As previously reported (https://patchwork.kernel.org/patch/8642031/)
it's possible to call shrink_dentry_list with a large number of dentries
(> 10000). This, in turn, could trigger the softlockup detector and
possibly trigger a panic. In addition to the unmount path being
vulnerable to this scenario, at SuSE we've observed similar situation
happening during process exit on processes that touch a lot of dentries.
Here is an excerpt from a crash dump. The number after the colon are
the number of dentries on the list passed to shrink_dentry_list:
PID 99760: 10722
PID 107530: 215
PID 108809: 24134
PID 108877: 21331
PID 141708: 16487
So we want to kill between 15k-25k dentries without yielding.
And one possible call stack looks like:
4 [ffff8839ece41db0] _raw_spin_lock at ffffffff8152a5f8
5 [ffff8839ece41db0] evict at ffffffff811c3026
6 [ffff8839ece41dd0] __dentry_kill at ffffffff811bf258
7 [ffff8839ece41df0] shrink_dentry_list at ffffffff811bf593
8 [ffff8839ece41e18] shrink_dcache_parent at ffffffff811bf830
9 [ffff8839ece41e50] proc_flush_task at ffffffff8120dd61
10 [ffff8839ece41ec0] release_task at ffffffff81059ebd
11 [ffff8839ece41f08] do_exit at ffffffff8105b8ce
12 [ffff8839ece41f78] sys_exit at ffffffff8105bd53
13 [ffff8839ece41f80] system_call_fastpath at ffffffff81532909
While some of the callers of shrink_dentry_list do use cond_resched,
this is not sufficient to prevent softlockups. So just move
cond_resched into shrink_dentry_list from its callers.
David said: I've found hundreds of occurrences of warnings that we emit
when need_resched stays set for a prolonged period of time with the
stack trace that is included in the change log.
Link: http://lkml.kernel.org/r/1521718946-31521-1-git-send-email-nborisov@suse.com
Signed-off-by: Nikolay Borisov <nborisov@suse.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Acked-by: David Rientjes <rientjes@google.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Goldwyn Rodrigues <rgoldwyn@suse.de>
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Davidlohr Bueso <dave@stgolabs.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
I received a report about suspicious growth of unreclaimable slabs on
some machines. I've found that it happens on machines with low memory
pressure, and these unreclaimable slabs are external names attached to
dentries.
External names are allocated using generic kmalloc() function, so they
are accounted as unreclaimable. But they are held by dentries, which
are reclaimable, and they will be reclaimed under the memory pressure.
In particular, this breaks MemAvailable calculation, as it doesn't take
unreclaimable slabs into account. This leads to a silly situation, when
a machine is almost idle, has no memory pressure and therefore has a big
dentry cache. And the resulting MemAvailable is too low to start a new
workload.
To address the issue, the NR_INDIRECTLY_RECLAIMABLE_BYTES counter is
used to track the amount of memory, consumed by external names. The
counter is increased in the dentry allocation path, if an external name
structure is allocated; and it's decreased in the dentry freeing path.
To reproduce the problem I've used the following Python script:
import os
for iter in range (0, 10000000):
try:
name = ("/some_long_name_%d" % iter) + "_" * 220
os.stat(name)
except Exception:
pass
Without this patch:
$ cat /proc/meminfo | grep MemAvailable
MemAvailable: 7811688 kB
$ python indirect.py
$ cat /proc/meminfo | grep MemAvailable
MemAvailable: 2753052 kB
With the patch:
$ cat /proc/meminfo | grep MemAvailable
MemAvailable: 7809516 kB
$ python indirect.py
$ cat /proc/meminfo | grep MemAvailable
MemAvailable: 7749144 kB
[guro@fb.com: fix indirectly reclaimable memory accounting for CONFIG_SLOB]
Link: http://lkml.kernel.org/r/20180312194140.19517-1-guro@fb.com
[guro@fb.com: fix indirectly reclaimable memory accounting]
Link: http://lkml.kernel.org/r/20180313125701.7955-1-guro@fb.com
Link: http://lkml.kernel.org/r/20180305133743.12746-5-guro@fb.com
Signed-off-by: Roman Gushchin <guro@fb.com>
Reviewed-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Mel Gorman <mgorman@techsingularity.net>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Currently d_move(from, to) does the following:
* name/parent of from <- old name/parent of to, from hashed there
* to is unhashed
* name of to is preserved
* if from used to be detached, to gets detached
* if from used to be attached, parent of to <- old parent of from.
That's both user-visibly bogus and complicates reasoning a lot.
