When do_balance() balances the tree, a trick is performed to
provide the ability for other tree writers/readers to check whether
do_balance() is executing concurrently (requires CONFIG_REISERFS_CHECK).
This is done to protect concurrent accesses to the tree. The trick
is the following:
When do_balance is called, a unique global variable called cur_tb
takes a pointer to the current tree to be rebalanced.
Once do_balance finishes its work, cur_tb takes the NULL value.
Then, concurrent tree readers/writers just have to check the value
of cur_tb to ensure do_balance isn't executing concurrently.
If it is, then it proves that schedule() occured on do_balance(),
which then relaxed the bkl that protected the tree.
Now that the bkl has be turned into a mutex, this check is still
fine even though do_balance() becomes preemptible: the write lock
will not be automatically released on schedule(), so the tree is
still protected.
But this is only fine if we have a single reiserfs mountpoint.
Indeed, because the bkl is a global lock, it didn't allowed
concurrent executions between a tree reader/writer in a mount point
and a do_balance() on another tree from another mountpoint.
So assuming all these readers/writers weren't supposed to be
reentrant, the current check now sometimes detect false positives with
the current per-superblock mutex which allows this reentrancy.
This patch keeps the concurrent tree accesses check but moves it
per superblock, so that only trees from a same mount point are
checked to be not accessed concurrently.
[ Impact: fix spurious panic while running several reiserfs mount-points ]
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
While searching a pathname, an inode mutex can be acquired
in do_lookup() which calls reiserfs_lookup() which in turn
acquires the write lock.
On the other side reiserfs_fill_super() can acquire the write_lock
and then call reiserfs_lookup_privroot() which can acquire an
inode mutex (the root of the mount point).
So we theoretically risk an AB - BA lock inversion that could lead
to a deadlock.
As for other lock dependencies found since the bkl to mutex
conversion, the fix is to use reiserfs_mutex_lock_safe() which
drops the lock dependency to the write lock.
[ Impact: fix a possible deadlock with reiserfs ]
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
The goal of fs_changed() is to check whether the tree changed during a
schedule(). This is a BKL legacy.
A recent patch added an explicit unconditional release/reacquire of the
write lock around the cond_resched() called inside fs_changed.
But it's wasteful to unconditionally do that, we are creating superfluous
lock contention in !TIF_NEED_RESCHED case.
This patch manage that by calling reiserfs_cond_resched() from fs_changed()
which only releases the lock if we are going to reschedule.
[ Impact: inject less lock contention and tree job retries ]
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Usually, when we call cond_resched(), we want the write lock
to be released and then reacquired once we return from scheduling.
Not only does it follow the previous bkl based locking scheme, but
it also let other waiters to get the lock.
But if we aren't going to reschedule(), such as in !TIF_NEED_RESCHED
case, it's useless to release the lock. Worse, if we release and reacquire
the lock whereas it is not needed, we create useless contentions. Also
if someone takes the lock while we are modifying or reading the tree,
there are good chances we'll have to retry our operation, eg if the
block we were seeeking has moved.
So this patch introduces a helper which only unlock the write lock
if we are going to schedule.
[ Impact: prepare to inject less lock contention and less tree operation attempts ]
Reported-by: Andi Kleen <andi@firstfloor.org>
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
fs_changed() is a macro used by reiserfs to check whether its tree has been
rebalanced. It has been designed to check parallel changes on the tree after
calling a sleeping function, which released the Bkl.
fs_changed() also calls cond_resched(), so that if rescheduling is needed,
we are in the best place to do that, since we check if the tree has changed
just after (because of the bkl release on schedule()).
Even if we are not anymore using the Bkl, we still want to release the lock
while we reschedule, so that other waiters for the lock can acquire it safely,
because of the following __fs_changed() check.
[ Impact: release the reiserfs write lock when it is not needed ]
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Chris Mason <chris.mason@oracle.com>
Cc: Alexander Beregalov <a.beregalov@gmail.com>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Sometimes we don't want to recursively hold the per superblock write
lock because we want to be sure it is actually released when we come
to sleep.
This patch introduces the necessary tools for that.
reiserfs_write_lock_once() does the same job than reiserfs_write_lock()
except that it won't try to acquire recursively the lock if the current
task already owns it. Also the lock_depth before the call of this function
is returned.
reiserfs_write_unlock_once() unlock only if reiserfs_write_lock_once()
returned a depth equal to -1, ie: only if it actually locked.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Alessio Igor Bogani <abogani@texware.it>
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Alexander Beregalov <a.beregalov@gmail.com>
Cc: Chris Mason <chris.mason@oracle.com>
LKML-Reference: <1239680065-25013-2-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch is an attempt to remove the Bkl based locking scheme from
reiserfs and is intended.
