2019-04-10 14:32:41 +03:00
===============
Locking lessons
===============
2009-12-13 23:12:46 +03:00
Lesson 1: Spin locks
2019-04-10 14:32:41 +03:00
====================
2005-04-17 02:20:36 +04:00
2019-04-10 14:32:41 +03:00
The most basic primitive for locking is spinlock::
2005-04-17 02:20:36 +04:00
2019-04-10 14:32:41 +03:00
static DEFINE_SPINLOCK(xxx_lock);
2005-04-17 02:20:36 +04:00
unsigned long flags;
spin_lock_irqsave(&xxx_lock, flags);
... critical section here ..
spin_unlock_irqrestore(&xxx_lock, flags);
2009-12-13 23:12:46 +03:00
The above is always safe. It will disable interrupts _locally_, but the
2005-04-17 02:20:36 +04:00
spinlock itself will guarantee the global lock, so it will guarantee that
there is only one thread-of-control within the region(s) protected by that
2011-07-11 22:04:58 +04:00
lock. This works well even under UP also, so the code does _not_ need to
worry about UP vs SMP issues: the spinlocks work correctly under both.
2009-12-13 23:12:46 +03:00
NOTE! Implications of spin_locks for memory are further described in:
2005-04-17 02:20:36 +04:00
2009-12-13 23:12:46 +03:00
Documentation/memory-barriers.txt
2019-04-10 14:32:41 +03:00
2020-01-31 23:52:33 +03:00
(5) ACQUIRE operations.
2019-04-10 14:32:41 +03:00
2020-01-31 23:52:33 +03:00
(6) RELEASE operations.
2005-04-17 02:20:36 +04:00
The above is usually pretty simple (you usually need and want only one
spinlock for most things - using more than one spinlock can make things a
lot more complex and even slower and is usually worth it only for
2019-04-10 14:32:41 +03:00
sequences that you **know** need to be split up: avoid it at all cost if you
2011-07-11 22:04:58 +04:00
aren't sure).
2005-04-17 02:20:36 +04:00
This is really the only really hard part about spinlocks: once you start
using spinlocks they tend to expand to areas you might not have noticed
before, because you have to make sure the spinlocks correctly protect the
2019-04-10 14:32:41 +03:00
shared data structures **everywhere** they are used. The spinlocks are most
2009-12-13 23:12:46 +03:00
easily added to places that are completely independent of other code (for
example, internal driver data structures that nobody else ever touches).
2019-04-10 14:32:41 +03:00
NOTE! The spin-lock is safe only when you **also** use the lock itself
2009-12-13 23:12:46 +03:00
to do locking across CPU's, which implies that EVERYTHING that
touches a shared variable has to agree about the spinlock they want
to use.
2005-04-17 02:20:36 +04:00
----
Lesson 2: reader-writer spinlocks.
2019-04-10 14:32:41 +03:00
==================================
2005-04-17 02:20:36 +04:00
If your data accesses have a very natural pattern where you usually tend
to mostly read from the shared variables, the reader-writer locks
2009-12-13 23:12:46 +03:00
(rw_lock) versions of the spinlocks are sometimes useful. They allow multiple
2005-04-17 02:20:36 +04:00
readers to be in the same critical region at once, but if somebody wants
2009-12-13 23:12:46 +03:00
to change the variables it has to get an exclusive write lock.
2005-04-17 02:20:36 +04:00
2009-12-13 23:12:46 +03:00
NOTE! reader-writer locks require more atomic memory operations than
simple spinlocks. Unless the reader critical section is long, you
are better off just using spinlocks.
2005-04-17 02:20:36 +04:00
2019-04-10 14:32:41 +03:00
The routines look the same as above::
2009-12-13 23:12:46 +03:00
2011-01-23 17:30:09 +03:00
rwlock_t xxx_lock = __RW_LOCK_UNLOCKED(xxx_lock);
2005-04-17 02:20:36 +04:00
unsigned long flags;
read_lock_irqsave(&xxx_lock, flags);
.. critical section that only reads the info ...
read_unlock_irqrestore(&xxx_lock, flags);
write_lock_irqsave(&xxx_lock, flags);
.. read and write exclusive access to the info ...
write_unlock_irqrestore(&xxx_lock, flags);
2009-12-13 23:12:46 +03:00
The above kind of lock may be useful for complex data structures like
linked lists, especially searching for entries without changing the list
itself. The read lock allows many concurrent readers. Anything that
2019-04-10 14:32:41 +03:00
**changes** the list will have to get the write lock.
2009-12-13 23:12:46 +03:00
NOTE! RCU is better for list traversal, but requires careful
2019-07-15 11:31:06 +03:00
attention to design detail (see Documentation/RCU/listRCU.rst).
