From owner-freebsd-current Sat Feb 8 18:45:57 1997 Return-Path: Received: (from root@localhost) by freefall.freebsd.org (8.8.5/8.8.5) id SAA19863 for current-outgoing; Sat, 8 Feb 1997 18:45:57 -0800 (PST) Received: from sendero.i-connect.net ([206.190.144.100]) by freefall.freebsd.org (8.8.5/8.8.5) with ESMTP id SAA19830 for ; Sat, 8 Feb 1997 18:45:53 -0800 (PST) Received: (from shimon@localhost) by sendero.i-connect.net (8.8.5/8.8.4) id TAA00781; Sat, 8 Feb 1997 19:43:42 -0800 (PST) Message-ID: X-Mailer: XFMail 1.1-alpha [p0] on FreeBSD Content-Type: text/plain; charset=iso-8859-8 Content-Transfer-Encoding: 8bit MIME-Version: 1.0 In-Reply-To: <199702072223.JAA19401@godzilla.zeta.org.au> Date: Sat, 08 Feb 1997 15:06:07 -0800 (PST) Organization: iConnect Corp. From: Simon Shapiro To: Bruce Evans Subject: Re: Pppd Drops Under Heavy Load Cc: freebsd-current@freebsd.org Sender: owner-current@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Hi Bruce Evans; On 07-Feb-97 you wrote: > >a. A USR Sportster 33.6 and a on the MB 16550-like UART running at > > 115,200 baud. (ccua1) > > > >b. Motorola BitSRFR-Pro ISDN on a 16550 card, running at 230,400 (frequency > > doubler; it appears 115,200 to the kernel). (ccua2) > > > >Symptoms: As the Subject says. After a while, under heavy load, the line > >silently drops. Rarely, I get in the log: > > Some buffers are scaled for 115200 bps max. Double RS_IBUFSIZE and TTYHOG > for 230400 bps (the latter is optional and best not done if you only use > kernel ppp). Also, perhaps user mode ppp depends on 115200 really being > 115200. Thanx. I found these in /usr/src/sys/sys/tty.h and /usr/src/sys/i386/isa/cy.c. The way the board provides 230,400 is by using a faster crystal. There is no way for software to know how fast the bits flow in the wire. Although larger buffers are better for faster wires, why the disconnects? There is flow control in the RS-232 and it is actually active. Simon P.S. Neophite question? How do I maintain these deltas on my local tree, so that the next cvs checkout does not wipe them out? In this case it looks as if the total increase in memory footprint will be 1.25KB... but I would like to knowwhat is the best rule to follow. Simon