… | |
… | |
18 | }; |
18 | }; |
19 | |
19 | |
20 | # version 2+ has request and group objects |
20 | # version 2+ has request and group objects |
21 | use IO::AIO 2; |
21 | use IO::AIO 2; |
22 | |
22 | |
|
|
23 | aioreq_pri 4; # give next request a very high priority |
23 | my $req = aio_unlink "/tmp/file", sub { }; |
24 | my $req = aio_unlink "/tmp/file", sub { }; |
24 | $req->cancel; # cancel request if still in queue |
25 | $req->cancel; # cancel request if still in queue |
25 | |
26 | |
26 | my $grp = aio_group sub { print "all stats done\n" }; |
27 | my $grp = aio_group sub { print "all stats done\n" }; |
27 | add $grp aio_stat "..." for ...; |
28 | add $grp aio_stat "..." for ...; |
… | |
… | |
50 | =head1 DESCRIPTION |
51 | =head1 DESCRIPTION |
51 | |
52 | |
52 | This module implements asynchronous I/O using whatever means your |
53 | This module implements asynchronous I/O using whatever means your |
53 | operating system supports. |
54 | operating system supports. |
54 | |
55 | |
55 | Currently, a number of threads are started that execute your read/writes |
56 | In this version, a number of threads are started that execute your |
56 | and signal their completion. You don't need thread support in your libc or |
57 | requests and signal their completion. You don't need thread support |
57 | perl, and the threads created by this module will not be visible to the |
58 | in perl, and the threads created by this module will not be visible |
58 | pthreads library. In the future, this module might make use of the native |
59 | to perl. In the future, this module might make use of the native aio |
59 | aio functions available on many operating systems. However, they are often |
60 | functions available on many operating systems. However, they are often |
60 | not well-supported (Linux doesn't allow them on normal files currently, |
61 | not well-supported or restricted (Linux doesn't allow them on normal |
61 | for example), and they would only support aio_read and aio_write, so the |
62 | files currently, for example), and they would only support aio_read and |
62 | remaining functionality would have to be implemented using threads anyway. |
63 | aio_write, so the remaining functionality would have to be implemented |
|
|
64 | using threads anyway. |
63 | |
65 | |
64 | Although the module will work with in the presence of other threads, it is |
66 | Although the module will work with in the presence of other (Perl-) |
65 | currently not reentrant, so use appropriate locking yourself, always call |
67 | threads, it is currently not reentrant in any way, so use appropriate |
66 | C<poll_cb> from within the same thread, or never call C<poll_cb> (or other |
68 | locking yourself, always call C<poll_cb> from within the same thread, or |
67 | C<aio_> functions) recursively. |
69 | never call C<poll_cb> (or other C<aio_> functions) recursively. |
|
|
70 | |
|
|
71 | =head1 REQUEST ANATOMY AND LIFETIME |
|
|
72 | |
|
|
73 | Every C<aio_*> function creates a request. which is a C data structure not |
|
|
74 | directly visible to Perl. |
|
|
75 | |
|
|
76 | If called in non-void context, every request function returns a Perl |
|
|
77 | object representing the request. In void context, nothing is returned, |
|
|
78 | which saves a bit of memory. |
|
|
79 | |
|
|
80 | The perl object is a fairly standard ref-to-hash object. The hash contents |
|
|
81 | are not used by IO::AIO so you are free to store anything you like in it. |
|
|
82 | |
|
|
83 | During their existance, aio requests travel through the following states, |
|
|
84 | in order: |
|
|
85 | |
|
|
86 | =over 4 |
|
|
87 | |
|
|
88 | =item ready |
|
|
89 | |
|
|
90 | Immediately after a request is created it is put into the ready state, |
|
|
91 | waiting for a thread to execute it. |
|
|
92 | |
|
|
93 | =item execute |
|
|
94 | |
|
|
95 | A thread has accepted the request for processing and is currently |
