Not OK: perl v5.8.0 on PA-RISC1.1 11.00 #5526

p5pRT · 2002-06-03T09:34:20Z

Migrated from rt.perl.org#9456 (status was 'resolved')

Searchable as RT9456$

p5pRT · 2002-06-03T09:34:20Z

From philip.newton@datenrevision.de

Four failures:
[...]
t/comp/require.......................FAILED at test 10
[...]
ext/Time/HiRes/HiRes.................FAILED at test 20
[...]
lib/ExtUtils/t/basic.................Make: Unknown flag argument . Stop.
# Failed test (../lib/ExtUtils/t/basic.t at line 90)
FAILED at test 7
lib/ExtUtils/t/Command...............ok
lib/ExtUtils/t/Constant..............Make: Unknown flag argument . Stop.
FAILED at test 3
[...]

Plus a little bit of whinging:

[...]
lib/File/Temp/t/security.............uid=109 topuid=10 $<=110
path='/bfs/tmp/link/' at ../lib/File/Temp.pm line 659
File::Temp::_is_safe('/bfs/tmp/link/','SCALAR(0x40117f7c)') called
at ../lib/File/Temp.pm line 424

File::Temp::_gettemp('/bfs/tmp/link/tmpXXXXX','open',1,'mkdir',0,'unlink_on_
close',0,'suffixlen',0,...) called at ../lib/File/Temp.pm line 1057
File::Temp::tempfile('tmpXXXXX','DIR','/bfs/tmp/link','UNLINK',1)
called at ../lib/File/Temp/t/security.t line 100
eval {...} called at ../lib/File/Temp/t/security.t line 100
main::test_security(0) called at ../lib/File/Temp/t/security.t line
63
uid=109 topuid=10 $<=110 path='/bfs/tmp/link/' at ../lib/File/Temp.pm line
659
File::Temp::_is_safe('/bfs/tmp/link/','SCALAR(0x40118048)') called
at ../lib/File/Temp.pm line 719
File::Temp::_is_verysafe('/bfs/tmp/link/','SCALAR(0x40118048)')
called at ../lib/File/Temp.pm line 430

File::Temp::_gettemp('/bfs/tmp/link/tmpXXXXX','open',1,'mkdir',0,'unlink_on_
close',0,'suffixlen',0,...) called at ../lib/File/Temp.pm line 1057
File::Temp::tempfile('tmpXXXXX','DIR','/bfs/tmp/link','UNLINK',1)
called at ../lib/File/Temp/t/security.t line 100
eval {...} called at ../lib/File/Temp/t/security.t line 100
main::test_security(0) called at ../lib/File/Temp/t/security.t line
72
ok
[...]
lib/Term/Cap.........................tput: unknown terminal "foo"
tput: unknown terminal "foo"
tput: unknown terminal "foo"
ok
[...]
Failed 4 test scripts out of 653, 99.39% okay.

I didn't want to `./perl harness` since 'make test' already took three hours
(it's a slooow machine). However, here are the two non-ExtUtils failed
tests, run manually:

DAB@idndb@pne /tk/users/pne/tmp/install/perl-5.8.0-RC1/t > ./perl -Ilib
-I../lib comp/require.t
1..23
ok 1
ok 2
ok 3
ok 4
ok 5
ok 6
ok 7
ok 8
ok 9
# Perl v10.199.999 required--this is only v5.8.0, stopped at comp/require.t
line 77.

not ok 10
# Perl v10.0.19 required--this is only v5.8.0, stopped at comp/require.t
line 82.

not ok 11
ok 12
ok 13
ok 14
ok 15
ok 16
ok 17
ok 18
ok 19
ok 20
ok 21
ok 22
ok 23

Looks as if the machine fell afoul of the "check inaccurate fp" checks :)
The first should have been 10.2 (aka v10.200.0) but became v10.199.999, and
the second should have been 10.000_02 (aka v10.0.20) but became v10.0.19.

