Mailing List Archive: Perl: porters

Perl 5.16.0 RC0

1 2

View All

Index | Next | Previous | View Threaded

perl.p5p at rjbs

May 10, 2012, 8:38 PM

Post #1 of 28 (326 views)
Permalink

Perl 5.16.0 RC0

Cerebrotonic Cato may
Extol the Ancient Disciplines,
But the muscle-bound Marines
Mutiny for food and pay.

Caesar's double-bed is warm
As an unimportant clerk
Writes I DO NOT LIKE MY WORK
On a pink official form.

-- W. H. Auden, The Fall of Rome

I've just uploaded the first release candidate of Perl 5.16.0 to PAUSE.

The files should be available soon, with the following SHA sums:

3f08f8e00adc215ccbf6aff38c58eef58f4df102 perl-5.16.0-RC0.tar.bz2
99ce35a3e1a48b6936e3e2e56966cb6736181875 perl-5.16.0-RC0.tar.gz

While we go to lengths to ensure that new versions of Perl don't break
existing programs, it does happen. It's really, really important that
we catch unintentional breakage BEFORE we release Perl 5.16.0.
It is imperative that you test this release candidate with any software
written in Perl which you use or maintain.

Similarly, we test Perl on a variety of operating systems on a number of
platforms with several different compilers. If you're not 100% certain
that we're testing yours, NOW is the time to make sure that Perl 5.16.0
builds and passes its tests on your platform.

If no "showstopper" class bugs are found in the next 7 days, we will release
a virtually identical tarball as Perl 5.16.0 on or soon after Thursday, May
17, 2012.

--
rjbs

Attachments:

signature.asc (0.48 KB)

sprout at cpan

May 10, 2012, 9:57 PM

Post #2 of 28 (327 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

rjbs said:
> I've just uploaded the first release candidate of Perl 5.16.0 to PAUSE.

I see typos and formatting errors in the modules section of perldelta.
Where am I supposed to patch it? Are you going to merge it to blead,
to make things less confusing?

What happened with commit 1061b56a7b2cc? You seem to have reverted my
commit 1405d8c986, but in my name.

lespea at gmail

May 10, 2012, 11:37 PM

Post #3 of 28 (325 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

Awesome news!

One question, I was going through perldelta and I noticed that CPANPLUS is
scheduled to be removed from core?

~Adam~

h.m.brand at xs4all

May 11, 2012, 12:28 AM

Post #4 of 28 (329 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

On Thu, 10 May 2012 23:38:23 -0400, Ricardo Signes
<perl.p5p [at] rjbs> wrote:

>
> Cerebrotonic Cato may
> Extol the Ancient Disciplines,
> But the muscle-bound Marines
> Mutiny for food and pay.
>
> Caesar's double-bed is warm
> As an unimportant clerk
> Writes I DO NOT LIKE MY WORK
> On a pink official form.
>
> -- W. H. Auden, The Fall of Rome
>
> I've just uploaded the first release candidate of Perl 5.16.0 to PAUSE.
>
> The files should be available soon, with the following SHA sums:
>
> 3f08f8e00adc215ccbf6aff38c58eef58f4df102 perl-5.16.0-RC0.tar.bz2

Won't unpack

> 99ce35a3e1a48b6936e3e2e56966cb6736181875 perl-5.16.0-RC0.tar.gz

works fine

tested on Linux 3.1.10-1.9-desktop
i386 Core(TM) i7-2620M CPU @2.70GHz/2701(4) i686 7951 Mb

all OK for i686-linux-64int
and i686-linux-thread-multi-64int-ld

with LWP YAML HTTP::Date Expect CPAN::SQLite YAML::Syck YAML::Tiny
YAML::XS JSON::PP JSON Digest::SHA File::Temp Test::Harness
Test::More File::Temp ExtUtils::MakeMaker V Test::NoWarnings
Data::Peek Text::CSV_XS plus dependencies
--
H.Merijn Brand http://tux.nl Perl Monger http://amsterdam.pm.org/
using perl5.00307 .. 5.14 porting perl5 on HP-UX, AIX, and openSUSE
http://mirrors.develooper.com/hpux/ http://www.test-smoke.org/
http://qa.perl.org http://www.goldmark.org/jeff/stupid-disclaimers/

