Bug #112021 for Math-MPFR: libmath-mpfr-perl: FTBFS on most architectures

Sun Feb 14 13:22:55 2016 gregoa [...] cpan.org - Ticket created

From:	gregoa [...] cpan.org
Subject:	libmath-mpfr-perl: FTBFS on most architectures

We have the following bug reported to the Debian package of Math-MPFR (https://bugs.debian.org/814730): It doesn't seem to be a bug in the packaging, so you may want to take a look. Thanks! ------8<-----------8<-----------8<-----------8<-----------8<----- Source: libmath-mpfr-perl Version: 3.29-1 Severity: serious Justification: fails to build from source (but built successfully in the past) libmath-mpfr-perl 3.29-1 and 3.30-1 don't build on most architectures: https://buildd.debian.org/status/logs.php?pkg=libmath-mpfr-perl Cheers, gregor ------8<-----------8<-----------8<-----------8<-----------8<----- Thanks for considering, gregor herrmann, Debian Perl Group

Sun Feb 14 18:22:21 2016 sisyphus [...] cpan.org - Correspondence added

Thanks Gregor. The failures I've looked at have all been the same - with DoubleDouble2.t and bytes.t being the 2 failing test scripts. Are you aware of any other failing tests ? This is a bug which should affect any machine that uses the double-double configuration when (and only when) perl's nvtype is "double". They pertain to a feature that was added in 3.29. I do have a ppc64 box running Debian Wheezy - but, stupidly, I've only been testing Math-MPFR on perls whose nvtype is "long double". When, on this ppc64 box, I switch to a perl whose nvtype is "double" then I see exactly the same failures. In future I'll make a point of checking Math-MPFR against a "double" build on my ppc64 machine - sorry for the inconvenience to you. The t/DoubleDouble2.t failure is a bug in the test script - and the t/bytes.t failures might also be a bug in that test script (not sure yet). I'll get this fixed, upload a Math-MPFR-3.31 and announce here when that's done. Thanks for reporting. Cheers, Rob

Sun Feb 14 18:22:21 2016 The RT System itself - Status changed from 'new' to 'open'

Mon Feb 15 05:24:20 2016 sisyphus [...] cpan.org - Correspondence added

Math-MPFR-3.31 has now been uploaded to CPAN. It gets rid of the failures (for me, at least). Only changes from 3.30 to 3.31 are in t/DoubleDouble.t (minor cleanup), t/DoubleDouble2.t (bugfix for this ticket) and t/bytes.t (bugfix for this ticket). I'll leave this ticket open until it's verified that these changes have, in fact, resolved this ticket. Cheers, Rob

Mon Feb 15 11:49:59 2016 gregoa [...] debian.org - Correspondence added

Subject:	Re: [rt.cpan.org #112021] libmath-mpfr-perl: FTBFS on most architectures
Date:	Mon, 15 Feb 2016 17:49:40 +0100
To:	Sisyphus via RT <bug-Math-MPFR [...] rt.cpan.org>
From:	gregor herrmann <gregoa [...] debian.org>

On Mon, 15 Feb 2016 05:24:21 -0500, Sisyphus via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=112021 > > > Math-MPFR-3.31 has now been uploaded to CPAN.

Thanks, much appreciated! Show quoted text

> It gets rid of the failures (for me, at least). > Only changes from 3.30 to 3.31 are in t/DoubleDouble.t (minor > cleanup), t/DoubleDouble2.t (bugfix for this ticket) and t/bytes.t > (bugfix for this ticket). > > I'll leave this ticket open until it's verified that these changes > have, in fact, resolved this ticket.

I've uploaded 3.31 to Debian now, and we can watch the build results at https://buildd.debian.org/status/logs.php?pkg=libmath-mpfr-perl or https://buildd.debian.org/status/package.php?p=libmath-mpfr-perl (I also left the Debian bug report open until we have the results.) Thanks again, gregor -- .''`. Homepage https://info.comodo.priv.at/ - OpenPGP key 0xBB3A68018649AA06 : :' : Debian GNU/Linux user, admin, and developer - https://www.debian.org/ `. `' Member of VIBE!AT & SPI, fellow of the Free Software Foundation Europe `- NP: Novi Sad: On The Ground

Download signature.asc
application/pgp-signature 949b

Message body not shown because it is not plain text.

