So I am new to using gdb and gdbserver. And I'm trying to use it on a pretty
complex program. But that's okay, I can't get it to run on a simple system,
so let's talk about that.
I created the usual c++ Hello World. I have access to two identical Centos
5 development systems connected with tcp.
#include <iostream>
int main(int argc, char *argv[])
{
std::cout << "Hello World." << std::endl;
return -1;
}
If I compile my hello world with:
$ g++ -g hello.cpp -o hello
Then everything works pretty much as expected. I can run it on either
system. I can use gdb on one system, connect to gdbserver on the other
system, and everything seems to work fine. I set a break at main on my
local system, gdbserver the program on the target, use gdb to tell it to
run, and Hello World prints out on the gdbserver system.
But if I compile it with -m32, and repeat the process, then the two systems
seem to be fighting over architecture issues. I receive messages like
register badly formatted. Or "warning: Selected architecture i386 is not
compatible with reported target architecture i386:x86-64". This seems to
happen regardless of how I tell gdb to set the architecture, either i386 or
i386:x86-64.
g++ -g -m32 hello.cpp -o hello
So I gather much of my problem is because I'm ignorant. And I don't really
understand what -m32 does, although I know our hideously complex system IS
compiled that way.
Apart from that, I am using:
$ g++ --version
g++ (GCC) 4.1.2 20080704 (Red Hat 4.1.2-46)
$ gdbserver --version
GNU gdbserver Fedora (6.8-37.el5)
This gdbserver was configured as "x86_64-redhat-linux-gnu"
$ cat /etc/redhat-release
CentOS release 5.4 (Final)
Can someone help clue me in?
Thank you