Re: [PATCH] Fix Gold/strip discrepancies for PR 11786

[Doug Evans <dje@google.com>]

On Mon, Nov 4, 2013 at 6:34 PM, Tom Tromey <tromey@redhat.com> wrote:
>>> (void)
>
> Doug> We don't have any such style rules for testcases.
> Doug> But ok, done.
>
> Doug> Going forward though, for my own patch reviews of other people's code,
> Doug> what's the story here?
>
> In this case, "()" is not idiomatic C, whereas "(void)" is.
> This is not the same as a coding style rule.

I'm not sure how to read this.

Is this an argument for requiring (void) in all testsuite C function
definitions?
It's ok by me, but it seems to me it's not a requirement today as
there are plenty of existing examples, including recent ones.  OTOH,
if there is such a requirement we'd better get it written down so we
can refer to it when requesting corrections in patches, and so people
can know ahead of time what's expected.  Obviously the rules state
this for gdb itself, but it's been my understanding that these rules
explicitly do not apply to the testsuite, and this understanding has
been affirmed from time to time.  All I'm asking for is clarity and
consistency.

Or is this just a point about a bad use of the word "style"?
By itself "style" is a pretty nondescript word.
Alas I'm not one for always assigning names with precision.
If this isn't a style issue, coding *or* otherwise, let me know what to call it.

If this is something else, let me know that too. :-)