[ViewVC] Diff of: cvs/IO-AIO/AIO.pm

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.290 by root, Sun Aug 12 06:03:20 2018 UTC vs.
Revision 1.309 by root, Tue Dec 29 15:20:12 2020 UTC

 use common::sense;
 use base 'Exporter';
 BEGIN {
-   our $VERSION = 4.51;
+   our $VERSION = 4.74;
    our @AIO_REQ = qw(aio_sendfile aio_seek aio_read aio_write aio_open aio_close
                      aio_stat aio_lstat aio_unlink aio_rmdir aio_readdir aio_readdirx
                      aio_scandir aio_symlink aio_readlink aio_realpath aio_fcntl aio_ioctl
                      aio_sync aio_fsync aio_syncfs aio_fdatasync aio_sync_file_range
    IO::AIO::idle_timeout $seconds
    IO::AIO::max_outstanding $maxreqs
    IO::AIO::nreqs
    IO::AIO::nready
    IO::AIO::npending
+   IO::AIO::reinit
-   $nfd = IO::AIO::get_fdlimit [EXPERIMENTAL]
+   $nfd = IO::AIO::get_fdlimit
-   IO::AIO::min_fdlimit $nfd [EXPERIMENTAL]
+   IO::AIO::min_fdlimit $nfd
    IO::AIO::sendfile $ofh, $ifh, $offset, $count
    IO::AIO::fadvise $fh, $offset, $len, $advice
    IO::AIO::mmap $scalar, $length, $prot, $flags[, $fh[, $offset]]
    IO::AIO::munmap $scalar
    IO::AIO::mremap $scalar, $new_length, $flags[, $new_address]
    IO::AIO::madvise $scalar, $offset, $length, $advice
    IO::AIO::mprotect $scalar, $offset, $length, $protect
    IO::AIO::munlock $scalar, $offset = 0, $length = undef
    IO::AIO::munlockall
+   # stat extensions
+   $counter = IO::AIO::st_gen
+   $seconds = IO::AIO::st_atime, IO::AIO::st_mtime, IO::AIO::st_ctime, IO::AIO::st_btime
+   ($atime, $mtime, $ctime, $btime, ...) = IO::AIO::st_xtime
+   $nanoseconds = IO::AIO::st_atimensec, IO::AIO::st_mtimensec, IO::AIO::st_ctimensec, IO::AIO::st_btimensec
+   $seconds = IO::AIO::st_btimesec
+   ($atime, $mtime, $ctime, $btime, ...) = IO::AIO::st_xtimensec
+   # very much unportable syscalls
+   IO::AIO::accept4 $r_fh, $sockaddr, $sockaddr_len, $flags
+   IO::AIO::splice $r_fh, $r_off, $w_fh, $w_off, $length, $flags
+   IO::AIO::tee $r_fh, $w_fh, $length, $flags
+   $actual_size = IO::AIO::pipesize $r_fh[, $new_size]
+   ($rfh, $wfh) = IO::AIO::pipe2 [$flags]
+   $fh = IO::AIO::memfd_create $pathname[, $flags]
+   $fh = IO::AIO::eventfd [$initval, [$flags]]
+   $fh = IO::AIO::timerfd_create $clockid[, $flags]
+   ($cur_interval, $cur_value) = IO::AIO::timerfd_settime $fh, $flags, $new_interval, $nbw_value
+   ($cur_interval, $cur_value) = IO::AIO::timerfd_gettime $fh
 =head2 API NOTES
 All the C<aio_*> calls are more or less thin wrappers around the syscall
 with the same name (sans C<aio_>). The arguments are similar or identical,
 =item aio_stat  $fh_or_path, $callback->($status)
 =item aio_lstat $fh, $callback->($status)
-Works like perl's C<stat> or C<lstat> in void context. The callback will
+Works almost exactly like perl's C<stat> or C<lstat> in void context. The
-be called after the stat and the results will be available using C<stat _>
+callback will be called after the stat and the results will be available
-or C<-s _> etc...
