ViewVC Help
View File | Revision Log | Show Annotations | Download File
/cvs/cvsroot/Async-Interrupt/Interrupt.pm
(Generate patch)

Comparing cvsroot/Async-Interrupt/Interrupt.pm (file contents):
Revision 1.12 by root, Tue Jul 14 19:51:24 2009 UTC vs.
Revision 1.19 by root, Tue Jul 28 13:17:05 2009 UTC

53 53
54For example the deliantra game server uses a variant of this technique 54For example the deliantra game server uses a variant of this technique
55to interrupt background processes regularly to send map updates to game 55to interrupt background processes regularly to send map updates to game
56clients. 56clients.
57 57
58Or L<EV::Loop::Async> uses an interrupt object to wake up perl when new
59events have arrived.
60
58L<IO::AIO> and L<BDB> could also use this to speed up result reporting. 61L<IO::AIO> and L<BDB> could also use this to speed up result reporting.
59 62
60=item Speedy event loop invocation 63=item Speedy event loop invocation
61 64
62One could use this module e.g. in L<Coro> to interrupt a running coro-thread 65One could use this module e.g. in L<Coro> to interrupt a running coro-thread
88I<running> interpreter, there is optional support for signalling a pipe 91I<running> interpreter, there is optional support for signalling a pipe
89- that means you can also wait for the pipe to become readable (e.g. via 92- that means you can also wait for the pipe to become readable (e.g. via
90L<EV> or L<AnyEvent>). This, of course, incurs the overhead of a C<read> 93L<EV> or L<AnyEvent>). This, of course, incurs the overhead of a C<read>
91and C<write> syscall. 94and C<write> syscall.
92 95
96=head1 USAGE EXAMPLES
97
98=head2 Implementing race-free signal handling
99
100This example uses a single event pipe for all signals, and one
101Async::Interrupt per signal. This code is actually what the L<AnyEvent>
102module uses itself when Async::Interrupt is available.
103
104First, create the event pipe and hook it into the event loop
105
106 $SIGPIPE = new Async::Interrupt::EventPipe;
107 $SIGPIPE_W = AnyEvent->io (
108 fh => $SIGPIPE->fileno,
109 poll => "r",
110 cb => \&_signal_check, # defined later
111 );
112
113Then, for each signal to hook, create an Async::Interrupt object. The
114callback just sets a global variable, as we are only interested in
115synchronous signals (i.e. when the event loop polls), which is why the
116pipe draining is not done automatically.
117
118 my $interrupt = new Async::Interrupt
119 cb => sub { undef $SIGNAL_RECEIVED{$signum} }
120 signal => $signum,
121 pipe => [$SIGPIPE->filenos],
122 pipe_autodrain => 0,
123 ;
124
125Finally, the I/O callback for the event pipe handles the signals:
126
127 sub _signal_check {
128 # drain the pipe first
129 $SIGPIPE->drain;
130
131 # two loops, just to be sure
132 while (%SIGNAL_RECEIVED) {
133 for (keys %SIGNAL_RECEIVED) {
134 delete $SIGNAL_RECEIVED{$_};
135 warn "signal $_ received\n";
136 }
137 }
138 }
139
140=head2 Interrupt perl from another thread
141
142This example interrupts the Perl interpreter from another thread, via the
143XS API. This is used by e.g. the L<EV::Loop::Async> module.
144
145#TODO#
146
147=head1 THE Async::Interrupt CLASS
148
93=over 4 149=over 4
94 150
95=cut 151=cut
96 152
97package Async::Interrupt; 153package Async::Interrupt;
98 154
99use common::sense; 155use common::sense;
100 156
101BEGIN { 157BEGIN {
102 # the next line forces initialisation of internal 158 # the next line forces initialisation of internal
103 # signal handling # variables 159 # signal handling variables, otherwise, PL_sig_pending
160 # etc. will be null pointers.