Steve.Hay at verosoftware

May 11, 2012, 2:10 AM

Post #5 of 28 (318 views)
Permalink

RE: Perl 5.16.0 RC0 [In reply to]

Ricardo Signes wrote on 2012-05-11:
> I've just uploaded the first release candidate of Perl 5.16.0 to
PAUSE.
>
> The files should be available soon, with the following SHA sums:
>
> 3f08f8e00adc215ccbf6aff38c58eef58f4df102 perl-5.16.0-RC0.tar.bz2
> 99ce35a3e1a48b6936e3e2e56966cb6736181875 perl-5.16.0-RC0.tar.gz

Default build out-of-the-box on Windows 7 x64 with VC++ 2010 SP1 is
sadly not a pleasant experience for me. As reported previously, numerous
tests hang and have to be killed, yielding the test results below. As
noted before, however, all is well aside from the intermittent fork.t
test 24 failure if STDERR is redirected somewhere.

It's a great shame that I've not been able to fix this. Is anyone else
seeing this, or is it just me? I get the same thing on two different
machines, so it's at least not just one particular machine.

Test Summary Report
-------------------
op/fork.t (Wstat:
0 Tests: 25 Failed: 1)
Failed test: 24
../cpan/CGI/t/tmpdir.t (Wstat:
0 Tests: 9 Failed: 0)
TODO passed: 3-9
../cpan/CPANPLUS/t/19_CPANPLUS-Dist.t (Wstat:
256 Tests: 43 Failed: 0)
Non-zero exit status: 1
Parse errors: No plan found in TAP output
../cpan/IPC-Cmd/t/01_IPC-Cmd.t (Wstat:
256 Tests: 18 Failed: 0)
Non-zero exit status: 1
Parse errors: No plan found in TAP output
../cpan/Module-Build/t/perl_mb_opt.t (Wstat:
512 Tests: 8 Failed: 2)
Failed tests: 4, 8
Non-zero exit status: 2
../cpan/Module-Build/t/runthrough.t (Wstat:
256 Tests: 8 Failed: 0)
Non-zero exit status: 1
Parse errors: Bad plan. You planned 29 tests but ran 8.
../cpan/Module-Build/t/test_types.t (Wstat:
256 Tests: 1 Failed: 0)
Non-zero exit status: 1
Parse errors: Bad plan. You planned 25 tests but ran 1.
../cpan/Module-Build/t/xs.t (Wstat:
256 Tests: 9 Failed: 0)
Non-zero exit status: 1
Parse errors: Bad plan. You planned 20 tests but ran 9.
../dist/IO/t/io_sock.t (Wstat:
0 Tests: 25 Failed: 0)
Parse errors: Tests out of sequence. Found (12) but expected (11)
Tests out of sequence. Found (13) but expected (12)
Tests out of sequence. Found (14) but expected (13)
Tests out of sequence. Found (15) but expected (14)
Tests out of sequence. Found (16) but expected (15)
Displayed the first 5 of 16 TAP syntax errors.
Re-run prove with the -p option to see them all.
Files=2339, Tests=527532, 2241 wallclock secs (51.56 usr + 5.87 sys =
57.42 CPU)
Result: FAIL
NMAKE : fatal error U1077: '..\perl.exe' : return code '0x3'
Stop.

tony at develop-help

May 11, 2012, 2:29 AM

Post #6 of 28 (316 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

On Fri, May 11, 2012 at 10:10:08AM +0100, Steve Hay wrote:
> Ricardo Signes wrote on 2012-05-11:
> > I've just uploaded the first release candidate of Perl 5.16.0 to
> PAUSE.
> >
> > The files should be available soon, with the following SHA sums:
> >
> > 3f08f8e00adc215ccbf6aff38c58eef58f4df102 perl-5.16.0-RC0.tar.bz2
> > 99ce35a3e1a48b6936e3e2e56966cb6736181875 perl-5.16.0-RC0.tar.gz
>
> Default build out-of-the-box on Windows 7 x64 with VC++ 2010 SP1 is
> sadly not a pleasant experience for me. As reported previously, numerous
> tests hang and have to be killed, yielding the test results below. As
> noted before, however, all is well aside from the intermittent fork.t
> test 24 failure if STDERR is redirected somewhere.

I get similar failures with the SDK compiler.

io_sock.t has been behaving for me recently though.

