README.md

   1 Userspace RCU Implementation
   2 ============================
   3
   4 by Mathieu Desnoyers and Paul E. McKenney
   5
   6
   7 Building
   8 --------
   9
  10     ./bootstrap # skip if using tarball
  11     ./configure
  12     make
  13     make install
  14     ldconfig
  15
  16 Hints:
  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
  35 Architectures supported
  36 -----------------------
  37
  38 Currently, the following architectures are supported:
  39
  40   - x86 (i386, i486, i586, i686)
  41   - amd64 / x86_64
  42   - PowerPC 32/64
  43   - S390, S390x
  44   - ARM 32/64
  45   - MIPS
  46   - NIOS2
  47   - Alpha
  48   - ia64
  49   - Sparcv9 32/64
  50   - Tilera
  51   - hppa/PA-RISC
  52   - m68k
  53   - RISC-V
  54
  55 Tested on:
  56
  57   - Linux all architectures
  58   - FreeBSD 8.2/8.3/9.0/9.1/10.0 i386/amd64
  59   - Solaris 10/11 i386
  60   - Cygwin i386/amd64
  61   - MacOSX amd64
  62
  63 Should also work on:
  64
  65   - Android
  66   - NetBSD 5
  67   - OpenBSD
  68   - Darwin
  69
  70 (more testing needed before claiming support for these OS).
  71
  72 Linux ARM depends on running a Linux kernel 2.6.15 or better, GCC 4.4 or
  73 better.
  74
  75 The GCC compiler versions 3.3, 3.4, 4.0, 4.1, 4.2, 4.3, 4.4 and 4.5 are
  76 supported, with the following exceptions:
  77
  78   - GCC 3.3 and 3.4 have a bug that prevents them from generating volatile
  79     accesses to offsets in a TLS structure on 32-bit x86. These versions are
  80     therefore not compatible with `liburcu` on x86 32-bit
  81     (i386, i486, i586, i686).
  82     The problem has been reported to the GCC community:
  83     http://www.mail-archive.com/gcc-bugs@gcc.gnu.org/msg281255.html
  84   - GCC 3.3 cannot match the "xchg" instruction on 32-bit x86 build.
  85     See http://kerneltrap.org/node/7507
  86   - Alpha, ia64 and ARM architectures depend on GCC 4.x with atomic builtins
  87     support. For ARM this was introduced with GCC 4.4:
  88     http://gcc.gnu.org/gcc-4.4/changes.html.
  89
  90 Clang version 3.0 (based on LLVM 3.0) is supported.
  91
  92 Building on MacOS X (Darwin) requires a work-around for processor
  93 detection:
  94
  95   - 32-bit:
  96
  97         ./configure --build=i686-apple-darwin11
  98
  99   - 64-bit:
 100
 101         ./configure --build=x86_64-apple-darwin11
 102
 103 For developers using the Git tree:
 104
 105 This source tree is based on the autotools suite from GNU to simplify
 106 portability. Here are some things you should have on your system in order to
 107 compile the git repository tree :
 108
 109   - GNU autotools (automake >=1.10, autoconf >=2.50, autoheader >=2.50)
 110     (make sure your system wide `automake` points to a recent version!)
 111   - GNU Libtool >=2.2
 112     (for more information, go to http://www.gnu.org/software/autoconf/)
 113
 114 If you get the tree from the repository, you will need to use the `bootstrap`
 115 script in the root of the tree. It calls all the GNU tools needed to prepare
 116 the tree configuration.
 117
 118 Test scripts provided in the `tests/` directory of the source tree depend
 119 on `bash` and the `seq` program.
 120
 121
 122 API
 123 ---
 124
 125 See the relevant API documentation files in `doc/`. The APIs provided by
 126 Userspace RCU are, by prefix:
 127
 128   - `rcu_`: Read-Copy Update (see [`doc/rcu-api.md`](doc/rcu-api.md))
 129   - `cmm_`: Concurrent Memory Model
 130   - `caa_`: Concurrent Architecture Abstraction
 131   - `cds_`: Concurrent Data Structures
 132     (see [`doc/cds-api.md`](doc/cds-api.md))
 133   - `uatomic_`: Userspace Atomic
 134     (see [`doc/uatomic-api.md`](doc/uatomic-api.md))
 135
 136
 137 Quick start guide
 138 -----------------
 139
 140 ### Usage of all urcu libraries:
 141
 142   - Define `_LGPL_SOURCE` (only) if your code is LGPL or GPL compatible
 143     before including the `urcu.h` or `urcu-qsbr.h` header. If your application
 144     is distributed under another license, function calls will be generated
 145     instead of inlines, so your application can link with the library.
