Re: Performance tweaks

From: Jamie Lokier (lk@tantalophile.demon.co.uk)
Date: Sun Feb 27 2000 - 16:24:39 EST


D. Hugh Redelmeier wrote (about C):
> If you define a variable to be const, it must not be changed by the
> program. If, however, you have a pointer-to-const, it is possible
> that the object pointed to may change due to other parts of the
> program having non-const access.
>
> In other words, a pointer-to-const can point to something that isn't
> const. But something that is const must not be changed.
>
> So, in the quoted example, gcc could assume (and even enforce) that
> the variable isn't changed during its lifetime.

You are right. I just tried some examples with GCC 2.95.1 using C and
C++, and they both inline the value of a const int, if it is defined.

I am fairly confident that when I read the first edition of Stroustrup's
book, the use of "const int" to declare constants was touted as a C++
feature, as a key difference from C which required preprocessor macros
to get equivalent results.

So I'm puzzled.

-- Jamie

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/



This archive was generated by hypermail 2b29 : Tue Feb 29 2000 - 21:00:18 EST