Tony

zefram at fysh

May 11, 2012, 4:15 AM

Post #7 of 28 (333 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

Ricardo Signes wrote:
> 3f08f8e00adc215ccbf6aff38c58eef58f4df102 perl-5.16.0-RC0.tar.bz2

Downloaded a file by this name that had a different SHA1 sum:

5cb530735dca0d180e5ea9fbbee99197bd7cccb0 perl-5.16.0-RC0.tar.bz2

> 99ce35a3e1a48b6936e3e2e56966cb6736181875 perl-5.16.0-RC0.tar.gz

Got this one intact. Built and tested fine for i686-linux-64int-ld.
Failed to install mod_perl from CPAN:

|Running install for module 'APR'
|Running make for P/PH/PHRED/mod_perl-2.0.6.tar.gz
|Checksum for /opt/perl-5.16.0-RC0/cpan/sources/authors/id/P/PH/PHRED/mod_perl-2.0.6.tar.gz ok
|
| CPAN.pm: Building P/PH/PHRED/mod_perl-2.0.6.tar.gz
|
|Please repair your Module::CoreList at lib/Apache2/Build.pm line 49.
|BEGIN failed--compilation aborted at lib/Apache2/Build.pm line 64.
|Compilation failed in require at Makefile.PL line 37.
|BEGIN failed--compilation aborted at Makefile.PL line 37.
|Warning: No success on command[/opt/perl-5.16.0-RC0/bin/perl Makefile.PL]
|'YAML' not installed, will not store persistent state
| PHRED/mod_perl-2.0.6.tar.gz
| /opt/perl-5.16.0-RC0/bin/perl Makefile.PL -- NOT OK
|Running make test
| Make had some problems, won't test
|Running make install
| Make had some problems, won't install

The problem is that %Module::CoreList::version contains the key "5.016"
whereas the string value of $] is "5.016000". Did we change how this
works somewhere?

-zefram

rgs at consttype

May 11, 2012, 4:25 AM

Post #8 of 28 (330 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

On 11 May 2012 13:15, Zefram <zefram [at] fysh> wrote:
> Ricardo Signes wrote:
>> �3f08f8e00adc215ccbf6aff38c58eef58f4df102 �perl-5.16.0-RC0.tar.bz2
>
> Downloaded a file by this name that had a different SHA1 sum:
>
> 5cb530735dca0d180e5ea9fbbee99197bd7cccb0 �perl-5.16.0-RC0.tar.bz2
>
>> �99ce35a3e1a48b6936e3e2e56966cb6736181875 �perl-5.16.0-RC0.tar.gz
>
> Got this one intact. �Built and tested fine for i686-linux-64int-ld.
> Failed to install mod_perl from CPAN:
>
> |Running install for module 'APR'
> |Running make for P/PH/PHRED/mod_perl-2.0.6.tar.gz
> |Checksum for /opt/perl-5.16.0-RC0/cpan/sources/authors/id/P/PH/PHRED/mod_perl-2.0.6.tar.gz ok
> |
> | �CPAN.pm: Building P/PH/PHRED/mod_perl-2.0.6.tar.gz
> |
> |Please repair your Module::CoreList at lib/Apache2/Build.pm line 49.
> |BEGIN failed--compilation aborted at lib/Apache2/Build.pm line 64.
> |Compilation failed in require at Makefile.PL line 37.
> |BEGIN failed--compilation aborted at Makefile.PL line 37.
> |Warning: No success on command[/opt/perl-5.16.0-RC0/bin/perl Makefile.PL]
> |'YAML' not installed, will not store persistent state
> | �PHRED/mod_perl-2.0.6.tar.gz
> | �/opt/perl-5.16.0-RC0/bin/perl Makefile.PL -- NOT OK
> |Running make test
> | �Make had some problems, won't test
> |Running make install
> | �Make had some problems, won't install
>
> The problem is that %Module::CoreList::version contains the key "5.016"
> whereas the string value of $] is "5.016000". �Did we change how this
> works somewhere?

