Bug #130890 for Crypt-MagicSignatures-Key: perl-5.31.5: test failure

Sun Nov 03 17:09:08 2019 jkeenan [...] cpan.org - Ticket created

Subject:

perl-5.31.5: test failure

As reported in the Perl 5 core distribution's issue tracker: https://github.com/Perl/perl5/issues/17247 ##### Description The test suite of Crypt-MagicSignatures-Key-0.21 fails with perl 5.31.5. 5.31.4 and older passes. Overview of test reports: http://fast-matrix.cpantesters.org/?dist=Crypt-MagicSignatures-Key-0.21 ##### This may be Perl 5 blead's responsibility -- but it also may be that an improvement in the Perl 5 core distribution has exposed some sub-optimal code in this distribution. Not yet further analyzed. Thank you very much. Jim Keenan

Sun Nov 03 17:09:20 2019 jkeenan [...] cpan.org - Reference to https://github.com/Perl/perl5/issues/17247 added

Mon Nov 04 21:41:41 2019 ANDK [...] cpan.org - Correspondence added

RT-Send-CC:

tony [...] develop-help.com

Bisect reveals: : There are only 'skip'ped commits left to test. : The first bad commit could be any of: : e31f3d3767e7d53f56f4bf96ac71a6e8dcec3e95 : e8864dba80952684bf3afe83438d4eee0c3939a9 : 06e6ef8e32e28d977557b12e340a09bf453c69c0 : ee9ac1cd8eb988fea70841eae211b11355711416 : 686a07f48a98cac22ae2aac42df706b176496784 : We cannot bisect more! : bisect run cannot continue any more That's the streak of the Digest-MD5 rollout. Should probably be reported to p5p, trying to CC Tony Cook

Mon Nov 04 21:41:42 2019 The RT System itself - Status changed from 'new' to 'open'

Sat Dec 07 00:24:26 2019 ANDK [...] cpan.org - Correspondence added

I retried the bisect with a different smoker and without attemtping to generate a CPAN::Reporter for each perl and this time I found a single commit: e8864dba80952684bf3afe83438d4eee0c3939a9 is the first bad commit commit e8864dba80952684bf3afe83438d4eee0c3939a9 Author: Matt Turner <mattst88@gmail.com> Date: Wed Sep 4 21:48:56 2019 -0700 Clean up U8TO*_LE macro implementations The code guarded by #ifndef U32_ALIGNMENT_REQUIRED attempts to optimize byte-swapping by doing unaligned loads, but accessing data through unaligned pointers is undefined behavior in C. Moreover, compilers are more than capable of recognizing these open-coded byte-swap patterns and emitting a bswap instruction, or an unaligned load instruction, or a combined load, etc. There's no need for multiple paths to attain the desired result. See https://rt.perl.org/Ticket/Display.html?id=133495 hv_macro.h | 31 ++++++++++++------------------- stadtx_hash.h | 52 ---------------------------------------------------- zaphod32_hash.h | 35 ----------------------------------- 3 files changed, 12 insertions(+), 106 deletions(-)

Sat Dec 07 03:35:09 2019 ANDK [...] cpan.org - Correspondence added

Xref: https://github.com/Perl/perl5/issues/17247

Sat Dec 07 09:51:55 2019 ANDK [...] cpan.org - Correspondence added

Apparently resolved by bleadperl commit v5.31.5-172-g783bd734e8

Sat Dec 07 09:51:56 2019 ANDK [...] cpan.org - Status changed from 'open' to 'resolved'