Allow forcing the use of sys membarrier
[urcu.git] / README.md
CommitLineData
d589a916
PP
1Userspace RCU Implementation
2============================
3
4by Mathieu Desnoyers and Paul E. McKenney
5
6
7Building
8--------
9
10 ./bootstrap # skip if using tarball
11 ./configure
12 make
13 make install
14 ldconfig
15
16Hints:
17
18 - Forcing 32-bit build:
19
20 CFLAGS="-m32 -g -O2" ./configure
21
22 - Forcing 64-bit build:
23
24 CFLAGS="-m64 -g -O2" ./configure
25
26 - Forcing a 32-bit build with 386 backward compatibility:
27
28 CFLAGS="-m32 -g -O2" ./configure --host=i386-pc-linux-gnu
29
30 - Forcing a 32-bit build for Sparcv9 (typical for Sparc v9)
31
32 CFLAGS="-m32 -Wa,-Av9a -g -O2" ./configure
33
34
35Architectures supported
36-----------------------
37
38Currently, the following architectures are supported:
39
40 - Linux x86 (i386, i486, i586, i686)
41 - x86 64-bit
42 - PowerPC 32/64
43 - S390, S390x
44 - ARM 32/64
45 - MIPS
859050b3 46 - NIOS2
d589a916
PP
47 - Alpha
48 - ia64
49 - Sparcv9 32/64
50 - Tilera
51 - hppa/PA-RISC
52
53Tested on Linux, FreeBSD 8.2/8.3/9.0/9.1/10.0 i386/amd64, and Cygwin.
54Should also work on:
55
56 - Android
57 - NetBSD 5
58 - OpenBSD
59 - Darwin
60
61(more testing needed before claiming support for these OS).
62
63Linux ARM depends on running a Linux kernel 2.6.15 or better, GCC 4.4 or
64better.
65
66The GCC compiler versions 3.3, 3.4, 4.0, 4.1, 4.2, 4.3, 4.4 and 4.5 are
67supported, with the following exceptions:
68
69 - GCC 3.3 and 3.4 have a bug that prevents them from generating volatile
70 accesses to offsets in a TLS structure on 32-bit x86. These versions are
71 therefore not compatible with `liburcu` on x86 32-bit
72 (i386, i486, i586, i686).
73 The problem has been reported to the GCC community:
74 http://www.mail-archive.com/gcc-bugs@gcc.gnu.org/msg281255.html
75 - GCC 3.3 cannot match the "xchg" instruction on 32-bit x86 build.
76 See http://kerneltrap.org/node/7507
77 - Alpha, ia64 and ARM architectures depend on GCC 4.x with atomic builtins
78 support. For ARM this was introduced with GCC 4.4:
79 http://gcc.gnu.org/gcc-4.4/changes.html.
80
81Clang version 3.0 (based on LLVM 3.0) is supported.
82
83Building on MacOS X (Darwin) requires a work-around for processor
84detection:
85
86 - 32-bit:
87
88 ./configure --build=i686-apple-darwin11
89
90 - 64-bit:
91
92 ./configure --build=x86_64-apple-darwin11
93
94For developers using the Git tree:
95
96This source tree is based on the autotools suite from GNU to simplify
97portability. Here are some things you should have on your system in order to
98compile the git repository tree :
99
100 - GNU autotools (automake >=1.10, autoconf >=2.50, autoheader >=2.50)
101 (make sure your system wide `automake` points to a recent version!)
102 - GNU Libtool >=2.2
103 (for more information, go to http://www.gnu.org/software/autoconf/)
104
105If you get the tree from the repository, you will need to use the `bootstrap`
106script in the root of the tree. It calls all the GNU tools needed to prepare
107the tree configuration.
108
109Test scripts provided in the `tests/` directory of the source tree depend
110on `bash` and the `seq` program.
