This is a live mirror of the Perl 5 development currently hosted at https://github.com/perl/perl5
perl5.git
9 years agoRevert "cpan/Time-HiRes/t/itimer.t: better diagnostics"
David Mitchell [Thu, 21 Aug 2014 15:50:39 +0000 (16:50 +0100)]
Revert "cpan/Time-HiRes/t/itimer.t: better diagnostics"

This reverts commit 65dd9533e90d8e1ea84c091ff43133132c667a1c.

That commit was just temporary to help debug a smoke issue.
Since Time::Hires is a CPAN-first module, back the patch out from blead,
and leave it to the module's owner whether they want to add it back in.

9 years agodump.c: eliminate DUMP_OP_FLAGS/PRIVATE macros
David Mitchell [Thu, 21 Aug 2014 15:22:35 +0000 (16:22 +0100)]
dump.c: eliminate DUMP_OP_FLAGS/PRIVATE macros

With the MAD code, these macros were each used in two places to
symbolically dump op_flags and op_private. Now that the MAD code
has been removed, they are each only used once, so inline them.

9 years agoUpdate Pod-Perldoc to CPAN version 3.24
Chris 'BinGOs' Williams [Thu, 21 Aug 2014 10:35:53 +0000 (11:35 +0100)]
Update Pod-Perldoc to CPAN version 3.24

  [DELTA]

3.24 - Tue Aug 19 03:38:07 UTC 2014
    * Release 3.24

    Make sure when we open a filehandle for reading
    or writing, we set ':encoding(UTF-8)' on it
    everywhere.  Closes RT#98019.

3.23_01 - Sat Aug 16 16:47:45 UTC 2014
    * Test release to test UTF8 filehandles.
    * Happy CPAN Day!

9 years agoUpdate Socket version in Maintainers.pl
Chris 'BinGOs' Williams [Thu, 21 Aug 2014 10:34:14 +0000 (11:34 +0100)]
Update Socket version in Maintainers.pl

9 years agoUpdate Test-Harness to CPAN version 3.33
Chris 'BinGOs' Williams [Thu, 21 Aug 2014 10:32:26 +0000 (11:32 +0100)]
Update Test-Harness to CPAN version 3.33

  [DELTA]

3.33
        - Various documentation fixes (Leon Timmermans, Justin Cook)

9 years agoUpdate experimental to CPAN version 0.010
Chris 'BinGOs' Williams [Thu, 21 Aug 2014 10:07:03 +0000 (11:07 +0100)]
Update experimental to CPAN version 0.010

  [DELTA]

0.010     2014-08-19 17:47:20+02:00 Europe/Amsterdam
          Add returning 1's to evals in tests

0.009     2014-08-16 13:43:53+02:00 Europe/Amsterdam
          Hardcode features for perl < 5.15.7

9 years agoIEEE quadruple exponents were off.
Jarkko Hietaniemi [Thu, 21 Aug 2014 02:48:05 +0000 (22:48 -0400)]
IEEE quadruple exponents were off.

9 years agoModule::CoreList version number bump
Peter Martini [Thu, 21 Aug 2014 11:23:39 +0000 (07:23 -0400)]
Module::CoreList version number bump

9 years agoVersion bump addendum; changed a 5.21.1 to .3
Peter Martini [Thu, 21 Aug 2014 06:35:17 +0000 (02:35 -0400)]
Version bump addendum; changed a 5.21.1 to .3

The note on backwards compatibility notes that this
may not be binary compatible with version 5.21.1 or earlier.
That should float to always be one subversion behind the current,
so bumping it to 5.21.3.

9 years agoBump version number from 5.21.3 to 5.21.4
Peter Martini [Thu, 21 Aug 2014 06:32:55 +0000 (02:32 -0400)]
Bump version number from 5.21.3 to 5.21.4

9 years agoNew perldelta.
Peter Martini [Thu, 21 Aug 2014 06:02:48 +0000 (02:02 -0400)]
New perldelta.

