On 03/28/2018 09:47 PM, Karl Williamson wrote:
On 03/28/2018 05:44 AM, Yaroslav Kuzmin wrote:
Output :
(16:42) RS12 : PDKUZM | ~/ussport/perl/perl-build
:> ./perl -V
Summary of my perl5 (revision 5 version 27 subversion 11) configuration:
Snapshot of: 3a7f897ad3c330a4e99ba46255bb72e89c43cd08
Platform:
osname=os390
osvers=25.00
archname=os390
uname='os390 rs12 25.00 04 2964 '
config_args='-Dusedevel -Dusedl -d -s'
hint=recommended
useposix=true
d_sigaction=define
useithreads=undef
usemultiplicity=undef
use64bitint=undef
use64bitall=undef
uselongdouble=undef
usemymalloc=n
default_inc_excludes_dot=define
bincompat5005=undef
Compiler:
cc='c99'
ccflags ='-qlanglvl=extended:extc89:extc99 -qlongname -qxplink
-qdll -qfloat=ieee -qhaltonmsg=3296:4108 -DMAXSIG=39 -DOEMVS
-D_OE_SOCKETS -D_XOPEN_SOURCE_EXTENDED -D_ALL_SOURCE -DYYDYNAMIC
-D_POSIX_SOURCE=1 -D_SHR_ENVIRON -DPERL_EXTERNAL_GLOB -qexportall
-qdll -qxplink'
optimize=' '
cppflags='-DMAXSIG=39 -DOEMVS -D_OE_SOCKETS
-D_XOPEN_SOURCE_EXTENDED -D_ALL_SOURCE -DYYDYNAMIC -D_POSIX_SOURCE=1
-D_SHR_ENVIRON -DPERL_EXTERNAL_GLOB'
ccversion=''
gccversion=''
gccosandvers=''
intsize=4
longsize=4
ptrsize=4
doublesize=8
byteorder=4321
doublekind=4
d_longlong=define
longlongsize=8
d_longdbl=define
longdblsize=16
longdblkind=2
ivtype='long'
ivsize=4
nvtype='double'
nvsize=8
Off_t='off_t'
lseeksize=4
alignbytes=8
prototype=define
Linker and Libraries:
ld='c99'
ldflags ='-qxplink -qdll'
libpth=/lib /usr/lib
libs=-lm -lc
perllibs=-lm -lc
libc=
so=so
useshrplib=true
libperl=libperl.so
gnulibc_version=''
Dynamic Linking:
dlsrc=dl_dllload.xs
dlext=so
d_dlsymun=undef
ccdlflags='-qxplink -qdll
/u/pdkuzm/ussport/perl/perl-build/libperl.x'
cccdlflags='-c -qexportall -qxplink -qdll'
lddlflags='-qxplink -qdll
/u/pdkuzm/ussport/perl/perl-build/libperl.x'
Characteristics of this binary (from libperl):
Compile-time options:
HAS_TIMES
PERL_COPY_ON_WRITE
PERL_DONT_CREATE_GVSV
PERL_EXTERNAL_GLOB
PERL_MALLOC_WRAP
PERL_OP_PARENT
PERL_PRESERVE_IVUV
PERL_USE_DEVEL
PERLIO_LAYERS
USE_LARGE_FILES
USE_LOCALE
USE_LOCALE_COLLATE
USE_LOCALE_CTYPE
USE_LOCALE_NUMERIC
USE_LOCALE_TIME
USE_PERL_ATOF
USE_PERLIO
Locally applied patches:
SMOKE3a7f897ad3c330a4e99ba46255bb72e89c43cd08
Built under os390
Compiled at Mar 28 2018 12:12:09
%ENV:
PERL5LIB="/u/pdkuzm/ussport/perl/perl-build/lib"
@INC:
/u/pdkuzm/ussport/perl/perl-build/lib
/usr/local/lib/perl5/site_perl/5.27.11/os390
/usr/local/lib/perl5/site_perl/5.27.11
/usr/local/lib/perl5/5.27.11/os390
/usr/local/lib/perl5/5.27.11
(16:43) RS12 : PDKUZM | ~/ussport/perl/perl-build
:> ./perl -le 'my $a = "\x{587}"; my $b = uc($a); use bytes; print
map { ^Jsprintf " %x", ord } split "", $a; print map { sprintf " %x",
ord } ^Jsplit "", $b'
b8 53 48
b8 53 48
The first sequence is the correct UTF-EBCDIC for \x{587}, but it has an
uppercase value that is different.
I think the first thing is for you to package up the directory
lib/unicore and send a copy to me.
If you have the machine available, you might want to try testing
https://perl5.git.perl.org/perl.git/shortlog/refs/heads/smoke-me/khw-core
which has a somewhat different casing implementation
For example, a function that is segfaulting in your report is completely
removed.
I also looked through the .25 million lines of code that got changed
between successful smokes. The likely perpetrators are changes to
multiconcat and/or regular expression pattern matching.
В Вс., 25/03/2018 в 16:05 -0600, Karl Williamson пишет:
On 03/21/2018 07:09 AM, Yaroslav Kuzmin wrote:
Smoke log at available
https://na01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fdrive.google.com%2Ffile%2Fd%2F1XHUAwq74zUMfS3Ps3WnMAvpSagcXFWxJ&data=02%7C01%7Cykuzmin%40rocketsoftware.com%
7C797e757095254fbcdbc408d5929c90fa%7C79544c1eed224879a082b67a9a672aae%7C0%7C1%7C636576123523851610&sdata=n0upDMJ9%2B5AW5XqESiqK2lWUP3FXk%2BcsVa0exwoztSE%3D&reserved=0
Automated smoke report for branch blead 5.27.10 patch
8d8484540d380c09fb40616135be00825189f578 v5.27.8-546-g8d8484540d
RS12: 2964 (2964/)
on os/390 - 25.00
using c99 version
smoketime 2 hours 53 minutes (average 1 hour 26 minutes)
Summary: FAIL(XM)
The many failures in this report indicates that something fundamental
got broken. So to start finding it, I think we need to go back to the
basics.
Try this one-liner, and send me the results:
./perl -le 'my $a = "\x{587}"; my $b = uc($a); use bytes; print map {
sprintf " %x", ord } split "", $a; print map { sprintf " %x", ord }
split "", $b'
================================
Rocket Software, Inc. and subsidiaries ■ 77 Fourth Avenue, Waltham MA
02451 ■ Main Office Toll Free Number: +1 855.577.4323
Contact Customer Support:
https://my.rocketsoftware.com/RocketCommunity/RCEmailSupport
Unsubscribe from Marketing Messages/Manage Your Subscription
Preferences - http://www.rocketsoftware.com/manage-your-email-preferences
Privacy Policy -
http://www.rocketsoftware.com/company/legal/privacy-policy
================================
This communication and any attachments may contain confidential
information of Rocket Software, Inc. All unauthorized use, disclosure
or distribution is prohibited. If you are not the intended recipient,
please notify Rocket Software immediately and destroy all copies of
this communication. Thank you.