Make a few functions work with base_breakpoint instead of breakpoint
authorPedro Alves <pedro@palves.net>
Fri, 6 May 2022 22:11:34 +0000 (23:11 +0100)
committerPedro Alves <pedro@palves.net>
Fri, 20 May 2022 19:41:00 +0000 (20:41 +0100)
commitff733ec228e3f4f1c1607fe50aeb944505105fde
treeb6f5ae051a9aa5b251a7c012b10b9f696dd2834d
parentb925bf21e073f0d9531385e8cfc3a825887c0778
Make a few functions work with base_breakpoint instead of breakpoint

This makes tracepoints inherit from base_breakpoint, since their
locations are code locations.  If we do that, then we can eliminate
tracepoint::re_set and tracepoint::decode_location, as they are doing
the same as the base_breakpoint implementations.

With this, all breakpoint types created by new_breakpoint_from_type
are code breakpoints, i.e., base_breakpoint subclasses, and thus we
can make it return a base_breakpoint pointer.

Finally, init_breakpoint_sal can take a base_breakpoint pointer as
"self" pointer too.  This will let us convert this function to a
base_breakpoint ctor in a following patch.

Change-Id: I3a4073ff1a4c865f525588095c18dc42b744cb54
gdb/breakpoint.c
gdb/breakpoint.h