9 years agoModule-CoreList CPAN version synchronisation
Chris 'BinGOs' Williams [Thu, 21 Aug 2014 09:50:39 +0000 (10:50 +0100)]
Module-CoreList CPAN version synchronisation

9 years agoEpigraph for 5.21.3
Peter Martini [Thu, 21 Aug 2014 05:39:04 +0000 (01:39 -0400)]
Epigraph for 5.21.3

9 years agoMerge branch 'release-5.21.3' into blead
Peter Martini [Thu, 21 Aug 2014 05:17:36 +0000 (01:17 -0400)]
Merge branch 'release-5.21.3' into blead

9 years agoScowl. 19 should have been '[VERSION]'
Peter Martini [Thu, 21 Aug 2014 02:47:01 +0000 (22:47 -0400)]
Scowl.  19 should have been '[VERSION]'

And I should have noticed that before I hit 'send'!

9 years agoperldelta: cleanup per JHI's comment origin/release-5.21.3 v5.21.3
Peter Martini [Wed, 20 Aug 2014 20:20:05 +0000 (16:20 -0400)]
perldelta: cleanup per JHI's comment

9 years agoAdd new release to perlhist
Peter Martini [Wed, 20 Aug 2014 08:23:26 +0000 (04:23 -0400)]
Add new release to perlhist

9 years agoFinal perldelta change for 5.21.3
Peter Martini [Wed, 20 Aug 2014 08:21:39 +0000 (04:21 -0400)]
Final perldelta change for 5.21.3

9 years agoUpdate Module::CoreList for 5.21.3
Peter Martini [Wed, 20 Aug 2014 08:21:14 +0000 (04:21 -0400)]
Update Module::CoreList for 5.21.3

9 years agoperldelta suggested by Module-CoreList
Peter Martini [Wed, 20 Aug 2014 07:49:31 +0000 (03:49 -0400)]
perldelta suggested by Module-CoreList

9 years agoperldelta: document the ExtUtils move to cpan/
Peter Martini [Wed, 20 Aug 2014 07:40:53 +0000 (03:40 -0400)]
perldelta: document the ExtUtils move to cpan/

9 years agoperldelta: fix my bad pod
Peter Martini [Wed, 20 Aug 2014 07:15:28 +0000 (03:15 -0400)]
perldelta: fix my bad pod

9 years agoperldelta for hexadecimal floats.
Jarkko Hietaniemi [Wed, 20 Aug 2014 13:45:50 +0000 (09:45 -0400)]
perldelta for hexadecimal floats.

9 years agomagic.t: android: bypass two $0 tests and add one
Alexandre (Midnite) Jousset [Wed, 20 Aug 2014 19:24:21 +0000 (21:24 +0200)]
magic.t: android: bypass two $0 tests and add one

"ps" on Android has different syntaxes depending on the tool used.
It can be toolbox or busybox. Bypass the "ps" tests and enable
the /proc/$$/cmdline one.

9 years agoAndroid hints: Always append /system/lib and /vendor/lib in libpth
Brian Fraser [Wed, 20 Aug 2014 18:25:18 +0000 (20:25 +0200)]
Android hints: Always append /system/lib and /vendor/lib in libpth

These are the canonical directories for libraries on Android.
This is only mildly important for native builds, since
whatever toolchain was installed will likely provide
their own /lib, but is quite important for cross builds.

9 years agoAndroid hints: define d_procselfexe and procselfexe if needed
Brian Fraser [Wed, 20 Aug 2014 17:21:30 +0000 (19:21 +0200)]
Android hints: define d_procselfexe and procselfexe if needed

Only affects cross-compile builds; currently the tests
in Configure for this feature will be run on the host,
not the target system, which makes the detection of the
feature unreliable.

Since we know that /proc/self/exe exists, just define it
and set $procselfexe to that.

9 years agoConfigure: Improve the d_prctl_set_name detection for Android
Brian Fraser [Wed, 20 Aug 2014 17:19:13 +0000 (19:19 +0200)]
Configure: Improve the d_prctl_set_name detection for Android

On some versions of Android (e.g., 4.2.2, API level 17)
the header that defines the constants used for this test
will not work if unistd.h isn't already included.

