Bug #117800 for ExtUtils-MakeMaker: don't know how to make dynamic

Mon Sep 12 15:27:02 2016 SREZIC [...] cpan.org - Ticket created

Subject:

don't know how to make dynamic

On my freebsd 10 smoker I observe some CPAN distributions failing like this: Running make test "/usr/perl5.18.4p/bin/perl5.18.4" -MExtUtils::Command::MM -e 'cp_nonempty' -- XS.bs blib/arch/auto/CGI/Deurl/XS/XS.bs 644 make[1]: don't know how to make dynamic. Stop make[1]: stopped in /home/cpansand/.cpan/build/2016091221/CGI-Deurl-XS-0.08-eSw2Lg/parser *** Error code 2 Stop. make: stopped in /home/cpansand/.cpan/build/2016091221/CGI-Deurl-XS-0.08-eSw2Lg A couple of distributions are affected: * NEZUMI/Unicode-LineBreak-2016.003.tar.gz (sample report from my freebsd system: http://www.cpantesters.org/cpan/report/233e64c2-6796-11e6-af57-ad669edbf766 sample report from Bingos' netbsd system: http://www.cpantesters.org/cpan/report/36e9485e-2969-11e6-ab41-c893a58a4b8c ) * ATHOMASON/CGI-Deurl-XS-0.08.tar.g (sample report from a windows system: http://www.cpantesters.org/cpan/report/7c1f8cbe-6bf9-1014-96ea-4dcac45e766f ) * KROKI/Cache-Memcached-Fast-0.23.tar.gz * GRAY/Compress-LZ4-0.22.tar.gz * AJGB/Crypt-Curve25519-0.05.tar.gz * BORISZ/Crypt-UnixCrypt_XS-0.10.tar.gz (sample report from a netbsd system: http://www.cpantesters.org/cpan/report/14155f50-5cd3-11e6-9fc5-14bfbd15da07 ) * MDEHOON/Algorithm-Cluster-1.52.tar.gz * TPODER/Net-NfDump-1.21.tar.gz * TPODER/Net-NfDump-1.22.tar.gz * CALLAHAN/Math-CDF-0.1.tar.gz (sample report from an openbsd system: http://www.cpantesters.org/cpan/report/5804b198-56f7-11e6-b1db-3e644bb50351 ) GRAY/Compress-LZ4-0.23.tar.gz has a workaround for the problem, and refers to https://rt.cpan.org/Ticket/Display.html?id=7417 ("SKIP doesn't work on NetBSD make") The problem does not happen with EUMM 7.10.

Mon Sep 12 15:54:18 2016 LEONT [...] cpan.org - Correspondence added

On Mon Sep 12 15:27:02 2016, SREZIC wrote: Show quoted text

> On my freebsd 10 smoker I observe some CPAN distributions failing like > this: > > Running make test > "/usr/perl5.18.4p/bin/perl5.18.4" -MExtUtils::Command::MM -e > 'cp_nonempty' -- XS.bs blib/arch/auto/CGI/Deurl/XS/XS.bs 644 > make[1]: don't know how to make dynamic. Stop > > make[1]: stopped in /home/cpansand/.cpan/build/2016091221/CGI-Deurl- > XS-0.08-eSw2Lg/parser > *** Error code 2 > > Stop. > make: stopped in /home/cpansand/.cpan/build/2016091221/CGI-Deurl-XS- > 0.08-eSw2Lg > > > A couple of distributions are affected: > * NEZUMI/Unicode-LineBreak-2016.003.tar.gz > (sample report from my freebsd system: > http://www.cpantesters.org/cpan/report/233e64c2-6796-11e6-af57- > ad669edbf766 > sample report from Bingos' netbsd system: > http://www.cpantesters.org/cpan/report/36e9485e-2969-11e6-ab41- > c893a58a4b8c > ) > * ATHOMASON/CGI-Deurl-XS-0.08.tar.g > (sample report from a windows system: > http://www.cpantesters.org/cpan/report/7c1f8cbe-6bf9-1014-96ea- > 4dcac45e766f > ) > * KROKI/Cache-Memcached-Fast-0.23.tar.gz > * GRAY/Compress-LZ4-0.22.tar.gz > * AJGB/Crypt-Curve25519-0.05.tar.gz > * BORISZ/Crypt-UnixCrypt_XS-0.10.tar.gz > (sample report from a netbsd system: > http://www.cpantesters.org/cpan/report/14155f50-5cd3-11e6-9fc5- > 14bfbd15da07 ) > * MDEHOON/Algorithm-Cluster-1.52.tar.gz > * TPODER/Net-NfDump-1.21.tar.gz > * TPODER/Net-NfDump-1.22.tar.gz > * CALLAHAN/Math-CDF-0.1.tar.gz > (sample report from an openbsd system: > http://www.cpantesters.org/cpan/report/5804b198-56f7-11e6-b1db- > 3e644bb50351 ) > > GRAY/Compress-LZ4-0.23.tar.gz has a workaround for the problem, and > refers to https://rt.cpan.org/Ticket/Display.html?id=7417 ("SKIP > doesn't work on NetBSD make") > > The problem does not happen with EUMM 7.10.

