Bug 115

Summary: f2c gives segmentation fault when initializing
Product: [Retired] Red Hat Linux Reporter: nolty_r
Component: fort77Assignee: Trond Eivind Glomsrxd <teg>
Status: CLOSED CURRENTRELEASE QA Contact:
Severity: medium Docs Contact:
Priority: medium    
Version: 5.1   
Target Milestone: ---   
Target Release: ---   
Hardware: i386   
OS: Linux   
Whiteboard:
Fixed In Version: Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 1999-05-16 21:25:59 UTC Type: ---
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:

Description nolty_r 1998-11-18 02:56:20 UTC
(I got this RPM from the contrib area, but the RPM info
lists `bugs' as the packager).

I am attempting to use f2c (f2c-19970805-3) on RedHat 5.1.
So far all
my programs compile OK, but one of my executables is
segmentation
faulting during initialization, before control transfers to
MAIN__.
The call stack after the crash is:

#0 malloc_extend_top
#1 chunk_alloc
#2 __libc_malloc
#3 malloc_hook_ini
#4 __libc_malloc
#5 f_init
#6 main

Routines #0-4 are from a malloc.c file that was apparently
compiled
debug (file and line number information is printed in gdb)
but I don't
have the .c file on my system.

Does anyone have any idea what could cause this?  This is a
very large
FORTRAN program linked against lots of large FORTRAN
libraries, but I
also compiled and linked a different program with 90% the
same code,
which initialized OK.

In /usr/doc/f2c-19970805/README, it says that the malloc.c
supplied
with f2c may not work on some systems; does anyone know if
this rpm
was compiled to use f2c's malloc.c?  Could that be the
problem?

Incidentally, I had this program working using f2c with an
old
slackware installation a few years ago.

Bob

Comment 1 Jeff Johnson 1999-05-14 18:52:59 UTC
There's a newer patched and maintained f2c that's part of PowerTools.
See if that fixes your problem. Please reopen this bug if not.