 146   - Linking with one of the libraries below is always necessary even for
 147     LGPL and GPL applications.
 148   - Define `URCU_INLINE_SMALL_FUNCTIONS` before including Userspace RCU
 149     headers if you want Userspace RCU to inline small functions (10
 150     lines or less) into the application. It can be used by applications
 151     distributed under any kind of license, and does *not* make the
 152     application a derived work of Userspace RCU.
 153
 154 Those small inlined functions are guaranteed to match the library
 155 content as long as the library major version is unchanged.
 156 Therefore, the application *must* be compiled with headers matching
 157 the library major version number. Applications using
 158 `URCU_INLINE_SMALL_FUNCTIONS` may be unable to use debugging
 159 features of Userspace RCU without being recompiled.
 160
 161 There are multiple flavors of liburcu available:
 162
 163   - `memb`,
 164   - `qsbr`,
 165   - `mb`,
 166   - `signal`,
 167   - `bp`.
 168
 169 The API members start with the prefix "urcu_<flavor>_", where
 170 <flavor> is the chosen flavor name.
 171
 172
 173 ### Usage of `liburcu-memb`
 174
 175   1. `#include <urcu/urcu-memb.h>`
 176   2. Link the application with `-lurcu-memb`
 177
 178 This is the preferred version of the library, in terms of
 179 grace-period detection speed, read-side speed and flexibility.
 180 Dynamically detects kernel support for `sys_membarrier()`. Falls back
 181 on `urcu-mb` scheme if support is not present, which has slower
 182 read-side. Use the --disable-sys-membarrier-fallback configure option
 183 to disable the fall back, thus requiring `sys_membarrier()` to be
 184 available. This gives a small speedup when `sys_membarrier()` is
 185 supported by the kernel, and aborts in the library constructor if not
 186 supported.
 187
 188
 189 ### Usage of `liburcu-qsbr`
 190
 191   1. `#include <urcu/urcu-qsbr.h>`
 192   2. Link with `-lurcu-qsbr`
 193
 194 The QSBR flavor of RCU needs to have each reader thread executing
 195 `rcu_quiescent_state()` periodically to progress. `rcu_thread_online()`
 196 and `rcu_thread_offline()` can be used to mark long periods for which
 197 the threads are not active. It provides the fastest read-side at the
 198 expense of more intrusiveness in the application code.
 199
 200
 201 ### Usage of `liburcu-mb`
 202
 203   1. `#include <urcu/urcu-mb.h>`
 204   2. Link with `-lurcu-mb`
 205
 206 This version of the urcu library uses memory barriers on the writer
 207 and reader sides. This results in faster grace-period detection, but
 208 results in slower reads.
 209
 210
 211 ### Usage of `liburcu-signal`
 212
 213   1. `#include <urcu/urcu-signal.h>`
 214   2. Link the application with `-lurcu-signal`
 215
 216 Version of the library that requires a signal, typically `SIGUSR1`. Can
 217 be overridden with `-DSIGRCU` by modifying `Makefile.build.inc`.
 218
 219
 220 ### Usage of `liburcu-bp`
 221
 222   1. `#include <urcu/urcu-bp.h>`
 223   2. Link with `-lurcu-bp`
 224
 225 The BP library flavor stands for "bulletproof". It is specifically
 226 designed to help tracing library to hook on applications without
 227 requiring to modify these applications. `urcu_bp_init()`,
 228 `urcu_bp_register_thread()` and `urcu_bp_unregister_thread()` all become
 229 nops. The state is dealt with by the library internally at the expense
 230 of read-side and write-side performance.
 231
 232
 233 ### Initialization
 234
 235 Each thread that has reader critical sections (that uses
 236 `urcu_<flavor>_read_lock()`/`urcu_<flavor>_read_unlock()` must first
 237 register to the URCU library. This is done by calling
 238 `urcu_<flavor>_register_thread()`. Unregistration must be performed
 239 before exiting the thread by using `urcu_<flavor>_unregister_thread()`.
 240
 241
 242 ### Reading
 243
 244 Reader critical sections must be protected by locating them between
 245 calls to `urcu_<flavor>_read_lock()` and `urcu_<flavor>_read_unlock()`.
 246 Inside that lock, `rcu_dereference()` may be called to read an RCU
 247 protected pointer.
 248
 249
 250 ### Writing
 251
 252 `rcu_assign_pointer()` and `rcu_xchg_pointer()` may be called anywhere.
