X-pop3-spooler: POP3MAIL 2.1.0 b 4 980420 -bs- Message-ID: <19980908054110.62662@cerebro.laendle> Date: Tue, 8 Sep 1998 05:41:10 +0200 From: Marc Lehmann To: beastium-list AT Desk DOT nl Subject: Re: Bug on P6 identified... not good... Mail-Followup-To: beastium-list AT Desk DOT nl References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: ; from Krzysztof Strasburger on Mon, Sep 07, 1998 at 11:22:00AM +0000 X-Operating-System: Linux version 2.1.120 (root AT cerebro) (gcc version pgcc-2.91.57 19980901 (egcs-1.1 release)) Status: RO Content-Length: 848 Lines: 16 On Mon, Sep 07, 1998 at 11:22:00AM +0000, Krzysztof Strasburger wrote: > I just found the option, which corrupted the code on PPro. > Unfortunately, it is -ffast-math combined with any non-zero it could be a case where rounding actually matters (and -ffast-math does break some rules). pgcc is actually more aggressive than egcs with -ffast-math, so programs that happen to work with egcs might not work with pgcc. -----==- | ----==-- _ | ---==---(_)__ __ ____ __ Marc Lehmann +-- --==---/ / _ \/ // /\ \/ / pcg AT goof DOT com |e| -=====/_/_//_/\_,_/ /_/\_\ --+ The choice of a GNU generation | |