From owner-freebsd-hackers Sat Dec 4 13:51:31 1999 Delivered-To: freebsd-hackers@freebsd.org Received: from apollo.backplane.com (apollo.backplane.com [216.240.41.2]) by hub.freebsd.org (Postfix) with ESMTP id A42AF14C1F; Sat, 4 Dec 1999 13:51:27 -0800 (PST) (envelope-from dillon@apollo.backplane.com) Received: (from dillon@localhost) by apollo.backplane.com (8.9.3/8.9.1) id NAA57545; Sat, 4 Dec 1999 13:49:53 -0800 (PST) (envelope-from dillon) Date: Sat, 4 Dec 1999 13:49:53 -0800 (PST) From: Matthew Dillon Message-Id: <199912042149.NAA57545@apollo.backplane.com> To: "Jordan K. Hubbard" Cc: Dennis , Mike Smith , Kris Kennaway , freebsd-hackers@FreeBSD.ORG Subject: Re: PCI DMA lockups in 3.2 (3.3 maybe?) References: <41264.944342473@zippy.cdrom.com> Sender: owner-freebsd-hackers@FreeBSD.ORG Precedence: bulk X-Loop: FreeBSD.ORG : :> There was a time that when someone reported a problem there was interest in :> finding out what it might be. : :Bah, this is a shameless attempt to inflame emotions as a substitute :for having an actual logical point and you know it. Save it for the :presidential debates! : :There is ALWAYS interest in finding out what a problem is when it's :reported in such a way that the effort is worth the potential reward. :Having someone walk up and say, in effect, "Dudes, your system is :broken. Fix it!" is a content-free statement and does not qualify as This is complete bullshit. Read through the thread from day 1... starting around Nov 20th. Dennis started out posting a bug report. A very reasonable bug report whos only fault might be that it was lacking in detail. The very first response was a snide, unhelpful "Not enough information" which did not even explain what type of information was needed, so Dennis's second response, again a very reasonable "what kind of information is needed?" was obvious. The later responses, mainly by existing FreeBSD people, was to essentially roast him over the coals. By the third message the thread turned into an emotional mush, and *NONE* of it was Dennis's doing. Dennis is fully within his rights to become annoyed at that. You people (though not mainly Jordan) essentially roasted him over the coals unrelentingly until he got angry enough to post something a tad over the top and then you people (including Jordan) roasted him for that as well. If you want an example of the *PROPER* way to deal with a bug report containing insufficient information, look at some of my past responses to bug reports related to VM or NFS that contained insufficient information. The proper way to go about it is to say "Well, I can't figure out anything from that, perhaps if you do A, B, and C it may give us enough information to work on". --- Now, with all that said, going back to Dennis's original bug report posted Nov 20th, I would like to note that I had a similar problem with PCI overloading for a few days what, a month ago? When one of the Adaptec drivers was temporarily changed to turn off a special PCI DMA mode which the author had no documentation on, causing the Adaptec to blow up when the 100BaseTX ethernet was loaded down. That particular bug was fixed within a few days, but the author noted that the special PCI mode required to avoid the lockup seemed only to exist in a few adaptec cards and he wasn't sure what the other cards did. But it points to a possible line of direction in regards to locating the current bug. It's possible that another similar bug has been found, and if the conversation had not degraded into a flame war perhaps it would be fixed by now. -Matt To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message