|
|
96 | executing it (e.g. blocking in read). |
|
|
97 | |
|
|
98 | =item pending |
|
|
99 | |
|
|
100 | The request has been executed and is waiting for result processing. |
|
|
101 | |
|
|
102 | While request submission and execution is fully asynchronous, result |
|
|
103 | processing is not and relies on the perl interpreter calling C<poll_cb> |
|
|
104 | (or another function with the same effect). |
|
|
105 | |
|
|
106 | =item result |
|
|
107 | |
|
|
108 | The request results are processed synchronously by C<poll_cb>. |
|
|
109 | |
|
|
110 | The C<poll_cb> function will process all outstanding aio requests by |
|
|
111 | calling their callbacks, freeing memory associated with them and managing |
|
|
112 | any groups they are contained in. |
|
|
113 | |
|
|
114 | =item done |
|
|
115 | |
|
|
116 | Request has reached the end of its lifetime and holds no resources anymore |
|
|
117 | (except possibly for the Perl object, but its connection to the actual |
|
|
118 | aio request is severed and calling its methods will either do nothing or |
|
|
119 | result in a runtime error). |
68 | |
120 | |
69 | =cut |
121 | =cut |
70 | |
122 | |
71 | package IO::AIO; |
123 | package IO::AIO; |
72 | |
124 | |
… | |
… | |
76 | use base 'Exporter'; |
128 | use base 'Exporter'; |
77 | |
129 | |
78 | BEGIN { |
130 | BEGIN { |
79 | our $VERSION = '2.0'; |
131 | our $VERSION = '2.0'; |
80 | |
132 | |
81 | our @EXPORT = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat |
133 | our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close aio_stat |
82 | aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink |
134 | aio_lstat aio_unlink aio_rmdir aio_readdir aio_scandir aio_symlink |
83 | aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move |
135 | aio_fsync aio_fdatasync aio_readahead aio_rename aio_link aio_move |
84 | aio_group aio_nop); |
136 | aio_group aio_nop); |
85 | our @EXPORT_OK = qw(poll_fileno poll_cb min_parallel max_parallel max_outstanding nreqs); |
137 | our @EXPORT = (@AIO_REQ, qw(aioreq_pri aioreq_nice)); |
|
|
138 | our @EXPORT_OK = qw(poll_fileno poll_cb poll_wait flush |
|
|
139 | min_parallel max_parallel max_outstanding nreqs); |
86 | |
140 | |
87 | @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; |
141 | @IO::AIO::GRP::ISA = 'IO::AIO::REQ'; |
88 | |
142 | |
89 | require XSLoader; |
143 | require XSLoader; |
90 | XSLoader::load ("IO::AIO", $VERSION); |
144 | XSLoader::load ("IO::AIO", $VERSION); |
… | |
… | |
121 | environment, d) use Glib::filename_from_unicode on unicode filenames or e) |
175 | environment, d) use Glib::filename_from_unicode on unicode filenames or e) |
122 | use something else. |
176 | use something else. |
123 | |
177 | |
124 | =over 4 |
178 | =over 4 |
125 | |
179 | |
|
|
180 | =item aioreq_pri $pri |
|
|
181 | |
|
|
182 | Sets the priority for the next aio request. The default priority |
|
|
183 | is C<0>, the minimum and maximum priorities are C<-4> and C<4>, |
|
|
184 | respectively. Requests with higher priority will be serviced first. |
|
|
185 | |
|
|
186 | The priority will be reset to C<0> after each call to one of the C<aio_> |
|
|
187 | functions. |
|
|
188 | |
|
|
189 | Example: open a file with low priority, then read something from it with |
|
|
190 | higher priority so the read request is serviced before other low priority |
|
|
191 | open requests (potentially spamming the cache): |
|
|
192 | |
|
|
193 | aioreq_pri -3; |
|
|
194 | aio_open ..., sub { |
|
|
195 | return unless $_[0]; |
|
|
196 | |
|
|
197 | aioreq_pri -2; |
|
|
198 | aio_read $_[0], ..., sub { |
|
|
199 | ... |
|
|
200 | }; |
|
|
201 | }; |
|
|
202 | |
|
|
203 | =item aioreq_nice $pri_adjust |
|
|
204 | |
|
|
205 | Similar to C<aioreq_pri>, but subtracts the given value from the current |
|
|
206 | priority, so effects are cumulative. |
|
|
207 | |
