put perl5240delta into place

[perl5.git] / pod / perldelta.pod

=encoding utf8

=head1 NAME

perldelta - what is new for perl v5.24.0

=head1 DESCRIPTION

This document describes the differences between the 5.22.0 release and the
5.24.0 release.

=head1 Core Enhancements

=head2 Postfix dereferencing is no longer experimental

Using the C<postderef> and C<postderef_qq> features no longer emits a
warning. Existing code that disables the C<experimental::postderef> warning
category that they previously used will continue to work. The C<postderef>
feature has no effect; all Perl code can use postfix dereferencing,
regardless of what feature declarations are in scope. The C<5.24> feature
bundle now includes the C<postderef_qq> feature.

=head2 Unicode 8.0 is now supported

For details on what is in this release, see
L<http://www.unicode.org/versions/Unicode8.0.0/>.

=head2 perl will now croak when closing an in-place output file fails

Until now, failure to close the output file for an in-place edit was not
detected, meaning that the input file could be clobbered without the edit being
successfully completed.  Now, when the output file cannot be closed
successfully, an exception is raised.

=head2 New C<\b{lb}> boundary in regular expressions

C<lb> stands for Line Break.  It is a Unicode property
that determines where a line of text is suitable to break (typically so
that it can be output without overflowing the available horizontal
space).  This capability has long been furnished by the
L<Unicode::LineBreak> module, but now a light-weight, non-customizable
version that is suitable for many purposes is in core Perl.

=head2 C<qr/(?[ ])/> now works in UTF-8 locales

L<Extended Bracketed Character Classes|perlrecharclass/Extended Bracketed Character Classes>
now will successfully compile when S<C<use locale>> is in effect.  The compiled
pattern will use standard Unicode rules.  If the runtime locale is not a
UTF-8 one, a warning is raised and standard Unicode rules are used
anyway.  No tainting is done since the outcome does not actually depend
on the locale.

=head2 Integer shift (C<< << >> and C<< >> >>) now more explicitly defined

Negative shifts are reverse shifts: left shift becomes right shift,
and right shift becomes left shift.

Shifting by the number of bits in a native integer (or more) is zero,
except when the "overshift" is right shifting a negative value under
C<use integer>, in which case the result is -1 (arithmetic shift).

Until now negative shifting and overshifting have been undefined
because they have relied on whatever the C implementation happens
to do.  For example, for the overshift a common C behavior is
"modulo shift":

  1 >> 64 == 1 >> (64 % 64) == 1 >> 0 == 1  # Common C behavior.

  # And the same for <<, while Perl now produces 0 for both.

Now these behaviors are well-defined under Perl, regardless of what
the underlying C implementation does.  Note, however, that you cannot
escape the native integer width, you need to know how far left you
can go.  You can use for example:

  use Config;
  my $wordbits = $Config{uvsize} * 8;  # Or $Config{uvsize} << 3.

If you need a more bits on the left shift, you can use for example
the C<bigint> pragma, or the C<Bit::Vector> module from CPAN.

=head2 printf and sprintf now allow reordered precision arguments

That is, C<< sprintf '|%.*2$|', 2, 3 >> now returns C<|002|>. This extends
the existing reordering mechanism (which allows reordering for arguments
that are used as format fields, widths, and vector separators).

=head2 More fields provided to C<sigaction> callback with C<SA_SIGINFO>

When passing the C<SA_SIGINFO> flag to L<sigaction|POSIX/sigaction>, the
C<errno>, C<status>, C<uid>, C<pid>, C<addr> and C<band> fields are now
included in the hash passed to the handler, if supported by the
platform.

=head1 Security

=head2 Set proper umask before calling C<mkstemp(3)>

In 5.22 perl started setting umask to 0600 before calling C<mkstemp(3)>
and restoring it afterwards. This wrongfully tells open(2) to strip
the owner read and write bits from the given mode before applying it,
rather than the intended negation of leaving only those bits in place.

Systems that use mode 0666 in C<mkstemp(3)> (like old versions of
glibc) create a file with permissions 0066, leaving world read and
write permissions regardless of current umask.

This has been fixed by using umask 0177 instead. [perl #127322]

=head2 fix out of boundary access in Win32 path handling

This is CVE-2015-8608.  For more information see
L<[perl #126755]|https://rt.perl.org/Ticket/Display.html?id=126755>

=head2 fix loss of taint in canonpath

This is CVE-2015-8607.  For more information see
L<[perl #126862]|https://rt.perl.org/Ticket/Display.html?id=126862>

=head2 Avoid accessing uninitialized memory in win32 C<crypt()>

Added validation that will detect both a short salt and invalid characters
in the salt.
L<[perl #126922]|https://rt.perl.org/Ticket/Display.html?id=126922>

=head2 Remove duplicate environment variables from C<environ>

Previously, if an environment variable appeared more than once in
C<environ[]>, C<%ENV> would contain the last entry for that name,
while a typical C<getenv()> would return the first entry. We now
make sure C<%ENV> contains the same as what C<getenv> returns.

Second, we remove duplicates from C<environ[]>, so if a setting
with that name is set in C<%ENV> we won't pass an unsafe value
to a child process.

[CVE-2016-2381]

=head1 Incompatible Changes

=head2 The C<autoderef> feature has been removed

The experimental C<autoderef> feature (which allowed calling C<push>,
C<pop>, C<shift>, C<unshift>, C<splice>, C<keys>, C<values>, and C<each> on
a scalar argument) has been deemed unsuccessful. It has now been removed;
trying to use the feature (or to disable the C<experimental::autoderef>
warning it previously triggered) now yields an exception.

=head2 Lexical $_ has been removed

C<my $_> was introduced in Perl 5.10, and subsequently caused much confusion
with no obvious solution.  In Perl 5.18.0, it was made experimental on the
theory that it would either be removed or redesigned in a less confusing (but
backward-incompatible) way.  Over the following years, no alternatives were
proposed.  The feature has now been removed and will fail to compile.

=head2 C<qr/\b{wb}/> is now tailored to Perl expectations

This is now more suited to be a drop-in replacement for plain C<\b>, but
giving better results for parsing natural language.  Previously it
strictly followed the current Unicode rules which calls for it to match
between each white space character.  Now it doesn't generally match
within spans of white space, behaving like C<\b> does.  See
L<perlrebackslash/\b{wb}>

=head2 Regular expression compilation errors

Some regular expression patterns that had runtime errors now
don't compile at all.

Almost all Unicode properties using the C<\p{}> and C<\P{}> regular
expression pattern constructs are now checked for validity at pattern
compilation time, and invalid ones will cause the program to not
compile.  In earlier releases, this check was often deferred until run
time.  Whenever an error check is moved from run- to compile time,
erroneous code is caught 100% of the time, whereas before it would only
get caught if and when the offending portion actually gets executed,
which for unreachable code might be never.

