5.28.1 with -Duseshrplib issues

--Sig_/xmyKaSO1ATHnhYNZ=j2vpnH
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

I built, tested and installed 5.28.1 on my laptop using

../Configure -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads -Duses=
hrplib -des

I installed it into my usual place: /pro/bin/
the I linked /pro/bin/perl to /pro/bin/perl5.28.0 again to not disrupt
my other tasks while installing all the modules in the 5.28.1 tree

I always forbid to use modules from the previous installs:

$ perl -V:.*inc_version.*
d_inc_version_list=3D'undef';
inc_version_list=3D' ';
inc_version_list_init=3D'0';

What I then not is that when using CPAN in a script like

 $ perl5.28.1 -MCPAN::Shell -we 'CPAN::Shell->install ($_) for qw( ... )'

all modules that use ExtUtil::MakeMaker install as expected, as the usual r=
iddle is

 perl5.28.1 Makefile.PL && make && make test && make install

where make invokes the perl used to build the Makefile: perl5.28.1

and all modules that use Module::Build fail, as the use

 perl5.28.1 Build.PL && ./Build && ./Build test && ./Build install

and the shebang of Build is /pro/bin/perl which points to the previous perl

My previous perl (5.28.0) was NOT built with -Duseshrplib

I (by now) do not know if having the previous perl not having a shared
plib has any influence on the bad behavior of Module::Build and if the
causes havoc on my current perl, but in the new perl I now get some
fails I never saw before:

Image-Scale-0.14-0 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

t/stringify.t ...... perl5.28.1: src/image.c:48: image_init: Assertion `PL_=
valid_types_PVX[SvTYPE(_svpvx) & SVt_MASK]' failed.
t/stringify.t ...... Failed 3/3 subtests

indirect-0.38-0 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

t/40-threads.t ............. # Using threads 2.22
# Using threads::shared 1.58
perl5.28.1: indirect.xs:264: indirect_map_store: Assertion `((void*)(oi->bu=
f)) !=3D 0' failed.
t/40-threads.t ............. No subtests run

t/42-threads-global.t ...... # Using threads 2.22
# Using threads::shared 1.58
perl5.28.1: indirect.xs:264: indirect_map_store: Assertion `((void*)(oi->bu=
f)) !=3D 0' failed.
t/42-threads-global.t ...... No subtests run

Text-Aspell-0.09-0 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

t/05-core.t .. 1/19 panic: XSUB Text::Aspell::list_dictionaries (Aspell.c) =
failed to extend arg stack: base=3D694100, sp=3D6942a8, hwm=3D694110
# Looks like your test exited with 2 just after 15.
t/05-core.t .. Dubious, test returned 2 (wstat 512, 0x200)
Failed 4/19 subtests

Math-Cephes-0.5305-0 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D

t/elliptics.t ........ 1/10 panic: XSUB Math::Cephesc::ellpj (Cephes_wrap.c=
) failed to extend arg stack: base=3Df5e100, sp=3Df5e128, hwm=3Df5e118
t/elliptics.t ........ Dubious, test returned 255 (wstat 65280, 0xff00)
Failed 5/10 subtests

t/misc.t ............. 1/33 panic: XSUB Math::Cephesc::fresnl (Cephes_wrap.=
c) failed to extend arg stack: base=3D1bd0100, sp=3D1bd0118, hwm=3D1bd0110
t/misc.t ............. Dubious, test returned 255 (wstat 65280, 0xff00)
Failed 29/33 subtests

probably some more.

Does this mean I have to start all over without shared plib or did I
surface a bug in perl core?

--=20
H.Merijn Brand  http://tux.nl   Perl Monger  http://amsterdam.pm.org/
using perl5.00307 .. 5.29   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/

--Sig_/xmyKaSO1ATHnhYNZ=j2vpnH
Content-Type: application/pgp-signature
Content-Description: OpenPGP digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJcB5QYAAoJEAOhR6E+XcCY+lAIAIqv3WpVXgZTp3GiJqN+7RZN
9litQMsamyEZFJiMWMBBYj7s3T1gs+pSnWFQvNImUP1//Vi4tDcWLEJwtB4h5y2W
uc5FtbuvTdUmIUimOtG/LKYYfO+7VtqJzPaW/xTmLl1iP3LxLkleY5JYZEpPNX3p
hoBxcH/smjNCyyUJgQotNt/tP3lFHJpVHOI6w+V9USSvqEegFxh9qI7eF5k0hnvU
3ClKe7a7jsr/rX3IRViHeDewzN4qGUKBR4l0PcWFxKNWRN8SMSIKyi3/PAwSl72j
bW5JS91EBwbpnhLEuesNlMcZVaGrLHyXCxRqirHFP/cJ2UYRIjXvuJ65c4EOdYo=
=q+5P
-----END PGP SIGNATURE-----

--Sig_/xmyKaSO1ATHnhYNZ=j2vpnH--
0
h
12/5/2018 9:02:07 AM
perl.perl5.porters 47504 articles. 0 followers. Follow

10 Replies
37 Views

Similar Articles

[PageSpeed] 19
See related articles to this posting 


--Sig_/J.9.=9fdiLcm5gLbsIBO4hl
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

On Wed, 5 Dec 2018 10:02:07 +0100, "H.Merijn Brand"
<h.m.brand@xs4all.nl> wrote:

> I built, tested and installed 5.28.1 on my laptop using
>=20
> ./Configure -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads -Dus=
eshrplib -des
>=20
> I installed it into my usual place: /pro/bin/
> the I linked /pro/bin/perl to /pro/bin/perl5.28.0 again to not disrupt
> my other tasks while installing all the modules in the 5.28.1 tree
>
> [...]
>=20
> probably some more.

One that blocks quite a few:

Scope-Upper-0.31-0 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D

t/60-uplevel-target.t ............ 1/93 perl5.28.1: Upper.xs:2030: su_uplev=
el_runops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D =
0' failed.
t/60-uplevel-target.t ............ Failed 40/93 subtests
t/61-uplevel-args.t .............. 1/297 perl5.28.1: Upper.xs:2030: su_uple=
vel_runops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D=
 0' failed.
t/61-uplevel-args.t .............. Failed 257/297 subtests
t/62-uplevel-return.t ............ 1/69 perl5.28.1: Upper.xs:2030: su_uplev=
el_runops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D =
0' failed.
t/62-uplevel-return.t ............ Failed 25/69 subtests
t/63-uplevel-ctl.t ............... 1/51 perl5.28.1: Upper.xs:2030: su_uplev=
el_runops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D =
0' failed.
t/63-uplevel-ctl.t ............... Failed 47/51 subtests
t/64-uplevel-caller.t ............ 1/32 perl5.28.1: Upper.xs:2030: su_uplev=
el_runops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D =
0' failed.
t/64-uplevel-caller.t ............ Failed 8/32 subtests
t/65-uplevel-multi.t ............. ok
t/66-uplevel-context.t ........... perl5.28.1: Upper.xs:2030: su_uplevel_ru=
nops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fa=
iled.
t/66-uplevel-context.t ........... Failed 6/6 subtests
t/67-uplevel-scope.t ............. perl5.28.1: Upper.xs:2030: su_uplevel_ru=
nops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fa=
iled.
t/67-uplevel-scope.t ............. Failed 20/20 subtests
t/69-uplevel-threads.t ........... # Using threads 2.22
# Using threads::shared 1.58
# Using usleep() from Time::HiRes 1.9759
t/69-uplevel-threads.t ........... 1/? perl5.28.1: Upper.xs:2030: su_upleve=
l_runops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0=
' failed.
t/69-uplevel-threads.t ........... All 3 subtests passed

t/75-uid-uplevel.t ............... perl5.28.1: Upper.xs:2030: su_uplevel_ru=
nops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fa=
iled.
t/75-uid-uplevel.t ............... Failed 106/106 subtests

t/87-stress-uid.t ................ 1/? perl5.28.1: Upper.xs:2030: su_upleve=
l_runops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0=
' failed.

> Does this mean I have to start all over without shared plib or did I
> surface a bug in perl core?

--=20
H.Merijn Brand  http://tux.nl   Perl Monger  http://amsterdam.pm.org/
using perl5.00307 .. 5.29   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/

--Sig_/J.9.=9fdiLcm5gLbsIBO4hl
Content-Type: application/pgp-signature
Content-Description: OpenPGP digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJcB5qhAAoJEAOhR6E+XcCYts0H/jD7f7luLXqEK5En87cvqBLz
3vU7xqszhx3r3D82hX1WF6OG43EFBucHEeYwHxQXtM5pYMbs0OEWsKQYCjuNrCDI
sM7Q48ub8yGBphLxspcx7OOL/fPA4kWCGnZ0S+YKJJDkn6MqUfA0yXK9KnflJGKZ
g8x9wd7BzFJgJkwks67vxG0i0GUuerVk0LpqPxYO3fD2BMhBIYPcf3N0tEREMx5J
HX43SRqap5upmioO4BDO/Ni8r/6h4KHQufZCuDc0CBIoFF9I/Gb0PGQPeKu2zYn1
MFHjheW/rJPjjiEU/LXoek2Bv7TOrhY+OvrqnesQHVu6TkreEFNF99F38hZkAjo=
=Fa89
-----END PGP SIGNATURE-----

--Sig_/J.9.=9fdiLcm5gLbsIBO4hl--
0
h
12/5/2018 9:30:02 AM 
On Wed, Dec 05, 2018 at 10:02:07AM +0100, H.Merijn Brand wrote:
> I built, tested and installed 5.28.1 on my laptop using
> 
> ./Configure -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads -Duseshrplib -des

Did you supply -Dprefix too?

> I installed it into my usual place: /pro/bin/
> the I linked /pro/bin/perl to /pro/bin/perl5.28.0 again to not disrupt
> my other tasks while installing all the modules in the 5.28.1 tree
> 
> I always forbid to use modules from the previous installs:
> 
> $ perl -V:.*inc_version.*

Is that the new or old perl?

