blocks past i_size") causes a 14.5% slowdown in the AIM9 creat-clo
benchmark.
creat-clo does just what you'd expect from the name, and creat's O_TRUNC
on 0-length file does indeed get into more overhead now shmem_setattr()
tests "0 <= 0" instead of "0 < 0".
I'm not sure how much we care, but I think it would not be too VW-like to
add in a check for whether any pages (or swap) are allocated: if none are
allocated, there's none to remove from the radix_tree. At first I thought
that check would be good enough for the unmaps too, but no: we should not
skip the unlikely case of unmapping pages beyond the new EOF, which were
COWed from holes which have now been reclaimed, leaving none.
This gives me an 8.5% speedup: on Haswell instead of LKP's Westmere, and
running a debug config before and after: I hope those account for the
lesser speedup.
And probably someone has a benchmark where a thousand threads keep on
stat'ing the same file repeatedly: forestall that report by adjusting v4.3
commit 44a30220bc ("shmem: recalculate file inode when fstat") not to
take the spinlock in shmem_getattr() when there's no work to do.
Signed-off-by: Hugh Dickins <hughd@google.com>
Reported-by: Ying Huang <ying.huang@linux.intel.com>
Tested-by: Ying Huang <ying.huang@linux.intel.com>
Cc: Josef Bacik <jbacik@fb.com>
Cc: Yu Zhao <yuzhao@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
After v4.3's commit 0610c25daa ("memcg: fix dirty page migration")
mem_cgroup_migrate() doesn't have much to offer in page migration: convert
migrate_misplaced_transhuge_page() to set_page_memcg() instead.
Then rename mem_cgroup_migrate() to mem_cgroup_replace_page(), since its
remaining callers are replace_page_cache_page() and shmem_replace_page():
both of whom passed lrucare true, so just eliminate that argument.
Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: Christoph Lameter <cl@linux.com>
Cc: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Cc: Rik van Riel <riel@redhat.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Oleg Nesterov <oleg@redhat.com>
Cc: Sasha Levin <sasha.levin@oracle.com>
Cc: Dmitry Vyukov <dvyukov@google.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Shmem uses shmem_recalc_inode to update i_blocks when it allocates page,
undoes range or swaps. But mm can drop clean page without notifying
shmem. This makes fstat sometimes return out-of-date block size.
The problem can be partially solved when we add
inode_operations->getattr which calls shmem_recalc_inode to update
i_blocks for fstat.
shmem_recalc_inode also updates counter used by statfs and
vm_committed_as. For them the situation is not changed. They still
suffer from the discrepancy after dropping clean page and before the
function is called by aforementioned triggers.
Signed-off-by: Yu Zhao <yuzhao@google.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The shm implementation internally uses shmem or hugetlbfs inodes for shm
segments. As these inodes are never directly exposed to userspace and
only accessed through the shm operations which are already hooked by
security modules, mark the inodes with the S_PRIVATE flag so that inode
security initialization and permission checking is skipped.
This was motivated by the following lockdep warning:
======================================================
[ INFO: possible circular locking dependency detected ]
4.2.0-0.rc3.git0.1.fc24.x86_64+debug #1 Tainted: G W
-------------------------------------------------------
httpd/1597 is trying to acquire lock:
(&ids->rwsem){+++++.}, at: shm_close+0x34/0x130
but task is already holding lock:
(&mm->mmap_sem){++++++}, at: SyS_shmdt+0x4b/0x180
which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #3 (&mm->mmap_sem){++++++}:
lock_acquire+0xc7/0x270
__might_fault+0x7a/0xa0
filldir+0x9e/0x130
xfs_dir2_block_getdents.isra.12+0x198/0x1c0 [xfs]
xfs_readdir+0x1b4/0x330 [xfs]
xfs_file_readdir+0x2b/0x30 [xfs]
iterate_dir+0x97/0x130
SyS_getdents+0x91/0x120
entry_SYSCALL_64_fastpath+0x12/0x76
-> #2 (&xfs_dir_ilock_class){++++.+}:
lock_acquire+0xc7/0x270
down_read_nested+0x57/0xa0
xfs_ilock+0x167/0x350 [xfs]
xfs_ilock_attr_map_shared+0x38/0x50 [xfs]
xfs_attr_get+0xbd/0x190 [xfs]
xfs_xattr_get+0x3d/0x70 [xfs]
generic_getxattr+0x4f/0x70
inode_doinit_with_dentry+0x162/0x670
sb_finish_set_opts+0xd9/0x230
selinux_set_mnt_opts+0x35c/0x660
superblock_doinit+0x77/0xf0
delayed_superblock_init+0x10/0x20
iterate_supers+0xb3/0x110
selinux_complete_init+0x2f/0x40
security_load_policy+0x103/0x600
sel_write_load+0xc1/0x750
__vfs_write+0x37/0x100
vfs_write+0xa9/0x1a0
SyS_write+0x58/0xd0
entry_SYSCALL_64_fastpath+0x12/0x76
...