104 $SIG{KILL} = sub { }; 161 $SIG{KILL} = sub { };
105 162
106 our $VERSION = '0.042'; 163 our $VERSION = '1.0';
107 164
108 require XSLoader; 165 require XSLoader;
109 XSLoader::load ("Async::Interrupt", $VERSION); 166 XSLoader::load ("Async::Interrupt", $VERSION);
110} 167}
111 168
134The exceptions are C<$!> and C<$@>, which are saved and restored by 191The exceptions are C<$!> and C<$@>, which are saved and restored by
135Async::Interrupt. 192Async::Interrupt.
136 193
137If the callback should throw an exception, then it will be caught, 194If the callback should throw an exception, then it will be caught,
138and C<$Async::Interrupt::DIED> will be called with C<$@> containing 195and C<$Async::Interrupt::DIED> will be called with C<$@> containing
139the exception. The default will simply C<warn> about the message and 196the exception. The default will simply C<warn> about the message and
140continue. 197continue.
141 198
142=item c_cb => [$c_func, $c_arg] 199=item c_cb => [$c_func, $c_arg]
143 200
144Registers a C callback the be invoked whenever the async interrupt is 201Registers a C callback the be invoked whenever the async interrupt is
157might use (the exception is C<errno>, which is saved and restored by 214might use (the exception is C<errno>, which is saved and restored by
158Async::Interrupt). The callback itself runs as part of the perl context, 215Async::Interrupt). The callback itself runs as part of the perl context,
159so you can call any perl functions and modify any perl data structures (in 216so you can call any perl functions and modify any perl data structures (in
160which case the requirements set out for C<cb> apply as well). 217which case the requirements set out for C<cb> apply as well).
161 218
219=item var => $scalar_ref
220
221When specified, then the given argument must be a reference to a
222scalar. The scalar will be set to C<0> initially. Signalling the interrupt
223object will set it to the passed value, handling the interrupt will reset
224it to C<0> again.
225
226Note that the only thing you are legally allowed to do is to is to check
227the variable in a boolean or integer context (e.g. comparing it with a
228string, or printing it, will I<destroy> it and might cause your program to
229crash or worse).
230
162=item signal => $signame_or_value 231=item signal => $signame_or_value
163 232
164When this parameter is specified, then the Async::Interrupt will hook the 233When this parameter is specified, then the Async::Interrupt will hook the
165given signal, that is, it will effectively call C<< ->signal (0) >> each time 234given signal, that is, it will effectively call C<< ->signal (0) >> each time
166the given signal is caught by the process. 235the given signal is caught by the process.
167 236
168Only one async can hook a given signal, and the signal will be restored to 237Only one async can hook a given signal, and the signal will be restored to
169defaults when the Async::Interrupt object gets destroyed. 238defaults when the Async::Interrupt object gets destroyed.
239
240=item signal_hysteresis => $boolean
241
242Sets the initial signal hysteresis state, see the C<signal_hysteresis>
243method, below.
170 244
171=item pipe => [$fileno_or_fh_for_reading, $fileno_or_fh_for_writing] 245=item pipe => [$fileno_or_fh_for_reading, $fileno_or_fh_for_writing]
172 246
173Specifies two file descriptors (or file handles) that should be signalled 247Specifies two file descriptors (or file handles) that should be signalled
174whenever the async interrupt is signalled. This means a single octet will 248whenever the async interrupt is signalled. This means a single octet will
175be written to it, and before the callback is being invoked, it will be 249be written to it, and before the callback is being invoked, it will be
176read again. Due to races, it is unlikely but possible that multiple octets 250read again. Due to races, it is unlikely but possible that multiple octets
177are written. It is required that the file handles are both in nonblocking 251are written. It is required that the file handles are both in nonblocking
178mode. 252mode.
179 253
180You can get a portable pipe and set non-blocking mode portably by using
181e.g. L<AnyEvent::Util> from the L<AnyEvent> distribution.