Much saner semantics would be
* name/parent of from <- name/parent of to, from hashed there.
* to is unhashed
* name/parent of to is unchanged.
The price, of course, is that old parent of from might lose a reference.
However,
* all potentially cross-directory callers of d_move() have both
parents pinned directly; typically, dentries themselves are grabbed
only after we have grabbed and locked both parents. IOW, the decrement
of old parent's refcount in case of d_move() won't reach zero.
* __d_move() from d_splice_alias() is done to detached alias.
No refcount decrements in that case
* __d_move() from __d_unalias() *can* get the refcount to zero.
So let's grab a reference to alias' old parent before calling __d_unalias()
and dput() it after we'd dropped rename_lock.
That does make d_splice_alias() potentially blocking. However, it has
no callers in non-sleepable contexts (and the case where we'd grown
that dget/dput pair is _very_ rare, so performance is not an issue).
Another thing that needs adjustment is unlocking in the end of __d_move();
folded it in. And cleaned the remnants of bogus ordering from the
"lock them in the beginning" counterpart - it's never been right and
now (well, for 7 years now) we have that thing always serialized on
rename_lock anyway.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
shrink_dentry_list() holds dentry->d_lock and needs to acquire
dentry->d_inode->i_lock. This cannot be done with a spin_lock()
operation because it's the reverse of the regular lock order.
To avoid ABBA deadlocks it is done with a trylock loop.
Trylock loops are problematic in two scenarios:
1) PREEMPT_RT converts spinlocks to 'sleeping' spinlocks, which are
preemptible. As a consequence the i_lock holder can be preempted
by a higher priority task. If that task executes the trylock loop
it will do so forever and live lock.
2) In virtual machines trylock loops are problematic as well. The
VCPU on which the i_lock holder runs can be scheduled out and a
task on a different VCPU can loop for a whole time slice. In the
worst case this can lead to starvation. Commits 47be61845c
("fs/dcache.c: avoid soft-lockup in dput()") and 046b961b45
("shrink_dentry_list(): take parent's d_lock earlier") are
addressing exactly those symptoms.
Avoid the trylock loop by using dentry_kill(). When pruning ancestors,
the same code applies that is used to kill a dentry in dput(). This
also has the benefit that the locking order is now the same. First
the inode is locked, then the parent.
Signed-off-by: John Ogness <john.ogness@linutronix.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
In case when trylock in there fails, deal with it directly in
dentry_kill(). Note that in cases when we drop and retake
->d_lock, we need to recheck whether to retain the dentry.
Another thing is that dropping/retaking ->d_lock might have
ended up with negative dentry turning into positive; that,
of course, can happen only once...
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
In case of trylock failure don't re-add to the list - drop the locks
and carefully get them in the right order. For shrink_dentry_list(),
somebody having grabbed a reference to dentry means that we can
kick it off-list, so if we find dentry being modified under us we
don't need to play silly buggers with retries anyway - off the list
it is.
The locking logics taken out into a helper of its own; lock_parent()
is no longer used for dentries that can be killed under us.
[fix from Eric Biggers folded]
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
A subsequent patch will modify dentry_kill() to call lock_parent().
Move the dentry_kill() implementation "as is" below lock_parent()
first. This will help simplify the review of the subsequent patch
with dentry_kill() changes.
Signed-off-by: John Ogness <john.ogness@linutronix.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Commit 0d98439ea3 ("vfs: use lockred "dead" flag to mark unrecoverably
dead dentries") removed the `ref' parameter in dentry_kill() but its
documentation remained. Remove it.
Signed-off-by: John Ogness <john.ogness@linutronix.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
i_dir_seq is subject to concurrent modification by a cmpxchg or
store-release operation, so ensure that the relaxed access in
d_alloc_parallel uses READ_ONCE.
Reported-by: Peter Zijlstra <peterz@infradead.org>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
If d_alloc_parallel runs concurrently with __d_add, it is possible for
d_alloc_parallel to continuously retry whilst i_dir_seq has been
incremented to an odd value by __d_add:
CPU0:
__d_add
n = start_dir_add(dir);
cmpxchg(&dir->i_dir_seq, n, n + 1) == n
CPU1:
d_alloc_parallel
retry:
seq = smp_load_acquire(&parent->d_inode->i_dir_seq) & ~1;
hlist_bl_lock(b);
bit_spin_lock(0, (unsigned long *)b); // Always succeeds
CPU0:
__d_lookup_done(dentry)
hlist_bl_lock
bit_spin_lock(0, (unsigned long *)b); // Never succeeds
CPU1:
if (unlikely(parent->d_inode->i_dir_seq != seq)) {
hlist_bl_unlock(b);
goto retry;
}
Since the simple bit_spin_lock used to implement hlist_bl_lock does not
provide any fairness guarantees, then CPU1 can starve CPU0 of the lock
and prevent it from reaching end_dir_add(dir), therefore CPU1 cannot
exit its retry loop because the sequence number always has the bottom
bit set.