Signed-off-by: Stephen Smalley <sds@tycho.nsa.gov>
Reported-by: Morten Stevens <mstevens@fedoraproject.org>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Paul Moore <paul@paul-moore.com>
Cc: Manfred Spraul <manfred@colorfullife.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Cc: Prarit Bhargava <prarit@redhat.com>
Cc: Eric Paris <eparis@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
One of the rocksdb people noticed that when you do something like this
fallocate(fd, FALLOC_FL_KEEP_SIZE, 0, 10M)
pwrite(fd, buf, 5M, 0)
ftruncate(5M)
on tmpfs, the file would still take up 10M: which led to super fun
issues because we were getting ENOSPC before we thought we should be
getting ENOSPC. This patch fixes the problem, and mirrors what all the
other fs'es do (and was agreed to be the correct behaviour at LSF).
I tested it locally to make sure it worked properly with the following
xfs_io -f -c "falloc -k 0 10M" -c "pwrite 0 5M" -c "truncate 5M" file
Without the patch we have "Blocks: 20480", with the patch we have the
correct value of "Blocks: 10240".
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull vfs updates from Al Viro:
"In this pile: pathname resolution rewrite.
- recursion in link_path_walk() is gone.
- nesting limits on symlinks are gone (the only limit remaining is
that the total amount of symlinks is no more than 40, no matter how
nested).
- "fast" (inline) symlinks are handled without leaving rcuwalk mode.
- stack footprint (independent of the nesting) is below kilobyte now,
about on par with what it used to be with one level of nested
symlinks and ~2.8 times lower than it used to be in the worst case.
- struct nameidata is entirely private to fs/namei.c now (not even
opaque pointers are being passed around).
- ->follow_link() and ->put_link() calling conventions had been
changed; all in-tree filesystems converted, out-of-tree should be
able to follow reasonably easily.
For out-of-tree conversions, see Documentation/filesystems/porting
for details (and in-tree filesystems for examples of conversion).
That has sat in -next since mid-May, seems to survive all testing
without regressions and merges clean with v4.1"
* 'for-linus-1' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (131 commits)
turn user_{path_at,path,lpath,path_dir}() into static inlines
namei: move saved_nd pointer into struct nameidata
inline user_path_create()
inline user_path_parent()
namei: trim do_last() arguments
namei: stash dfd and name into nameidata
namei: fold path_cleanup() into terminate_walk()
namei: saner calling conventions for filename_parentat()
namei: saner calling conventions for filename_create()
namei: shift nameidata down into filename_parentat()
namei: make filename_lookup() reject ERR_PTR() passed as name
namei: shift nameidata inside filename_lookup()
namei: move putname() call into filename_lookup()
namei: pass the struct path to store the result down into path_lookupat()
namei: uninline set_root{,_rcu}()
namei: be careful with mountpoint crossings in follow_dotdot_rcu()
Documentation: remove outdated information from automount-support.txt
get rid of assorted nameidata-related debris
lustre: kill unused helper
lustre: kill unused macro (LOOKUP_CONTINUE)
...
It appears that, at some point last year, XFS made directory handling
changes which bring it into lockdep conflict with shmem_zero_setup():
it is surprising that mmap() can clone an inode while holding mmap_sem,
but that has been so for many years.
Since those few lockdep traces that I've seen all implicated selinux,
I'm hoping that we can use the __shmem_file_setup(,,,S_PRIVATE) which
v3.13's commit c727709092 ("security: shmem: implement kernel private
shmem inodes") introduced to avoid LSM checks on kernel-internal inodes:
the mmap("/dev/zero") cloned inode is indeed a kernel-internal detail.
This also covers the !CONFIG_SHMEM use of ramfs to support /dev/zero
(and MAP_SHARED|MAP_ANONYMOUS). I thought there were also drivers
which cloned inode in mmap(), but if so, I cannot locate them now.