DAB@idndb@pne /tk/users/pne/tmp/install/perl-5.8.0-RC1/t > ./perl -Ilib
-I../lib ../ext/Time/HiRes/HiRes.t
1..25
ok 1
ok 2
ok 3
ok 4
ok 5
ok 6
ok 7
ok 8
ok 9
ok 10
ok 11
ok 12
ok 13
ok 14
# s = 25.7409472465515, n = 100, s/n = 0.257409472465515
# time...1023120813.27085
ok 15
# sleep...0.508374
ok 16
# Tick! 4 0.308931
# Select returned! 4 0.310855
# 0.896381666666667
# Tick! 3 0.618865
# Select returned! 3 0.620527
# 0.793157666666667
# Tick! 2 0.928888
# Select returned! 2 0.930549
# 0.689817
# Tick! 1 1.238943
# Select returned! 1 1.240637
# 0.586454333333333
# Tick! 0 1.548865
# Select returned! 0 1.550503
# 0.483165666666667
ok 17 # 0
# setitimer: 0 0
ok 18
# getitimer: 0 0
# getitimer: 0 0
ok 19
not ok 20 # 1.51989006996155 went by while sleeping 1.5, ratio
1.33333333333333.
ok 21 # 1.51839601993561 went by while sleeping 1.5, ratio 1.01194266666667.
ok 22
ok 23
ok 24
ok 25

This might just be a speed-of-machine problem? Or a maths problem? Or a
return-value-of-sleep() problem? After all, tests 20 and 21 had similar
elapsed time, but the ratio calculated is quite different. (Note that #20
tests sleep() and #21 tests usleep().)

I'm not going to try to reproduce the ExtUtils ones.

Cheers,
Philip

Perl Info


Flags:
    category=install
    severity=none

Site configuration information for perl v5.8.0:

Configured by pne at Mon Jun  3 14:25:53 MESZ 2002.

Summary of my perl5 (revision 5.0 version 8 subversion 0) configuration:
  Platform:
    osname=hpux, osvers=11.00, archname=PA-RISC1.1
    uname='hp-ux idndb b.11.00 a 9000715 2004727219 two-user license '
    config_args='-Duselargefiles'
    hint=recommended, useposix=true, d_sigaction=define
    usethreads=undef use5005threads=undef useithreads=undef
usemultiplicity=undef
    useperlio=define d_sfio=undef uselargefiles=define usesocks=undef
    use64bitint=undef use64bitall=undef uselongdouble=undef
    usemymalloc=n, bincompat5005=undef
  Compiler:
    cc='cc', ccflags =' -Ae -D_HPUX_SOURCE -Wl,+vnocompatwarnings
-I/usr/local/include -D_LARGEFILE_SOURCE -D_FILE_OFFSET_BITS=64 ',
    optimize='',
    cppflags='-Aa -D__STDC_EXT__ -D_HPUX_SOURCE -Ae -D_HPUX_SOURCE
-Wl,+vnocompatwarnings -I/usr/local/include'
    ccversion='B.11.11.04', gccversion='', gccosandvers=''
    intsize=4, longsize=4, ptrsize=4, doublesize=8, byteorder=4321
    d_longlong=define, longlongsize=8, d_longdbl=define, longdblsize=16
    ivtype='long', ivsize=4, nvtype='double', nvsize=8, Off_t='off_t',
lseeksize=8
    alignbytes=8, prototype=define
  Linker and Libraries:
    ld='/usr/bin/ld', ldflags =' -L/usr/local/lib'
    libpth=/usr/local/lib /lib /usr/lib /usr/ccs/lib
    libs=-lnsl -lnm -lndbm -lmalloc -ldld -lm -lc -lndir -lcrypt -lsec
    perllibs=-lnsl -lnm -lmalloc -ldld -lm -lc -lndir -lcrypt -lsec
    libc=/lib/libc.sl, so=sl, useshrplib=false, libperl=libperl.a
  Dynamic Linking:
    dlsrc=dl_hpux.xs, dlext=sl, d_dlsymun=undef, ccdlflags='-Wl,-E
-Wl,-B,deferred '
    cccdlflags='+Z', lddlflags='-b +vnocompatwarnings -L/usr/local/lib'

Locally applied patches:
    


@INC for perl v5.8.0:
    lib
    /opt/perl5/lib/5.8.0/PA-RISC1.1
    /opt/perl5/lib/5.8.0
    /opt/perl5/lib/site_perl/5.8.0/PA-RISC1.1
    /opt/perl5/lib/site_perl/5.8.0
    /opt/perl5/lib/site_perl
    .


