From owner-freebsd-hackers Thu Oct 26 16:45:17 1995 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id QAA06534 for hackers-outgoing; Thu, 26 Oct 1995 16:45:17 -0700 Received: from hda.com (hda.com [199.232.40.182]) by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id QAA06505 for ; Thu, 26 Oct 1995 16:45:13 -0700 Received: (from dufault@localhost) by hda.com (8.6.11/8.6.9) id TAA07049; Thu, 26 Oct 1995 19:47:44 -0400 From: Peter Dufault Message-Id: <199510262347.TAA07049@hda.com> Subject: Re: Physically contig memory in user process To: tinguely@plains.nodak.edu (Mark Tinguely) Date: Thu, 26 Oct 1995 19:47:43 -0400 (EDT) Cc: hackers@freebsd.org In-Reply-To: <199510262312.SAA01148@plains.nodak.edu> from "Mark Tinguely" at Oct 26, 95 06:12:26 pm X-Mailer: ELM [version 2.4 PL24] Content-Type: text Content-Length: 554 Sender: owner-hackers@freebsd.org Precedence: bulk > why not allocate the buffers as part of the device's driver and mmap them into > user space instead of allocating them into users space directly? I thought one may not usually care about going through bounce buffers or splitting up the transfer, and that when you did you may as well have a generic way of doing it by mmapping /dev/mem and not having it in the individual drivers. -- Peter Dufault Real Time Machine Control and Simulation HD Associates, Inc. Voice: 508 433 6936 dufault@hda.com Fax: 508 433 5267