> d_inc_version_list='undef';
> inc_version_list=' ';
> inc_version_list_init='0';
> 
> What I then not is that when using CPAN in a script like
> 
>  $ perl5.28.1 -MCPAN::Shell -we 'CPAN::Shell->install ($_) for qw( ... )'
> 
> all modules that use ExtUtil::MakeMaker install as expected, as the usual riddle is
> 
>  perl5.28.1 Makefile.PL && make && make test && make install
> 
> where make invokes the perl used to build the Makefile: perl5.28.1
> 
> and all modules that use Module::Build fail, as the use
> 
>  perl5.28.1 Build.PL && ./Build && ./Build test && ./Build install
> 
> and the shebang of Build is /pro/bin/perl which points to the previous perl

If your new perl is also being installed as /pro/bin/perl the module
build process might be confused.

> My previous perl (5.28.0) was NOT built with -Duseshrplib
> 
> I (by now) do not know if having the previous perl not having a shared
> plib has any influence on the bad behavior of Module::Build and if the
> causes havoc on my current perl, but in the new perl I now get some
> fails I never saw before:

For a Makefile.PL module build try checking the values of PERL and
FULLPERL in the generated Makefile to see if it points at
/opt/bin/perl (bad) or /opt/bin/perl5.28.1 (good).

> Does this mean I have to start all over without shared plib or did I
> surface a bug in perl core?

I'm pretty sure you just confused the build process.

Tony
0
tony
12/5/2018 9:40:47 AM 
--Sig_/lY4_/+H6TopAxECR7BBkMe2
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

On Wed, 5 Dec 2018 20:40:47 +1100, Tony Cook <tony@develop-help.com>
wrote:

> On Wed, Dec 05, 2018 at 10:02:07AM +0100, H.Merijn Brand wrote:
> > I built, tested and installed 5.28.1 on my laptop using
> >=20
> > ./Configure -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads -D=
useshrplib -des =20
>=20
> Did you supply -Dprefix too?

Of course :)

> > I installed it into my usual place: /pro/bin/
> > the I linked /pro/bin/perl to /pro/bin/perl5.28.0 again to not disrupt
> > my other tasks while installing all the modules in the 5.28.1 tree
> >=20
> > I always forbid to use modules from the previous installs:
> >=20
> > $ perl -V:.*inc_version.* =20
>=20
> Is that the new or old perl?

both

> > d_inc_version_list=3D'undef';
> > inc_version_list=3D' ';
> > inc_version_list_init=3D'0';
> >=20
> > What I then not is that when using CPAN in a script like
> >=20
> >  $ perl5.28.1 -MCPAN::Shell -we 'CPAN::Shell->install ($_) for qw( ... =
)'
> >=20
> > all modules that use ExtUtil::MakeMaker install as expected, as the usu=
al riddle is
> >=20
> >  perl5.28.1 Makefile.PL && make && make test && make install
> >=20
> > where make invokes the perl used to build the Makefile: perl5.28.1
> >=20
> > and all modules that use Module::Build fail, as the use
> >=20
> >  perl5.28.1 Build.PL && ./Build && ./Build test && ./Build install
> >=20
> > and the shebang of Build is /pro/bin/perl which points to the previous =
perl =20
>=20
> If your new perl is also being installed as /pro/bin/perl the module
> build process might be confused.

$ cd perl-5.28.1
$ make install
$ cd /pro/bin
$ rm perl
$ ln -s perl5.28.0 perl

/pro/bin/perl symlinks to my "current" perl

> > My previous perl (5.28.0) was NOT built with -Duseshrplib
> >=20
> > I (by now) do not know if having the previous perl not having a shared
> > plib has any influence on the bad behavior of Module::Build and if the
> > causes havoc on my current perl, but in the new perl I now get some
> > fails I never saw before: =20
>=20
> For a Makefile.PL module build try checking the values of PERL and
> FULLPERL in the generated Makefile to see if it points at
> /opt/bin/perl (bad) or /opt/bin/perl5.28.1 (good).

All ExtUtils::MakeMaker build succeed as the Makefile.PL is invoked
with perl5.28.1

> > Does this mean I have to start all over without shared plib or did I
> > surface a bug in perl core? =20
>=20
> I'm pretty sure you just confused the build process.

Ok=C3=A9, I'll start afresh and make this perl the default perl

> Tony


--=20
H.Merijn Brand  http://tux.nl   Perl Monger  http://amsterdam.pm.org/
using perl5.00307 .. 5.29   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/

--Sig_/lY4_/+H6TopAxECR7BBkMe2
Content-Type: application/pgp-signature
Content-Description: OpenPGP digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJcB7W1AAoJEAOhR6E+XcCY2UMH/Rc+U/BYfA343iAf1GjWhVs0
sqC2ix8xvx7XV5cD0keUYlyoYYhe1pCxBmWxF8ctMGnYvTaMWzL6DVCwWHPwYLTG
uxtlan+PfrTPQBNXGn7Kp9NsO1PlDXwgg4NAef3ogADxHNIjDBwa7uZiPGyr8/hA
/4Asi9RWg6dmEsk558CNRMD4lndEfEU7Dmvl8rCD4BUCysBUsoTUDc9/egas9q9p
Io9tNQTUcBVwp2XnFVqmfAYTcUb250XC2TFNdbRbc9EJdNZLXD/2GK9AD33o4KJb
kHlp1U2U2VE6mN8SmLphtQcu+SSqy9e9d9GbHDD6vGdJ2yAqJGXobZ7i/qZfXQY=
=GZOo
-----END PGP SIGNATURE-----

--Sig_/lY4_/+H6TopAxECR7BBkMe2--
0
h
12/5/2018 11:25:24 AM 
--Sig_/kTKmV=6TGeLtP0XjnOjOl1A
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

On Wed, 5 Dec 2018 20:40:47 +1100, Tony Cook <tony@develop-help.com>
wrote:

> > ./Configure -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads -D=
useshrplib -des =20
>=20
> Did you supply -Dprefix too?

Plain install, no shadow perls

$ perl -v
This is perl 5, version 28, subversion 1 (v5.28.1) built for x86_64-linux-t=
hread-multi-ld

$ cpan Scope::Upper
:
:
Checking if your kit is complete...
Looks good
Generating a Unix-style Makefile
Writing Makefile for Scope::Upper
Writing MYMETA.yml and MYMETA.json
  VPIT/Scope-Upper-0.31.tar.gz
  /pro/bin/perl Makefile.PL -- OK
Running make for V/VP/VPIT/Scope-Upper-0.31.tar.gz
cp lib/Scope/Upper.pm blib/lib/Scope/Upper.pm
Running Mkbootstrap for Upper ()
chmod 644 "Upper.bs"
"/pro/bin/perl" -MExtUtils::Command::MM -e 'cp_nonempty' -- Upper.bs blib/a=
rch/auto/Scope/Upper/Upper.bs 644
"/pro/bin/perl" "/pro/lib/perl5/5.28.1/ExtUtils/xsubpp"  -typemap '/pro/lib=
/perl5/5.28.1/ExtUtils/typemap'  Upper.xs > Upper.xsc
mv Upper.xsc Upper.c
cc -c   -D_REENTRANT -D_GNU_SOURCE -fPIC -DDEBUGGING -fwrapv -fno-strict-al=
iasing -pipe -fstack-protector-strong -I/pro/local/include -D_LARGEFILE_SOU=
RCE -D_FILE_OFFSET_BITS=3D64 -O2   -DVERSION=3D\"0.31\" -DXS_VERSION=3D\"0.=
31\" -fPIC "-I/pro/lib/perl5/5.28.1/x86_64-linux-thread-multi-ld/CORE"  -DS=
U_RELEASE=3D1 Upper.c
rm -f blib/arch/auto/Scope/Upper/Upper.so
cc  -shared -O2 -L/pro/local/lib -fstack-protector-strong  Upper.o  -o blib=
/arch/auto/Scope/Upper/Upper.so  \
      \