182
183It is also possible to pass in a linux eventfd as both read and write
184handle (which is faster than a pipe).
185
186The object will keep a reference to the file handles. 254The object will keep a reference to the file handles.
187 255
188This can be used to ensure that async notifications will interrupt event 256This can be used to ensure that async notifications will interrupt event
189frameworks as well. 257frameworks as well.
190 258
259Note that C<Async::Interrupt> will create a suitable signal fd
260automatically when your program requests one, so you don't have to specify
261this argument when all you want is an extra file descriptor to watch.
262
263If you want to share a single event pipe between multiple Async::Interrupt
264objects, you can use the C<Async::Interrupt::EventPipe> class to manage
265those.
266
267=item pipe_autodrain => $boolean
268
269Sets the initial autodrain state, see the C<pipe_autodrain> method, below.
270
191=back 271=back
192 272
193=cut 273=cut
194 274
195sub new { 275sub new {
196 my ($class, %arg) = @_; 276 my ($class, %arg) = @_;
197 277
198 bless \(_alloc $arg{cb}, @{$arg{c_cb}}[0,1], @{$arg{pipe}}[0,1], $arg{signal}), $class 278 my $self = bless \(_alloc $arg{cb}, @{$arg{c_cb}}[0,1], @{$arg{pipe}}[0,1], $arg{signal}, $arg{var}), $class;
279
280 # urgs, reminds me of Event
281 for my $attr (qw(pipe_autodrain signal_hysteresis)) {
282 $self->$attr ($arg{$attr}) if exists $arg{$attr};
283 }
284
285 $self
199} 286}
200 287
201=item ($signal_func, $signal_arg) = $async->signal_func 288=item ($signal_func, $signal_arg) = $async->signal_func
202 289
203Returns the address of a function to call asynchronously. The function has 290Returns the address of a function to call asynchronously. The function
204the following prototype and needs to be passed the specified C<$c_arg>, 291has the following prototype and needs to be passed the specified
205which is a C<void *> cast to C<IV>: 292C<$signal_arg>, which is a C<void *> cast to C<IV>:
206 293
207 void (*signal_func) (void *signal_arg, int value) 294 void (*signal_func) (void *signal_arg, int value)
208 295
209An example call would look like: 296An example call would look like:
210 297
211 signal_func (signal_arg, 0); 298 signal_func (signal_arg, 0);
212 299
213The function is safe to call from within signal and thread contexts, at 300The function is safe to call from within signal and thread contexts, at
214any time. The specified C<value> is passed to both C and Perl callback. 301any time. The specified C<value> is passed to both C and Perl callback.
215 302
216C<$value> must be in the valid range for a C<sig_atomic_t> (0..127 is 303C<$value> must be in the valid range for a C<sig_atomic_t>, except C<0>
217portable). 304(1..127 is portable).
218 305
219If the function is called while the Async::Interrupt object is already 306If the function is called while the Async::Interrupt object is already
220signaled but before the callbacks are being executed, then the stored 307signaled but before the callbacks are being executed, then the stored
221C<value> is either the old or the new one. Due to the asynchronous 308C<value> is either the old or the new one. Due to the asynchronous
222nature of the code, the C<value> can even be passed to two consecutive 309nature of the code, the C<value> can even be passed to two consecutive
223invocations of the callback. 310invocations of the callback.
224 311
312=item $address = $async->c_var
313
314Returns the address (cast to IV) of an C<IV> variable. The variable is set
315to C<0> initially and gets set to the passed value whenever the object
316gets signalled, and reset to C<0> once the interrupt has been handled.
317
318Note that it is often beneficial to just call C<PERL_ASYNC_CHECK ()> to
319handle any interrupts.
320
321Example: call some XS function to store the address, then show C code
322waiting for it.