+using C<stat _> or C<-s _> and other tests (with the exception of C<-B>
+and C<-T>).
 The pathname passed to C<aio_stat> must be absolute. See API NOTES, above,
 for an explanation.
 Currently, the stats are always 64-bit-stats, i.e. instead of returning an
 Works like perl's C<utime> function (including the special case of $atime
 and $mtime being undef). Fractional times are supported if the underlying
 syscalls support them.
-When called with a pathname, uses utimes(2) if available, otherwise
+When called with a pathname, uses utimensat(2) or utimes(2) if available,
-utime(2). If called on a file descriptor, uses futimes(2) if available,
+otherwise utime(2). If called on a file descriptor, uses futimens(2)
-otherwise returns ENOSYS, so this is not portable.
+or futimes(2) if available, otherwise returns ENOSYS, so this is not
+portable.
 Examples:
    # set atime and mtime to current time (basically touch(1)):
    aio_utime "path", undef, undef;
       aioreq_pri $pri;
       add $grp aio_stat $wd, sub {
          return $grp->result () if $_[0];
          my $now = time;
          my $hash1 = join ":", (stat _)[0,1,3,7,9];
+         my $rdxflags = READDIR_DIRS_FIRST;
+         if ((stat _)[3] < 2) {
+            # at least one non-POSIX filesystem exists
+            # that returns useful DT_type values: btrfs,
+            # so optimise for this here by requesting dents
+            $rdxflags |= READDIR_DENTS;
+         }
          # read the directory entries
          aioreq_pri $pri;
-         add $grp aio_readdirx $wd, READDIR_DIRS_FIRST, sub {
+         add $grp aio_readdirx $wd, $rdxflags, sub {
-            my $entries = shift
+            my ($entries, $flags) = @_
                or return $grp->result ();
+            if ($rdxflags & READDIR_DENTS) {
+               # if we requested type values, see if we can use them directly.
+               # if there were any DT_UNKNOWN entries then we assume we
+               # don't know. alternatively, we could assume that if we get
+               # one DT_DIR, then all directories are indeed marked with
+               # DT_DIR, but this seems not required for btrfs, and this
+               # is basically the "btrfs can't get it's act together" code
+               # branch.
+               unless ($flags & READDIR_FOUND_UNKNOWN) {
+                  # now we have valid DT_ information for all entries,
+                  # so use it as an optimisation without further stat's.
+                  # they must also all be at the beginning of @$entries
+                  # by now.
+                  my $dirs;
+                  if (@$entries) {
+                     for (0 .. $#$entries) {
+                        if ($entries->[$_][1] != DT_DIR) {
+                           # splice out directories
+                           $dirs = [splice @$entries, 0, $_];
+                           last;
+                        }
+                     }
+                     # if we didn't find any non-dir, then all entries are dirs
+                     unless ($dirs) {
+                        ($dirs, $entries) = ($entries, []);
+                     }
+                  } else {
+                     # directory is empty, so there are no sbdirs
+                     $dirs = [];
+                  }
+                  # either splice'd the directories out or the dir was empty.
+                  # convert dents to filenames
+                  $_ = $_->[0] for @$dirs;
+                  $_ = $_->[0] for @$entries;
+                  return $grp->result ($dirs, $entries);
+               }
+               # cannot use, so return to our old ways
+               # by pretending we only scanned for names.
+               $_ = $_->[0] for @$entries;
+            }
             # stat the dir another time
             aioreq_pri $pri;
             add $grp aio_stat $wd, sub {
                my $hash2 = join ":", (stat _)[0,1,3,7,9];
 So in general, you should only use these calls for things that do
 (filesystem) I/O, not for things that wait for other events (network,
 other processes), although if you are careful and know what you are doing,
 you still can.
-The following constants are available (missing ones are, as usual C<0>):
+The following constants are available and can be used for normal C<ioctl>
+and C<fcntl> as well (missing ones are, as usual C<0>):
 C<F_DUPFD_CLOEXEC>,
 C<F_OFD_GETLK>, C<F_OFD_SETLK>, C<F_OFD_GETLKW>,
 C<FIFREEZE>, C<FITHAW>, C<FITRIM>, C<FICLONE>, C<FICLONERANGE>, C<FIDEDUPERANGE>.