You need both entries in the hashes, see patch below. I did not dare
pushing it yet. That part should probably have tests, too.

diff --git a/dist/Module-CoreList/lib/Module/CoreList.pm
b/dist/Module-CoreList/lib/Module/CoreList.pm
index 9b5b07a..ec6e1aa 100644
--- a/dist/Module-CoreList/lib/Module/CoreList.pm
+++ b/dist/Module-CoreList/lib/Module/CoreList.pm
@@ -34687,6 +34687,8 @@ $released{'5.011000'} = $released{5.011};
$released{'5.012000'} = $released{5.012};
$released{'5.013000'} = $released{5.013};
$released{'5.014000'} = $released{5.014};
+$released{'5.015000'} = $released{5.015};
+$released{'5.016000'} = $released{5.016};

$version{'5.000'} = $version{5};
$version{'5.010000'} = $version{5.01};
@@ -34694,11 +34696,15 @@ $version{'5.011000'} = $version{5.011};
$version{'5.012000'} = $version{5.012};
$version{'5.013000'} = $version{5.013};
$version{'5.014000'} = $version{5.014};
+$version{'5.015000'} = $version{5.015};
+$version{'5.016000'} = $version{5.016};

$deprecated{'5.011000'} = $deprecated{5.011};
$deprecated{'5.012000'} = $deprecated{5.012};
$deprecated{'5.013000'} = $deprecated{5.013};
$deprecated{'5.014000'} = $deprecated{5.014};
+$deprecated{'5.015000'} = $deprecated{5.015};
+$deprecated{'5.016000'} = $deprecated{5.016};

1;
__END__

perl.p5p at rjbs

May 11, 2012, 5:02 AM

Post #9 of 28 (317 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

* Zefram <zefram [at] fysh> [2012-05-11T07:15:21]
> Ricardo Signes wrote:
> > 3f08f8e00adc215ccbf6aff38c58eef58f4df102 perl-5.16.0-RC0.tar.bz2
>
> Downloaded a file by this name that had a different SHA1 sum:
>
> 5cb530735dca0d180e5ea9fbbee99197bd7cccb0 perl-5.16.0-RC0.tar.bz2

PAUSE seems to have failed to correctly retrieve the file to which I pointed
it.

~/Dropbox/Public/perl$ shasum *
3f08f8e00adc215ccbf6aff38c58eef58f4df102 perl-5.16.0-RC0.tar.bz2

You can get the file at

http://dl.dropbox.com/u/88746/perl/perl-5.16.0-RC0.tar.bz2

I will ask Andreas to replace it.

> The problem is that %Module::CoreList::version contains the key "5.016"
> whereas the string value of $] is "5.016000". Did we change how this
> works somewhere?

Thanks. I will fix this and will fix the corelist updater to add it. (I
relied on the automation too heavily.)

--
rjbs

Attachments:

signature.asc (0.48 KB)

perl.p5p at rjbs

May 11, 2012, 5:10 AM

Post #10 of 28 (316 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

* Father Chrysostomos <sprout [at] cpan> [2012-05-11T00:57:45]
> rjbs said:
> > I've just uploaded the first release candidate of Perl 5.16.0 to PAUSE.
>
> I see typos and formatting errors in the modules section of perldelta.
> Where am I supposed to patch it? Are you going to merge it to blead,
> to make things less confusing?

Yes, those are rough. They're mostly C&P directly from the previous deltas.
Dave Rolsky has already made some fixes, but more are welcome. Now that the
world didn't end, blead now descends from rjbs/release-5.16 and you should feel
free to make those perldelta fixes directly. Thanks.

> What happened with commit 1061b56a7b2cc? You seem to have reverted my
> commit 1405d8c986, but in my name.

During my repeated rebasing of rjbs/release-5.16 on blead, I encountered a few
perldelta changes and rebuild them onto my branch. I clearly screwed that one
up, and I'm really sorry! I've corrected it.

--
rjbs

Attachments:

signature.asc (0.48 KB)

perl.p5p at rjbs

May 11, 2012, 5:11 AM

Post #11 of 28 (316 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

* Adam <lespea [at] gmail> [2012-05-11T02:37:43]
> Awesome news!
>
> One question, I was going through perldelta and I noticed that CPANPLUS is
> scheduled to be removed from core?

It's not scheduled yet, but the tentative plan is deprecation in 5.18 and
removal in 5.20. Feel free to provide compelling arguments against (in a new
thread!).