323
324 my_xs_func $async->c_var;
325
326 static IV *valuep;
327
328 void
329 my_xs_func (void *addr)
330 CODE:
331 valuep = (IV *)addr;
332
333 // code in a loop, waiting
334 while (!*valuep)
335 ; // do something
336
225=item $async->signal ($value=0) 337=item $async->signal ($value=1)
226 338
227This signals the given async object from Perl code. Semi-obviously, this 339This signals the given async object from Perl code. Semi-obviously, this
228will instantly trigger the callback invocation. 340will instantly trigger the callback invocation (it does not, as the name
341might imply, do anything with POSIX signals).
229 342
230C<$value> must be in the valid range for a C<sig_atomic_t> (0..127 is 343C<$value> must be in the valid range for a C<sig_atomic_t>, except C<0>
231portable). 344(1..127 is portable).
345
346=item $async->signal_hysteresis ($enable)
347
348Enables or disables signal hysteresis (default: disabled). If a POSIX
349signal is used as a signal source for the interrupt object, then enabling
350signal hysteresis causes Async::Interrupt to reset the signal action to
351C<SIG_IGN> in the signal handler and restore it just before handling the
352interruption.
353
354When you expect a lot of signals (e.g. when using SIGIO), then enabling
355signal hysteresis can reduce the number of handler invocations
356considerably, at the cost of two extra syscalls.
357
358Note that setting the signal to C<SIG_IGN> can have unintended side
359effects when you fork and exec other programs, as often they do nto expect
360signals to be ignored by default.
232 361
233=item $async->block 362=item $async->block
234 363
235=item $async->unblock 364=item $async->unblock
236 365
251This call C<< $async->block >> and installs a handler that is called when 380This call C<< $async->block >> and installs a handler that is called when
252the current scope is exited (via an exception, by canceling the Coro 381the current scope is exited (via an exception, by canceling the Coro
253thread, by calling last/goto etc.). 382thread, by calling last/goto etc.).
254 383
255This is the recommended (and fastest) way to implement critical sections. 384This is the recommended (and fastest) way to implement critical sections.
385
386=item ($block_func, $block_arg) = $async->scope_block_func
387
388Returns the address of a function that implements the C<scope_block>
389functionality.
390
391It has the following prototype and needs to be passed the specified
392C<$block_arg>, which is a C<void *> cast to C<IV>:
393
394 void (*block_func) (void *block_arg)
395
396An example call would look like:
397
398 block_func (block_arg);
399
400The function is safe to call only from within the toplevel of a perl XS
401function and will call C<LEAVE> and C<ENTER> (in this order!).
256 402
257=item $async->pipe_enable 403=item $async->pipe_enable
258 404
259=item $async->pipe_disable 405=item $async->pipe_disable
260 406
262enabled). Writing to a pipe is relatively expensive, so it can be disabled 408enabled). Writing to a pipe is relatively expensive, so it can be disabled
263when you know you are not waiting for it (for example, with L<EV> you 409when you know you are not waiting for it (for example, with L<EV> you
264could disable the pipe in a check watcher, and enable it in a prepare 410could disable the pipe in a check watcher, and enable it in a prepare
265watcher). 411watcher).
266 412
267Note that when C<fd_disable> is in effect, no attempt to read from the 413Note that currently, while C<pipe_disable> is in effect, no attempt to
268pipe will be done. 414read from the pipe will be done when handling events. This might change as
415soon as I realize why this is a mistake.
416
417=item $fileno = $async->pipe_fileno
418
419Returns the reading side of the signalling pipe. If no signalling pipe is
420currently attached to the object, it will dynamically create one.
421
422Note that the only valid oepration on this file descriptor is to wait
423until it is readable. The fd might belong currently to a pipe, a tcp
424socket, or an eventfd, depending on the platform, and is guaranteed to be
425C<select>able.
426
427=item $async->pipe_autodrain ($enable)
428
429Enables (C<1>) or disables (C<0>) automatic draining of the pipe (default:
430enabled). When automatic draining is enabled, then Async::Interrupt will
431automatically clear the pipe. Otherwise the user is responsible for this
432draining.