+C<F_ADD_SEALS>, C<F_GET_SEALS>, C<F_SEAL_SEAL>, C<F_SEAL_SHRINK>, C<F_SEAL_GROW> and
+C<F_SEAL_WRITE>.
 C<FS_IOC_GETFLAGS>, C<FS_IOC_SETFLAGS>, C<FS_IOC_GETVERSION>, C<FS_IOC_SETVERSION>,
 C<FS_IOC_FIEMAP>.
 C<FS_IOC_FSGETXATTR>, C<FS_IOC_FSSETXATTR>, C<FS_IOC_SET_ENCRYPTION_POLICY>,
    IO::AIO::mmap $data, -s $fh, IO::AIO::PROT_READ, IO::AIO::MAP_SHARED, $fh;
    aio_mlock $data; # mlock in background
 =item aio_mlockall $flags, $callback->($status)
-Calls the C<mlockall> function with the given C<$flags> (a combination of
+Calls the C<mlockall> function with the given C<$flags> (a
-C<IO::AIO::MCL_CURRENT> and C<IO::AIO::MCL_FUTURE>).
+combination of C<IO::AIO::MCL_CURRENT>, C<IO::AIO::MCL_FUTURE> and
+C<IO::AIO::MCL_ONFAULT>).
 On systems that do not implement C<mlockall>, this function returns C<-1>
-and sets errno to C<ENOSYS>.
+and sets errno to C<ENOSYS>. Similarly, flag combinations not supported
+by the system result in a return value of C<-1> with errno being set to
+C<EINVAL>.
 Note that the corresponding C<munlockall> is synchronous and is
 documented under L<MISCELLANEOUS FUNCTIONS>.
 Example: asynchronously lock all current and future pages into memory.
 C<aio_wd> callback, as future requests using the value will fail in the
 expected way.
 =item IO::AIO::CWD
-This is a compiletime constant (object) that represents the process
+This is a compile time constant (object) that represents the process
 current working directory.
 Specifying this object as working directory object for a pathname is as if
 the pathname would be specified directly, without a directory object. For
 example, these calls are functionally identical:
 The default value for the limit is C<0>, but note that setting a feeder
 automatically bumps it up to C<2>.
 =back
 =head2 SUPPORT FUNCTIONS
 =head3 EVENT PROCESSING AND EVENT LOOP INTEGRATION
 =over 4
 Strictly equivalent to:
    IO::AIO::poll_wait, IO::AIO::poll_cb
       while IO::AIO::nreqs;
+This function can be useful at program aborts, to make sure outstanding
+I/O has been done (C<IO::AIO> uses an C<END> block which already calls
+this function on normal exits), or when you are merely using C<IO::AIO>
+for its more advanced functions, rather than for async I/O, e.g.:
+   my ($dirs, $nondirs);
+   IO::AIO::aio_scandir "/tmp", 0, sub { ($dirs, $nondirs) = @_ };
+   IO::AIO::flush;
+   # $dirs, $nondirs are now set
 =item IO::AIO::max_poll_reqs $nreqs
 =item IO::AIO::max_poll_time $seconds
 These set the maximum number of requests (default C<0>, meaning infinity)
               poll => 'r', nice => 1,
               cb => &IO::AIO::poll_cb);
 =back
 =head3 CONTROLLING THE NUMBER OF THREADS
 =over
 =item IO::AIO::min_parallel $nthreads
 The default value for C<max_outstanding> is very large, so there is no
 practical limit on the number of outstanding requests.
 =back
 =head3 STATISTICAL INFORMATION
 =over
 =item IO::AIO::nreqs
 Returns the number of requests currently in the pending state (executed,
 but not yet processed by poll_cb).
 =back
 =head3 SUBSECOND STAT TIME ACCESS
 Both C<aio_stat>/C<aio_lstat> and perl's C<stat>/C<lstat> functions can
 generally find access/modification and change times with subsecond time
 not supported or could not be detected, a fractional part of C<0> is
 returned, so it is always safe to call these functions.
