Content-Type: text/plain; charset=US-ASCII
From: Denis Vlasenko <vda@port.imtp.ilyichevsk.odessa.ua>
To: William Lee Irwin III <wli@holomorphy.com>, Matt Mackall <mpm@selenic.com>
Subject: Re: [1/2] use const in time.h unit conversion functions
Date: Thu, 3 Jun 2004 11:17:09 +0300
Cc: Andrew Morton <akpm@osdl.org>, linux-kernel@vger.kernel.org,
       suparna@in.ibm.com, linux-aio@kvack.org
References: <20040601021539.413a7ad7.akpm@osdl.org> <20040602184335.GC5414@waste.org> <20040602185205.GX21007@holomorphy.com>
In-Reply-To: <20040602185205.GX21007@holomorphy.com>
MIME-Version: 1.0
Content-Transfer-Encoding: 7BIT
Message-Id: <200406031117.09678.vda@port.imtp.ilyichevsk.odessa.ua>
Sender: linux-kernel-owner@vger.kernel.org
Content-Length: 945
Lines: 34

On Wednesday 02 June 2004 21:52, William Lee Irwin III wrote:
> On Wed, Jun 02, 2004 at 01:43:35PM -0500, Matt Mackall wrote:
> > This is the second const-correctness patch I've seen in a couple days,
> > and I'd like to point out that while it's a noble cause, retrofitting
> > const decls onto interfaces is notorious for causing ripple effects in
> > APIs.
>
> There's a point to this one. A warning got tripped when const stuff was
> passed to it in patch #2, hence this as a preparatory cleanup.

If you get warnings in this:

const int c=5;
void f(int i);
void g() {
    f(c);
}

then you see compiler bug. Warning is ok in

const int c=5;
void f(int* i);
void g() {
    f(&c);
}

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