111
112
113API
114---
115
116See the relevant API documentation files in `doc/`. The APIs provided by
117Userspace RCU are, by prefix:
118
dcb9c05a 119 - `rcu_`: Read-Copy Update (see [`doc/rcu-api.md`](doc/rcu-api.md))
d589a916
PP
120 - `cmm_`: Concurrent Memory Model
121 - `caa_`: Concurrent Architecture Abstraction
122 - `cds_`: Concurrent Data Structures
dcb9c05a 123 (see [`doc/cds-api.md`](doc/cds-api.md))
d589a916 124 - `uatomic_`: Userspace Atomic
dcb9c05a 125 (see [`doc/uatomic-api.md`](doc/uatomic-api.md))
d589a916
PP
126
127
128Quick start guide
129-----------------
130
131### Usage of all urcu libraries:
132
133 - Define `_LGPL_SOURCE` (only) if your code is LGPL or GPL compatible
134 before including the `urcu.h` or `urcu-qsbr.h` header. If your application
135 is distributed under another license, function calls will be generated
136 instead of inlines, so your application can link with the library.
137 - Linking with one of the libraries below is always necessary even for
138 LGPL and GPL applications.
139 - Define `URCU_INLINE_SMALL_FUNCTIONS` before including Userspace RCU
140 headers if you want Userspace RCU to inline small functions (10
141 lines or less) into the application. It can be used by applications
142 distributed under any kind of license, and does *not* make the
143 application a derived work of Userspace RCU.
144
145Those small inlined functions are guaranteed to match the library
146content as long as the library major version is unchanged.
147Therefore, the application *must* be compiled with headers matching
148the library major version number. Applications using
149`URCU_INLINE_SMALL_FUNCTIONS` may be unable to use debugging
150features of Userspace RCU without being recompiled.
151
152
153### Usage of `liburcu`
154
155 1. `#include <urcu.h>`
156 2. Link the application with `-lurcu`
157
158This is the preferred version of the library, in terms of
159grace-period detection speed, read-side speed and flexibility.
160Dynamically detects kernel support for `sys_membarrier()`. Falls back
161on `urcu-mb` scheme if support is not present, which has slower
d8d9a340
MD
162read-side. Use the --disable-sys-membarrier-fallback configure option
163to disable the fall back, thus requiring `sys_membarrier()` to be
164available. This gives a small speedup when `sys_membarrier()` is
165supported by the kernel, and aborts in the library constructor if not
166supported.
d589a916
PP
167
168
169### Usage of `liburcu-qsbr`
170
171 1. `#include <urcu-qsbr.h>`
172 2. Link with `-lurcu-qsbr`
173
174The QSBR flavor of RCU needs to have each reader thread executing
175`rcu_quiescent_state()` periodically to progress. `rcu_thread_online()`
176and `rcu_thread_offline()` can be used to mark long periods for which
177the threads are not active. It provides the fastest read-side at the
178expense of more intrusiveness in the application code.
179
180
181### Usage of `liburcu-mb`
182
183 1. `#include <urcu.h>`
184 2. Compile any `_LGPL_SOURCE` code using this library with `-DRCU_MB`
185 3. Link with `-lurcu-mb`
186
187This version of the urcu library uses memory barriers on the writer
188and reader sides. This results in faster grace-period detection, but
189results in slower reads.
190
191
192### Usage of `liburcu-signal`
193
194 1. `#include <urcu.h>`
195 2. Compile any `_LGPL_SOURCE` code using this library with `-DRCU_SIGNAL`
196 3. Link the application with `-lurcu-signal`
197
198Version of the library that requires a signal, typically `SIGUSR1`. Can
199be overridden with `-DSIGRCU` by modifying `Makefile.build.inc`.
200
201
202### Usage of `liburcu-bp`
203
204 1. `#include <urcu-bp.h>`
205 2. Link with `-lurcu-bp`
206
207The BP library flavor stands for "bulletproof". It is specifically
208designed to help tracing library to hook on applications without
209requiring to modify these applications. `rcu_init()`,
210`rcu_register_thread()` and `rcu_unregister_thread()` all become nops.