=head2 C<qr/\N{}/> now disallowed under C<use re "strict">

An empty C<\N{}> makes no sense, but for backwards compatibility is
silently accepted as doing nothing.  But now this is a fatal error under
the experimental feature L<re/'strict' mode>.

=head2 Nested declarations are now disallowed

A C<my>, C<our>, or C<state> declaration is no longer allowed inside
of another C<my>, C<our>, or C<state> declaration.

For example, these are now fatal:

   my ($x, my($y));
   our (my $x);

L<[perl #125587]|https://rt.perl.org/Ticket/Display.html?id=125587>

L<[perl #121058]|https://rt.perl.org/Ticket/Display.html?id=121058>

=head2 The C</\C/> character class has been removed.

This regular expression character class was deprecated in v5.20.0 and has
produced a deprecation warning since v5.22.0. It is now a compile-time
error. If you need to examine the individual bytes that make up a
UTF8-encoded character, then use C<utf8::encode()> on the string (or a
copy) first.

=head2 C<chdir('')> no longer chdirs home

Using C<chdir('')> or C<chdir(undef)> to chdir home has been deprecated since
perl v5.8, and will now fail.  Use C<chdir()> instead.

=head2 ASCII characters in variable names must now be all visible

It was legal until now on ASCII platforms for variable names to contain
non-graphical ASCII control characters (ordinals 0 through 31, and 127,
which are the C0 controls and C<DELETE>).  This usage has been
deprecated since v5.20, and as of now causes a syntax error.  The
variables these names referred to are special, reserved by Perl for
whatever use it may choose, now, or in the future.  Each such variable
has an alternative way of spelling it.  Instead of the single
non-graphic control character, a two character sequence beginning with a
caret is used, like C<$^]> and C<${^GLOBAL_PHASE}>.  Details are at
L<perlvar>.   It remains legal, though unwise and deprecated (raising a
deprecation warning), to use certain non-graphic non-ASCII characters in
variables names when not under S<C<use utf8>>.  No code should do this,
as all such variables are reserved by Perl, and Perl doesn't currently
define any of them (but could at any time, without notice).

=head2 An off by one issue in C<$Carp::MaxArgNums> has been fixed

C<$Carp::MaxArgNums> is supposed to be the number of arguments to display.
Prior to this version, it was instead showing C<$Carp::MaxArgNums> + 1 arguments,
contrary to the documentation.

=head2 Only blanks and tabs are now allowed within C<[...]> within C<(?[...])>.

The experimental Extended Bracketed Character Classes can contain regular
bracketed character classes within them.  These differ from regular ones in
that white space is generally ignored, unless escaped by preceding it with a
backslash.  The white space that is ignored is now limited to just tab C<\t>
and SPACE characters.  Previously, it was any white space.  See
L<perlrecharclass/Extended Bracketed Character Classes>.

=head1 Deprecations

=head2 Using code points above the platform's C<IV_MAX> is now deprecated

Unicode defines code points in the range C<0..0x10FFFF>.  Some standards
at one time defined them up to 2**31 - 1, but Perl has allowed them to
be as high as anything that will fit in a word on the platform being
used.  However, use of those above the platform's C<IV_MAX> is broken in
some constructs, notably C<tr///>, regular expression patterns involving
quantifiers, and in some arithmetic and comparison operations, such as
being the upper limit of a loop.  Now the use of such code points raises
a deprecation warning, unless that warning category is turned off.
C<IV_MAX> is typically 2**31 -1 on 32-bit platforms, and 2**63-1 on
64-bit ones.

=head2 Doing bitwise operations on strings containing code points above
0xFF is deprecated

The string bitwise operators treat their operands as strings of bytes,
and values beyond 0xFF are nonsensical in this context.  To operate on
encoded bytes, first encode the strings.  To operate on code points'
numeric values, use C<split> and C<map ord>.  In the future, this
warning will be replaced by an exception.

=head2 sysread(), syswrite(), recv() and send() are deprecated on
:utf8 handles

The sysread(), recv(), syswrite() and send() operators
are deprecated on handles that have the C<:utf8> layer, either
explicitly, or implicitly, eg., with the C<:encoding(UTF-16LE)> layer.

Both sysread() and recv() currently use only the C<:utf8> flag for the
stream, ignoring the actual layers.  Since sysread() and recv() do no
UTF-8 validation they can end up creating invalidly encoded scalars.

Similarly, syswrite() and send() use only the C<:utf8> flag, otherwise
ignoring any layers.  If the flag is set, both write the value UTF-8
encoded, even if the layer is some different encoding, such as the
example above.

Ideally, all of these operators would completely ignore the C<:utf8>
state, working only with bytes, but this would result in silently
breaking existing code.  To avoid this a future version of perl will
throw an exception when any of sysread(), recv(), syswrite() or send()
are called on handle with the C<:utf8> layer.

=head1 Performance Enhancements

=over 4

=item *

The overhead of scope entry and exit has been considerably reduced, so
for example subroutine calls, loops and basic blocks are all faster now.
This empty function call now takes about a third less time to execute:

    sub f{} f();

=item *

Many languages, such as Chinese, are caseless.  Perl now knows about
most modern commercially important ones, and skips much of the work when
a program tries to change case in them (like C<ucfirst()>) or match
caselessly (C<qr//i>).  This will speed up a program, such as a web
server, that can operate on multiple languages, while operating on a
caseless one.

=item *

C</fixed-substr/> has been made much faster.

On platforms with a libc memchr() implementation which makes good use of
underlying hardware support, patterns which include fixed substrings will now
often be much faster; for example with glibc on a recent x86_64 CPU, this:

    $s = "a" x 1000 . "wxyz";
    $s =~ /wxyz/ for 1..30000

is now about 7 times faster.  On systems with slow memchr(), e.g. 32-bit ARM
Raspberry Pi, there will be a small or little speedup.  Conversely, some
pathological cases, such as C<"ab" x 1000 =~ /aa/> will be slower now; up to 3
times slower on the rPi, 1.5x slower on x86_64.

=item *

Faster addition, subtraction and multiplication.

Since 5.8.0, arithmetic became slower due to the need to support
64-bit integers. To deal with 64-bit integers, a lot more corner
cases need to be checked, which adds time. We now detect common
cases where there is no need to check for those corner cases,
and special-case them.

=item *

Preincrement, predecrement, postincrement, and postdecrement have been
made faster by internally splitting the functions which handled multiple
cases into different functions.

=item *

Creating Perl debugger data structures (see L<perldebguts/"Debugger Internals">)
for XSUBs and const subs has been removed.  This removed one glob/scalar combo
for each unique C<.c> file that XSUBs and const subs came from.  On startup
(C<perl -e"0">) about half a dozen glob/scalar debugger combos were created.
Loading XS modules created more glob/scalar combos.  These things were created
regardless if the perl debugger was being used or not, unlike for pure perl
subs, and ignores that the perl debugger can not debug C code.