This patch resolves the livelock by not taking hlist_bl_lock in
d_alloc_parallel if the sequence counter is odd, since any subsequent
masked comparison with i_dir_seq will fail anyway.
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Reported-by: Naresh Madhusudana <naresh.madhusudana@arm.com>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Matthew Wilcox <mawilcox@microsoft.com>
Signed-off-by: Will Deacon <will.deacon@arm.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
In case when dentry passed to lock_parent() is protected from freeing only
by the fact that it's on a shrink list and trylock of parent fails, we
could get hit by __dentry_kill() (and subsequent dentry_kill(parent))
between unlocking dentry and locking presumed parent. We need to recheck
that dentry is alive once we lock both it and parent *and* postpone
rcu_read_unlock() until after that point. Otherwise we could return
a pointer to struct dentry that already is rcu-scheduled for freeing, with
->d_lock held on it; caller's subsequent attempt to unlock it can end
up with memory corruption.
Cc: stable@vger.kernel.org # 3.12+, counting backports
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Pull overlayfs updates from Miklos Szeredi:
"This work from Amir adds NFS export capability to overlayfs. NFS
exporting an overlay filesystem is a challange because we want to keep
track of any copy-up of a file or directory between encoding the file
handle and decoding it.
This is achieved by indexing copied up objects by lower layer file
handle. The index is already used for hard links, this patchset
extends the use to NFS file handle decoding"
* 'overlayfs-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs: (51 commits)
ovl: check ERR_PTR() return value from ovl_encode_fh()
ovl: fix regression in fsnotify of overlay merge dir
ovl: wire up NFS export operations
ovl: lookup indexed ancestor of lower dir
ovl: lookup connected ancestor of dir in inode cache
ovl: hash non-indexed dir by upper inode for NFS export
ovl: decode pure lower dir file handles
ovl: decode indexed dir file handles
ovl: decode lower file handles of unlinked but open files
ovl: decode indexed non-dir file handles
ovl: decode lower non-dir file handles
ovl: encode lower file handles
ovl: copy up before encoding non-connectable dir file handle
ovl: encode non-indexed upper file handles
ovl: decode connected upper dir file handles
ovl: decode pure upper file handles
ovl: encode pure upper file handles
ovl: document NFS export
vfs: factor out helpers d_instantiate_anon() and d_alloc_anon()
ovl: store 'has_upper' and 'opaque' as bit flags
...
cache objects. This is good, but still leaves a lot of kernel memory
available to be copied to/from userspace in the face of bugs. To further
restrict what memory is available for copying, this creates a way to
whitelist specific areas of a given slab cache object for copying to/from
userspace, allowing much finer granularity of access control. Slab caches
that are never exposed to userspace can declare no whitelist for their
objects, thereby keeping them unavailable to userspace via dynamic copy
operations. (Note, an implicit form of whitelisting is the use of constant
sizes in usercopy operations and get_user()/put_user(); these bypass all
hardened usercopy checks since these sizes cannot change at runtime.)
This new check is WARN-by-default, so any mistakes can be found over the
next several releases without breaking anyone's system.
The series has roughly the following sections:
- remove %p and improve reporting with offset
- prepare infrastructure and whitelist kmalloc
- update VFS subsystem with whitelists
- update SCSI subsystem with whitelists
- update network subsystem with whitelists
- update process memory with whitelists
- update per-architecture thread_struct with whitelists
- update KVM with whitelists and fix ioctl bug
- mark all other allocations as not whitelisted
- update lkdtm for more sensible test overage
-----BEGIN PGP SIGNATURE-----
Version: GnuPG v1
Comment: Kees Cook <kees@outflux.net>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=tzmJ
-----END PGP SIGNATURE-----
Merge tag 'usercopy-v4.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux
Pull hardened usercopy whitelisting from Kees Cook:
"Currently, hardened usercopy performs dynamic bounds checking on slab
cache objects. This is good, but still leaves a lot of kernel memory
available to be copied to/from userspace in the face of bugs.
To further restrict what memory is available for copying, this creates
a way to whitelist specific areas of a given slab cache object for
copying to/from userspace, allowing much finer granularity of access
control.