Reported-and-tested-by: Prarit Bhargava <prarit@redhat.com>
Reported-and-tested-by: Daniel Wagner <wagi@monom.org>
Reported-and-tested-by: Morten Stevens <mstevens@fedoraproject.org>
Signed-off-by: Hugh Dickins <hughd@google.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
a) instead of storing the symlink body (via nd_set_link()) and returning
an opaque pointer later passed to ->put_link(), ->follow_link() _stores_
that opaque pointer (into void * passed by address by caller) and returns
the symlink body. Returning ERR_PTR() on error, NULL on jump (procfs magic
symlinks) and pointer to symlink body for normal symlinks. Stored pointer
is ignored in all cases except the last one.
Storing NULL for opaque pointer (or not storing it at all) means no call
of ->put_link().
b) the body used to be passed to ->put_link() implicitly (via nameidata).
Now only the opaque pointer is. In the cases when we used the symlink body
to free stuff, ->follow_link() now should store it as opaque pointer in addition
to returning it.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
All places outside of core VFS that checked ->read and ->write for being NULL or
called the methods directly are gone now, so NULL {read,write} with non-NULL
{read,write}_iter will do the right thing in all cases.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
struct kiocb now is a generic I/O container, so move it to fs.h.
Also do a #include diet for aio.h while we're at it.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Convert the following where appropriate:
(1) S_ISLNK(dentry->d_inode) to d_is_symlink(dentry).
(2) S_ISREG(dentry->d_inode) to d_is_reg(dentry).
(3) S_ISDIR(dentry->d_inode) to d_is_dir(dentry). This is actually more
complicated than it appears as some calls should be converted to
d_can_lookup() instead. The difference is whether the directory in
question is a real dir with a ->lookup op or whether it's a fake dir with
a ->d_automount op.
In some circumstances, we can subsume checks for dentry->d_inode not being
NULL into this, provided we the code isn't in a filesystem that expects
d_inode to be NULL if the dirent really *is* negative (ie. if we're going to
use d_inode() rather than d_backing_inode() to get the inode pointer).
Note that the dentry type field may be set to something other than
DCACHE_MISS_TYPE when d_inode is NULL in the case of unionmount, where the VFS
manages the fall-through from a negative dentry to a lower layer. In such a
case, the dentry type of the negative union dentry is set to the same as the
type of the lower dentry.
However, if you know d_inode is not NULL at the call site, then you can use
the d_is_xxx() functions even in a filesystem.
There is one further complication: a 0,0 chardev dentry may be labelled
DCACHE_WHITEOUT_TYPE rather than DCACHE_SPECIAL_TYPE. Strictly, this was
intended for special directory entry types that don't have attached inodes.
The following perl+coccinelle script was used:
use strict;
my @callers;
open($fd, 'git grep -l \'S_IS[A-Z].*->d_inode\' |') ||
die "Can't grep for S_ISDIR and co. callers";
@callers = <$fd>;
close($fd);
unless (@callers) {
print "No matches\n";
exit(0);
}
my @cocci = (
'@@',
'expression E;',
'@@',
'',
'- S_ISLNK(E->d_inode->i_mode)',
'+ d_is_symlink(E)',
'',
'@@',
'expression E;',
'@@',
'',
'- S_ISDIR(E->d_inode->i_mode)',
'+ d_is_dir(E)',
'',
'@@',
'expression E;',
'@@',
'',
'- S_ISREG(E->d_inode->i_mode)',
'+ d_is_reg(E)' );
my $coccifile = "tmp.sp.cocci";
open($fd, ">$coccifile") || die $coccifile;
print($fd "$_\n") || die $coccifile foreach (@cocci);
close($fd);
foreach my $file (@callers) {
chomp $file;
print "Processing ", $file, "\n";
system("spatch", "--sp-file", $coccifile, $file, "--in-place", "--no-show-diff") == 0 ||
die "spatch failed";
}
[AV: overlayfs parts skipped]
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull backing device changes from Jens Axboe:
"This contains a cleanup of how the backing device is handled, in
preparation for a rework of the life time rules. In this part, the
most important change is to split the unrelated nommu mmap flags from
it, but also removing a backing_dev_info pointer from the
address_space (and inode), and a cleanup of other various minor bits.