=item *

On Win32, C<stat>ing or C<-X>ing a path, if the file or directory does not
exist, is now 3.5x faster than before.

=item *

Single arguments in list assign are now slightly faster:

  ($x) = (...);
  (...) = ($x);

=item *

Less peak memory is now used when compiling regular expression patterns.

=back

=head1 Modules and Pragmata

=head2 Updated Modules and Pragmata

XXX: todo

=head1 Documentation

=head2 Changes to Existing Documentation

=head3 L<perlapi>

=over 4

=item *

The process of using undocumented globals has been documented, namely, that one
should send email to L<perl5-porters@perl.org|mailto:perl5-porters@perl.org>
first to get the go-ahead for documenting and using an undocumented function or
global variable.

=back

=head3 L<perlcall>

=over 4

=item *

A number of cleanups have been made to perlcall, including:

=over 4

=item *

use EXTEND(SP, n) and PUSHs() instead of XPUSHs() where applicable
and update prose to match

=item *

add POPu, POPul and POPpbytex to the "complete list of POP macros"
and clarify the documentation for some of the existing entries, and
a note about side-effects

=item *

add API documentation for POPu and POPul

=item *

use ERRSV more efficiently

=item *

approaches to thread-safety storage of SVs.

=back

=back

=head3 L<perlfunc>

=over 4

=item *

The documentation of C<hex> has been revised to clarify valid inputs.

=item *

Better explain meaning of negative PIDs in C<waitpid>.
L<[perl #127080]|https://rt.perl.org/Ticket/Display.html?id=127080>

=item *

General cleanup: there's more consistency now (in POD usage, grammar, code
examples), better practices in code examples (use of C<my>, removal of bareword
filehandles, dropped usage of C<&> when calling subroutines, ...), etc.

=back

=head3 L<perlguts>

=over 4

=item *

A new section has been added, L<perlguts/"Dynamic Scope and the Context
Stack">, which explains how the perl context stack works.

=back

=head3 L<perlmodlib>

=over 4

=item *

We now recommend contacting the module-authors list or PAUSE in seeking
guidance on the naming of modules.

=back

=head3 L<perlop>

=over 4

=item *

The documentation of C<qx//> now describes how C<$?> is affected.

=back

=head3 L<perlpolicy>

=over 4

=item *

This note has been added to perlpolicy:

  While civility is required, kindness is encouraged; if you have any doubt
  about whether you are being civil, simply ask yourself, "Am I being kind?"
  and aspire to that.

=back

=head3 L<perlreftut>

=over 4

=item *

Fix some examples to be L<strict> clean.

=back

=head3 L<perlrebackslash>

=over 4

=item *

Clarify that in languages like Japanese and Thai, dictionary lookup
is required to determine word boundaries.

=back

=head3 L<perlsub>

=over 4

=item *

Updated to note that anonymous subroutines can have signatures.

=back

=head3 L<perlsyn>

=over 4

=item *

Fixed a broken example where C<=> was used instead of
C<==> in conditional in do/while example.

=back

=head3 L<perltie>

=over 4

=item *

The usage of C<FIRSTKEY> and C<NEXTKEY> has been clarified.

=back

=head3 L<perlunicode>

=over 4

=item *

Discourage use of 'In' prefix for Unicode Block property.

=back

=head3 L<perlvar>

=over 4

=item *

The documentation of C<$@> was reworded to clarify that it is not just for
syntax errors in C<eval>.
L<[perl #124034]|https://rt.perl.org/Ticket/Display.html?id=124034>

=item *

The specific true value of C<$!{E...}> is now documented, noting that it is
subject to change and not guaranteed.

=item *

Use of C<$OLD_PERL_VERSION> is now discouraged.

=back

=head3 L<perlxs>

=over 4

=item *

The documentation of C<PROTOTYPES> has been corrected; they are I<disabled>
by default, not I<enabled>.

=back

=head1 Diagnostics

The following additions or changes have been made to diagnostic output,
including warnings and fatal error messages.  For the complete list of
diagnostic messages, see L<perldiag>.

=head2 New Diagnostics

=head3 New Errors

=over 4

=item *

L<%s must not be a named sequence in transliteration operator|perldiag/"%s must not be a named sequence in transliteration operator">

(F) Transliteration (C<tr///> and C<y///>) transliterates individual
characters.  But a named sequence by definition is more than an
individual charater, and hence doing this operation on it doesn't make
sense.

=item *

L<Can't find Unicode property definition "%s" in regex;|perldiag/"Can't find Unicode property definition "%s" in regex; marked by <-- HERE in m/%s/">

=item *

L<Can't redeclare "%s" in "%s"|perldiag/"Can't redeclare "%s" in "%s"">

(F) A "my", "our" or "state" declaration was found within another declaration,
such as C<my ($x, my($y), $z)> or C<our (my $x)>.

=item *

L<Character following \p must be '{' or a single-character Unicode property name in regex;|perldiag/"Character following \%c must be '{' or a single-character Unicode property name in regex; marked by <-- HERE in m/%s/">

=item *

L<Empty \%c in regex; marked by E<lt>-- HERE in mE<sol>%sE<sol>
|perldiag/"Empty \%c in regex; marked by <-- HERE in mE<sol>%sE<sol>">

=item *

L<Illegal user-defined property name|perldiag/"Illegal user-defined property name">

=item *

L<Invalid number '%s' for -C option.|perldiag/"Invalid number '%s' for -C option.">

(F) You supplied a number to the -C option that either has extra leading
zeroes or overflows perl's unsigned integer representation.

=item *

L<<< Sequence (?... not terminated in regex; marked by S<<-- HERE> in mE<sol>%sE<sol>|perldiag/"Sequence (?... not terminated in regex; marked by <-- HERE in mE<sol>%sE<sol>" >>>

=item *

L<<< Sequence (?PE<lt>... not terminated in regex; marked by E<lt>-- HERE in mE<sol>%sE<sol>
|perldiag/"Sequence (?PE<lt>... not terminated in regex; marked by <-- HERE in mE<sol>%sE<sol>" >>>

=item *

L<Sequence (?PE<gt>... not terminated in regex; marked by E<lt>-- HERE in mE<sol>%sE<sol>
|perldiag/"Sequence (?PE<gt>... not terminated in regex; marked by <-- HERE in mE<sol>%sE<sol>">

=back

=head3 New Warnings

=over 4

=item *

L<Assuming NOT a POSIX class since %s in regex; marked by E<lt>-- HERE in mE<sol>%sE<sol>|
perldiag/Assuming NOT a POSIX class since %s in regex; marked by <-- HERE in mE<sol>%sE<sol>>

=item *

L<%s() is deprecated on :utf8 handles|perldiag/"%s() is deprecated on :utf8 handles">

(W deprecated) The sysread(), recv(), syswrite() and send() operators
are deprecated on handles that have the C<:utf8> layer, either
explicitly, or implicitly, eg., with the C<:encoding(UTF-16LE)> layer.