126 | =item aio_open $pathname, $flags, $mode, $callback->($fh) |
208 | =item aio_open $pathname, $flags, $mode, $callback->($fh) |
127 | |
209 | |
128 | Asynchronously open or create a file and call the callback with a newly |
210 | Asynchronously open or create a file and call the callback with a newly |
129 | created filehandle for the file. |
211 | created filehandle for the file. |
130 | |
212 | |
… | |
… | |
181 | $_[0] > 0 or die "read error: $!"; |
263 | $_[0] > 0 or die "read error: $!"; |
182 | print "read $_[0] bytes: <$buffer>\n"; |
264 | print "read $_[0] bytes: <$buffer>\n"; |
183 | }; |
265 | }; |
184 | |
266 | |
185 | =item aio_move $srcpath, $dstpath, $callback->($status) |
267 | =item aio_move $srcpath, $dstpath, $callback->($status) |
186 | |
|
|
187 | [EXPERIMENTAL due to internal aio_group use] |
|
|
188 | |
268 | |
189 | Try to move the I<file> (directories not supported as either source or |
269 | Try to move the I<file> (directories not supported as either source or |
190 | destination) from C<$srcpath> to C<$dstpath> and call the callback with |
270 | destination) from C<$srcpath> to C<$dstpath> and call the callback with |
191 | the C<0> (error) or C<-1> ok. |
271 | the C<0> (error) or C<-1> ok. |
192 | |
272 | |
… | |
… | |
344 | |
424 | |
345 | The callback a single argument which is either C<undef> or an array-ref |
425 | The callback a single argument which is either C<undef> or an array-ref |
346 | with the filenames. |
426 | with the filenames. |
347 | |
427 | |
348 | =item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) |
428 | =item aio_scandir $path, $maxreq, $callback->($dirs, $nondirs) |
349 | |
|
|
350 | [EXPERIMENTAL due to internal aio_group use] |
|
|
351 | |
429 | |
352 | Scans a directory (similar to C<aio_readdir>) but additionally tries to |
430 | Scans a directory (similar to C<aio_readdir>) but additionally tries to |
353 | separate the entries of directory C<$path> into two sets of names, ones |
431 | separate the entries of directory C<$path> into two sets of names, ones |
354 | you can recurse into (directories or links to them), and ones you cannot |
432 | you can recurse into (directories or links to them), and ones you cannot |
355 | recurse into (everything else). |
433 | recurse into (everything else). |
… | |
… | |
517 | |
595 | |
518 | [EXPERIMENTAL] |
596 | [EXPERIMENTAL] |
519 | |
597 | |
520 | This is a very special aio request: Instead of doing something, it is a |
598 | This is a very special aio request: Instead of doing something, it is a |
521 | container for other aio requests, which is useful if you want to bundle |
599 | container for other aio requests, which is useful if you want to bundle |
522 | many requests into a single, composite, request. |
600 | many requests into a single, composite, request with a definite callback |
|
|
601 | and the ability to cancel the whole request with its subrequests. |
523 | |
602 | |
524 | Returns an object of class L<IO::AIO::GRP>. See its documentation below |
603 | Returns an object of class L<IO::AIO::GRP>. See its documentation below |
525 | for more info. |
604 | for more info. |
526 | |
605 | |
527 | Example: |
606 | Example: |
… | |
… | |
540 | This is a special request - it does nothing in itself and is only used for |
619 | This is a special request - it does nothing in itself and is only used for |
541 | side effects, such as when you want to add a dummy request to a group so |
620 | side effects, such as when you want to add a dummy request to a group so |
542 | that finishing the requests in the group depends on executing the given |
621 | that finishing the requests in the group depends on executing the given |
543 | code. |
622 | code. |
544 | |
623 | |
|
|
624 | While this request does nothing, it still goes through the execution |
|
|
625 | phase and still requires a worker thread. Thus, the callback will not |
