1. SIGBUS does not cause core dump
I am having a programming problem under Linux that I'm sure someone
else has already encountered and resolved. My problem is simply this:
when a process receives the SIGBUS signal (signal 7), it terminates
without a core image file.
The executable compiled from the following source module easily
replicates the problem on my system. It is compiled with gcc 2.6.3 as
follows: gcc -static -g foo.c -o foo
------------------- foo.c ----------------------------
#include <stdio.h>
int main(int argc, char **argv) {
int i;
int foo[10];
for(i=0; i < 1024; i++)
foo[i] = i;
------------------------------------------------------
When I run the executable generated from foo.c, the process gets a
SIGBUS (signal 7), which causes it to terminate with no core file.
The SIGSEGV signal does cause receiving processes terminate with
core image files. I have removed the default suppression of core files
under "bash" by running the command: ulimit -c unlimited.
I tried running "foo" in several different shells, and got the same
result. In each shell, I suppressed the shell limits associated with
the production of core files.
I suspect that the solution to the problem I am having has something
to do with the way the kernel on my system is configured.
I am posting this question here as a last resort. I have carefully
searched the FAQs and HOWTOs for an answer, and found nothing.
How do you get a process that receives SIGBUS to dump core in a
meaningful way? I tried installing a signal handler for SIGBUS which
calls abort(). This gets me a core file which is useless, since the
stack frame of the offending code is not accessible in the debugger
due to the invocation of the signal handler.
Here is some information about my system configuration that may help:
Kernel: Version 1.2.8, built from the kernel sources that came with
the Slackware 2.3 distribution. The kernel configuration is very
plain.
Hardware: Intel 486 DX/33, 16MB ram, EIDE hard drive and controller
You can email suggestions, comments, or flames directly to me at:
Thanks in advance,
Eddie Parker
2. ST-01 Seagate controller
3. httpd: caught SIGBUS, attempting to dump core in /usr/local/wwws
4. Xfree86 RPM Problem
5. Apache problem - core dumps (httpd.core) all over the place
6. need help with lilo
7. dump core or not dump core
8. best pop server?
9. core dumps core dumps everywhere...
10. Apple RTSP Proxy core dump problems
11. Apache-1.2.6 proxy-server with non-standard HTTPS port ?
12. proxy-server for ftp, gopher, wais (Apache)?
13. Apache-1.2b1 as a proxy-server for gopher/WAIS ?
by