From owner-freebsd-hackers Sat Dec 2 15:28:11 1995 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id PAA14816 for hackers-outgoing; Sat, 2 Dec 1995 15:28:11 -0800 Received: from phaeton.artisoft.com (phaeton.Artisoft.COM [198.17.250.211]) by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id PAA14684 for ; Sat, 2 Dec 1995 15:28:02 -0800 Received: (from terry@localhost) by phaeton.artisoft.com (8.6.11/8.6.9) id QAA06602; Sat, 2 Dec 1995 16:25:00 -0700 From: Terry Lambert Message-Id: <199512022325.QAA06602@phaeton.artisoft.com> Subject: Re: ft (Qic 40/80) To: trangmar@gnsnet.com (Robert Trangmar) Date: Sat, 2 Dec 1995 16:25:00 -0700 (MST) Cc: hackers@FreeBSD.ORG In-Reply-To: from "Robert Trangmar" at Dec 2, 95 00:59:15 am X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1051 Sender: owner-hackers@FreeBSD.ORG Precedence: bulk > Does anyone know why this this wasn't written as a *complete* tape driver? > It would be really useful for me to have this as a device i could use > dump and tar on and i am considering looking further into it with a view > to enhancing it to do just that. > Just thought i'd cast around to see if anyone had any input before i > expended/wasted valuable time. Yeah. Even with double buffering to allow non-streaming access to resync correctly, you have only a 200uS window in which to respond. Process quantum @ 100HZ is 10mS, 50 times to large to guarantee the thing will operate correctly on a loaded system without some higher resoloution kernel timers and scheduling mechanisms in the kernel itself. Then either the timer has to be handled at interrupt level, or it has to add a preeemption point for involuntary preemption of the process that happens to be running when it needs to do its thing. Terry Lambert terry@lambert.org --- Any opinions in this posting are my own and not those of my present or previous employers.