Slab caches that are never exposed to userspace can declare no
whitelist for their objects, thereby keeping them unavailable to
userspace via dynamic copy operations. (Note, an implicit form of
whitelisting is the use of constant sizes in usercopy operations and
get_user()/put_user(); these bypass all hardened usercopy checks since
these sizes cannot change at runtime.)
This new check is WARN-by-default, so any mistakes can be found over
the next several releases without breaking anyone's system.
The series has roughly the following sections:
- remove %p and improve reporting with offset
- prepare infrastructure and whitelist kmalloc
- update VFS subsystem with whitelists
- update SCSI subsystem with whitelists
- update network subsystem with whitelists
- update process memory with whitelists
- update per-architecture thread_struct with whitelists
- update KVM with whitelists and fix ioctl bug
- mark all other allocations as not whitelisted
- update lkdtm for more sensible test overage"
* tag 'usercopy-v4.16-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/kees/linux: (38 commits)
lkdtm: Update usercopy tests for whitelisting
usercopy: Restrict non-usercopy caches to size 0
kvm: x86: fix KVM_XEN_HVM_CONFIG ioctl
kvm: whitelist struct kvm_vcpu_arch
arm: Implement thread_struct whitelist for hardened usercopy
arm64: Implement thread_struct whitelist for hardened usercopy
x86: Implement thread_struct whitelist for hardened usercopy
fork: Provide usercopy whitelisting for task_struct
fork: Define usercopy region in thread_stack slab caches
fork: Define usercopy region in mm_struct slab caches
net: Restrict unwhitelisted proto caches to size 0
sctp: Copy struct sctp_sock.autoclose to userspace using put_user()
sctp: Define usercopy region in SCTP proto slab cache
caif: Define usercopy region in caif proto slab cache
ip: Define usercopy region in IP proto slab cache
net: Define usercopy region in struct proto slab cache
scsi: Define usercopy region in scsi_sense_cache slab cache
cifs: Define usercopy region in cifs_request slab cache
vxfs: Define usercopy region in vxfs_inode slab cache
ufs: Define usercopy region in ufs_inode_cache slab cache
...
Merge KASAN word-at-a-time fixups from Andrey Ryabinin.
The word-at-a-time optimizations have caused headaches for KASAN, since
the whole point is that we access byte streams in bigger chunks, and
KASAN can be unhappy about the potential extra access at the end of the
string.
We used to have a horrible hack in dcache, and then people got
complaints from the strscpy() case. This fixes it all up properly, by
adding an explicit helper for the "access byte stream one word at a
time" case.
* emailed patches from Andrey Ryabinin <aryabinin@virtuozzo.com>:
fs: dcache: Revert "manually unpoison dname after allocation to shut up kasan's reports"
fs/dcache: Use read_word_at_a_time() in dentry_string_cmp()
lib/strscpy: Shut up KASAN false-positives in strscpy()
compiler.h: Add read_word_at_a_time() function.
compiler.h, kasan: Avoid duplicating __read_once_size_nocheck()
This reverts commit df4c0e36f1.
It's no longer needed since dentry_string_cmp() now uses
read_word_at_a_time() to avoid kasan's reports.
Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
dentry_string_cmp() performs the word-at-a-time reads from 'cs' and may
read slightly more than it was requested in kmallac(). Normally this
would make KASAN to report out-of-bounds access, but this was
workarounded by commit df4c0e36f1 ("fs: dcache: manually unpoison
dname after allocation to shut up kasan's reports").
This workaround is not perfect, since it allows out-of-bounds access to
dentry's name for all the code, not just in dentry_string_cmp().
So it would be better to use read_word_at_a_time() instead and revert
commit df4c0e36f1.
Signed-off-by: Andrey Ryabinin <aryabinin@virtuozzo.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull dcache updates from Al Viro:
"Neil Brown's d_move()/d_path() race fix"
* 'work.dcache' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
VFS: close race between getcwd() and d_move()
Pull misc vfs updates from Al Viro:
"All kinds of misc stuff, without any unifying topic, from various
people.
Neil's d_anon patch, several bugfixes, introduction of kvmalloc
analogue of kmemdup_user(), extending bitfield.h to deal with
fixed-endians, assorted cleanups all over the place..."
