From owner-freebsd-current Tue Jun 13 11:36:21 1995 Return-Path: current-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id LAA02443 for current-outgoing; Tue, 13 Jun 1995 11:36:21 -0700 Received: from cs.weber.edu (cs.weber.edu [137.190.16.16]) by freefall.cdrom.com (8.6.10/8.6.6) with SMTP id LAA02437 for ; Tue, 13 Jun 1995 11:36:20 -0700 Received: by cs.weber.edu (4.1/SMI-4.1.1) id AA05399; Tue, 13 Jun 95 12:29:04 MDT From: terry@cs.weber.edu (Terry Lambert) Message-Id: <9506131829.AA05399@cs.weber.edu> Subject: Re: GENERIC kernel & some basic UNIX pointers To: wollman@halloran-eldar.lcs.mit.edu (Garrett Wollman) Date: Tue, 13 Jun 95 12:29:03 MDT Cc: bde@zeta.org.au, current@FreeBSD.ORG In-Reply-To: <9506131626.AA01992@halloran-eldar.lcs.mit.edu> from "Garrett Wollman" at Jun 13, 95 12:26:01 pm X-Mailer: ELM [version 2.4dev PL52] Sender: current-owner@FreeBSD.ORG Precedence: bulk > > The context switches don't stop zmodem from being faster than the kernel ppp > > even when the packet sizes are large. > > Zmodem doesn't have to do any context switches in order to operate. > IIJ-PPP by definition does. Kernel PPP does not. Zmodem is also unidirectional sliding window and doesn't have the normal request/response latency of a TCP connection over PPP. Actually, we are starting more and more to need an async I/O mechanism to cause kernel calls to be non-blocking to avoid context switches and latency as a result of blocking operations. Terry Lambert terry@cs.weber.edu --- Any opinions in this posting are my own and not those of my present or previous employers.