develooper Front page | perl.perl5.porters | Postings from July 2001

report on BS2000 with perl@11006

Thread Next
From:
Dorner Thomas
Date:
July 5, 2001 00:35
Subject:
report on BS2000 with perl@11006
Message ID:
6727B1DACFCDD311A757009027CA8D690257A973@Ex02.inhouse.start.de
Hi fellow porters,

here are the latest tests from BS2000 Posix Subsystem using the 11006:


Failed 31/422 test scripts, 92.65% okay. 248/23008 subtests failed, 98.92%
okay.
Failed Test                     Stat Wstat Total Fail  Failed  List of
Failed
----------------------------------------------------------------------------
---
../ext/B/Deparse.t                            14    1   7.14%  14
../ext/B/Showlex.t                             1    1 100.00%  1
../ext/Digest/MD5/t/files.t                    2    2 100.00%  1-2
../ext/Encode/Encode/Tcl.t                   590    5   0.85%  2 4 6 8 10
../ext/IO/lib/IO/t/io_sock.t     204 52224    20    9  45.00%  12-20
../ext/IO/lib/IO/t/io_udp.t        0    15     7    4  57.14%  4-7
../ext/POSIX/POSIX.t                          29    1   3.45%  14
../ext/Time/HiRes/HiRes.t          0    25    19    4  21.05%  7-8 11-12
../ext/XS/Typemap/Typemap.t        0   132    84   30  35.71%  55-84
../lib/charnames.t                            16    2  12.50%  11-12
../lib/overload.t                            222    4   1.80%  152-153
170-171
../lib/warnings.t                            434    1   0.23%  75
../lib/ExtUtils.t                             27   24  88.89%  4-27
../lib/File/Find/taint.t           8  2048    45   45 100.00%  1-45
../lib/Math/BigInt/t/bigintpm.t             1190   12   1.01%  207 806 819
916
                                                               918 924 926
928
                                                               930 1145
1182-
                                                               1183
../lib/Math/Complex.t                       1020    2   0.20%  267 487
comp/require.t                                20    1   5.00%  18
op/bop.t                                      44    1   2.27%  37
op/concat.t                      255 65280    11    8  72.73%  4-11
op/magic.t                       232 59392    41    5  12.20%  37-41
op/misc.t                                     83    2   2.41%  12 70
op/numconvert.t                             1440   32   2.22%  1409-1440
op/pat.t                                     639    9   1.41%  234-235 238
240
                                                               242-243
424-425
                                                               626
op/regexp.t                                  791    3   0.38%  483 496 500
op/regexp_noamp.t                            791    3   0.38%  483 496 500
op/sprintf.t                                 221   30  13.57%  21 27 30-31
35-
                                                               41 170-172
188-
                                                               194 197-205
op/stat.t                                     58    3   5.17%  36-37 39
op/study.t                                    ??   ??       %  ??
op/taint.t                       232 59392    ??   ??       %  ??
op/tr.t                                       70    3   4.29%  50 58-59
op/vec.t                                      30    1   3.33%  27
 (3 subtests UNEXPECTEDLY SUCCEEDED), 16 tests and 83 subtests skipped.


Now lets get into some details - my major concerns at the moment are some
(probably) infinte loops and a few core dumps:

op/concat.t:
1..11
ok 1
ok 2
ok 3
Malformed UTF-8 character (unexpected continuation byte 0x48) in regexp
compilation at op/concat.t line 29.
Malformed UTF-8 character (unexpected continuation byte 0x59) in regexp
compilation at op/concat.t line 29.
Unmatched [ in regex; marked by <-- HERE in m/([ <-- HERE ??)/ at
op/concat.t line 29.
Here are two Latin-1 characters, ccedile and ssharp:      ^^

Seems like unicode kills the test script.

op/magic.t:
1..41:
ok 1
[...]
ok 31
ok 32 # skipped: no caseless %ENV support
ok 33 # skipped: no caseless %ENV support
ok 34 # skipped: no caseless %ENV support
ok 35 # skipped: no caseless %ENV support
ok 36
%  CCM0996 NO SPACE AVAILABLE FOR RUNTIME STACK
%  IDA0N51 PROGRAM INTERRUPT AT LOCATION '01A52BC8 (IC@MAI3@), (CDUMP),
EC=90, CODE=NRT0101'

The system wrote a core dump of about 520 MB (the executable itself is about
12 MB) so this looks like an infinite recursion in test 37.  What is so
special about that eval with the '%!' variable?

op/study.t:
1..26
ok 1
[...]
ok 24
%  EXC0072 TIME LIMIT FOR TASK/PROGRAM EXCEEDED. PROGRAM TERMINATED
%  IDA0N51 PROGRAM INTERRUPT AT LOCATION '017A61A8 (REGEXEC&), (CDUMP),
EC=80, CODE=ETMSE07'

Well, the test says "may loop" - it seems to do that here (note that it was
caught somewhere in regexec.o).

op/taint.t:
%  CCM0996 NO SPACE AVAILABLE FOR RUNTIME STACK
%  IDA0N51 PROGRAM INTERRUPT AT LOCATION '01A52BC8 (IC@MAI3@), (CDUMP),
EC=90, CODE=NRT0101'

This one dumps before the very first output.

../lib/Math/BigInt/t/bigintpm.t:

All errors are NaN errors (not a number), are they similar to the ones on
IBM's 390?

../ext/XS/Typemap/Typemap:
1..84
ok 1
[...]
ok 54
# T_OPAQUEPTR with a struct
%  IDA0N51 PROGRAM INTERRUPT AT LOCATION '012DF250 (TYPEMAP&), (CDUMP),
EC=58, CODE=ETMSE05'

The first "normal" core dump (due to some bad pointer or something similar).

../ext/IO/lib/IO/t/io_udp.t:
1..7
ok 1
ok 2
ok 3

This one seemed to hang after test 3 (I had to kill the interpreter after 10
minutes or so).  There is no possible "not ok 4" in the test, so what may
that mean?

Regards, Thomas

Thread Next


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