Christoph did all the work here, I just fixed an oops with pages that
have a swap backing. Arnd fixed a missing export, and Oleg killed the
lustre backing_dev_info from staging. Last patch was from Al,
unexporting parts that are now no longer needed outside"
* 'for-3.20/bdi' of git://git.kernel.dk/linux-block:
Make super_blocks and sb_lock static
mtd: export new mtd_mmap_capabilities
fs: make inode_to_bdi() handle NULL inode
staging/lustre/llite: get rid of backing_dev_info
fs: remove default_backing_dev_info
fs: don't reassign dirty inodes to default_backing_dev_info
nfs: don't call bdi_unregister
ceph: remove call to bdi_unregister
fs: remove mapping->backing_dev_info
fs: export inode_to_bdi and use it in favor of mapping->backing_dev_info
nilfs2: set up s_bdi like the generic mount_bdev code
block_dev: get bdev inode bdi directly from the block device
block_dev: only write bdev inode on close
fs: introduce f_op->mmap_capabilities for nommu mmap support
fs: kill BDI_CAP_SWAP_BACKED
fs: deduplicate noop_backing_dev_info
The body of this function was removed by commit 0a31bc97c8 ("mm:
memcontrol: rewrite uncharge API").
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It has been reported that 965GM might trigger
VM_BUG_ON_PAGE(!lrucare && PageLRU(oldpage), oldpage)
in mem_cgroup_migrate when shmem wants to replace a swap cache page
because of shmem_should_replace_page (the page is allocated from an
inappropriate zone). shmem_replace_page expects that the oldpage is not
on LRU list and calls mem_cgroup_migrate without lrucare. This is
obviously incorrect because swapcache pages might be on the LRU list
(e.g. swapin readahead page).
Fix this by enabling lrucare for the migration in shmem_replace_page.
Also clarify that lrucare should be used even if one of the pages might
be on LRU list.
The BUG_ON will trigger only when CONFIG_DEBUG_VM is enabled but even
without that the migration code might leave the old page on an
inappropriate memcg' LRU which is not that critical because the page
would get removed with its last reference but it is still confusing.
Fixes: 0a31bc97c8 ("mm: memcontrol: rewrite uncharge API")
Signed-off-by: Michal Hocko <mhocko@suse.cz>
Reported-by: Chris Wilson <chris@chris-wilson.co.uk>
Reported-by: Dave Airlie <airlied@gmail.com>
Acked-by: Hugh Dickins <hughd@google.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: <stable@vger.kernel.org> [3.17+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Now that we never use the backing_dev_info pointer in struct address_space
we can simply remove it and save 4 to 8 bytes in every inode.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Acked-by: Ryusuke Konishi <konishi.ryusuke@lab.ntt.co.jp>
Reviewed-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jens Axboe <axboe@fb.com>
This bdi flag isn't too useful - we can determine that a vma is backed by
either swap or shmem trivially in the caller.
This also allows removing the backing_dev_info instaces for swap and shmem
in favor of noop_backing_dev_info.
Signed-off-by: Christoph Hellwig <hch@lst.de>
Reviewed-by: Tejun Heo <tj@kernel.org>
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Jens Axboe <axboe@fb.com>
Allocate a dentry, initialize it with a whiteout and hash it in the place
of the old dentry. Later the old dentry will be moved away and the
whiteout will remain.
i_mutex protects agains concurrent readdir.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: Hugh Dickins <hughd@google.com>
Pull percpu updates from Tejun Heo:
"A lot of activities on percpu front. Notable changes are...
- percpu allocator now can take @gfp. If @gfp doesn't contain
GFP_KERNEL, it tries to allocate from what's already available to
the allocator and a work item tries to keep the reserve around
certain level so that these atomic allocations usually succeed.
This will replace the ad-hoc percpu memory pool used by
blk-throttle and also be used by the planned blkcg support for
writeback IOs.
Please note that I noticed a bug in how @gfp is interpreted while
preparing this pull request and applied the fix 6ae833c7fe
("percpu: fix how @gfp is interpreted by the percpu allocator")
just now.
- percpu_ref now uses longs for percpu and global counters instead of
ints. It leads to more sparse packing of the percpu counters on
64bit machines but the overhead should be negligible and this
allows using percpu_ref for refcnting pages and in-memory objects
directly.
- The switching between percpu and single counter modes of a
percpu_ref is made independent of putting the base ref and a
percpu_ref can now optionally be initialized in single or killed
mode. This allows avoiding percpu shutdown latency for cases where
the refcounted objects may be synchronously created and destroyed
in rapid succession with only a fraction of them reaching fully
operational status (SCSI probing does this when combined with
blk-mq support). It's also planned to be used to implement forced
single mode to detect underflow more timely for debugging.
