Re: software single step - feature request

[Andrew Cagney <ac131313@cygnus.com>]

> Hi,
> 
> 
> There are a few nits associated with soft single step support, which are
> easy to work around but annoying enough to justify a fix.


Your not alone.  This is taken from gdbarch.sh:

# FIXME/cagney/2001-01-18: This should be split in two.  A target method 
that indicat
es if
# the target needs software single step.  An ISA method to implement it.
#
# FIXME/cagney/2001-01-18: This should be replaced with something that 
inserts breakp
oints
# using the breakpoint system instead of blatting memory directly (as 
with rs6000).
#
# FIXME/cagney/2001-01-18: The logic is backwards.  It should be asking 
if the target
  can
# single step.  If not, then implement single step using breakpoints.
F:2:SOFTWARE_SINGLE_STEP:void:software_single_step:enum target_signal 
sig, int insert
_breakpoints_p:sig, insert_breakpoints_p::0:0


> 1. Soft single step is configured per platform (e.g. arm-wince has one
>    and generic arm does not), IMHO, it should be provided on per CPU
>    basis.
> 
> 2. It has to be provided for every CPU - may be as part of multi-arch
>    transition.
> 
> 3. It should be run-time configurable - so it should be possible to
>    use the same gdb image both with and without soft single step.
> 
> 4. If target supports hw single step it should be possible to configure
>    removal of breakpoints before stepping.
> 
> 
> Thanks,
> 
> Aleksey


see gdb/120