… | |
… | |
168 | use common::sense; |
168 | use common::sense; |
169 | |
169 | |
170 | use base 'Exporter'; |
170 | use base 'Exporter'; |
171 | |
171 | |
172 | BEGIN { |
172 | BEGIN { |
173 | our $VERSION = '3.72'; |
173 | our $VERSION = '3.9'; |
174 | |
174 | |
175 | our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close |
175 | our @AIO_REQ = qw(aio_sendfile aio_read aio_write aio_open aio_close |
176 | aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx |
176 | aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx |
177 | aio_scandir aio_symlink aio_readlink aio_sync aio_fsync |
177 | aio_scandir aio_symlink aio_readlink aio_sync aio_fsync |
178 | aio_fdatasync aio_sync_file_range aio_pathsync aio_readahead |
178 | aio_fdatasync aio_sync_file_range aio_pathsync aio_readahead |
… | |
… | |
368 | } else { |
368 | } else { |
369 | die "open failed: $!\n"; |
369 | die "open failed: $!\n"; |
370 | } |
370 | } |
371 | }; |
371 | }; |
372 | |
372 | |
|
|
373 | In addition to all the common open modes/flags (C<O_RDONLY>, C<O_WRONLY>, |
|
|
374 | C<O_RDWR>, C<O_CREAT>, C<O_TRUNC>, C<O_EXCL> and C<O_APPEND>), the |
|
|
375 | following POSIX and non-POSIX constants are available (missing ones on |
|
|
376 | your system are, as usual, C<0>): |
|
|
377 | |
|
|
378 | C<O_ASYNC>, C<O_DIRECT>, C<O_NOATIME>, C<O_CLOEXEC>, C<O_NOCTTY>, C<O_NOFOLLOW>, |
|
|
379 | C<O_NONBLOCK>, C<O_EXEC>, C<O_SEARCH>, C<O_DIRECTORY>, C<O_DSYNC>, |
|
|
380 | C<O_RSYNC>, C<O_SYNC> and C<O_TTY_INIT>. |
|
|
381 | |
373 | |
382 | |
374 | =item aio_close $fh, $callback->($status) |
383 | =item aio_close $fh, $callback->($status) |
375 | |
384 | |
376 | Asynchronously close a file and call the callback with the result |
385 | Asynchronously close a file and call the callback with the result |
377 | code. |
386 | code. |
… | |
… | |
427 | |
436 | |
428 | Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts |
437 | Tries to copy C<$length> bytes from C<$in_fh> to C<$out_fh>. It starts |
429 | reading at byte offset C<$in_offset>, and starts writing at the current |
438 | reading at byte offset C<$in_offset>, and starts writing at the current |
430 | file offset of C<$out_fh>. Because of that, it is not safe to issue more |
439 | file offset of C<$out_fh>. Because of that, it is not safe to issue more |
431 | than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each |
440 | than one C<aio_sendfile> per C<$out_fh>, as they will interfere with each |
432 | other. |
441 | other. The same C<$in_fh> works fine though, as this function does not |
|
|
442 | move or use the file offset of C<$in_fh>. |
433 | |
443 | |
434 | Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than |
444 | Please note that C<aio_sendfile> can read more bytes from C<$in_fh> than |
435 | are written, and there is no way to find out how many bytes have been read |
445 | are written, and there is no way to find out how many more bytes have been |
436 | from C<aio_sendfile> alone, as C<aio_sendfile> only provides the number of |
446 | read from C<aio_sendfile> alone, as C<aio_sendfile> only provides the |
437 | bytes written to C<$out_fh>. Only if the result value equals C<$length> |
447 | number of bytes written to C<$out_fh>. Only if the result value equals |
438 | one can assume that C<$length> bytes have been read. |
448 | C<$length> one can assume that C<$length> bytes have been read. |
439 | |
449 | |
440 | Unlike with other C<aio_> functions, it makes a lot of sense to use |
450 | Unlike with other C<aio_> functions, it makes a lot of sense to use |
441 | C<aio_sendfile> on non-blocking sockets, as long as one end (typically |
451 | C<aio_sendfile> on non-blocking sockets, as long as one end (typically |
442 | the C<$in_fh>) is a file - the file I/O will then be asynchronous, while |