* 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs: (28 commits)
alpha: osf_sys.c: use timespec64 where appropriate
alpha: osf_sys.c: fix put_tv32 regression
jffs2: Fix use-after-free bug in jffs2_iget()'s error handling path
dcache: delete unused d_hash_mask
dcache: subtract d_hash_shift from 32 in advance
fs/buffer.c: fold init_buffer() into init_page_buffers()
fs: fold __inode_permission() into inode_permission()
fs: add RWF_APPEND
sctp: use vmemdup_user() rather than badly open-coding memdup_user()
snd_ctl_elem_init_enum_names(): switch to vmemdup_user()
replace_user_tlv(): switch to vmemdup_user()
new primitive: vmemdup_user()
memdup_user(): switch to GFP_USER
eventfd: fold eventfd_ctx_get() into eventfd_ctx_fileget()
eventfd: fold eventfd_ctx_read() into eventfd_read()
eventfd: convert to use anon_inode_getfd()
nfs4file: get rid of pointless include of btrfs.h
uvc_v4l2: clean copyin/copyout up
vme_user: don't use __copy_..._user()
usx2y: don't bother with memdup_user() for 16-byte structure
...
Those helpers are going to be used by overlayfs to implement
NFS export decode.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
Directory index entries should have 'upper' xattr pointing to the real
upper dir. Verifying that the upper dir file handle is not stale is
expensive, so only verify stale directory index entries on mount if
NFS export feature is enabled.
Signed-off-by: Amir Goldstein <amir73il@gmail.com>
Signed-off-by: Miklos Szeredi <mszeredi@redhat.com>
VFS pathnames are stored in the names_cache slab cache, either inline
or across an entire allocation entry (when approaching PATH_MAX). These
are copied to/from userspace, so they must be entirely whitelisted.
cache object allocation:
include/linux/fs.h:
#define __getname() kmem_cache_alloc(names_cachep, GFP_KERNEL)
example usage trace:
strncpy_from_user+0x4d/0x170
getname_flags+0x6f/0x1f0
user_path_at_empty+0x23/0x40
do_mount+0x69/0xda0
SyS_mount+0x83/0xd0
fs/namei.c:
getname_flags(...):
...
result = __getname();
...
kname = (char *)result->iname;
result->name = kname;
len = strncpy_from_user(kname, filename, EMBEDDED_NAME_MAX);
...
if (unlikely(len == EMBEDDED_NAME_MAX)) {
const size_t size = offsetof(struct filename, iname[1]);
kname = (char *)result;
result = kzalloc(size, GFP_KERNEL);
...
result->name = kname;
len = strncpy_from_user(kname, filename, PATH_MAX);
In support of usercopy hardening, this patch defines the entire cache
object in the names_cache slab cache as whitelisted, since it may entirely
hold name strings to be copied to/from userspace.
This patch is verbatim from Brad Spengler/PaX Team's PAX_USERCOPY
whitelisting code in the last public patch of grsecurity/PaX based on my
understanding of the code. Changes or omissions from the original code are
mine and don't reflect the original grsecurity/PaX code.
Signed-off-by: David Windsor <dave@nullcore.net>
[kees: adjust commit log, add usage trace]
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
When a dentry name is short enough, it can be stored directly in the
dentry itself (instead in a separate kmalloc allocation). These dentry
short names, stored in struct dentry.d_iname and therefore contained in
the dentry_cache slab cache, need to be coped to userspace.
cache object allocation:
fs/dcache.c:
__d_alloc(...):
...
dentry = kmem_cache_alloc(dentry_cache, ...);
...
dentry->d_name.name = dentry->d_iname;
example usage trace:
filldir+0xb0/0x140
dcache_readdir+0x82/0x170
iterate_dir+0x142/0x1b0
SyS_getdents+0xb5/0x160
fs/readdir.c:
(called via ctx.actor by dir_emit)
filldir(..., const char *name, ...):
...
copy_to_user(..., name, namlen)
fs/libfs.c:
dcache_readdir(...):
...
next = next_positive(dentry, p, 1)
...
dir_emit(..., next->d_name.name, ...)
In support of usercopy hardening, this patch defines a region in the
dentry_cache slab cache in which userspace copy operations are allowed.
This region is known as the slab cache's usercopy region. Slab caches can
now check that each dynamic copy operation involving cache-managed memory
falls entirely within the slab's usercopy region.
This patch is modified from Brad Spengler/PaX Team's PAX_USERCOPY
whitelisting code in the last public patch of grsecurity/PaX based on my
understanding of the code. Changes or omissions from the original code are
mine and don't reflect the original grsecurity/PaX code.
Signed-off-by: David Windsor <dave@nullcore.net>
[kees: adjust hunks for kmalloc-specific things moved later]
[kees: adjust commit log, provide usage trace]
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Kees Cook <keescook@chromium.org>
d_move() will call __d_drop() and then __d_rehash()
on the dentry being moved. This creates a small window
when the dentry appears to be unhashed. Many tests
of d_unhashed() are made under ->d_lock and so are safe
from racing with this window, but some aren't.