chmod 755 blib/arch/auto/Scope/Upper/Upper.so
Manifying 1 pod document
  VPIT/Scope-Upper-0.31.tar.gz
  /usr/bin/make -- OK
Running make test
"/pro/bin/perl" -MExtUtils::Command::MM -e 'cp_nonempty' -- Upper.bs blib/a=
rch/auto/Scope/Upper/Upper.bs 644
PERL_DL_NONLAZY=3D1 "/pro/bin/perl" "-MExtUtils::Command::MM" "-MTest::Harn=
ess" "-e" "undef *Test::Harness::Switches; test_harness(0, 'blib/lib', 'bli=
b/arch')" t/*.t
t/00-load.t ...................... 1/1 # Testing Scope::Upper 0.31, Perl 5.=
028001, /pro/bin/perl
t/00-load.t ...................... ok
t/01-import.t .................... ok
t/05-words.t ..................... ok
t/06-want_at.t ................... ok
t/07-context_info.t .............. ok
t/09-load-threads.t .............. # Using threads 2.22
# Using threads::shared 1.58
t/09-load-threads.t .............. ok
t/11-reap-level.t ................ ok
t/12-reap-block.t ................ ok
t/13-reap-ctl.t .................. ok
t/15-reap-multi.t ................ ok
t/16-reap-numerous.t ............. ok
t/20-localize-target.t ........... ok
t/21-localize-level.t ............ ok
t/22-localize-block.t ............ ok
t/23-localize-ctl.t .............. ok
t/24-localize-magic.t ............ ok
t/25-localize-multi.t ............ ok
t/26-localize-numerous.t ......... ok
t/30-localize_elem-target.t ...... ok
t/31-localize_elem-level.t ....... ok
t/32-localize_elem-block.t ....... ok
t/34-localize_elem-magic.t ....... ok
t/36-localize_elem-numerous.t .... ok
t/40-localize_delete-target.t .... ok
t/41-localize_delete-level.t ..... ok
t/44-localize_delete-magic.t ..... ok
t/46-localize_delete-numerous.t .. ok
t/50-unwind-target.t ............. ok
t/51-unwind-multi.t .............. ok
t/52-unwind-context.t ............ ok
t/53-unwind-misc.t ............... ok
t/54-unwind-threads.t ............ # Using threads 2.22
# Using threads::shared 1.58
# Using usleep() from Time::HiRes 1.9759
t/54-unwind-threads.t ............ ok
t/55-yield-target.t .............. ok
t/57-yield-context.t ............. ok
t/58-yield-misc.t ................ ok
t/59-yield-threads.t ............. # Using threads 2.22
# Using threads::shared 1.58
# Using usleep() from Time::HiRes 1.9759
t/59-yield-threads.t ............. ok
t/60-uplevel-target.t ............ 1/93 perl: Upper.xs:2030: su_uplevel_run=
ops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fai=
led.
t/60-uplevel-target.t ............ Failed 40/93 subtests
t/61-uplevel-args.t .............. 1/297 perl: Upper.xs:2030: su_uplevel_ru=
nops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fa=
iled.
t/61-uplevel-args.t .............. Failed 257/297 subtests
t/62-uplevel-return.t ............ 1/69 perl: Upper.xs:2030: su_uplevel_run=
ops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fai=
led.
t/62-uplevel-return.t ............ Failed 25/69 subtests
t/63-uplevel-ctl.t ............... 1/51 perl: Upper.xs:2030: su_uplevel_run=
ops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fai=
led.
t/63-uplevel-ctl.t ............... Failed 47/51 subtests
t/64-uplevel-caller.t ............ 1/32 perl: Upper.xs:2030: su_uplevel_run=
ops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fai=
led.
t/64-uplevel-caller.t ............ Failed 8/32 subtests
t/65-uplevel-multi.t ............. ok
t/66-uplevel-context.t ........... perl: Upper.xs:2030: su_uplevel_runops_h=
ook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' failed.
t/66-uplevel-context.t ........... Failed 6/6 subtests
t/67-uplevel-scope.t ............. perl: Upper.xs:2030: su_uplevel_runops_h=
ook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' failed.
t/67-uplevel-scope.t ............. Failed 20/20 subtests
t/69-uplevel-threads.t ........... # Using threads 2.22
# Using threads::shared 1.58
# Using usleep() from Time::HiRes 1.9759
t/69-uplevel-threads.t ........... 1/? perl: Upper.xs:2030: su_uplevel_runo=
ps_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fail=
ed.
t/69-uplevel-threads.t ........... All 3 subtests passed
t/70-uid-target.t ................ ok
t/74-uid-validate.t .............. ok
t/75-uid-uplevel.t ............... perl: Upper.xs:2030: su_uplevel_runops_h=
ook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' failed.
t/75-uid-uplevel.t ............... Failed 106/106 subtests
t/79-uid-threads.t ............... # Using threads 2.22
# Using threads::shared 1.58
# Using usleep() from Time::HiRes 1.9759
t/79-uid-threads.t ............... ok
t/81-stress-level.t .............. ok
t/84-stress-unwind.t ............. ok
t/85-stress-yield.t .............. ok
t/86-stress-uplevel.t ............ ok
t/87-stress-uid.t ................ 1/? perl: Upper.xs:2030: su_uplevel_runo=
ps_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fail=
ed.
t/87-stress-uid.t ................ All 385 subtests passed

Test Summary Report
-------------------
t/60-uplevel-target.t          (Wstat: 134 Tests: 53 Failed: 0)
  Non-zero wait status: 134
  Parse errors: Bad plan.  You planned 93 tests but ran 53.
t/61-uplevel-args.t            (Wstat: 134 Tests: 40 Failed: 0)
  Non-zero wait status: 134
  Parse errors: Bad plan.  You planned 297 tests but ran 40.
t/62-uplevel-return.t          (Wstat: 134 Tests: 44 Failed: 0)
  Non-zero wait status: 134
  Parse errors: Bad plan.  You planned 69 tests but ran 44.
t/63-uplevel-ctl.t             (Wstat: 134 Tests: 4 Failed: 0)
  Non-zero wait status: 134
  Parse errors: Bad plan.  You planned 51 tests but ran 4.
t/64-uplevel-caller.t          (Wstat: 134 Tests: 24 Failed: 0)
  Non-zero wait status: 134
  Parse errors: Bad plan.  You planned 32 tests but ran 24.
t/66-uplevel-context.t         (Wstat: 134 Tests: 0 Failed: 0)
  Non-zero wait status: 134
  Parse errors: Bad plan.  You planned 6 tests but ran 0.
t/67-uplevel-scope.t           (Wstat: 134 Tests: 0 Failed: 0)
  Non-zero wait status: 134
  Parse errors: Bad plan.  You planned 20 tests but ran 0.
t/69-uplevel-threads.t         (Wstat: 134 Tests: 3 Failed: 0)
  Non-zero wait status: 134
  Parse errors: No plan found in TAP output
t/75-uid-uplevel.t             (Wstat: 134 Tests: 0 Failed: 0)
  Non-zero wait status: 134
  Parse errors: Bad plan.  You planned 106 tests but ran 0.
t/87-stress-uid.t              (Wstat: 134 Tests: 385 Failed: 0)
  Non-zero wait status: 134
  Parse errors: No plan found in TAP output
Files=3D54, Tests=3D101395, 17 wallclock secs ( 8.33 usr  0.08 sys + 12.13 =
cusr  1.00 csys =3D 21.54 CPU)
Result: FAIL
Failed 10/54 test programs. 0/101395 subtests failed.
make: *** [Makefile:1067: test_dynamic] Error 255
  VPIT/Scope-Upper-0.31.tar.gz
  /usr/bin/make test -- NOT OK
//hint// to see the cpan-testers results for installing this module, try:
  reports VPIT/Scope-Upper-0.31.tar.gz

$ perl -V
Summary of my perl5 (revision 5 version 28 subversion 1) configuration:

  Platform:
    osname=3Dlinux
    osvers=3D4.19.5-1-default
    archname=3Dx86_64-linux-thread-multi-ld
    uname=3D'linux lx09 4.19.5-1-default #1 smp preempt tue nov 27 19:56:09=
 utc 2018 (6210279) x86_64 x86_64 x86_64 gnulinux '
    config_args=3D'-Duse64bitall -Duselongdouble -Dusethreads -Duseithreads=
 -Duseshrplib -des'
    hint=3Drecommended
    useposix=3Dtrue
    d_sigaction=3Ddefine
    useithreads=3Ddefine
    usemultiplicity=3Ddefine
    use64bitint=3Ddefine
    use64bitall=3Ddefine
    uselongdouble=3Ddefine
    usemymalloc=3Dn
    default_inc_excludes_dot=3Ddefine
    bincompat5005=3Dundef
  Compiler:
    cc=3D'cc'
    ccflags =3D'-D_REENTRANT -D_GNU_SOURCE -fPIC -DDEBUGGING -fwrapv -fno-s=
trict-aliasing -pipe -fstack-protector-strong -I/pro/local/include -D_LARGE=
FILE_SOURCE -D_FILE_OFFSET_BITS=3D64'
    optimize=3D'-O2'
    cppflags=3D'-D_REENTRANT -D_GNU_SOURCE -fPIC -DDEBUGGING -fwrapv -fno-s=
trict-aliasing -pipe -fstack-protector-strong -I/pro/local/include'
    ccversion=3D''
    gccversion=3D'8.2.1 20181108 [gcc-8-branch revision 265914]'
    gccosandvers=3D''
    intsize=3D4
    longsize=3D8
    ptrsize=3D8
    doublesize=3D8
    byteorder=3D12345678
    doublekind=3D3
    d_longlong=3Ddefine
    longlongsize=3D8
    d_longdbl=3Ddefine
    longdblsize=3D16
    longdblkind=3D3
    ivtype=3D'long'
    ivsize=3D8
    nvtype=3D'long double'
    nvsize=3D16
    Off_t=3D'off_t'
    lseeksize=3D8
    alignbytes=3D16
    prototype=3Ddefine
  Linker and Libraries:
    ld=3D'cc'
    ldflags =3D'-L/pro/local/lib -fstack-protector-strong'
    libpth=3D/usr/local/lib /usr/lib64/gcc/x86_64-suse-linux/8/include-fixe=
d /usr/lib64/gcc/x86_64-suse-linux/8/../../../../x86_64-suse-linux/lib /usr=
/lib /pro/local/lib /lib/../lib64 /usr/lib/../lib64 /lib /lib64 /usr/lib64 =
/usr/local/lib64
    libs=3D-lpthread -lnsl -lgdbm -ldb -ldl -lm -lcrypt -lutil -lc -lgdbm_c=
ompat
    perllibs=3D-lpthread -lnsl -ldl -lm -lcrypt -lutil -lc
    libc=3Dlibc-2.27.so
    so=3Dso
    useshrplib=3Dtrue
    libperl=3Dlibperl.so
    gnulibc_version=3D'2.27'
  Dynamic Linking:
    dlsrc=3Ddl_dlopen.xs
    dlext=3Dso
    d_dlsymun=3Dundef
    ccdlflags=3D'-Wl,-E -Wl,-rpath,/pro/lib/perl5/5.28.1/x86_64-linux-threa=
d-multi-ld/CORE'
    cccdlflags=3D'-fPIC'
    lddlflags=3D'-shared -O2 -L/pro/local/lib -fstack-protector-strong'


Characteristics of this binary (from libperl):
  Compile-time options:
    DEBUGGING
    HAS_TIMES
    MULTIPLICITY
    PERLIO_LAYERS
    PERL_COPY_ON_WRITE
    PERL_DONT_CREATE_GVSV
    PERL_IMPLICIT_CONTEXT
    PERL_MALLOC_WRAP
    PERL_OP_PARENT
    PERL_PRESERVE_IVUV
    PERL_TRACK_MEMPOOL
    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_LONG_DOUBLE
    USE_PERLIO
    USE_PERL_ATOF
    USE_REENTRANT_API
  Built under linux
  Compiled at Dec  4 2018 15:28:05
  %ENV:
    PERL6LIB=3D"inst#/pro/3gl/CPAN/rakudo/install"
  @INC:
    /pro/lib/perl5/site_perl/5.28.1/x86_64-linux-thread-multi-ld
    /pro/lib/perl5/site_perl/5.28.1
    /pro/lib/perl5/5.28.1/x86_64-linux-thread-multi-ld
    /pro/lib/perl5/5.28.1