 253 After, `urcu_<flavor>_synchronize_rcu()` must be called. When it
 254 returns, the old values are not in usage anymore.
 255
 256
 257 ### Usage of `liburcu-defer`
 258
 259   - Follow instructions for either `liburcu-memb`, `liburcu-qsbr`,
 260     `liburcu-mb`, `liburcu-signal`, or `liburcu-bp` above.
 261     The `liburcu-defer` functionality is pulled into each of
 262     those library modules.
 263   - Provides `urcu_<flavor>_defer_rcu()` primitive to enqueue delayed
 264     callbacks. Queued callbacks are executed in batch periodically after
 265     a grace period.  Do _not_ use `urcu_<flavor>_defer_rcu()` within a
 266     read-side critical section, because it may call
 267     `urcu_<flavor>_synchronize_rcu()` if the thread queue is full.  This
 268     can lead to deadlock or worse.
 269   - Requires that `urcu_<flavor>_defer_barrier()` must be called in
 270     library destructor if a library queues callbacks and is expected to
 271     be unloaded with `dlclose()`.
 272
 273 Its API is currently experimental. It may change in future library releases.
 274
 275
 276 ### Usage of `urcu-call-rcu`
 277
 278   - Follow instructions for either `liburcu-memb`, `liburcu-qsbr`,
 279     `liburcu-mb`, `liburcu-signal`, or `liburcu-bp` above.
 280     The `urcu-call-rcu` functionality is pulled into each of
 281     those library modules.
 282   - Provides the `urcu_<flavor>_call_rcu()` primitive to enqueue delayed
 283     callbacks in a manner similar to `urcu_<flavor>_defer_rcu()`, but
 284     without ever delaying for a grace period.  On the other hand,
 285     `urcu_<flavor>_call_rcu()`'s best-case overhead is not quite as good
 286     as that of `urcu_<flavor>_defer_rcu()`.
 287   - Provides `urcu_<flavor>_call_rcu()` to allow asynchronous handling
 288     of RCU grace periods.  A number of additional functions are provided
 289     to manage the helper threads used by `urcu_<flavor>_call_rcu()`, but
 290     reasonable defaults are used if these additional functions are not
 291     invoked. See [`doc/rcu-api.md`](doc/rcu-api.md) in userspace-rcu
 292     documentation for more details.
 293
 294
 295 ### Being careful with signals
 296
 297 The `liburcu-signal` library uses signals internally. The signal handler is
 298 registered with the `SA_RESTART` flag. However, these signals may cause
 299 some non-restartable system calls to fail with `errno = EINTR`. Care
 300 should be taken to restart system calls manually if they fail with this
 301 error. A list of non-restartable system calls may be found in
 302 `signal(7)`.
 303
 304 Read-side critical sections are allowed in a signal handler,
 305 except those setup with `sigaltstack(2)`, with `liburcu-memb` and
 306 `liburcu-mb`. Be careful, however, to disable these signals
 307 between thread creation and calls to `urcu_<flavor>_register_thread()`,
 308 because a signal handler nesting on an unregistered thread would not be
 309 allowed to call `urcu_<flavor>_read_lock()`.
 310
 311 Read-side critical sections are _not_ allowed in a signal handler with
 312 `liburcu-qsbr`, unless signals are disabled explicitly around each
 313 `urcu_qsbr_quiescent_state()` calls, when threads are put offline and around
 314 calls to `urcu_qsbr_synchronize_rcu()`. Even then, we do not recommend it.
 315
 316
 317 ### Interaction with mutexes
 318
 319 One must be careful to do not cause deadlocks due to interaction of
 320 `urcu_<flavor>_synchronize_rcu()` and RCU read-side with mutexes. If
 321 `urcu_<flavor>_synchronize_rcu()` is called with a mutex held, this
 322 mutex (or any mutex which has this mutex in its dependency chain) should
 323 not be acquired from within a RCU read-side critical section.
 324
 325 This is especially important to understand in the context of the
 326 QSBR flavor: a registered reader thread being "online" by
 327 default should be considered as within a RCU read-side critical
 328 section unless explicitly put "offline". Therefore, if
 329 `urcu_qsbr_synchronize_rcu()` is called with a mutex held, this mutex,
 330 as well as any mutex which has this mutex in its dependency chain should
 331 only be taken when the RCU reader thread is "offline" (this can be
 332 performed by calling `urcu_qsbr_thread_offline()`).