9 years agoMention the possibility of fp rounding modes' effects.
Jarkko Hietaniemi [Wed, 20 Aug 2014 15:31:15 +0000 (11:31 -0400)]
Mention the possibility of fp rounding modes' effects.

9 years agoTry addressing perl #122578, low-order fp diffs.
Jarkko Hietaniemi [Wed, 20 Aug 2014 15:19:39 +0000 (11:19 -0400)]
Try addressing perl #122578, low-order fp diffs.

9 years agoPass $ENV{HARNESS_TIMER} to Test::Harness from t/harness.
Jarkko Hietaniemi [Wed, 20 Aug 2014 13:48:26 +0000 (09:48 -0400)]
Pass $ENV{HARNESS_TIMER} to Test::Harness from t/harness.

(One would think it worked as-is, but didn't.  Maybe T::H resets it somehow.)

9 years agoperldelta for hexadecimal floats.
Jarkko Hietaniemi [Wed, 20 Aug 2014 13:45:50 +0000 (09:45 -0400)]
perldelta for hexadecimal floats.

9 years agoSeparate grok_infnan() from grok_number().
Jarkko Hietaniemi [Mon, 18 Aug 2014 16:41:41 +0000 (12:41 -0400)]
Separate grok_infnan() from grok_number().

Remaining issues:

(1) would need tests, but there are two problems: [a] generating inf/nan
    reliably and testing for it from Perl level is hard (see items (2) and
    (3) below), and [b] the behavior of various systems with especially NaN
    differs (some platforms might throw SIGFPEs).

(2) toke.c:scan_number() will not call this code (via grok_number)
    because "NaN" or "Inf" do not look at all like floats to it.

(3) Even as we now recognize these forms, the native strtod()
    might not (problem of cross-portability of these exceptional
    forms: Win32 outputs e.g. "1.#INF", what Linux reading this should do,
    or conversely Linux outputs "Inf", what should Win32 do?)

9 years agoPerl_signbit should return non-zero for -0.
Jarkko Hietaniemi [Mon, 18 Aug 2014 15:31:50 +0000 (11:31 -0400)]
Perl_signbit should return non-zero for -0.

9 years agoWin32 apparently has _isnan() and _finite(), and _fpclass().
Jarkko Hietaniemi [Mon, 18 Aug 2014 14:55:13 +0000 (10:55 -0400)]
Win32 apparently has _isnan() and _finite(), and _fpclass().

(At least, if one believes the Oracle of Google.)

9 years agoUse Perl_isfinite().
Jarkko Hietaniemi [Mon, 18 Aug 2014 13:12:09 +0000 (09:12 -0400)]
Use Perl_isfinite().

Also kill obsolete/wrong comment: Perl_frexp is not casting its
argument to NV, and if USE_LONG_DOUBLE Perl_frexp is frexpl.
And if we do have long double and don't have frexpl, it is unclear
what should we do? (in this particular case, casting to double might
be the best possible thing to do anyway.)

9 years agoPerl_isinf enhancements.
Jarkko Hietaniemi [Mon, 18 Aug 2014 12:23:17 +0000 (08:23 -0400)]
Perl_isinf enhancements.

Try isfinitel first, and require Perl_isnan.

9 years agoOSX does have isfinite(), and 10.9+ prefers it over finite().
Jarkko Hietaniemi [Mon, 18 Aug 2014 12:56:48 +0000 (08:56 -0400)]
OSX does have isfinite(), and 10.9+ prefers it over finite().

isfinite() is a macro+inline while Configure tests it as a library symbol.

9 years agoPerl_isfinite() enhancements.
Jarkko Hietaniemi [Mon, 18 Aug 2014 12:55:23 +0000 (08:55 -0400)]
Perl_isfinite() enhancements.

Use isfinitel() if available, prefer C99 isfinite() over BSD finite().

9 years agoConfigure for isfinitel / isinfl.
Jarkko Hietaniemi [Mon, 18 Aug 2014 13:57:06 +0000 (09:57 -0400)]
Configure for isfinitel / isinfl.

