We are going to make FS shrinkers memcg-aware. To achieve that, we will
have to pass the memcg to scan to the nr_cached_objects and
free_cached_objects VFS methods, which currently take only the NUMA node
to scan. Since the shrink_control structure already holds the node, and
the memcg to scan will be added to it when we introduce memcg-aware
vmscan, let us consolidate the methods' arguments in this structure to
keep things clean.
Signed-off-by: Vladimir Davydov <vdavydov@parallels.com>
Suggested-by: Dave Chinner <david@fromorbit.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@suse.cz>
Cc: Greg Thelen <gthelen@google.com>
Cc: Glauber Costa <glommer@gmail.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
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>
Cc: Tejun Heo <tj@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull nfsd updates from Bruce Fields:
"The main change is the pNFS block server support from Christoph, which
allows an NFS client connected to shared disk to do block IO to the
shared disk in place of NFS reads and writes. This also requires xfs
patches, which should arrive soon through the xfs tree, barring
unexpected problems. Support for other filesystems is also possible
if there's interest.
Thanks also to Chuck Lever for continuing work to get NFS/RDMA into
shape"
* 'for-3.20' of git://linux-nfs.org/~bfields/linux: (32 commits)
nfsd: default NFSv4.2 to on
nfsd: pNFS block layout driver
exportfs: add methods for block layout exports
nfsd: add trace events
nfsd: update documentation for pNFS support
nfsd: implement pNFS layout recalls
nfsd: implement pNFS operations
nfsd: make find_any_file available outside nfs4state.c
nfsd: make find/get/put file available outside nfs4state.c
nfsd: make lookup/alloc/unhash_stid available outside nfs4state.c
nfsd: add fh_fsid_match helper
nfsd: move nfsd_fh_match to nfsfh.h
fs: add FL_LAYOUT lease type
fs: track fl_owner for leases
nfs: add LAYOUT_TYPE_MAX enum value
nfsd: factor out a helper to decode nfstime4 values
sunrpc/lockd: fix references to the BKL
nfsd: fix year-2038 nfs4 state problem
svcrdma: Handle additional inline content
svcrdma: Move read list XDR round-up logic
...
Merge misc updates from Andrew Morton:
"Bite-sized chunks this time, to avoid the MTA ratelimiting woes.
- fs/notify updates
- ocfs2
- some of MM"
That laconic "some MM" is mainly the removal of remap_file_pages(),
which is a big simplification of the VM, and which gets rid of a *lot*
of random cruft and special cases because we no longer support the
non-linear mappings that it used.
From a user interface perspective, nothing has changed, because the
remap_file_pages() syscall still exists, it's just done by emulating the
old behavior by creating a lot of individual small mappings instead of
one non-linear one.
The emulation is slower than the old "native" non-linear mappings, but
nobody really uses or cares about remap_file_pages(), and simplifying
the VM is a big advantage.
* emailed patches from Andrew Morton <akpm@linux-foundation.org>: (78 commits)
memcg: zap memcg_slab_caches and memcg_slab_mutex
memcg: zap memcg_name argument of memcg_create_kmem_cache
memcg: zap __memcg_{charge,uncharge}_slab
mm/page_alloc.c: place zone_id check before VM_BUG_ON_PAGE check
mm: hugetlb: fix type of hugetlb_treat_as_movable variable
mm, hugetlb: remove unnecessary lower bound on sysctl handlers"?
mm: memory: merge shared-writable dirtying branches in do_wp_page()
mm: memory: remove ->vm_file check on shared writable vmas
xtensa: drop _PAGE_FILE and pte_file()-related helpers
x86: drop _PAGE_FILE and pte_file()-related helpers
unicore32: drop pte_file()-related helpers
um: drop _PAGE_FILE and pte_file()-related helpers
tile: drop pte_file()-related helpers
sparc: drop pte_file()-related helpers
sh: drop _PAGE_FILE and pte_file()-related helpers
score: drop _PAGE_FILE and pte_file()-related helpers
s390: drop pte_file()-related helpers
parisc: drop _PAGE_FILE and pte_file()-related helpers
openrisc: drop _PAGE_FILE and pte_file()-related helpers
nios2: drop _PAGE_FILE and pte_file()-related helpers
...
We don't create non-linear mappings anymore. Let's drop code which
handles them in rmap.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
remap_file_pages(2) was invented to be able efficiently map parts of
huge file into limited 32-bit virtual address space such as in database
workloads.
Nonlinear mappings are pain to support and it seems there's no
legitimate use-cases nowadays since 64-bit systems are widely available.
Let's drop it and get rid of all these special-cased code.