Environment for perl v5.8.0:
    HOME=/tk/users/pne
    LANG=german
    LANGUAGE (unset)
    LC_ALL=de_DE.roman8
    LC_COLLATE=de_DE.roman8
    LC_CTYPE=de_DE.roman8
    LC_MESSAGES=de_DE.roman8
    LC_MONETARY=de_DE.roman8
    LC_NUMERIC=de_DE.roman8
    LC_TIME=de_DE.roman8
    LD_LIBRARY_PATH=/tk/users/pne/tmp/install/perl-5.8.0-RC1
    LOGDIR (unset)
 
PATH=/usr/bin:/opt/ansic/bin:/usr/ccs/bin:/usr/contrib/bin:/opt/nettladm/bin
:/opt/fc/bin:/opt/fcms/bin:/opt/upgrade/bin:/opt/pd/bin:/usr/bin/X11:/usr/co
ntrib/bin/X11:/opt/hparray/bin:/opt/ignite/bin:/opt/langtools/bin:/opt/imake
/bin:.:.:/tk/tk_skripte:/tk/users/pne/skripte:/bfs/tools:/tk/dab/scripte:/tk
/jam5.04g/util:/opt/vim/bin:/usr/local/bin
    PERL_BADLANG (unset)
    SHELL=/usr/bin/ksh


-- 
Philip Newton                  Philip.Newton@datenrevision.de
datenrevision GmbH & Co. OHG      http://www.datenrevision.de
a gedas company                         TEL +49-40-797 007-37
Cuxhavener Str. 36, D-21149 Hamburg     FAX +49-40-797 007-10

p5pRT · 2002-06-03T13:36:04Z

From @rgarcia

On 2002.06.03 18:34 "Newton, Philip" wrote:

DAB@idndb@pne /tk/users/pne/tmp/install/perl-5.8.0-RC1/t > ./perl -Ilib
-I../lib comp/require.t
1..23
[...]
# Perl v10.199.999 required--this is only v5.8.0, stopped at comp/require.t
line 77.

not ok 10
# Perl v10.0.19 required--this is only v5.8.0, stopped at comp/require.t
line 82.

not ok 11
[...]

Looks as if the machine fell afoul of the "check inaccurate fp" checks :)
The first should have been 10.2 (aka v10.200.0) but became v10.199.999, and
the second should have been 10.000_02 (aka v10.0.20) but became v10.0.19.

Yes, maths problem. IIRC that's why v-strings were introduced.
Not really surprising when you look at the code (in pp_ctl.c) :

else if (!SvPOKp(sv)) { /* require 5.005_03 */
if ((NV)PERL_REVISION + ((NV)PERL_VERSION/(NV)1000)
+ ((NV)PERL_SUBVERSION/(NV)1000000)
+ 0.00000099 < SvNV(sv))
[etc.]

Should the test be made more lenient ?

p5pRT · 2002-06-03T16:52:55Z

From @schwern

On Mon, Jun 03, 2002 at 10:38:17PM +0200, Rafael Garcia-Suarez wrote:

On 2002.06.03 18:34 "Newton, Philip" wrote:

DAB@idndb@pne /tk/users/pne/tmp/install/perl-5.8.0-RC1/t > ./perl -Ilib
-I../lib comp/require.t
1..23
[...]
# Perl v10.199.999 required--this is only v5.8.0, stopped at comp/require.t
line 77.

not ok 10
# Perl v10.0.19 required--this is only v5.8.0, stopped at comp/require.t
line 82.

not ok 11
[...]

Looks as if the machine fell afoul of the "check inaccurate fp" checks :)
The first should have been 10.2 (aka v10.200.0) but became v10.199.999, and
the second should have been 10.000_02 (aka v10.0.20) but became v10.0.19.

