… | |
… | |
380 | |
380 | |
381 | There is a slight catch to child watchers, however: you usually start |
381 | There is a slight catch to child watchers, however: you usually start |
382 | them *after* the child process was created, and this means the process |
382 | them *after* the child process was created, and this means the process |
383 | could have exited already (and no SIGCHLD will be sent anymore). |
383 | could have exited already (and no SIGCHLD will be sent anymore). |
384 | |
384 | |
385 | Not all event models handle this correctly (POE doesn't), but even for |
385 | Not all event models handle this correctly (neither POE nor IO::Async |
|
|
386 | do, see their AnyEvent::Impl manpages for details), but even for event |
386 | event models that *do* handle this correctly, they usually need to be |
387 | models that *do* handle this correctly, they usually need to be loaded |
387 | loaded before the process exits (i.e. before you fork in the first |
388 | before the process exits (i.e. before you fork in the first place). |
388 | place). |
389 | AnyEvent's pure perl event loop handles all cases correctly regardless |
|
|
390 | of when you start the watcher. |
389 | |
391 | |
390 | This means you cannot create a child watcher as the very first thing in |
392 | This means you cannot create a child watcher as the very first thing in |
391 | an AnyEvent program, you *have* to create at least one watcher before |
393 | an AnyEvent program, you *have* to create at least one watcher before |
392 | you "fork" the child (alternatively, you can call "AnyEvent::detect"). |
394 | you "fork" the child (alternatively, you can call "AnyEvent::detect"). |
393 | |
395 | |
394 | Example: fork a process and wait for it |
396 | Example: fork a process and wait for it |
395 | |
397 | |
396 | my $done = AnyEvent->condvar; |
398 | my $done = AnyEvent->condvar; |
397 | |
399 | |
398 | my $pid = fork or exit 5; |
400 | my $pid = fork or exit 5; |
399 | |
401 | |
400 | my $w = AnyEvent->child ( |
402 | my $w = AnyEvent->child ( |
401 | pid => $pid, |
403 | pid => $pid, |
402 | cb => sub { |
404 | cb => sub { |
403 | my ($pid, $status) = @_; |
405 | my ($pid, $status) = @_; |
404 | warn "pid $pid exited with status $status"; |
406 | warn "pid $pid exited with status $status"; |
405 | $done->send; |
407 | $done->send; |
406 | }, |
408 | }, |
407 | ); |
409 | ); |
408 | |
410 | |
409 | # do something else, then wait for process exit |
411 | # do something else, then wait for process exit |
410 | $done->recv; |
412 | $done->recv; |
411 | |
413 | |
412 | IDLE WATCHERS |
414 | IDLE WATCHERS |
413 | Sometimes there is a need to do something, but it is not so important to |
415 | Sometimes there is a need to do something, but it is not so important to |
414 | do it instantly, but only when there is nothing better to do. This |
416 | do it instantly, but only when there is nothing better to do. This |
… | |
… | |
697 | AnyEvent::Impl::Tk based on Tk, very bad choice. |
699 | AnyEvent::Impl::Tk based on Tk, very bad choice. |
698 | AnyEvent::Impl::Qt based on Qt, cannot be autoprobed (see its docs). |
700 | AnyEvent::Impl::Qt based on Qt, cannot be autoprobed (see its docs). |
699 | AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse. |
701 | AnyEvent::Impl::EventLib based on Event::Lib, leaks memory and worse. |
700 | AnyEvent::Impl::POE based on POE, not generic enough for full support. |
702 | AnyEvent::Impl::POE based on POE, not generic enough for full support. |
701 | |
703 | |
|
|
704 | # warning, support for IO::Async is only partial, as it is too broken |
|
|
705 | # and limited toe ven support the AnyEvent API. See AnyEvent::Impl::Async. |
|
|
706 | AnyEvent::Impl::IOAsync based on IO::Async, cannot be autoprobed (see its docs). |
|
|
707 | |
