Re: [PATCH 2/3] gdb/python: fix use of frame_info_ptr within pending_frame_object

[Tom Tromey <tom@tromey.com>]

>>>>> "Andrew" == Andrew Burgess <aburgess@redhat.com> writes:

Andrew> The problem with frame_info_ptr is that its constructor registers the
Andrew> new frame_info_ptr with the global frame_list.  It is by this
Andrew> registration that invalidation of frame_info_ptr objects is performed.

Thanks for finding & fixing this.

Andrew> +  frame_info_ptr frame (*pending_frame->frame_info);
Andrew>    return PyUnicode_FromFormat ("<%s frame #%d, saved_regs=(%s)>",
Andrew>  			       gdbpy_py_obj_tp_name (self),
Andrew>  			       frame_relative_level (frame),

This could be 'const frame_info_ptr &frame (...)' and avoid some updates
to globals, since no copy will be done.

I didn't look closely but perhaps other spots could do the same.

It may not matter so if you want to ignore this that's fine by me.

Approved-By: Tom Tromey <tom@tromey.com>

Tom