From owner-freebsd-hackers Wed Jan 14 21:59:11 1998 Return-Path: Received: (from majordom@localhost) by hub.freebsd.org (8.8.8/8.8.8) id VAA28693 for hackers-outgoing; Wed, 14 Jan 1998 21:59:11 -0800 (PST) (envelope-from owner-freebsd-hackers@FreeBSD.ORG) Received: from wcc.wcc.net (wcc.wcc.net [208.6.232.10]) by hub.freebsd.org (8.8.8/8.8.8) with ESMTP id VAA28657; Wed, 14 Jan 1998 21:58:52 -0800 (PST) (envelope-from piquan@wcc.wcc.net) Received: from detlev.UUCP (tnt219.wcc.net [208.10.139.219]) by wcc.wcc.net (8.8.7/8.8.7) with ESMTP id XAA08259; Wed, 14 Jan 1998 23:55:26 -0600 (CST) Received: (from joelh@localhost) by detlev.UUCP (8.8.8/8.8.7) id XAA04566; Wed, 14 Jan 1998 23:58:00 -0600 (CST) (envelope-from joelh) Date: Wed, 14 Jan 1998 23:58:00 -0600 (CST) Message-Id: <199801150558.XAA04566@detlev.UUCP> To: lyndon@ve7tcp.ampr.org CC: john@mailhost.cas.unt.edu, skafte@worldgate.com, freebsd-stable@FreeBSD.ORG, freebsd-hackers@FreeBSD.ORG In-reply-to: <199801140514.WAA11772@ve7tcp.ampr.org> (message from Lyndon Nerenberg on Tue, 13 Jan 1998 22:14:40 -0700) Subject: Re: 3com 3C509B Combo card From: Joel Ray Holveck Reply-to: joelh@gnu.org References: <199801140514.WAA11772@ve7tcp.ampr.org> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk > Okay, I'm stumped. I've run 509B's (and plain old 509's) since they > came out, in a lot of busy production critical servers, running FreeBSD > and just about anything else for x86 that networked. I've found them to > be nothing but rock solid reliable. > Do we know who originated the "buggy" comment in the kernel config? > Would someone like to volunteer to track back through CVS and see > who originated it? Would the originator care to explain the comment? > Personally, I think the whole "buggy" statement is bogus, and has been > for quite some time. Lacking evidence to the contrary the comment should > be squelched before this business of buggy 509's reaches the status > of urban legend :-) I can personally vouch that the 3C509 driver in -current was buggy at the year's beginning. At the beginning of the year, I installed a small LAN in my apartment: one W95 machine with a SVEC FD0290, and one FreeBSD machine with a 509. I never had usable network operations. A tcpdump showed that pings (and other packets) would not show as being returned until the ping process ended. (Simultanious pings would show responses from each when that process ended.) At that point, the appropriate number of ECHO_REPLYs would be generated. The computers operated normally when both were running Windows 95. I switched to an NE2000 clone and it worked fine. I haven't repeated the experiment since the recent changes in if_ep. Happy hacking, joelh -- Joel Ray Holveck - joelh@gnu.org - http://www.wp.com/piquan Fourth law of programming: Anything that can go wrong wi sendmail: segmentation violation - core dumped