Both sysread() and recv() currently use only the C<:utf8> flag for the
stream, ignoring the actual layers.  Since sysread() and recv() do no
UTF-8 validation they can end up creating invalidly encoded scalars.

Similarly, syswrite() and send() use only the C<:utf8> flag, otherwise
ignoring any layers.  If the flag is set, both write the value UTF-8
encoded, even if the layer is some different encoding, such as the
example above.

Ideally, all of these operators would completely ignore the C<:utf8>
state, working only with bytes, but this would result in silently
breaking existing code.  To avoid this a future version of perl will
throw an exception when any of sysread(), recv(), syswrite() or send()
are called on handle with the C<:utf8> layer.

=back

=head2 Changes to Existing Diagnostics

=over 4

=item *

Accessing the C<IO> part of a glob as C<FILEHANDLE> instead of C<IO> is no
longer deprecated.  It is discouraged to encourage uniformity (so that, for
example, one can grep more easily) but it will not be removed.
L<[perl #127060]|https://rt.perl.org/Ticket/Display.html?id=127060>

=item *

The diagnostic C<< Hexadecimal float: internal error >> has been changed to
C<< Hexadecimal float: internal error (%s) >> to include more information.

=item *

L<Can't modify non-lvalue subroutine call of &%s|perldiag/"Can't modify non-lvalue subroutine call of &%s">

This error now reports the name of the non-lvalue subroutine you attempted to
use as an lvalue.

=item *

When running out of memory during an attempt the increase the stack
size, previously, perl would die using the cryptic message
C<< panic: av_extend_guts() negative count (-9223372036854775681) >>.
This has been fixed to show the prettier message:
L<< Out of memory during stack extend|perldiag/"Out of memory during %s extend" >>

=back

=head1 Configuration and Compilation

=over 4

=item *

C<Configure> now acts as if the C<-O> option is always passed, allowing command
line options to override saved configuration.  This should eliminate confusion
when command line options are ignored for no obvious reason.  C<-O> is now
permitted, but ignored.

=item *

Bison 3.0 is now supported.

=item *

F<Configure> no longer probes for F<libnm> by default.  Originally
this was the "New Math" library, but the name has been re-used by the
GNOME NetworkManager.
L<[perl #127131]|https://rt.perl.org/Ticket/Display.html?id=127131>

=item *

Added F<Configure> probes for C<newlocale>, C<freelocale>, and C<uselocale>.

=item *

C<< PPPort.so/PPPort.dll >> no longer get installed, as they are
not used by C<< PPPort.pm >>, only by its test files.

=item *

It is now possible to specify which compilation date to show on
C<< perl -V >> output, by setting the macro C<< PERL_BUILD_DATE >>.

=item *

Using the C<NO_HASH_SEED> define in combination with the default hash algorithm
C<PERL_HASH_FUNC_ONE_AT_A_TIME_HARD> resulted in a fatal error while compiling
the interpreter, since Perl 5.17.10.  This has been fixed.

=item *

F<Configure> should handle spaces in paths a little better.

=item *

No longer generate EBCDIC POSIX-BC tables.  We don't believe anyone is
using Perl and POSIX-BC at this time, and by not generating these tables
it saves time during development, and makes the resulting tar ball smaller.

=item *

The GNU Make makefile for Win32 now supports parallel builds.  [perl #126632]

=item *

You can now build perl with MSVC++ on Win32 using GNU Make.  [perl #126632]

=item *

The Win32 miniperl now has a real C<getcwd> which increases build performance
resulting in C<getcwd()> being 605x faster in Win32 miniperl.

=back

=head1 Testing

=over 4

=item *

A new test (F<t/op/aassign.t>) has been added to test the list assignment operator
C<OP_AASSIGN>.

=item *

Parallel building has been added to the dmake C<makefile.mk> makefile. All
Win32 compilers are supported.

=back

=head1 Platform Support

=head2 Platform-Specific Notes

=over 4

=item AmigaOS

The AmigaOS port has been reintegrated into the main tree, based off of
Perl 5.22.1.

=item Cygwin

Tests are more robust against unusual cygdrive prefixes.
L<[perl #126834]|https://rt.perl.org/Ticket/Display.html?id=126834>

=item EBCDIC

=over 4

=item UTF-EBCDIC extended

UTF-EBCDIC is like UTF-8, but for EBCDIC platforms.  It now has been
extended so that it can represent code points up to 2 ** 64 - 1 on
platforms with 64-bit words.  This brings it into parity with UTF-8.
This enhancement requires an incompatible change to the representation
of code points in the range 2 ** 30 to 2 ** 31 -1 (the latter was the
previous maximum representable code point).  This means that a file that
contains one of these code points, written out with previous versions of
perl cannot be read in, without conversion, by a perl containing this
change.  We do not believe any such files are in existence, but if you
do have one, submit a ticket at L<perlbug@perl.org|mailto:perlbug@perl.org>,
and we will write a conversion script for you.

=item EBCDIC C<cmp()> and C<sort()> fixed for UTF-EBCDIC strings

Comparing two strings that were both encoded in UTF-8 (or more
precisely, UTF-EBCDIC) did not work properly until now.  Since C<sort()>
uses C<cmp()>, this fixes that as well.

=item EBCDIC C<tr///> and C<y///> fixed for C<\N{}>, and C<S<use utf8>> ranges

Perl v5.22 introduced the concept of portable ranges to regular
expression patterns.  A portable range matches the same set of
characters no matter what platform is being run on.  This concept is now
extended to C<tr///>.  See
C<L<trE<sol>E<sol>E<sol>|perlop/trE<sol>SEARCHLISTE<sol>REPLACEMENTLISTE<sol>cdsr>>.

There were also some problems with these operations under S<C<use
utf8>>, which are now fixed

=back

=item FreeBSD

=over

=item *

Use the C<fdclose()> function from FreeBSD if it is available.
L<[perl #126847]|https://rt.perl.org/Ticket/Display.html?id=126847>

=back

=item IRIX

=over

=item *

Under some circumstances IRIX stdio fgetc() and fread() set the errno to
C<ENOENT>, which made no sense according to either IRIX or POSIX docs.  Errno
is now cleared in such cases.
L<[perl #123977]|https://rt.perl.org/Ticket/Display.html?id=123977>

=item *

Problems when multiplying long doubles by infinity have been fixed.
L<[perl #126396]|https://rt.perl.org/Ticket/Display.html?id=126396>

=back

=item MacOS X

=over

=item *

Until now OS X builds of perl have specified a link target of 10.3 (Panther,
2003) but have not specified a compiler target.  From now on, builds of perl on
OS X 10.6 or later (Snow Leopard, 2008) by default capture the current OS X
version and specify that as the explicit build target in both compiler and
linker flags, thus preserving binary compatibility for extensions built later
regardless of changes in OS X, SDK, or compiler and linker versions.  To
override the default value used in the build and preserved in the flags,
specify C<export MACOSX_DEPLOYMENT_TARGET=10.N> before configuring and building
perl, where 10.N is the version of OS X you wish to target.  In OS X 10.5 or
earlier there is no change to the behavior present when those systems were
current; the link target is still OS X 10.3 and there is no explicit compiler
target.