--=20
H.Merijn Brand  http://tux.nl   Perl Monger  http://amsterdam.pm.org/
using perl5.00307 .. 5.29   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/

--Sig_/kTKmV=6TGeLtP0XjnOjOl1A
Content-Type: application/pgp-signature
Content-Description: OpenPGP digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJcB8D3AAoJEAOhR6E+XcCYZjQIAIT4S+NraLzPIxEhimx48Ltw
1wQugquKHQurDbCfdCCOQvktB5q4CvqIs4izwok8alrcGeageHifxpp2s9kEZCvc
w2gzDLdWkkkgP0lY56gg2y5ipv9abSLX9TlPjFmNa/MbunujwcOC8rnqZJH3uvIH
vn8bIqcxOFZAW7LYCwmNbAro+RG3tAWQrNHQ8u4kqf7PGtdSzvIEEtwvHDDuIskl
poHWaoTOFiiNU6WlrOaXlNQN4jp16SGBSAqa6Q1r1yKklDfwkvYpnlZyUX2A8uA5
i9ezk4bz56qjjojl2zJuVyn08VvVN8YJU3TCPvm0iYA+jG9KPj1F8TFUK8Ajvn8=
=abZ8
-----END PGP SIGNATURE-----

--Sig_/kTKmV=6TGeLtP0XjnOjOl1A--
0
h
12/5/2018 12:13:39 PM 
--000000000000f12b91057c4551f6
Content-Type: text/plain; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

Tux (or Merijn?),

This sounds a lot to me like a bug in Module::Build. If you look at the
very top of the "Build" file produced when you run Build.PL, what does it
say?

My initial though was that it is taking the name of the perl interpreter
from a Config entry rather than $^X (though why it would do that is a
mystery to me). When I looked into the source code for Module::Build, I was
left scratching my head. Module::Build::Base constructs the shebang line (
https://metacpan.org/source/Module::Build::Base#L1824) by appending to '#!'
whatever is returned by the object's _startperl method. As a grep shows (
https://grep.metacpan.org/search?qd=3DModule-Build&source=3Dmetacpan&q=3Dst=
artperl),
the default executable comes from config->('startperl'), but I cannot find
where that is defined. Furthermore, Unix platforms override _startperl to
return whatever is returned by the "perl" method. However, I cannot find
that method defined anywhere.

So, it seems to me like a bug in M::B, but it's really a mystery.

David

On Wed, Dec 5, 2018 at 6:26 AM H.Merijn Brand <h.m.brand@xs4all.nl> wrote:

> On Wed, 5 Dec 2018 20:40:47 +1100, Tony Cook <tony@develop-help.com>
> wrote:
>
> > On Wed, Dec 05, 2018 at 10:02:07AM +0100, H.Merijn Brand wrote:
> > > I built, tested and installed 5.28.1 on my laptop using
> > >
> > > ./Configure -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads
> -Duseshrplib -des
> >
> > Did you supply -Dprefix too?
>
> Of course :)
>
> > > I installed it into my usual place: /pro/bin/
> > > the I linked /pro/bin/perl to /pro/bin/perl5.28.0 again to not disrup=
t
> > > my other tasks while installing all the modules in the 5.28.1 tree
> > >
> > > I always forbid to use modules from the previous installs:
> > >
> > > $ perl -V:.*inc_version.*
> >
> > Is that the new or old perl?
>
> both
>
> > > d_inc_version_list=3D'undef';
> > > inc_version_list=3D' ';
> > > inc_version_list_init=3D'0';
> > >
> > > What I then not is that when using CPAN in a script like
> > >
> > >  $ perl5.28.1 -MCPAN::Shell -we 'CPAN::Shell->install ($_) for qw( ..=
..
> )'
> > >
> > > all modules that use ExtUtil::MakeMaker install as expected, as the
> usual riddle is
> > >
> > >  perl5.28.1 Makefile.PL && make && make test && make install
> > >
> > > where make invokes the perl used to build the Makefile: perl5.28.1
> > >
> > > and all modules that use Module::Build fail, as the use
> > >
> > >  perl5.28.1 Build.PL && ./Build && ./Build test && ./Build install
> > >
> > > and the shebang of Build is /pro/bin/perl which points to the previou=
s
> perl
> >
> > If your new perl is also being installed as /pro/bin/perl the module
> > build process might be confused.
>
> $ cd perl-5.28.1
> $ make install
> $ cd /pro/bin
> $ rm perl
> $ ln -s perl5.28.0 perl
>
> /pro/bin/perl symlinks to my "current" perl
>
> > > My previous perl (5.28.0) was NOT built with -Duseshrplib
> > >
> > > I (by now) do not know if having the previous perl not having a share=
d
> > > plib has any influence on the bad behavior of Module::Build and if th=
e
> > > causes havoc on my current perl, but in the new perl I now get some
> > > fails I never saw before:
> >
> > For a Makefile.PL module build try checking the values of PERL and
> > FULLPERL in the generated Makefile to see if it points at
> > /opt/bin/perl (bad) or /opt/bin/perl5.28.1 (good).
>
> All ExtUtils::MakeMaker build succeed as the Makefile.PL is invoked
> with perl5.28.1
>
> > > Does this mean I have to start all over without shared plib or did I
> > > surface a bug in perl core?
> >
> > I'm pretty sure you just confused the build process.
>
> Ok=C3=A9, I'll start afresh and make this perl the default perl
>
> > Tony
>
>
> --
> H.Merijn Brand  http://tux.nl   Perl Monger  http://amsterdam.pm.org/
> using perl5.00307 .. 5.29   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/
>


--=20
 "Debugging is twice as hard as writing the code in the first place.
  Therefore, if you write the code as cleverly as possible, you are,
  by definition, not smart enough to debug it." -- Brian Kernighan

--000000000000f12b91057c4551f6
Content-Type: text/html; charset="UTF-8"
Content-Transfer-Encoding: quoted-printable