 =over 4
-=item $seconds = IO::AIO::st_atime, IO::AIO::st_mtime, IO::AIO::st_ctime
+=item $seconds = IO::AIO::st_atime, IO::AIO::st_mtime, IO::AIO::st_ctime, IO::AIO::st_btime
-Return the access, modication or change time, respectively, including
+Return the access, modication, change or birth time, respectively,
-fractional part. Due to the limited precision of floating point, the
+including fractional part. Due to the limited precision of floating point,
-accuracy on most platforms is only a bit better than milliseconds for
+the accuracy on most platforms is only a bit better than milliseconds
-times around now - see the I<nsec> function family, below, for full
+for times around now - see the I<nsec> function family, below, for full
 accuracy.
+File birth time is only available when the OS and perl support it (on
+FreeBSD and NetBSD at the time of this writing, although support is
+adaptive, so if your OS/perl gains support, IO::AIO can take advantage of
+it). On systems where it isn't available, C<0> is currently returned, but
+this might change to C<undef> in a future version.
-=item ($atime, $mtime, $ctime, ...) = IO::AIO::st_xtime
+=item ($atime, $mtime, $ctime, $btime, ...) = IO::AIO::st_xtime
-Returns access, modification and change time all in one go, and maybe more
+Returns access, modification, change and birth time all in one go, and
-times in the future version.
+maybe more times in the future version.
-=item $nanoseconds = IO::AIO::st_atimensec, IO::AIO::st_mtimensec, IO::AIO::st_ctimensec
+=item $nanoseconds = IO::AIO::st_atimensec, IO::AIO::st_mtimensec, IO::AIO::st_ctimensec, IO::AIO::st_btimensec
-Return the fractional access, modifcation or change time, in nanoseconds,
+Return the fractional access, modifcation, change or birth time, in nanoseconds,
 as an integer in the range C<0> to C<999999999>.
+Note that no accessors are provided for access, modification and
+change times - you need to get those from C<stat _> if required (C<int
+IO::AIO::st_atime> and so on will I<not> generally give you the correct
+value).
+=item $seconds = IO::AIO::st_btimesec
+The (integral) seconds part of the file birth time, if available.
-=item ($atime, $mtime, $ctime, ...) = IO::AIO::st_xtimensec
+=item ($atime, $mtime, $ctime, $btime, ...) = IO::AIO::st_xtimensec
-Like the functions above, but returns all three times in one go (and maybe
+Like the functions above, but returns all four times in one go (and maybe
 more in future versions).
+=item $counter = IO::AIO::st_gen
+Returns the generation counter (in practice this is just a random number)
+of the file. This is only available on platforms which have this member in
+their C<struct stat> (most BSDs at the time of this writing) and generally
+only to the root usert. If unsupported, C<0> is returned, but this might
+change to C<undef> in a future version.
 =back
 Example: print the high resolution modification time of F</etc>, using
 C<stat>, and C<IO::AIO::aio_stat>.
 Output of the awbove on my system, showing reduced and full accuracy:
    stat(/etc) mtime: 1534043702.020808
    aio_stat(/etc) mtime: 1534043702.020807792
 =head3 MISCELLANEOUS FUNCTIONS
 IO::AIO implements some functions that are useful when you want to use
 some "Advanced I/O" function not available to in Perl, without going the
 =over 4
 =item $numfd = IO::AIO::get_fdlimit
-This function is I<EXPERIMENTAL> and subject to change.
 Tries to find the current file descriptor limit and returns it, or
 C<undef> and sets C<$!> in case of an error. The limit is one larger than
 the highest valid file descriptor number.
 =item IO::AIO::min_fdlimit [$numfd]
-This function is I<EXPERIMENTAL> and subject to change.
 Try to increase the current file descriptor limit(s) to at least C<$numfd>
 by changing the soft or hard file descriptor resource limit. If C<$numfd>
 is missing, it will try to set a very high limit, although this is not
 recommended when you know the actual minimum that you require.
 implemented, but not supported and might go away in a future version.
