LZOP for HP-UX

[Mihai Lazar]

Hi

I am looking for lzop for HP-UX

I am trying to compile lzop on HP-UX 11.11 (hppa64-hp-hpux11.11)
I am getting the following error: cannot compute sizeof (short), 77
See `config.log' for more details.

I have the sources for lzop 1.02rc1 and lzo-2.03 installed form the depo.
I have tried to use HP ANSI C++ B3910B A.03.60 and gcc-4.2.3 no luck.

Please see attached the config.log.

If you have a copy of lzop already precompiled that will be of great help.
I got lzo precompiled from http://hpux.cs.utah.edu.

If you have any suggestions that will be of big help.

Best regards

[Steven Schweda]

Works fine for me:

dy # uname -a
HP-UX dy B.11.11 U 9000/785 2012616114 unlimited-user license

dy # gcc --version
gcc (GCC) 4.3.0
[...]

config.log:
[...]
configure:6105: checking for short
configure:6129: gcc -c -g -O2 conftest.c >&5
configure:6135: $? = 0
configure:6138: test -z || test ! -s conftest.err
configure:6141: $? = 0
configure:6144: test -s conftest.o
configure:6147: $? = 0
configure:6158: result: yes
configure:6161: checking size of short
configure:6473: gcc -o conftest -g -O2 conftest.c -llzo2 >&5
configure:6476: $? = 0
configure:6478: ./conftest
configure:6481: $? = 0
configure:6504: result: 2
[...]

No "Bus error(coredump)" here.

I extracted the "configure: failed program
was:" stuff, and looked at it. Seems pretty
simple. Even runs on my VMS system. Gets
"2" there, too:

alp $ cc short
alp $ link short
alp $ run short
alp $ type conftest.val
2

There seems to be something seriously wrong
with your system, or compiler, or something.
Luckily, HP hasn't offered me their C or C++
compiler at a competitive price ($0), so I
can't compare using one of those.

The rest of the build seems to work as well:

dy # ./src/lzop --version
lzop 1.02rc1
LZO library 2.03
Copyright (C) 1996-2005 Markus Franz Xaver Johannes Oberhumer

> I have tried to use HP ANSI C++ B3910B
> A.03.60 and gcc-4.2.3 no luck.

A transcript showing the actual commands and
actual output might be more helpful than a
vague description of what you did ("I tried
[...] gcc-4.2.3"), and of what happened when
you did it ("no luck").

Have you tried using a C compiler instead of
a C++ compiler?

[Dennis Handly]

I have no problems compiling that fragment with aCC -Ae. Of course I don't have: -I/usr/local/include/lzo -L/usr/local/lib -llzo2

Can you use a debugger to find where that signal 10 is occurring?

>Steven: Have you tried using a C compiler instead of a C++ compiler?

The use of "aCC -Ae" should be fine.