Re^2: Advice wanted for debugging CPAN Testers failures

Replies are listed 'Best First'.
Re^3: Advice wanted for debugging CPAN Testers failures by syphilis (Archbishop) on Aug 24, 2016 at 02:59 UTC
apparently the same system, with two different perl versions but same gcc version and same libc version, got different results Are you sure ? It's no big deal but I'd be interested to see 2 reports indicative of that. Cheers, Rob	[reply]
Re^4: Advice wanted for debugging CPAN Testers failures by stevieb (Canon) on Aug 24, 2016 at 12:54 UTC
For me, on the same system, it passes on 5.22+, but fails on everything below that. I'm wondering if it may be related to Infinity and NaN (not-a-number) handling improved changes mentioned in perldelta-5.22.	[reply]
Re^4: Advice wanted for debugging CPAN Testers failures by pryrt (Abbot) on Aug 24, 2016 at 13:59 UTC
I'd be interested to see 2 reports indicative of that. The two reports linked in my original post: both showed `linux 3.16.0-4-amd64 debian 3.16.7-ckt25-2 (2016-04-08)` with `gccversion='4.9.2'`, `libc=libc-2.19.so`, and `gnulibc_version='2.19'`, the only difference being 5.20.1 vs 5.22.0. But as stevieb replied, it's probably the improved NaN handling in 5.22 which fixed the bug. (Which then surprises me that my ancient linux perl v5.8.5 passed... but then again, there were probably other NaN handling changes in Perl or differences between the systems/compilers). excerpts from the fail reports in the spoiler <Reveal this spoiler or all in this thread>	[reply] [d/l] [select]
Re^5: Advice wanted for debugging CPAN Testers failures by syphilis (Archbishop) on Aug 24, 2016 at 23:26 UTC
Interesting. I didn't realise that perl had got actively involved in standardising the signedness of NaNs. Thanks for the heads up stevieb, pryrt. Cheers, Rob	[reply]