There's a separate branch percpu/for-3.18-consistent-ops which cleans
up the duplicate percpu accessors. That branch causes a number of
conflicts with s390 and other trees. I'll send a separate pull
request w/ resolutions once other branches are merged"
* 'for-3.18' of git://git.kernel.org/pub/scm/linux/kernel/git/tj/percpu: (33 commits)
percpu: fix how @gfp is interpreted by the percpu allocator
blk-mq, percpu_ref: start q->mq_usage_counter in atomic mode
percpu_ref: make INIT_ATOMIC and switch_to_atomic() sticky
percpu_ref: add PERCPU_REF_INIT_* flags
percpu_ref: decouple switching to percpu mode and reinit
percpu_ref: decouple switching to atomic mode and killing
percpu_ref: add PCPU_REF_DEAD
percpu_ref: rename things to prepare for decoupling percpu/atomic mode switch
percpu_ref: replace pcpu_ prefix with percpu_
percpu_ref: minor code and comment updates
percpu_ref: relocate percpu_ref_reinit()
Revert "blk-mq, percpu_ref: implement a kludge for SCSI blk-mq stall during probe"
Revert "percpu: free percpu allocation info for uniprocessor system"
percpu-refcount: make percpu_ref based on longs instead of ints
percpu-refcount: improve WARN messages
percpu: fix locking regression in the failure path of pcpu_alloc()
percpu-refcount: add @gfp to percpu_ref_init()
proportions: add @gfp to init functions
percpu_counter: add @gfp to percpu_counter_init()
percpu_counter: make percpu_counters_lock irq-safe
...
This is designed to avoid a few ifdefs in .c files but it's obnoxious
because it can cause unsuspecting "migrate_page" symbols to get turned into
"NULL".
Just nuke it and use the ifdefs.
Cc: Konstantin Khlebnikov <k.khlebnikov@samsung.com>
Cc: Rafael Aquini <aquini@redhat.com>
Cc: Andrey Ryabinin <ryabinin.a.a@gmail.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If overwriting an empty directory with rename, then need to drop the extra
nlink.
Test prog:
#include <stdio.h>
#include <fcntl.h>
#include <err.h>
#include <sys/stat.h>
int main(void)
{
const char *test_dir1 = "test-dir1";
const char *test_dir2 = "test-dir2";
int res;
int fd;
struct stat statbuf;
res = mkdir(test_dir1, 0777);
if (res == -1)
err(1, "mkdir(\"%s\")", test_dir1);
res = mkdir(test_dir2, 0777);
if (res == -1)
err(1, "mkdir(\"%s\")", test_dir2);
fd = open(test_dir2, O_RDONLY);
if (fd == -1)
err(1, "open(\"%s\")", test_dir2);
res = rename(test_dir1, test_dir2);
if (res == -1)
err(1, "rename(\"%s\", \"%s\")", test_dir1, test_dir2);
res = fstat(fd, &statbuf);
if (res == -1)
err(1, "fstat(%i)", fd);
if (statbuf.st_nlink != 0) {
fprintf(stderr, "nlink is %lu, should be 0\n", statbuf.st_nlink);
return 1;
}
return 0;
}
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Cc: stable@vger.kernel.org
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Percpu allocator now supports allocation mask. Add @gfp to
percpu_counter_init() so that !GFP_KERNEL allocation masks can be used
with percpu_counters too.
We could have left percpu_counter_init() alone and added
percpu_counter_init_gfp(); however, the number of users isn't that
high and introducing _gfp variants to all percpu data structures would
be quite ugly, so let's just do the conversion. This is the one with
the most users. Other percpu data structures are a lot easier to
convert.
This patch doesn't make any functional difference.
Signed-off-by: Tejun Heo <tj@kernel.org>
Acked-by: Jan Kara <jack@suse.cz>
Acked-by: "David S. Miller" <davem@davemloft.net>
Cc: x86@kernel.org
Cc: Jens Axboe <axboe@kernel.dk>
Cc: "Theodore Ts'o" <tytso@mit.edu>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>
Pull vfs updates from Al Viro:
"Stuff in here:
- acct.c fixes and general rework of mnt_pin mechanism. That allows
to go for delayed-mntput stuff, which will permit mntput() on deep
stack without worrying about stack overflows - fs shutdown will
happen on shallow stack. IOW, we can do Eric's umount-on-rmdir
series without introducing tons of stack overflows on new mntput()
call chains it introduces.
- Bruce's d_splice_alias() patches
- more Miklos' rename() stuff.
- a couple of regression fixes (stable fodder, in the end of branch)
and a fix for API idiocy in iov_iter.c.