--
rjbs

Attachments:

signature.asc (0.48 KB)

shlomif at shlomifish

May 11, 2012, 5:16 AM

Post #12 of 28 (315 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

Hi all,

On Thu, 10 May 2012 23:38:23 -0400
Ricardo Signes <perl.p5p [at] rjbs> wrote:

>
> Cerebrotonic Cato may
> Extol the Ancient Disciplines,
> But the muscle-bound Marines
> Mutiny for food and pay.
>
> Caesar's double-bed is warm
> As an unimportant clerk
> Writes I DO NOT LIKE MY WORK
> On a pink official form.
>
> -- W. H. Auden, The Fall of Rome
>
> I've just uploaded the first release candidate of Perl 5.16.0 to PAUSE.
>
> The files should be available soon, with the following SHA sums:
>
> 3f08f8e00adc215ccbf6aff38c58eef58f4df102 perl-5.16.0-RC0.tar.bz2
> 99ce35a3e1a48b6936e3e2e56966cb6736181875 perl-5.16.0-RC0.tar.gz
>
> While we go to lengths to ensure that new versions of Perl don't break
> existing programs, it does happen. It's really, really important that
> we catch unintentional breakage BEFORE we release Perl 5.16.0.
> It is imperative that you test this release candidate with any software
> written in Perl which you use or maintain.

OK, trying to install my CPAN distributions (and their dependencies), I found
that:

1. IO::All tests now emit a lot of warnings (due to a stray \E after a \u). I
fixed it in the main IO-All repository, but Ingy needs to release a new CPAN
distribution:

https://github.com/ingydotnet/io-all-pm

2. The tests of URI::Fetch are now broken on my machine (but also on
perl-5.14.2). I reported this bug report about it:

https://rt.cpan.org/Ticket/Display.html?id=77131

3. Most of my other modules were built and installed fine.

----

Next, I'm going to try to force install URI::Fetch and then try to build my
home site with it.

Regards,

Shlomi Fish

>
> Similarly, we test Perl on a variety of operating systems on a number of
> platforms with several different compilers. If you're not 100% certain
> that we're testing yours, NOW is the time to make sure that Perl 5.16.0
> builds and passes its tests on your platform.
>
> If no "showstopper" class bugs are found in the next 7 days, we will release
> a virtually identical tarball as Perl 5.16.0 on or soon after Thursday, May
> 17, 2012.
>

--
-----------------------------------------------------------------
Shlomi Fish http://www.shlomifish.org/
Optimising Code for Speed - http://shlom.in/optimise

You name it — COBOL does not have it.

Please reply to list if it's a mailing list post - http://shlom.in/reply .

zefram at fysh

May 11, 2012, 5:53 AM

Post #13 of 28 (313 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

After working round the Module::CoreList bug, it turns out mod_perl needs
an update for the removal of PL_uid et al. Reported as [rt.cpan.org
#77129]. Apart from that, all the CPAN deps of $ork codebase install
fine.

I'm getting some "Subroutine redefined" warnings from a non-CPAN module,
which didn't occur with 5.14.2 and fails some Test::NoWarnings tests.
I don't recall any deliberate change around this warning. Looking into
it.

-zefram

zefram at fysh

May 11, 2012, 6:12 AM

Post #14 of 28 (316 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

I wrote:
>I'm getting some "Subroutine redefined" warnings from a non-CPAN module,
>which didn't occur with 5.14.2

The warnings show up if I XSLoader::load() an XS module more than once.
Can reproduce it perfectly well with a CPAN XS module, if I manually do
the XSLoader::load(). Doesn't occur in 5.14.2. I'm not clear on what's
meant to happen here: who is responsible for making sure a module's BOOT
section only runs once? I guess the CPAN smokes won't have caught if
this affects anything on CPAN, since it's only a warning.

Anyway, I can certainly fix it for this module by adding guards to the
multiple load() calls. I already have such guards in XML-Easy, which
would otherwise exhibit this problem.

-zefram

dmcbride at cpan

May 11, 2012, 7:37 AM

Post #15 of 28 (313 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

On Thursday May 10 2012 11:38:23 PM Ricardo Signes wrote:
> I've just uploaded the first release candidate of Perl 5.16.0 to PAUSE.