452 | the C<$in_fh>) is a file - the file I/O will then be asynchronous, while |
443 | the socket I/O will be non-blocking. Note, however, that you can run into |
453 | the socket I/O will be non-blocking. Note, however, that you can run |
444 | a trap where C<aio_sendfile> reads some data with readahead, then fails |
454 | into a trap where C<aio_sendfile> reads some data with readahead, then |
445 | to write all data, and when the socket is ready the next time, the data |
455 | fails to write all data, and when the socket is ready the next time, the |
446 | in the cache is already lost, forcing C<aio_sendfile> to again hit the |
456 | data in the cache is already lost, forcing C<aio_sendfile> to again hit |
447 | disk. Explicit C<aio_read> + C<aio_write> let's you control resource usage |
457 | the disk. Explicit C<aio_read> + C<aio_write> let's you better control |
448 | much better. |
458 | resource usage. |
449 | |
459 | |
450 | This call tries to make use of a native C<sendfile> syscall to provide |
460 | This call tries to make use of a native C<sendfile>-like syscall to |
451 | zero-copy operation. For this to work, C<$out_fh> should refer to a |
461 | provide zero-copy operation. For this to work, C<$out_fh> should refer to |
452 | socket, and C<$in_fh> should refer to an mmap'able file. |
462 | a socket, and C<$in_fh> should refer to an mmap'able file. |
453 | |
463 | |
454 | If a native sendfile cannot be found or it fails with C<ENOSYS>, |
464 | If a native sendfile cannot be found or it fails with C<ENOSYS>, |
455 | C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or C<ENOTSOCK>, |
465 | C<EINVAL>, C<ENOTSUP>, C<EOPNOTSUPP>, C<EAFNOSUPPORT>, C<EPROTOTYPE> or |
456 | it will be emulated, so you can call C<aio_sendfile> on any type of |
466 | C<ENOTSOCK>, it will be emulated, so you can call C<aio_sendfile> on any |
457 | filehandle regardless of the limitations of the operating system. |
467 | type of filehandle regardless of the limitations of the operating system. |
|
|
468 | |
|
|
469 | As native sendfile syscalls (as practically any non-POSIX interface hacked |
|
|
470 | together in a hurry to improve benchmark numbers) tend to be rather buggy |
|
|
471 | on many systems, this implementation tries to work around some known bugs |
|
|
472 | in Linux and FreeBSD kernels (probably others, too), but that might fail, |
|
|
473 | so you really really should check the return value of C<aio_sendfile> - |
|
|
474 | fewre bytes than expected might have been transferred. |
458 | |
475 | |
459 | |
476 | |
460 | =item aio_readahead $fh,$offset,$length, $callback->($retval) |
477 | =item aio_readahead $fh,$offset,$length, $callback->($retval) |
461 | |
478 | |
462 | C<aio_readahead> populates the page cache with data from a file so that |
479 | C<aio_readahead> populates the page cache with data from a file so that |
… | |
… | |
668 | |
685 | |
669 | =over 4 |
686 | =over 4 |
670 | |
687 | |
671 | =item IO::AIO::READDIR_DENTS |
688 | =item IO::AIO::READDIR_DENTS |
672 | |
689 | |
673 | When this flag is off, then the callback gets an arrayref with of names |
690 | When this flag is off, then the callback gets an arrayref consisting of |
674 | only (as with C<aio_readdir>), otherwise it gets an arrayref with |
691 | names only (as with C<aio_readdir>), otherwise it gets an arrayref with |
675 | C<[$name, $type, $inode]> arrayrefs, each describing a single directory |
692 | C<[$name, $type, $inode]> arrayrefs, each describing a single directory |
676 | entry in more detail. |
693 | entry in more detail. |
677 | |
694 | |
678 | C<$name> is the name of the entry. |
695 | C<$name> is the name of the entry. |
679 | |
696 | |
… | |
… | |
692 | systems that do not deliver the inode information. |
709 | systems that do not deliver the inode information. |