<div dir=3D"ltr"><div dir=3D"ltr"><div dir=3D"ltr"><div dir=3D"ltr"><div>Tu=
x (or Merijn?),</div><div><br></div><div>This sounds a lot to me like a bug=
 in Module::Build. If you look at the very top of the &quot;Build&quot; fil=
e produced when you run Build.PL, what does it say?</div><div><br></div><di=
v>My initial though was that  it is taking the name of the perl interpreter=
 from a Config entry rather than $^X (though why it would do that is a myst=
ery to me). When I looked into the source code for Module::Build, I was lef=
t scratching my head. Module::Build::Base constructs the shebang line (<a h=
ref=3D"https://metacpan.org/source/Module::Build::Base#L1824">https://metac=
pan.org/source/Module::Build::Base#L1824</a>) by appending to &#39;#!&#39; =
whatever is returned by the object&#39;s _startperl method. As a grep shows=
 (<a href=3D"https://grep.metacpan.org/search?qd=3DModule-Build&amp;source=
=3Dmetacpan&amp;q=3Dstartperl">https://grep.metacpan.org/search?qd=3DModule=
-Build&amp;source=3Dmetacpan&amp;q=3Dstartperl</a>), the default executable=
 comes from config-&gt;(&#39;startperl&#39;), but I cannot find where that =
is defined. Furthermore,  Unix platforms override _startperl to return what=
ever is returned by the &quot;perl&quot; method. However, I cannot find tha=
t method defined anywhere.<br></div><div><br></div>So, it seems to me like =
a bug in M::B, but it&#39;s really a mystery.</div><div dir=3D"ltr"><br></d=
iv><div>David<br></div></div></div></div><br><div class=3D"gmail_quote"><di=
v dir=3D"ltr">On Wed, Dec 5, 2018 at 6:26 AM H.Merijn Brand &lt;<a href=3D"=
mailto:h.m.brand@xs4all.nl">h.m.brand@xs4all.nl</a>&gt; wrote:<br></div><bl=
ockquote class=3D"gmail_quote" style=3D"margin:0 0 0 .8ex;border-left:1px #=
ccc solid;padding-left:1ex">On Wed, 5 Dec 2018 20:40:47 +1100, Tony Cook &l=
t;<a href=3D"mailto:tony@develop-help.com" target=3D"_blank">tony@develop-h=
elp.com</a>&gt;<br>
wrote:<br>
<br>
&gt; On Wed, Dec 05, 2018 at 10:02:07AM +0100, H.Merijn Brand wrote:<br>
&gt; &gt; I built, tested and installed 5.28.1 on my laptop using<br>
&gt; &gt; <br>
&gt; &gt; ./Configure -Duse64bitall -Duselongdouble -Dusethreads -Duseithre=
ads -Duseshrplib -des=C2=A0 <br>
&gt; <br>
&gt; Did you supply -Dprefix too?<br>
<br>
Of course :)<br>
<br>
&gt; &gt; I installed it into my usual place: /pro/bin/<br>
&gt; &gt; the I linked /pro/bin/perl to /pro/bin/perl5.28.0 again to not di=
srupt<br>
&gt; &gt; my other tasks while installing all the modules in the 5.28.1 tre=
e<br>
&gt; &gt; <br>
&gt; &gt; I always forbid to use modules from the previous installs:<br>
&gt; &gt; <br>
&gt; &gt; $ perl -V:.*inc_version.*=C2=A0 <br>
&gt; <br>
&gt; Is that the new or old perl?<br>
<br>
both<br>
<br>
&gt; &gt; d_inc_version_list=3D&#39;undef&#39;;<br>
&gt; &gt; inc_version_list=3D&#39; &#39;;<br>
&gt; &gt; inc_version_list_init=3D&#39;0&#39;;<br>
&gt; &gt; <br>
&gt; &gt; What I then not is that when using CPAN in a script like<br>
&gt; &gt; <br>
&gt; &gt;=C2=A0 $ perl5.28.1 -MCPAN::Shell -we &#39;CPAN::Shell-&gt;install=
 ($_) for qw( ... )&#39;<br>
&gt; &gt; <br>
&gt; &gt; all modules that use ExtUtil::MakeMaker install as expected, as t=
he usual riddle is<br>
&gt; &gt; <br>
&gt; &gt;=C2=A0 perl5.28.1 Makefile.PL &amp;&amp; make &amp;&amp; make test=
 &amp;&amp; make install<br>
&gt; &gt; <br>
&gt; &gt; where make invokes the perl used to build the Makefile: perl5.28.=
1<br>
&gt; &gt; <br>
&gt; &gt; and all modules that use Module::Build fail, as the use<br>
&gt; &gt; <br>
&gt; &gt;=C2=A0 perl5.28.1 Build.PL &amp;&amp; ./Build &amp;&amp; ./Build t=
est &amp;&amp; ./Build install<br>
&gt; &gt; <br>
&gt; &gt; and the shebang of Build is /pro/bin/perl which points to the pre=
vious perl=C2=A0 <br>
&gt; <br>
&gt; If your new perl is also being installed as /pro/bin/perl the module<b=
r>
&gt; build process might be confused.<br>
<br>
$ cd perl-5.28.1<br>
$ make install<br>
$ cd /pro/bin<br>
$ rm perl<br>
$ ln -s perl5.28.0 perl<br>
<br>
/pro/bin/perl symlinks to my &quot;current&quot; perl<br>
<br>
&gt; &gt; My previous perl (5.28.0) was NOT built with -Duseshrplib<br>
&gt; &gt; <br>
&gt; &gt; I (by now) do not know if having the previous perl not having a s=
hared<br>
&gt; &gt; plib has any influence on the bad behavior of Module::Build and i=
f the<br>
&gt; &gt; causes havoc on my current perl, but in the new perl I now get so=
me<br>
&gt; &gt; fails I never saw before:=C2=A0 <br>
&gt; <br>
&gt; For a Makefile.PL module build try checking the values of PERL and<br>
&gt; FULLPERL in the generated Makefile to see if it points at<br>
&gt; /opt/bin/perl (bad) or /opt/bin/perl5.28.1 (good).<br>
<br>
All ExtUtils::MakeMaker build succeed as the Makefile.PL is invoked<br>
with perl5.28.1<br>
<br>
&gt; &gt; Does this mean I have to start all over without shared plib or di=
d I<br>
&gt; &gt; surface a bug in perl core?=C2=A0 <br>
&gt; <br>
&gt; I&#39;m pretty sure you just confused the build process.<br>
<br>
Ok=C3=A9, I&#39;ll start afresh and make this perl the default perl<br>
<br>
&gt; Tony<br>
<br>
<br>
-- <br>
H.Merijn Brand=C2=A0 <a href=3D"http://tux.nl" rel=3D"noreferrer" target=3D=
"_blank">http://tux.nl</a>=C2=A0 =C2=A0Perl Monger=C2=A0 <a href=3D"http://=
amsterdam.pm.org/" rel=3D"noreferrer" target=3D"_blank">http://amsterdam.pm=
..org/</a><br>
using perl5.00307 .. 5.29=C2=A0 =C2=A0porting perl5 on HP-UX, AIX, and open=
SUSE<br>
<a href=3D"http://mirrors.develooper.com/hpux/" rel=3D"noreferrer" target=
=3D"_blank">http://mirrors.develooper.com/hpux/</a>=C2=A0 =C2=A0 =C2=A0 =C2=
=A0 <a href=3D"http://www.test-smoke.org/" rel=3D"noreferrer" target=3D"_bl=
ank">http://www.test-smoke.org/</a><br>
<a href=3D"http://qa.perl.org" rel=3D"noreferrer" target=3D"_blank">http://=
qa.perl.org</a>=C2=A0 =C2=A0<a href=3D"http://www.goldmark.org/jeff/stupid-=
disclaimers/" rel=3D"noreferrer" target=3D"_blank">http://www.goldmark.org/=
jeff/stupid-disclaimers/</a><br>
</blockquote></div><br clear=3D"all"><br>-- <br><div dir=3D"ltr" class=3D"g=
mail_signature" data-smartmail=3D"gmail_signature">=C2=A0&quot;Debugging is=
 twice as hard as writing the code in the first place.<br>
 =C2=A0 Therefore, if you write the code as cleverly as possible, you are,<=
br>
 =C2=A0 by definition, not smart enough to debug it.&quot; -- Brian Kernigh=
an<br></div>

--000000000000f12b91057c4551f6--
0
dcmertens
12/5/2018 12:13:58 PM 
--Sig_/yJ407heC1wVf=iCc/sWzGKV
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

On Wed, 5 Dec 2018 10:02:07 +0100, "H.Merijn Brand"
<h.m.brand@xs4all.nl> wrote:

> indirect-0.38-0 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
> t/40-threads.t ............. # Using threads 2.22
> # Using threads::shared 1.58
> perl5.28.1: indirect.xs:264: indirect_map_store: Assertion `((void*)(oi->=
buf)) !=3D 0' failed.
> t/40-threads.t ............. No subtests run
>=20
> t/42-threads-global.t ...... # Using threads 2.22
> # Using threads::shared 1.58
> perl5.28.1: indirect.xs:264: indirect_map_store: Assertion `((void*)(oi->=
buf)) !=3D 0' failed.
> t/42-threads-global.t ...... No subtests run

If I do not install any module (start from a plain 5.28.1),
test and installation for indirect passes

When I try to install indirect after a whole truckload of other
modules, these tests fail

indirect-0.38-0 542 > perl -Mblib -d:TraceLoad t/40-threads.t
[  0.0000] blib
[  0.0002] Cwd
[  0.0039] File::Spec
[  0.0041] File::Spec::Unix
[  0.0047] constant
[  0.0049] warnings::register
[  0.0084] lib
[  0.0086] Config
[  0.0361] POSIX
[  0.0365] Fcntl
[  0.0414] Tie::Hash
[  0.0416] Carp
[  0.0430] overloading
[  0.0482] threads
[  0.0491] overload
[  0.0508] Test::More
[  0.0510] Test::Builder::Module
[  0.0511] Test::Builder
[  0.0514] Scalar::Util
[  0.0518] List::Util
[  0.0527] Test2::Util
[  0.0530] PerlIO
[  0.0545] Config_heavy.pl
[  0.0564] Config_git.pl
[  0.0572] Test2::API
[  0.0575] Test2::API::Instance
[  0.0578] Test2::EventFacet::Trace
[  0.0580] Test2::EventFacet
[  0.0581] Test2::Util::HashBase
[  0.0585] mro
[  0.0608] Test2::API::Stack
[  0.0610] Test2::Hub
[  0.0614] Test2::Util::ExternalMeta
[  0.0701] Test2::Util::Trace
[  0.0703] Test2::Hub::Subtest
[  0.0711] Test2::Hub::Interceptor
[  0.0713] Test2::Hub::Interceptor::Terminator
[  0.0718] Test2::Event::Ok
[  0.0721] Test2::Event
[  0.0726] Test2::EventFacet::About
[  0.0730] Test2::EventFacet::Amnesty
[  0.0734] Test2::EventFacet::Assert
[  0.0737] Test2::EventFacet::Control
[  0.0741] Test2::EventFacet::Error
[  0.0744] Test2::EventFacet::Info
[  0.0747] Test2::EventFacet::Meta
[  0.0750] vars
[  0.0758] Test2::EventFacet::Parent
[  0.0762] Test2::EventFacet::Plan
[  0.0766] Test2::EventFacet::Hub
[  0.0790] Test2::Event::Diag
[  0.0794] Test2::Event::Note
[  0.0799] Test2::Event::Plan
[  0.0808] Test2::Event::Bail
[  0.0812] Test2::Event::Exception
[  0.0817] Test2::Event::Waiting
[  0.0821] Test2::Event::Skip
[  0.0826] Test2::Event::Subtest
[  0.0870] Test2::API::Context
[  0.0900] Test2::Event::Pass
[  0.0907] Test2::Event::Fail
[  0.0913] Test2::Event::V2
[  0.0917] Test2::Util::Facets2Legacy
[  0.0919] base
[  0.0949] Test2::IPC
[  0.0958] Test2::IPC::Driver::Files
[  0.0960] Test2::IPC::Driver
[  0.0969] File::Temp
[  0.0972] File::Path
[  0.0975] File::Basename
[  0.1026] IO::Seekable
[  0.1029] IO::Handle
[  0.1032] Symbol
[  0.1041] SelectSaver
[  0.1044] IO
[  0.1074] Errno
[  0.1087] parent
[  0.1148] Carp::Heavy
[  0.1153] Storable
[  0.1157] Log::Agent
[  0.1181] Storable::Limit
[  0.1211] IPC::SysV
[  0.1217] DynaLoader
[  0.1238] Test::Builder::Formatter
[  0.1240] Test2::Formatter::TAP
[  0.1243] Test2::Formatter
[  0.1275] Test::Builder::TodoDiag
# Using threads 2.22
[  0.1426] threads::shared
# Using threads::shared 1.58
[  0.1443] Test::Leaner
[  0.1489] indirect
perl: indirect.xs:264: indirect_map_store: Assertion `((void*)(oi->buf)) !=
=3D 0' failed.
Abort (core dumped)

The fix is this:

--- indirect.xs.org     2018-12-09 16:18:19.342479799 +0100
+++ indirect.xs 2018-12-09 16:18:21.362533538 +0100
@@ -261,7 +261,7 @@ static void indirect_map_store(pTHX_ con
   XSH_LOCAL_REALLOC(oi->buf, oi->size, len, char);
   oi->size =3D len;
  }
- Copy(s, oi->buf, len, char);
+ if (oi->buf) Copy(s, oi->buf, len, char);

  oi->len  =3D len;
  oi->pos  =3D pos;

And all tests pass

This is *way* beyond my competence area, so please speak up if you
understand


FWIW I used this to check:
---8<--- Devel/TraceLoad.pm
package Devel::TraceLoad;

use 5.14.1;
use warnings;

use Time::HiRes qw( gettimeofday tv_interval );