The thing that seems to trigger it is using sub-Makefile.PLs with both a SKIP hash and a MY::top_targets. I'm guessing the other part of this is bsdmake being pickier than gmake about something. A related bug was fixed in 7.16 (it may be worthwhile to check if this isn't a regression between 7.14 and 7.16, but I would expect not) Leon

Mon Sep 12 15:54:20 2016 The RT System itself - Status changed from 'new' to 'open'

Mon Sep 12 16:00:24 2016 SREZIC [...] cpan.org - Correspondence added

On 2016-09-12 15:54:18, LEONT wrote: Show quoted text

> On Mon Sep 12 15:27:02 2016, SREZIC wrote:

> > On my freebsd 10 smoker I observe some CPAN distributions failing > > like > > this: > > > > Running make test > > "/usr/perl5.18.4p/bin/perl5.18.4" -MExtUtils::Command::MM -e > > 'cp_nonempty' -- XS.bs blib/arch/auto/CGI/Deurl/XS/XS.bs 644 > > make[1]: don't know how to make dynamic. Stop > > > > make[1]: stopped in /home/cpansand/.cpan/build/2016091221/CGI-Deurl- > > XS-0.08-eSw2Lg/parser > > *** Error code 2 > > > > Stop. > > make: stopped in /home/cpansand/.cpan/build/2016091221/CGI-Deurl-XS- > > 0.08-eSw2Lg > > > > > > A couple of distributions are affected: > > * NEZUMI/Unicode-LineBreak-2016.003.tar.gz > > (sample report from my freebsd system: > > http://www.cpantesters.org/cpan/report/233e64c2-6796-11e6-af57- > > ad669edbf766 > > sample report from Bingos' netbsd system: > > http://www.cpantesters.org/cpan/report/36e9485e-2969-11e6-ab41- > > c893a58a4b8c > > ) > > * ATHOMASON/CGI-Deurl-XS-0.08.tar.g > > (sample report from a windows system: > > http://www.cpantesters.org/cpan/report/7c1f8cbe-6bf9-1014-96ea- > > 4dcac45e766f > > ) > > * KROKI/Cache-Memcached-Fast-0.23.tar.gz > > * GRAY/Compress-LZ4-0.22.tar.gz > > * AJGB/Crypt-Curve25519-0.05.tar.gz > > * BORISZ/Crypt-UnixCrypt_XS-0.10.tar.gz > > (sample report from a netbsd system: > > http://www.cpantesters.org/cpan/report/14155f50-5cd3-11e6-9fc5- > > 14bfbd15da07 ) > > * MDEHOON/Algorithm-Cluster-1.52.tar.gz > > * TPODER/Net-NfDump-1.21.tar.gz > > * TPODER/Net-NfDump-1.22.tar.gz > > * CALLAHAN/Math-CDF-0.1.tar.gz > > (sample report from an openbsd system: > > http://www.cpantesters.org/cpan/report/5804b198-56f7-11e6-b1db- > > 3e644bb50351 ) > > > > GRAY/Compress-LZ4-0.23.tar.gz has a workaround for the problem, and > > refers to https://rt.cpan.org/Ticket/Display.html?id=7417 ("SKIP > > doesn't work on NetBSD make") > > > > The problem does not happen with EUMM 7.10.

> > The thing that seems to trigger it is using sub-Makefile.PLs with both > a SKIP hash and a MY::top_targets. I'm guessing the other part of this > is bsdmake being pickier than gmake about something. A related bug was > fixed in 7.16 (it may be worthwhile to check if this isn't a > regression between 7.14 and 7.16, but I would expect not) > > Leon

I checked with CGI-Deurl-XS-0.08: 7.10 & 7.14 -> PASS 7.16 -> FAIL

Mon Sep 12 16:34:37 2016 LEONT [...] cpan.org - Correspondence added

On Mon Sep 12 16:00:24 2016, SREZIC wrote: Show quoted text

> On 2016-09-12 15:54:18, LEONT wrote:

> > On Mon Sep 12 15:27:02 2016, SREZIC wrote:

> > > On my freebsd 10 smoker I observe some CPAN distributions failing > > > like > > > this: > > > > > > Running make test > > > "/usr/perl5.18.4p/bin/perl5.18.4" -MExtUtils::Command::MM -e > > > 'cp_nonempty' -- XS.bs blib/arch/auto/CGI/Deurl/XS/XS.bs 644 > > > make[1]: don't know how to make dynamic. Stop > > > > > > make[1]: stopped in /home/cpansand/.cpan/build/2016091221/CGI-Deurl- > > > XS-0.08-eSw2Lg/parser > > > *** Error code 2 > > > > > > Stop. > > > make: stopped in /home/cpansand/.cpan/build/2016091221/CGI-Deurl-XS- > > > 0.08-eSw2Lg > > > > > > > > > A couple of distributions are affected: > > > * NEZUMI/Unicode-LineBreak-2016.003.tar.gz > > > (sample report from my freebsd system: > > > http://www.cpantesters.org/cpan/report/233e64c2-6796-11e6-af57- > > > ad669edbf766 > > > sample report from Bingos' netbsd system: > > > http://www.cpantesters.org/cpan/report/36e9485e-2969-11e6-ab41- > > > c893a58a4b8c > > > ) > > > * ATHOMASON/CGI-Deurl-XS-0.08.tar.g > > > (sample report from a windows system: > > > http://www.cpantesters.org/cpan/report/7c1f8cbe-6bf9-1014-96ea- > > > 4dcac45e766f > > > ) > > > * KROKI/Cache-Memcached-Fast-0.23.tar.gz > > > * GRAY/Compress-LZ4-0.22.tar.gz > > > * AJGB/Crypt-Curve25519-0.05.tar.gz > > > * BORISZ/Crypt-UnixCrypt_XS-0.10.tar.gz > > > (sample report from a netbsd system: > > > http://www.cpantesters.org/cpan/report/14155f50-5cd3-11e6-9fc5- > > > 14bfbd15da07 ) > > > * MDEHOON/Algorithm-Cluster-1.52.tar.gz > > > * TPODER/Net-NfDump-1.21.tar.gz > > > * TPODER/Net-NfDump-1.22.tar.gz > > > * CALLAHAN/Math-CDF-0.1.tar.gz > > > (sample report from an openbsd system: > > > http://www.cpantesters.org/cpan/report/5804b198-56f7-11e6-b1db- > > > 3e644bb50351 ) > > > > > > GRAY/Compress-LZ4-0.23.tar.gz has a workaround for the problem, and > > > refers to https://rt.cpan.org/Ticket/Display.html?id=7417 ("SKIP > > > doesn't work on NetBSD make") > > > > > > The problem does not happen with EUMM 7.10.

> > > > The thing that seems to trigger it is using sub-Makefile.PLs with both > > a SKIP hash and a MY::top_targets. I'm guessing the other part of this > > is bsdmake being pickier than gmake about something. A related bug was > > fixed in 7.16 (it may be worthwhile to check if this isn't a > > regression between 7.14 and 7.16, but I would expect not) > > > > Leon

> > I checked with CGI-Deurl-XS-0.08: > > 7.10 & 7.14 -> PASS > 7.16 -> FAIL

Thanks, that is most informative! Leon

Mon Sep 12 19:26:35 2016 ETJ [...] cpan.org - Correspondence added

It would be very helpful if someone with a BSD setup could attach a Makefile that fails, so I can compare it with one that my setup generates.

Tue Sep 13 01:21:34 2016 SREZIC [...] cpan.org - Correspondence added

On 2016-09-12 19:26:35, ETJ wrote: Show quoted text

> It would be very helpful if someone with a BSD setup could attach a > Makefile that fails, so I can compare it with one that my setup > generates.

Done, the both Makefiles created in CGI-Deurl-XS-0.08 (root Makefile + the one in the "parser" subdir). This is on a FreeBSD 10.1 system.

Subject:

Makefile

Download Makefile
application/octet-stream 43.6k

Message body not shown because it is not plain text.

Subject:

parser-Makefile

Download parser-Makefile
application/octet-stream 34.5k

Message body not shown because it is not plain text.