 On systems where this call is not supported or is not emulated, this call
 returns falls and sets C<$!> to C<ENOSYS>.
+=item IO::AIO::mlockall $flags
+Calls the C<eio_mlockall_sync> function, which is like C<aio_mlockall>,
+but is blocking.
 =item IO::AIO::munlock $scalar, $offset = 0, $length = undef
 Calls the C<munlock> function, undoing the effects of a previous
 C<aio_mlock> call (see its description for details).
 Calls the C<munlockall> function.
 On systems that do not implement C<munlockall>, this function returns
 ENOSYS, otherwise the return value of C<munlockall>.
+=item $fh = IO::AIO::accept4 $r_fh, $sockaddr, $sockaddr_maxlen, $flags
+Uses the GNU/Linux C<accept4(2)> syscall, if available, to accept a socket
+and return the new file handle on success, or sets C<$!> and returns
+C<undef> on error.
+The remote name of the new socket will be stored in C<$sockaddr>, which
+will be extended to allow for at least C<$sockaddr_maxlen> octets. If the
+socket name does not fit into C<$sockaddr_maxlen> octets, this is signaled
+by returning a longer string in C<$sockaddr>, which might or might not be
+truncated.
+To accept name-less sockets, use C<undef> for C<$sockaddr> and C<0> for
+C<$sockaddr_maxlen>.
+The main reasons to use this syscall rather than portable C<accept(2)>
+are that you can specify C<SOCK_NONBLOCK> and/or C<SOCK_CLOEXEC>
+flags and you can accept name-less sockets by specifying C<0> for
+C<$sockaddr_maxlen>, which is sadly not possible with perl's interface to
+C<accept>.
 =item IO::AIO::splice $r_fh, $r_off, $w_fh, $w_off, $length, $flags
 Calls the GNU/Linux C<splice(2)> syscall, if available. If C<$r_off> or
 C<$w_off> are C<undef>, then C<NULL> is passed for these, otherwise they
 Example: create a pipe race-free w.r.t. threads and fork:
    my ($rfh, $wfh) = IO::AIO::pipe2 IO::AIO::O_CLOEXEC
       or die "pipe2: $!\n";
+=item $fh = IO::AIO::memfd_create $pathname[, $flags]
+This is a direct interface to the Linux L<memfd_create(2)> system
+call. The (unhelpful) default for C<$flags> is C<0>, but your default
+should be C<IO::AIO::MFD_CLOEXEC>.
+On success, the new memfd filehandle is returned, otherwise returns
+C<undef>. If the memfd_create syscall is missing, fails with C<ENOSYS>.
+Please refer to L<memfd_create(2)> for more info on this call.
+The following C<$flags> values are available: C<IO::AIO::MFD_CLOEXEC>,
+C<IO::AIO::MFD_ALLOW_SEALING> and C<IO::AIO::MFD_HUGETLB>.
+Example: create a new memfd.
+   my $fh = IO::AIO::memfd_create "somenameforprocfd", IO::AIO::MFD_CLOEXEC
+      or die "memfd_create: $!\n";
+=item $fh = IO::AIO::pidfd_open $pid[, $flags]
+This is an interface to the Linux L<pidfd_open(2)> system call. The
+default for C<$flags> is C<0>.
+On success, a new pidfd filehandle is returned (that is already set to
+close-on-exec), otherwise returns C<undef>. If the syscall is missing,
+fails with C<ENOSYS>.
+Example: open pid 6341 as pidfd.
+   my $fh = IO::AIO::pidfd_open 6341
+      or die "pidfd_open: $!\n";
+=item $status = IO::AIO::pidfd_send_signal $pidfh, $signal[, $siginfo[, $flags]]
+This is an interface to the Linux L<pidfd_send_signal> system call. The
+default for C<$siginfo> is C<undef> and the default for C<$flags> is C<0>.
+Returns the system call status.  If the syscall is missing, fails with
+C<ENOSYS>.