Mon Feb 15 21:01:46 2016 sisyphus [...] cpan.org - Correspondence added

On Mon Feb 15 11:49:59 2016, gregoa@debian.org wrote: Show quoted text

> I've uploaded 3.31 to Debian now, and we can watch the build results > at > https://buildd.debian.org/status/logs.php?pkg=libmath-mpfr-perl or > https://buildd.debian.org/status/package.php?p=libmath-mpfr-perl

Damn - looks like there's another architecture (that I didn't notice) throwing essentially the same errors. Looking at https://buildd.debian.org/status/fetch.php?pkg=libmath-mpfr-perl&arch=arm64&ver=3.31-1&stamp=1455556770 it seems you guys have machines with 128-bit (113-bit precision) long doubles. Is that right ? Where can I find the 'perl -V' output of these perls ? Again, it's just going to be a matter of fixing DoubleDouble2.t and bytes.t to accommodate this and releasing Math-MPFR-3.32 (which I'll hopefully get done today). Cheers, Rob

Mon Feb 15 21:49:13 2016 sisyphus [...] cpan.org - Correspondence added

On Mon Feb 15 21:01:46 2016, SISYPHUS wrote: Show quoted text

> Damn - looks like there's another architecture (that I didn't notice) > throwing essentially the same errors.

Actually, another *2* architectures, as there's also the machines with the 53-bit precision long double that also suffer from the mistakes I've made in t/bytes.t. I'm wondering why that hasn't been picked up by cpantesters as I know there's some testers running with 53-bit precision long doubles - perhaps those guys don't have mpfr installed. Getting there .... I hope ;-) Not related, but it would be nice to have a machine sitting here that provided me with a 128-bit long double. Are any such machines generally available reasonably cheaply - ie second hand and cheap ? (I do have machines that provide me with a 128-bit __float128, so it's probably a luxury that I don't really need, especially now that perl can be built with nvtype of __float128.) Cheers, Rob

Tue Feb 16 02:26:20 2016 gregoa [...] debian.org - Correspondence added

CC:	perl [...] packages.debian.org
Subject:	Re: [rt.cpan.org #112021] libmath-mpfr-perl: FTBFS on most architectures
Date:	Tue, 16 Feb 2016 08:25:56 +0100
To:	Sisyphus via RT <bug-Math-MPFR [...] rt.cpan.org>
From:	gregor herrmann <gregoa [...] debian.org>

On Mon, 15 Feb 2016 21:01:47 -0500, Sisyphus via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=112021 > > > On Mon Feb 15 11:49:59 2016, gregoa@debian.org wrote: >

> > I've uploaded 3.31 to Debian now, and we can watch the build results > > at > > https://buildd.debian.org/status/logs.php?pkg=libmath-mpfr-perl or > > https://buildd.debian.org/status/package.php?p=libmath-mpfr-perl

> > Damn - looks like there's another architecture (that I didn't > notice) throwing essentially the same errors.

Yeah, there's quite a few architectures out there :) Show quoted text

> Looking at https://buildd.debian.org/status/fetch.php?pkg=libmath-mpfr-perl&arch=arm64&ver=3.31-1&stamp=1455556770 > it seems you guys have machines with 128-bit (113-bit precision) > long doubles. Is that right ?

I'm cc'ing the Debian perl maintainers, Niko and Dom probably know more about how perl is configured on which architecture. Show quoted text

> Where can I find the 'perl -V' output of these perls ?

Does https://anonscm.debian.org/cgit/perl/perl.git/tree/debian/cross help? That's the config.sh files generated during build, collected for some architectures. Otherwise I need to log into the porterboxes to run perl -V there. Cheers, gregor -- .''`. Homepage: http://info.comodo.priv.at/ - OpenPGP key 0xBB3A68018649AA06 : :' : Debian GNU/Linux user, admin, and developer - https://www.debian.org/ `. `' Member of VIBE!AT & SPI, fellow of the Free Software Foundation Europe `- BOFH excuse #407: Route flapping at the NAP.

Tue Feb 16 02:35:59 2016 ntyni [...] debian.org - Correspondence added

CC:	Sisyphus via RT <bug-Math-MPFR [...] rt.cpan.org>, perl [...] packages.debian.org
Subject:	Re: [rt.cpan.org #112021] libmath-mpfr-perl: FTBFS on most architectures
Date:	Tue, 16 Feb 2016 09:35:31 +0200
To:	gregor herrmann <gregoa [...] debian.org>
From:	Niko Tyni <ntyni [...] debian.org>

On Tue, Feb 16, 2016 at 08:25:56AM +0100, gregor herrmann wrote: Show quoted text

> On Mon, 15 Feb 2016 21:01:47 -0500, Sisyphus via RT wrote:

> > <URL: https://rt.cpan.org/Ticket/Display.html?id=112021 >

Show quoted text

> > Looking at https://buildd.debian.org/status/fetch.php?pkg=libmath-mpfr-perl&arch=arm64&ver=3.31-1&stamp=1455556770 > > it seems you guys have machines with 128-bit (113-bit precision) > > long doubles. Is that right ?