|
|
626 | be executed immediately but only after other requests in the queue have |
|
|
627 | entered their execution phase. This can be used to measure request |
|
|
628 | latency. |
|
|
629 | |
545 | =item IO::AIO::aio_sleep $fractional_seconds, $callback->() *NOT EXPORTED* |
630 | =item IO::AIO::aio_busy $fractional_seconds, $callback->() *NOT EXPORTED* |
546 | |
631 | |
547 | Mainly used for debugging and benchmarking, this aio request puts one of |
632 | Mainly used for debugging and benchmarking, this aio request puts one of |
548 | the request workers to sleep for the given time. |
633 | the request workers to sleep for the given time. |
549 | |
634 | |
550 | While it is theoretically handy to have simple I/O scheduling requests |
635 | While it is theoretically handy to have simple I/O scheduling requests |
551 | like sleep and file handle readable/writable, the overhead this creates |
636 | like sleep and file handle readable/writable, the overhead this creates is |
552 | is immense, so do not use this function except to put your application |
637 | immense (it blocks a thread for a long time) so do not use this function |
553 | under artificial I/O pressure. |
638 | except to put your application under artificial I/O pressure. |
554 | |
639 | |
555 | =back |
640 | =back |
556 | |
641 | |
557 | =head2 IO::AIO::REQ CLASS |
642 | =head2 IO::AIO::REQ CLASS |
558 | |
643 | |
559 | All non-aggregate C<aio_*> functions return an object of this class when |
644 | All non-aggregate C<aio_*> functions return an object of this class when |
560 | called in non-void context. |
645 | called in non-void context. |
561 | |
646 | |
562 | A request always moves through the following five states in its lifetime, |
|
|
563 | in order: B<ready> (request has been created, but has not been executed |
|
|
564 | yet), B<execute> (request is currently being executed), B<pending> |
|
|
565 | (request has been executed but callback has not been called yet), |
|
|
566 | B<result> (results are being processed synchronously, includes calling the |
|
|
567 | callback) and B<done> (request has reached the end of its lifetime and |
|
|
568 | holds no resources anymore). |
|
|
569 | |
|
|
570 | =over 4 |
647 | =over 4 |
571 | |
648 | |
572 | =item $req->cancel |
649 | =item cancel $req |
573 | |
650 | |
574 | Cancels the request, if possible. Has the effect of skipping execution |
651 | Cancels the request, if possible. Has the effect of skipping execution |
575 | when entering the B<execute> state and skipping calling the callback when |
652 | when entering the B<execute> state and skipping calling the callback when |
576 | entering the the B<result> state, but will leave the request otherwise |
653 | entering the the B<result> state, but will leave the request otherwise |
577 | untouched. That means that requests that currently execute will not be |
654 | untouched. That means that requests that currently execute will not be |
578 | stopped and resources held by the request will not be freed prematurely. |
655 | stopped and resources held by the request will not be freed prematurely. |
579 | |
656 | |
|
|
657 | =item cb $req $callback->(...) |
|
|
658 | |
|
|
659 | Replace (or simply set) the callback registered to the request. |
|
|
660 | |
580 | =back |
661 | =back |
581 | |
662 | |
582 | =head2 IO::AIO::GRP CLASS |
663 | =head2 IO::AIO::GRP CLASS |
583 | |
664 | |
584 | This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to |
665 | This class is a subclass of L<IO::AIO::REQ>, so all its methods apply to |
… | |
… | |
641 | group. And only when all those requests have finished will the the group |
722 | group. And only when all those requests have finished will the the group |
642 | itself finish. |
723 | itself finish. |
643 | |
724 | |
644 | =over 4 |
725 | =over 4 |
645 | |
726 | |
|
|
727 | =item add $grp ... |
|
|
728 | |
