166 lines · plain
1===============2Locking lessons3===============4 5Lesson 1: Spin locks6====================7 8The most basic primitive for locking is spinlock::9 10 static DEFINE_SPINLOCK(xxx_lock);11 12 unsigned long flags;13 14 spin_lock_irqsave(&xxx_lock, flags);15 ... critical section here ..16 spin_unlock_irqrestore(&xxx_lock, flags);17 18The above is always safe. It will disable interrupts _locally_, but the19spinlock itself will guarantee the global lock, so it will guarantee that20there is only one thread-of-control within the region(s) protected by that21lock. This works well even under UP also, so the code does _not_ need to22worry about UP vs SMP issues: the spinlocks work correctly under both.23 24 NOTE! Implications of spin_locks for memory are further described in:25 26 Documentation/memory-barriers.txt27 28 (5) ACQUIRE operations.29 30 (6) RELEASE operations.31 32The above is usually pretty simple (you usually need and want only one33spinlock for most things - using more than one spinlock can make things a34lot more complex and even slower and is usually worth it only for35sequences that you **know** need to be split up: avoid it at all cost if you36aren't sure).37 38This is really the only really hard part about spinlocks: once you start39using spinlocks they tend to expand to areas you might not have noticed40before, because you have to make sure the spinlocks correctly protect the41shared data structures **everywhere** they are used. The spinlocks are most42easily added to places that are completely independent of other code (for43example, internal driver data structures that nobody else ever touches).44 45 NOTE! The spin-lock is safe only when you **also** use the lock itself46 to do locking across CPU's, which implies that EVERYTHING that47 touches a shared variable has to agree about the spinlock they want48 to use.49 50----51 52Lesson 2: reader-writer spinlocks.53==================================54 55If your data accesses have a very natural pattern where you usually tend56to mostly read from the shared variables, the reader-writer locks57(rw_lock) versions of the spinlocks are sometimes useful. They allow multiple58readers to be in the same critical region at once, but if somebody wants59to change the variables it has to get an exclusive write lock.60 61 NOTE! reader-writer locks require more atomic memory operations than62 simple spinlocks. Unless the reader critical section is long, you63 are better off just using spinlocks.64 65The routines look the same as above::66 67 rwlock_t xxx_lock = __RW_LOCK_UNLOCKED(xxx_lock);68 69 unsigned long flags;70 71 read_lock_irqsave(&xxx_lock, flags);72 .. critical section that only reads the info ...73 read_unlock_irqrestore(&xxx_lock, flags);74 75 write_lock_irqsave(&xxx_lock, flags);76 .. read and write exclusive access to the info ...77 write_unlock_irqrestore(&xxx_lock, flags);78 79The above kind of lock may be useful for complex data structures like80linked lists, especially searching for entries without changing the list81itself. The read lock allows many concurrent readers. Anything that82**changes** the list will have to get the write lock.83 84 NOTE! RCU is better for list traversal, but requires careful85 attention to design detail (see Documentation/RCU/listRCU.rst).86 87Also, you cannot "upgrade" a read-lock to a write-lock, so if you at _any_88time need to do any changes (even if you don't do it every time), you have89to get the write-lock at the very beginning.90 91 NOTE! We are working hard to remove reader-writer spinlocks in most92 cases, so please don't add a new one without consensus. (Instead, see93 Documentation/RCU/rcu.rst for complete information.)94 95----96 97Lesson 3: spinlocks revisited.98==============================99 100The single spin-lock primitives above are by no means the only ones. They101are the most safe ones, and the ones that work under all circumstances,102but partly **because** they are safe they are also fairly slow. They are slower103than they'd need to be, because they do have to disable interrupts104(which is just a single instruction on a x86, but it's an expensive one -105and on other architectures it can be worse).106 107If you have a case where you have to protect a data structure across108several CPU's and you want to use spinlocks you can potentially use109cheaper versions of the spinlocks. IFF you know that the spinlocks are110never used in interrupt handlers, you can use the non-irq versions::111 112 spin_lock(&lock);113 ...114 spin_unlock(&lock);115 116(and the equivalent read-write versions too, of course). The spinlock will117guarantee the same kind of exclusive access, and it will be much faster.118This is useful if you know that the data in question is only ever119manipulated from a "process context", ie no interrupts involved.120 121The reasons you mustn't use these versions if you have interrupts that122play with the spinlock is that you can get deadlocks::123 124 spin_lock(&lock);125 ...126 <- interrupt comes in:127 spin_lock(&lock);128 129where an interrupt tries to lock an already locked variable. This is ok if130the other interrupt happens on another CPU, but it is _not_ ok if the131interrupt happens on the same CPU that already holds the lock, because the132lock will obviously never be released (because the interrupt is waiting133for the lock, and the lock-holder is interrupted by the interrupt and will134not continue until the interrupt has been processed).135 136(This is also the reason why the irq-versions of the spinlocks only need137to disable the _local_ interrupts - it's ok to use spinlocks in interrupts138on other CPU's, because an interrupt on another CPU doesn't interrupt the139CPU that holds the lock, so the lock-holder can continue and eventually140releases the lock).141 142 Linus143 144----145 146Reference information:147======================148 149For dynamic initialization, use spin_lock_init() or rwlock_init() as150appropriate::151 152 spinlock_t xxx_lock;153 rwlock_t xxx_rw_lock;154 155 static int __init xxx_init(void)156 {157 spin_lock_init(&xxx_lock);158 rwlock_init(&xxx_rw_lock);159 ...160 }161 162 module_init(xxx_init);163 164For static initialization, use DEFINE_SPINLOCK() / DEFINE_RWLOCK() or165__SPIN_LOCK_UNLOCKED() / __RW_LOCK_UNLOCKED() as appropriate.166