develooper Front page | perl.perl5.porters | Postings from March 2003

Re: [req #800]Perl 5.8.1 closing in

From:
Len Makin
Date:
March 13, 2003 02:08
Subject:
Re: [req #800]Perl 5.8.1 closing in
Message ID:
200303130519.QAA27158@trout.hpc.CSIRO.AU
Hi Jarkko,
	Thought you'd retired from perl porting. ;-)
On Monday, 10 March 2003, at 0:34,
Jarkko Hietaniemi <jhi@iki.fi> wrote:
> Hi, I want to consider Perl 5.8.1 soon-ish.  Here's a Perl snapshot
...
> If you could consider trying it out soon on SUPER-UX, I would be grateful.
OK, tested perl@18872 on super-ux (and unicos):
Testing is as follows. I run "gmake test-notty" (or "cd t;./perl harness")
as a batch job, since there are memory restrictions on interactive sessions
and if I use gmake test interactively I get many "not enough space" errors.
Some interactive tests may be skipped or fail when tested in batch.
I then run failed tests interactively using
	export PERL_CORE=1; cd t; ./perl -I../lib misc/failedtest
If there are any important tests which are missed by this procedure,
let me know and I can request a temporary increase in interactive limits
to get it done. As previously, results below are (almost)all from failed tests.
All test failures same as for perl 5.8.0 ;-(

1. Super-UX 10.2:

Failed 12/763 test scripts, 98.43% okay. 5864/68451 subtests failed, 91.43% okay
.
Failed Test                   Stat Wstat Total Fail  Failed  List of Failed
-------------------------------------------------------------------------------
../ext/IO/lib/IO/t/io_poll.t                10    2  20.00%  3-4
../ext/IPC/SysV/ipcsysv.t                   16    3  18.75%  2 5-6
../ext/IPC/SysV/t/msg.t                      9    4  44.44%  2 4-6
../ext/Socket/socketpair.t       0     2    45   35  77.78%  11-45
../lib/IPC/SysV.t                           16    3  18.75%  2 5-6
../lib/Net/Ping/t/190_alarm.t  255 65280     6    1  16.67%  6
../lib/warnings.t                          485    4   0.82%  115-116 118-119
op/64bitint.t                               67    6   8.96%  29-30 32-33 35-36
op/arith.t                                 134    1   0.75%  128
op/pack.t                      255 65280  5826 5802  99.59%  25-5826
op/pow.t                         2   512    ??   ??       %  ??
op/stat.t                                   73    3   4.11%  41-42 44
56 tests and 227 subtests skipped.

../ext/IO/lib/IO/t/io_poll.t (ignore)
OK interactively

../ext/IPC/SysV/ipcsysv.t
not ok 2
#
# The failure of the subtest #2 may indicate that the message queue
# resource limits either of the system or of the testing account
# have been reached.  Error message "Operating would block" is
# usually indicative of this situation.  The error message was now:
# "Invalid argument"
#
# You can check the message queues with the 'ipcs' command and
# you can remove unneeded queues with the 'ipcrm -q id' command.
# You may also consider configuring your system or account
# to have more message queue resources.
#
# Because of the subtest #2 failing also the substests #5 and #6 will
# very probably also fail.
#
not ok 5
#
# This failure was to be expected because the subtest #2 failed.
#
not ok 6
#
# This failure was to be expected because the subtest #2 failed.
#

../ext/IPC/SysV/t/msg.t (not very informative)
1..9
ok 1
not ok 2
ok 3
not ok 4
not ok 5
not ok 6
ok 7
ok 8
ok 9

../ext/Socket/socketpair.t (same as Cray)
ok 10 - shutdown left for writing
EOF on right took over 3 seconds at ../ext/Socket/socketpair.t line 118.
Something unexpectedly hung during testing at ../ext/Socket/socketpair.t line 39.

../lib/IPC/SysV.t
not ok 2
#
# The failure of the subtest #2 may indicate that the message queue
# resource limits either of the system or of the testing account
# have been reached.  Error message "Operating would block" is
# usually indicative of this situation.  The error message was now:
# "Invalid argument"
#
# You can check the message queues with the 'ipcs' command and
# you can remove unneeded queues with the 'ipcrm -q id' command.
# You may also consider configuring your system or account
# to have more message queue resources.
#
# Because of the subtest #2 failing also the substests #5 and #6 will
# very probably also fail.
#
not ok 5
#
# This failure was to be expected because the subtest #2 failed.
#
not ok 6
#
# This failure was to be expected because the subtest #2 failed.
#

