This is the mail archive of the
gdb-patches@sourceware.org
mailing list for the GDB project.
Re: [PATCH] Fix Gold/strip discrepancies for PR 11786
- From: Doug Evans <dje at google dot com>
- To: Jan Kratochvil <jan dot kratochvil at redhat dot com>
- Cc: Tom Tromey <tromey at redhat dot com>, gdb-patches <gdb-patches at sourceware dot org>, Cary Coutant <ccoutant at google dot com>
- Date: Tue, 5 Nov 2013 09:56:29 -0800
- Subject: Re: [PATCH] Fix Gold/strip discrepancies for PR 11786
- Authentication-results: sourceware.org; auth=none
- References: <yjt24n85x78h dot fsf at ruffy dot mtv dot corp dot google dot com> <20131031154957 dot GA11260 at host2 dot jankratochvil dot net> <CADPb22QKBpYpmmZzeKJy7JWukpfkTQcYZDm+KeEkr6K_92LJ2A at mail dot gmail dot com> <87li13shk2 dot fsf at fleche dot redhat dot com> <CADPb22QNaGzvagsDwgt2mAVOQw9kQxtKbnHKtnTbUMy-7xaJhw at mail dot gmail dot com> <20131105172219 dot GA21529 at host2 dot jankratochvil dot net>
On Tue, Nov 5, 2013 at 9:22 AM, Jan Kratochvil
<jan.kratochvil@redhat.com> wrote:
> On Tue, 05 Nov 2013 18:04:38 +0100, Doug Evans wrote:
>> I'm not sure how to read this.
>
> Primarily I do not understand if you really defend "()". "()" is wrong and
> obsolete syntax. Reason is that it does not check against passed parameters,
> making bugs more difficult to catch and therefore the code expensive to
> maintain.
All I'm asking for is clarity and consistency.
[I'm not asking for perfection, but IWBN to make sure we're all on the
same page.]
> The real bug is that gdb/testsuite/ does not use the same CFLAGS like gdb/
> does, including -Wmissing-prototypes.
>
> It would be nice to enable it there one day. But nowadays nobody is going to
> fix all the gdb/testsuite/ sources to make them compliant. But it is not
> right to (1) make the testsuite code more expensive to maintain,
> (2) needlessly different from the main GDB codebase code, (3) making the
> future work of enabling -Wmissing-prototypes for gdb/testsuite/ more expensive
> (if it ever happens).
New tests go in all the time that are needlessly different from GDB.
>> It's ok by me, but it seems to me it's not a requirement today as
>> there are plenty of existing examples,
>
> Any existing code should be irrelevant, existing GDB code base is in a worse
> state than what should be required for new commits.
If the decision is to be more strict with the rules for testcases
that's fine by me.
Let's write it down, then discussions like these will become a *lot* shorter.
>> including recent ones.
>
> This is worse, I am aware of it. Just I do not want to spend more time
> catching such nitpicks when GDB has in several orders of magnitude more
> serious problems.
We spend time pointing out rule violations all the time.
Why is this any different?
Point (3) above says don't make the problem worse, but if I hadn't
raised this issue I'm quite sure tests will continue to go in that
violate any or all of 1,2,3.
> Discussing such a clear thing like that "()" is forbidden
> seems also as a loss of time to me. C++ even does not know K&R prototype
> anymore so it will not even be possible in GDB anymore in some time,
> hopefully.
We also spend time discussing things from time to time. Why should
this be any different?