Lesson 1: Spin locks¶

The most basic primitive for locking is spinlock:

static DEFINE_SPINLOCK(xxx_lock);      unsigned long flags;      spin_lock_irqsave(&xxx_lock, flags);      ... critical section here ..      spin_unlock_irqrestore(&xxx_lock, flags);

The above is always safe. It will disable interrupts _locally_, but thespinlock itself will guarantee the global lock, so it will guarantee thatthere is only one thread-of-control within the region(s) protected by thatlock. This works well even under UP also, so the code does _not_ need toworry about UP vs SMP issues: the spinlocks work correctly under both.

NOTE! Implications of spin_locks for memory are further described in:

Documentation/memory-barriers.txt

5. ACQUIRE operations.

6. RELEASE operations.

The above is usually pretty simple (you usually need and want only onespinlock for most things - using more than one spinlock can make things alot more complex and even slower and is usually worth it only forsequences that youknow need to be split up: avoid it at all cost if youaren’t sure).

This is really the only really hard part about spinlocks: once you startusing spinlocks they tend to expand to areas you might not have noticedbefore, because you have to make sure the spinlocks correctly protect theshared data structureseverywhere they are used. The spinlocks are mosteasily added to places that are completely independent of other code (forexample, internal driver data structures that nobody else ever touches).

NOTE! The spin-lock is safe only when youalso use the lock itselfto do locking across CPU’s, which implies that EVERYTHING thattouches a shared variable has to agree about the spinlock they wantto use.

Lesson 2: reader-writer spinlocks.¶

If your data accesses have a very natural pattern where you usually tendto mostly read from the shared variables, the reader-writer locks(rw_lock) versions of the spinlocks are sometimes useful. They allow multiplereaders to be in the same critical region at once, but if somebody wantsto change the variables it has to get an exclusive write lock.

NOTE! reader-writer locks require more atomic memory operations thansimple spinlocks. Unless the reader critical section is long, youare better off just using spinlocks.

The routines look the same as above:

rwlock_t xxx_lock = __RW_LOCK_UNLOCKED(xxx_lock);     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);

The above kind of lock may be useful for complex data structures likelinked lists, especially searching for entries without changing the listitself. The read lock allows many concurrent readers. Anything thatchanges the list will have to get the write lock.

NOTE! RCU is better for list traversal, but requires carefulattention to design detail (seeUsing RCU to Protect Read-Mostly Linked Lists).

Also, you cannot “upgrade” a read-lock to a write-lock, so if you at _any_time need to do any changes (even if you don’t do it every time), you haveto get the write-lock at the very beginning.

NOTE! We are working hard to remove reader-writer spinlocks in mostcases, so please don’t add a new one without consensus. (Instead, seeRCU Concepts for complete information.)

Lesson 3: spinlocks revisited.¶

The single spin-lock primitives above are by no means the only ones. Theyare the most safe ones, and the ones that work under all circumstances,but partlybecause they are safe they are also fairly slow. They are slowerthan 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).

If you have a case where you have to protect a data structure acrossseveral CPU’s and you want to use spinlocks you can potentially usecheaper versions of the spinlocks. IFF you know that the spinlocks arenever used in interrupt handlers, you can use the non-irq versions:

spin_lock(&lock);...spin_unlock(&lock);

(and the equivalent read-write versions too, of course). The spinlock willguarantee the same kind of exclusive access, and it will be much faster.This is useful if you know that the data in question is only evermanipulated from a “process context”, ie no interrupts involved.

The reasons you mustn’t use these versions if you have interrupts thatplay with the spinlock is that you can get deadlocks:

spin_lock(&lock);...        <- interrupt comes in:                spin_lock(&lock);

where an interrupt tries to lock an already locked variable. This is ok ifthe other interrupt happens on another CPU, but it is _not_ ok if theinterrupt happens on the same CPU that already holds the lock, because thelock will obviously never be released (because the interrupt is waitingfor the lock, and the lock-holder is interrupted by the interrupt and willnot continue until the interrupt has been processed).

(This is also the reason why the irq-versions of the spinlocks only needto disable the _local_ interrupts - it’s ok to use spinlocks in interruptson other CPU’s, because an interrupt on another CPU doesn’t interrupt theCPU that holds the lock, so the lock-holder can continue and eventuallyreleases the lock).

Linus

Reference information:¶

For dynamic initialization, usespin_lock_init() orrwlock_init() asappropriate:

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, useDEFINE_SPINLOCK() /DEFINE_RWLOCK() or__SPIN_LOCK_UNLOCKED() /__RW_LOCK_UNLOCKED() as appropriate.