=item *

Builds with both -DDEBUGGING and threading enabled would fail with a
"panic: free from wrong pool" error when built or tested from Terminal
on OS X.  This was caused by perl's internal management of the
environment conflicting with an atfork handler using the libc
setenv() function to update the environment.

Perl now uses setenv()/unsetenv() to update the environment on OS X.
L<[perl #126240]|https://rt.perl.org/Ticket/Display.html?id=126240>

=back

=item Solaris

All Solaris now builds shared libperl.

Solaris and variants like OpenIndiana now always build with the shared
Perl library (Configure -Duseshrplib).  This was required for the
OpenIndiana builds, but this has also been the setting for Oracle/Sun
Perl builds for several years.

=item Tru64

Workaround where Tru64 balks when prototypes are listed as
C<< PERL_STATIC_INLINE >>, but where the test is build with
C<< -DPERL_NO_INLINE_FUNCTIONS >>.

=item VMS

=over

=item *

On VMS, the math function prototypes in C<math.h> are now visible under C++.
Now building the POSIX extension with C++ will no longer crash.

=item *

VMS has had C<setenv/unsetenv> since v7.0 (released in 1996), C<Perl_vmssetenv>
now always uses C<setenv/unsetenv>.

=item *

Perl now implements its own C<killpg> by scanning for processes in the
specified process group, which may not mean exactly the same thing as a Unix
process group, but allows us to send a signal to a parent (or master) process
and all of its sub-processes.  At the perl level, this means we can now send a
negative pid like so:

    kill SIGKILL, -$pid;

to signal all processes in the same group as C<$pid>.

=item *

For those C<%ENV> elements based on the CRTL environ array, we've always
preserved case when setting them but did look-ups only after upcasing the
key first, which made lower- or mixed-case entries go missing. This problem
has been corrected by making C<%ENV> elements derived from the environ array
case-sensitive on look-up as well as case-preserving on store.

=item *

Environment look-ups for C<PERL5LIB> and C<PERLLIB> previously only
considered logical names, but now consider all sources of C<%ENV> as
determined by C<PERL_ENV_TABLES> and as documented in L<perlvms/%ENV>.

=item *

The minimum supported version of VMS is now v7.3-2, released in 2003.  As a
side effect of this change, VAX is no longer supported as the terminal
release of OpenVMS VAX was v7.3 in 2001.

=back

=item Win32

=over

=item *

A new build option C<USE_NO_REGISTRY> has been added to the makefiles.  This
option is off by default, meaning the default is to do Windows registry
lookups.  This option stops Perl from looking inside the registry for anything.
For what values are looked up in the registry see L<perlwin32>.  Internally, in
C, the name of this option is C<WIN32_NO_REGISTRY>.

=item *

The behavior of Perl using C<HKEY_CURRENT_USER\Software\Perl> and
C<HKEY_LOCAL_MACHINE\Software\Perl> to lookup certain values, including C<%ENV>
vars starting with C<PERL> has changed.  Previously, the 2 keys were checked
for entries at all times through Perl processes life time even if they did not
exist.  For performance reasons, now, if the root key (i.e.
C<HKEY_CURRENT_USER\Software\Perl> or C<HKEY_LOCAL_MACHINE\Software\Perl>) does
not exist at process start time, it will not be checked again for C<%ENV>
override entries for the remainder of the Perl processes life.  This more
closely matches Unix behaviour in that the environment is copied or inherited
on startup and changing the variable in the parent process or another process
or editing F<.bashrc> will not change the environmental variable in other
existing, running, processes.

=item *

One glob fetch was removed for each C<-X> or C<stat> call whether done from
Perl code or internally from Perl's C code.  The glob being looked up was
C<${^WIN32_SLOPPY_STAT}> which is a special variable.  This makes C<-X> and
C<stat> slightly faster.

=item *

During miniperl's process startup, during the build process, 4 to 8 IO calls
related to the process starting F<.pl> and the F<buildcustomize.pl> file were
removed from the code opening and executing the first 1 or 2 F<.pl> files.

=item *

Builds using Microsoft Visual C++ 2003 and earlier no longer produce
an "INTERNAL COMPILER ERROR" message.  [perl #126045]

=item *

Visual C++ 2013 builds will now execute on XP and higher. Previously they would
only execute on Vista and higher.

=item *

You can now build perl with GNU Make and GCC.  [perl #123440]

=item *

C<truncate($filename, $size)> now works for files over 4GB in size.
[perl #125347]

=item *

Parallel building has been added to the dmake C<makefile.mk> makefile. All
Win32 compilers are supported.

=item *

Building a 64-bit perl with a 64-bit GCC but a 32-bit gmake would
result in an invalid C<$Config{archname}> for the resulting perl.
[perl #127584]

=item *

Errors set by Winsock functions are now put directly into C<$^E>, and the
relevant C<WSAE*> error codes are now exported from the L<Errno> and L<POSIX>
modules for testing this against.

The previous behaviour of putting the errors (converted to POSIX-style C<E*>
error codes since Perl 5.20.0) into C<$!> was buggy due to the non-equivalence
of like-named Winsock and POSIX error constants, a relationship between which
has unfortunately been established in one way or another since Perl 5.8.0.

The new behaviour provides a much more robust solution for checking Winsock
errors in portable software without accidentally matching POSIX tests that were
intended for other OSes and may have different meanings for Winsock.

The old behaviour is currently retained, warts and all, for backwards
compatibility, but users are encouraged to change any code that tests C<$!>
against C<E*> constants for Winsock errors to instead test C<$^E> against
C<WSAE*> constants.  After a suitable deprecation period, the old behaviour may
be removed, leaving C<$!> unchanged after Winsock function calls, to avoid any
possible confusion over which error variable to check.

=back

=item ppc64el floating point

The floating point format of ppc64el (Debian naming for little-endian
PowerPC) is now detected correctly.

=back

=head1 Internal Changes

=over 4

=item *

The implementation of perl's context stack system, and its internal API,
have been heavily reworked. Note that no significant changes have been
made to any external APIs, but XS code which relies on such internal
details may need to be fixed. The main changes are:

=over 4

=item *

The C<PUSHBLOCK()>, C<POPSUB()> etc. macros have been replaced with static
inline functions such as C<cx_pushblock()>, C<cx_popsub()> etc. These use
function args rather than implicitly relying on local vars such as
C<gimme> and C<newsp> being available. Also their functionality has
changed: in particular, C<cx_popblock()> no longer decrements
C<cxstack_ix>. The ordering of the steps in the C<pp_leave*> functions
involving C<cx_popblock()>, C<cx_popsub()> etc. has changed. See the new
documentation, L<perlguts/"Dynamic Scope and the Context Stack">, for
details on how to use them.