9 years agoAdd todo comment about special hexfp values.
Jarkko Hietaniemi [Tue, 19 Aug 2014 22:44:30 +0000 (18:44 -0400)]
Add todo comment about special hexfp values.

9 years agoQuadruple float BE test triggering had a typo.
Jarkko Hietaniemi [Wed, 20 Aug 2014 02:16:54 +0000 (22:16 -0400)]
Quadruple float BE test triggering had a typo.

9 years agoReduce double-double %a to single-double for now.
Jarkko Hietaniemi [Mon, 18 Aug 2014 16:56:51 +0000 (12:56 -0400)]
Reduce double-double %a to single-double for now.

How to exactly handle the second double is unknown, so for
now extract the mantissa bits only from the first double.
(Known platforms aix-ppc and linux-ppc.)

9 years agodouble-double needs the first mantissa bits of doubles.
Jarkko Hietaniemi [Mon, 18 Aug 2014 21:04:06 +0000 (17:04 -0400)]
double-double needs the first mantissa bits of doubles.

9 years agoModify the hexfp exponent explicitly in each long double case.
Jarkko Hietaniemi [Mon, 18 Aug 2014 19:27:58 +0000 (15:27 -0400)]
Modify the hexfp exponent explicitly in each long double case.

9 years agoComment: scan_num also handles hexadecimal floats now.
Jarkko Hietaniemi [Mon, 18 Aug 2014 12:13:18 +0000 (08:13 -0400)]
Comment: scan_num also handles hexadecimal floats now.

9 years agoConfigure backports cause reorder
H.Merijn Brand [Wed, 20 Aug 2014 09:07:25 +0000 (11:07 +0200)]
Configure backports cause reorder

9 years agoperldelta: fix my bad pod
Peter Martini [Wed, 20 Aug 2014 07:15:28 +0000 (03:15 -0400)]
perldelta: fix my bad pod

9 years agoMake S_method_common use gv_stashpvn instead of copypasted cache usage
syber [Fri, 8 Aug 2014 17:06:10 +0000 (21:06 +0400)]
Make S_method_common use gv_stashpvn instead of copypasted cache usage

The previous commit copied the PL_stashcache handling code from
S_method_common() to gv_stashpvn(), so now we can call that function
directly rather than doing it ourselves.

9 years agoMake gv_stashpvn() use PL_stashcache
syber [Fri, 8 Aug 2014 15:03:03 +0000 (19:03 +0400)]
Make gv_stashpvn() use PL_stashcache

perl has a stash name lookup cache, which is currently just used for
looking up up class names in Some::Class->method calls. This means
that a lookup of the class name 'A::B::C' gets reduced from doing
several stash lookups (e.g. $::{'A::'}{'B::'}{'C::'}) to a single
cache lookup (e.g. $PL_stashcache{'A::B::C::'}, so to speak).

Make gv_stashpvn() use this cache too, which means that all package
name lookups benefit, not just class method calls. Among other things,
this also indirectly affects bless operations both from Perl (OP_BLESS)
and XS.

9 years agoperldelta: finished the stragglers
Peter Martini [Wed, 20 Aug 2014 06:55:31 +0000 (02:55 -0400)]
perldelta: finished the stragglers

Cleaned up remaining XXX entries, checked formatting,
all that's left is contributors.

9 years agoperldelta for 7e0dd61, MurmurHash64[AB]
Peter Martini [Wed, 20 Aug 2014 06:42:00 +0000 (02:42 -0400)]
perldelta for 7e0dd61, MurmurHash64[AB]

9 years agoperldelta for f04d2c3
Peter Martini [Wed, 20 Aug 2014 06:39:44 +0000 (02:39 -0400)]
perldelta for f04d2c3

9 years agoperldelta: put fatal perldiags in the right spot
Peter Martini [Wed, 20 Aug 2014 06:33:15 +0000 (02:33 -0400)]
perldelta: put fatal perldiags in the right spot

9 years agoperldelta: No deprecations or perf enhancements
Peter Martini [Wed, 20 Aug 2014 06:29:48 +0000 (02:29 -0400)]
perldelta: No deprecations or perf enhancements

