Pull locking updates from Ingo Molnar:
"The main changes in this cycle were:
- Implement wraparound-safe refcount_t and kref_t types based on
generic atomic primitives (Peter Zijlstra)
- Improve and fix the ww_mutex code (Nicolai Hähnle)
- Add self-tests to the ww_mutex code (Chris Wilson)
- Optimize percpu-rwsems with the 'rcuwait' mechanism (Davidlohr
Bueso)
- Micro-optimize the current-task logic all around the core kernel
(Davidlohr Bueso)
- Tidy up after recent optimizations: remove stale code and APIs,
clean up the code (Waiman Long)
- ... plus misc fixes, updates and cleanups"
* 'locking-core-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: (50 commits)
fork: Fix task_struct alignment
locking/spinlock/debug: Remove spinlock lockup detection code
lockdep: Fix incorrect condition to print bug msgs for MAX_LOCKDEP_CHAIN_HLOCKS
lkdtm: Convert to refcount_t testing
kref: Implement 'struct kref' using refcount_t
refcount_t: Introduce a special purpose refcount type
sched/wake_q: Clarify queue reinit comment
sched/wait, rcuwait: Fix typo in comment
locking/mutex: Fix lockdep_assert_held() fail
locking/rtmutex: Flip unlikely() branch to likely() in __rt_mutex_slowlock()
locking/rwsem: Reinit wake_q after use
locking/rwsem: Remove unnecessary atomic_long_t casts
jump_labels: Move header guard #endif down where it belongs
locking/atomic, kref: Implement kref_put_lock()
locking/ww_mutex: Turn off __must_check for now
locking/atomic, kref: Avoid more abuse
locking/atomic, kref: Use kref_get_unless_zero() more
locking/atomic, kref: Kill kref_sub()
locking/atomic, kref: Add kref_read()
locking/atomic, kref: Add KREF_INIT()
...
Add regular waiters in stamp order. Keep adding waiters that have no
context in FIFO order and take care not to starve them.
While adding our task as a waiter, back off if we detect that there is
a waiter with a lower stamp in front of us.
Make sure to call lock_contended even when we back off early.
For w/w mutexes, being first in the wait list is only stable when
taking the lock without a context. Therefore, the purpose of the first
flag is split into two: 'first' remains to indicate whether we want to
spin optimistically, while 'handoff' indicates that we should be
prepared to accept a handoff.
For w/w locking with a context, we always accept handoffs after the
first schedule(), to handle the following sequence of events:
1. Task #0 unlocks and hands off to Task #2 which is first in line
2. Task #1 adds itself in front of Task #2
3. Task #2 wakes up and must accept the handoff even though it is no
longer first in line
Signed-off-by: Nicolai Hähnle <nicolai.haehnle@amd.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: =?UTF-8?q?Nicolai=20H=C3=A4hnle?= <Nicolai.Haehnle@amd.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel@ffwll.ch>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Maarten Lankhorst <dev@mblankhorst.nl>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dri-devel@lists.freedesktop.org
Link: http://lkml.kernel.org/r/1482346000-9927-7-git-send-email-nhaehnle@gmail.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
While reviewing the ww_mutex patches, I noticed that it was still
possible to (incorrectly) succeed for (incorrect) code like:
mutex_lock(&a);
mutex_lock(&a);
This was possible if the second mutex_lock() would block (as expected)
but then receive a spurious wakeup. At that point it would find itself
at the front of the queue, request a handoff and instantly claim
ownership and continue, since owner would point to itself.
Avoid this scenario and simplify the code by introducing a third low
bit to signal handoff pickup. So once we request handoff, unlock
clears the handoff bit and sets the pickup bit along with the new
owner.
This also removes the need for the .handoff argument to
__mutex_trylock(), since that becomes superfluous with PICKUP.
In order to guarantee enough low bits, ensure task_struct alignment is
at least L1_CACHE_BYTES (which seems a good ideal regardless).
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
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>
Fixes: 9d659ae14b ("locking/mutex: Add lock handoff to avoid starvation")
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The current mutex implementation has an atomic lock word and a
non-atomic owner field.
This disparity leads to a number of issues with the current mutex code
as it means that we can have a locked mutex without an explicit owner
(because the owner field has not been set, or already cleared).
This leads to a number of weird corner cases, esp. between the
optimistic spinning and debug code. Where the optimistic spinning
code needs the owner field updated inside the lock region, the debug
code is more relaxed because the whole lock is serialized by the
wait_lock.
Also, the spinning code itself has a few corner cases where we need to
deal with a held lock without an owner field.
Furthermore, it becomes even more of a problem when trying to fix
starvation cases in the current code. We end up stacking special case
on special case.
To solve this rework the basic mutex implementation to be a single
atomic word that contains the owner and uses the low bits for extra
state.
This matches how PI futexes and rt_mutex already work. By having the
owner an integral part of the lock state a lot of the problems
dissapear and we get a better option to deal with starvation cases,
direct owner handoff.
Changing the basic mutex does however invalidate all the arch specific
mutex code; this patch leaves that unused in-place, a later patch will
remove that.
