This is the mail archive of the
gdb-patches@sources.redhat.com
mailing list for the GDB project.
Re: dejagnu MMIX port
- From: Andrew Cagney <ac131313 at cygnus dot com>
- To: Hans-Peter Nilsson <hp at bitrange dot com>
- Cc: gdb-patches at sources dot redhat dot com
- Date: Sun, 20 Jan 2002 19:54:29 -0500
- Subject: Re: dejagnu MMIX port
- References: <Pine.BSF.4.30.0111101129120.83243-100000@dair.pair.com>
> Please consider accepting this port.
>
> 2001-11-10 Hans-Peter Nilsson <hp@bitrange.com>
>
> * baseboards/mmixware-sim.exp: New file.
For the src/dejagnu repository, please feel free to just check it in. I
don't think anyone cares if you add support for a new base board.
Andrew
> *** /dev/null Tue Jan 1 05:00:00 1980
> --- baseboards/mmixware-sim.exp Sat Nov 10 17:24:54 2001
> ***************
> *** 0 ****
> --- 1,59 ----
> + # This is a list of toolchains that are supported on this board.
> + set_board_info target_install {mmix-knuth-mmixware}
> +
> + # Load the generic configuration for this board. This will define any
> + # routines needed to communicate with the board.
> + load_generic_config "sim"
> +
> + # Unset the slow_simulator flag; this simulator isn't slow. (I haven't
> + # found documentation as to what constitutes a "slow" simulator and
> + # there's no reference to that variable in the GCC test-suite, but all
> + # tests I've found work.)
> + unset_board_info slow_simulator
> +
> + # We don't use basic-sim.exp and setup_sim because the simulator, called
> + # "mmix", is external to the build tree. If nobody set the simulator
> + # name, we set it to "mmix", so the rest of the framework works. Let $SIM
> + # override.
> + if ![board_info $board exists sim] {
> + if [info exists SIM] {
> + set_board_info sim $SIM
> + } else {
> + # Has to exist in the users path.
> + set_board_info sim mmix
> + }
> + }
> +
> + # No default multilib options are needed for this board.
> + process_multilib_options ""
> +
> + set_board_info is_simulator 1
> +
> + # It isn't visible in the target triple, so we set it explicitly. The
> + # *linked* format is currently mmo, but those cases where this matters we
> + # can deal with separately.
> + set_board_info obj_format "elf"
> +
> + # The compiler used to build for this board. This has *nothing* to do
> + # with what compiler is tested if we're testing gcc.
> + set_board_info compiler "[find_gcc]"
> +
> + # We only support newlib on this target. We assume that all multilib
> + # options have been specified before we get here.
> + set_board_info cflags "[newlib_include_flags]"
> +
> + # We put the stack in the code segment, so trampolines work; the mmix
> + # simulator does not allow code to be executed in the "normal" documented
> + # data and stack segments. Use a somewhat magic number, so stack addresses
> + # are easily recognized when debugging.
> + set_board_info ldflags "[newlib_link_flags] -Wl,--defsym,__Stack_start=0x1680a00000000000"
> +
> + # No linker script needed - the simulator is the main environment for this
> + # target.
> + set_board_info ldscript ""
> +
> + # Return codes aren't supported by the simulator.
> + set_board_info needs_status_wrapper 1
> +
> + # FIXME: Some gdb board variables are usually here. I currently have no clue what to
> + # set them to, so I won't.
>
> brgds, H-P
>
>
>