In particular, getcwd() calls d_unlinked() (which calls
d_unhashed()) without d_lock protection, so it can race.
This races has been seen in practice with lustre, which uses d_move() as
part of name lookup. See:
https://jira.hpdd.intel.com/browse/LU-9735
It could race with a regular rename(), and result in ENOENT instead
of either the 'before' or 'after' name.
The race can be demonstrated with a simple program which
has two threads, one renaming a directory back and forth
while another calls getcwd() within that directory: it should never
fail, but does. See:
https://patchwork.kernel.org/patch/9455345/
We could fix this race by taking d_lock and rechecking when
d_unhashed() reports true. Alternately when can remove the window,
which is the approach this patch takes.
___d_drop() is introduce which does *not* clear d_hash.pprev
so the dentry still appears to be hashed. __d_drop() calls
___d_drop(), then clears d_hash.pprev.
__d_move() now uses ___d_drop() and only clears d_hash.pprev
when not rehashing.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The original purpose of the per-superblock d_anon list was to
keep disconnected dentries in the cache between consecutive
requests to the NFS server. Dentries can be disconnected if
a client holds a file open and repeatedly performs IO on it,
and if the server drops the dentry, whether due to memory
pressure, server restart, or "echo 3 > /proc/sys/vm/drop_caches".
This purpose was thwarted by commit 75a6f82a0d ("freeing unlinked
file indefinitely delayed") which caused disconnected dentries
to be freed as soon as their refcount reached zero.
This means that, when a dentry being used by nfsd gets disconnected, a
new one needs to be allocated for every request (unless requests
overlap). As the dentry has no name, no parent, and no children,
there is little of value to cache. As small memory allocations are
typically fast (from per-cpu free lists) this likely has little cost.
This means that the original purpose of s_anon is no longer relevant:
there is no longer any need to keep disconnected dentries on a list so
they appear to be hashed.
However, s_anon now has a new use. When you mount an NFS filesystem,
the dentry stored in s_root is just a placebo. The "real" root dentry
is allocated using d_obtain_root() and so it kept on the s_anon list.
I don't know the reason for this, but suspect it related to NFSv4
where a mount of "server:/some/path" require NFS to look up the root
filehandle on the server, then walk down "/some" and "/path" to get
the filehandle to mount.
Whatever the reason, NFS depends on the s_anon list and on
shrink_dcache_for_umount() pruning all dentries on this list. So we
cannot simply remove s_anon.
We could just leave the code unchanged, but apart from that being
potentially confusing, the (unfair) bit-spin-lock which protects
s_anon can become a bottle neck when lots of disconnected dentries are
being created.
So this patch renames s_anon to s_roots, and stops storing
disconnected dentries on the list. Only dentries obtained with
d_obtain_root() are now stored on this list. There are many fewer of
these (only NFS and NILFS2 use the call, and only during filesystem
mount) so contention on the bit-lock will not be a problem.
Possibly an alternate solution should be found for NFS and NILFS2, but
that would require understanding their needs first.
Signed-off-by: NeilBrown <neilb@suse.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Preempt counter APIs have been split out, currently, hardirq.h just
includes irq_enter/exit APIs which are not used by vfs at all.
So, remove the unused hardirq.h.
Signed-off-by: Yang Shi <yang.s@alibaba-inc.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
The code in __d_alloc() carefully orders filling in the NUL character
of the name (and the length, hash, and the name itself) with assigning
of the name itself. However, prepend_name() does not order the accesses
to the ->name and ->len fields, other than on TSO systems. This commit
therefore replaces prepend_name()'s READ_ONCE() of ->name with an
smp_load_acquire(), which orders against the subsequent READ_ONCE() of
->len. Because READ_ONCE() now incorporates smp_read_barrier_depends(),
prepend_name()'s smp_read_barrier_depends() is removed. Finally,
to save a line, the smp_wmb()/store pair in __d_alloc() is replaced
by smp_store_release().
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: <linux-fsdevel@vger.kernel.org>
Patch series "kmemcheck: kill kmemcheck", v2.
As discussed at LSF/MM, kill kmemcheck.
KASan is a replacement that is able to work without the limitation of
kmemcheck (single CPU, slow). KASan is already upstream.
We are also not aware of any users of kmemcheck (or users who don't
consider KASan as a suitable replacement).
The only objection was that since KASAN wasn't supported by all GCC
versions provided by distros at that time we should hold off for 2
years, and try again.