The patch replaces the syscall with emulation which creates new VMA on
each remap_file_pages(), unless they it can be merged with an adjacent
one.
I didn't find *any* real code that uses remap_file_pages(2) to test
emulation impact on. I've checked Debian code search and source of all
packages in ALT Linux. No real users: libc wrappers, mentions in
strace, gdb, valgrind and this kind of stuff.
There are few basic tests in LTP for the syscall. They work just fine
with emulation.
To test performance impact, I've written small test case which
demonstrate pretty much worst case scenario: map 4G shmfs file, write to
begin of every page pgoff of the page, remap pages in reverse order,
read every page.
The test creates 1 million of VMAs if emulation is in use, so I had to
set vm.max_map_count to 1100000 to avoid -ENOMEM.
Before: 23.3 ( +- 4.31% ) seconds
After: 43.9 ( +- 0.85% ) seconds
Slowdown: 1.88x
I believe we can live with that.
Test case:
#define _GNU_SOURCE
#include <assert.h>
#include <stdlib.h>
#include <stdio.h>
#include <sys/mman.h>
#define MB (1024UL * 1024)
#define SIZE (4096 * MB)
int main(int argc, char **argv)
{
unsigned long *p;
long i, pass;
for (pass = 0; pass < 10; pass++) {
p = mmap(NULL, SIZE, PROT_READ|PROT_WRITE,
MAP_SHARED | MAP_ANONYMOUS, -1, 0);
if (p == MAP_FAILED) {
perror("mmap");
return -1;
}
for (i = 0; i < SIZE / 4096; i++)
p[i * 4096 / sizeof(*p)] = i;
for (i = 0; i < SIZE / 4096; i++) {
if (remap_file_pages(p + i * 4096 / sizeof(*p), 4096,
0, (SIZE - 4096 * (i + 1)) >> 12, 0)) {
perror("remap_file_pages");
return -1;
}
}
for (i = SIZE / 4096 - 1; i >= 0; i--)
assert(p[i * 4096 / sizeof(*p)] == SIZE / 4096 - i - 1);
munmap(p, SIZE);
}
return 0;
}
[akpm@linux-foundation.org: fix spello]
[sasha.levin@oracle.com: initialize populate before usage]
[sasha.levin@oracle.com: grab file ref to prevent race while mmaping]
Signed-off-by: "Kirill A. Shutemov" <kirill@shutemov.name>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Ingo Molnar <mingo@kernel.org>
Cc: Dave Jones <davej@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Armin Rigo <arigo@tunes.org>
Signed-off-by: Sasha Levin <sasha.levin@oracle.com>
Cc: Hugh Dickins <hughd@google.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This (ab-)uses the file locking code to allow filesystems to recall
outstanding pNFS layouts on a file. This new lease type is similar but
not quite the same as FL_DELEG. A FL_LAYOUT lease can always be granted,
an a per-filesystem lock (XFS iolock for the initial implementation)
ensures not FL_LAYOUT leases granted when we would need to recall them.
Also included are changes that allow multiple outstanding read
leases of different types on the same file as long as they have a
differnt owner. This wasn't a problem until now as nfsd never set
FL_LEASE leases, and no one else used FL_DELEG leases, but given that
nfsd will also issues FL_LAYOUT leases we will have to handle it now.
Signed-off-by: Christoph Hellwig <hch@lst.de>
This makes things a bit more efficient in the cifs and ceph lock
pushing code.
Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Now that we use standard list_heads for tracking leases, we can have
lm_change take a pointer to the lease to be modified instead of a
double pointer.
Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Acked-by: Christoph Hellwig <hch@lst.de>
We can now add a dedicated spinlock without expanding struct inode.
Change to using that to protect the various i_flctx lists.
Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Nothing uses it anymore. Also add a forward declaration for struct
file_lock to silence some compiler warnings that the removal triggers.
Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Acked-by: Christoph Hellwig <hch@lst.de>
The current scheme of using the i_flock list is really difficult to
manage. There is also a legitimate desire for a per-inode spinlock to
manage these lists that isn't the i_lock.
Start conversion to a new scheme to eventually replace the old i_flock
list with a new "file_lock_context" object.
We start by adding a new i_flctx to struct inode. For now, it lives in
parallel with i_flock list, but will eventually replace it. The idea is
to allocate a structure to sit in that pointer and act as a locus for
all things file locking.
We allocate a file_lock_context for an inode when the first lock is
added to it, and it's only freed when the inode is freed. We use the
i_lock to protect the assignment, but afterward it should mostly be
accessed locklessly.
Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Acked-by: Christoph Hellwig <hch@lst.de>
...that we can use to queue file_locks to per-ctx list_heads. Go ahead
and convert locks_delete_lock and locks_dispose_list to use it instead
of the fl_block list.
