This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH v6 2/9] Explicit locations: introduce new struct event_location-based API
- From: Doug Evans <xdje42 at gmail dot com>
- To: Keith Seitz <keiths at redhat dot com>
- Cc: gdb-patches at sourceware dot org
- Date: Mon, 10 Aug 2015 12:58:33 -0700
- Subject: Re: [PATCH v6 2/9] Explicit locations: introduce new struct event_location-based API
- Authentication-results: sourceware.org; auth=none
- References: <20150805232802 dot 21646 dot 88440 dot stgit at valrhona dot uglyboxes dot com> <20150805232913 dot 21646 dot 96976 dot stgit at valrhona dot uglyboxes dot com> <m3d1yvdp5b dot fsf at sspiff dot org> <55C8E7FB dot 3030106 at redhat dot com>
Keith Seitz <keiths@redhat.com> writes:
> On 08/10/2015 10:33 AM, Doug Evans wrote:
>> Later (not now, let's get this sucker checked in, unless of course
>> you really want to), let's make event_location_to_string take a
>> const struct event_location *, and make the cached copy
>> "mutable in the c++ sense".
>
> Bah. You got me on that one. This was used as part of the
> pending-location hack ("append extra_string to the location") in
> create_breakpoint:
>
> if (extra_string != NULL)
> {
> char *new = xstrprintf ("%s %s",
> event_location_to_string_const (location),
> extra_string);
>
> set_event_location_string (b->location, new);
> xfree (new);
> }
>
> ... which was removed in this revision, and I forgot to check if it was
> being used at all anymore.
>
> Answer: No, it is not used anymore. I could just remove it entirely.
> Which is what I've done in my local copy. [Do you want me to repost this?]
>
> Thanks!
>
> Keith
Well, event_location_to_string_const is used internally in one location
in the v6 patch set, but if you've removed it completely that's fine by me.
As for reposting, in this particular case (2/9) I suppose you should,
but only for informational sake. No need for waiting for another review.