9 years agoperldelta for 842a9d2, [perl #122460]
Peter Martini [Wed, 20 Aug 2014 06:22:45 +0000 (02:22 -0400)]
perldelta for 842a9d2, [perl #122460]

9 years agoperldelta: platform specific notes
Peter Martini [Wed, 20 Aug 2014 06:16:34 +0000 (02:16 -0400)]
perldelta: platform specific notes

9 years agoperldelta: document grok_atou
Peter Martini [Wed, 20 Aug 2014 05:53:59 +0000 (01:53 -0400)]
perldelta: document grok_atou

9 years agoperldelta for f57000b
Peter Martini [Wed, 20 Aug 2014 05:49:31 +0000 (01:49 -0400)]
perldelta for f57000b

9 years agoperldelta: Added documentation changes
Peter Martini [Wed, 20 Aug 2014 05:32:35 +0000 (01:32 -0400)]
perldelta: Added documentation changes

9 years agoperldelta remove (most) boilerplate
Peter Martini [Wed, 20 Aug 2014 05:05:48 +0000 (01:05 -0400)]
perldelta remove (most) boilerplate

9 years agoperldelta for Safe changes
Peter Martini [Wed, 20 Aug 2014 04:53:26 +0000 (00:53 -0400)]
perldelta for Safe changes

Safe was bumped to 2.38 after a mechanism to replace subs was
discovered and fixed.

9 years agoperldelta for 57d69a4
Peter Martini [Wed, 20 Aug 2014 04:33:49 +0000 (00:33 -0400)]
perldelta for 57d69a4

HTTP::Tiny version bump

9 years agoperldelta for f4ef132
Peter Martini [Wed, 20 Aug 2014 04:31:07 +0000 (00:31 -0400)]
perldelta for f4ef132

defined(AASSIGN) is not fatalized

9 years agoperldelta for 60df683
Peter Martini [Wed, 20 Aug 2014 04:16:34 +0000 (00:16 -0400)]
perldelta for 60df683

Config::Perl::V version bump

9 years agoperldelta for 5f8324b
Peter Martini [Wed, 20 Aug 2014 04:14:32 +0000 (00:14 -0400)]
perldelta for 5f8324b

CPAN::Meta version bump

9 years agoperldelta for 54b7cb3
Peter Martini [Wed, 20 Aug 2014 04:13:22 +0000 (00:13 -0400)]
perldelta for 54b7cb3

Delta for CPAN::Meta::Requirements version bump

9 years agoperldelta for 438f0014a812, revised
Peter Martini [Wed, 20 Aug 2014 03:59:06 +0000 (23:59 -0400)]
perldelta for 438f0014a812, revised

It was actually 1.19 to 1.21, missed the earlier one.

9 years agoperldelta for 438f0014a812
Peter Martini [Wed, 20 Aug 2014 03:55:46 +0000 (23:55 -0400)]
perldelta for 438f0014a812

9 years agoperldelta for jhi double/float work
Peter Martini [Wed, 20 Aug 2014 03:48:48 +0000 (23:48 -0400)]
perldelta for jhi double/float work

9 years agoperldelta for 4cedf205aa6e
Peter Martini [Wed, 20 Aug 2014 02:58:33 +0000 (22:58 -0400)]
perldelta for 4cedf205aa6e

9 years agoperldelta for 3699825258c
Peter Martini [Wed, 20 Aug 2014 02:40:08 +0000 (22:40 -0400)]
perldelta for 3699825258c

9 years agoperldelta typo fix: change a ',' to a '.'
Peter Martini [Wed, 20 Aug 2014 02:16:52 +0000 (22:16 -0400)]
perldelta typo fix: change a ',' to a '.'

9 years agoRevert Test::Builder changes in prep for 5.21.3
Peter Martini [Wed, 20 Aug 2014 01:23:39 +0000 (21:23 -0400)]
Revert Test::Builder changes in prep for 5.21.3

This reverts commit 6bdb88770f849c47b5c09e733ac460ce3e9dbc97
and commit 59c96aebdd56baf5d1979103046b1c6d4e308aa2, which
are causing the test suite to run much slower and which are
failing on some platforms.

http://nntp.perl.org/group/perl.perl5.porters/218678 has
further details, with RT 122538 being the ticket number.