It is a bit inspired from an old attempt by Peter Zijlstra:
http://lkml.indiana.edu/hypermail/linux/kernel/0704.2/2174.html
The bkl is heavily used in this filesystem to prevent from
concurrent write accesses on the filesystem.
Reiserfs makes a deep use of the specific properties of the Bkl:
- It can be acqquired recursively by a same task
- It is released on the schedule() calls and reacquired when schedule() returns
The two properties above are a roadmap for the reiserfs write locking so it's
very hard to simply replace it with a common mutex.
- We need a recursive-able locking unless we want to restructure several blocks
of the code.
- We need to identify the sites where the bkl was implictly relaxed
(schedule, wait, sync, etc...) so that we can in turn release and
reacquire our new lock explicitly.
Such implicit releases of the lock are often required to let other
resources producer/consumer do their job or we can suffer unexpected
starvations or deadlocks.
So the new lock that replaces the bkl here is a per superblock mutex with a
specific property: it can be acquired recursively by a same task, like the
bkl.
For such purpose, we integrate a lock owner and a lock depth field on the
superblock information structure.
The first axis on this patch is to turn reiserfs_write_(un)lock() function
into a wrapper to manage this mutex. Also some explicit calls to
lock_kernel() have been converted to reiserfs_write_lock() helpers.
The second axis is to find the important blocking sites (schedule...(),
wait_on_buffer(), sync_dirty_buffer(), etc...) and then apply an explicit
release of the write lock on these locations before blocking. Then we can
safely wait for those who can give us resources or those who need some.
Typically this is a fight between the current writer, the reiserfs workqueue
(aka the async commiter) and the pdflush threads.
The third axis is a consequence of the second. The write lock is usually
on top of a lock dependency chain which can include the journal lock, the
flush lock or the commit lock. So it's dangerous to release and trying to
reacquire the write lock while we still hold other locks.
This is fine with the bkl:
T1 T2
lock_kernel()
mutex_lock(A)
unlock_kernel()
// do something
lock_kernel()
mutex_lock(A) -> already locked by T1
schedule() (and then unlock_kernel())
lock_kernel()
mutex_unlock(A)
....
This is not fine with a mutex:
T1 T2
mutex_lock(write)
mutex_lock(A)
mutex_unlock(write)
// do something
mutex_lock(write)
mutex_lock(A) -> already locked by T1
schedule()
mutex_lock(write) -> already locked by T2
deadlock
The solution in this patch is to provide a helper which releases the write
lock and sleep a bit if we can't lock a mutex that depend on it. It's another
simulation of the bkl behaviour.
The last axis is to locate the fs callbacks that are called with the bkl held,
according to Documentation/filesystem/Locking.
Those are:
- reiserfs_remount
- reiserfs_fill_super
- reiserfs_put_super
Reiserfs didn't need to explicitly lock because of the context of these callbacks.
But now we must take care of that with the new locking.
After this patch, reiserfs suffers from a slight performance regression (for now).
On UP, a high volume write with dd reports an average of 27 MB/s instead
of 30 MB/s without the patch applied.
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Reviewed-by: Ingo Molnar <mingo@elte.hu>
Cc: Jeff Mahoney <jeffm@suse.com>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Bron Gondwana <brong@fastmail.fm>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Alexander Viro <viro@zeniv.linux.org.uk>
LKML-Reference: <1239070789-13354-1-git-send-email-fweisbec@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* git://git.kernel.org/pub/scm/linux/kernel/git/agk/linux-2.6-dm:
dm snapshot: fix on disk chunk size validation
dm exception store: split set_chunk_size
dm snapshot: fix header corruption race on invalidation
dm snapshot: refactor zero_disk_area to use chunk_io
dm log: userspace add luid to distinguish between concurrent log instances
dm raid1: do not allow log_failure variable to unset after being set
dm log: remove incorrect field from userspace table output
dm log: fix userspace status output
dm stripe: expose correct io hints
dm table: add more context to terse warning messages
dm table: fix queue_limit checking device iterator
dm snapshot: implement iterate devices
dm multipath: fix oops when request based io fails when no paths
Tom Horsley reports that his debugger hangs when it tries to read
/proc/pid_of_tracee/maps, this happens since
"mm_for_maps: take ->cred_guard_mutex to fix the race with exec"
04b836cbf19e885f8366bccb2e4b0474346c02d
commit in 2.6.31.
But the root of the problem lies in the fact that do_execve() path calls
tracehook_report_exec() which can stop if the tracer sets PT_TRACE_EXEC.
The tracee must not sleep in TASK_TRACED holding this mutex. Even if we
remove ->cred_guard_mutex from mm_for_maps() and proc_pid_attr_write(),
another task doing PTRACE_ATTACH should not hang until it is killed or the
tracee resumes.