Signed-off-by: Jeff Layton <jlayton@primarydata.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Fix clashing values for O_PATH and FMODE_NONOTIFY on sparc. The
clashing O_PATH value was added in commit 5229645bdc ("vfs: add
nonconflicting values for O_PATH") but this can't be changed as it is
user-visible.
FMODE_NONOTIFY is only used internally in the kernel, but it is in the
same numbering space as the other O_* flags, as indicated by the comment
at the top of include/uapi/asm-generic/fcntl.h (and its use in
fs/notify/fanotify/fanotify_user.c). So renumber it to avoid the clash.
All of this has happened before (commit 12ed2e36c98a: "fanotify:
FMODE_NONOTIFY and __O_SYNC in sparc conflict"), and all of this will
happen again -- so update the uniqueness check in fcntl_init() to
include __FMODE_NONOTIFY.
Signed-off-by: David Drysdale <drysdale@google.com>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: Jan Kara <jack@suse.cz>
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Cc: Eric Paris <eparis@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull vfs pile #2 from Al Viro:
"Next pile (and there'll be one or two more).
The large piece in this one is getting rid of /proc/*/ns/* weirdness;
among other things, it allows to (finally) make nameidata completely
opaque outside of fs/namei.c, making for easier further cleanups in
there"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
coda_venus_readdir(): use file_inode()
fs/namei.c: fold link_path_walk() call into path_init()
path_init(): don't bother with LOOKUP_PARENT in argument
fs/namei.c: new helper (path_cleanup())
path_init(): store the "base" pointer to file in nameidata itself
make default ->i_fop have ->open() fail with ENXIO
make nameidata completely opaque outside of fs/namei.c
kill proc_ns completely
take the targets of /proc/*/ns/* symlinks to separate fs
bury struct proc_ns in fs/proc
copy address of proc_ns_ops into ns_common
new helpers: ns_alloc_inum/ns_free_inum
make proc_ns_operations work with struct ns_common * instead of void *
switch the rest of proc_ns_operations to working with &...->ns
netns: switch ->get()/->put()/->install()/->inum() to working with &net->ns
make mntns ->get()/->put()/->install()/->inum() work with &mnt_ns->ns
common object embedded into various struct ....ns
Pull nfsd updates from Bruce Fields:
"A comparatively quieter cycle for nfsd this time, but still with two
larger changes:
- RPC server scalability improvements from Jeff Layton (using RCU
instead of a spinlock to find idle threads).
- server-side NFSv4.2 ALLOCATE/DEALLOCATE support from Anna
Schumaker, enabling fallocate on new clients"
* 'for-3.19' of git://linux-nfs.org/~bfields/linux: (32 commits)
nfsd4: fix xdr4 count of server in fs_location4
nfsd4: fix xdr4 inclusion of escaped char
sunrpc/cache: convert to use string_escape_str()
sunrpc: only call test_bit once in svc_xprt_received
fs: nfsd: Fix signedness bug in compare_blob
sunrpc: add some tracepoints around enqueue and dequeue of svc_xprt
sunrpc: convert to lockless lookup of queued server threads
sunrpc: fix potential races in pool_stats collection
sunrpc: add a rcu_head to svc_rqst and use kfree_rcu to free it
sunrpc: require svc_create callers to pass in meaningful shutdown routine
sunrpc: have svc_wake_up only deal with pool 0
sunrpc: convert sp_task_pending flag to use atomic bitops
sunrpc: move rq_cachetype field to better optimize space
sunrpc: move rq_splice_ok flag into rq_flags
sunrpc: move rq_dropme flag into rq_flags
sunrpc: move rq_usedeferral flag to rq_flags
sunrpc: move rq_local field to rq_flags
sunrpc: add a generic rq_flags field to svc_rqst and move rq_secure to it
nfsd: minor off by one checks in __write_versions()
sunrpc: release svc_pool_map reference when serv allocation fails
...
Pull security layer updates from James Morris:
"In terms of changes, there's general maintenance to the Smack,
SELinux, and integrity code.