../lib/Net/Ping/t/190_alarm.t
1..6
# Running under perl version 5.008001 for super-ux
# Current time local: Thu Mar 13 11:42:54 2003
# Current time GMT:   Thu Mar 13 00:42:54 2003
# Using Test.pm version 1.23
not ok 6
# Failed test 6 in ../lib/Net/Ping/t/190_alarm.t at line 48
#  ../lib/Net/Ping/t/190_alarm.t line 48 is: ok $@ =~ /alarm works/ or die $@;
Can't get tcp echo port by name at ../lib/Net/Ping/t/190_alarm.t line 37

../lib/warnings.t
ok 114
PROG: 
# all warnings should be displayed 
use warnings ;
my $a = oct "0xfffffffffffffffffg" ;
EXPECTED:
Integer overflow in hexadecimal number at - line 3.
Illegal hexadecimal digit 'g' ignored at - line 3.
Hexadecimal number > 0xffffffff non-portable at - line 3.
GOT:
Integer overflow in hexadecimal number at - line 3.
Illegal hexadecimal digit 'g' ignored at - line 3.
not ok 115
PROG: 
# all warnings should be displayed 
use warnings ;
my $a = oct "0b111111111111111111111111111111111111111111111111111111111111111112";
EXPECTED:
Integer overflow in binary number at - line 3.
Illegal binary digit '2' ignored at - line 3.
Binary number > 0b11111111111111111111111111111111 non-portable at - line 3.
GOT:
Integer overflow in binary number at - line 3.
Illegal binary digit '2' ignored at - line 3.
not ok 116
ok 117
PROG: 
 
# Check scope of pragma with eval
use warnings;
{
    no warnings ;
    eval q[ 
        use warnings ;
        my $a = oct "0xfffffffffffffffffg" ;
    ]; print STDERR $@;
    my $a = oct "0xfffffffffffffffffg" ;
}
EXPECTED:
Integer overflow in hexadecimal number at (eval 1) line 3.
Illegal hexadecimal digit 'g' ignored at (eval 1) line 3.
Hexadecimal number > 0xffffffff non-portable at (eval 1) line 3.
GOT:
Integer overflow in hexadecimal number at (eval 1) line 3.
Illegal hexadecimal digit 'g' ignored at (eval 1) line 3.
not ok 118
PROG: 
 
# Check scope of pragma with eval
no warnings;
{
    use warnings ;
    eval '
        my $a = oct "0xfffffffffffffffffg" ;
    '; print STDERR $@ ;
}
EXPECTED:
Integer overflow in hexadecimal number at (eval 1) line 2.
Illegal hexadecimal digit 'g' ignored at (eval 1) line 2.
Hexadecimal number > 0xffffffff non-portable at (eval 1) line 2.
GOT:
Integer overflow in hexadecimal number at (eval 1) line 2.
Illegal hexadecimal digit 'g' ignored at (eval 1) line 2.
not ok 119

op/64bitint.t
1..67
# This perl's maths preserves all bits of a UV.
not ok 29
not ok 30
ok 31
not ok 32
not ok 33
ok 34
not ok 35
not ok 36

op/arith.t
not ok 128 # 9.22337203685478e+18 != 9223372036854775808
ok 129
ok 130 # -2 is close to 2, $^O eq super-ux

op/pack.t
ok 24
Cannot compress negative numbers at op/pack.t line 126.
# Looks like you planned 5826 tests but ran 24.

op/pow.t
Can't take log of 0 at op/pow.t line 11.

op/stat.t (ignore)
All ok when run interactively


2. Cray J90se;Unicos 10.0.1.1;Programming Environment 3.4

Failed 5/763 test scripts, 99.34% okay. 41/70072 subtests failed, 99.94% okay.
Failed Test                   Stat Wstat Total Fail  Failed  List of Failed
-------------------------------------------------------------------------------
../ext/Encode/t/mime-header.t    1   256     7    1  14.29%  7
../ext/Socket/socketpair.t       0     2    45   35  77.78%  11-45
op/pack.t                                 5826    1   0.02%  1354
op/pwent.t                                   2    1  50.00%  1
op/stat.t                                   73    3   4.11%  41-42 44
54 tests and 309 subtests skipped.