Sun Sep 18 12:06:26 2016 LEONT [...] cpan.org - Correspondence added

On Mon Sep 12 16:00:24 2016, SREZIC wrote: Show quoted text

> I checked with CGI-Deurl-XS-0.08: > > 7.10 & 7.14 -> PASS > 7.16 -> FAIL >

What happens is that test defined a dependency on test_dynamic (or actually test_$(LINKTYPE)), which depends on dynamic even when SKIP=>dynamic is set (this may or may not be correct), but those Makefile.PLs don't define such a target in their MY::top_targets. bmake is perfectly correct to conclude a dependency isn't defined so it bails out. gmake accepts this because dynamic is mentioned in the .PHONY section (if you take it out it will complain loudly). I guess test_$(LINKTYPE) needs to depend on something else (pure_all?) when SKIP=> $LINKTYPE is present. Leon

Sat Feb 25 14:27:25 2017 SREZIC [...] cpan.org - Reference by ticket #120370 added

Sat Feb 25 14:28:09 2017 SREZIC [...] cpan.org - Correspondence added

On 2016-09-18 12:06:26, LEONT wrote: Show quoted text

> On Mon Sep 12 16:00:24 2016, SREZIC wrote:

> > I checked with CGI-Deurl-XS-0.08: > > > > 7.10 & 7.14 -> PASS > > 7.16 -> FAIL > >

> > What happens is that test defined a dependency on test_dynamic (or > actually test_$(LINKTYPE)), which depends on dynamic even when > SKIP=>dynamic is set (this may or may not be correct), but those > Makefile.PLs don't define such a target in their MY::top_targets. > bmake is perfectly correct to conclude a dependency isn't defined so > it bails out. gmake accepts this because dynamic is mentioned in the > .PHONY section (if you take it out it will complain loudly). > > I guess test_$(LINKTYPE) needs to depend on something else (pure_all?) > when SKIP=> $LINKTYPE is present. > > Leon

I created a patch for Unicode-Linebreak: https://rt.cpan.org/Ticket/Display.html?id=120370

Mon Feb 27 01:28:51 2017 SREZIC [...] cpan.org - Reference by ticket #120407 added

Mon Feb 27 14:14:11 2017 SREZIC [...] cpan.org - Reference by ticket #120432 added

Mon Feb 27 15:42:51 2017 SREZIC [...] cpan.org - Reference by ticket #120440 added

Mon Feb 27 15:59:40 2017 SREZIC [...] cpan.org - Reference by ticket #120442 added

Mon Feb 27 17:10:28 2017 SREZIC [...] cpan.org - Reference by ticket #120447 added

Wed Mar 01 15:06:18 2017 SREZIC [...] cpan.org - Reference by https://github.com/ajgb/crypt-curve25519/issues/5 added

Wed Mar 01 15:08:15 2017 SREZIC [...] cpan.org - Correspondence added

On 2017-02-25 14:28:09, SREZIC wrote: Show quoted text

> On 2016-09-18 12:06:26, LEONT wrote:

> > On Mon Sep 12 16:00:24 2016, SREZIC wrote:

> > > I checked with CGI-Deurl-XS-0.08: > > > > > > 7.10 & 7.14 -> PASS > > > 7.16 -> FAIL > > >

> > > > What happens is that test defined a dependency on test_dynamic (or > > actually test_$(LINKTYPE)), which depends on dynamic even when > > SKIP=>dynamic is set (this may or may not be correct), but those > > Makefile.PLs don't define such a target in their MY::top_targets. > > bmake is perfectly correct to conclude a dependency isn't defined so > > it bails out. gmake accepts this because dynamic is mentioned in the > > .PHONY section (if you take it out it will complain loudly). > > > > I guess test_$(LINKTYPE) needs to depend on something else > > (pure_all?) > > when SKIP=> $LINKTYPE is present. > > > > Leon

> > I created a patch for Unicode-Linebreak: > https://rt.cpan.org/Ticket/Display.html?id=120370

All of the mentioned CPAN modules have now proposed patches (see links), or have later versions with a fix for the problem.

Sat Mar 04 02:40:35 2017 SREZIC [...] cpan.org - Reference by ticket #120495 added

Fri Mar 24 15:04:06 2017 SREZIC [...] cpan.org - Reference by ticket #120746 added

Mon Jul 24 17:36:54 2017 SREZIC [...] cpan.org - Reference by ticket #122593 added

Mon Jan 15 14:09:18 2018 SREZIC [...] cpan.org - Reference by ticket #124090 added

Mon Jan 15 14:10:09 2018 SREZIC [...] cpan.org - Correspondence added

On 2017-03-01 15:08:15, SREZIC wrote: Show quoted text

> On 2017-02-25 14:28:09, SREZIC wrote:

> > On 2016-09-18 12:06:26, LEONT wrote:

> > > On Mon Sep 12 16:00:24 2016, SREZIC wrote:

> > > > I checked with CGI-Deurl-XS-0.08: > > > > > > > > 7.10 & 7.14 -> PASS > > > > 7.16 -> FAIL > > > >

> > > > > > What happens is that test defined a dependency on test_dynamic (or > > > actually test_$(LINKTYPE)), which depends on dynamic even when > > > SKIP=>dynamic is set (this may or may not be correct), but those > > > Makefile.PLs don't define such a target in their MY::top_targets. > > > bmake is perfectly correct to conclude a dependency isn't defined > > > so > > > it bails out. gmake accepts this because dynamic is mentioned in > > > the > > > .PHONY section (if you take it out it will complain loudly). > > > > > > I guess test_$(LINKTYPE) needs to depend on something else > > > (pure_all?) > > > when SKIP=> $LINKTYPE is present. > > > > > > Leon

> > > > I created a patch for Unicode-Linebreak: > > https://rt.cpan.org/Ticket/Display.html?id=120370

> > All of the mentioned CPAN modules have now proposed patches (see > links), or have later versions with a fix for the problem.

