Skip Menu |
Report information
Id: 132913
Status: open
Priority: 0/
Queue: perl5

Owner: Nobody
Requestors: jhi <jhi [at] iki.fi>
jkeenan [at] pobox.com
Cc:
AdminCc:

Operating System: (no value)
PatchStatus: (no value)
Severity: low
Type: unknown
Perl Version: (no value)
Fixed In: (no value)



To: perlbug [...] perl.org
Subject: t/op/sprintf2.t: 6 test failures when compiling with g++7
Date: Mon, 26 Feb 2018 12:32:13 -0500
From: James E Keenan <jkeenan [...] pobox.com>
Download (untitled) / with headers
text/plain 792b
Today I installed the gcc7 package on both FreeBSD-10.3 and 11.0. I then compiled and build perl configuring like this: ##### sh ./Configure -des -Dusedevel -Duseithreads \ -Doptimize="-O2 -pipe -fstack-protector -fno-strict-aliasing" \ -Dcc="g++7" \ -Accflags="-Wl,-rpath=/usr/local/lib/gcc7" \ -Aldflags="-Wl,-rpath=/usr/local/lib/gcc7" ##### There were plenty of warnings during 'make', but since we have not yet heavily explored building perl with gcc7 or g++7, I'll defer discussion of them; 'make' completed successfully. During 'make test_harness', however, I got failures in t/op/sprintf2.t. Please see attachments. Note that we have one other long-standing Open ticket for t/op/sprintf2.t: https://rt.perl.org/Ticket/Display.html?id=122591 Thank you very much. Jim Keenan

Message body is not shown because sender requested not to inline it.

Message body is not shown because sender requested not to inline it.