I think most of the architectures have those, but we're not configuring perl with -Duselongdouble anywhere. Show quoted text

> > Where can I find the 'perl -V' output of these perls ?

https://buildd.debian.org/status/package.php?p=perl&suite=unstable has the build logs, and we run 'perl -V' as part of the build. Look for 'Summary of my perl5' in the logs. For arm64: Summary of my perl5 (revision 5 version 22 subversion 1) configuration: Platform: osname=linux, osvers=3.16.0, archname=aarch64-linux-gnu-thread-multi uname='linux localhost 3.16.0 #1 smp debian 3.16.0 aarch64 gnulinux ' config_args='-Dusethreads -Duselargefiles -Dcc=aarch64-linux-gnu-gcc -Dcpp=aarch64-linux-gnu-cpp -Dld=aarch64-linux-gnu-gcc -Dccflags=-DDEBIAN -Wdate-time -D_FORTIFY_SOURCE=2 -g -O2 -fstack-protector-strong -Wformat -Werror=format-security -Dldflags= -Wl,-z,relro -Dlddlflags=-shared -Wl,-z,relro -Dcccdlflags=-fPIC -Darchname=aarch64-linux-gnu -Dprefix=/usr -Dprivlib=/usr/share/perl/5.22 -Darchlib=/usr/lib/aarch64-linux-gnu/perl/5.22 -Dvendorprefix=/usr -Dvendorlib=/usr/share/perl5 -Dvendorarch=/usr/lib/aarch64-linux-gnu/perl5/5.22 -Dsiteprefix=/usr/local -Dsitelib=/usr/local/share/perl/5.22.1 -Dsitearch=/usr/local/lib/aarch64-linux-gnu/perl/5.22.1 -Dman1dir=/usr/share/man/man1 -Dman3dir=/usr/share/man/man3 -Dsiteman1dir=/usr/local/man/man1 -Dsiteman3dir=/usr/local/man/man3 -Duse64bitint -Dman1ext=1 -Dman3ext=3perl -Dpager=/usr/bin/sensible-pager -Uafs -Ud_csh -Ud_ualarm -Uusesfio -Uusenm -Ui_libutil -Uversiononly -DDEBUGGING=-g -Doptimize=-O2 -dEs -Uuseshrplib -Dotherlibdirs=/usr/lib/aarch64-linux-gnu/perl-base' hint=recommended, useposix=true, d_sigaction=define useithreads=define, usemultiplicity=define use64bitint=define, use64bitall=define, uselongdouble=undef usemymalloc=n, bincompat5005=undef Compiler: cc='aarch64-linux-gnu-gcc', ccflags ='-D_REENTRANT -D_GNU_SOURCE -DDEBIAN -Wdate-time -D_FORTIFY_SOURCE=2 -g -O2 -fstack-protector-strong -Wformat -Werror=format-security -fwrapv -fno-strict-aliasing -pipe -I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64', optimize='-O2 -g', cppflags='-D_REENTRANT -D_GNU_SOURCE -DDEBIAN -Wdate-time -D_FORTIFY_SOURCE=2 -g -O2 -fstack-protector-strong -Wformat -Werror=format-security -fwrapv -fno-strict-aliasing -pipe -I/usr/local/include' ccversion='', gccversion='5.3.1 20160205', gccosandvers='' intsize=4, longsize=8, ptrsize=8, doublesize=8, byteorder=12345678, doublekind=3 d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16, longdblkind=1 ivtype='long', ivsize=8, nvtype='double', nvsize=8, Off_t='off_t', lseeksize=8 alignbytes=8, prototype=define Linker and Libraries: ld='aarch64-linux-gnu-gcc', ldflags =' -Wl,-z,relro -fstack-protector-strong -L/usr/local/lib' libpth=/usr/local/lib /usr/lib/gcc/aarch64-linux-gnu/5/include-fixed /usr/include/aarch64-linux-gnu /usr/lib /lib/aarch64-linux-gnu /lib/../lib /usr/lib/aarch64-linux-gnu /usr/lib/../lib /lib libs=-lgdbm -lgdbm_compat -ldb -ldl -lm -lpthread -lc -lcrypt perllibs=-ldl -lm -lpthread -lc -lcrypt libc=libc-2.21.so, so=so, useshrplib=false, libperl=libperl.a gnulibc_version='2.21' Dynamic Linking: dlsrc=dl_dlopen.xs, dlext=so, d_dlsymun=undef, ccdlflags='-Wl,-E' cccdlflags='-fPIC', lddlflags='-shared -Wl,-z,relro -L/usr/local/lib -fstack-protector-strong' Characteristics of this binary (from libperl): Compile-time options: HAS_TIMES MULTIPLICITY PERLIO_LAYERS PERL_DONT_CREATE_GVSV PERL_HASH_FUNC_ONE_AT_A_TIME_HARD PERL_IMPLICIT_CONTEXT PERL_MALLOC_WRAP PERL_NEW_COPY_ON_WRITE PERL_PRESERVE_IVUV USE_64_BIT_ALL USE_64_BIT_INT USE_ITHREADS USE_LARGE_FILES USE_LOCALE USE_LOCALE_COLLATE USE_LOCALE_CTYPE USE_LOCALE_NUMERIC USE_LOCALE_TIME USE_PERLIO USE_PERL_ATOF USE_REENTRANT_API [...] Hope this helps, -- Niko Tyni ntyni@debian.org