# Based on http://www.perladvent.org/2018/2018-12-06.html

# BEGIN blocks and @INC dealing honestly stolen from Devel::TraceUse
# created and maintained by BooK
# Devel::TraceUse prints at the end, which won't happen on a core dump

# detect being loaded via -d:TraceLoad and disable the debugger features we
# don't need. better names for evals (0x100) and anon subs (0x200).
BEGIN {
    !defined &DB::DB && $^P & 0x02 and $^P =3D 0x100 | 0x200;
    unshift @INC, \&trace_load unless grep { "$_" eq \&trace_use."" } @INC;
    };

my $time =3D [ gettimeofday ];
my %seen;

sub trace_load {
    my ($code, $filename) =3D @_;

    # ensure our hook remains first in @INC
    $INC[0] eq $code or @INC =3D ($code, grep { $_ ne $code } @INC);

    my $module =3D $filename =3D~ s{/}{::}gr =3D~ s{\.pm$}{}r;
    $seen{$module}++ and return;

    printf STDERR "[%8.4f] %s\n", tv_interval ($time), $module;
    return;
    } # trace_load

1;
-->8---

/me now continues to find the other fails

--=20
H.Merijn Brand  http://tux.nl   Perl Monger  http://amsterdam.pm.org/
using perl5.00307 .. 5.29   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/

--Sig_/yJ407heC1wVf=iCc/sWzGKV
Content-Type: application/pgp-signature
Content-Description: OpenPGP digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJcDTL8AAoJEAOhR6E+XcCYBTYH/0GWgmtLaU2qtAboiXq7D3zM
sFMfS8KRLF8ddkaPI4qhK/WJW993azf5mSOGvdoPRA2Wd/PfqfPduL9btH5mWVX4
lDz9B+NM2Su/wv4Q3QvJ5koRUZmoErXPR04C65e+UL9Lr8xJ4gjWzaCHzEpe1QLp
zrWNFlvxNqOnwENpTaIjS53saAB8XKAFfFgM9pBQa7pdl5lfwYjrnURS67cnjOhV
2cFSkcD6OG4Y3ojMuK4hJK1+vsGuo1jixEpNX003ocsi4AGuJx0GwnO1d5TVq2ZF
ApZiG79hDgVU7r3i5JyMbo6bBYIRnyrGBrzFo1R5pGS7Xkz27s70XQEecvbEQhE=
=M515
-----END PGP SIGNATURE-----

--Sig_/yJ407heC1wVf=iCc/sWzGKV--
0
h
12/9/2018 3:21:23 PM 
--Sig_/H4Gndhbu6so4Xq_g2p/oRgZ
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

On Wed, 5 Dec 2018 10:02:07 +0100, "H.Merijn Brand"
<h.m.brand@xs4all.nl> wrote:

> Math-Cephes-0.5305-0 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
> t/elliptics.t ........ 1/10 panic: XSUB Math::Cephesc::ellpj (Cephes_wrap=
..c) failed to extend arg stack: base=3Df5e100, sp=3Df5e128, hwm=3Df5e118
> t/elliptics.t ........ Dubious, test returned 255 (wstat 65280, 0xff00)
> Failed 5/10 subtests
>=20
> t/misc.t ............. 1/33 panic: XSUB Math::Cephesc::fresnl (Cephes_wra=
p.c) failed to extend arg stack: base=3D1bd0100, sp=3D1bd0118, hwm=3D1bd0110
> t/misc.t ............. Dubious, test returned 255 (wstat 65280, 0xff00)
> Failed 29/33 subtests

This is bad XS code in the module itself. I attached a patch to the RT
ticket: https://rt.cpan.org/Ticket/Display.html?id=3D122912

--=20
H.Merijn Brand  http://tux.nl   Perl Monger  http://amsterdam.pm.org/
using perl5.00307 .. 5.29   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/

--Sig_/H4Gndhbu6so4Xq_g2p/oRgZ
Content-Type: application/pgp-signature
Content-Description: OpenPGP digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJcDWRbAAoJEAOhR6E+XcCYpyIIAIHlOr2c6E55ygo8u000z2C/
oI7V90ggmg3VS88RSxyZ/4w5QOfRWinlqQIUAXGg9tYl+22t/KLpkS8G9Fd4uOa2
1wOVsP7JZFqich7TkEQazP+5ODqZz3VnNpNxKC0IZIulrlwi5ekPAinS2NgNwHbb
mwDTUmdIn/petb/QfynASzf5YcZc4SAjWvOproP2dqnlXxquvzErL1QDDIHsDtgI
KctODSwFHerb2s8YEQwOEUuRQ1AhO61DlJqIgDOjxk4RBbVmGZoSI7fwdUPvepqJ
rCoRpvhVRBRcVrssQpsIUaOrQbay11tgzLtL8FuuGnXjPL/vyvCFVcI6iEEua80=
=3ZI9
-----END PGP SIGNATURE-----

--Sig_/H4Gndhbu6so4Xq_g2p/oRgZ--
0
h
12/9/2018 6:52:01 PM 
--Sig_/ro_iasDsYk7b=ATo3BFIe67
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

On Wed, 5 Dec 2018 10:02:07 +0100, "H.Merijn Brand" <h.m.brand@xs4all.nl> w=
rote:

> Text-Aspell-0.09-0 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
> t/05-core.t .. 1/19 panic: XSUB Text::Aspell::list_dictionaries (Aspell.c=
) failed to extend arg stack: base=3D694100, sp=3D6942a8, hwm=3D694110
> # Looks like your test exited with 2 just after 15.
> t/05-core.t .. Dubious, test returned 2 (wstat 512, 0x200)
> Failed 4/19 subtests

This is an error in the module's XS code. TonyC already suggested a
patch in an RT ticket, which proves to work correctly.

I upped the severity to critical.

--=20
H.Merijn Brand  http://tux.nl   Perl Monger  http://amsterdam.pm.org/
using perl5.00307 .. 5.29   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/

--Sig_/ro_iasDsYk7b=ATo3BFIe67
Content-Type: application/pgp-signature
Content-Description: OpenPGP digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJcDh7+AAoJEAOhR6E+XcCYu54H/2FumjzwjStMse0XMrLB2QFG
YJYehH+luLLMtMa3I/v1EOurH1BhfRg+MqTR1loejLtCo8GwECT8bOklZ5MV+itS
NdYQ4D1SWmKKyFhgpur1r5pOst/jZKOxVutSSLYObAEn2+/ArcPVEc2Qm1VpIbg0
OMJDqZcAT5Nv6c6G7DiOZhg0sRDgY24aVYSjfZiQgoeLhhdoctrEa1vWqEzuGU+9
vclA9Zery/yWUcRY5zdmdkGX7pqBB/ypKYo1NdCyugCzxUVMi6Lagt4tJ1TbyjTf
GiJrsgFmz036LoPtpesmJq5a5zcfhv30fnmUwGgVhGFvYw4M9s5yFkI0mFzqUyM=
=/2Ln
-----END PGP SIGNATURE-----

--Sig_/ro_iasDsYk7b=ATo3BFIe67--
0
h
12/10/2018 8:08:22 AM 
--Sig_/gKBHa7bq53_ibV3OnLCub6+
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: quoted-printable

Interesting case. Important question at the end

On Wed, 5 Dec 2018 10:02:07 +0100, "H.Merijn Brand" <h.m.brand@xs4all.nl> w=
rote:

> Image-Scale-0.14-0 =3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=
=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D=3D
>=20
> t/stringify.t ...... perl5.28.1: src/image.c:48: image_init: Assertion `P=
L_valid_types_PVX[SvTYPE(_svpvx) & SVt_MASK]' failed.
> t/stringify.t ...... Failed 3/3 subtests

This is Path::Tiny related

The path followed is ...

    my $im =3D Image::Scale->new( Path::Tiny::path( _f("jpg/rgb.jpg") ) );

which ends up here

  if (my_hv_exists(self, "file")) {
    // Input from file
    SV *path =3D *(my_hv_fetch(self, "file"));
    file =3D SvPVX(path);
    im->fh =3D IoIFP(sv_2io(*(my_hv_fetch(self, "_fh"))));
    im->path =3D newSVsv(path);
  }

an sv_dump of path shows

SV =3D IV(0x23c4bd0) at 0x23c4be0
  REFCNT =3D 1
  FLAGS =3D (ROK)
  RV =3D 0x2d475f0
    SV =3D PVAV(0x2cf2b40) at 0x2d475f0
      REFCNT =3D 3
      FLAGS =3D (OBJECT)
      STASH =3D 0x23e9b70 "Path::Tiny"
      ARRAY =3D 0x2d61b00
      FILL =3D 1
      MAX =3D 1
      FLAGS =3D (REAL)
        Elt No. 0
        SV =3D PV(0x23b23a0) at 0x2c98f88
          REFCNT =3D 1
          FLAGS =3D (POK,IsCOW,pPOK)
          PV =3D 0x2d8d140 "/data/home/merijn/.cpan/build/Image-Scale-0.14/=
t/images/jpg/rgb.jpg"\0
          CUR =3D 67
          LEN =3D 69
          COW_REFCNT =3D 1
        Elt No. 1
        SV =3D PV(0x2cc7800) at 0x2d475a8
          REFCNT =3D 1
          FLAGS =3D (POK,IsCOW,pPOK)
          PV =3D 0x2d8d140 "/data/home/merijn/.cpan/build/Image-Scale-0.14/=
t/images/jpg/rgb.jpg"\0
          CUR =3D 67
          LEN =3D 69
          COW_REFCNT =3D 1

That shows that string-overloading does not happen from XS
If I change the .t invocation to

    my $im =3D Image::Scale->new( Path::Tiny::path( _f("jpg/rgb.jpg") )->st=
ringify );
    $im->save_jpeg( Path::Tiny::path($outjpg)->stringify );
    $im->save_png( Path::Tiny::path($outpng)->stringify );

in all locations where this happens, all tests pass.
I have updated the ticket, but is this a incompatibility change
somewhere?