Tested-by: Jason Low <jason.low2@hpe.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Will Deacon <will.deacon@arm.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
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>
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
We will need the MCS lock code for doing optimistic spinning for rwsem
and queued rwlock. Extracting the MCS code from mutex.c and put into
its own file allow us to reuse this code easily.
We also inline mcs_spin_lock and mcs_spin_unlock functions
for better efficiency.
Note that using the smp_load_acquire/smp_store_release pair used in
mcs_lock and mcs_unlock is not sufficient to form a full memory barrier
across cpus for many architectures (except x86). For applications that
absolutely need a full barrier across multiple cpus with mcs_unlock and
mcs_lock pair, smp_mb__after_unlock_lock() should be used after mcs_lock.
Reviewed-by: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Signed-off-by: Tim Chen <tim.c.chen@linux.intel.com>
Signed-off-by: Davidlohr Bueso <davidlohr@hp.com>
Signed-off-by: Peter Zijlstra <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/1390347360.3138.63.camel@schen9-DESK
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Linus suggested to replace
#ifndef CONFIG_HAVE_ARCH_MUTEX_CPU_RELAX
#define arch_mutex_cpu_relax() cpu_relax()
#endif
with just a simple
#ifndef arch_mutex_cpu_relax
# define arch_mutex_cpu_relax() cpu_relax()
#endif
to get rid of CONFIG_HAVE_CPU_RELAX_SIMPLE. So architectures can
simply define arch_mutex_cpu_relax if they want an architecture
specific function instead of having to add a select statement in
their Kconfig in addition.
Suggested-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: Heiko Carstens <heiko.carstens@de.ibm.com>
Injects EDEADLK conditions at pseudo-random interval, with
exponential backoff up to UINT_MAX (to ensure that every lock
operation still completes in a reasonable time).
This way we can test the wound slowpath even for ww mutex users
where contention is never expected, and the ww deadlock
avoidance algorithm is only needed for correctness against
malicious userspace. An example would be protecting kernel
modesetting properties, which thanks to single-threaded X isn't
really expected to contend, ever.
I've looked into using the CONFIG_FAULT_INJECTION
infrastructure, but decided against it for two reasons:
- EDEADLK handling is mandatory for ww mutex users and should
never affect the outcome of a syscall. This is in contrast to -ENOMEM
injection. So fine configurability isn't required.
- The fault injection framework only allows to set a simple
probability for failure. Now the probability that a ww mutex acquire
stage with N locks will never complete (due to too many injected
EDEADLK backoffs) is zero. But the expected number of ww_mutex_lock
operations for the completely uncontended case would be O(exp(N)).
The per-acuiqire ctx exponential backoff solution choosen here only
results in O(log N) overhead due to injection and so O(log N * N)
lock operations. This way we can fail with high probability (and so
have good test coverage even for fancy backoff and lock acquisition
paths) without running into patalogical cases.
Note that EDEADLK will only ever be injected when we managed to
acquire the lock. This prevents any behaviour changes for users
which rely on the EALREADY semantics.
Signed-off-by: Daniel Vetter <daniel.vetter@ffwll.ch>
Signed-off-by: Maarten Lankhorst <maarten.lankhorst@canonical.com>
Acked-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: dri-devel@lists.freedesktop.org
Cc: linaro-mm-sig@lists.linaro.org
Cc: rostedt@goodmis.org
Cc: daniel@ffwll.ch
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: http://lkml.kernel.org/r/20130620113117.4001.21681.stgit@patser
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The current mutex spinning code (with MUTEX_SPIN_ON_OWNER option
turned on) allow multiple tasks to spin on a single mutex
concurrently. A potential problem with the current approach is
that when the mutex becomes available, all the spinning tasks
will try to acquire the mutex more or less simultaneously. As a
result, there will be a lot of cacheline bouncing especially on
systems with a large number of CPUs.
This patch tries to reduce this kind of contention by putting
the mutex spinners into a queue so that only the first one in
the queue will try to acquire the mutex. This will reduce
contention and allow all the tasks to move forward faster.
The queuing of mutex spinners is done using an MCS lock based
implementation which will further reduce contention on the mutex
cacheline than a similar ticket spinlock based implementation.
This patch will add a new field into the mutex data structure
for holding the MCS lock. This expands the mutex size by 8 bytes
for 64-bit system and 4 bytes for 32-bit system. This overhead
will be avoid if the MUTEX_SPIN_ON_OWNER option is turned off.
The following table shows the jobs per minute (JPM) scalability
data on an 8-node 80-core Westmere box with a 3.7.10 kernel. The
numactl command is used to restrict the running of the fserver
workloads to 1/2/4/8 nodes with hyperthreading off.