With this patch do_execve() does not use ->cred_guard_mutex directly and
we do not hold it throughout, instead:
- introduce prepare_bprm_creds() helper, it locks the mutex
and calls prepare_exec_creds() to initialize bprm->cred.
- install_exec_creds() drops the mutex after commit_creds(),
and thus before tracehook_report_exec()->ptrace_stop().
or, if exec fails,
free_bprm() drops this mutex when bprm->cred != NULL which
indicates install_exec_creds() was not called.
Reported-by: Tom Horsley <tom.horsley@att.net>
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Acked-by: David Howells <dhowells@redhat.com>
Cc: Roland McGrath <roland@redhat.com>
Cc: James Morris <jmorris@namei.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
cancel_delayed_work() has to use del_timer_sync() to guarantee the timer
function is not running after return. But most users doesn't actually
need this, and del_timer_sync() has problems: it is not useable from
interrupt, and it depends on every lock which could be taken from irq.
Introduce __cancel_delayed_work() which calls del_timer() instead.
The immediate reason for this patch is
http://bugzilla.kernel.org/show_bug.cgi?id=13757
but hopefully this helper makes sense anyway.
As for 13757 bug, actually we need requeue_delayed_work(), but its
semantics are not yet clear.
Merge this patch early to resolves cross-tree interdependencies between
input and infiniband.
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Cc: Dmitry Torokhov <dmitry.torokhov@gmail.com>
Cc: Roland Dreier <rdreier@cisco.com>
Cc: Stefan Richter <stefanr@s5r6.in-berlin.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Device-mapper userspace logs (like the clustered log) are
identified by a universally unique identifier (UUID). This
identifier is used to associate requests from the kernel to
a specific log in userspace. The UUID must be unique everywhere,
since multiple machines may use this identifier when communicating
about a particular log, as is the case for cluster logs.
Sometimes, device-mapper/LVM may re-use a UUID. This is the
case during pvmoves, when moving from one segment of an LV
to another, or when resizing a mirror, etc. In these cases,
a new log is created with the same UUID and loaded in the
"inactive" slot. When a device-mapper "resume" is issued,
the "live" table is deactivated and the new "inactive" table
becomes "live". (The "inactive" table can also be removed
via a device-mapper 'clear' command.)
The above two issues were colliding. More than one log was being
created with the same UUID, and there was no way to distinguish
between them. So, sometimes the wrong log would be swapped
out during the exchange.
The solution is to create a locally unique identifier,
'luid', to go along with the UUID. This new identifier is used
to determine exactly which log is being referenced by the kernel
when the log exchange is made. The identifier is not
universally safe, but it does not need to be, since
create/destroy/suspend/resume operations are bound to a specific
machine; and these are the operations that make up the exchange.
Signed-off-by: Jonathan Brassow <jbrassow@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
Set sensible I/O hints for striped DM devices in the topology
infrastructure added for 2.6.31 for userspace tools to
obtain via sysfs.
Add .io_hints to 'struct target_type' to allow the I/O hints portion
(io_min and io_opt) of the 'struct queue_limits' to be set by each
target and implement this for dm-stripe.
Signed-off-by: Mike Snitzer <snitzer@redhat.com>
Signed-off-by: Alasdair G Kergon <agk@redhat.com>
These are full of unresolved problems, mainly that conversions don't
work 1-1 from hrtimers to tasklet_hrtimers because unlike hrtimers
tasklets can't be killed from softirq context.
And when a qdisc gets reset, that's exactly what we need to do here.
We'll work this out in the net-next-2.6 tree and if warranted we'll
backport that work to -stable.
This reverts the following 3 changesets:
a2cb6a4dd4
("pkt_sched: Fix bogon in tasklet_hrtimer changes.")
38acce2d79
("pkt_sched: Convert CBQ to tasklet_hrtimer.")
ee5f9757ea
("pkt_sched: Convert qdisc_watchdog to tasklet_hrtimer")
Signed-off-by: David S. Miller <davem@davemloft.net>
My previous patch (commit 4f8ee2c9cc: "lmb: Remove __init from
lmb_end_of_DRAM()") removed __init in lmb.c but missed the fact that it
was also marked as such in the .h
Signed-off-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
As struct skcipher_givcrypt_request includes struct crypto_request
at a non-zero offset, testing for NULL after converting the pointer
returned by crypto_dequeue_request does not work. This can result
in IPsec crashes when the queue is depleted.
This patch fixes it by doing the pointer conversion only when the
return value is non-NULL. In particular, we create a new function
__crypto_dequeue_request that does the pointer conversion.
Reported-by: Brad Bosch <bradbosch@comcast.net>
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au>
If the BIOS reports an invalid throttling state (which seems to be
fairly common after system boot), a reset is done to state T0.