211The state is dealt with by the library internally at the expense of
212read-side and write-side performance.
213
214
215### Initialization
216
217Each thread that has reader critical sections (that uses
218`rcu_read_lock()`/`rcu_read_unlock()` must first register to the URCU
219library. This is done by calling `rcu_register_thread()`. Unregistration
220must be performed before exiting the thread by using
221`rcu_unregister_thread()`.
222
223
224### Reading
225
226Reader critical sections must be protected by locating them between
227calls to `rcu_read_lock()` and `rcu_read_unlock()`. Inside that lock,
228`rcu_dereference()` may be called to read an RCU protected pointer.
229
230
231### Writing
232
233`rcu_assign_pointer()` and `rcu_xchg_pointer()` may be called anywhere.
234After, `synchronize_rcu()` must be called. When it returns, the old
235values are not in usage anymore.
236
237
238### Usage of `liburcu-defer`
239
240 - Follow instructions for either `liburcu`, `liburcu-qsbr`,
241 `liburcu-mb`, `liburcu-signal`, or `liburcu-bp` above.
242 The `liburcu-defer` functionality is pulled into each of
243 those library modules.
244 - Provides `defer_rcu()` primitive to enqueue delayed callbacks. Queued
245 callbacks are executed in batch periodically after a grace period.
246 Do _not_ use `defer_rcu()` within a read-side critical section, because
247 it may call `synchronize_rcu()` if the thread queue is full.
248 This can lead to deadlock or worse.
249 - Requires that `rcu_defer_barrier()` must be called in library destructor
250 if a library queues callbacks and is expected to be unloaded with
251 `dlclose()`.
252
253Its API is currently experimental. It may change in future library releases.
254
255
256### Usage of `urcu-call-rcu`
257
258 - Follow instructions for either `liburcu`, `liburcu-qsbr`,
259 `liburcu-mb`, `liburcu-signal`, or `liburcu-bp` above.
260 The `urcu-call-rcu` functionality is pulled into each of
261 those library modules.
262 - Provides the `call_rcu()` primitive to enqueue delayed callbacks
263 in a manner similar to `defer_rcu()`, but without ever delaying
264 for a grace period. On the other hand, `call_rcu()`'s best-case
265 overhead is not quite as good as that of `defer_rcu()`.
266 - Provides `call_rcu()` to allow asynchronous handling of RCU
267 grace periods. A number of additional functions are provided
268 to manage the helper threads used by `call_rcu()`, but reasonable
269 defaults are used if these additional functions are not invoked.
dcb9c05a 270 See [`doc/rcu-api.md`](doc/rcu-api.md) in userspace-rcu documentation
d589a916
PP
271 for more details.
272
273
274### Being careful with signals
275
276The `liburcu` library uses signals internally. The signal handler is
277registered with the `SA_RESTART` flag. However, these signals may cause
278some non-restartable system calls to fail with `errno = EINTR`. Care
279should be taken to restart system calls manually if they fail with this
280error. A list of non-restartable system calls may be found in
281`signal(7)`. The `liburcu-mb` and `liburcu-qsbr` versions of the Userspace RCU
282library do not require any signal.
283
284Read-side critical sections are allowed in a signal handler,
285except those setup with `sigaltstack(2)`, with `liburcu` and
286`liburcu-mb`. Be careful, however, to disable these signals
287between thread creation and calls to `rcu_register_thread()`, because a
288signal handler nesting on an unregistered thread would not be
289allowed to call `rcu_read_lock()`.
290
291Read-side critical sections are _not_ allowed in a signal handler with
292`liburcu-qsbr`, unless signals are disabled explicitly around each
293`rcu_quiescent_state()` calls, when threads are put offline and around
294calls to `synchronize_rcu()`. Even then, we do not recommend it.
295
296
297### Interaction with mutexes
298
299One must be careful to do not cause deadlocks due to interaction of
300`synchronize_rcu()` and RCU read-side with mutexes. If `synchronize_rcu()`
301is called with a mutex held, this mutex (or any mutex which has this
302mutex in its dependency chain) should not be acquired from within a RCU
303read-side critical section.