Another distribution with this problem (seen on FreeBSD 9 & 10 systems): XML-CanonicalizeXML-0.09

Sun Mar 25 17:19:13 2018 SREZIC [...] cpan.org - Correspondence added

On 2018-01-15 14:10:09, SREZIC wrote: Show quoted text

> On 2017-03-01 15:08:15, SREZIC wrote:

> > On 2017-02-25 14:28:09, SREZIC wrote:

> > > On 2016-09-18 12:06:26, LEONT wrote:

> > > > On Mon Sep 12 16:00:24 2016, SREZIC wrote:

> > > > > I checked with CGI-Deurl-XS-0.08: > > > > > > > > > > 7.10 & 7.14 -> PASS > > > > > 7.16 -> FAIL > > > > >

> > > > > > > > What happens is that test defined a dependency on test_dynamic > > > > (or > > > > actually test_$(LINKTYPE)), which depends on dynamic even when > > > > SKIP=>dynamic is set (this may or may not be correct), but those > > > > Makefile.PLs don't define such a target in their MY::top_targets. > > > > bmake is perfectly correct to conclude a dependency isn't defined > > > > so > > > > it bails out. gmake accepts this because dynamic is mentioned in > > > > the > > > > .PHONY section (if you take it out it will complain loudly). > > > > > > > > I guess test_$(LINKTYPE) needs to depend on something else > > > > (pure_all?) > > > > when SKIP=> $LINKTYPE is present. > > > > > > > > Leon

> > > > > > I created a patch for Unicode-Linebreak: > > > https://rt.cpan.org/Ticket/Display.html?id=120370

> > > > All of the mentioned CPAN modules have now proposed patches (see > > links), or have later versions with a fix for the problem.

> > Another distribution with this problem (seen on FreeBSD 9 & 10 > systems): XML-CanonicalizeXML-0.09

Another victim: Tk-804.034. Detected by JKEENAN: https://rt.cpan.org/Ticket/Display.html?id=124910

Sun Mar 25 17:19:34 2018 SREZIC [...] cpan.org - Reference to ticket #124910 added

Sun Mar 25 20:26:25 2018 jkeenan [...] cpan.org - Correspondence added

On Sun Mar 25 17:19:13 2018, SREZIC wrote: Show quoted text

> On 2018-01-15 14:10:09, SREZIC wrote:

> > On 2017-03-01 15:08:15, SREZIC wrote:

> > > On 2017-02-25 14:28:09, SREZIC wrote:

> > > > On 2016-09-18 12:06:26, LEONT wrote:

> > > > > On Mon Sep 12 16:00:24 2016, SREZIC wrote:

> > > > > > I checked with CGI-Deurl-XS-0.08: > > > > > > > > > > > > 7.10 & 7.14 -> PASS > > > > > > 7.16 -> FAIL > > > > > >

> > > > > > > > > > What happens is that test defined a dependency on test_dynamic > > > > > (or > > > > > actually test_$(LINKTYPE)), which depends on dynamic even when > > > > > SKIP=>dynamic is set (this may or may not be correct), but > > > > > those > > > > > Makefile.PLs don't define such a target in their > > > > > MY::top_targets. > > > > > bmake is perfectly correct to conclude a dependency isn't > > > > > defined > > > > > so > > > > > it bails out. gmake accepts this because dynamic is mentioned > > > > > in > > > > > the > > > > > .PHONY section (if you take it out it will complain loudly). > > > > > > > > > > I guess test_$(LINKTYPE) needs to depend on something else > > > > > (pure_all?) > > > > > when SKIP=> $LINKTYPE is present. > > > > > > > > > > Leon

> > > > > > > > I created a patch for Unicode-Linebreak: > > > > https://rt.cpan.org/Ticket/Display.html?id=120370

> > > > > > All of the mentioned CPAN modules have now proposed patches (see > > > links), or have later versions with a fix for the problem.

