Re: [PATCH 1/2] Support the new PPC476 processor -- Arch Independent

[Joel Brobecker <brobecker at adacore dot com>]

> > > +	    /* We have to temporary disable this watchpoint, otherwise
> > > +	       we will count it twice (once as being inserted, and once
> > > +	       as a watchpoint that we want to insert).  */
> > > +	    e = b->enable_state;
> > > +	    b->enable_state = bp_disabled;
> > 
> > What's the story behind the need to temporarily disabling this
> > watchpoint?
> 
> This is a fix for an actual bug in the current GDB code. Jan suggested an
> improved fix. I'll post this chunk as a separate patch, and explain the
> issue there.

Please do! :-)

I just noticed that bug yesterday, and came up with yet another fix
(which is the worse of all suggestions made so far, so I won't send it,
after all [had everything ready to go, I was just waiting for our nightly
build results to come back and some of them take more than 24h]). You
and I are probably the only ones who care, because most of the
watchpoint implementations I looked at were just ignoring the count
and always returning either 1 (ok) or 0 (no more resource).

-- 
Joel