Yes, maths problem. IIRC that's why v-strings were introduced.
Not really surprising when you look at the code (in pp_ctl.c) :
else if $\!SvPOKp\(sv$\) \{            /\* require 5\.005\_03 \*/
    if $\(NV$PERL\_REVISION \+ $\(NV$PERL\_VERSION/$NV$1000\)
    \+ $\(NV$PERL\_SUBVERSION/$NV$1000000\)
    \+ 0\.00000099 \< SvNV$sv$\)
\[etc\.\]
Should the test be made more lenient ?

10.2 is surely a small enough float that it shouldn't fall afoul of floating
point error?

--
This sig file temporarily out of order.

p5pRT · 2002-06-03T18:18:25Z

From @JohnPeacock

Michael G Schwern wrote:

10.2 is surely a small enough float that it shouldn't fall afoul of floating
point error?

Repeat after me: SIZE DOESN'T MATTER... ;~)

Certain decimals do not translate cleanly into binary notation; when they are
entered as literals, they will appear fine, but when they are reached as a
result of a mathematical operation, they can lose accuracy. This is one of
those instances. I'm sorry that my version object patch was not able to be
integrated into 5.8.0, since that would have dealt with it cleanly.

John

p5pRT · 2002-06-04T03:22:08Z

From @Tux

On Mon 03 Jun 2002 18:34, "Newton, Philip" <Philip.Newton@datenrevision.de> wrote:

This is a build failure report for perl from Philip.Newton@datenrevision.de,
generated with the help of perlbug 1.33 running under perl v5.8.0.

How up-to-date is /this/ system? Is it Y2K compliant?

Summary of my perl5 (revision 5.0 version 8 subversion 0) configuration:
Platform:
osname=hpux, osvers=11.00, archname=PA-RISC1.1
uname='hp-ux idndb b.11.00 a 9000715 2004727219 two-user license '
config_args='-Duselargefiles'
hint=recommended, useposix=true, d_sigaction=define

--
H.Merijn Brand Amsterdam Perl Mongers (http://amsterdam.pm.org/)
using perl-5.6.1, 5.8.0 & 631 on HP-UX 10.20 & 11.00, AIX 4.2, AIX 4.3,
WinNT 4, Win2K pro & WinCE 2.11. Smoking perl CORE: smokers@perl.org
http://archives.develooper.com/daily-build@perl.org/ perl-qa@perl.org
send smoke reports to: smokers-reports@perl.org, QA: http://qa.perl.org

p5pRT · 2003-07-07T00:05:12Z

From @JohnPeacock

This was 5.8.0-RC1 on a very old machine.

p5pRT · 2003-07-07T00:05:12Z

@JohnPeacock - Status changed from 'open' to 'resolved'

p5pRT closed this as completed Jul 7, 2003

p5pRT added the affects-5.8 label Oct 18, 2019

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

Not OK: perl v5.8.0 on PA-RISC1.1 11.00 #5526

Not OK: perl v5.8.0 on PA-RISC1.1 11.00 #5526

p5pRT commented Jun 3, 2002

p5pRT commented Jun 3, 2002

p5pRT commented Jun 3, 2002

p5pRT commented Jun 3, 2002

p5pRT commented Jun 3, 2002

p5pRT commented Jun 4, 2002

p5pRT commented Jul 7, 2003

p5pRT commented Jul 7, 2003

Not OK: perl v5.8.0 on PA-RISC1.1 11.00 #5526

Not OK: perl v5.8.0 on PA-RISC1.1 11.00 #5526

Comments

p5pRT commented Jun 3, 2002

p5pRT commented Jun 3, 2002

From philip.newton@datenrevision.de

p5pRT commented Jun 3, 2002

From @rgarcia

p5pRT commented Jun 3, 2002

From @schwern

p5pRT commented Jun 3, 2002

From @JohnPeacock

p5pRT commented Jun 4, 2002

From @Tux

p5pRT commented Jul 7, 2003

From @JohnPeacock

p5pRT commented Jul 7, 2003