=item *

Various macros, which now consistently have a CX_ prefix, have been added:

  CX_CUR(), CX_LEAVE_SCOPE(), CX_POP()

or renamed:

  CX_POP_SAVEARRAY(), CX_DEBUG(), CX_PUSHSUBST(), CX_POPSUBST()

=item *

C<cx_pushblock()> now saves C<PL_savestack_ix> and C<PL_tmps_floor>, so
C<pp_enter*> and C<pp_leave*> no longer do

  ENTER; SAVETMPS; ....; LEAVE

=item *

C<cx_popblock()> now also restores C<PL_curpm>.

=item *

In C<dounwind()> for every context type, the current savestack frame is
now processed before each context is popped; formerly this was only done
for sub-like context frames. This action has been removed from
C<cx_popsub()> and placed into its own macro, C<CX_LEAVE_SCOPE(cx)>, which
must be called before C<cx_popsub()> etc.

C<dounwind()> now also does a C<cx_popblock()> on the last popped frame
(formerly it only did the C<cx_popsub()> etc. actions on each frame).

=item *

The temps stack is now freed on scope exit; previously, temps created
during the last statement of a block wouldn't be freed until the next
C<nextstate> following the block (apart from an existing hack that did
this for recursive subs in scalar context); and in something like
C<f(g())>, the temps created by the last statement in C<g()> would
formerly not be freed until the statement following the return from
C<f()>.

=item *

Most values that were saved on the savestack on scope entry are now
saved in suitable new fields in the context struct, and saved and
restored directly by C<cx_pushfoo()> and C<cx_popfoo()>, which is much
faster.

=item *

Various context struct fields have been added, removed or modified.

=item *

The handling of C<@_> in C<cx_pushsub()> and C<cx_popsub()> has been
considerably tidied up, including removing the C<argarray> field from the
context struct, and extracting out some common (but rarely used) code into
a separate function, C<clear_defarray()>. Also, useful subsets of
C<cx_popsub()> which had been unrolled in places like C<pp_goto> have been
gathered into the new functions C<cx_popsub_args()> and
C<cx_popsub_common()>.

=item *

C<pp_leavesub> and C<pp_leavesublv> now use the same function as the rest
of the C<pp_leave*>'s to process return args.

=item *

C<CXp_FOR_PAD> and C<CXp_FOR_GV> flags have been added, and
C<CXt_LOOP_FOR> has been split into C<CXt_LOOP_LIST>, C<CXt_LOOP_ARY>.

=item *

Some variables formerly declared by C<dMULTICALL> (but not documented) have
been removed.

=back

=item *

The obscure C<PL_timesbuf> variable, effectively a vestige of Perl 1, has
been removed. It was documented as deprecated in Perl 5.20, with a statement
that it would be removed early in the 5.21.x series; that has now finally
happened.
L<[perl #121351]|https://rt.perl.org/Ticket/Display.html?id=121351>

=item *

Remove unwarranted assertion in C<Perl_newATTRSUB_x()>. If a stub subroutine
definition with a prototype has been seen, then any subsequent stub (or
definition) of the same subroutine with an attribute was causing an assertion
failure because of a null pointer.
L<[perl #126845]|https://rt.perl.org/Ticket/Display.html?id=126845>

=item *

Replace C<::> with C<__> in C<ExtUtils::ParseXS> like it's done for
parameters/return values. This is more consistent, and simplifies writing XS
code wrapping C++ classes into a nested Perl namespace (it requires only
a typedef for C<Foo__Bar> rather than two, one for C<Foo_Bar> and the other
for C<Foo::Bar>).

=item *

Deprecate the C<to_utf8_case()> function, see
L<http://nntp.perl.org/group/perl.perl5.porters/233287>.

=item *

Perl core code and the threads extension have been annotated so that,
if Perl is configured to use threads, then during compile-time clang (3.6
or later) will warn about suspicious uses of mutexes.
See L<http://clang.llvm.org/docs/ThreadSafetyAnalysis.html> for more
information.

=item *

The C<signbit()> emulation has been enhanced.  This will help older
and/or more exotic platforms or configurations.

=item *

The C<to_utf8_case> function is discouraged in favor of C<toUPPER_utf8>,
C<toTITLE_utf8>, C<toLOWER_utf8>, and C<toFOLD_utf8>.

=item *

EBCDIC code paths have largely been unified to avoid repetition.

=item *

MSWin32 code for C<$^X> has been moved out of the F<win32> directory to
F<caretx.c>, where other operating systems set that variable.

=item *

C<< sv_ref() >> is now part of the API.

=item *

L<perlapi/sv_backoff> had its return type changed from C<int> to C<void>.  It
previously has always returned C<0> since Perl 5.000 stable but that was
undocumented.  Although C<sv_backoff> is marked as public API, XS code is not
expected to be impacted since the proper API call would be through public API
C<sv_setsv(sv, &PL_sv_undef)>, or quasi-public C<SvOOK_off>, or non-public
C<SvOK_off> calls, and the return value of C<sv_backoff> was previously a
meaningless constant that can be rewritten as C<(sv_backoff(sv),0)>.

=item *

The C<EXTEND> and C<MEXTEND> macros have been improved to avoid various issues
with integer truncation and wrapping.  In particular, some casts formerly used
within the macros have been removed.  This means for example that passing an
unsigned nitems argument is likely to raise a compiler warning now (it's always
been documented to require a signed value; formerly int, lately SSize_t).

=item *

C<PL_sawalias> and C<GPf_ALIASED_SV> have been removed.

=item *

C<GvASSIGN_GENERATION> and C<GvASSIGN_GENERATION_set> have been removed.

=back

=head1 Selected Bug Fixes

=over 4

=item *

It now works properly to specify a user-defined property, such as

 qr/\p{mypkg1::IsMyProperty}/i

with C</i> caseless matching, an explicit package name, and
I<IsMyProperty> not defined at the time of the pattern compilation.

=item *

Perl's memcpy(), memmove(), memset() and memcmp() fallbacks are now
more compatible with the originals.  [perl #127619]

=item *

Fixed the issue where a C<< s///r >>) with B<< -DPERL_NO_COW >> attempts
to modify the source SV, resulting in the program dying. [perl #127635]

=item *

Fixed a spurious warning about posix character classes. [perl #127581]

=item *

Fixed an obscure case where a pattern could fail to match. This only 
occurred when matching characters from the set of C1 controls, when
the target matched string was in UTF-8, and only on EBCDIC platforms.

=item *

Fixed over eager warnings for C<< /[.foo.]/ >>.

This prevents Perl from warning about constructs like C<< /[.].*[.]/ >>.
[perl #127582, #127604]

=item *

Narrow the filename check in F<strict.pm> and F<warnings.pm>. Previously,
it assumed that if the filename (without the F<.pmc?> extension) differed
from the package name, if was a misspelled use statement (i.e. C<use Strict>
instead of C<use strict>). We now check whether there's really a 
miscapitalization happening, and not another issue.

