[Bug libc/11120] strncmp() will be broken if used in ld.so on x86-64

["dhatch at ilm dot com" <sourceware-bugzilla at sourceware dot org>]

http://sourceware.org/bugzilla/show_bug.cgi?id=11120

--- Comment #11 from Don Hatch <dhatch at ilm dot com> 2013-03-22 22:21:35 UTC ---
(In reply to comment #10)
> (In reply to comment #9)
> > Can you comment on the fact that, prior to your latest change
> > which now prevents calling it altogether,
> > strncmp was returning inconsistent results?
> > This is contrary to what I would expect from Mark Seaborn's original
> > description "If strncmp() is ever used in ld.so on x86-64 it will
> > silently be implemented as strcmp()".
> 
> There is no code in rtld that directly or indirectly calls strncmp.
> The change I committed is fixed safeguards that in case rtld is changed to use
> strncmp, x86-64 build would fail until strncmp is implemented for rtld.
> 
> It's certainly better to let rtld build fail rather than allow strncmp to be
> implemented as strcmp.

Hi Dmitry,

I understand, and I completely agree with you,
as I've tried to say already.
There is absolutely no disagreement or question there.
After your latest change, I observe that any attempt to call strncmp directly,
or indirectly (e.g. by attempting to call getenv), fails at compile time.
That's a very good thing.  Thank you.

However, you have not answered my question.

My question is about the fact that,
contrary to what both you and Mark Seaborn have said,
prior to your change,
strncmp was *not* acting like strcmp, at least not consistently.
On one call it acted like strcmp, on the next call
(with exactly the same effective input) it didn't.
For a simple example of this,
please look carefully at my test case above (in my note of 2013-03-22
02:48:03).

Is this a surprise or not?
My concern is that your disabling of the function inside rtld
may be pushing more mistakes or rottenness under the rug.

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.