develooper Front page | perl.perl5.porters | Postings from June 2019

[perl #134182] Fix test failure in POSIX/t/mb.t with semi-brokenlocales

Thread Previous | Thread Next
From:
James E Keenan via RT
Date:
June 8, 2019 22:32
Subject:
[perl #134182] Fix test failure in POSIX/t/mb.t with semi-brokenlocales
Message ID:
rt-4.0.24-5721-1560033110-1120.134182-15-0@perl.org
On Sat, 08 Jun 2019 21:43:16 GMT, jkeenan wrote:
> Dom,
> 
> Unfortunately I have to call your attention to 2 smoke-test failures
> in ext/POSIX/t/mb.t which were recorded *after* I applied your patch.
> 
> http://perl5.test-smoke.org/report/89146
> logs at: http://perl5.test-smoke.org/logfile/89146
> 
> http://perl5.test-smoke.org/report/89211
> logs at: http://perl5.test-smoke.org/logfile/89211
> 
> (These can be tracked via this search:
> http://perl5.test-
> smoke.org/submatrix?test=../ext/POSIX/t/mb.t&pversion=5.31.1)
> 
> In each case the failures in mb.t occurred when blead was configured
> as follows:
> 
> [stdio] -Dcc=clang -Accflags="-Werror=declaration-after-statement -g
> -fno-omit-frame-pointer -fsanitize=address  -fno-common -fsanitize-
> blacklist=`pwd`/asan_ignore" -Aldflags="-fsanitize=address"
> 
> With and without -DDEBUGGING.
> 
> A couple of points:
> 
> 1. Tester is using what I would guess is an advanced version of the
> Linux kernel: 5.0.9-200 versus my own 4.15.0-51 (Ubuntu 18.04 LTS).
> OTOH, we are getting smoke-test reports from rigs with even higher-
> numbered Linux kernels.
> 
> 2. I myself don't understand all those compiler switches the tester is
> using.  In particular, 'make' fails for me on FreeBSD-11.2 when I use
> those compiler switches.
> 
> 3. Nonetheless, when I build a perl with all those switches (except
> -DDEBUGGING), I get those test same failures.  See attachments.
> 
> 4. When I build blead with those same compiler switches at the commit
> immediately prior to the one where I applied your patch, I get a PASS.
> 
> #####
> $ git show | head -1
> commit fb55ce6b7596b9e94f941cf83eac5ff84f760ea2
> $ cd t;./perl harness -v ../ext/POSIX/t/mb.t; cd -
> 
> ok 1 - mblen() basically works
> ok 2 - MB_CUR_MAX is at least 4 in a UTF-8 locale
> ok 3 - mblen() recognizes invalid multibyte characters
> ok 4 - mblen() works on UTF-8 characters
> ok
> All tests successful.
> Files=1, Tests=4,  0 wallclock secs ( 0.02 usr  0.00 sys +  0.26 cusr
> 0.12 csys =  0.40 CPU)
> Result: PASS
> #####
> 
> So your patch has triggered test failures, albeit under these very
> obscure conditions.  I'm going to revert your patch from blead and
> then re-apply it in a branch so that we can continue to gather smoke-
> test reports.
> 
> Thank you very much.

The smoke-test branch is:

smoke-me/jkeenan/dom/134182-mb

-- 
James E Keenan (jkeenan@cpan.org)

---
via perlbug:  queue: perl5 status: open
https://rt.perl.org/Ticket/Display.html?id=134182

Thread Previous | Thread Next


nntp.perl.org: Perl Programming lists via nntp and http.
Comments to Ask Bjørn Hansen at ask@perl.org | Group listing | About