2005-04-17 02:20:36 +04:00
2009-12-13 23:12:46 +03:00
Also, you cannot "upgrade" a read-lock to a write-lock, so if you at _any_
2005-04-17 02:20:36 +04:00
time need to do any changes (even if you don't do it every time), you have
2009-12-13 23:12:46 +03:00
to get the write-lock at the very beginning.
NOTE! We are working hard to remove reader-writer spinlocks in most
cases, so please don't add a new one without consensus. (Instead, see
2019-07-15 11:31:06 +03:00
Documentation/RCU/rcu.rst for complete information.)
2005-04-17 02:20:36 +04:00
----
Lesson 3: spinlocks revisited.
2019-04-10 14:32:41 +03:00
==============================
2005-04-17 02:20:36 +04:00
The single spin-lock primitives above are by no means the only ones. They
are the most safe ones, and the ones that work under all circumstances,
2019-04-10 14:32:41 +03:00
but partly **because** they are safe they are also fairly slow. They are slower
2011-07-11 22:04:58 +04:00
than they'd need to be, because they do have to disable interrupts
(which is just a single instruction on a x86, but it's an expensive one -
and on other architectures it can be worse).
2005-04-17 02:20:36 +04:00
If you have a case where you have to protect a data structure across
several CPU's and you want to use spinlocks you can potentially use
cheaper versions of the spinlocks. IFF you know that the spinlocks are
2019-04-10 14:32:41 +03:00
never used in interrupt handlers, you can use the non-irq versions::
2005-04-17 02:20:36 +04:00
spin_lock(&lock);
...
spin_unlock(&lock);
(and the equivalent read-write versions too, of course). The spinlock will
2014-07-31 00:41:55 +04:00
guarantee the same kind of exclusive access, and it will be much faster.
2005-04-17 02:20:36 +04:00
This is useful if you know that the data in question is only ever
2014-07-31 00:41:55 +04:00
manipulated from a "process context", ie no interrupts involved.
2005-04-17 02:20:36 +04:00
The reasons you mustn't use these versions if you have interrupts that
2019-04-10 14:32:41 +03:00
play with the spinlock is that you can get deadlocks::
2005-04-17 02:20:36 +04:00
spin_lock(&lock);
...
<- interrupt comes in:
spin_lock(&lock);
where an interrupt tries to lock an already locked variable. This is ok if
the other interrupt happens on another CPU, but it is _not_ ok if the
interrupt happens on the same CPU that already holds the lock, because the
lock will obviously never be released (because the interrupt is waiting
for the lock, and the lock-holder is interrupted by the interrupt and will
2014-07-31 00:41:55 +04:00
not continue until the interrupt has been processed).
2005-04-17 02:20:36 +04:00
(This is also the reason why the irq-versions of the spinlocks only need
to disable the _local_ interrupts - it's ok to use spinlocks in interrupts
on other CPU's, because an interrupt on another CPU doesn't interrupt the
CPU that holds the lock, so the lock-holder can continue and eventually
2014-07-31 00:41:55 +04:00
releases the lock).
2005-04-17 02:20:36 +04:00
Linus
2009-12-13 23:12:46 +03:00
----
Reference information:
2019-04-10 14:32:41 +03:00
======================
2009-12-13 23:12:46 +03:00
For dynamic initialization, use spin_lock_init() or rwlock_init() as
2019-04-10 14:32:41 +03:00
appropriate::
2009-12-13 23:12:46 +03:00
spinlock_t xxx_lock;
rwlock_t xxx_rw_lock;
static int __init xxx_init(void)
{
spin_lock_init(&xxx_lock);
rwlock_init(&xxx_rw_lock);
...
}
module_init(xxx_init);
For static initialization, use DEFINE_SPINLOCK() / DEFINE_RWLOCK() or
__SPIN_LOCK_UNLOCKED() / __RW_LOCK_UNLOCKED() as appropriate.