[Bug 828320] New: perl 5.18 is broken, that tests with UTF-8 strings fails randomly
https://bugzilla.novell.com/show_bug.cgi?id=828320 https://bugzilla.novell.com/show_bug.cgi?id=828320#c0 Summary: perl 5.18 is broken, that tests with UTF-8 strings fails randomly Classification: openSUSE Product: openSUSE Factory Version: 13.1 Milestone 2 Platform: x86-64 OS/Version: openSUSE 12.3 Status: NEW Keywords: dogfood Severity: Critical Priority: P5 - None Component: Basesystem AssignedTo: mls@suse.com ReportedBy: werner@suse.com QAContact: qa-bugs@suse.de CC: coolo@suse.com Found By: Development Blocker: --- For a history see https://github.com/plk/biber/issues/6 as this is my bug report for biblatex-biber. What bothers is that the errors occur randomly which may idicate a bug in perl 5.18.0 as otherwise an error should be well repeatable. You may report/forward this to upstream perl -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=828320 https://bugzilla.novell.com/show_bug.cgi?id=828320#c1 --- Comment #1 from Stephan Kulow <coolo@suse.com> 2013-07-05 14:46:00 CEST --- See http://search.cpan.org/dist/perl-5.18.0/pod/perldelta.pod#Hash_randomization and perl.changes - update to 5.18.0 .. * better hash randomization .. I'd say INVALID -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=828320 https://bugzilla.novell.com/show_bug.cgi?id=828320#c2 --- Comment #2 from Dr. Werner Fink <werner@suse.com> 2013-07-05 13:21:56 UTC --- Using PERL_HASH_SEED=0x0 PERL_HASH_SEED_DEBUG=1 does *not* help! IMHO this has nothing to do with hash randomization. AFAIK the internal strings of perl 5.18 are not hashes and random messages like Wide character in print at /usr/lib/perl5/5.18.0/Test/Builder.pm line 1759. on one large UTF-8 string has nothing to with hashes. The test uses the perl core module Encode(3pm) with decode_utf8() -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=828320 https://bugzilla.novell.com/show_bug.cgi?id=828320#c3 --- Comment #3 from Michael Schröder <mls@suse.com> 2013-07-05 14:50:45 UTC --- "Wide character in print" just means that the output stream is not utf8 safe. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=828320 https://bugzilla.novell.com/show_bug.cgi?id=828320#c4 --- Comment #4 from Dr. Werner Fink <werner@suse.com> 2013-07-09 08:35:04 UTC ---
From upstream developer of biblatex-biber:
| bitstreamout commented |> On Mon, Jul 08, 2013 at 11:35:53PM -0700, plk wrote: |> I have just run the test | suite under Windows Strawberry perl 5.18 and it's |> all fine for me ... | | And your're sure that the perl 5.18 on Windows[tm] has enabled randomization | of the hash table implementation. Do have tried my test perl script? If there | are no random results the major feature of perl 5.18 is not enabled. | plk commented | Yes, I get random results on the perl 5.18 with which I ran the tests. I don't | think this is the problem - I don't rely on the order of hashes in biber - no | perl programmer who has a clue has ever done that anyway - this is why it's | possible for perl 5.18 to make this change at all ... In other words our perl installation seems to broken ... maybe one or more of the required perl modules below devel:languages:perl and devel:languages:perl:CPAN-* do not work with perl 5.18 Beside this the upstream developer of perl(Data::Diver) has written | I'll try to release a new dist this weekend. | | Tye on tex-live@tug.org on a mail with subject: [tex-live] TL13: biblatex-biber using perl(Data::Diver) without license information -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=828320 https://bugzilla.novell.com/show_bug.cgi?id=828320#c5 --- Comment #5 from Dr. Werner Fink <werner@suse.com> 2013-07-10 14:11:38 UTC --- To be noted: I've run several test with a self compiled perl 5.18 ... one with diabled hash seed feature by adding both -DUSE_HASH_SEED_EXPLICIT and -DNO_HASH_SEED to configure line to both the normal configure options as well as to the -Accflags= option. After changing make test to make test || : I see [ 250s] t/run/noswitch ................................................ ok [ 250s] t/run/runenv .................................................. # Failed test 46 - PERL_HASH_SEED => 0,PERL_HASH_SEED_DEBUG => 1,-e,1 stderr at run/runenv.t line 218 [ 250s] # Failed test 48 - PERL_HASH_SEED => 0x0,PERL_HASH_SEED_DEBUG => 1,-e,1 stderr at run/runenv.t line 225 [ 250s] FAILED at test 46 [ 250s] t/run/script .................................................. ok which is expected and get in the `perl -V' the output Characteristics of this binary (from libperl): Compile-time options: HAS_TIMES MULTIPLICITY NO_HASH_SEED PERLIO_LAYERS PERL_DONT_CREATE_GVSV PERL_HASH_FUNC_ONE_AT_A_TIME_HARD PERL_IMPLICIT_CONTEXT PERL_MALLOC_WRAP PERL_PRESERVE_IVUV PERL_SAWAMPERSAND PERL_USE_SAFE_PUTENV USE_64_BIT_ALL USE_64_BIT_INT USE_HASH_SEED_EXPLICIT USE_ITHREADS USE_LARGE_FILES USE_LOCALE USE_LOCALE_COLLATE USE_LOCALE_CTYPE USE_LOCALE_NUMERIC USE_PERLIO USE_PERL_ATOF USE_REENTRANT_API .. beside this I've switched back to gcc 4.7 by using %build ln -sf /usr/bin/gcc-4.7 gcc ln -sf /usr/bin/gcc-4.7 cc ln -sf /usr/bin/g++-4.7 g++ ln -sf /usr/bin/g++-4.7 c++ ln -sf /usr/bin/cpp-4.7 cpp PATH=$PWD:$PATH gcc --version to be sure not to run into a compiler issue. Nevertheless the problem remains that is the results of biblatex-biber are random. IMHO the perl 5.18 this could be a race condition on Linux ... as upstream of biblatex-biber has run all tests below perl 5.18 but on Windows Strawberry (see https://github.com/plk/biber/issues/6#issuecomment-20655986) -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=828320 https://bugzilla.novell.com/show_bug.cgi?id=828320#c6 Ismail Donmez <idonmez@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO CC| |idonmez@suse.com InfoProvider| |werner@suse.com --- Comment #6 from Ismail Donmez <idonmez@suse.com> 2013-09-27 16:09:09 UTC --- This is now fixed upstream it seems, please confirm & close this bug. -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=828320 https://bugzilla.novell.com/show_bug.cgi?id=828320#c7 Dr. Werner Fink <werner@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |NEW InfoProvider|werner@suse.com | --- Comment #7 from Dr. Werner Fink <werner@suse.com> 2013-09-27 16:22:02 UTC --- Correct -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
https://bugzilla.novell.com/show_bug.cgi?id=828320 https://bugzilla.novell.com/show_bug.cgi?id=828320#c8 Dr. Werner Fink <werner@suse.com> changed: What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |RESOLVED Resolution| |FIXED --- Comment #8 from Dr. Werner Fink <werner@suse.com> 2013-09-27 16:22:23 UTC --- done -- Configure bugmail: https://bugzilla.novell.com/userprefs.cgi?tab=email ------- You are receiving this mail because: ------- You are on the CC list for the bug.
participants (1)
-
bugzilla_noreply＠novell.com