> > > > Another distribution with this problem (seen on FreeBSD 9 & 10 > > systems): XML-CanonicalizeXML-0.09

> > Another victim: Tk-804.034. Detected by JKEENAN: > https://rt.cpan.org/Ticket/Display.html?id=124910

Setting aside the problem I found with Tk-804.034, it appears that most of the distributions cited in this RT have been fixed. The two exceptions I found today in testing against perl 5 blead (commit 811612a11e) on FreeBSD-11.0 with 'cpanm' were: Net::NfDump Math::CDF build.log available upon request. Thank you very much. Jim Keenan

Sun Jul 08 18:21:32 2018 jkeenan [...] cpan.org - Correspondence added

On Sun Mar 25 20:26:25 2018, JKEENAN wrote: Show quoted text

> On Sun Mar 25 17:19:13 2018, SREZIC wrote:

> > On 2018-01-15 14:10:09, SREZIC wrote:

> > > On 2017-03-01 15:08:15, SREZIC wrote:

> > > > On 2017-02-25 14:28:09, SREZIC wrote:

> > > > > On 2016-09-18 12:06:26, LEONT wrote:

> > > > > > On Mon Sep 12 16:00:24 2016, SREZIC wrote:

> > > > > > > I checked with CGI-Deurl-XS-0.08: > > > > > > > > > > > > > > 7.10 & 7.14 -> PASS > > > > > > > 7.16 -> FAIL > > > > > > >

> > > > > > > > > > > > What happens is that test defined a dependency on > > > > > > test_dynamic > > > > > > (or > > > > > > actually test_$(LINKTYPE)), which depends on dynamic even > > > > > > when > > > > > > SKIP=>dynamic is set (this may or may not be correct), but > > > > > > those > > > > > > Makefile.PLs don't define such a target in their > > > > > > MY::top_targets. > > > > > > bmake is perfectly correct to conclude a dependency isn't > > > > > > defined > > > > > > so > > > > > > it bails out. gmake accepts this because dynamic is mentioned > > > > > > in > > > > > > the > > > > > > .PHONY section (if you take it out it will complain loudly). > > > > > > > > > > > > I guess test_$(LINKTYPE) needs to depend on something else > > > > > > (pure_all?) > > > > > > when SKIP=> $LINKTYPE is present. > > > > > > > > > > > > Leon

> > > > > > > > > > I created a patch for Unicode-Linebreak: > > > > > https://rt.cpan.org/Ticket/Display.html?id=120370

> > > > > > > > All of the mentioned CPAN modules have now proposed patches (see > > > > links), or have later versions with a fix for the problem.

> > > > > > Another distribution with this problem (seen on FreeBSD 9 & 10 > > > systems): XML-CanonicalizeXML-0.09

> > > > Another victim: Tk-804.034. Detected by JKEENAN: > > https://rt.cpan.org/Ticket/Display.html?id=124910

> > Setting aside the problem I found with Tk-804.034, it appears that > most of the distributions cited in this RT have been fixed. The two > exceptions I found today in testing against perl 5 blead (commit > 811612a11e) on FreeBSD-11.0 with 'cpanm' were: > > Net::NfDump > Math::CDF > > build.log available upon request. > > Thank you very much. > Jim Keenan

Today I again attempted to install Net::NfDump and Math::CDF against perl 5 blead using cpanm as the installer and ExtUtils::MakeMaker version 7.34. Both are still failing. ##### $ ./bin/perl -Ilib -v |head -2 | tail -1 This is perl 5, version 29, subversion 1 (v5.29.1 (v5.29.0-19-g4c821bdad5)) built for amd64-freebsd $ ./bin/perl -Ilib -MExtUtils::MakeMaker -E 'say $ExtUtils::MakeMaker::VERSION;' 7.34 $ ./bin/cpanm Net::NfDump Math::CDF ... --> Working on Net::NfDump Fetching http://www.cpan.org/authors/id/T/TP/TPODER/Net-NfDump-1.25.tar.gz ... OK Configuring Net-NfDump-1.25 ... OK Building and testing Net-NfDump-1.25 ... FAIL ! Installing Net::NfDump failed. See /home/jkeenan/.cpanm/work/1531087641.67094/build.log for details. Retry with --force to force install it. --> Working on Math::CDF Fetching http://www.cpan.org/authors/id/C/CA/CALLAHAN/Math-CDF-0.1.tar.gz ... OK Configuring Math-CDF-0.1 ... OK Building and testing Math-CDF-0.1 ... FAIL ! Installing Math::CDF failed. See /home/jkeenan/.cpanm/work/1531087641.67094/build.log for details. Retry with --force to force install it. 1 distribution installed ##### Attaching excerpts from the cpan build.log Thank you very much. Jim Keenan