--=20
H.Merijn Brand  http://tux.nl   Perl Monger  http://amsterdam.pm.org/
using perl5.00307 .. 5.29   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/

--Sig_/gKBHa7bq53_ibV3OnLCub6+
Content-Type: application/pgp-signature
Content-Description: OpenPGP digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJcDjXOAAoJEAOhR6E+XcCYzcAH/R2gXbPTexEd+LaS27pbKlWc
tBWdxdJaP95xd2Q6cB9DTxC73zVEPnnQWywgrAj2lbgw1e38Ui0t7PiYWgVEKleU
wt5v1aTXK4HHU3XAn9I+r6YAKFXn8ZbYE6j43DEDcyjrzUGLIlPVh8KLutGAa84V
z6CcTokE+1AimzSPmbpoOqOcxqdmD1S6cefBv7rhZD04pNunBDS6SENNb195PXjQ
R9fM4eSwBZl9uAHNyfJQHwS7D5Y/4k4EOsPk2WZI5Keg7Yyr/OqBgPqUhPJstNFi
1nSun0nRaOYiRwUhzwvPdEzmp7HAiAI8r4PHK/9KL9+5EH84dugvWZn2YrKL22k=
=qVBY
-----END PGP SIGNATURE-----

--Sig_/gKBHa7bq53_ibV3OnLCub6+--
0
h
12/10/2018 9:45:40 AM 
--Sig_/xfoNcdJzmT/XJPvPgt_=Rs.
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: quoted-printable

On Wed, 5 Dec 2018 13:13:39 +0100, "H.Merijn Brand"
<h.m.brand@xs4all.nl> wrote:

> On Wed, 5 Dec 2018 20:40:47 +1100, Tony Cook <tony@develop-help.com>
> wrote:
>=20
> > > ./Configure -Duse64bitall -Duselongdouble -Dusethreads -Duseithreads =
-Duseshrplib -des   =20
> >=20
> > Did you supply -Dprefix too? =20
>=20
> Plain install, no shadow perls

av_extend(av, X); will cause real PAIN id X < 0

--8<---
--- Scope-Upper-0.31-0/Upper.xs 2018-08-26 21:09:25.000000000 +0200
+++ Scope-Upper-0.31-1/Upper.xs 2018-12-10 17:32:25.115448191 +0100
@@ -2015,17 +2015,18 @@ static int su_uplevel_runops_hook_enters
  OP *op =3D PL_op;
  dXSH_CXT;
  su_uplevel_ud *sud =3D XSH_CXT.uplevel_storage.top;
+ int avmax;

  /* Create a new array containing a copy of the original sub's call args,
   * then stick it in PL_curpad[0] of the current running sub so that
   * thay will be seen by caller().
   */
  assert(sud);
- if (sud->argarray) {
+ if (sud->argarray && ((avmax =3D AvMAX (sud->argarray)) >=3D 0)) {
   AV *av =3D newAV();
   AvREAL_off(av);
   AvREIFY_on(av);
-  av_extend(av, AvMAX(sud->argarray));
+  av_extend(av, avmax);
   AvFILLp(av) =3D AvFILLp(sud->argarray);
   Copy(AvARRAY(sud->argarray), AvARRAY(av), AvFILLp(av) + 1, SV *);

-->8---

=E2=86=92

All tests successful.
Files=3D54, Tests=3D102379, 16 wallclock secs ( 7.29 usr  0.08 sys + 11.76 =
cusr  0.79 csys =3D 19.92 CPU)
Result: PASS