The IMA code adds a new kconfig option, IMA_APPRAISE_SIGNED_INIT,
which allows IMA appraisal to require signatures. Support for reading
keys from rootfs before init is call is also added"
* 'next' of git://git.kernel.org/pub/scm/linux/kernel/git/jmorris/linux-security: (23 commits)
selinux: Remove security_ops extern
security: smack: fix out-of-bounds access in smk_parse_smack()
VFS: refactor vfs_read()
ima: require signature based appraisal
integrity: provide a hook to load keys when rootfs is ready
ima: load x509 certificate from the kernel
integrity: provide a function to load x509 certificate from the kernel
integrity: define a new function integrity_read_file()
Security: smack: replace kzalloc with kmem_cache for inode_smack
Smack: Lock mode for the floor and hat labels
ima: added support for new kernel cmdline parameter ima_template_fmt
ima: allocate field pointers array on demand in template_desc_init_fields()
ima: don't allocate a copy of template_fmt in template_desc_init_fields()
ima: display template format in meas. list if template name length is zero
ima: added error messages to template-related functions
ima: use atomic bit operations to protect policy update interface
ima: ignore empty and with whitespaces policy lines
ima: no need to allocate entry for comment
ima: report policy load status
ima: use path names cache
...
Pull aio updates from Benjamin LaHaise.
* git://git.kvack.org/~bcrl/aio-next:
aio: Skip timer for io_getevents if timeout=0
aio: Make it possible to remap aio ring
There are actually two issues this patch addresses. Let me start with
the one I tried to solve in the beginning.
So, in the checkpoint-restore project (criu) we try to dump tasks'
state and restore one back exactly as it was. One of the tasks' state
bits is rings set up with io_setup() call. There's (almost) no problems
in dumping them, there's a problem restoring them -- if I dump a task
with aio ring originally mapped at address A, I want to restore one
back at exactly the same address A. Unfortunately, the io_setup() does
not allow for that -- it mmaps the ring at whatever place mm finds
appropriate (it calls do_mmap_pgoff() with zero address and without
the MAP_FIXED flag).
To make restore possible I'm going to mremap() the freshly created ring
into the address A (under which it was seen before dump). The problem is
that the ring's virtual address is passed back to the user-space as the
context ID and this ID is then used as search key by all the other io_foo()
calls. Reworking this ID to be just some integer doesn't seem to work, as
this value is already used by libaio as a pointer using which this library
accesses memory for aio meta-data.
So, to make restore work we need to make sure that
a) ring is mapped at desired virtual address
b) kioctx->user_id matches this value
Having said that, the patch makes mremap() on aio region update the
kioctx's user_id and mmap_base values.
Here appears the 2nd issue I mentioned in the beginning of this mail.
If (regardless of the C/R dances I do) someone creates an io context
with io_setup(), then mremap()-s the ring and then destroys the context,
the kill_ioctx() routine will call munmap() on wrong (old) address.
This will result in a) aio ring remaining in memory and b) some other
vma get unexpectedly unmapped.
What do you think?
Signed-off-by: Pavel Emelyanov <xemul@parallels.com>
Acked-by: Dmitry Monakhov <dmonakhov@openvz.org>
Signed-off-by: Benjamin LaHaise <bcrl@kvack.org>
This patchset adds execveat(2) for x86, and is derived from Meredydd
Luff's patch from Sept 2012 (https://lkml.org/lkml/2012/9/11/528).
The primary aim of adding an execveat syscall is to allow an
implementation of fexecve(3) that does not rely on the /proc filesystem,
at least for executables (rather than scripts). The current glibc version
of fexecve(3) is implemented via /proc, which causes problems in sandboxed
or otherwise restricted environments.
Given the desire for a /proc-free fexecve() implementation, HPA suggested
(https://lkml.org/lkml/2006/7/11/556) that an execveat(2) syscall would be
an appropriate generalization.
Also, having a new syscall means that it can take a flags argument without
back-compatibility concerns. The current implementation just defines the
AT_EMPTY_PATH and AT_SYMLINK_NOFOLLOW flags, but other flags could be
added in future -- for example, flags for new namespaces (as suggested at
https://lkml.org/lkml/2006/7/11/474).
Related history:
- https://lkml.org/lkml/2006/12/27/123 is an example of someone
realizing that fexecve() is likely to fail in a chroot environment.
- http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=514043 covered
documenting the /proc requirement of fexecve(3) in its manpage, to
"prevent other people from wasting their time".
- https://bugzilla.redhat.com/show_bug.cgi?id=241609 described a
problem where a process that did setuid() could not fexecve()
because it no longer had access to /proc/self/fd; this has since
been fixed.
This patch (of 4):
Add a new execveat(2) system call. execveat() is to execve() as openat()
is to open(): it takes a file descriptor that refers to a directory, and
resolves the filename relative to that.
In addition, if the filename is empty and AT_EMPTY_PATH is specified,
execveat() executes the file to which the file descriptor refers. This
replicates the functionality of fexecve(), which is a system call in other
UNIXen, but in Linux glibc it depends on opening "/proc/self/fd/<fd>" (and
so relies on /proc being mounted).
