From owner-freebsd-questions Tue Jan 31 12:41:02 1995 Return-Path: questions-owner Received: (from root@localhost) by freefall.cdrom.com (8.6.9/8.6.6) id MAA28388 for questions-outgoing; Tue, 31 Jan 1995 12:41:02 -0800 Received: from hermes.cybernetics.net (hermes.cybernetics.net [198.80.51.103]) by freefall.cdrom.com (8.6.9/8.6.6) with ESMTP id MAA28379 for ; Tue, 31 Jan 1995 12:40:57 -0800 Received: (from james@localhost) by hermes.cybernetics.net (8.6.8/8.6.6) id PAA01420; Tue, 31 Jan 1995 15:50:44 -0500 From: James Robinson Message-Id: <199501312050.PAA01420@hermes.cybernetics.net> Subject: Re: Discouraged... To: nelson@seahunt.imat.com Date: Tue, 31 Jan 1995 15:50:43 -0500 (EST) Cc: questions@FreeBSD.org In-Reply-To: <199501311704.JAA00408@seahunt.imat.com> from "Michael Nelson" at Jan 31, 95 09:04:52 am X-Mailer: ELM [version 2.4 PL23] Content-Type: text Content-Length: 3361 Sender: questions-owner@FreeBSD.org Precedence: bulk One would think that such a high caliber computer professional would approach a volenteer group with a little less of an inflammatory attitude. Remember -- no one made you do this, an nothing makes us want to help aside from our good nature. You did not purchase a product from the FreeBSD team (whom I do not represent -- I am just a happy user). Now lets see if we can get some pertinent info and try and fix your problems :-) . > > Machine: AMI 486/33 > Ram: 16MB > SCSI: 2 ea CDC 320MB drives on an Adaptec AHA-1740 > 1 ea Archive Viper 150 Tape > 1 ea Toshiba 3101 CDROM > Ethernet: WD (SMC) 8013 > Video: ATI Graphics Ultra DX or SX chip? > Secondly, I had gotten disenchanted with some of the little niggling > problems with Linux's networking code, and had heard that FreeBSD had the > best networking code of all, bar none. I am certainly NOT finding that > to be the case! For _ME_, on _MY_ machine, FreeBSD's networking code has > been very disappointing and troublesome. "niggling"? Nice phrase there. > > EXAMPLE - Networking > -------------------- > [sup deleted] Can't help with that one -- don't run sup, as that I am still running 1.1.5.1, the last really stable release. 2.0R and the two snaps do seem to have problems, which 2.1 is looking to fix by around the middle of March. > > Using ncftp, I get about the same transfer speeds from that site, > although other sites seem to be OK (1.5KB/sec... we're on a SLIP). Ping > times and traceroutes look normal between here and FreeBSD.ORG. I have > no idea what's wrong. What speed -- 16 cps ? Are you saying that you get 1.5K/s from your FreeBSD machine to other machines in the real world, but really slow ones to freefall.cdrom.com? Or do you get 1.5K/s from a peer Linux machine to the real world? Can you capture some documentary output describing your situation, as well as your network setup? > > And, at the same time, I can telnet to my roommate's Linux machine and > get very good transfer speeds using ncftp from the same site, through the > same slip connection. So I doubt it's network connections between our > slip and FreeBSD.ORG. Ok -- what is acting as the gateway machine? What does the FreeBSD box's routing table look like? Do you get good pure ethernet transfer times? > > Example - HARDWARE SUPPORT > -------------------------- > > Toshiba 3101 CDROM Drive: [...] > describing this problem in any of the FAQs, which I read prior to > switching over to FreeBSD. A newly developed documentation team has taken over the FAQ writing job, but it is easier to write about the features that we have, as opposed to the ones that we don't :). Suggestion noted, though. The reason that the ioctls aren't there is because no one has written them. Not a crime, really. > > EXAMPLE - XWindows > ------------------ > > Common programs such as "xload" periodically blow up and exit with a > floating point exception error. I've been through the xload code with a > fine tooth comb, and finally got it to compile absolutely clean with the > gcc -Wall switch, but it STILL blows up. I don't think the problem is > with xload, but with FreeBSD. Do you have a math co-processor? Sounds like a math emulation problem -- try configuring a kernel with the options GPL_MATH_EMULATE as opposed to options MATH_EMULATE Or installing a DX chip.