Test Summary Report
-------------------
../cpan/CGI/t/tmpdir.t (Wstat: 0
Tests: 9 Failed: 0)
TODO passed: 3-9
Files=2333, Tests=530355, 671 wallclock secs (182.79 usr 59.90 sys + 1418.87
cusr 439.70 csys = 2101.26 CPU)
Result: PASS
Configured with: ../perl-5.16.0-RC0/Configure -des -Dcc=xlC_r -Duseshrplib -
Dusethreads -Dmksymlinks -Duse64bitall -Dprefix=/opt/darin/perl5_64

Initial testing of AIX (64-bit) seems okay. Will have to try a few actual
programs later :-)

Attachments:

signature.asc (0.19 KB)

sprout at cpan

May 11, 2012, 8:28 AM

Post #16 of 28 (313 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

Zeframe wrote:
> I wrote:
> >I'm getting some "Subroutine redefined" warnings from a non-CPAN module,
> >which didn't occur with 5.14.2
>
> > The warnings show up if I XSLoader::load() an XS module more than once.

This was a regression from 5.8, which I fixed. A consting patch in
5.10, believe it or not, reversed the logic in a if() condition.

jand at activestate

May 11, 2012, 1:09 PM

Post #17 of 28 (314 views)
Permalink

RE: Perl 5.16.0 RC0 [In reply to]

On Fri, 11 May 2012, Steve Hay wrote:
> > 99ce35a3e1a48b6936e3e2e56966cb6736181875 perl-5.16.0-RC0.tar.gz
>
> Default build out-of-the-box on Windows 7 x64 with VC++ 2010 SP1 is
> sadly not a pleasant experience for me. As reported previously, numerous
> tests hang and have to be killed, yielding the test results below. As
> noted before, however, all is well aside from the intermittent fork.t
> test 24 failure if STDERR is redirected somewhere.
>
> It's a great shame that I've not been able to fix this. Is anyone else
> seeing this, or is it just me? I get the same thing on two different
> machines, so it's at least not just one particular machine.

I don't get any hangs with my VC6 + 2003 PSDK builds, and just some
passing TODO tests as well as a single CPAN module failure:

Test Summary Report
-------------------
../cpan/CGI/t/tmpdir.t (Wstat: 0 Tests: 9 Failed: 0)
TODO passed: 3-9
../cpan/IPC-Cmd/t/01_IPC-Cmd.t (Wstat: 256 Tests: 459 Failed: 1)
Failed test: 20
Non-zero exit status: 1
Files=2339, Tests=528066, 1906 wallclock secs (27.03 usr + 90.13 sys = 117.16 CPU)
Result: FAIL

I've been offline for the last 3.5 weeks and can't remember if
the IPC-Cmd failure is new or not; the passing tmpdir.t TODO tests
have been happening for a while though.

Cheers,
-Jan

dennis.kaarsemaker at booking

May 11, 2012, 2:01 PM

Post #18 of 28 (314 views)
Permalink

RE: Perl 5.16.0 RC0 [In reply to]

On vr, 2012-05-11 at 10:10 +0100, Steve Hay wrote:
>
> Default build out-of-the-box on Windows 7 x64 with VC++ 2010 SP1 is
> sadly not a pleasant experience for me. As reported previously, numerous
> tests hang and have to be killed, yielding the test results below. As
> noted before, however, all is well aside from the intermittent fork.t
> test 24 failure if STDERR is redirected somewhere.
>
> It's a great shame that I've not been able to fix this. Is anyone else
> seeing this, or is it just me? I get the same thing on two different
> machines, so it's at least not just one particular machine.

I'm seeing the same on the win7 64/vc2010 express box I'm building for
p5p and blamed it on my inexperience with building perl on windows 7.

--
Dennis Kaarsemaker, Team Leader Unix Administrators
Booking.com
Herengracht 597, 1017 CE Amsterdam
Tel external +31 (0) 20 715 3409
Tel internal (7207) 3409