Subject:

cpan-117800.Math-CDF.build.log

Searching Math::CDF () on cpanmetadb ... --> Working on Math::CDF Fetching http://www.cpan.org/authors/id/C/CA/CALLAHAN/Math-CDF-0.1.tar.gz -> OK Unpacking Math-CDF-0.1.tar.gz Entering Math-CDF-0.1 META.yml/json not found. Creating skeleton for it. Configuring Math-CDF-0.1 Running Makefile.PL Checking if your kit is complete... Looks good Writing MYMETA.yml and MYMETA.json Generating a Unix-style Makefile Writing Makefile for Math::CDF Writing MYMETA.yml and MYMETA.json -> OK Checking dependencies from MYMETA.json ... Checking if you have ExtUtils::MakeMaker 0 ... Yes (7.34) Building and testing Math-CDF-0.1 cp CDF.pm blib/lib/Math/CDF.pm AutoSplitting blib/lib/Math/CDF.pm (blib/lib/auto/Math/CDF) cd cdflib && make LIBPERL_A="libperl.a" LINKTYPE="dynamic" OPTIMIZE="-O2" PREFIX="/home/jkeenan/testing/blead" PASTHRU_DEFINE=' ' PASTHRU_INC=' ' cc -c -DHAS_FPSETMASK -DHAS_FLOATINGPOINT_H -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include -D_FORTIFY_SOURCE=2 -O2 -DVERSION=\"\" -DXS_VERSION=\"\" -DPIC -fPIC "-I/home/jkeenan/testing/blead/lib/perl5/5.29.1/amd64-freebsd/CORE" dcdflib.c dcdflib.c:1889:36: warning: '&&' within '||' [-Wlogical-op-parentheses] if(!(*s < 0.0e0 || *which != 3 && *s > *xn)) goto S160; ~~ ~~~~~~~~~~~~^~~~~~~~~~~ dcdflib.c:1889:36: note: place parentheses around the '&&' expression to silence this warning if(!(*s < 0.0e0 || *which != 3 && *s > *xn)) goto S160; ^ ( ) dcdflib.c:3551:20: warning: '&&' within '||' [-Wlogical-op-parentheses] if(!(qporq && cum > 1.5e0 || !qporq && ccum > 1.5e0)) goto S280; ~~~~~~^~~~~~~~~~~~~~ ~~ dcdflib.c:3551:20: note: place parentheses around the '&&' expression to silence this warning if(!(qporq && cum > 1.5e0 || !qporq && ccum > 1.5e0)) goto S280; ^ ( ) dcdflib.c:3551:45: warning: '&&' within '||' [-Wlogical-op-parentheses] if(!(qporq && cum > 1.5e0 || !qporq && ccum > 1.5e0)) goto S280; ~~ ~~~~~~~^~~~~~~~~~~~~~~ dcdflib.c:3551:45: note: place parentheses around the '&&' expression to silence this warning if(!(qporq && cum > 1.5e0 || !qporq && ccum > 1.5e0)) goto S280; ^ ( ) dcdflib.c:6410:17: warning: '&&' within '||' [-Wlogical-op-parentheses] qup = qincr && yy < 0.0e0 || !qincr && yy > 0.0e0; ~~~~~~^~~~~~~~~~~~~ ~~ dcdflib.c:6410:17: note: place parentheses around the '&&' expression to silence this warning qup = qincr && yy < 0.0e0 || !qincr && yy > 0.0e0; ^ ( ) dcdflib.c:6410:41: warning: '&&' within '||' [-Wlogical-op-parentheses] qup = qincr && yy < 0.0e0 || !qincr && yy > 0.0e0; ~~ ~~~~~~~^~~~~~~~~~~~~ dcdflib.c:6410:41: note: place parentheses around the '&&' expression to silence this warning qup = qincr && yy < 0.0e0 || !qincr && yy > 0.0e0; ^ ( ) dcdflib.c:6434:18: warning: '&&' within '||' [-Wlogical-op-parentheses] qbdd = qincr && yy >= 0.0e0 || !qincr && yy <= 0.0e0; ~~~~~~^~~~~~~~~~~~~~ ~~ dcdflib.c:6434:18: note: place parentheses around the '&&' expression to silence this warning qbdd = qincr && yy >= 0.0e0 || !qincr && yy <= 0.0e0; ^ ( ) dcdflib.c:6434:43: warning: '&&' within '||' [-Wlogical-op-parentheses] qbdd = qincr && yy >= 0.0e0 || !qincr && yy <= 0.0e0; ~~ ~~~~~~~^~~~~~~~~~~~~~ dcdflib.c:6434:43: note: place parentheses around the '&&' expression