futex(4) -- Linux man page
NAMEfutex - Fast Userspace Locking
The Linux kernel provides futexes ('Fast Userspace muTexes') as a building block for fast userspace locking and semaphores. Futexes are very basic and lend themselves well for building higher level locking abstractions such as POSIX mutexes.
This page does not set out to document all design decisions but restricts itself to issues relevant for application and library development. Most programmers will in fact not be using futexes directly but instead rely on system libraries built on them, such as the NPTL pthreads implementation.
A futex is identified by a piece of memory which can be shared between different processes. In these different processes, it need not have identical addresses. In its bare form, a futex has semaphore semantics; it is a counter that can be incremented and decremented atomically; processes can wait for the value to become positive.
Futex operation is entirely userspace for the non-contended case. The kernel is only involved to arbitrate the contended case. As any sane design will strive for non-contension, futexes are also optimised for this situation.
In its bare form, a futex is an aligned integer which is only touched by atomic assembler instructions. Processes can share this integer over mmap, via shared segments or because they share memory space, in which case the application is commonly called multithreaded.
Any futex operation starts in userspace, but it may necessary to communicate with the kernel using the futex(2) system call.
To 'up' a futex, execute the proper assembler instructions that will cause the host CPU to atomically increment the integer. Afterwards, check if it has in fact changed from 0 to 1, in which case there were no waiters and the operation is done. This is the non-contended case which is fast and should be common.
In the contended case, the atomic increment changed the counter from -1 (or some other negative number). If this is detected, there are waiters. Userspace should now set the counter to 1 and instruct the kernel to wake up any waiters using the FUTEX_WAKE operation.
Waiting on a futex, to 'down' it, is the reverse operation. Atomically decrement the counter and check if it changed to 0, in which case the operation is done and the futex was uncontended. In all other circumstances, the process should set the counter to -1 and request that the kernel wait for another process to up the futex. This is done using the FUTEX_WAIT operation.
The futex system call can optionally be passed a timeout specifying how long the kernel should wait for the futex to be upped. In this case, semantics are more complex and the programmer is referred to futex(2) for more details. The same holds for asynchronous futex waiting.
To reiterate, bare futexes are not intended as an easy to use abstraction for end-users. Implementors are expected to be assembly literate and to have read the sources of the futex userspace library referenced below.
This man page illustrates the most common use of the futex(2) primitives: it is by no means the only one.
Futexes were designed and worked on by Hubertus Franke (IBM Thomas J. Watson Research Center), Matthew Kirkwood, Ingo Molnar (Red Hat) and Rusty Russell (IBM Linux Technology Center). This page written by bert hubert.
futex(2), `Fuss, Futexes and Furwocks: Fast Userlevel Locking in Linux' (proceedings of the Ottawa Linux Symposium 2002), futex example library, futex-*.tar.bz2 <URL:ftp://ftp.kernel.org:/pub/linux/kernel/people/rusty/>.