Now that 2 years have passed, and all distros provide gcc that supports
KASAN, kill kmemcheck again for the very same reasons.
This patch (of 4):
Remove kmemcheck annotations, and calls to kmemcheck from the kernel.
[alexander.levin@verizon.com: correctly remove kmemcheck call from dma_map_sg_attrs]
Link: http://lkml.kernel.org/r/20171012192151.26531-1-alexander.levin@verizon.com
Link: http://lkml.kernel.org/r/20171007030159.22241-2-alexander.levin@verizon.com
Signed-off-by: Sasha Levin <alexander.levin@verizon.com>
Cc: Alexander Potapenko <glider@google.com>
Cc: Eric W. Biederman <ebiederm@xmission.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: Steven Rostedt <rostedt@goodmis.org>
Cc: Tim Hansen <devtimhansen@gmail.com>
Cc: Vegard Nossum <vegardno@ifi.uio.no>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
For several reasons, it is desirable to use {READ,WRITE}_ONCE() in
preference to ACCESS_ONCE(), and new code is expected to use one of the
former. So far, there's been no reason to change most existing uses of
ACCESS_ONCE(), as these aren't currently harmful.
However, for some features it is necessary to instrument reads and
writes separately, which is not possible with ACCESS_ONCE(). This
distinction is critical to correct operation.
It's possible to transform the bulk of kernel code using the Coccinelle
script below. However, this doesn't handle comments, leaving references
to ACCESS_ONCE() instances which have been removed. As a preparatory
step, this patch converts the dcache code and comments to use
{READ,WRITE}_ONCE() consistently.
----
virtual patch
@ depends on patch @
expression E1, E2;
@@
- ACCESS_ONCE(E1) = E2
+ WRITE_ONCE(E1, E2)
@ depends on patch @
expression E;
@@
- ACCESS_ONCE(E)
+ READ_ONCE(E)
----
Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Signed-off-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: davem@davemloft.net
Cc: linux-arch@vger.kernel.org
Cc: mpe@ellerman.id.au
Cc: shuah@kernel.org
Cc: snitzer@redhat.com
Cc: thor.thayer@linux.intel.com
Cc: tj@kernel.org
Cc: will.deacon@arm.com
Link: http://lkml.kernel.org/r/1508792849-3115-4-git-send-email-paulmck@linux.vnet.ibm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
READ_ONCE() now has an implicit smp_read_barrier_depends() call, so it
can be used instead of lockless_dereference() without any change in
semantics.
Signed-off-by: Will Deacon <will.deacon@arm.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1508840570-22169-4-git-send-email-will.deacon@arm.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Pull ->s_options removal from Al Viro:
"Preparations for fsmount/fsopen stuff (coming next cycle). Everything
gets moved to explicit ->show_options(), killing ->s_options off +
some cosmetic bits around fs/namespace.c and friends. Basically, the
stuff needed to work with fsmount series with minimum of conflicts
with other work.
It's not strictly required for this merge window, but it would reduce
the PITA during the coming cycle, so it would be nice to have those
bits and pieces out of the way"
* 'work.mount' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
isofs: Fix isofs_show_options()
VFS: Kill off s_options and helpers
orangefs: Implement show_options
9p: Implement show_options
isofs: Implement show_options
afs: Implement show_options
affs: Implement show_options
befs: Implement show_options
spufs: Implement show_options
bpf: Implement show_options
ramfs: Implement show_options
pstore: Implement show_options
omfs: Implement show_options
hugetlbfs: Implement show_options
VFS: Don't use save/replace_mount_options if not using generic_show_options
VFS: Provide empty name qstr
VFS: Make get_filesystem() return the affected filesystem
VFS: Clean up whitespace in fs/namespace.c and fs/super.c
Provide a function to create a NUL-terminated string from unterminated data
__list_lru_walk_one() acquires nlru spin lock (nlru->lock) for longer
duration if there are more number of items in the lru list. As per the
current code, it can hold the spin lock for upto maximum UINT_MAX
entries at a time. So if there are more number of items in the lru
list, then "BUG: spinlock lockup suspected" is observed in the below
path:
spin_bug+0x90
do_raw_spin_lock+0xfc
_raw_spin_lock+0x28
list_lru_add+0x28
dput+0x1c8
path_put+0x20
terminate_walk+0x3c
path_lookupat+0x100
filename_lookup+0x6c
user_path_at_empty+0x54
SyS_faccessat+0xd0
el0_svc_naked+0x24
This nlru->lock is acquired by another CPU in this path -
d_lru_shrink_move+0x34
dentry_lru_isolate_shrink+0x48
__list_lru_walk_one.isra.10+0x94
list_lru_walk_node+0x40
shrink_dcache_sb+0x60
do_remount_sb+0xbc
do_emergency_remount+0xb0
process_one_work+0x228
worker_thread+0x2e0
kthread+0xf4
ret_from_fork+0x10
Fix this lockup by reducing the number of entries to be shrinked from
the lru list to 1024 at once. Also, add cond_resched() before
processing the lru list again.