> $ perl -v
> This is perl 5, version 28, subversion 1 (v5.28.1) built for x86_64-linux=
-thread-multi-ld
>=20
> $ cpan Scope::Upper
> :
> :
> Checking if your kit is complete...
> Looks good
> Generating a Unix-style Makefile
> Writing Makefile for Scope::Upper
> Writing MYMETA.yml and MYMETA.json
>   VPIT/Scope-Upper-0.31.tar.gz
>   /pro/bin/perl Makefile.PL -- OK
> Running make for V/VP/VPIT/Scope-Upper-0.31.tar.gz
> cp lib/Scope/Upper.pm blib/lib/Scope/Upper.pm
> Running Mkbootstrap for Upper ()
> chmod 644 "Upper.bs"
> "/pro/bin/perl" -MExtUtils::Command::MM -e 'cp_nonempty' -- Upper.bs blib=
/arch/auto/Scope/Upper/Upper.bs 644
> "/pro/bin/perl" "/pro/lib/perl5/5.28.1/ExtUtils/xsubpp"  -typemap '/pro/l=
ib/perl5/5.28.1/ExtUtils/typemap'  Upper.xs > Upper.xsc
> mv Upper.xsc Upper.c
> cc -c   -D_REENTRANT -D_GNU_SOURCE -fPIC -DDEBUGGING -fwrapv -fno-strict-=
aliasing -pipe -fstack-protector-strong -I/pro/local/include -D_LARGEFILE_S=
OURCE -D_FILE_OFFSET_BITS=3D64 -O2   -DVERSION=3D\"0.31\" -DXS_VERSION=3D\"=
0.31\" -fPIC "-I/pro/lib/perl5/5.28.1/x86_64-linux-thread-multi-ld/CORE"  -=
DSU_RELEASE=3D1 Upper.c
> rm -f blib/arch/auto/Scope/Upper/Upper.so
> cc  -shared -O2 -L/pro/local/lib -fstack-protector-strong  Upper.o  -o bl=
ib/arch/auto/Scope/Upper/Upper.so  \
>       \
>=20
> chmod 755 blib/arch/auto/Scope/Upper/Upper.so
> Manifying 1 pod document
>   VPIT/Scope-Upper-0.31.tar.gz
>   /usr/bin/make -- OK
> Running make test
> "/pro/bin/perl" -MExtUtils::Command::MM -e 'cp_nonempty' -- Upper.bs blib=
/arch/auto/Scope/Upper/Upper.bs 644
> PERL_DL_NONLAZY=3D1 "/pro/bin/perl" "-MExtUtils::Command::MM" "-MTest::Ha=
rness" "-e" "undef *Test::Harness::Switches; test_harness(0, 'blib/lib', 'b=
lib/arch')" t/*.t
> t/00-load.t ...................... 1/1 # Testing Scope::Upper 0.31, Perl =
5.028001, /pro/bin/perl
> t/00-load.t ...................... ok
> t/01-import.t .................... ok
> t/05-words.t ..................... ok
> t/06-want_at.t ................... ok
> t/07-context_info.t .............. ok
> t/09-load-threads.t .............. # Using threads 2.22
> # Using threads::shared 1.58
> t/09-load-threads.t .............. ok
> t/11-reap-level.t ................ ok
> t/12-reap-block.t ................ ok
> t/13-reap-ctl.t .................. ok
> t/15-reap-multi.t ................ ok
> t/16-reap-numerous.t ............. ok
> t/20-localize-target.t ........... ok
> t/21-localize-level.t ............ ok
> t/22-localize-block.t ............ ok
> t/23-localize-ctl.t .............. ok
> t/24-localize-magic.t ............ ok
> t/25-localize-multi.t ............ ok
> t/26-localize-numerous.t ......... ok
> t/30-localize_elem-target.t ...... ok
> t/31-localize_elem-level.t ....... ok
> t/32-localize_elem-block.t ....... ok
> t/34-localize_elem-magic.t ....... ok
> t/36-localize_elem-numerous.t .... ok
> t/40-localize_delete-target.t .... ok
> t/41-localize_delete-level.t ..... ok
> t/44-localize_delete-magic.t ..... ok
> t/46-localize_delete-numerous.t .. ok
> t/50-unwind-target.t ............. ok
> t/51-unwind-multi.t .............. ok
> t/52-unwind-context.t ............ ok
> t/53-unwind-misc.t ............... ok
> t/54-unwind-threads.t ............ # Using threads 2.22
> # Using threads::shared 1.58
> # Using usleep() from Time::HiRes 1.9759
> t/54-unwind-threads.t ............ ok
> t/55-yield-target.t .............. ok
> t/57-yield-context.t ............. ok
> t/58-yield-misc.t ................ ok
> t/59-yield-threads.t ............. # Using threads 2.22
> # Using threads::shared 1.58
> # Using usleep() from Time::HiRes 1.9759
> t/59-yield-threads.t ............. ok
> t/60-uplevel-target.t ............ 1/93 perl: Upper.xs:2030: su_uplevel_r=
unops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' f=
ailed.
> t/60-uplevel-target.t ............ Failed 40/93 subtests
> t/61-uplevel-args.t .............. 1/297 perl: Upper.xs:2030: su_uplevel_=
runops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' =
failed.
> t/61-uplevel-args.t .............. Failed 257/297 subtests
> t/62-uplevel-return.t ............ 1/69 perl: Upper.xs:2030: su_uplevel_r=
unops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' f=
ailed.
> t/62-uplevel-return.t ............ Failed 25/69 subtests
> t/63-uplevel-ctl.t ............... 1/51 perl: Upper.xs:2030: su_uplevel_r=
unops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' f=
ailed.
> t/63-uplevel-ctl.t ............... Failed 47/51 subtests
> t/64-uplevel-caller.t ............ 1/32 perl: Upper.xs:2030: su_uplevel_r=
unops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' f=
ailed.
> t/64-uplevel-caller.t ............ Failed 8/32 subtests
> t/65-uplevel-multi.t ............. ok
> t/66-uplevel-context.t ........... perl: Upper.xs:2030: su_uplevel_runops=
_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' failed.
> t/66-uplevel-context.t ........... Failed 6/6 subtests
> t/67-uplevel-scope.t ............. perl: Upper.xs:2030: su_uplevel_runops=
_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' failed.
> t/67-uplevel-scope.t ............. Failed 20/20 subtests
> t/69-uplevel-threads.t ........... # Using threads 2.22
> # Using threads::shared 1.58
> # Using usleep() from Time::HiRes 1.9759
> t/69-uplevel-threads.t ........... 1/? perl: Upper.xs:2030: su_uplevel_ru=
nops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fa=
iled.
> t/69-uplevel-threads.t ........... All 3 subtests passed
> t/70-uid-target.t ................ ok
> t/74-uid-validate.t .............. ok
> t/75-uid-uplevel.t ............... perl: Upper.xs:2030: su_uplevel_runops=
_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' failed.
> t/75-uid-uplevel.t ............... Failed 106/106 subtests
> t/79-uid-threads.t ............... # Using threads 2.22
> # Using threads::shared 1.58
> # Using usleep() from Time::HiRes 1.9759
> t/79-uid-threads.t ............... ok
> t/81-stress-level.t .............. ok
> t/84-stress-unwind.t ............. ok
> t/85-stress-yield.t .............. ok
> t/86-stress-uplevel.t ............ ok
> t/87-stress-uid.t ................ 1/? perl: Upper.xs:2030: su_uplevel_ru=
nops_hook_entersub: Assertion `((void*)(((av)->sv_u.svu_array))) !=3D 0' fa=
iled.
> t/87-stress-uid.t ................ All 385 subtests passed
>=20
> Test Summary Report
> -------------------
> t/60-uplevel-target.t          (Wstat: 134 Tests: 53 Failed: 0)
>   Non-zero wait status: 134
>   Parse errors: Bad plan.  You planned 93 tests but ran 53.
> t/61-uplevel-args.t            (Wstat: 134 Tests: 40 Failed: 0)
>   Non-zero wait status: 134
>   Parse errors: Bad plan.  You planned 297 tests but ran 40.
> t/62-uplevel-return.t          (Wstat: 134 Tests: 44 Failed: 0)
>   Non-zero wait status: 134
>   Parse errors: Bad plan.  You planned 69 tests but ran 44.
> t/63-uplevel-ctl.t             (Wstat: 134 Tests: 4 Failed: 0)
>   Non-zero wait status: 134
>   Parse errors: Bad plan.  You planned 51 tests but ran 4.
> t/64-uplevel-caller.t          (Wstat: 134 Tests: 24 Failed: 0)
>   Non-zero wait status: 134
>   Parse errors: Bad plan.  You planned 32 tests but ran 24.
> t/66-uplevel-context.t         (Wstat: 134 Tests: 0 Failed: 0)
>   Non-zero wait status: 134
>   Parse errors: Bad plan.  You planned 6 tests but ran 0.
> t/67-uplevel-scope.t           (Wstat: 134 Tests: 0 Failed: 0)
>   Non-zero wait status: 134
>   Parse errors: Bad plan.  You planned 20 tests but ran 0.
> t/69-uplevel-threads.t         (Wstat: 134 Tests: 3 Failed: 0)
>   Non-zero wait status: 134
>   Parse errors: No plan found in TAP output
> t/75-uid-uplevel.t             (Wstat: 134 Tests: 0 Failed: 0)
>   Non-zero wait status: 134
>   Parse errors: Bad plan.  You planned 106 tests but ran 0.
> t/87-stress-uid.t              (Wstat: 134 Tests: 385 Failed: 0)
>   Non-zero wait status: 134
>   Parse errors: No plan found in TAP output
> Files=3D54, Tests=3D101395, 17 wallclock secs ( 8.33 usr  0.08 sys + 12.1=
3 cusr  1.00 csys =3D 21.54 CPU)
> Result: FAIL
> Failed 10/54 test programs. 0/101395 subtests failed.
> make: *** [Makefile:1067: test_dynamic] Error 255
>   VPIT/Scope-Upper-0.31.tar.gz
>   /usr/bin/make test -- NOT OK
> //hint// to see the cpan-testers results for installing this module, try:
>   reports VPIT/Scope-Upper-0.31.tar.gz
>=20
> $ perl -V
> Summary of my perl5 (revision 5 version 28 subversion 1) configuration:
>=20
>   Platform:
>     osname=3Dlinux
>     osvers=3D4.19.5-1-default
>     archname=3Dx86_64-linux-thread-multi-ld
>     uname=3D'linux lx09 4.19.5-1-default #1 smp preempt tue nov 27 19:56:=
09 utc 2018 (6210279) x86_64 x86_64 x86_64 gnulinux '
>     config_args=3D'-Duse64bitall -Duselongdouble -Dusethreads -Duseithrea=
ds -Duseshrplib -des'
>     hint=3Drecommended
>     useposix=3Dtrue
>     d_sigaction=3Ddefine
>     useithreads=3Ddefine
>     usemultiplicity=3Ddefine
>     use64bitint=3Ddefine
>     use64bitall=3Ddefine
>     uselongdouble=3Ddefine
>     usemymalloc=3Dn
>     default_inc_excludes_dot=3Ddefine
>     bincompat5005=3Dundef
>   Compiler:
>     cc=3D'cc'
>     ccflags =3D'-D_REENTRANT -D_GNU_SOURCE -fPIC -DDEBUGGING -fwrapv -fno=
-strict-aliasing -pipe -fstack-protector-strong -I/pro/local/include -D_LAR=
GEFILE_SOURCE -D_FILE_OFFSET_BITS=3D64'
>     optimize=3D'-O2'
>     cppflags=3D'-D_REENTRANT -D_GNU_SOURCE -fPIC -DDEBUGGING -fwrapv -fno=
-strict-aliasing -pipe -fstack-protector-strong -I/pro/local/include'
>     ccversion=3D''
>     gccversion=3D'8.2.1 20181108 [gcc-8-branch revision 265914]'
>     gccosandvers=3D''
>     intsize=3D4
>     longsize=3D8
>     ptrsize=3D8
>     doublesize=3D8
>     byteorder=3D12345678
>     doublekind=3D3
>     d_longlong=3Ddefine
>     longlongsize=3D8
>     d_longdbl=3Ddefine
>     longdblsize=3D16
>     longdblkind=3D3
>     ivtype=3D'long'
>     ivsize=3D8
>     nvtype=3D'long double'
>     nvsize=3D16
>     Off_t=3D'off_t'
>     lseeksize=3D8
>     alignbytes=3D16
>     prototype=3Ddefine
>   Linker and Libraries:
>     ld=3D'cc'
>     ldflags =3D'-L/pro/local/lib -fstack-protector-strong'
>     libpth=3D/usr/local/lib /usr/lib64/gcc/x86_64-suse-linux/8/include-fi=
xed /usr/lib64/gcc/x86_64-suse-linux/8/../../../../x86_64-suse-linux/lib /u=
sr/lib /pro/local/lib /lib/../lib64 /usr/lib/../lib64 /lib /lib64 /usr/lib6=
4 /usr/local/lib64
>     libs=3D-lpthread -lnsl -lgdbm -ldb -ldl -lm -lcrypt -lutil -lc -lgdbm=
_compat
>     perllibs=3D-lpthread -lnsl -ldl -lm -lcrypt -lutil -lc
>     libc=3Dlibc-2.27.so
>     so=3Dso
>     useshrplib=3Dtrue
>     libperl=3Dlibperl.so
>     gnulibc_version=3D'2.27'
>   Dynamic Linking:
>     dlsrc=3Ddl_dlopen.xs
>     dlext=3Dso
>     d_dlsymun=3Dundef
>     ccdlflags=3D'-Wl,-E -Wl,-rpath,/pro/lib/perl5/5.28.1/x86_64-linux-thr=
ead-multi-ld/CORE'
>     cccdlflags=3D'-fPIC'
>     lddlflags=3D'-shared -O2 -L/pro/local/lib -fstack-protector-strong'
>=20
>=20
> Characteristics of this binary (from libperl):
>   Compile-time options:
>     DEBUGGING
>     HAS_TIMES
>     MULTIPLICITY
>     PERLIO_LAYERS
>     PERL_COPY_ON_WRITE
>     PERL_DONT_CREATE_GVSV
>     PERL_IMPLICIT_CONTEXT
>     PERL_MALLOC_WRAP
>     PERL_OP_PARENT
>     PERL_PRESERVE_IVUV
>     PERL_TRACK_MEMPOOL
>     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_LONG_DOUBLE
>     USE_PERLIO
>     USE_PERL_ATOF
>     USE_REENTRANT_API
>   Built under linux
>   Compiled at Dec  4 2018 15:28:05
>   %ENV:
>     PERL6LIB=3D"inst#/pro/3gl/CPAN/rakudo/install"
>   @INC:
>     /pro/lib/perl5/site_perl/5.28.1/x86_64-linux-thread-multi-ld
>     /pro/lib/perl5/site_perl/5.28.1
>     /pro/lib/perl5/5.28.1/x86_64-linux-thread-multi-ld
>     /pro/lib/perl5/5.28.1
>=20


--=20
H.Merijn Brand  http://tux.nl   Perl Monger  http://amsterdam.pm.org/
using perl5.00307 .. 5.29   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/

--Sig_/xfoNcdJzmT/XJPvPgt_=Rs.
Content-Type: application/pgp-signature
Content-Description: OpenPGP digital signature

-----BEGIN PGP SIGNATURE-----
Version: GnuPG v2

iQEcBAEBAgAGBQJcDpZAAAoJEAOhR6E+XcCYXr0H/0xOkR26aShsMCXF4K7wuL1H
4TleVrqWtEjZ/yw3EhA8S/DDkAuI+o56vKEsnVzppUVDXQnu56q1+aRSrbvrCRoy
z22CoTQGST74OgurQIq+uhCQKJwtY8wOJC1zGFKxi9CSoVgAxXdcJADW8oUpwqP9
BajhMKrRoiB+p7GaNczaXh0jz3+uGgdpbETvBiJiJj9kDNkH8DpSvfzOfChCfFvz
XFH7QM0Mk611teqnZuoCUG66RTxQBJw4Zf2J/TOrj6NicXf6nsFGHyE08Hwx83V3
2PIHCQr+XFUlBeZ22HlHhdg50htDy0TTrgJpRnvmT7fbzZ3KaBBF4vZ28QbWHbY=
=hnlp
-----END PGP SIGNATURE-----

--Sig_/xfoNcdJzmT/XJPvPgt_=Rs.--
0
h
12/10/2018 4:37:15 PM
Reply: