README for GDBserver
by Stu Grossman
Introduction:
This is GDBserver, a remote server for Un*x-like systems. It can be used to
control the execution of a program on a target host from a GDB on a different
host. GDB and GDBserver communicate using the standard remote serial protocol
implemented in remote.c, and various *-stub.c files. They can communicate via
either a serial line or a TCP connection.
Usage (server (target) side):
First, you will need to have a copy of the program to be debugged put onto
the target system. It can be stripped if you need to save space. This is ok
because GDBserver doesn't care about symbols, all of that stuff is taken care
of by the GDB running on the host system.
To use the server, you will need to log on to the target system, and run the
server program. You will need to tell it how to communicate with GDB, the
name of the program to be debugged, and it's arguments. For example, using a
serial port, you might say:
target> gdbserver /dev/com1 emacs foo.txt
This tells gdbserver to debug emacs with an argument of foo.txt. The server
will communicate with GDB via /dev/com1. GDBserver will now wait patiently
for GDB to communicate with it.
To use a TCP connection, you could say:
target> gdbserver host:2345 emacs foo.txt
This says pretty much the same thing as the last example, except that we are
now going to communicate with GDB via TCP. The `host:2345' argument means that
we are expecting to see a TCP connection from `host' to local TCP port 2345.
Currently, the host part is ignored. You can choose any number you want for
the port number as long as it does not conflict with any existing ports on your
system. This same port number will also be used in the GDB `target remote'
command, which we will discuss later. Note that it's safe to chose a number
that conflicts, gdbserver will just print an error message and exit.
Usage (host side):
You should have a copy of the target program on your host system, since GDB
will need it to examine symbol tables and such. You should start up GDB just
as you normally would, with the target program as the first argument. Ie:
`gdb target-prog'. After that, you will only need to know about one new
command. This is `target remote'. It's argument is either a device name
(preferably of a serial device, like /dev/ttyb), or a host:port descriptor.
For example:
(gdb) target remote /dev/ttyb
will communicate with the server via the hardware serial line /dev/ttyb, and:
(gdb) target remote the-target:2345
will communicate via a TCP connection to port 2345 on host `the-target', where
you have already started up gdbserver with the same port number. Note that you
must start up gdbserver prior to using the target command, otherwise you will
get an error that looks something like `Connection refused'.
Building:
Currently, the only target system supported by the server is Lynx. To build
the server for Lynx, make a new copy of the distribution onto a disk that is
NFS shared with the Lynx system. Lets say that's in a directory called xyzzy.
Then, follow these steps under the host system:
1) cd xyzzy/gdb/gdbserver
2) ../../configure --target i386-none-lynx
When that completes, do the following on the Lynx system:
3) cd xyzzy/gdb/gdbserver
4) make CC=gcc
It should build with only a minor complaint about NULL being redefined. That's
a LynxOS problem, and can be ignored.
It's also possible that you may have a cross-compiler to Lynx. In that case,
you can skip the stuff about NFS. You would replace steps 3 & 4 with:
make CC=lynx-target-compiler...
