From owner-freebsd-questions Wed Oct 25 20:00:08 1995 Return-Path: owner-questions Received: (from root@localhost) by freefall.freebsd.org (8.6.12/8.6.6) id UAA20604 for questions-outgoing; Wed, 25 Oct 1995 20:00:08 -0700 Received: from genesis.atrad.adelaide.edu.au (genesis.atrad.adelaide.edu.au [129.127.96.120]) by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id UAA20575 for ; Wed, 25 Oct 1995 20:00:03 -0700 Received: from msmith@localhost by genesis.atrad.adelaide.edu.au (8.6.9/8.6.9) id MAA09505; Thu, 26 Oct 1995 12:26:37 +0930 From: Michael Smith Message-Id: <199510260256.MAA09505@genesis.atrad.adelaide.edu.au> Subject: Re: Too many UUCP checksum errors! Help! To: raoul@cssc-syd.tansu.com.au (Raoul Golan) Date: Thu, 26 Oct 1995 12:26:36 +0930 (CST) Cc: freebsd-questions@freebsd.org In-Reply-To: <199510260209.MAA03703@kiwi.ind.tansu.com.au> from "Raoul Golan" at Oct 26, 95 12:09:50 pm Content-Type: text Content-Length: 2425 Sender: owner-questions@freebsd.org Precedence: bulk Raoul Golan stands accused of saying: > This is my situation. I'm using 2.0.5-RELEASE, and I'm getting lots > of checksum errors on UUCP over my 16550A UART, mostly due to lost > bytes. I swap cards to a 16450 UART, and the problem disappears. > I tried using the 16550A UART again, this time with the FIFO > disabled in the kernel, but the checksum errors were still there. This is bizarre, but as I've commented previously, I have a local customer that has a similar problem. Characters are lost incoming and outgoing, but only when FIFOs are enabled. > I have a noisy line, so I always enable the modem's error correction. > I don't get a single checksum error over the 16450, but lots of lost > bytes over the 16550A. Are your '550s individual devices, or are they buried in a blob? What brand are they/is the blob? > One more thing: the 16550A is on a VESA slot, wheras my 16450 is > on an ISA slot. I have a DX4-100, i.e. my clock speed is 33, so > I don't really expect the bus speed to be causing any problems > (especially since Linux works at the same speed). I usually The '550s aren't actually accessed across the VESA bus, so this isn't an issue. > connect at 14.4K, but the situation is the same regardless of > the modem speed I connect at. Funny thing is that I can run > Netscape quite happily over the same UART without any problems - > perhaps SLIP handles lost bytes better than the UUCP protocols? > (BTW, I don't run UUCP over tcpip, but through direct dialup) Yes, SLIP will often survive incredible data loss. > It's no big deal for me to use a 16450, just to get FreeBSD UUCP > to run. It sounds like the 16550A driver could be dropping bytes, > but I'm not sure. I'm just interested in what could be going wrong. > I was hoping that this additional information may give someone a > hint as to what's happening. I'm puzzled as to what the problem is, but until I can get one of these malfunctioning devices and mail it all expenses paid to Bruce, I'm not expecting any real answers... > Raoul -- ]] Mike Smith, Software Engineer msmith@atrad.adelaide.edu.au [[ ]] Genesis Software genesis@atrad.adelaide.edu.au [[ ]] High-speed data acquisition and [[ ]] realtime instrument control (ph/fax) +61-8-267-3039 [[ ]] My car has "demand start" -Terry Lambert UNIX: live FreeBSD or die! [[