There definitely will be another pile, maybe even two. I'd like to
get Eric's series in this time, but even if we miss it, it'll go right
in the beginning of for-next in the next cycle - the tricky part of
prereqs is in this pile"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (40 commits)
fix copy_tree() regression
__generic_file_write_iter(): fix handling of sync error after DIO
switch iov_iter_get_pages() to passing maximal number of pages
fs: mark __d_obtain_alias static
dcache: d_splice_alias should detect loops
exportfs: update Exporting documentation
dcache: d_find_alias needn't recheck IS_ROOT && DCACHE_DISCONNECTED
dcache: remove unused d_find_alias parameter
dcache: d_obtain_alias callers don't all want DISCONNECTED
dcache: d_splice_alias should ignore DCACHE_DISCONNECTED
dcache: d_splice_alias mustn't create directory aliases
dcache: close d_move race in d_splice_alias
dcache: move d_splice_alias
namei: trivial fix to vfs_rename_dir comment
VFS: allow ->d_manage() to declare -EISDIR in rcu_walk mode.
cifs: support RENAME_NOREPLACE
hostfs: support rename flags
shmem: support RENAME_EXCHANGE
shmem: support RENAME_NOREPLACE
btrfs: add RENAME_NOREPLACE
...
If we set SEAL_WRITE on a file, we must make sure there cannot be any
ongoing write-operations on the file. For write() calls, we simply lock
the inode mutex, for mmap() we simply verify there're no writable
mappings. However, there might be pages pinned by AIO, Direct-IO and
similar operations via GUP. We must make sure those do not write to the
memfd file after we set SEAL_WRITE.
As there is no way to notify GUP users to drop pages or to wait for them
to be done, we implement the wait ourself: When setting SEAL_WRITE, we
check all pages for their ref-count. If it's bigger than 1, we know
there's some user of the page. We then mark the page and wait for up to
150ms for those ref-counts to be dropped. If the ref-counts are not
dropped in time, we refuse the seal operation.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Ryan Lortie <desrt@desrt.ca>
Cc: Lennart Poettering <lennart@poettering.net>
Cc: Daniel Mack <zonque@gmail.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
memfd_create() is similar to mmap(MAP_ANON), but returns a file-descriptor
that you can pass to mmap(). It can support sealing and avoids any
connection to user-visible mount-points. Thus, it's not subject to quotas
on mounted file-systems, but can be used like malloc()'ed memory, but with
a file-descriptor to it.
memfd_create() returns the raw shmem file, so calls like ftruncate() can
be used to modify the underlying inode. Also calls like fstat() will
return proper information and mark the file as regular file. If you want
sealing, you can specify MFD_ALLOW_SEALING. Otherwise, sealing is not
supported (like on all other regular files).
Compared to O_TMPFILE, it does not require a tmpfs mount-point and is not
subject to a filesystem size limit. It is still properly accounted to
memcg limits, though, and to the same overcommit or no-overcommit
accounting as all user memory.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Ryan Lortie <desrt@desrt.ca>
Cc: Lennart Poettering <lennart@poettering.net>
Cc: Daniel Mack <zonque@gmail.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
If two processes share a common memory region, they usually want some
guarantees to allow safe access. This often includes:
- one side cannot overwrite data while the other reads it
- one side cannot shrink the buffer while the other accesses it
- one side cannot grow the buffer beyond previously set boundaries
If there is a trust-relationship between both parties, there is no need
for policy enforcement. However, if there's no trust relationship (eg.,
for general-purpose IPC) sharing memory-regions is highly fragile and
often not possible without local copies. Look at the following two
use-cases:
1) A graphics client wants to share its rendering-buffer with a
graphics-server. The memory-region is allocated by the client for
read/write access and a second FD is passed to the server. While
scanning out from the memory region, the server has no guarantee that
the client doesn't shrink the buffer at any time, requiring rather
cumbersome SIGBUS handling.
2) A process wants to perform an RPC on another process. To avoid huge
bandwidth consumption, zero-copy is preferred. After a message is
assembled in-memory and a FD is passed to the remote side, both sides
want to be sure that neither modifies this shared copy, anymore. The
source may have put sensible data into the message without a separate
copy and the target may want to parse the message inline, to avoid a
local copy.
While SIGBUS handling, POSIX mandatory locking and MAP_DENYWRITE provide
ways to achieve most of this, the first one is unproportionally ugly to
use in libraries and the latter two are broken/racy or even disabled due
to denial of service attacks.
This patch introduces the concept of SEALING. If you seal a file, a
specific set of operations is blocked on that file forever. Unlike locks,
seals can only be set, never removed. Hence, once you verified a specific
set of seals is set, you're guaranteed that no-one can perform the blocked
operations on this file, anymore.
An initial set of SEALS is introduced by this patch:
- SHRINK: If SEAL_SHRINK is set, the file in question cannot be reduced
in size. This affects ftruncate() and open(O_TRUNC).
