Date: Tue, 7 Mar 2000 10:30:43 +0200 (IST)
From: Eli Zaretskii <eliz AT is DOT elta DOT co DOT il>
X-Sender: eliz AT is
To: Dieter Buerssner <buers AT gmx DOT de>
cc: djgpp AT delorie DOT com
Subject: Re: [long] gcc performance and possible bug
In-Reply-To: <8a1b91$33j7m$1@fu-berlin.de>
Message-ID: <Pine.SUN.3.91.1000307103019.21628J-100000@is>
MIME-Version: 1.0
Content-Type: TEXT/PLAIN; charset=US-ASCII
Reply-To: djgpp AT delorie DOT com
Errors-To: dj-admin AT delorie DOT com
X-Mailing-List: djgpp AT delorie DOT com
X-Unsubscribes-To: listserv AT delorie DOT com
Precedence: bulk


On 6 Mar 2000, Dieter Buerssner wrote:

> You will note, that there sometimes is almost an order of
> magnitude difference between the performance of mwc32 and
> mwc32c. The only difference between these functions is
> the type of the variable mul (static unsigned long vs.
> static const unsigned long). mwc32c is always slower,
> when there is a significant performance difference.

Did you look at the generated assembly?  That could provide important
clues.

>     (null): s=3051870873, used 3.077 CPU seconds 0.02292 usec/call
>     (null): s=3051870873, used 25.934 CPU seconds 0.19322 usec/call
>     ^^^^^^

Doesn't this bomb on plain DOS with CWSDPMI?

Anyway, this is a clear bug which should be reported to gnu.gcc.bug.