linux-stable.git/include/linux/lglock.h, branch master

locking/lglock: Remove lglock implementation

2016-09-22T13:25:56+00:00

It is now unused, remove it before someone else thinks its a good idea
to use this.

Signed-off-by: Peter Zijlstra (Intel) 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Thomas Gleixner 
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Ingo Molnar

stop_machine: Remove stop_cpus_lock and lg_double_lock/unlock()

2016-09-22T13:25:55+00:00

stop_two_cpus() and stop_cpus() use stop_cpus_lock to avoid the deadlock,
we need to ensure that the stopper functions can't be queued "backwards"
from one another. This doesn't look nice; if we use lglock then we do not
really need stopper->lock, cpu_stop_queue_work() could use lg_local_lock()
under local_irq_save().

OTOH it would be even better to avoid lglock in stop_machine.c and remove
lg_double_lock(). This patch adds "bool stop_cpus_in_progress" set/cleared
by queue_stop_cpus_work(), and changes cpu_stop_queue_two_works() to busy
wait until it is cleared.

queue_stop_cpus_work() sets stop_cpus_in_progress = T lockless, but after
it queues a work on CPU1 it must be visible to stop_two_cpus(CPU1, CPU2)
which checks it under the same lock. And since stop_two_cpus() holds the
2nd lock too, queue_stop_cpus_work() can not clear stop_cpus_in_progress
if it is also going to queue a work on CPU2, it needs to take that 2nd
lock to do this.

Signed-off-by: Oleg Nesterov 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Linus Torvalds 
Cc: Peter Zijlstra 
Cc: Rik van Riel 
Cc: Tejun Heo 
Cc: Thomas Gleixner 
Link: http://lkml.kernel.org/r/20151121181148.GA433@redhat.com
Signed-off-by: Ingo Molnar

sched/stop_machine: Fix deadlock between multiple stop_two_cpus()

2015-06-19T08:03:12+00:00

Jiri reported a machine stuck in multi_cpu_stop() with
migrate_swap_stop() as function and with the following src,dst cpu
pairs: {11,  4} {13, 11} { 4, 13}

                        4       11      13

cpuM: queue(4 ,13)
                        *Ma
cpuN: queue(13,11)
                                *N      Na
                        *M              Mb
cpuO: queue(11, 4)
                        *O      Oa
                                *Nb
                        *Ob

Where *X denotes the cpu running the queueing of cpu-X and X[ab] denotes
the first/second queued work.

You'll observe the top of the workqueue for each cpu: 4,11,13 to be work
from cpus: M, O, N resp. IOW. deadlock.

Do away with the queueing trickery and introduce lg_double_lock() to
lock both CPUs and fully serialize the stop_two_cpus() callers instead
of the partial (and buggy) serialization we have now.

Reported-by: Jiri Olsa 
Signed-off-by: Peter Zijlstra (Intel) 
Cc: Andrew Morton 
Cc: Borislav Petkov 
Cc: H. Peter Anvin 
Cc: Linus Torvalds 
Cc: Oleg Nesterov 
Cc: Peter Zijlstra 
Cc: Rik van Riel 
Cc: Thomas Gleixner 
Link: http://lkml.kernel.org/r/20150605153023.GH19282@twins.programming.kicks-ass.net
Signed-off-by: Ingo Molnar

lglock: map to spinlock when !CONFIG_SMP

2014-04-07T23:36:14+00:00

When the system has only one CPU, lglock is effectively a spinlock; map
it directly to spinlock to eliminate the indirection and duplicate code.

In addition to removing overhead, this drops 1.6k of code with a
defconfig modified to have !CONFIG_SMP, and 1.1k with a minimal config.

Signed-off-by: Josh Triplett 
Cc: Rusty Russell 
Cc: Michal Marek 
Cc: Thomas Gleixner 
Cc: David Howells 
Cc: "H. Peter Anvin" 
Cc: Nick Piggin 
Cc: Peter Zijlstra 
Signed-off-by: Andrew Morton 
Signed-off-by: Linus Torvalds

no need to keep brlock macros anymore...

2013-11-09T05:16:21+00:00

Signed-off-by: Al Viro

lglock: add DEFINE_STATIC_LGLOCK()

2012-10-10T05:15:44+00:00

When the lglock doesn't need to be exported we can use
DEFINE_STATIC_LGLOCK().

Signed-off-by: Lai Jiangshan 
Cc: Alexander Viro 
Cc: Rusty Russell 
Cc: Andi Kleen 
Signed-off-by: Andrew Morton 
Signed-off-by: Al Viro

lglock: make the per_cpu locks static

2012-10-10T05:15:44+00:00

The per_cpu locks are not used outside the file which contains the
DEFINE_LGLOCK(), so we can make these symbols static.

Signed-off-by: Lai Jiangshan 
Cc: Alexander Viro 
Cc: Rusty Russell 
Cc: Andi Kleen 
Signed-off-by: Andrew Morton 
Signed-off-by: Al Viro

lglock: remove unused DEFINE_LGLOCK_LOCKDEP()

2012-10-10T05:15:44+00:00

struct lglocks use their own lock_key/lock_dep_map which are defined in
struct lglock.  DEFINE_LGLOCK_LOCKDEP() is unused, so remove it and save a
small piece of memory.

Signed-off-by: Lai Jiangshan 
Cc: Alexander Viro 
Cc: Rusty Russell 
Cc: Andi Kleen 
Signed-off-by: Andrew Morton 
Signed-off-by: Al Viro

brlocks/lglocks: turn into functions

2012-05-30T03:28:41+00:00

lglocks and brlocks are currently generated with some complicated macros
in lglock.h.  But there's no reason to not just use common utility
functions and put all the data into a common data structure.

Since there are at least two users it makes sense to share this code in a
library.  This is also easier maintainable than a macro forest.

This will also make it later possible to dynamically allocate lglocks and
also use them in modules (this would both still need some additional, but
now straightforward, code)

[akpm@linux-foundation.org: checkpatch fixes]
Signed-off-by: Andi Kleen 
Cc: Al Viro 
Cc: Rusty Russell 
Signed-off-by: Andrew Morton 
Signed-off-by: Rusty Russell 
Signed-off-by: Al Viro

lglock: remove online variants of lock

2012-05-30T03:28:41+00:00

Optimizing the slow paths adds a lot of complexity.  If you need to
grab every lock often, you have other problems.

Signed-off-by: Rusty Russell 
Acked-by: Nick Piggin 
Signed-off-by: Al Viro