The filename fed to the executed program as argv[0] (or the name of the
script fed to a script interpreter) will be of the form "/dev/fd/<fd>"
(for an empty filename) or "/dev/fd/<fd>/<filename>", effectively
reflecting how the executable was found. This does however mean that
execution of a script in a /proc-less environment won't work; also, script
execution via an O_CLOEXEC file descriptor fails (as the file will not be
accessible after exec).
Based on patches by Meredydd Luff.
Signed-off-by: David Drysdale <drysdale@google.com>
Cc: Meredydd Luff <meredydd@senatehouse.org>
Cc: Shuah Khan <shuah.kh@samsung.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Andy Lutomirski <luto@amacapital.net>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: "H. Peter Anvin" <hpa@zytor.com>
Cc: Kees Cook <keescook@chromium.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Rich Felker <dalias@aerifal.cx>
Cc: Christoph Hellwig <hch@infradead.org>
Cc: Michael Kerrisk <mtk.manpages@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Similarly to the anon memory counterpart, we can share the mapping's lock
ownership as the interval tree is not modified when doing doing the walk,
only the file page.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Acked-by: Rik van Riel <riel@redhat.com>
Acked-by: "Kirill A. Shutemov" <kirill@shutemov.name>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The i_mmap_mutex is a close cousin of the anon vma lock, both protecting
similar data, one for file backed pages and the other for anon memory. To
this end, this lock can also be a rwsem. In addition, there are some
important opportunities to share the lock when there are no tree
modifications.
This conversion is straightforward. For now, all users take the write
lock.
[sfr@canb.auug.org.au: update fremap.c]
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Acked-by: "Kirill A. Shutemov" <kirill@shutemov.name>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This series is a continuation of the conversion of the i_mmap_mutex to
rwsem, following what we have for the anon memory counterpart. With
Hugh's feedback from the first iteration.
Ultimately, the most obvious paths that require exclusive ownership of the
lock is when we modify the VMA interval tree, via
vma_interval_tree_insert() and vma_interval_tree_remove() families. Cases
such as unmapping, where the ptes content is changed but the tree remains
untouched should make it safe to share the i_mmap_rwsem.
As such, the code of course is straightforward, however the devil is very
much in the details. While its been tested on a number of workloads
without anything exploding, I would not be surprised if there are some
less documented/known assumptions about the lock that could suffer from
these changes. Or maybe I'm just missing something, but either way I
believe its at the point where it could use more eyes and hopefully some
time in linux-next.
Because the lock type conversion is the heart of this patchset,
its worth noting a few comparisons between mutex vs rwsem (xadd):
(i) Same size, no extra footprint.
(ii) Both have CONFIG_XXX_SPIN_ON_OWNER capabilities for
exclusive lock ownership.
(iii) Both can be slightly unfair wrt exclusive ownership, with
writer lock stealing properties, not necessarily respecting
FIFO order for granting the lock when contended.
(iv) Mutexes can be slightly faster than rwsems when
the lock is non-contended.
(v) Both suck at performance for debug (slowpaths), which
shouldn't matter anyway.
Sharing the lock is obviously beneficial, and sem writer ownership is
close enough to mutexes. The biggest winner of these changes is
migration.
As for concrete numbers, the following performance results are for a
4-socket 60-core IvyBridge-EX with 130Gb of RAM.
Both alltests and disk (xfs+ramdisk) workloads of aim7 suite do quite well
with this set, with a steady ~60% throughput (jpm) increase for alltests
and up to ~30% for disk for high amounts of concurrency. Lower counts of
workload users (< 100) does not show much difference at all, so at least
no regressions.
