From owner-freebsd-hackers Wed May 30 8:21: 5 2001 Delivered-To: freebsd-hackers@freebsd.org Received: from silby.com (cb34181-a.mdsn1.wi.home.com [24.14.173.39]) by hub.freebsd.org (Postfix) with ESMTP id D5C0D37B423 for ; Wed, 30 May 2001 08:21:02 -0700 (PDT) (envelope-from silby@silby.com) Received: (qmail 71499 invoked by uid 1000); 30 May 2001 15:21:01 -0000 Received: from localhost (sendmail-bs@127.0.0.1) by localhost with SMTP; 30 May 2001 15:21:01 -0000 Date: Wed, 30 May 2001 10:21:01 -0500 (CDT) From: Mike Silbersack To: Terry Lambert Cc: Terry Lambert , Subject: Re: general speed differences between 4.1.1-RELEASE and 4.3-RELEASE In-Reply-To: <3B14D2AF.47CD9ECB@mindspring.com> Message-ID: <20010530101415.D71465-100000@achilles.silby.com> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG On Wed, 30 May 2001, Terry Lambert wrote: > Mike Silbersack wrote: > > 1. Have the ata driver leave the write cache setting > > alone by default, providing a sysctl which can cause > > disabled or enabled if requested. When the default is > > allowed, put something in dmesg which says "Note: Write > > caching may be enabled. See ata(4) for the reliability > > implications of this." > > You need to look at the code; it would be relatively hard > to make this runtime tunable instead of boot-time tunable. I didn't intend to imply that it needed to be runtime tuneable. And of course, it could be runtime tuneable if desired. There's nothing in the spec which says it shouldn't be. > > 2. Leave the default to be 0, but instead print "Please > > see ata(4) to choose your write cache preferences." Also, > > put something in the release notes or src/updating which > > mentions this. > > I think the answer is "reliable by default". > > As a friend of mine says "I can make it go as fast as you > want, if it doesn't have to work"... > > > -- Terry You entirely missed my point. Yes, we could leave it at 0. But if so, we should tell people so that they can make an informed choice. If we don't make the choice obvious, people are going to continue to be alarmed and confused as to why their Linux boxes now run circles around their FreeBSD boxes. This speed difference would be ok _if they knew why_. But they don't, at present. Mike "Silby" Silbersack To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message