+-----------------+-----------+-----------+-------------+----------+
| Configuration | Mean JPM | Mean JPM | Mean JPM | % Change |
| | w/o patch | patch 1 | patches 1&2 | 1->1&2 |
+-----------------+------------------------------------------------+
| | User Range 1100 - 2000 |
+-----------------+------------------------------------------------+
| 8 nodes, HT off | 227972 | 227237 | 305043 | +34.2% |
| 4 nodes, HT off | 393503 | 381558 | 394650 | +3.4% |
| 2 nodes, HT off | 334957 | 325240 | 338853 | +4.2% |
| 1 node , HT off | 198141 | 197972 | 198075 | +0.1% |
+-----------------+------------------------------------------------+
| | User Range 200 - 1000 |
+-----------------+------------------------------------------------+
| 8 nodes, HT off | 282325 | 312870 | 332185 | +6.2% |
| 4 nodes, HT off | 390698 | 378279 | 393419 | +4.0% |
| 2 nodes, HT off | 336986 | 326543 | 340260 | +4.2% |
| 1 node , HT off | 197588 | 197622 | 197582 | 0.0% |
+-----------------+-----------+-----------+-------------+----------+
At low user range 10-100, the JPM differences were within +/-1%.
So they are not that interesting.
The fserver workload uses mutex spinning extensively. With just
the mutex change in the first patch, there is no noticeable
change in performance. Rather, there is a slight drop in
performance. This mutex spinning patch more than recovers the
lost performance and show a significant increase of +30% at high
user load with the full 8 nodes. Similar improvements were also
seen in a 3.8 kernel.
The table below shows the %time spent by different kernel
functions as reported by perf when running the fserver workload
at 1500 users with all 8 nodes.
+-----------------------+-----------+---------+-------------+
| Function | % time | % time | % time |
| | w/o patch | patch 1 | patches 1&2 |
+-----------------------+-----------+---------+-------------+
| __read_lock_failed | 34.96% | 34.91% | 29.14% |
| __write_lock_failed | 10.14% | 10.68% | 7.51% |
| mutex_spin_on_owner | 3.62% | 3.42% | 2.33% |
| mspin_lock | N/A | N/A | 9.90% |
| __mutex_lock_slowpath | 1.46% | 0.81% | 0.14% |
| _raw_spin_lock | 2.25% | 2.50% | 1.10% |
+-----------------------+-----------+---------+-------------+
The fserver workload for an 8-node system is dominated by the
contention in the read/write lock. Mutex contention also plays a
role. With the first patch only, mutex contention is down (as
shown by the __mutex_lock_slowpath figure) which help a little
bit. We saw only a few percents improvement with that.
By applying patch 2 as well, the single mutex_spin_on_owner
figure is now split out into an additional mspin_lock figure.
The time increases from 3.42% to 11.23%. It shows a great
reduction in contention among the spinners leading to a 30%
improvement. The time ratio 9.9/2.33=4.3 indicates that there
are on average 4+ spinners waiting in the spin_lock loop for
each spinner in the mutex_spin_on_owner loop. Contention in
other locking functions also go down by quite a lot.
The table below shows the performance change of both patches 1 &
2 over patch 1 alone in other AIM7 workloads (at 8 nodes,
hyperthreading off).
+--------------+---------------+----------------+-----------------+
| Workload | mean % change | mean % change | mean % change |
| | 10-100 users | 200-1000 users | 1100-2000 users |
+--------------+---------------+----------------+-----------------+
| alltests | 0.0% | -0.8% | +0.6% |
| five_sec | -0.3% | +0.8% | +0.8% |
| high_systime | +0.4% | +2.4% | +2.1% |
| new_fserver | +0.1% | +14.1% | +34.2% |
| shared | -0.5% | -0.3% | -0.4% |
| short | -1.7% | -9.8% | -8.3% |
+--------------+---------------+----------------+-----------------+
The short workload is the only one that shows a decline in
performance probably due to the spinner locking and queuing
overhead.
Signed-off-by: Waiman Long <Waiman.Long@hp.com>
Reviewed-by: Davidlohr Bueso <davidlohr.bueso@hp.com>
Acked-by: Rik van Riel <riel@redhat.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Chandramouleeswaran Aswin <aswin@hp.com>
Cc: Norton Scott J <scott.norton@hp.com>
Cc: Paul E. McKenney <paulmck@linux.vnet.ibm.com>
Cc: David Howells <dhowells@redhat.com>
Cc: Dave Jones <davej@redhat.com>
Cc: Clark Williams <williams@redhat.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Link: http://lkml.kernel.org/r/1366226594-5506-4-git-send-email-Waiman.Long@hp.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The non-debug variant of mutex_destroy is a no-op, currently
implemented as a macro which does nothing. This approach fails
to check the type of the parameter, so an error would only show
when debugging gets enabled. Using an inline function instead,
offers type checking for earlier bug catching.
Signed-off-by: Jean Delvare <khali@linux-fr.org>
Cc: Peter Zijlstra <a.p.zijlstra@chello.nl>
Link: http://lkml.kernel.org/r/20110716174200.41002352@endymion.delvare
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Linus Torvalds
Ingo Molnar
Tejun Heo
Nicolai Hähnle
Peter Zijlstra
Adrien Schildknecht
Davidlohr Bueso
Jason Low
Tim Chen
Heiko Carstens
Maarten Lankhorst
Daniel Vetter
Waiman Long
Arun Sharma
Jean Delvare
Peter Zijlstra