- GROW: If SEAL_GROW is set, the file in question cannot be increased
in size. This affects ftruncate(), fallocate() and write().
- WRITE: If SEAL_WRITE is set, no write operations (besides resizing)
are possible. This affects fallocate(PUNCH_HOLE), mmap() and
write().
- SEAL: If SEAL_SEAL is set, no further seals can be added to a file.
This basically prevents the F_ADD_SEAL operation on a file and
can be set to prevent others from adding further seals that you
don't want.
The described use-cases can easily use these seals to provide safe use
without any trust-relationship:
1) The graphics server can verify that a passed file-descriptor has
SEAL_SHRINK set. This allows safe scanout, while the client is
allowed to increase buffer size for window-resizing on-the-fly.
Concurrent writes are explicitly allowed.
2) For general-purpose IPC, both processes can verify that SEAL_SHRINK,
SEAL_GROW and SEAL_WRITE are set. This guarantees that neither
process can modify the data while the other side parses it.
Furthermore, it guarantees that even with writable FDs passed to the
peer, it cannot increase the size to hit memory-limits of the source
process (in case the file-storage is accounted to the source).
The new API is an extension to fcntl(), adding two new commands:
F_GET_SEALS: Return a bitset describing the seals on the file. This
can be called on any FD if the underlying file supports
sealing.
F_ADD_SEALS: Change the seals of a given file. This requires WRITE
access to the file and F_SEAL_SEAL may not already be set.
Furthermore, the underlying file must support sealing and
there may not be any existing shared mapping of that file.
Otherwise, EBADF/EPERM is returned.
The given seals are _added_ to the existing set of seals
on the file. You cannot remove seals again.
The fcntl() handler is currently specific to shmem and disabled on all
files. A file needs to explicitly support sealing for this interface to
work. A separate syscall is added in a follow-up, which creates files that
support sealing. There is no intention to support this on other
file-systems. Semantics are unclear for non-volatile files and we lack any
use-case right now. Therefore, the implementation is specific to shmem.
Signed-off-by: David Herrmann <dh.herrmann@gmail.com>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Cc: Ryan Lortie <desrt@desrt.ca>
Cc: Lennart Poettering <lennart@poettering.net>
Cc: Daniel Mack <zonque@gmail.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The memcg uncharging code that is involved towards the end of a page's
lifetime - truncation, reclaim, swapout, migration - is impressively
complicated and fragile.
Because anonymous and file pages were always charged before they had their
page->mapping established, uncharges had to happen when the page type
could still be known from the context; as in unmap for anonymous, page
cache removal for file and shmem pages, and swap cache truncation for swap
pages. However, these operations happen well before the page is actually
freed, and so a lot of synchronization is necessary:
- Charging, uncharging, page migration, and charge migration all need
to take a per-page bit spinlock as they could race with uncharging.
- Swap cache truncation happens during both swap-in and swap-out, and
possibly repeatedly before the page is actually freed. This means
that the memcg swapout code is called from many contexts that make
no sense and it has to figure out the direction from page state to
make sure memory and memory+swap are always correctly charged.
- On page migration, the old page might be unmapped but then reused,
so memcg code has to prevent untimely uncharging in that case.
Because this code - which should be a simple charge transfer - is so
special-cased, it is not reusable for replace_page_cache().
But now that charged pages always have a page->mapping, introduce
mem_cgroup_uncharge(), which is called after the final put_page(), when we
know for sure that nobody is looking at the page anymore.
For page migration, introduce mem_cgroup_migrate(), which is called after
the migration is successful and the new page is fully rmapped. Because
the old page is no longer uncharged after migration, prevent double
charges by decoupling the page's memcg association (PCG_USED and
pc->mem_cgroup) from the page holding an actual charge. The new bits
PCG_MEM and PCG_MEMSW represent the respective charges and are transferred
to the new page during migration.
mem_cgroup_migrate() is suitable for replace_page_cache() as well,
which gets rid of mem_cgroup_replace_page_cache(). However, care
needs to be taken because both the source and the target page can
already be charged and on the LRU when fuse is splicing: grab the page
lock on the charge moving side to prevent changing pc->mem_cgroup of a
page under migration. Also, the lruvecs of both pages change as we
uncharge the old and charge the new during migration, and putback may
race with us, so grab the lru lock and isolate the pages iff on LRU to
prevent races and ensure the pages are on the right lruvec afterward.