433
434This is useful when you want to share one pipe among many Async::Interrupt
435objects.
436
437=item $async->post_fork
438
439The object will not normally be usable after a fork (as the pipe fd is
440shared between processes). Calling this method after a fork in the child
441ensures that the object will work as expected again. It only needs to be
442called when the async object is used in the child.
443
444This only works when the pipe was created by Async::Interrupt.
445
446Async::Interrupt ensures that the reading file descriptor does not change
447it's value.
448
449=item $signum = Async::Interrupt::sig2num $signame_or_number
450
451=item $signame = Async::Interrupt::sig2name $signame_or_number
452
453These two convenience functions simply convert a signal name or number to
454the corresponding name or number. They are not used by this module and
455exist just because perl doesn't have a nice way to do this on its own.
456
457They will return C<undef> on illegal names or numbers.
458
459=back
460
461=head1 THE Async::Interrupt::EventPipe CLASS
462
463Pipes are the predominent utility to make asynchronous signals
464synchronous. However, pipes are hard to come by: they don't exist on the
465broken windows platform, and on GNU/Linux systems, you might want to use
466an C<eventfd> instead.
467
468This class creates selectable event pipes in a portable fashion: on
469windows, it will try to create a tcp socket pair, on GNU/Linux, it will
470try to create an eventfd and everywhere else it will try to use a normal
471pipe.
472
473=over 4
474
475=item $epipe = new Async::Interrupt::EventPipe
476
477This creates and returns an eventpipe object. This object is simply a
478blessed array reference:
479
480=item ($r_fd, $w_fd) = $epipe->filenos
481
482Returns the read-side file descriptor and the write-side file descriptor.
483
484Example: pass an eventpipe object as pipe to the Async::Interrupt
485constructor, and create an AnyEvent watcher for the read side.
486
487 my $epipe = new Async::Interrupt::EventPipe;
488 my $asy = new Async::Interrupt pipe => [$epipe->filenos];
489 my $iow = AnyEvent->io (fh => $epipe->fileno, poll => 'r', cb => sub { });
490
491=item $r_fd = $epipe->fileno
492
493Return only the reading/listening side.
494
495=item $epipe->signal
496
497Write something to the pipe, in a portable fashion.
498
499=item $epipe->drain
500
501Drain (empty) the pipe.
502
503=item $epipe->renew
504
505Recreates the pipe (useful after a fork). The reading side will not change
506it's file descriptor number, but the writing side might.
507
508=back
269 509
270=cut 510=cut
271 511
2721; 5121;
273
274=back
275 513
276=head1 EXAMPLE 514=head1 EXAMPLE
277 515
278There really should be a complete C/XS example. Bug me about it. Better 516There really should be a complete C/XS example. Bug me about it. Better
279yet, create one. 517yet, create one.
287then intercepts the interpreter handling it. This makes normal signal 525then intercepts the interpreter handling it. This makes normal signal
288handling slower (probably unmeasurably, though), but has the advantage 526handling slower (probably unmeasurably, though), but has the advantage
289of not requiring a special runops function, nor slowing down normal perl 527of not requiring a special runops function, nor slowing down normal perl
290execution a bit. 528execution a bit.
291 529
292It assumes that C<sig_atomic_t> and C<int> are both async-safe to modify 530It assumes that C<sig_atomic_t>, C<int> and C<IV> are all async-safe to
293(C<sig_atomic_> is used by this module, and perl itself uses C<int>, so we 531modify.
294can assume that this is quite portable, at least w.r.t. signals).
295 532
296=head1 AUTHOR 533=head1 AUTHOR
297 534
298 Marc Lehmann <schmorp@schmorp.de> 535 Marc Lehmann <schmorp@schmorp.de>
299 http://home.schmorp.de/ 536 http://home.schmorp.de/

Diff Legend

Removed lines
+ Added lines
< Changed lines
> Changed lines