From owner-cvs-all Sun Jan 25 12:32:03 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id MAA16329 for cvs-all-outgoing; Sun, 25 Jan 1998 12:32:03 -0800 (PST) (envelope-from owner-cvs-all@FreeBSD.ORG) Received: from dyson.iquest.net (dyson.iquest.net [198.70.144.127]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id MAA16310; Sun, 25 Jan 1998 12:31:59 -0800 (PST) (envelope-from toor@dyson.iquest.net) Received: (from root@localhost) by dyson.iquest.net (8.8.8/8.8.8) id PAA00343; Sun, 25 Jan 1998 15:31:43 -0500 (EST) (envelope-from toor) From: "John S. Dyson" Message-Id: <199801252031.PAA00343@dyson.iquest.net> Subject: Re: cvs commit: src/sys/i386/i386 machdep.c src/sys/i386/include param.h src/sys/i386/isa wd.c src/sys/kern kern_physio. In-Reply-To: <19980125130216.61126@klemm.gtn.com> from Andreas Klemm at "Jan 25, 98 01:02:16 pm" To: andreas@klemm.gtn.com (Andreas Klemm) Date: Sun, 25 Jan 1998 15:31:43 -0500 (EST) Cc: toor@dyson.iquest.net, wpaul@FreeBSD.ORG, dyson@FreeBSD.ORG, committers@hub.freebsd.org X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-cvs-all@FreeBSD.ORG Precedence: bulk Andreas Klemm said: > > Would be great, if this change could be isolated in a manner, so > that it could be merged into stable. So we would be able to read > dumps from other architectures (Solaris 2,...) that doesn't have > such a limit ... If I remember right, it's not unusual > praxis, to dump with 96, 126 and 128 blocks ( <= 64KB) and we > currently had <=32 KB. > The major complication is that the various drivers have to be able to pass the maximum size back to the upper level physio code. It isn't even done in -current (I only did an example implementation for IDE, and the mechanism is going to change anyway.) If someone upgrades the various SCSI drivers in -current, I'll probably end up modifying the implementation to the eventual (architecturally clean) design. The implementation in -current is only to create some interest among the other driver developers to pass the info upwards, so I can use it. Then, we'll probably create a more complete implementation. At least, now, IDE HD's can maximize their transfer rates. The hooks just do not exist in 2.2.X and are only evolving in 3.0. To do correctly, it would require touching alot of (probably about 10-20) files. I think that it is best to be risk-adverse in -stable. You definitely don't want me hacking on -stable (and I really do mean it as a threat!!! :-)). -- John | Never try to teach a pig to sing, dyson@freebsd.org | it just makes you look stupid, jdyson@nc.com | and it irritates the pig.