# New Ticket Created by Petr Pisar # Please include the string: [perl #123784] # in the subject line of all future correspondence about this issue. # <URL: https://rt.perl.org/Ticket/Display.html?id=123784 > Hello, I don't know if this is the best news before releasing perl 5.20.2, but none of the perl branches (including blead) does not pass tests when built with GCC 5.0: t/run/switches ................................................. "EACCES" is not exported by the Errno module "EISDIR" is not exported by the Errno module Can't continue after import errors at run/switches.t line 126. # Looks like you planned 115 tests but ran 8. FAILED--expected 115 tests, saw 8 etc. This is caused by wrongly generated Errno.h which is missing all the constants (e.g. Errno::EISDIR): --- /usr/lib64/perl5/Errno.pm 2015-01-23 15:15:43.000000000 +0100 +++ ext/Errno/Errno.pm 2015-02-10 16:06:47.831000000 +0100 @@ -9,10 +9,10 @@ use strict; "$Config{'archname'}-$Config{'osvers'}" eq -"x86_64-linux-thread-multi-3.17.8-300.bz1178975.fc21.x86_64" or - die "Errno architecture (x86_64-linux-thread-multi-3.17.8-300.bz1178975.fc21.x86_64) does not match executable architecture ($Config{'archname'}-$Config{'osvers'})"; +"x86_64-linux-thread-multi-3.19.0-0.rc7.git2.1.fc22.x86_64" or + die "Errno architecture (x86_64-linux-thread-multi-3.19.0-0.rc7.git2.1.fc22.x86_64) does not match executable architecture ($Config{'archname'}-$Config{'osvers'})"; -our $VERSION = "1.20_03"; +our $VERSION = "1.22"; $VERSION = eval $VERSION; our @ISA = 'Exporter'; @@ -20,140 +20,6 @@ BEGIN { %err = ( - EPERM => 1, - ENOENT => 2, - ESRCH => 3, [...] That's reportedly caused by a change in GCC 5.0's preprocesor as described in: <https://lists.fedoraproject.org/pipermail/devel/2015-February/207549.html> <https://gcc.gnu.org/bugzilla/show_bug.cgi?id=60723#c27> The preprocessed output contains a new line with current working directory directive which make troubles to the Errno_pm.PL parser. Simple work-around is to call the preprocessor with -P option (inhibit generation of linemarkers). Once I find a proper fix, I will post it here. -- PetrThread Previous | Thread Next