3.18-rc1 3.18-rc1-i_mmap_rwsem
alltests-100 17918.72 ( 0.00%) 28417.97 ( 58.59%)
alltests-200 16529.39 ( 0.00%) 26807.92 ( 62.18%)
alltests-300 16591.17 ( 0.00%) 26878.08 ( 62.00%)
alltests-400 16490.37 ( 0.00%) 26664.63 ( 61.70%)
alltests-500 16593.17 ( 0.00%) 26433.72 ( 59.30%)
alltests-600 16508.56 ( 0.00%) 26409.20 ( 59.97%)
alltests-700 16508.19 ( 0.00%) 26298.58 ( 59.31%)
alltests-800 16437.58 ( 0.00%) 26433.02 ( 60.81%)
alltests-900 16418.35 ( 0.00%) 26241.61 ( 59.83%)
alltests-1000 16369.00 ( 0.00%) 26195.76 ( 60.03%)
alltests-1100 16330.11 ( 0.00%) 26133.46 ( 60.03%)
alltests-1200 16341.30 ( 0.00%) 26084.03 ( 59.62%)
alltests-1300 16304.75 ( 0.00%) 26024.74 ( 59.61%)
alltests-1400 16231.08 ( 0.00%) 25952.35 ( 59.89%)
alltests-1500 16168.06 ( 0.00%) 25850.58 ( 59.89%)
alltests-1600 16142.56 ( 0.00%) 25767.42 ( 59.62%)
alltests-1700 16118.91 ( 0.00%) 25689.58 ( 59.38%)
alltests-1800 16068.06 ( 0.00%) 25599.71 ( 59.32%)
alltests-1900 16046.94 ( 0.00%) 25525.92 ( 59.07%)
alltests-2000 16007.26 ( 0.00%) 25513.07 ( 59.38%)
disk-100 7582.14 ( 0.00%) 7257.48 ( -4.28%)
disk-200 6962.44 ( 0.00%) 7109.15 ( 2.11%)
disk-300 6435.93 ( 0.00%) 6904.75 ( 7.28%)
disk-400 6370.84 ( 0.00%) 6861.26 ( 7.70%)
disk-500 6353.42 ( 0.00%) 6846.71 ( 7.76%)
disk-600 6368.82 ( 0.00%) 6806.75 ( 6.88%)
disk-700 6331.37 ( 0.00%) 6796.01 ( 7.34%)
disk-800 6324.22 ( 0.00%) 6788.00 ( 7.33%)
disk-900 6253.52 ( 0.00%) 6750.43 ( 7.95%)
disk-1000 6242.53 ( 0.00%) 6855.11 ( 9.81%)
disk-1100 6234.75 ( 0.00%) 6858.47 ( 10.00%)
disk-1200 6312.76 ( 0.00%) 6845.13 ( 8.43%)
disk-1300 6309.95 ( 0.00%) 6834.51 ( 8.31%)
disk-1400 6171.76 ( 0.00%) 6787.09 ( 9.97%)
disk-1500 6139.81 ( 0.00%) 6761.09 ( 10.12%)
disk-1600 4807.12 ( 0.00%) 6725.33 ( 39.90%)
disk-1700 4669.50 ( 0.00%) 5985.38 ( 28.18%)
disk-1800 4663.51 ( 0.00%) 5972.99 ( 28.08%)
disk-1900 4674.31 ( 0.00%) 5949.94 ( 27.29%)
disk-2000 4668.36 ( 0.00%) 5834.93 ( 24.99%)
In addition, a 67.5% increase in successfully migrated NUMA pages, thus
improving node locality.
The patch layout is simple but designed for bisection (in case reversion
is needed if the changes break upstream) and easier review:
o Patches 1-4 convert the i_mmap lock from mutex to rwsem.
o Patches 5-10 share the lock in specific paths, each patch
details the rationale behind why it should be safe.
This patchset has been tested with: postgres 9.4 (with brand new hugetlb
support), hugetlbfs test suite (all tests pass, in fact more tests pass
with these changes than with an upstream kernel), ltp, aim7 benchmarks,
memcached and iozone with the -B option for mmap'ing. *Untested* paths
are nommu, memory-failure, uprobes and xip.
This patch (of 8):
Various parts of the kernel acquire and release this mutex, so add
i_mmap_lock_write() and immap_unlock_write() helper functions that will
encapsulate this logic. The next patch will make use of these.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Acked-by: "Kirill A. Shutemov" <kirill@shutemov.name>
Acked-by: Hugh Dickins <hughd@google.com>
Cc: Oleg Nesterov <oleg@redhat.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Srikar Dronamraju <srikar@linux.vnet.ibm.com>
Acked-by: Mel Gorman <mgorman@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
As it is, default ->i_fop has NULL ->open() (along with all other methods).
The only case where it matters is reopening (via procfs symlink) a file that
didn't get its ->f_op from ->i_fop - anything else will have ->i_fop assigned
to something sane (default would fail on read/write/ioctl/etc.).
Unfortunately, such case exists - alloc_file() users, especially
anon_get_file() ones. There we have tons of opened files of very different
kinds sharing the same inode. As the result, attempt to reopen those via
procfs succeeds and you get a descriptor you can't do anything with.
Moreover, in case of sockets we set ->i_fop that will only be used
on such reopen attempts - and put a failing ->open() into it to make sure
those do not succeed.
It would be simpler to put such ->open() into default ->i_fop and leave
it unchanged both for anon inode (as we do anyway) and for socket ones. Result:
* everything going through do_dentry_open() works as it used to
* sock_no_open() kludge is gone
* attempts to reopen anon-inode files fail as they really ought to
* ditto for aio_private_file()
* ditto for perfmon - this one actually tried to imitate sock_no_open()
trick, but failed to set ->i_fop, so in the current tree reopens succeed and
yield completely useless descriptor. Intent clearly had been to fail with
-ENXIO on such reopens; now it actually does.