646 | =item $grp->add (...) |
729 | =item $grp->add (...) |
647 | |
|
|
648 | =item add $grp ... |
|
|
649 | |
730 | |
650 | Add one or more requests to the group. Any type of L<IO::AIO::REQ> can |
731 | Add one or more requests to the group. Any type of L<IO::AIO::REQ> can |
651 | be added, including other groups, as long as you do not create circular |
732 | be added, including other groups, as long as you do not create circular |
652 | dependencies. |
733 | dependencies. |
653 | |
734 | |
… | |
… | |
656 | =item $grp->result (...) |
737 | =item $grp->result (...) |
657 | |
738 | |
658 | Set the result value(s) that will be passed to the group callback when all |
739 | Set the result value(s) that will be passed to the group callback when all |
659 | subrequests have finished. By default, no argument will be passed. |
740 | subrequests have finished. By default, no argument will be passed. |
660 | |
741 | |
661 | =item $grp->set_feeder ($callback->($grp)) |
742 | =item feed $grp $callback->($grp) |
662 | |
743 | |
663 | [VERY EXPERIMENTAL] |
744 | [VERY EXPERIMENTAL] |
664 | |
745 | |
665 | Sets a feeder/generator on this group: every group can have an attached |
746 | Sets a feeder/generator on this group: every group can have an attached |
666 | generator that generates requests if idle. The idea behind this is that, |
747 | generator that generates requests if idle. The idea behind this is that, |
… | |
… | |
669 | example, C<aio_scandir> might generate hundreds of thousands C<aio_stat> |
750 | example, C<aio_scandir> might generate hundreds of thousands C<aio_stat> |
670 | requests, delaying any later requests for a long time. |
751 | requests, delaying any later requests for a long time. |
671 | |
752 | |
672 | To avoid this, and allow incremental generation of requests, you can |
753 | To avoid this, and allow incremental generation of requests, you can |
673 | instead a group and set a feeder on it that generates those requests. The |
754 | instead a group and set a feeder on it that generates those requests. The |
674 | feeder will be called whenever there are few enough (see C<feeder_limit>, |
755 | feed callback will be called whenever there are few enough (see C<limit>, |
675 | below) requests active in the group itself and is expected to queue more |
756 | below) requests active in the group itself and is expected to queue more |
676 | requests. |
757 | requests. |
677 | |
758 | |
678 | The feeder can queue as many requests as it likes (i.e. C<add> does not |
759 | The feed callback can queue as many requests as it likes (i.e. C<add> does |
679 | impose any limits). |
760 | not impose any limits). |
680 | |
761 | |
681 | If the feeder does not queue more requests when called, it will be |
762 | If the feed does not queue more requests when called, it will be |
682 | automatically removed from the group. |
763 | automatically removed from the group. |
683 | |
764 | |
684 | If the feeder limit is C<0>, it will be set to C<2> automatically. |
765 | If the feed limit is C<0>, it will be set to C<2> automatically. |
685 | |
766 | |
686 | Example: |
767 | Example: |
687 | |
768 | |
688 | # stat all files in @files, but only ever use four aio requests concurrently: |
769 | # stat all files in @files, but only ever use four aio requests concurrently: |
689 | |
770 | |
690 | my $grp = aio_group sub { print "finished\n" }; |
771 | my $grp = aio_group sub { print "finished\n" }; |
691 | $grp->feeder_limit (4); |
772 | limit $grp 4; |
692 | $grp->set_feeder (sub { |
773 | feed $grp sub { |
693 | my $file = pop @files |
774 | my $file = pop @files |
694 | or return; |
775 | or return; |
695 | |
776 | |
696 | add $grp aio_stat $file, sub { ... }; |
777 | add $grp aio_stat $file, sub { ... }; |
697 | }); |
778 | }; |
698 | |
779 | |
699 | =item $grp->feeder_limit ($num) |
780 | =item limit $grp $num |
700 | |
781 | |