9 years agospelling: till -> until
Karen Etheridge [Wed, 20 Aug 2014 00:21:09 +0000 (17:21 -0700)]
spelling: till -> until

9 years agodocument what version \K was added in
Karen Etheridge [Wed, 20 Aug 2014 00:20:01 +0000 (17:20 -0700)]
document what version \K was added in

9 years agoWhen cross-compiling, copy cpan/Module-Metadata/
Brian Fraser [Tue, 29 Jul 2014 18:30:44 +0000 (20:30 +0200)]
When cross-compiling, copy cpan/Module-Metadata/

cpan/Module-Metadata/t/encoding.t needs some files in the
distribution to pass.

9 years agoFix finite/finitel/isfinite in configure.com.
Craig A. Berry [Tue, 19 Aug 2014 11:53:42 +0000 (06:53 -0500)]
Fix finite/finitel/isfinite in configure.com.

VMS has had these for a very long time but we haven't been
admitting it.  Spotted by Jarkko.

9 years agoUpdate blead to Test::Builder alpha 40
Chad Granum [Mon, 18 Aug 2014 21:04:14 +0000 (14:04 -0700)]
Update blead to Test::Builder alpha 40

Fixes performance issues
Fixes a test failure specific to certain utf8 conditions
Fixes a broken win32+fork test

Committer's notes: 2nd Test-Simple patch applied;
corresponds to $Test::Simple::VERSION 1.301001_040

9 years agoperldelta for b5b6210a9/#46947
Father Chrysostomos [Mon, 18 Aug 2014 00:26:47 +0000 (17:26 -0700)]
perldelta for b5b6210a9/#46947

9 years agoInclude longdblkind in perl -V output.
Jarkko Hietaniemi [Sun, 17 Aug 2014 20:47:01 +0000 (16:47 -0400)]
Include longdblkind in perl -V output.

9 years agoComment tweaks.
Jarkko Hietaniemi [Sun, 17 Aug 2014 17:51:01 +0000 (13:51 -0400)]
Comment tweaks.

9 years agoLittle-endian double-double detection was wrong.
Jarkko Hietaniemi [Sun, 17 Aug 2014 17:50:08 +0000 (13:50 -0400)]
Little-endian double-double detection was wrong.

9 years agoAdd quadruple precision tests, verified by Craig Berry.
Jarkko Hietaniemi [Sun, 17 Aug 2014 14:00:06 +0000 (10:00 -0400)]
Add quadruple precision tests, verified by Craig Berry.

Also tentatively adding double-double tests, but those are probably bogus.

Also comment tweaks.

9 years agoSeparate the extraction of hex values.
Jarkko Hietaniemi [Sat, 16 Aug 2014 15:27:52 +0000 (11:27 -0400)]
Separate the extraction of hex values.

This way we can call it twice: first to count the size of output (and
sanity check the extraction indices), and the second time to actually
extract.  Otherwise with bad extraction code we will smash the stack.

Furthermore, this makes for cleaner code since the large #ifdef forest
(about differing long double formats, and different uv/nv/quad sizes)
is moved within the helper function.

Furthermore: IEEE 754 128-bit extraction was extracting wrong:
we need only the mantissa bits, and in the right nybble order.

Last but not least, handle the 'implicit bit' of IEEE 754 quadruple
and double-double (but which is explicitly *not* in the 80-bit format).
(If we handled the standard 64-bit format by explicit byte extraction,
instead of frexp+ldexp the hidden bit would be needed there, too.)

9 years agoUpdate to include latest Test::Builder alpha
Chad Granum [Fri, 15 Aug 2014 15:40:10 +0000 (08:40 -0700)]
Update to include latest Test::Builder alpha

Also updated some tests that the new builder broke

9 years agoUpdate exp_name in toke.c, following b5b6210a976
Father Chrysostomos [Sat, 16 Aug 2014 04:52:12 +0000 (21:52 -0700)]
Update exp_name in toke.c, following b5b6210a976

Ooops.

