From owner-freebsd-chat Sat Oct 4 19:51:25 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id TAA20962 for chat-outgoing; Sat, 4 Oct 1997 19:51:25 -0700 (PDT) Received: from obie.softweyr.ml.org ([199.104.124.49]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id TAA20955 for ; Sat, 4 Oct 1997 19:51:20 -0700 (PDT) Received: (from wes@localhost) by obie.softweyr.ml.org (8.7.5/8.6.12) id UAA07165; Sat, 4 Oct 1997 20:55:47 -0600 (MDT) Date: Sat, 4 Oct 1997 20:55:47 -0600 (MDT) Message-Id: <199710050255.UAA07165@obie.softweyr.ml.org> From: Wes Peters To: Tom CC: chat@freebsd.org Subject: Re: supermicro p6sns/p6sas In-Reply-To: References: <199709300503.WAA12579@MindBender.serv.net> Sender: owner-freebsd-chat@freebsd.org X-Loop: FreeBSD.org Precedence: bulk tom@sdf.com writes: > Yes, I was just thinking about that. If gcc is randomly dying on older > K6 processors, is it actually producing correct object code when it > doesn't crash? Yes. It is producing correct code when it crashes, too. It just jumps to the wrong address, which quickly leads to a crash. > AMD and Cyrix scare me. Intel's true competition is DEC, Motorola/IBM, > Sparc, and MIPS. And you assume they are perfect? Hah! The only way to trust any silicon is to read the errata sheets in great detail, and then disbelieve 90% of what you read. These chips are more complex than software, and their designers don't have any better idea what's happening inside the chips than you do inside the FreeBSD VM management code. -- "Where am I, and what am I doing in this handbasket?" Wes Peters Softweyr LLC http://www.xmission.com/~softweyr softweyr@xmission.com