RT-Send-CC: perl5-porters [...] perl.org
Download (untitled) / with headers
text/plain 1021b
On Mon, 26 Feb 2018 17:32:31 GMT, jkeenan@pobox.com wrote: Show quoted text
> Today I installed the gcc7 package on both FreeBSD-10.3 and 11.0. I > then compiled and build perl configuring like this: > > ##### > sh ./Configure -des -Dusedevel -Duseithreads \ > -Doptimize="-O2 -pipe -fstack-protector -fno-strict-aliasing" \ > -Dcc="g++7" \ > -Accflags="-Wl,-rpath=/usr/local/lib/gcc7" \ > -Aldflags="-Wl,-rpath=/usr/local/lib/gcc7" > ##### > > There were plenty of warnings during 'make', but since we have not yet > heavily explored building perl with gcc7 or g++7, I'll defer discussion > of them; 'make' completed successfully. > > During 'make test_harness', however, I got failures in t/op/sprintf2.t. > Please see attachments. > > Note that we have one other long-standing Open ticket for t/op/sprintf2.t: > > https://rt.perl.org/Ticket/Display.html?id=122591 > > Thank you very much. > Jim Keenan
Curiously, if I configure with '-Dcc="gcc7" instead of g++7, the 6 tests now PASS. -- James E Keenan (jkeenan@cpan.org)
To: James E Keenan <jkeenan [...] pobox.com>, perlbug [...] perl.org
Subject: Re: t/op/sprintf2.t: 6 test failures when compiling with g++7
From: Jarkko Hietaniemi <jhi [...] iki.fi>
Date: Mon, 26 Feb 2018 20:27:02 +0200
CC: Zefram <zefram [...] fysh.org>
Download (untitled) / with headers
text/plain 1.1k
On 2/26/18 19:32, James E Keenan wrote: Show quoted text
> Today I installed the gcc7 package on both FreeBSD-10.3 and 11.0. I > then compiled and build perl configuring like this: > > ##### > sh ./Configure -des -Dusedevel -Duseithreads \ > -Doptimize="-O2 -pipe -fstack-protector -fno-strict-aliasing" \ > -Dcc="g++7" \ > -Accflags="-Wl,-rpath=/usr/local/lib/gcc7" \ > -Aldflags="-Wl,-rpath=/usr/local/lib/gcc7" > ##### > > There were plenty of warnings during 'make', but since we have not yet > heavily explored building perl with gcc7 or g++7, I'll defer discussion > of them; 'make' completed successfully. > > During 'make test_harness', however, I got failures in t/op/sprintf2.t. > Please see attachments.
Off-hand, since the only tests that started failing are to do with subnormal handling, and since the tests have previously passed (right?) with freebsd/gcc (some older gcc), maybe something has changed in the subnormal handling? CCing Zefram explicitly since he usually has a good hunch about these things. Show quoted text
> Note that we have one other long-standing Open ticket for t/op/sprintf2.t: > > https://rt.perl.org/Ticket/Display.html?id=122591 > > Thank you very much. > Jim Keenan >
RT-Send-CC: perl5-porters [...] perl.org
Download (untitled) / with headers
text/plain 1.6k
On Mon, 26 Feb 2018 18:27:16 GMT, jhi wrote: Show quoted text
> On 2/26/18 19:32, James E Keenan wrote:
> > Today I installed the gcc7 package on both FreeBSD-10.3 and 11.0. I > > then compiled and build perl configuring like this: > > > > ##### > > sh ./Configure -des -Dusedevel -Duseithreads \ > > -Doptimize="-O2 -pipe -fstack-protector -fno-strict-aliasing" \ > > -Dcc="g++7" \ > > -Accflags="-Wl,-rpath=/usr/local/lib/gcc7" \ > > -Aldflags="-Wl,-rpath=/usr/local/lib/gcc7" > > ##### > > > > There were plenty of warnings during 'make', but since we have not yet > > heavily explored building perl with gcc7 or g++7, I'll defer discussion > > of them; 'make' completed successfully. > > > > During 'make test_harness', however, I got failures in t/op/sprintf2.t. > > Please see attachments.
> > Off-hand, since the only tests that started failing are to do with > subnormal handling, and since the tests have previously passed (right?) > with freebsd/gcc (some older gcc), maybe something has changed in the > subnormal handling? >
I only installed gcc7 (and, hence, g++7) on FreeBSD-11.0 for the first time today. I previously installed gcc6, but where was some problem installing g++6, so I've never built with the latter on this platform. On IRC #p5p, Tux mentioned that he had observed g++7 test failures several months ago at the time of the core hackathon in Amsterdam (mid-October). Show quoted text
> CCing Zefram explicitly since he usually has a good hunch about these > things. >
> > Note that we have one other long-standing Open ticket for t/op/sprintf2.t: > > > > https://rt.perl.org/Ticket/Display.html?id=122591 > > > > Thank you very much. > > Jim Keenan > >
-- James E Keenan (jkeenan@cpan.org)
Subject: Re: [perl #132913] t/op/sprintf2.t: 6 test failures when compiling with g++7
Date: Mon, 26 Feb 2018 23:29:53 +0100
From: "H.Merijn Brand" <h.m.brand [...] xs4all.nl>
To: perl5-porters [...] perl.org
On Mon, 26 Feb 2018 10:57:52 -0800, "James E Keenan via RT" <perlbug-followup@perl.org> wrote: Show quoted text
> On Mon, 26 Feb 2018 18:27:16 GMT, jhi wrote:
> > On 2/26/18 19:32, James E Keenan wrote:
> > > Today I installed the gcc7 package on both FreeBSD-10.3 and 11.0. I > > > then compiled and build perl configuring like this: > > > > > > ##### > > > sh ./Configure -des -Dusedevel -Duseithreads \ > > > -Doptimize="-O2 -pipe -fstack-protector -fno-strict-aliasing" \ > > > -Dcc="g++7" \ > > > -Accflags="-Wl,-rpath=/usr/local/lib/gcc7" \ > > > -Aldflags="-Wl,-rpath=/usr/local/lib/gcc7" > > > ##### > > > > > > There were plenty of warnings during 'make', but since we have not yet > > > heavily explored building perl with gcc7 or g++7, I'll defer discussion > > > of them; 'make' completed successfully. > > > > > > During 'make test_harness', however, I got failures in t/op/sprintf2.t. > > > Please see attachments.
> > > > Off-hand, since the only tests that started failing are to do with > > subnormal handling, and since the tests have previously passed (right?) > > with freebsd/gcc (some older gcc), maybe something has changed in the > > subnormal handling?
> > I only installed gcc7 (and, hence, g++7) on FreeBSD-11.0 for the > first time today. > > I previously installed gcc6, but where was some problem installing > g++6, so I've never built with the latter on this platform. > > On IRC #p5p, Tux mentioned that he had observed g++7 test failures > several months ago at the time of the core hackathon in Amsterdam > (mid-October).
Several month being over a year now It is g++-6 + g++-7 specific (g++-5 is ok) Search on [All versions] for host [lx09] on http://perl5.test-smoke.org/search Last pass: 2016-06-02 13:31:52+00 v5.25.1-98-gad1bab9 http://perl5.test-smoke.org/report/47121 lx09: Intel(R) Core(TM) i7-2620M CPU @ 2.70GHz (GenuineIntel 1377MHz) (x86_64/4[8 cores] cpus) on linux 4.6.0-1-default [openSUSE Tumbleweed (20160422) (x86_64)] using g++ version 5.3.1 20160412 [gcc-5-branch revision 234894] (*1) smoketime 4 hours 18 minutes (average 32 minutes 21 seconds) First fail: 2016-08-27 19:55:14+00 v5.25.4-21-g51068c1 http://perl5.test-smoke.org/report/49402 lx09: Intel(R) Core(TM) i7-2620M CPU @ 2.70GHz (GenuineIntel 1385MHz) (x86_64/4[8 cores] cpus) on linux 4.7.2-1-default [openSUSE Tumbleweed (20160812) (x86_64)] using g++ version 6.1.1 20160815 [gcc-6-branch revision 239479] (*1) smoketime 2 hours 33 minutes (average 19 minutes 9 seconds) Show quoted text
> > CCing Zefram explicitly since he usually has a good hunch about these > > things. > >
> > > Note that we have one other long-standing Open ticket for t/op/sprintf2.t: > > > > > > https://rt.perl.org/Ticket/Display.html?id=122591 > > > > > > Jim Keenan
-- H.Merijn Brand http://tux.nl Perl Monger http://amsterdam.pm.org/ using perl5.00307 .. 5.27 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/
Download (untitled)
application/pgp-signature 473b

Message body not shown because it is not plain text.



This service is sponsored and maintained by Best Practical Solutions and runs on Perl.org infrastructure.

For issues related to this RT instance (aka "perlbug"), please contact perlbug-admin at perl.org