Link: http://marc.info/?t=149722864900001&r=1&w=2
Link: http://lkml.kernel.org/r/1498707575-2472-1-git-send-email-stummala@codeaurora.org
Signed-off-by: Sahitya Tummala <stummala@codeaurora.org>
Suggested-by: Jan Kara <jack@suse.cz>
Suggested-by: Vladimir Davydov <vdavydov.dev@gmail.com>
Acked-by: Vladimir Davydov <vdavydov.dev@gmail.com>
Cc: Alexander Polakov <apolyakov@beget.ru>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Pull misc filesystem updates from Al Viro:
"Assorted normal VFS / filesystems stuff..."
* 'work.misc' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/vfs:
dentry name snapshots
Make statfs properly return read-only state after emergency remount
fs/dcache: init in_lookup_hashtable
minix: Deinline get_block, save 2691 bytes
fs: Reorder inode_owner_or_capable() to avoid needless
fs: warn in case userspace lied about modprobe return
take_dentry_name_snapshot() takes a safe snapshot of dentry name;
if the name is a short one, it gets copied into caller-supplied
structure, otherwise an extra reference to external name is grabbed
(those are never modified). In either case the pointer to stable
string is stored into the same structure.
dentry must be held by the caller of take_dentry_name_snapshot(),
but may be freely dropped afterwards - the snapshot will stay
until destroyed by release_dentry_name_snapshot().
Intended use:
struct name_snapshot s;
take_dentry_name_snapshot(&s, dentry);
...
access s.name
...
release_dentry_name_snapshot(&s);
Replaces fsnotify_oldname_...(), gets used in fsnotify to obtain the name
to pass down with event.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Provide an empty name (ie. "") qstr for general use.
Signed-off-by: David Howells <dhowells@redhat.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
in_lookup_hashtable was introduced in commit 94bdd655ca ("parallel
lookups machinery, part 3") and never initialized but since it is in
the data it is all zeros. But we need this for -RT.
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
Cc: linux-fsdevel@vger.kernel.org
Signed-off-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
It's not hard to trigger a bunch of d_invalidate() on the same
dentry in parallel. They end up fighting each other - any
dentry picked for removal by one will be skipped by the rest
and we'll go for the next iteration through the entire
subtree, even if everything is being skipped. Morevoer, we
immediately go back to scanning the subtree. The only thing
we really need is to dissolve all mounts in the subtree and
as soon as we've nothing left to do, we can just unhash the
dentry and bugger off.
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
By default we set DCACHE_REFERENCED and I_REFERENCED on any dentry or
inode we create. This is problematic as this means that it takes two
trips through the LRU for any of these objects to be reclaimed,
regardless of their actual lifetime. With enough pressure from these
caches we can easily evict our working set from page cache with single
use objects. So instead only set *REFERENCED if we've already been
added to the LRU list. This means that we've been touched since the
first time we were accessed, and so more likely to need to hang out in
cache.
To illustrate this issue I wrote the following scripts
https://github.com/josefbacik/debug-scripts/tree/master/cache-pressure
on my test box. It is a single socket 4 core CPU with 16gib of RAM and
I tested on an Intel 2tib NVME drive. The cache-pressure.sh script
creates a new file system and creates 2 6.5gib files in order to take up
13gib of the 16gib of ram with pagecache. Then it runs a test program
that reads these 2 files in a loop, and keeps track of how often it has
to read bytes for each loop. On an ideal system with no pressure we
should have to read 0 bytes indefinitely. The second thing this script
does is start a fs_mark job that creates a ton of 0 length files,
putting pressure on the system with slab only allocations. On exit the
script prints out how many bytes were read by the read-file program.
The results are as follows
Without patch:
/mnt/btrfs-test/reads/file1: total read during loops 27262988288
/mnt/btrfs-test/reads/file2: total read during loops 27262976000
With patch:
/mnt/btrfs-test/reads/file2: total read during loops 18640457728
/mnt/btrfs-test/reads/file1: total read during loops 9565376512
This patch results in a 50% reduction of the amount of pages evicted
from our working set.
Signed-off-by: Josef Bacik <jbacik@fb.com>
Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