Because of a check in acpi_processor_get_throttling_ptc(), the reset
never actually gets executed, which results in the error reoccurring
on every access of for example /proc/acpi/processor/CPU0/throttling.
Add a 'force' option to acpi_processor_set_throttling() to ensure
the reset really takes effect.
Addresses http://bugzilla.kernel.org/show_bug.cgi?id=13389
This patch, together with the next one, fixes a regression introduced in
2.6.30, listed on the regression list. They have been available for 2.5
months now in bugzilla, but have not been picked up, despite various
reminders and without any reason given.
Google shows that numerous people are hitting this issue. The issue is in
itself relatively minor, but the bug in the code is clear.
The patches have been in all my kernels and today testing has shown that
throttling works correctly with the patches applied when the system
overheats (http://bugzilla.kernel.org/show_bug.cgi?id=13918#c14).
Signed-off-by: Frans Pop <elendil@planet.nl>
Acked-by: Zhang Rui <rui.zhang@intel.com>
Cc: Len Brown <lenb@kernel.org>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
It's problematic to allow signed element_nr's or total's to be passed as
part of the flex array API.
flex_array_alloc() allows total_nr_elements to be set to a negative
quantity, which is obviously erroneous.
flex_array_get() and flex_array_put() allows negative array indices in
dereferencing an array part, which could address memory mapped before
struct flex_array.
The fix is to convert all existing element_nr formals to be qualified as
unsigned. Existing checks to compare it to total_nr_elements or the max
array size based on element_size need not be changed.
Signed-off-by: David Rientjes <rientjes@google.com>
Cc: Dave Hansen <dave@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The `parts' member of struct flex_array should evaluate to an incomplete
type so that sizeof() cannot be used and C99 does not require the
zero-length specification.
Signed-off-by: David Rientjes <rientjes@google.com>
Acked-by: Dave Hansen <dave@linux.vnet.ibm.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
2.6.30's commit 8a0bdec194 removed
user_shm_lock() calls in hugetlb_file_setup() but left the
user_shm_unlock call in shm_destroy().
In detail:
Assume that can_do_hugetlb_shm() returns true and hence user_shm_lock()
is not called in hugetlb_file_setup(). However, user_shm_unlock() is
called in any case in shm_destroy() and in the following
atomic_dec_and_lock(&up->__count) in free_uid() is executed and if
up->__count gets zero, also cleanup_user_struct() is scheduled.
Note that sched_destroy_user() is empty if CONFIG_USER_SCHED is not set.
However, the ref counter up->__count gets unexpectedly non-positive and
the corresponding structs are freed even though there are live
references to them, resulting in a kernel oops after a lots of
shmget(SHM_HUGETLB)/shmctl(IPC_RMID) cycles and CONFIG_USER_SCHED set.
Hugh changed Stefan's suggested patch: can_do_hugetlb_shm() at the
time of shm_destroy() may give a different answer from at the time
of hugetlb_file_setup(). And fixed newseg()'s no_id error path,
which has missed user_shm_unlock() ever since it came in 2.6.9.
Reported-by: Stefan Huber <shuber2@gmail.com>
Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Tested-by: Stefan Huber <shuber2@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
* git://git.kernel.org/pub/scm/linux/kernel/git/davem/net-2.6:
smc91x: let smc91x work well under netpoll
pxaficp-ir: remove incorrect net_device_ops
NET: llc, zero sockaddr_llc struct
drivers/net: fixed drivers that support netpoll use ndo_start_xmit()
netpoll: warning for ndo_start_xmit returns with interrupts enabled
net: Fix Micrel KSZ8842 Kconfig description
netfilter: xt_quota: fix wrong return value (error case)
ipv6: Fix commit 63d9950b08 (ipv6: Make v4-mapped bindings consistent with IPv4)
E100: fix interaction with swiotlb on X86.
pkt_sched: Convert CBQ to tasklet_hrtimer.
pkt_sched: Convert qdisc_watchdog to tasklet_hrtimer
rtl8187: always set MSR_LINK_ENEDCA flag with RTL8187B
ibm_newemac: emac_close() needs to call netif_carrier_off()
net: fix ks8851 build errors
net: Rename MAC platform driver for w90p910 platform
yellowfin: Fix buffer underrun after dev_alloc_skb() failure
orinoco: correct key bounds check in orinoco_hw_get_tkip_iv
mac80211: fix todo lock
vfs_read() offset is defined as loff_t, but kernel_read()
offset is only defined as unsigned long. Redefine
kernel_read() offset as loff_t.
Cc: stable@kernel.org
Signed-off-by: Mimi Zohar <zohar@us.ibm.com>
Signed-off-by: James Morris <jmorris@namei.org>
None of this stuff should execute in hw IRQ context, therefore
use a tasklet_hrtimer so that it runs in softirq context.
Signed-off-by: David S. Miller <davem@davemloft.net>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