to silence this warning qbdd = qincr && yy >= 0.0e0 || !qincr && yy <= 0.0e0; ^ ( ) dcdflib.c:6475:18: warning: '&&' within '||' [-Wlogical-op-parentheses] qbdd = qincr && yy <= 0.0e0 || !qincr && yy >= 0.0e0; ~~~~~~^~~~~~~~~~~~~~ ~~ dcdflib.c:6475:18: note: place parentheses around the '&&' expression to silence this warning qbdd = qincr && yy <= 0.0e0 || !qincr && yy >= 0.0e0; ^ ( ) dcdflib.c:6475:43: warning: '&&' within '||' [-Wlogical-op-parentheses] qbdd = qincr && yy <= 0.0e0 || !qincr && yy >= 0.0e0; ~~ ~~~~~~~^~~~~~~~~~~~~~ dcdflib.c:6475:43: note: place parentheses around the '&&' expression to silence this warning qbdd = qincr && yy <= 0.0e0 || !qincr && yy >= 0.0e0; ^ ( ) dcdflib.c:6877:26: warning: '&&' within '||' [-Wlogical-op-parentheses] qrzero = fc >= 0.0e0 && fb <= 0.0e0 || fc < 0.0e0 && fb >= 0.0e0; ~~~~~~~~~~~~^~~~~~~~~~~~~~ ~~ dcdflib.c:6877:26: note: place parentheses around the '&&' expression to silence this warning qrzero = fc >= 0.0e0 && fb <= 0.0e0 || fc < 0.0e0 && fb >= 0.0e0; ^ ( ) dcdflib.c:6877:55: warning: '&&' within '||' [-Wlogical-op-parentheses] qrzero = fc >= 0.0e0 && fb <= 0.0e0 || fc < 0.0e0 && fb >= 0.0e0; ~~ ~~~~~~~~~~~^~~~~~~~~~~~~~ dcdflib.c:6877:55: note: place parentheses around the '&&' expression to silence this warning qrzero = fc >= 0.0e0 && fb <= 0.0e0 || fc < 0.0e0 && fb >= 0.0e0; ^ ( ) 11 warnings generated. cc -c -DHAS_FPSETMASK -DHAS_FLOATINGPOINT_H -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include -D_FORTIFY_SOURCE=2 -O2 -DVERSION=\"\" -DXS_VERSION=\"\" -DPIC -fPIC "-I/home/jkeenan/testing/blead/lib/perl5/5.29.1/amd64-freebsd/CORE" ipmpar.c ar cr libcdflib.a dcdflib.o ipmpar.o : libcdflib.a Running Mkbootstrap for CDF () chmod 644 "CDF.bs" "/usr/home/jkeenan/testing/blead/bin/perl" -MExtUtils::Command::MM -e 'cp_nonempty' -- CDF.bs blib/arch/auto/Math/CDF/CDF.bs 644 "/usr/home/jkeenan/testing/blead/bin/perl" "/home/jkeenan/testing/blead/lib/perl5/5.29.1/ExtUtils/xsubpp" -typemap '/home/jkeenan/testing/blead/lib/perl5/5.29.1/ExtUtils/typemap' CDF.xs > CDF.xsc mv CDF.xsc CDF.c cc -c -DHAS_FPSETMASK -DHAS_FLOATINGPOINT_H -fno-strict-aliasing -pipe -fstack-protector-strong -I/usr/local/include -D_FORTIFY_SOURCE=2 -O2 -DVERSION=\"0.1\" -DXS_VERSION=\"0.1\" -DPIC -fPIC "-I/home/jkeenan/testing/blead/lib/perl5/5.29.1/amd64-freebsd/CORE" CDF.c rm -f blib/arch/auto/Math/CDF/CDF.so cc -shared -L/usr/local/lib -fstack-protector-strong CDF.o -o blib/arch/auto/Math/CDF/CDF.so cdflib/libcdflib.a -lm chmod 755 blib/arch/auto/Math/CDF/CDF.so "/usr/home/jkeenan/testing/blead/bin/perl" -MExtUtils::Command::MM -e 'cp_nonempty' -- CDF.bs blib/arch/auto/Math/CDF/CDF.bs 644 make[1]: don't know how to make test_dynamic. Stop make[1]: stopped in /usr/home/jkeenan/.cpanm/work/1531087641.67094/Math-CDF-0.1/cdflib *** Error code 2 Stop. make: stopped in /usr/home/jkeenan/.cpanm/work/1531087641.67094/Math-CDF-0.1 -> FAIL Installing Math::CDF failed. See /home/jkeenan/.cpanm/work/1531087641.67094/build.log for details. Retry with --force to force install it.

Subject:

cpan-117800.Net-NfDump.build.log

Message body is not shown because it is too large.

Sun Jul 08 18:31:29 2018 jkeenan [...] cpan.org - Reference by ticket #125798 added