702 | There is no support for WxWidgets, as WxWidgets has no support for |
708 | There is no support for WxWidgets, as WxWidgets has no support for |
703 | watching file handles. However, you can use WxWidgets through the |
709 | watching file handles. However, you can use WxWidgets through the |
704 | POE Adaptor, as POE has a Wx backend that simply polls 20 times per |
710 | POE Adaptor, as POE has a Wx backend that simply polls 20 times per |
705 | second, which was considered to be too horrible to even consider for |
711 | second, which was considered to be too horrible to even consider for |
706 | AnyEvent. Likewise, other POE backends can be used by AnyEvent by |
712 | AnyEvent. Likewise, other POE backends can be used by AnyEvent by |
… | |
… | |
883 | "condvar->recv"), the Event and EV modules call "$Event/EV::DIED->()", |
889 | "condvar->recv"), the Event and EV modules call "$Event/EV::DIED->()", |
884 | Glib uses "install_exception_handler" and so on. |
890 | Glib uses "install_exception_handler" and so on. |
885 | |
891 | |
886 | ENVIRONMENT VARIABLES |
892 | ENVIRONMENT VARIABLES |
887 | The following environment variables are used by this module or its |
893 | The following environment variables are used by this module or its |
888 | submodules: |
894 | submodules. |
|
|
895 | |
|
|
896 | Note that AnyEvent will remove *all* environment variables starting with |
|
|
897 | "PERL_ANYEVENT_" from %ENV when it is loaded while taint mode is |
|
|
898 | enabled. |
889 | |
899 | |
890 | "PERL_ANYEVENT_VERBOSE" |
900 | "PERL_ANYEVENT_VERBOSE" |
891 | By default, AnyEvent will be completely silent except in fatal |
901 | By default, AnyEvent will be completely silent except in fatal |
892 | conditions. You can set this environment variable to make AnyEvent |
902 | conditions. You can set this environment variable to make AnyEvent |
893 | more talkative. |
903 | more talkative. |
… | |
… | |
902 | "PERL_ANYEVENT_STRICT" |
912 | "PERL_ANYEVENT_STRICT" |
903 | AnyEvent does not do much argument checking by default, as thorough |
913 | AnyEvent does not do much argument checking by default, as thorough |
904 | argument checking is very costly. Setting this variable to a true |
914 | argument checking is very costly. Setting this variable to a true |
905 | value will cause AnyEvent to load "AnyEvent::Strict" and then to |
915 | value will cause AnyEvent to load "AnyEvent::Strict" and then to |
906 | thoroughly check the arguments passed to most method calls. If it |
916 | thoroughly check the arguments passed to most method calls. If it |
907 | finds any problems it will croak. |
917 | finds any problems, it will croak. |
908 | |
918 | |
909 | In other words, enables "strict" mode. |
919 | In other words, enables "strict" mode. |
910 | |
920 | |
911 | Unlike "use strict", it is definitely recommended ot keep it off in |
921 | Unlike "use strict", it is definitely recommended to keep it off in |
912 | production. Keeping "PERL_ANYEVENT_STRICT=1" in your environment |
922 | production. Keeping "PERL_ANYEVENT_STRICT=1" in your environment |
913 | while developing programs can be very useful, however. |
923 | while developing programs can be very useful, however. |
914 | |
924 | |
915 | "PERL_ANYEVENT_MODEL" |
925 | "PERL_ANYEVENT_MODEL" |
916 | This can be used to specify the event model to be used by AnyEvent, |
926 | This can be used to specify the event model to be used by AnyEvent, |
… | |
… | |
1204 | EV/Any 100000 224 2.88 0.34 0.27 EV + AnyEvent watchers |
1214 | EV/Any 100000 224 2.88 0.34 0.27 EV + AnyEvent watchers |
1205 | CoroEV/Any 100000 224 2.85 0.35 0.28 coroutines + Coro::Signal |
1215 | CoroEV/Any 100000 224 2.85 0.35 0.28 coroutines + Coro::Signal |
1206 | Perl/Any 100000 452 4.13 0.73 0.95 pure perl implementation |
1216 | Perl/Any 100000 452 4.13 0.73 0.95 pure perl implementation |
1207 | Event/Event 16000 517 32.20 31.80 0.81 Event native interface |
1217 | Event/Event 16000 517 32.20 31.80 0.81 Event native interface |
1208 | Event/Any 16000 590 35.85 31.55 1.06 Event + AnyEvent watchers |
1218 | Event/Any 16000 590 35.85 31.55 1.06 Event + AnyEvent watchers |
|
|
1219 | IOAsync/Any 16000 989 38.10 32.77 11.13 via IO::Async::Loop::IO_Poll |
|
|
1220 | IOAsync/Any 16000 990 37.59 29.50 10.61 via IO::Async::Loop::Epoll |
1209 | Glib/Any 16000 1357 102.33 12.31 51.00 quadratic behaviour |
1221 | Glib/Any 16000 1357 102.33 12.31 51.00 quadratic behaviour |
1210 | Tk/Any 2000 1860 27.20 66.31 14.00 SEGV with >> 2000 watchers |
1222 | Tk/Any 2000 1860 27.20 66.31 14.00 SEGV with >> 2000 watchers |
1211 | POE/Event 2000 6328 109.99 751.67 14.02 via POE::Loop::Event |
1223 | POE/Event 2000 6328 109.99 751.67 14.02 via POE::Loop::Event |
1212 | POE/Select 2000 6027 94.54 809.13 579.80 via POE::Loop::Select |
1224 | POE/Select 2000 6027 94.54 809.13 579.80 via POE::Loop::Select |
1213 | |
1225 | |
… | |
… | |
1242 | few of them active), of course, but this was not subject of this |
1254 | few of them active), of course, but this was not subject of this |
1243 | benchmark. |
1255 | benchmark. |
1244 | |
1256 | |
1245 | The "Event" module has a relatively high setup and callback invocation |
1257 | The "Event" module has a relatively high setup and callback invocation |
1246 | cost, but overall scores in on the third place. |
1258 | cost, but overall scores in on the third place. |
|
|
1259 | |
|
|
1260 | "IO::Async" performs admirably well, about on par with "Event", even |
|
|
1261 | when using its pure perl backend. |
1247 | |
1262 | |
1248 | "Glib"'s memory usage is quite a bit higher, but it features a faster |
1263 | "Glib"'s memory usage is quite a bit higher, but it features a faster |
1249 | callback invocation and overall ends up in the same class as "Event". |
1264 | callback invocation and overall ends up in the same class as "Event". |
1250 | However, Glib scales extremely badly, doubling the number of watchers |
1265 | However, Glib scales extremely badly, doubling the number of watchers |
1251 | increases the processing time by more than a factor of four, making it |
1266 | increases the processing time by more than a factor of four, making it |
… | |
… | |
1322 | single "request", that is, reading the token from the pipe and |
1337 | single "request", that is, reading the token from the pipe and |
1323 | forwarding it to another server. This includes deleting the old timeout |
1338 | forwarding it to another server. This includes deleting the old timeout |
1324 | and creating a new one that moves the timeout into the future. |
1339 | and creating a new one that moves the timeout into the future. |
1325 | |
1340 | |
1326 | Results |
1341 | Results |
1327 | name sockets create request |
1342 | name sockets create request |
1328 | EV 20000 69.01 11.16 |
1343 | EV 20000 69.01 11.16 |
1329 | Perl 20000 73.32 35.87 |
1344 | Perl 20000 73.32 35.87 |
|
|
1345 | IOAsync 20000 157.00 98.14 epoll |
|
|
1346 | IOAsync 20000 159.31 616.06 poll |
1330 | Event 20000 212.62 257.32 |
1347 | Event 20000 212.62 257.32 |
1331 | Glib 20000 651.16 1896.30 |
1348 | Glib 20000 651.16 1896.30 |
1332 | POE 20000 349.67 12317.24 uses POE::Loop::Event |
1349 | POE 20000 349.67 12317.24 uses POE::Loop::Event |
1333 | |
1350 | |
1334 | Discussion |
1351 | Discussion |
1335 | This benchmark *does* measure scalability and overall performance of the |
1352 | This benchmark *does* measure scalability and overall performance of the |
1336 | particular event loop. |
1353 | particular event loop. |
1337 | |
1354 | |
1338 | EV is again fastest. Since it is using epoll on my system, the setup |
1355 | EV is again fastest. Since it is using epoll on my system, the setup |
1339 | time is relatively high, though. |
1356 | time is relatively high, though. |
1340 | |
1357 | |
1341 | Perl surprisingly comes second. It is much faster than the C-based event |
1358 | Perl surprisingly comes second. It is much faster than the C-based event |
1342 | loops Event and Glib. |
1359 | loops Event and Glib. |
|
|
1360 | |
|
|
1361 | IO::Async performs very well when using its epoll backend, and still |
|
|
1362 | quite good compared to Glib when using its pure perl backend. |
1343 | |
1363 | |
1344 | Event suffers from high setup time as well (look at its code and you |
1364 | Event suffers from high setup time as well (look at its code and you |
1345 | will understand why). Callback invocation also has a high overhead |
1365 | will understand why). Callback invocation also has a high overhead |
1346 | compared to the "$_->() for .."-style loop that the Perl event loop |
1366 | compared to the "$_->() for .."-style loop that the Perl event loop |
1347 | uses. Event uses select or poll in basically all documented |
1367 | uses. Event uses select or poll in basically all documented |
… | |
… | |
1398 | |
1418 | |
1399 | Summary |
1419 | Summary |
1400 | * C-based event loops perform very well with small number of watchers, |
1420 | * C-based event loops perform very well with small number of watchers, |
1401 | as the management overhead dominates. |
1421 | as the management overhead dominates. |
1402 | |
1422 | |
|
|
1423 | THE IO::Lambda BENCHMARK |
|
|
1424 | Recently I was told about the benchmark in the IO::Lambda manpage, which |
|
|
1425 | could be misinterpreted to make AnyEvent look bad. In fact, the |
|
|
1426 | benchmark simply compares IO::Lambda with POE, and IO::Lambda looks |
|
|
1427 | better (which shouldn't come as a surprise to anybody). As such, the |
|
|
1428 | benchmark is fine, and mostly shows that the AnyEvent backend from |
|
|
1429 | IO::Lambda isn't very optimal. But how would AnyEvent compare when used |
|
|
1430 | without the extra baggage? To explore this, I wrote the equivalent |
|
|
1431 | benchmark for AnyEvent. |
|
|
1432 | |
|
|
1433 | The benchmark itself creates an echo-server, and then, for 500 times, |
|
|
1434 | connects to the echo server, sends a line, waits for the reply, and then |
|
|
1435 | creates the next connection. This is a rather bad benchmark, as it |
|
|
1436 | doesn't test the efficiency of the framework or much non-blocking I/O, |
|
|
1437 | but it is a benchmark nevertheless. |
|
|
1438 | |
|
|
1439 | name runtime |
|
|
1440 | Lambda/select 0.330 sec |
|
|
1441 | + optimized 0.122 sec |
|
|
1442 | Lambda/AnyEvent 0.327 sec |
|
|
1443 | + optimized 0.138 sec |
|
|
1444 | Raw sockets/select 0.077 sec |
|
|
1445 | POE/select, components 0.662 sec |
|
|
1446 | POE/select, raw sockets 0.226 sec |
|
|
1447 | POE/select, optimized 0.404 sec |
|
|
1448 | |
|
|
1449 | AnyEvent/select/nb 0.085 sec |
|
|
1450 | AnyEvent/EV/nb 0.068 sec |
|
|
1451 | +state machine 0.134 sec |
|
|
1452 | |
|
|
1453 | The benchmark is also a bit unfair (my fault): the IO::Lambda/POE |
|
|
1454 | benchmarks actually make blocking connects and use 100% blocking I/O, |
|
|
1455 | defeating the purpose of an event-based solution. All of the newly |
|
|
1456 | written AnyEvent benchmarks use 100% non-blocking connects (using |
|
|
1457 | AnyEvent::Socket::tcp_connect and the asynchronous pure perl DNS |
|
|
1458 | resolver), so AnyEvent is at a disadvantage here, as non-blocking |
|
|
1459 | connects generally require a lot more bookkeeping and event handling |
|
|
1460 | than blocking connects (which involve a single syscall only). |
|
|
1461 | |
|
|
1462 | The last AnyEvent benchmark additionally uses AnyEvent::Handle, which |
|
|
1463 | offers similar expressive power as POE and IO::Lambda, using |
|
|
1464 | conventional Perl syntax. This means that both the echo server and the |
|
|
1465 | client are 100% non-blocking, further placing it at a disadvantage. |
|
|
1466 | |
|
|
1467 | As you can see, the AnyEvent + EV combination even beats the |
|
|
1468 | hand-optimised "raw sockets benchmark", while AnyEvent + its pure perl |
|
|
1469 | backend easily beats IO::Lambda and POE. |
|
|
1470 | |
|
|
1471 | And even the 100% non-blocking version written using the high-level (and |
|
|
1472 | slow :) AnyEvent::Handle abstraction beats both POE and IO::Lambda by a |
|
|
1473 | large margin, even though it does all of DNS, tcp-connect and socket I/O |
|
|
1474 | in a non-blocking way. |
|
|
1475 | |
|
|
1476 | The two AnyEvent benchmarks programs can be found as eg/ae0.pl and |
|
|
1477 | eg/ae2.pl in the AnyEvent distribution, the remaining benchmarks are |
|
|
1478 | part of the IO::lambda distribution and were used without any changes. |
|
|
1479 | |
1403 | SIGNALS |
1480 | SIGNALS |
1404 | AnyEvent currently installs handlers for these signals: |
1481 | AnyEvent currently installs handlers for these signals: |
1405 | |
1482 | |
1406 | SIGCHLD |
1483 | SIGCHLD |
1407 | A handler for "SIGCHLD" is installed by AnyEvent's child watcher |
1484 | A handler for "SIGCHLD" is installed by AnyEvent's child watcher |
1408 | emulation for event loops that do not support them natively. Also, |
1485 | emulation for event loops that do not support them natively. Also, |
1409 | some event loops install a similar handler. |
1486 | some event loops install a similar handler. |
|
|
1487 | |
|
|
1488 | If, when AnyEvent is loaded, SIGCHLD is set to IGNORE, then AnyEvent |
|
|
1489 | will reset it to default, to avoid losing child exit statuses. |
1410 | |
1490 | |
1411 | SIGPIPE |
1491 | SIGPIPE |
1412 | A no-op handler is installed for "SIGPIPE" when $SIG{PIPE} is |
1492 | A no-op handler is installed for "SIGPIPE" when $SIG{PIPE} is |
1413 | "undef" when AnyEvent gets loaded. |
1493 | "undef" when AnyEvent gets loaded. |
1414 | |
1494 | |
… | |
… | |
1442 | |
1522 | |
1443 | You can make AnyEvent completely ignore this variable by deleting it |
1523 | You can make AnyEvent completely ignore this variable by deleting it |
1444 | before the first watcher gets created, e.g. with a "BEGIN" block: |
1524 | before the first watcher gets created, e.g. with a "BEGIN" block: |
1445 | |
1525 | |
1446 | BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} } |
1526 | BEGIN { delete $ENV{PERL_ANYEVENT_MODEL} } |
1447 | |
1527 | |
1448 | use AnyEvent; |
1528 | use AnyEvent; |
1449 | |
1529 | |
1450 | Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can |
1530 | Similar considerations apply to $ENV{PERL_ANYEVENT_VERBOSE}, as that can |
1451 | be used to probe what backend is used and gain other information (which |
1531 | be used to probe what backend is used and gain other information (which |
1452 | is probably even less useful to an attacker than PERL_ANYEVENT_MODEL), |
1532 | is probably even less useful to an attacker than PERL_ANYEVENT_MODEL), |
1453 | and $ENV{PERL_ANYEGENT_STRICT}. |
1533 | and $ENV{PERL_ANYEVENT_STRICT}. |
|
|
1534 | |
|
|
1535 | Note that AnyEvent will remove *all* environment variables starting with |
|
|
1536 | "PERL_ANYEVENT_" from %ENV when it is loaded while taint mode is |
|
|
1537 | enabled. |
1454 | |
1538 | |
1455 | BUGS |
1539 | BUGS |
1456 | Perl 5.8 has numerous memleaks that sometimes hit this module and are |
1540 | Perl 5.8 has numerous memleaks that sometimes hit this module and are |
1457 | hard to work around. If you suffer from memleaks, first upgrade to Perl |
1541 | hard to work around. If you suffer from memleaks, first upgrade to Perl |
1458 | 5.10 and check wether the leaks still show up. (Perl 5.10.0 has other |
1542 | 5.10 and check wether the leaks still show up. (Perl 5.10.0 has other |