304
305This is especially important to understand in the context of the
306QSBR flavor: a registered reader thread being "online" by
307default should be considered as within a RCU read-side critical
308section unless explicitly put "offline". Therefore, if
309`synchronize_rcu()` is called with a mutex held, this mutex, as
310well as any mutex which has this mutex in its dependency chain
311should only be taken when the RCU reader thread is "offline"
312(this can be performed by calling `rcu_thread_offline()`).
313
314
315### Interaction with `fork()`
316
317Special care must be taken for applications performing `fork()` without
318any following `exec()`. This is caused by the fact that Linux only clones
319the thread calling `fork()`, and thus never replicates any of the other
320parent thread into the child process. Most `liburcu` implementations
321require that all registrations (as reader, `defer_rcu` and `call_rcu`
322threads) should be released before a `fork()` is performed, except for the
323rather common scenario where `fork()` is immediately followed by `exec()` in
324the child process. The only implementation not subject to that rule is
325`liburcu-bp`, which is designed to handle `fork()` by calling
326`rcu_bp_before_fork`, `rcu_bp_after_fork_parent` and
327`rcu_bp_after_fork_child`.
328
329Applications that use `call_rcu()` and that `fork()` without
330doing an immediate `exec()` must take special action. The parent
331must invoke `call_rcu_before_fork()` before the `fork()` and
332`call_rcu_after_fork_parent()` after the `fork()`. The child
333process must invoke `call_rcu_after_fork_child()`.
334Even though these three APIs are suitable for passing to
335`pthread_atfork()`, use of `pthread_atfork()` is **STRONGLY
336DISCOURAGED** for programs calling the glibc memory allocator
337(`malloc()`, `calloc()`, `free()`, ...) within `call_rcu` callbacks.
338This is due to limitations in the way glibc memory allocator
339handles calls to the memory allocator from concurrent threads
340while the `pthread_atfork()` handlers are executing.
341
342Combining e.g.:
343
344 - call to `free()` from callbacks executed within `call_rcu` worker
345 threads,
346 - executing `call_rcu` atfork handlers within the glibc pthread
347 atfork mechanism,
348
349will sometimes trigger interesting process hangs. This usually
350hangs on a memory allocator lock within glibc.
351
352
353### Thread Local Storage (TLS)
354
355Userspace RCU can fall back on `pthread_getspecific()` to emulate
356TLS variables on systems where it is not available. This behavior
357can be forced by specifying `--disable-compiler-tls` as configure
358argument.
359
360
361### Usage of `DEBUG_RCU`
362
363`DEBUG_RCU` is used to add internal debugging self-checks to the
364RCU library. This define adds a performance penalty when enabled.
365Can be enabled by uncommenting the corresponding line in
366`Makefile.build.inc`.
367
368
369### Usage of `DEBUG_YIELD`
370
371`DEBUG_YIELD` is used to add random delays in the code for testing
372purposes.
373
374
375### SMP support
376
377By default the library is configured to use synchronization primitives
378adequate for SMP systems. On uniprocessor systems, support for SMP
379systems can be disabled with:
380
381 ./configure --disable-smp-support
382
383theoretically yielding slightly better performance.
384
385
386Make targets
387------------
388
389In addition to the usual `make check` target, Userspace RCU features
390`make regtest` and `make bench` targets:
391
392 - `make check`: short tests, meant to be run when rebuilding or
393 porting Userspace RCU.
394 - `make regtest`: long (many hours) test, meant to be run when
395 modifying Userspace RCU or porting it to a new architecture or
396 operating system.
397 - `make bench`: long (many hours) benchmarks.
398
399
400Contacts
401--------
402
403You can contact the maintainers on the following mailing list:
dcb9c05a 404`lttng-dev@lists.lttng.org`.
This page took 0.05892 seconds and 4 git commands to generate.