Swap accounting is massively simplified: because the page is no longer
uncharged as early as swap cache deletion, a new mem_cgroup_swapout() can
transfer the page's memory+swap charge (PCG_MEMSW) to the swap entry
before the final put_page() in page reclaim.
Finally, page_cgroup changes are now protected by whatever protection the
page itself offers: anonymous pages are charged under the page table lock,
whereas page cache insertions, swapin, and migration hold the page lock.
Uncharging happens under full exclusion with no outstanding references.
Charging and uncharging also ensure that the page is off-LRU, which
serializes against charge migration. Remove the very costly page_cgroup
lock and set pc->flags non-atomically.
[mhocko@suse.cz: mem_cgroup_charge_statistics needs preempt_disable]
[vdavydov@parallels.com: fix flags definition]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Hugh Dickins <hughd@google.com>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Tested-by: Jet Chen <jet.chen@intel.com>
Acked-by: Michal Hocko <mhocko@suse.cz>
Tested-by: Felipe Balbi <balbi@ti.com>
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
These patches rework memcg charge lifetime to integrate more naturally
with the lifetime of user pages. This drastically simplifies the code and
reduces charging and uncharging overhead. The most expensive part of
charging and uncharging is the page_cgroup bit spinlock, which is removed
entirely after this series.
Here are the top-10 profile entries of a stress test that reads a 128G
sparse file on a freshly booted box, without even a dedicated cgroup (i.e.
executing in the root memcg). Before:
15.36% cat [kernel.kallsyms] [k] copy_user_generic_string
13.31% cat [kernel.kallsyms] [k] memset
11.48% cat [kernel.kallsyms] [k] do_mpage_readpage
4.23% cat [kernel.kallsyms] [k] get_page_from_freelist
2.38% cat [kernel.kallsyms] [k] put_page
2.32% cat [kernel.kallsyms] [k] __mem_cgroup_commit_charge
2.18% kswapd0 [kernel.kallsyms] [k] __mem_cgroup_uncharge_common
1.92% kswapd0 [kernel.kallsyms] [k] shrink_page_list
1.86% cat [kernel.kallsyms] [k] __radix_tree_lookup
1.62% cat [kernel.kallsyms] [k] __pagevec_lru_add_fn
After:
15.67% cat [kernel.kallsyms] [k] copy_user_generic_string
13.48% cat [kernel.kallsyms] [k] memset
11.42% cat [kernel.kallsyms] [k] do_mpage_readpage
3.98% cat [kernel.kallsyms] [k] get_page_from_freelist
2.46% cat [kernel.kallsyms] [k] put_page
2.13% kswapd0 [kernel.kallsyms] [k] shrink_page_list
1.88% cat [kernel.kallsyms] [k] __radix_tree_lookup
1.67% cat [kernel.kallsyms] [k] __pagevec_lru_add_fn
1.39% kswapd0 [kernel.kallsyms] [k] free_pcppages_bulk
1.30% cat [kernel.kallsyms] [k] kfree
As you can see, the memcg footprint has shrunk quite a bit.
text data bss dec hex filename
37970 9892 400 48262 bc86 mm/memcontrol.o.old
35239 9892 400 45531 b1db mm/memcontrol.o
This patch (of 4):
The memcg charge API charges pages before they are rmapped - i.e. have an
actual "type" - and so every callsite needs its own set of charge and
uncharge functions to know what type is being operated on. Worse,
uncharge has to happen from a context that is still type-specific, rather
than at the end of the page's lifetime with exclusive access, and so
requires a lot of synchronization.
Rewrite the charge API to provide a generic set of try_charge(),
commit_charge() and cancel_charge() transaction operations, much like
what's currently done for swap-in:
mem_cgroup_try_charge() attempts to reserve a charge, reclaiming
pages from the memcg if necessary.
mem_cgroup_commit_charge() commits the page to the charge once it
has a valid page->mapping and PageAnon() reliably tells the type.
mem_cgroup_cancel_charge() aborts the transaction.
This reduces the charge API and enables subsequent patches to
drastically simplify uncharging.
As pages need to be committed after rmap is established but before they
are added to the LRU, page_add_new_anon_rmap() must stop doing LRU
additions again. Revive lru_cache_add_active_or_unevictable().
[hughd@google.com: fix shmem_unuse]
[hughd@google.com: Add comments on the private use of -EAGAIN]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@suse.cz>
Cc: Tejun Heo <tj@kernel.org>
Cc: Vladimir Davydov <vdavydov@parallels.com>
Signed-off-by: Hugh Dickins <hughd@google.com>
Cc: Naoya Horiguchi <n-horiguchi@ah.jp.nec.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Linus Torvalds