This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [commit] Rename frame_pc_unwind and frame_unwind_id
> Date: Tue, 15 Jul 2008 15:22:13 -0400
> From: Daniel Jacobowitz <drow@false.org>
>
> On Tue, Jul 15, 2008 at 09:08:34PM +0200, Mark Kettenis wrote:
> > > Date: Tue, 15 Jul 2008 15:01:25 -0400
> > > From: Daniel Jacobowitz <drow@false.org>
> > >
> > > This patch is the first non-trivial change in inlining support, and
> > > readily separable.
> >
> > Non-trivial...
> >
> > > The users of frame_unwind_id and frame_pc_unwind are all either
> > > inferior control, trying to find the caller / return address of a new
> > > function, or trampoline handling. I audited all of the uses, and the
> > > right behavior in every one of them is to ignore any inlined functions
> > > at the current location. A future patch, the one adding inlined
> > > frames, will make the corresponding change to frame_unwind_caller_id
> > > and frame_unwind_caller_pc. For now, I've just renamed them to
> > > indicate the correct expectations.
> >
> > ...so cann't we discuss this first please?
>
> Sorry. Want me to back it out? I'm not going to commit anything
> further; I've just posted the meat of the patch, which is harder to
> separate.
If it's easy for you to back it out, I'd appreciate it. I really
don't think the new names are an improvement, and they are longer
makeing the code slightly less readable...
> I'm interested in your comments (about this patch or the larger one).
...but my main concern is that this diff and the larger one change the
meaning of a frame. It seems it gets us further away of what I
consider to be frame. Please give me a day or so to study the diff a
bit more, before I give a more detailed reaction.
Sorry that I didn't look closely enough at your earlier mails about
the inlining support. I thought they were mainly dealing with the
debug info side of things and didn't really affect the unwinders.