* everything else that used alloc_file() keeps working - it has ->i_fop
set for its inodes anyway
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull VFS changes from Al Viro:
"First pile out of several (there _definitely_ will be more). Stuff in
this one:
- unification of d_splice_alias()/d_materialize_unique()
- iov_iter rewrite
- killing a bunch of ->f_path.dentry users (and f_dentry macro).
Getting that completed will make life much simpler for
unionmount/overlayfs, since then we'll be able to limit the places
sensitive to file _dentry_ to reasonably few. Which allows to have
file_inode(file) pointing to inode in a covered layer, with dentry
pointing to (negative) dentry in union one.
Still not complete, but much closer now.
- crapectomy in lustre (dead code removal, mostly)
- "let's make seq_printf return nothing" preparations
- assorted cleanups and fixes
There _definitely_ will be more piles"
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (63 commits)
copy_from_iter_nocache()
new helper: iov_iter_kvec()
csum_and_copy_..._iter()
iov_iter.c: handle ITER_KVEC directly
iov_iter.c: convert copy_to_iter() to iterate_and_advance
iov_iter.c: convert copy_from_iter() to iterate_and_advance
iov_iter.c: get rid of bvec_copy_page_{to,from}_iter()
iov_iter.c: convert iov_iter_zero() to iterate_and_advance
iov_iter.c: convert iov_iter_get_pages_alloc() to iterate_all_kinds
iov_iter.c: convert iov_iter_get_pages() to iterate_all_kinds
iov_iter.c: convert iov_iter_npages() to iterate_all_kinds
iov_iter.c: iterate_and_advance
iov_iter.c: macros for iterating over iov_iter
kill f_dentry macro
dcache: fix kmemcheck warning in switch_names
new helper: audit_file()
nfsd_vfs_write(): use file_inode()
ncpfs: use file_inode()
kill f_dentry uses
lockd: get rid of ->f_path.dentry->d_sb
...
Pull quota updates from Jan Kara:
"Quota improvements and some minor cleanups.
The main portion in the pull request are changes which move i_dquot
array from struct inode into fs-private part of an inode which saves
memory for filesystems which don't use VFS quotas"
* 'for_linus' of git://git.kernel.org/pub/scm/linux/kernel/git/jack/linux-fs:
udf: One function call less in udf_fill_super() after error detection
udf: Deletion of unnecessary checks before the function call "iput"
jbd: Deletion of an unnecessary check before the function call "iput"
vfs: Remove i_dquot field from inode
jfs: Convert to private i_dquot field
reiserfs: Convert to private i_dquot field
ocfs2: Convert to private i_dquot field
ext4: Convert to private i_dquot field
ext3: Convert to private i_dquot field
ext2: Convert to private i_dquot field
quota: Use function to provide i_dquot pointers
xfs: Set allowed quota types
gfs2: Set allowed quota types
quota: Allow each filesystem to specify which quota types it supports
quota: Remove const from function declarations
quota: Add log level to printk
Pull the beginning of seq_file cleanup from Steven:
"I'm looking to clean up the seq_file code and to eventually merge the
trace_seq code with seq_file as well, since they basically do the same thing.
Part of this process is to remove the return code of seq_printf() and friends
as they are rather inconsistent. It is better to use the new function
seq_has_overflowed() if you want to stop processing when the buffer
is full. Note, if the buffer is full, the seq_file code will throw away
the contents, allocate a bigger buffer, and then call your code again
to fill in the data. The only thing that breaking out of the function
early does is to save a little time which is probably never noticed.
I started with patches from Joe Perches and modified them as well.
There's many more places that need to be updated before we can convert
seq_printf() and friends to return void. But this patch set introduces
the seq_has_overflowed() and does some initial updates."
integrity_kernel_read() duplicates the file read operations code
in vfs_read(). This patch refactors vfs_read() code creating a
helper function __vfs_read(). It is used by both vfs_read() and
integrity_kernel_read().
Signed-off-by: Dmitry Kasatkin <d.kasatkin@samsung.com>
Signed-off-by: Mimi Zohar <zohar@linux.vnet.ibm.com>
Currently, freezing a filesystem involves calling freeze_super, which locks
sb->s_umount and then calls the fs-specific freeze_fs hook. This makes it
hard for gfs2 (and potentially other cluster filesystems) to use the vfs
freezing code to do freezes on all the cluster nodes.