701 | Sets the feeder limit for the group: The feeder will be called whenever |
782 | Sets the feeder limit for the group: The feeder will be called whenever |
702 | the group contains less than this many requests. |
783 | the group contains less than this many requests. |
703 | |
784 | |
704 | Setting the limit to C<0> will pause the feeding process. |
785 | Setting the limit to C<0> will pause the feeding process. |
… | |
… | |
850 | This module should do "the right thing" when the process using it forks: |
931 | This module should do "the right thing" when the process using it forks: |
851 | |
932 | |
852 | Before the fork, IO::AIO enters a quiescent state where no requests |
933 | Before the fork, IO::AIO enters a quiescent state where no requests |
853 | can be added in other threads and no results will be processed. After |
934 | can be added in other threads and no results will be processed. After |
854 | the fork the parent simply leaves the quiescent state and continues |
935 | the fork the parent simply leaves the quiescent state and continues |
855 | request/result processing, while the child clears the request/result |
936 | request/result processing, while the child frees the request/result queue |
856 | queue (so the requests started before the fork will only be handled in |
937 | (so that the requests started before the fork will only be handled in the |
857 | the parent). Threads will be started on demand until the limit ste in the |
938 | parent). Threads will be started on demand until the limit set in the |
858 | parent process has been reached again. |
939 | parent process has been reached again. |
859 | |
940 | |
860 | In short: the parent will, after a short pause, continue as if fork had |
941 | In short: the parent will, after a short pause, continue as if fork had |
861 | not been called, while the child will act as if IO::AIO has not been used |
942 | not been called, while the child will act as if IO::AIO has not been used |
862 | yet. |
943 | yet. |
863 | |
944 | |
864 | =head2 MEMORY USAGE |
945 | =head2 MEMORY USAGE |
865 | |
946 | |
|
|
947 | Per-request usage: |
|
|
948 | |
866 | Each aio request uses - depending on your architecture - around 128 bytes |
949 | Each aio request uses - depending on your architecture - around 100-200 |
867 | of memory. In addition, stat requests need a stat buffer (possibly a few |
950 | bytes of memory. In addition, stat requests need a stat buffer (possibly |
868 | hundred bytes). Perl scalars and other data passed into aio requests will |
951 | a few hundred bytes), readdir requires a result buffer and so on. Perl |
869 | also be locked. |
952 | scalars and other data passed into aio requests will also be locked and |
|
|
953 | will consume memory till the request has entered the done state. |
870 | |
954 | |
871 | This is now awfully much, so queuing lots of requests is not usually a |
955 | This is now awfully much, so queuing lots of requests is not usually a |
872 | problem. |
956 | problem. |
873 | |
957 | |
874 | Each thread needs a stack area which is usually around 16k, sometimes much |
958 | Per-thread usage: |
875 | larger, depending on the OS. |
959 | |
|
|
960 | In the execution phase, some aio requests require more memory for |
|
|
961 | temporary buffers, and each thread requires a stack and other data |
|
|
962 | structures (usually around 16k-128k, depending on the OS). |
|
|
963 | |
|
|
964 | =head1 KNOWN BUGS |
|
|
965 | |
|
|
966 | Known bugs will be fixed in the next release. |
876 | |
967 | |
877 | =head1 SEE ALSO |
968 | =head1 SEE ALSO |
878 | |
969 | |
879 | L<Coro>, L<Linux::AIO> (obsolete). |
970 | L<Coro::AIO>. |
880 | |
971 | |
881 | =head1 AUTHOR |
972 | =head1 AUTHOR |
882 | |
973 | |
883 | Marc Lehmann <schmorp@schmorp.de> |
974 | Marc Lehmann <schmorp@schmorp.de> |
884 | http://home.schmorp.de/ |
975 | http://home.schmorp.de/ |