693 | |
710 | |
694 | =item IO::AIO::READDIR_DIRS_FIRST |
711 | =item IO::AIO::READDIR_DIRS_FIRST |
695 | |
712 | |
696 | When this flag is set, then the names will be returned in an order where |
713 | When this flag is set, then the names will be returned in an order where |
697 | likely directories come first. This is useful when you need to quickly |
714 | likely directories come first, in optimal stat order. This is useful when |
698 | find directories, or you want to find all directories while avoiding to |
715 | you need to quickly find directories, or you want to find all directories |
699 | stat() each entry. |
716 | while avoiding to stat() each entry. |
700 | |
717 | |
701 | If the system returns type information in readdir, then this is used |
718 | If the system returns type information in readdir, then this is used |
702 | to find directories directly. Otherwise, likely directories are files |
719 | to find directories directly. Otherwise, likely directories are names |
703 | beginning with ".", or otherwise files with no dots, of which files with |
720 | beginning with ".", or otherwise names with no dots, of which names with |
704 | short names are tried first. |
721 | short names are tried first. |
705 | |
722 | |
706 | =item IO::AIO::READDIR_STAT_ORDER |
723 | =item IO::AIO::READDIR_STAT_ORDER |
707 | |
724 | |
708 | When this flag is set, then the names will be returned in an order |
725 | When this flag is set, then the names will be returned in an order |
… | |
… | |
853 | if ($_[0] && $! == EXDEV) { |
870 | if ($_[0] && $! == EXDEV) { |
854 | aioreq_pri $pri; |
871 | aioreq_pri $pri; |
855 | add $grp aio_copy $src, $dst, sub { |
872 | add $grp aio_copy $src, $dst, sub { |
856 | $grp->result ($_[0]); |
873 | $grp->result ($_[0]); |
857 | |
874 | |
858 | if (!$_[0]) { |
875 | unless ($_[0]) { |
859 | aioreq_pri $pri; |
876 | aioreq_pri $pri; |
860 | add $grp aio_unlink $src; |
877 | add $grp aio_unlink $src; |
861 | } |
878 | } |
862 | }; |
879 | }; |
863 | } else { |
880 | } else { |
… | |
… | |
1406 | |
1423 | |
1407 | See C<poll_cb> for an example. |
1424 | See C<poll_cb> for an example. |
1408 | |
1425 | |
1409 | =item IO::AIO::poll_cb |
1426 | =item IO::AIO::poll_cb |
1410 | |
1427 | |
1411 | Process some outstanding events on the result pipe. You have to call this |
1428 | Process some outstanding events on the result pipe. You have to call |
1412 | regularly. Returns C<0> if all events could be processed, or C<-1> if it |
1429 | this regularly. Returns C<0> if all events could be processed (or there |
1413 | returned earlier for whatever reason. Returns immediately when no events |
1430 | were no events to process), or C<-1> if it returned earlier for whatever |
1414 | are outstanding. The amount of events processed depends on the settings of |
1431 | reason. Returns immediately when no events are outstanding. The amount of |
1415 | C<IO::AIO::max_poll_req> and C<IO::AIO::max_poll_time>. |
1432 | events processed depends on the settings of C<IO::AIO::max_poll_req> and |
|
|
1433 | C<IO::AIO::max_poll_time>. |
1416 | |
1434 | |
1417 | If not all requests were processed for whatever reason, the filehandle |
1435 | If not all requests were processed for whatever reason, the filehandle |
1418 | will still be ready when C<poll_cb> returns, so normally you don't have to |
1436 | will still be ready when C<poll_cb> returns, so normally you don't have to |
1419 | do anything special to have it called later. |
1437 | do anything special to have it called later. |
|
|
1438 | |
|
|
1439 | Apart from calling C<IO::AIO::poll_cb> when the event filehandle becomes |
|
|
1440 | ready, it can be beneficial to call this function from loops which submit |
|
|