steve.m.hay at googlemail

May 11, 2012, 3:26 PM

Post #19 of 28 (291 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

On 11 May 2012 21:09, Jan Dubois <jand [at] activestate> wrote:
> On Fri, 11 May 2012, Steve Hay wrote:
>> > 99ce35a3e1a48b6936e3e2e56966cb6736181875 perl-5.16.0-RC0.tar.gz
>>
>> Default build out-of-the-box on Windows 7 x64 with VC++ 2010 SP1 is
>> sadly not a pleasant experience for me. As reported previously, numerous
>> tests hang and have to be killed, yielding the test results below. As
>> noted before, however, all is well aside from the intermittent fork.t
>> test 24 failure if STDERR is redirected somewhere.
>>
>> It's a great shame that I've not been able to fix this. Is anyone else
>> seeing this, or is it just me? I get the same thing on two different
>> machines, so it's at least not just one particular machine.
>
> I don't get any hangs with my VC6 + 2003 PSDK builds, and just some
> passing TODO tests as well as a single CPAN module failure:
>
> Test Summary Report
> -------------------
> ../cpan/CGI/t/tmpdir.t (Wstat: 0 Tests: 9 Failed: 0)
> TODO passed: 3-9
> ../cpan/IPC-Cmd/t/01_IPC-Cmd.t (Wstat: 256 Tests: 459 Failed: 1)
> Failed test: 20
> Non-zero exit status: 1
> Files=2339, Tests=528066, 1906 wallclock secs (27.03 usr + 90.13 sys = 117.16 CPU)
> Result: FAIL
>
> I've been offline for the last 3.5 weeks and can't remember if
> the IPC-Cmd failure is new or not; the passing tmpdir.t TODO tests
> have been happening for a while though.
>
> Cheers,
> -Jan
>

What OS was that on?

I've just tried again with VC6 SP6, VC8 SP1 and VC10 SP1 and I get the
exact same set of tests hanging with all three of them. This is
WIndows 7 64-bit. I'm specifying WIN64=undef in the build for a 32-bit
build, but otherwise have changed nothing except the CCTYPE.

(The fork.t failure is interrmittent; so is the io_sock failure. But
the *hangs* are 100% reproducible every time on both machines
available to me, albeit both Win7 x64...)

jand at activestate

May 11, 2012, 3:51 PM

Post #20 of 28 (292 views)
Permalink

RE: Perl 5.16.0 RC0 [In reply to]

On Fri, 11 May 2012, Steve Hay wrote:
> On 11 May 2012 21:09, Jan Dubois <jand [at] activestate> wrote:
>> I don't get any hangs with my VC6 + 2003 PSDK builds, and just some
>> passing TODO tests as well as a single CPAN module failure:
>
> What OS was that on?
>
> I've just tried again with VC6 SP6, VC8 SP1 and VC10 SP1 and I get the
> exact same set of tests hanging with all three of them. This is
> WIndows 7 64-bit. I'm specifying WIN64=undef in the build for a 32-bit
> build, but otherwise have changed nothing except the CCTYPE.
>
> (The fork.t failure is interrmittent; so is the io_sock failure. But
> the *hangs* are 100% reproducible every time on both machines
> available to me, albeit both Win7 x64...)

It was on a lean little Win2K VM. I'll see if I can set up a Win7
machine for testing, but it may take me a little while. I much prefer
using the obsolete Windows versions as they run in comparatively tiny
amounts of memory. :)

Cheers,
-Jan

kmx at atlas

May 12, 2012, 5:55 AM

Post #21 of 28 (291 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

> It's a great shame that I've not been able to fix this. Is anyone else
> seeing this, or is it just me? I get the same thing on two different
> machines, so it's at least not just one particular machine.

I see more or less the same failures and hanging tests on Win7/64bit box
with 32bit perl-5.16-RC0 build (by 32bit gcc 4.6.3 + C-runtime from
mingw-w64.sf.net)

However, on the same box (Win7/64bit) with 64bit build of the same perl (by
64bit gcc 4.6.3 + C-runtime from mingw-w64.sf.net) surprisingly *no tests
hang* and the overall result is:

Test Summary Report
-------------------
../cpan/CGI/t/tmpdir.t (Wstat: 0 Tests: 9 Failed: 0)
TODO passed: 3-9
../dist/Net-Ping/t/510_ping_udp.t (Wstat: 256 Tests: 2 Failed: 1)
Failed test: 2
Non-zero exit status: 1
Files=2342, Tests=528963, 1552 wallclock secs (54.94 usr + 3.89 sys =
58.83 CPU)
Result: FAIL
dmake: Error code 129, while making 'test'

--
kmx

kmx at atlas

May 12, 2012, 6:16 AM

Post #22 of 28 (292 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

> I see more or less the same failures and hanging tests on Win7/64bit box
> with 32bit perl-5.16-RC0 build (by 32bit gcc 4.6.3 + C-runtime from
> mingw-w64.sf.net)

