support for ARM GNU/Linux
Richard Earnshaw
rearnsha@arm.com
Wed Apr 14 03:36:00 GMT 1999
ac131313@cygnus.com said:
> I'm slightly curious about what is going on with all these different
> breakpoints (I should hasten to point out that I'm _not_ the ARM
> expert/maintainer :-). Stan Shebs recently commited the change:
> 1999-03-02 Stan Shebs <shebs@andros.cygnus.com>
> From Gary Thomas <gthomas@cygnus.co.uk>:
> * arm-tdep.c (ARM_LE_BREAKPOINT, ARM_BE_BREAKPOINT,
> THUMB_LE_BREAKPOINT, THUMB_BE_BREAKPOINT): Use illegal instruction
> instead of SWI 24.
> * config/arm/tm-arm.h (CALL_DUMMY): Ditto.
> (IN_SIGTRAMP): Define.
A bit of history. SWI 24 was the breakpoint swi in RISC iX, which the
kernel understood (in some way). I don't know if it was necessary to use
exactly that SWI if you were using gdb or if the same effect could be
achieved with other illegal instructions.
A side-note, before ARM Architecture v4 many "undefined" instruction
formats aren't guaranteed to take the undefined instruction trap. Also
note that, in this respect, the ARM7TDMI is NOT fully ARM v4 compliant.
I don't know if anyone cares about RISC iX support any more (though gcc
still supports it).
Richard.
More information about the Gdb-patches
mailing list