[ python-Bugs-864379 ] Python 2.3.3 make test core dump on HP-UX11i

[SourceForge.net]

Bugs item #864379, was opened at 2003-12-22 12:40
Message generated for change (Comment added) made by loewis
You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105470&aid=864379&group_id=5470

Please note that this message will contain a full copy of the comment thread,
including the initial issue submission, for this request,
not just the latest update.
Category: Build
Group: Platform-specific
>Status: Closed
>Resolution: Wont Fix
Priority: 5
Submitted By: Richard Townsend (rptownsend)
Assigned to: Nobody/Anonymous (nobody)
Summary: Python 2.3.3 make test core dump on HP-UX11i

Initial Comment:
I built Python-2.3.3 on HP-UX11i in the same way that I 
built Python-2.3.2.

However, this time running 'make test' core dumps on 
the second pass with:

test_sax
*** Termination signal 138

Stop.

The output from test_sax.py is given below. Each test 
appears to fail, but the summary gives zero failures (this 
also happens with Python 2.3.2, but 'make test' doesn't 
core dump on that release).

> ./python ./Lib/test/test_sax.py
Failed test_attrs_empty
Failed test_attrs_wattr
Failed test_double_quoteattr
Failed test_escape_all
Failed test_escape_basic
Failed test_escape_extra
Failed test_expat_attrs_empty
Failed test_expat_attrs_wattr
Failed test_expat_dtdhandler
Failed test_expat_entityresolver
Failed test_expat_file
Failed test_expat_incomplete
Failed test_expat_incremental
Failed test_expat_incremental_reset
Failed test_expat_inpsource_filename
Failed test_expat_inpsource_location
Failed test_expat_inpsource_stream
Failed test_expat_inpsource_sysid
Failed test_expat_locator_noinfo
Failed test_expat_locator_withinfo
Failed test_expat_nsattrs_empty
Failed test_expat_nsattrs_wattr
Failed test_filter_basic
Failed test_make_parser
Failed test_make_parser2
Failed test_nsattrs_empty
Failed test_nsattrs_wattr
Failed test_quoteattr_basic
Failed test_single_double_quoteattr
Failed test_single_quoteattr
Failed test_unescape_all
Failed test_unescape_amp_extra
Failed test_unescape_basic
Failed test_unescape_extra
Failed test_xmlgen_attr_escape
Failed test_xmlgen_basic
Failed test_xmlgen_content
Failed test_xmlgen_content_escape
Failed test_xmlgen_ignorable
Failed test_xmlgen_ns
Failed test_xmlgen_pi
41 tests, 0 failures


----------------------------------------------------------------------

>Comment By: Martin v. Löwis (loewis)
Date: 2006-04-12 19:01

Message:
Logged In: YES 
user_id=21627

Sure. Closing it as "won't fix". If the problem persists for
2.5, please re-report.

----------------------------------------------------------------------

Comment By: Georg Brandl (gbrandl)
Date: 2006-04-12 16:39

Message:
Logged In: YES 
user_id=849994

As 2.3.x is not maintained anymore, should this be closed?

----------------------------------------------------------------------

Comment By: Martin v. Löwis (loewis)
Date: 2006-04-12 10:03

Message:
Logged In: YES 
user_id=21627

I don't think I will do anything about this anytime soon, so
unassigning myself.

----------------------------------------------------------------------

Comment By: Neal Norwitz (nnorwitz)
Date: 2003-12-24 14:23

Message:
Logged In: YES 
user_id=33168

I think I remember hearing something about an
incompatibility.  The two best people to discuss this are
probably Martin v. Loewis and Fred Drake.  I'm assigning to
Martin, perhaps he can shed some more light on the issues.

----------------------------------------------------------------------

Comment By: Richard Townsend (rptownsend)
Date: 2003-12-24 13:46

Message:
Logged In: YES 
user_id=200117

I have now installed Python-2.3.3 plus PyXML-0.8.3 and the 
core dump in 'make test' has returned.

So, it looks like a compatability problem with PyXML-0.8.3 ?



----------------------------------------------------------------------

Comment By: Richard Townsend (rptownsend)
Date: 2003-12-24 10:25

Message:
Logged In: YES 
user_id=200117

HP-UX capulet B.11.11 U 9000/785 2002932658 unlimited-user 
license

This is the configure command I used:

base_inst_dir=/opt
python_link=${base_inst_dir}/python

CC=cc \
OPT="+O2 -z +Onofltacc +ESlit +Oentrysched \
+Odataprefetch +Onolimit +DA2.0" \
INSTALL="./install-sh -c" \
./configure --with-threads --without-gcc \
--with-cxx=/opt/aCC/bin/aCC --without-libd \
--with-cycle-gc --prefix=$python_link \
--exec-prefix=$python_link

This is exactly the same command I have used on previous 
Python 2.3 releases which did not core dump in 'make test'.

Note it uses the native HP compiler.

I have now just rebuilt without optimisation, using:

CC=cc \
INSTALL="./install-sh -c" \
./configure --with-threads --without-gcc \
--with-cxx=/opt/aCC/bin/aCC --without-libd \
--with-cycle-gc --prefix=$python_link \
--exec-prefix=$python_link

It still core dumps at the same point in the second pass 
of 'make test'.

I have attached a stack trace from the core file. I noticed 
that this has references to /opt/python/lib/python2.3/site-
packages/_xmlplus/parsers/pyexpat.sl.  This is the previous 
installation of Python 2.3.2 plus PyXML-0.8.3.  

I wondered if I had forgotten to reset $PYTHONPATH before 
running 'make test', so I did that and tried again but it still 
core dumped.

So next I reset the symbolic link /opt/python to the empty 
directory /opt/python-2.3.3 and ran 'make test' again. This 
time it didn't core dump!

So why does the test suite access the previous installation 
when PYTHONPATH doesn't include it?




----------------------------------------------------------------------

Comment By: Neal Norwitz (nnorwitz)
Date: 2003-12-24 04:32

Message:
Logged In: YES 
user_id=33168

On an HP box I tried, I don't have any problems.
 * HP-UX taylor B.11.00 A 9000/829
 * gcc 3.2

What compiler are you using?  Have you tried on the HP test
drive machines?  Are you compiling with optimization?

----------------------------------------------------------------------

You can respond by visiting: 
https://sourceforge.net/tracker/?func=detail&atid=105470&aid=864379&group_id=5470