+When specified, C<$siginfo> must be a reference to a hash with one or more
+of the following members:
+=over
+=item code - the C<si_code> member
+=item pid - the C<si_pid> member
+=item uid - the C<si_uid> member
+=item value_int - the C<si_value.sival_int> member
+=item value_ptr - the C<si_value.sival_ptr> member, specified as an integer
+=back
+Example: send a SIGKILL to the specified process.
+   my $status = IO::AIO::pidfd_send_signal $pidfh, 9, undef
+      and die "pidfd_send_signal: $!\n";
+Example: send a SIGKILL to the specified process with extra data.
+   my $status = IO::AIO::pidfd_send_signal $pidfh, 9,  { code => -1, value_int => 7 }
+      and die "pidfd_send_signal: $!\n";
+=item $fh = IO::AIO::pidfd_getfd $pidfh, $targetfd[, $flags]
+This is an interface to the Linux L<pidfd_getfd> system call. The default
+for C<$flags> is C<0>.
+On success, returns a dup'ed copy of the target file descriptor (specified
+as an integer) returned (that is already set to close-on-exec), otherwise
+returns C<undef>. If the syscall is missing, fails with C<ENOSYS>.
+Example: get a copy of standard error of another process and print soemthing to it.
+   my $errfh = IO::AIO::pidfd_getfd $pidfh, 2
+      or die "pidfd_getfd: $!\n";
+   print $errfh "stderr\n";
 =item $fh = IO::AIO::eventfd [$initval, [$flags]]
 This is a direct interface to the Linux L<eventfd(2)> system call. The
 (unhelpful) defaults for C<$initval> and C<$flags> are C<0> for both.
 The following symbol flag values are available: C<IO::AIO::EFD_CLOEXEC>,
 C<IO::AIO::EFD_NONBLOCK> and C<IO::AIO::EFD_SEMAPHORE> (Linux 2.6.30).
 Example: create a new eventfd filehandle:
-   $fh = IO::AIO::eventfd 0, IO::AIO::O_CLOEXEC
+   $fh = IO::AIO::eventfd 0, IO::AIO::EFD_CLOEXEC
       or die "eventfd: $!\n";
 =item $fh = IO::AIO::timerfd_create $clockid[, $flags]
-This is a direct interface to the Linux L<timerfd_create(2)> system call. The
+This is a direct interface to the Linux L<timerfd_create(2)> system
-(unhelpful) default for C<$flags> is C<0>.
+call. The (unhelpful) default for C<$flags> is C<0>, but your default
+should be C<IO::AIO::TFD_CLOEXEC>.
 On success, the new timerfd filehandle is returned, otherwise returns
-C<undef>. If the eventfd syscall is missing, fails with C<ENOSYS>.
+C<undef>. If the timerfd_create syscall is missing, fails with C<ENOSYS>.
 Please refer to L<timerfd_create(2)> for more info on this call.
 The following C<$clockid> values are
 available: C<IO::AIO::CLOCK_REALTIME>, C<IO::AIO::CLOCK_MONOTONIC>
 known issue, rather than a bug.
 =head1 SEE ALSO
 L<AnyEvent::AIO> for easy integration into event loops, L<Coro::AIO> for a
-more natural syntax.
+more natural syntax and L<IO::FDPass> for file descriptor passing.
 =head1 AUTHOR
  Marc Lehmann <schmorp@schmorp.de>
  http://home.schmorp.de/

Diff Legend

-–
+Removed lines
-+
+Added lines
-<
+Changed lines
->
+Changed lines

Comparing IO-AIO/AIO.pm (file contents): Revision 1.290 by root, Sun Aug 12 06:03:20 2018 UTC vs. Revision 1.309 by root, Tue Dec 29 15:20:12 2020 UTC

Diff Legend

Comparing IO-AIO/AIO.pm (file contents):
Revision 1.290 by root, Sun Aug 12 06:03:20 2018 UTC vs.
Revision 1.309 by root, Tue Dec 29 15:20:12 2020 UTC