Re: [RFA] Don't reset watchpoint block on solib load.

[Jim Blandy <jimb at codesourcery dot com>]

Vladimir Prus <vladimir at codesourcery.com> writes:
>> Should we instead be saving the PC in whose scope we parsed the
>> expression?  That's something we could legitimately look up in
>> breakpoint_re_set_one after a symbol table change.  The test in
>> watchpoint_check could look up the PC's function and compare it to the
>> frame's function.
>
> Well, it does not work with address randomization. And I think that for
> global watchpoints, just re-parsing the expression works good
> enough.

Address randomization is a good point.  I think we should still be
able to get the right behavior somehow, though.