Tue Feb 16 05:23:30 2016 sisyphus [...] cpan.org - Correspondence added

On Tue Feb 16 02:35:59 2016, ntyni@debian.org wrote:> > > Where can I find the 'perl -V' output of these perls ? Show quoted text

> > https://buildd.debian.org/status/package.php?p=perl&suite=unstable > > has the build logs, and we run 'perl -V' as part of the build.

Thanks for that. This is a handy reference for future, too. Math-MPFR-3.32.tar.gz has been uploaded to CPAN, and version 3.31 has been marked for removal to backpan (and will be removed on 19th Feb). I *do* hope I've got it right for all architectures this time. I'll check the Debian build logs periodically to see how it's going. Cheers, Rob

Tue Feb 16 10:35:46 2016 gregoa [...] debian.org - Correspondence added

Subject:	Re: [rt.cpan.org #112021] libmath-mpfr-perl: FTBFS on most architectures
Date:	Tue, 16 Feb 2016 16:35:12 +0100
To:	Sisyphus via RT <bug-Math-MPFR [...] rt.cpan.org>
From:	gregor herrmann <gregoa [...] debian.org>

On Tue, 16 Feb 2016 05:23:36 -0500, Sisyphus via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=112021 >

Show quoted text

> Math-MPFR-3.32.tar.gz has been uploaded to CPAN, and version 3.31 has been marked for removal to backpan (and will be removed on 19th Feb).

Thank you! Uploaded to the Debian buildds now. Show quoted text

> I *do* hope I've got it right for all architectures this time. > I'll check the Debian build logs periodically to see how it's going.

Yeah, me too :) Cheers, gregor -- .''`. Homepage https://info.comodo.priv.at/ - OpenPGP key 0xBB3A68018649AA06 : :' : Debian GNU/Linux user, admin, and developer - https://www.debian.org/ `. `' Member of VIBE!AT & SPI, fellow of the Free Software Foundation Europe `- NP: Gypsy Sun And Rainbows: Star Spangled Banner

Download signature.asc
application/pgp-signature 949b

Message body not shown because it is not plain text.

Wed Feb 17 07:09:45 2016 sisyphus [...] cpan.org - Correspondence added

Seems ok now, so I'll mark this as resolved. Please re-open if I've missed something. Cheers, Rob

Wed Feb 17 07:15:29 2016 sisyphus [...] cpan.org - Correspondence added

Resolved. Cheers, Rob

Wed Feb 17 07:16:43 2016 sisyphus [...] cpan.org - Status changed from 'open' to 'resolved'

Wed Feb 17 11:03:11 2016 gregoa [...] debian.org - Correspondence added

Subject:	Re: [rt.cpan.org #112021] libmath-mpfr-perl: FTBFS on most architectures
Date:	Wed, 17 Feb 2016 17:02:51 +0100
To:	Sisyphus via RT <bug-Math-MPFR [...] rt.cpan.org>
From:	gregor herrmann <gregoa [...] debian.org>

On Wed, 17 Feb 2016 07:09:51 -0500, Sisyphus via RT wrote: Show quoted text

> <URL: https://rt.cpan.org/Ticket/Display.html?id=112021 > > > Seems ok now, so I'll mark this as resolved. > Please re-open if I've missed something.

Ack, the build logs present a nice piccture now. Thanks again for your work! Cheers, gregor -- .''`. Homepage https://info.comodo.priv.at/ - OpenPGP key 0xBB3A68018649AA06 : :' : Debian GNU/Linux user, admin, and developer - https://www.debian.org/ `. `' Member of VIBE!AT & SPI, fellow of the Free Software Foundation Europe `-

Download signature.asc
application/pgp-signature 949b

Message body not shown because it is not plain text.