Return to BSD News archive
Newsgroups: comp.os.386bsd.development
Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msuinfo!agate!howland.reston.ans.net!europa.eng.gtefsd.com!news.umbc.edu!haven.umd.edu!purdue!mozo.cc.purdue.edu!kidd.vet.purdue.edu!bj
From: bj@kidd.vet.purdue.edu (Ben Jackson)
Subject: Re: FreeBSD 1.1G - SLIP performance problems
Sender: news@mozo.cc.purdue.edu (USENET News)
Message-ID: <Cp15CK.qII@mozo.cc.purdue.edu>
Date: Fri, 29 Apr 1994 16:48:20 GMT
References: <2pqst6$g4s@opine.cs.umass.edu>
Organization: Purdue Daemons, West Lafayette, Indiana
Lines: 27
In article <2pqst6$g4s@opine.cs.umass.edu>,
Jim Doyle <doyle@cs.umass.edu> wrote:
>I've encountered receive-side throughput problems with the SLIP
>implementation in the latest 1.1Gamma FreeBSD release. It appears
>that the interface drops or misses a significant amount of the traffic
>presented to it on the receive side, . There are no throughput or reliability
>problems when the SLIP interface is transmitting data.
I think the "one-way" nature of your problem is the key...
>* The 386PC (386dx40) has a 16550A on the port used for the SLIP interface.
^^^^^^
> It is connected to the PCRoute box useing a hardwired null-modem
> SLIP link. The SLIP link ran at 38400 bps, I did not try 57600.
The 16550A has a FIFO which helps prevent data loss at high transfer
rates. As I understand it, though, there is just *one* FIFO, and you
have to choose which way it buffers (input or output).
>- FTP traffic from the DEC into the FreeBSD box died quickly. The SLIP
> link had alot of dead time, it appeared that traffic was being dropped.
I don't think it would be unreasonable to estimate that the FreeBSD
box is getting 16x the interrupt traffic in the non-buffered direction.
--Ben