*BSD News Article 92889


Return to BSD News archive

Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.ecn.uoknor.edu!feed1.news.erols.com!cpk-news-hub1.bbnplanet.com!news.bbnplanet.com!news.mindspring.com!jhsterne
From: jhsterne@mindspring.com.nospam (Jason S.)
Newsgroups: comp.unix.bsd.netbsd.misc
Subject: Re: Question about PPP performance
Date: Fri, 04 Apr 1997 21:13:08 -0500
Organization: MindSpring Enterprises
Lines: 32
Message-ID: <jhsterne-ya02408000R0404972113080001@news.mindspring.com>
References: <jhsterne-ya02408000R0304971945300001@news.mindspring.com>
NNTP-Posting-Host: ip96.buffalo3.ny.pub-ip.psi.net
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 8bit
X-Server-Date: 5 Apr 1997 02:08:30 GMT
X-No-Archive: yes
X-Newsreader: Yet Another NewsWatcher 2.4.0
Xref: euryale.cc.adfa.oz.au comp.unix.bsd.netbsd.misc:5757

In article <jhsterne-ya02408000R0304971945300001@news.mindspring.com>,
jhsterne@mindspring.com.nospam (Jason S.) wrote:

> I'm running NetBSD 1.2B on a Mac Performa 450 (aka LCIII - 25 MHz 68030)
> with 20 MB RAM (GENERICSBC-9 kernel).  For the most part, it has done what
> I've wanted it to (including X) but I have a question about the abysmal PPP
> throughput I've seen under NetBSD, namely, that running MacOS 7.5.5 with OT
> 1.1.1, the exact same hardware typically gave me 2.5 - 3.5 kilobytes per
> second on typical ftps, and under NetBSD I see 1 to 1.5 kilobytes per
> second (or less).  I have a Zoom V.34X modem (up to 33.6, although my ISPs
> all max out at 28.8) and have set my port speed to 57600.  The V.34 and DC
> (data compression) lights both light up when I dial up my ISP under NetBSD
> (as well as under MacOS (of course)).


[Replying to my own post - oh gawd!]

I got some helpful advice via email on this one - try a newer kernel.  I
upgraded to GENERICSBC-26 and while I haven't really checked its PPP
performance, it did exorcise one of the more annoying demons from my
machine - the proclivity to continue inverse text past the point when it
was supposed to stop (in, e.g., man pages and lynx).  That makes it worth
the price of admission.

New problem (with GENERICSBC-26):

When I use "ps", I now get "ps: process size mismatch (7824 total, 648
chunks)" and no process list.

Do I need another version of ps, or is this a kernel bug?

J.