$  ./miniperl -DT -e 'undef'
### 0:LEX_NORMAL/XSTATE "\n"
### <== UNIOP(ival=op_undef)

### 1:LEX_NORMAL/XNumber too long ";"  <------ Look here!
### <== ';'

### 1:LEX_NORMAL/XSTATE ""
### Tokener got EOF
### <== EOF

EXECUTING...

9 years agoSocket 2.015
Jarkko Hietaniemi [Sat, 16 Aug 2014 01:54:48 +0000 (21:54 -0400)]
Socket 2.015

9 years agoreg_posixcc.t: display chr(\\xXX) should be chr(0xXX)
Karl Williamson [Fri, 15 Aug 2014 16:34:53 +0000 (10:34 -0600)]
reg_posixcc.t: display chr(\\xXX) should be chr(0xXX)

The test name was misleading.  The operand to chr should be an ord.

9 years agoperldelta: Add entry for sync_locale()
Karl Williamson [Fri, 15 Aug 2014 15:05:28 +0000 (09:05 -0600)]
perldelta: Add entry for sync_locale()

9 years agoMake a wild guess at how to handle double-double.
Jarkko Hietaniemi [Fri, 15 Aug 2014 13:50:03 +0000 (09:50 -0400)]
Make a wild guess at how to handle double-double.

9 years agoLimit testing with 80-bit doubles.
Jarkko Hietaniemi [Fri, 15 Aug 2014 13:25:37 +0000 (09:25 -0400)]
Limit testing with 80-bit doubles.

(The test data would be wrong for 128-bit doubles.)

9 years ago80-bit big-endian extraction was also wrong.
Jarkko Hietaniemi [Fri, 15 Aug 2014 11:47:49 +0000 (07:47 -0400)]
80-bit big-endian extraction was also wrong.

Not that this format probably ever happens.

9 years agoTry preparing for weird UVSIZE or NV_MANT_DIG.
Jarkko Hietaniemi [Fri, 15 Aug 2014 02:20:16 +0000 (22:20 -0400)]
Try preparing for weird UVSIZE or NV_MANT_DIG.

(UVSIZE=4 and no quad, or NV_MANT_DIG=106, where quad would not help.)

9 years agoIn frexp+ldexp path simply prefer Uquad_t to UV.
Jarkko Hietaniemi [Fri, 15 Aug 2014 01:44:44 +0000 (21:44 -0400)]
In frexp+ldexp path simply prefer Uquad_t to UV.

9 years agoBig endian frexp+ldexp byte extraction was wrong.
Jarkko Hietaniemi [Fri, 15 Aug 2014 01:42:39 +0000 (21:42 -0400)]
Big endian frexp+ldexp byte extraction was wrong.

9 years agoOn literal hexadecimal floats use uquad, not UV.
Jarkko Hietaniemi [Thu, 14 Aug 2014 20:43:06 +0000 (16:43 -0400)]
On literal hexadecimal floats use uquad, not UV.

Since it is possible to have quads but not use64bitint.

9 years agoComment tweaks.
Jarkko Hietaniemi [Fri, 15 Aug 2014 01:06:06 +0000 (21:06 -0400)]
Comment tweaks.

9 years agoUse XDIGIT_VALUE().
Jarkko Hietaniemi [Thu, 14 Aug 2014 20:25:31 +0000 (16:25 -0400)]
Use XDIGIT_VALUE().

9 years agoZero is better longdblkind for "no long double" than -1.
Jarkko Hietaniemi [Thu, 14 Aug 2014 20:13:11 +0000 (16:13 -0400)]
Zero is better longdblkind for "no long double" than -1.

(zero = long double is double, -1 = unknown long double)

9 years agoWhitespace only: line breaks before elses.
Jarkko Hietaniemi [Thu, 14 Aug 2014 20:11:10 +0000 (16:11 -0400)]
Whitespace only: line breaks before elses.