This is the mail archive of the libc-alpha@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

Re: [patch] Fix mallinfo() to accumulate results for all arenas.


On Mon, Sep 5, 2011 at 5:55 PM, Ulrich Drepper <drepper@gmail.com> wrote:
> On Mon, Sep 5, 2011 at 17:43, Paul Pluzhnikov <ppluzhnikov@google.com> wrote:
>> That would appear to be a rather weak argument to keep it broken. If
>> it can be fixed, let's fix it.
>
> No, it's not. ?What is now counted is what people git so far and code expects.

IMHO that's bordering on insane.

libc can provide correct answer, or the wrong answer. You insist that
it must continue to forever give the wrong answer, because some
applications may expect that wrong answer.

But what possible use can an application have for the wrong answer?
Isn't that application broken (by definition), if providing correct
answer breaks it?

By that rethoric, why fix any libc bugs at all? I am sure there is
some application that might break when you fix any given bug.

Thanks,
-- 
Paul Pluzhnikov


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]