Lpc2000

Has anyone succeeded in getting hardware breakpoints
working with a Wiggler and gdb/insight?

Mine works perfectly for debugging code in ram (it
inserts software breakpoints perfectly), but I can't
set a breakpoint in  flash.  Neither "b main" or 
"hbreak main" work; both commands are accepted, but 
when I continue, the first gives an error writing to 
flash (no surprise) and the second warns that it could 
not insert the hardware breakpoint.

Any ideas?

At 08:24 PM 6/27/04 +0000, you wrote:
>Has anyone succeeded in getting hardware breakpoints
>working with a Wiggler and gdb/insight?
>
>Mine works perfectly for debugging code in ram (it
>inserts software breakpoints perfectly), but I can't
>set a breakpoint in  flash.  Neither "b main" or
>"hbreak main" work; both commands are accepted, but
>when I continue, the first gives an error writing to
>flash (no surprise) and the second warns that it could
>not insert the hardware breakpoint.
>
>Any ideas?

monitor set hbreak address

It works with numerics anyway, I haven't tried a symbolic address yet.

Robert


" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III

--- In lpc2000@yahoogroups.com, Robert Adsett <subscriptions@a...> wrote:
> At 08:24 PM 6/27/04 +0000, you wrote:
> >Has anyone succeeded in getting hardware breakpoints
> >working with a Wiggler and gdb/insight?
> >
> >Mine works perfectly for debugging code in ram (it
> >inserts software breakpoints perfectly), but I can't
> >set a breakpoint in  flash.  Neither "b main" or
> >"hbreak main" work; both commands are accepted, but
> >when I continue, the first gives an error writing to
> >flash (no surprise) and the second warns that it could
> >not insert the hardware breakpoint.
> >
> >Any ideas?
> 
> monitor set hbreak address
> 
> It works with numerics anyway, I haven't tried a symbolic address yet.
> 

Thanks Robert, the numeric address worked perfectly!  (Symbolic
addresses do not...they are accepted but don't actually cause a
break).  Is there a way to clear the hardware breakpoint once set (for
now I am just setting it somewhere else in memory, but it would be
nice to know how to just plain clear it).

At 05:58 AM 6/30/04 +0000, you wrote:
>--- In lpc2000@yahoogroups.com, Robert Adsett <subscriptions@a...> wrote:
> > monitor set hbreak address
> >
> > It works with numerics anyway, I haven't tried a symbolic address yet.
> >
>
>Thanks Robert, the numeric address worked perfectly!  (Symbolic
>addresses do not...they are accepted but don't actually cause a
>break).  Is there a way to clear the hardware breakpoint once set (for
>now I am just setting it somewhere else in memory, but it would be
>nice to know how to just plain clear it).

You're welcome.

How about using monitor set hbreak 0 as a 'clear'?  In most cases if you 
start executing from the reset vector that's an error anyway so it's a good 
point to trap.

Robert

" 'Freedom' has no meaning of itself.  There are always restrictions,
be they legal, genetic, or physical.  If you don't believe me, try to
chew a radio signal. "

                         Kelvin Throop, III