 333
 334
 335 ### Interaction with `fork()`
 336
 337 Special care must be taken for applications performing `fork()` without
 338 any following `exec()`. This is caused by the fact that Linux only clones
 339 the thread calling `fork()`, and thus never replicates any of the other
 340 parent thread into the child process. Most `liburcu` implementations
 341 require that all registrations (as reader, `defer_rcu` and `call_rcu`
 342 threads) should be released before a `fork()` is performed, except for the
 343 rather common scenario where `fork()` is immediately followed by `exec()` in
 344 the child process. The only implementation not subject to that rule is
 345 `liburcu-bp`, which is designed to handle `fork()` by calling
 346 `urcu_bp_before_fork`, `urcu_bp_after_fork_parent` and
 347 `urcu_bp_after_fork_child`.
 348
 349 Applications that use `urcu_<flavor>_call_rcu()` and that `fork()`
 350 without doing an immediate `exec()` must take special action.  The
 351 parent must invoke `urcu_<flavor>_call_rcu_before_fork()` before the
 352 `fork()` and `urcu_<flavor>_call_rcu_after_fork_parent()` after the
 353 `fork()`. The child process must invoke
 354 `urcu_<flavor>_call_rcu_after_fork_child()`.  Even though these three
 355 APIs are suitable for passing to `pthread_atfork()`, use of
 356 `pthread_atfork()` is **STRONGLY DISCOURAGED** for programs calling the
 357 glibc memory allocator (`malloc()`, `calloc()`, `free()`, ...) within
 358 `urcu_<flavor>_call_rcu` callbacks.  This is due to limitations in the
 359 way glibc memory allocator handles calls to the memory allocator from
 360 concurrent threads while the `pthread_atfork()` handlers are executing.
 361
 362 Combining e.g.:
 363
 364   - call to `free()` from callbacks executed within
 365     `urcu_<flavor>_call_rcu` worker threads,
 366   - executing `urcu_<flavor>_call_rcu` atfork handlers within the glibc
 367     pthread atfork mechanism,
 368
 369 will sometimes trigger interesting process hangs. This usually
 370 hangs on a memory allocator lock within glibc.
 371
 372
 373 ### Thread Local Storage (TLS)
 374
 375 Userspace RCU can fall back on `pthread_getspecific()` to emulate
 376 TLS variables on systems where it is not available. This behavior
 377 can be forced by specifying `--disable-compiler-tls` as configure
 378 argument.
 379
 380
 381 ### Usage of `DEBUG_RCU` & `--enable-rcu-debug`
 382
 383 By default the library is configured with internal debugging
 384 self-checks disabled.
 385
 386 For always-on debugging self-checks:
 387         ./configure --enable-rcu-debug
 388
 389 For fine grained enabling of debugging self-checks, build
 390 userspace-rcu with DEBUG_RCU defined and compile dependent
 391 applications with DEBUG_RCU defined when necessary.
 392
 393 Warning: Enabling this feature result in a performance penalty.
 394
 395
 396 ### Usage of `DEBUG_YIELD`
 397
 398 `DEBUG_YIELD` is used to add random delays in the code for testing
 399 purposes.
 400
 401
 402 ### SMP support
 403
 404 By default the library is configured to use synchronization primitives
 405 adequate for SMP systems. On uniprocessor systems, support for SMP
 406 systems can be disabled with:
 407
 408     ./configure --disable-smp-support
 409
 410 theoretically yielding slightly better performance.
 411
 412
 413 ### Usage of `--enable-cds-lfht-iter-debug`
 414
 415 By default the library is configured with extra debugging checks for
 416 lock-free hash table iterator traversal disabled.
 417
 418 Building liburcu with --enable-cds-lfht-iter-debug and rebuilding
 419 application to match the ABI change allows finding cases where the hash
 420 table iterator is re-purposed to be used on a different hash table while
 421 still being used to iterate on a hash table.
 422
 423 This option alters the rculfhash ABI. Make sure to compile both library
 424 and application with matching configuration.
 425
 426
 427 Make targets
 428 ------------
 429
 430 In addition to the usual `make check` target, Userspace RCU features
 431 `make regtest` and `make bench` targets:
 432
 433   - `make check`: short tests, meant to be run when rebuilding or
 434     porting Userspace RCU.
 435   - `make regtest`: long (many hours) test, meant to be run when
 436     modifying Userspace RCU or porting it to a new architecture or
 437     operating system.
 438   - `make bench`: long (many hours) benchmarks.
 439
 440
 441 Contacts
 442 --------
 443
 444 You can contact the maintainers on the following mailing list:
 445 `lttng-dev@lists.lttng.org`.