../ext/Encode/t/mime-header.t
not ok 7 - encode Q
#     Failed test (../ext/Encode/t/mime-header.t at line 93)
#          got: 'From:=?UTF-8?Q?=20=E5=B0=8F=E9=A3=BC=20=E5=BC=BE=20?=<dankogai@dan.co.jp>
# To: dankogai@dan.co.jp (=?UTF-8?Q?=E5=B0=8F=E9=A3=BC?==Kogai,
#  =?UTF-8?Q?=20=E5=BC=BE?==Dan)
# Subject:
#  =?UTF-8?Q?=20=E6=BC=A2=E5=AD=97=E3=80=81=E3=82=AB=E3=82=BF=E3=82=AB?=
#  =?UTF-8?Q?=E3=83=8A=E3=80=81=E3=81=B2=E3=82=89=E3=81=8C=E3=81=AA?=
#  =?UTF-8?Q?=E3=82=92=E5=90=AB=E3=82=80=E3=80=81=E9=9D=9E=E5=B8=B8?=
#  =?UTF-8?Q?=E3=81=AB=E9=95=B7=E3=81=84=E3=82=BF=E3=82=A4=E3=83=88?=
#  =?UTF-8?Q?=E3=83=AB=E8=A1=8C=E3=81=8C=E4=B8=80=E4=BD=93=E5=85=A8?=
#  =?UTF-8?Q?=E4=BD=93=E3=81=A9=E3=81=AE=E3=82=88=E3=81=86=E3=81=AB?=
#  =?UTF-8?Q?=E3=81=97=E3=81=A6Encode=E3=81=95=E3=82=8C?=
#  =?UTF-8?Q?=E3=82=8B=E3=81=AE=E3=81=8B=EF=BC=9F?=
# '
#     expected: 'From:=?UTF-8?Q?=20=E5=B0=8F=E9=A3=BC=20=E5=BC=BE=20?=<dankogai@dan.co.jp>
# To: dankogai@dan.co.jp (=?UTF-8?Q?=E5=B0=8F=E9=A3=BC?==Kogai,
#  =?UTF-8?Q?=20=E5=BC=BE?==Dan)
# Subject:
#  =?UTF-8?Q?=20=E6=BC=A2=E5=AD=97=E3=80=81=E3=82=AB=E3=82=BF=E3=82=AB?=
#  =?UTF-8?Q?=E3=83=8A=E3=80=81=E3=81=B2=E3=82=89=E3=81=8C=E3=81=AA=E3=82=92?=
#  =?UTF-8?Q?=E5=90=AB=E3=82=80=E3=80=81=E9=9D=9E=E5=B8=B8=E3=81=AB=E9=95=B7?=
#  =?UTF-8?Q?=E3=81=84=E3=82=BF=E3=82=A4=E3=83=88=E3=83=AB=E8=A1=8C=E3=81=8C?=
#  =?UTF-8?Q?=E4=B8=80=E4=BD=93=E5=85=A8=E4=BD=93=E3=81=A9=E3=81=AE=E3=82=88?=
#  =?UTF-8?Q?=E3=81=86=E3=81=AB=E3=81=97=E3=81=A6Encode=E3=81=95?=
#  =?UTF-8?Q?=E3=82=8C=E3=82=8B=E3=81=AE=E3=81=8B=EF=BC=9F?=
# '
# Looks like you failed 1 tests of 7.

../ext/Socket/socketpair.t
1..45
ok 1 - socketpair (LEFT, RIGHT, AF_UNIX, SOCK_STREAM, PF_UNSPEC)
ok 2 - syswrite to left
ok 3 - syswrite to left
ok 4 - syswrite to right
ok 5 - syswrite to right
ok 6 - read on left
ok 7 - content what we expected?
ok 8 - read on right
ok 9 - content what we expected?
ok 10 - shutdown left for writing
EOF on right took over 3 seconds at ../ext/Socket/socketpair.t line 118.
Something unexpectedly hung during testing at ../ext/Socket/socketpair.t line 39.

op/pack.t  (Is this still a problem?)
not ok 1354
# For list (0, 1, 9223372036854775807, 9223372036854775808, 18446744073709551615
) (total 0) packed with Q unpack '%65Q' gave 3.689348814742e+19, expected 0

op/pwent.t (ignore, we were expecting this due to using distributed
password scheme as per comment below. Minor problem with test is that
there is no check for Cray password administration system UDB)
not ok 1  
ok 2
# The failure of op/pwent test is not necessarily serious.
# It may fail due to local password administration conventions.
# If you are for example using both NIS and local passwords,
# test failure is possible.  Any distributed password scheme
# can cause such failures.

op/stat.t (ignore)
All ok when run interactively

Cheers,
	Len
-- 
Len.Makin@csiro.au:+61 3 9669 8109: CSIRO/Bureau of Meteorology
      High Performance Computing and Communications Centre
24th floor, 150 Lonsdale St | GPO Box 1289K, Melbourne, VIC 3001 Australia



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