1441 | a lot of requests, to make sure the results get processed when they become |
|
|
1442 | available and not just when the loop is finished and the event loop takes |
|
|
1443 | over again. This function returns very fast when there are no outstanding |
|
|
1444 | requests. |
1420 | |
1445 | |
1421 | Example: Install an Event watcher that automatically calls |
1446 | Example: Install an Event watcher that automatically calls |
1422 | IO::AIO::poll_cb with high priority (more examples can be found in the |
1447 | IO::AIO::poll_cb with high priority (more examples can be found in the |
1423 | SYNOPSIS section, at the top of this document): |
1448 | SYNOPSIS section, at the top of this document): |
1424 | |
1449 | |
… | |
… | |
1547 | Sets the minimum idle timeout (default 10) after which worker threads are |
1572 | Sets the minimum idle timeout (default 10) after which worker threads are |
1548 | allowed to exit. SEe C<IO::AIO::max_idle>. |
1573 | allowed to exit. SEe C<IO::AIO::max_idle>. |
1549 | |
1574 | |
1550 | =item IO::AIO::max_outstanding $maxreqs |
1575 | =item IO::AIO::max_outstanding $maxreqs |
1551 | |
1576 | |
|
|
1577 | Sets the maximum number of outstanding requests to C<$nreqs>. If |
|
|
1578 | you do queue up more than this number of requests, the next call to |
|
|
1579 | C<IO::AIO::poll_cb> (and other functions calling C<poll_cb>, such as |
|
|
1580 | C<IO::AIO::flush> or C<IO::AIO::poll>) will block until the limit is no |
|
|
1581 | longer exceeded. |
|
|
1582 | |
|
|
1583 | In other words, this setting does not enforce a queue limit, but can be |
|
|
1584 | used to make poll functions block if the limit is exceeded. |
|
|
1585 | |
1552 | This is a very bad function to use in interactive programs because it |
1586 | This is a very bad function to use in interactive programs because it |
1553 | blocks, and a bad way to reduce concurrency because it is inexact: Better |
1587 | blocks, and a bad way to reduce concurrency because it is inexact: Better |
1554 | use an C<aio_group> together with a feed callback. |
1588 | use an C<aio_group> together with a feed callback. |
1555 | |
1589 | |
1556 | Sets the maximum number of outstanding requests to C<$nreqs>. If you |
1590 | It's main use is in scripts without an event loop - when you want to stat |
1557 | do queue up more than this number of requests, the next call to the |
1591 | a lot of files, you can write somehting like this: |
1558 | C<poll_cb> (and C<poll_some> and other functions calling C<poll_cb>) |
|
|
1559 | function will block until the limit is no longer exceeded. |
|
|
1560 | |
1592 | |
1561 | The default value is very large, so there is no practical limit on the |
1593 | IO::AIO::max_outstanding 32; |
1562 | number of outstanding requests. |
|
|
1563 | |
1594 | |
1564 | You can still queue as many requests as you want. Therefore, |
1595 | for my $path (...) { |
1565 | C<max_outstanding> is mainly useful in simple scripts (with low values) or |
1596 | aio_stat $path , ...; |
1566 | as a stop gap to shield against fatal memory overflow (with large values). |
1597 | IO::AIO::poll_cb; |
|
|
1598 | } |
|
|
1599 | |
|
|
1600 | IO::AIO::flush; |
|
|
1601 | |
|
|
1602 | The call to C<poll_cb> inside the loop will normally return instantly, but |
|
|
1603 | as soon as more thna C<32> reqeusts are in-flight, it will block until |
|
|
1604 | some requests have been handled. This keeps the loop from pushing a large |
|
|
1605 | number of C<aio_stat> requests onto the queue. |
|
|
1606 | |
|
|
1607 | The default value for C<max_outstanding> is very large, so there is no |
|
|
1608 | practical limit on the number of outstanding requests. |
1567 | |
1609 | |
1568 | =back |
1610 | =back |
1569 | |
1611 | |
1570 | =head3 STATISTICAL INFORMATION |
1612 | =head3 STATISTICAL INFORMATION |
1571 | |
1613 | |