=item *

Turn an assertion into a more user friendly failure when parsing
regexes. [perl #127599]

=item *

Correctly raise an error when trying to compile patterns with 
unterminated character classes while there are trailing backslashes.
[perl #126141].

=item *

Line numbers larger than 2**31-1 but less than 2**32 are no longer
returned by caller() as negative numbers.  [perl #126991]

=item *

C<< unless ( I<assignment> ) >> now properly warns when syntax
warnings are enabled.  [perl #127122]

=item *

Setting an C<ISA> glob to an array reference now properly adds
C<isaelem> magic to any existing elements.  Previously modifying such
an element would not update the ISA cache, so method calls would call
the wrong function.  Perl would also crash if the C<ISA> glob was
destroyed, since new code added in 5.23.7 would try to release the
C<isaelem> magic from the elements.  [perl #127351]

=item *

If a here-doc was found while parsing another operator, the parser had
already read end of file, and the here-doc was not terminated, perl
could produce an assertion or a segmentation fault.  This now reliably
complains about the unterminated here-doc.  [perl #125540]

=item *

untie() would sometimes return the last value returned by the UNTIE()
handler as well as it's normal value, messing up the stack.  [perl
#126621]

=item *

Fixed an operator precedence problem when C< castflags & 2> is true.
[perl #127474]

=item *

Caching of DESTROY methods could result in a non-pointer or a
non-STASH stored in the SvSTASH() slot of a stash, breaking the B
STASH() method.  The DESTROY method is now cached in the MRO metadata
for the stash.  [perl #126410]

=item *

The AUTOLOAD method is now called when searching for a DESTROY method,
and correctly sets C<$AUTOLOAD> too.  [perl #124387]  [perl #127494]

=item *

Avoid parsing beyond the end of the buffer when processing a C<#line>
directive with no filename.  [perl #127334]

=item *

Perl now raises a warning when a regular expression pattern looks like
it was supposed to contain a POSIX class, like C<qr/[[:alpha:]]/>, but
there was some slight defect in its specification which causes it to
instead be treated as a regular bracketed character class.  An example
would be missing the second colon in the above like this:
C<qr/[[:alpha]]/>.  This compiles to match a sequence of two characters.
The second is C<"]">, and the first is any of: C<"[">, C<":">, C<"a">,
C<"h">, C<"l">, or C<"p">.   This is unlikely to be the intended
meaning, and now a warning is raised.  No warning is raised unless the
specification is very close to one of the 14 legal POSIX classes.  (See
L<perlrecharclass/POSIX Character Classes>.)
[perl #8904]

=item *

Certain regex patterns involving a complemented POSIX class in an
inverted bracketed character class, and matching something else
optionally would improperly fail to match.  An example of one that could
fail is C</qr/_?[^\Wbar]\x{100}/>.  This has been fixed.
[perl #127537]

=item *

Perl 5.22 added support to the C99 hexadecimal floating point notation,
but sometimes misparses hex floats. This has been fixed.
[perl #127183]

=item *

A regression that allowed undeclared barewords in hash keys to work despite
strictures has been fixed.
L<[perl #126981]|https://rt.perl.org/Ticket/Display.html?id=126981>

=item *

Calls to the placeholder C<&PL_sv_yes> used internally when an C<import()>
or C<unimport()> method isn't found now correctly handle scalar context.
L<[perl #126042]|https://rt.perl.org/Ticket/Display.html?id=126042>

=item *

Report more context when we see an array where we expect to see an
operator and avoid an assertion failure.
L<[perl #123737]|https://rt.perl.org/Ticket/Display.html?id=123737>

=item *

Modifying an array that was previously a package C<@ISA> no longer
causes assertion failures or crashes.
L<[perl #123788]|https://rt.perl.org/Ticket/Display.html?id=123788>

=item *

Retain binary compatibility across plain and DEBUGGING perl builds.
L<[perl #127212]|https://rt.perl.org/Ticket/Display.html?id=127212>

=item *

Avoid leaking memory when setting C<$ENV{foo}> on darwin.
L<[perl #126240]|https://rt.perl.org/Ticket/Display.html?id=126240>

=item *

C</...\G/> no longer crashes on utf8 strings. When C<\G> is a fixed number
of characters from the start of the regex, perl needs to count back that
many characters from the current C<pos()> position and start matching from
there. However, it was counting back bytes rather than characters, which
could lead to panics on utf8 strings.

=item *

In some cases operators that return integers would return negative
integers as large positive integers.
L<[perl #126635]|https://rt.perl.org/Ticket/Display.html?id=126635>

=item *

The C<pipe()> operator would assert for DEBUGGING builds instead of
producing the correct error message.  The condition asserted on is
detected and reported on correctly without the assertions, so the
assertions were removed.
L<[perl #126480]|https://rt.perl.org/Ticket/Display.html?id=126480>

=item *

In some cases, failing to parse a here-doc would attempt to use freed
memory.  This was caused by a pointer not being restored correctly.
L<[perl #126443]|https://rt.perl.org/Ticket/Display.html?id=126443>

=item *

C<< @x = sort { *a = 0; $a <=> $b } 0 .. 1 >> no longer frees the GP
for *a before restoring its SV slot.
L<[perl #124097]|https://rt.perl.org/Ticket/Display.html?id=124097>

=item *

Multiple problems with the new hexadecimal floating point printf
format C<%a> were fixed:
L<[perl #126582]|https://rt.perl.org/Ticket/Display.html?id=126582>,
L<[perl #126586]|https://rt.perl.org/Ticket/Display.html?id=126586>,
L<[perl #126822]|https://rt.perl.org/Ticket/Display.html?id=126822>

=item *

Calling mg_set() in leave_scope() no longer leaks.

=item *

A regression from Perl v5.20 was fixed in which debugging output of regular
expression compilation was wrong.  (The pattern was correctly compiled, but
what got displayed for it was wrong.)

=item *

C<\b{sb}> works much better.  In Perl v5.22.0, this new construct didn't
seem to give the expected results, yet passed all the tests in the
extensive suite furnished by Unicode.  It turns out that it was because
these were short input strings, and the failures had to do with longer
inputs.  This was fixed in Perl 5.23.4, but the improvement was not
noticed until after that was released, so is included here now.

=item *

Certain syntax errors in
L<perlrecharclass/Extended Bracketed Character Classes> caused panics
instead of the proper error message.  This has now been fixed. [perl
#126481]

=item *

An earlier commit added a message when a quantifier in a regular
expression was useless, but then caused the parser to skip it;
this caused the surplus quantifier to be silently ignored, instead
of throwing an error. This is now fixed. [perl #126253]

=item *

The switch to building non-XS modules last in win32/makefile.mk (introduced
by design as part of the changes to enable parallel building) caused the
build of POSIX to break due to problems with the version module. This
is now fixed.