In order to communicate that a freeze has been requested, and to make sure
that only one node is trying to freeze at a time, gfs2 uses a glock
(sd_freeze_gl). The problem is that there is no hook for gfs2 to acquire
this lock before calling freeze_super. This means that two nodes can
attempt to freeze the filesystem by both calling freeze_super, acquiring
the sb->s_umount lock, and then attempting to grab the cluster glock
sd_freeze_gl. Only one will succeed, and the other will be stuck in
freeze_super, making it impossible to finish freezing the node.
To solve this problem, this patch adds the freeze_super and thaw_super
hooks. If a filesystem implements these hooks, they are called instead of
the vfs freeze_super and thaw_super functions. This means that every
filesystem that implements these hooks must call the vfs freeze_super and
thaw_super functions itself within the hook function to make use of the vfs
freezing code.
Reviewed-by: Jan Kara <jack@suse.cz>
Signed-off-by: Benjamin Marzinski <bmarzins@redhat.com>
Signed-off-by: Steven Whitehouse <swhiteho@redhat.com>
All filesystems using VFS quotas are now converted to use their private
i_dquot fields. Remove the i_dquot field from generic inode structure.
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jan Kara <jack@suse.cz>
i_dquot array is used by relatively few filesystems (ext?, ocfs2, jfs,
reiserfs) so it is beneficial to move this array to fs-private part of
the inode. We cannot just pass quota pointers from filesystems to quota
functions because during quotaon and quotaoff we have to traverse list
of all inodes and manipulate i_dquot pointers for each inode. So we
provide a function which generic quota code can use to get pointer to
the i_dquot array from the filesystem.
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jan Kara <jack@suse.cz>
Currently all filesystems supporting VFS quota support user and group
quotas. With introduction of project quotas this is going to change so
make sure filesystem isn't called for quota type it doesn't support by
introduction of a bitmask determining which quota types each filesystem
supports.
Acked-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Jan Kara <jack@suse.cz>
This function needs to be exported so it can be used by the NFSD module
when responding to the new ALLOCATE and DEALLOCATE operations in NFS
v4.2. Christoph Hellwig suggested renaming the function to stay
consistent with how other vfs functions are named.
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: J. Bruce Fields <bfields@redhat.com>
seq_printf functions shouldn't really check the return value.
Checking seq_has_overflowed() occasionally is used instead.
Update vfs documentation.
Link: http://lkml.kernel.org/p/e37e6e7b76acbdcc3bb4ab2a57c8f8ca1ae11b9a.1412031505.git.joe@perches.com
Cc: David S. Miller <davem@davemloft.net>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Joe Perches <joe@perches.com>
[ did a few clean ups ]
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
Author: David Jeffery <djeffery@redhat.com>
Changes to the basic direct I/O code have broken the raw driver when reading
to the end of a raw device. Instead of returning a short read for a read that
extends partially beyond the device's end or 0 when at the end of the device,
these reads now return EIO.
The raw driver needs the same end of device handling as was added for normal
block devices. Using blkdev_read_iter, which has the needed size checks,
prevents the EIO conditions at the end of the device.
Signed-off-by: David Jeffery <djeffery@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
In an overlay directory that shadows an empty lower directory, say
/mnt/a/empty102, do:
touch /mnt/a/empty102/x
unlink /mnt/a/empty102/x
rmdir /mnt/a/empty102
It's actually harmless, but needs another level of nesting between
I_MUTEX_CHILD and I_MUTEX_NORMAL.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Tested-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Add a simple read-only counter to super_block that indicates how deep this
is in the stack of filesystems. Previously ecryptfs was the only stackable
filesystem and it explicitly disallowed multiple layers of itself.
Overlayfs, however, can be stacked recursively and also may be stacked
on top of ecryptfs or vice versa.
To limit the kernel stack usage we must limit the depth of the
filesystem stack. Initially the limit is set to 2.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Whiteout isn't actually a new file type, but is represented as a char
device (Linus's idea) with 0/0 device number.
This has several advantages compared to introducing a new whiteout file
type:
- no userspace API changes (e.g. trivial to make backups of upper layer
filesystem, without losing whiteouts)
- no fs image format changes (you can boot an old kernel/fsck without
whiteout support and things won't break)
- implementation is trivial
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
It's already duplicated in btrfs and about to be used in overlayfs too.
Move the sticky bit check to an inline helper and call the out-of-line
helper only in the unlikly case of the sticky bit being set.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
We need to be able to check inode permissions (but not filesystem implied
permissions) for stackable filesystems. Expose this interface for overlayfs.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Add a new inode operation i_op->dentry_open(). This is for stacked filesystems
that want to return a struct file from a different filesystem.
Signed-off-by: Miklos Szeredi <mszeredi@suse.cz>
Vladimir Davydov