[PATCHv5] gdb: prevent an assertion when computing the frame_id for an inline frame

[Andrew Burgess]

* Pedro Alves <pedro@palves.net> [2021-08-23 11:26:49 +0100]:

> Hi Andrew,
> 
> Sorry I hadn't responded yet.  I was out on vacation a few weeks ago, and thought I would be able to
> get to this once I got back, but I got immediately pulled to a tight internal deadline instead.  FWIW, I've
> been very frustrated about failing to reply to you.  I should be able to take a look this week, though.
> 
> On 2021-08-23 10:41 a.m., Andrew Burgess wrote:
> > Ping!
> > 
> > Simon, what are your thoughts on this approach?  From what I recall of
> > our discussion on IRC your concerns with my original patch were that
> > after my change the function implementation no longer matched with the
> > function name, this made GDB's internal APIs confusing and
> > inconsistent.
> > 
> > With this patch I've tried to avoid this by renaming the functions to
> > hopefully make the API clearer.
> > 
> > I'm just guessing, but I suspect it was the use of exceptions that
> > Pedro wasn't happy with, so maybe if you're happy with this latest
> > patch we could merge this change, given Pedro already approved the
> > pre-exception patch.
> 
> I don't recall the discussion with Simon, but yes, use custom exceptions
> for this doesn't sound great to me.  Let me take a better look.

Thanks for taking the time to reply.  I completely understand other
pressures getting in the way of being as active on the m/l.  Even just
knowing for sure which part of the patch you didn't like is great
information as I can start coming up with alternatives.  So I'm always
grateful for any feedback, no matter how terse.

Thanks again,
Andrew