=item *

Improved parsing of hex float constants.

=item *

Fixed an issue with C<< pack >> where C<< pack "H" >> (and C<< pack "h" >>)
could read past the source when given a non-utf8 source, and a utf8 target.
[perl #126325]

=item *

Fixed several cases where perl would abort due to a segmentation fault,
or a C-level assert. [perl #126615], [perl #126602], [perl #126193].

=item *

There were places in regular expression patterns where comments (C<(?#...)>)
weren't allowed, but should have been.  This is now fixed.
L<[perl #116639]|https://rt.perl.org/Ticket/Display.html?id=116639>

=item *

Some regressions from Perl 5.20 have been fixed, in which some syntax errors in
L<C<(?[...])>|perlrecharclass/Extended Bracketed Character Classes> constructs
within regular expression patterns could cause a segfault instead of a proper
error message.
L<[perl #126180]|https://rt.perl.org/Ticket/Display.html?id=126180>
L<[perl #126404]|https://rt.perl.org/Ticket/Display.html?id=126404>

=item *

Another problem with
L<C<(?[...])>|perlrecharclass/Extended Bracketed Character Classes>
constructs has been fixed wherein things like C<\c]> could cause panics.
L<[perl #126181]|https://rt.perl.org/Ticket/Display.html?id=126181>

=item *

Some problems with attempting to extend the perl stack to around 2G or 4G
entries have been fixed.  This was particularly an issue on 32-bit perls built
to use 64-bit integers, and was easily noticeable with the list repetition
operator, e.g.

    @a = (1) x $big_number

Formerly perl may have crashed, depending on the exact value of C<$big_number>;
now it will typically raise an exception.
L<[perl #125937]|https://rt.perl.org/Ticket/Display.html?id=125937>

=item *

In a regex conditional expression C<(?(condition)yes-pattern|no-pattern)>, if
the condition is C<(?!)> then perl failed the match outright instead of
matching the no-pattern.  This has been fixed.
L<[perl #126222]|https://rt.perl.org/Ticket/Display.html?id=126222>

=item *

The special backtracking control verbs C<(*VERB:ARG)> now all allow an optional
argument and set C<REGERROR>/C<REGMARK> appropriately as well.
L<[perl #126186]|https://rt.perl.org/Ticket/Display.html?id=126186>

=item *

Several bugs, including a segmentation fault, have been fixed with the bounds
checking constructs (introduced in Perl 5.22) C<\b{gcb}>, C<\b{sb}>, C<\b{wb}>,
C<\B{gcb}>, C<\B{sb}>, and C<\B{wb}>.  All the C<\B{}> ones now match an empty
string; none of the C<\b{}> ones do.
L<[perl #126319]|https://rt.perl.org/Ticket/Display.html?id=126319>

=item *

Duplicating a closed file handle for write no longer creates a
filename of the form F<GLOB(0xXXXXXXXX)>.  [perl #125115]

=item *

Warning fatality is now ignored when rewinding the stack.  This
prevents infinite recursion when the now fatal error also causes
rewinding of the stack.  [perl #123398]

=item * 

In perl v5.22.0, the logic changed when parsing a numeric parameter to the -C
option, such that the successfully parsed number was not saved as the option
value if it parsed to the end of the argument.  [perl #125381]

=item *

The PadlistNAMES macro is an lvalue again.

=item *

Zero -DPERL_TRACE_OPS memory for sub-threads.

perl_clone_using() was missing Zero init of PL_op_exec_cnt[].  This
caused sub-threads in threaded -DPERL_TRACE_OPS builds to spew exceedingly
large op-counts at destruct.  These counts would print %x as "ABABABAB",
clearly a mem-poison value.

=item *

A leak in the XS typemap caused one scalar to be leaked each time a C<FILE *>
or a C<PerlIO *> was C<OUTPUT:>ed or imported to Perl, since perl 5.000. These
particular typemap entries are thought to be extremely rarely used by XS
modules. [perl #124181]

=item *

C<alarm()> and C<sleep()> will now warn if the argument is a negative number
and return undef. Previously they would pass the negative value to the
underlying C function which may have set up a timer with a surprising value.

=item *

Perl can again be compiled with any Unicode version.  This used to
(mostly) work, but was lost in v5.18 through v5.20.  The property
C<Name_Alias> did not exist prior to Unicode 5.0.  L<Unicode::UCD>
incorrectly said it did.  This has been fixed.

=item *

Very large code-points (beyond Unicode) in regular expressions no
longer cause a buffer overflow in some cases when converted to UTF-8.
L<[perl #125826]|https://rt.perl.org/Ticket/Display.html?id=125826>

=item *

The integer overflow check for the range operator (...) in list
context now correctly handles the case where the size of the range is
larger than the address space.  This could happen on 32-bits with
-Duse64bitint.
L<[perl #125781]|https://rt.perl.org/Ticket/Display.html?id=125781>

=item *

A crash with C<< %::=(); J->${\"::"} >> has been fixed.
L<[perl #125541]|https://rt.perl.org/Ticket/Display.html?id=125541>

=item *

C<qr/(?[ () ])/> no longer segfaults, giving a syntax error message instead.
[perl #125805]

=item *

Regular expression possessive quantifier v5.20 regression now fixed.
C<qr/>I<PAT>C<{>I<min>,I<max>C<}+>C</> is supposed to behave identically
to C<qr/(?E<gt>>I<PAT>C<{>I<min>,I<max>C<})/>.  Since v5.20, this didn't
work if I<min> and I<max> were equal.  [perl #125825]

=item *

C<< BEGIN <> >> no longer segfaults and properly produces an error
message.  [perl #125341]

=item *

In C<tr///> an illegal backwards range like C<tr/\x{101}-\x{100}//> was
not always detected, giving incorrect results.  This is now fixed.

=back

=head1 Acknowledgements

[ XXX: generate this just in time, Ricardo! ]

=head1 Reporting Bugs

If you find what you think is a bug, you might check the articles recently
posted to the comp.lang.perl.misc newsgroup and the perl bug database at
https://rt.perl.org/ .  There may also be information at
http://www.perl.org/ , the Perl Home Page.

If you believe you have an unreported bug, please run the L<perlbug> program
included with your release.  Be sure to trim your bug down to a tiny but
sufficient test case.  Your bug report, along with the output of C<perl -V>,
will be sent off to perlbug@perl.org to be analysed by the Perl porting team.

If the bug you are reporting has security implications which make it
inappropriate to send to a publicly archived mailing list, then see
L<perlsec/SECURITY VULNERABILITY CONTACT INFORMATION>
for details of how to report the issue.

=head1 SEE ALSO

The F<Changes> file for an explanation of how to view exhaustive details on
what changed.

The F<INSTALL> file for how to build Perl.

The F<README> file for general stuff.

The F<Artistic> and F<Copying> files for copyright information.

=cut