Just to make it complete - with redirected stderr the result on 32bit
gcc-4.6.3 build was:

Test Summary Report
-------------------
op/fork.t (Wstat: 0 Tests: 25 Failed: 2)
Failed tests: 23-24
../cpan/CGI/t/tmpdir.t (Wstat: 0 Tests: 9 Failed: 0)
TODO passed: 3-9
../dist/IO/t/cachepropagate-tcp.t (Wstat: 256 Tests: 8 Failed: 1)
Failed test: 6
Non-zero exit status: 1
Files=2342, Tests=528728, 1679 wallclock secs (58.77 usr + 4.34 sys =
63.10 CPU)
Result: FAIL

--
kmx

paul at pjcj

May 12, 2012, 7:42 AM

Post #23 of 28 (292 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

On Thu, May 10, 2012 at 11:38:23PM -0400, Ricardo Signes wrote:

> I've just uploaded the first release candidate of Perl 5.16.0 to PAUSE.

> While we go to lengths to ensure that new versions of Perl don't break
> existing programs, it does happen. It's really, really important that
> we catch unintentional breakage BEFORE we release Perl 5.16.0.
> It is imperative that you test this release candidate with any software
> written in Perl which you use or maintain.

I suppose this is more of a BBC report, but the only problems I've found
with the application I'm running are to do with defined(@array) now
issuing a deprecation warning. I use three such modules, and two
already had an RT ticket with a patch. I've added a ticket for the
third. The modules are:

YAML - https://rt.cpan.org/Public/Bug/Display.html?id=74226
Module::Find - https://rt.cpan.org/Public/Bug/Display.html?id=74251
Net::Server - https://rt.cpan.org/Public/Bug/Display.html?id=77166

Otherwise 5.16 is looking good running against Dancer, DBIx::Class,
Moose, Plack, Starman, Devel::Cover and a bunch of related stuff.

--
Paul Johnson - paul [at] pjcj
http://www.pjcj.net

fraserbn at gmail

May 12, 2012, 2:45 PM

Post #24 of 28 (293 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

On Fri, May 11, 2012 at 9:02 AM, Ricardo Signes
<perl.p5p [at] rjbs>wrote:

> * Zefram <zefram [at] fysh> [2012-05-11T07:15:21]
> > Ricardo Signes wrote:
> > > 3f08f8e00adc215ccbf6aff38c58eef58f4df102 perl-5.16.0-RC0.tar.bz2
> >
> > Downloaded a file by this name that had a different SHA1 sum:
> >
> > 5cb530735dca0d180e5ea9fbbee99197bd7cccb0 perl-5.16.0-RC0.tar.bz2
>
> PAUSE seems to have failed to correctly retrieve the file to which I
> pointed
> it.
>
> ~/Dropbox/Public/perl$ shasum *
> 3f08f8e00adc215ccbf6aff38c58eef58f4df102 perl-5.16.0-RC0.tar.bz2
>
> You can get the file at
>
> http://dl.dropbox.com/u/88746/perl/perl-5.16.0-RC0.tar.bz2
>
>
Solaris 10/SPARC, suncc: All tests successful.

steve.m.hay at googlemail

May 13, 2012, 5:15 PM

Post #25 of 28 (291 views)
Permalink

Re: Perl 5.16.0 RC0 [In reply to]

On 12 May 2012 13:55, kmx <kmx [at] atlas> wrote:
>
>> It's a great shame that I've not been able to fix this. Is anyone else
>> seeing this, or is it just me? I get the same thing on two different
>> machines, so it's at least not just one particular machine.
>
> I see more or less the same failures and hanging tests on Win7/64bit box
> with 32bit perl-5.16-RC0 build (by 32bit gcc 4.6.3 + C-runtime from
> mingw-w64.sf.net)
>
> However, on the same box (Win7/64bit) with 64bit build of the same perl (by
> 64bit gcc 4.6.3 + C-runtime from mingw-w64.sf.net) surprisingly *no tests
> hang*
>

Interesting... However, I've just tried a 64-bit build using VC++ 2010
and I still get tests hanging as before :-(

1 2

View All

Index | Next | Previous | View Threaded

Interested in having your list archived? Contact Gossamer Threads