From owner-freebsd-hackers Tue Jun 3 23:25:55 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.5/8.8.5) id XAA28510 for hackers-outgoing; Tue, 3 Jun 1997 23:25:55 -0700 (PDT) Received: from ethanol.gnu.ai.mit.edu (joelh@ethanol.gnu.ai.mit.edu [128.52.46.64]) by hub.freebsd.org (8.8.5/8.8.5) with SMTP id XAA28496 for ; Tue, 3 Jun 1997 23:25:51 -0700 (PDT) Received: by ethanol.gnu.ai.mit.edu (8.6.12/8.6.12GNU) id CAA13877; Wed, 4 Jun 1997 02:25:35 -0400 Message-Id: <199706040625.CAA13877@ethanol.gnu.ai.mit.edu> From: Joel Ray Holveck Date: Wed, 4 Jun 1997 02:18:31 -0400 To: freebsd-hackers@freebsd.org Subject: aha1542 changes Reply-to: joelh@gnu.ai.mit.edu Sender: owner-hackers@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Recently, I built a new kernel for a old system (I think it was the 960612 2.2 snap), and added the TUNE_1542 option. After that, I've been getting a couple of bus errors from aha0 (about one every two hours sitting idle). (There were other changes to the kernel, so I'm still experimenting to see what's gone wrong, and what follows is part of how I'm going about these experiments.) I hacked in some changes to aha1542.c to add a new option, TUNE_1542_MARGIN. As we all know, the TUNE_1542 option backs off one notch from the last working speed. This sounds just fine, but concievably might (to my mind) not work for highly marginal cases. The new option, TUNE_1542_MARGIN, specifies how many notches to back off (instead of the default of one notch). I checked the -current aha1542.c, and the function that I hacked (aha_set_bus_speed) hasn't changed, so I made a set of diffs for the -current aha1542. Now, assuming that this solves all my problems and restores harmony and peace (as opposed to hard drive pieces), would anybody be interested in these diffs? Happy hacking, joelh PS: I hate to read what I'm about to say, but cope. I don't subscribe to -hackers, so please keep me in the cc: list. -- http://www.wp.com/piquan --- Joel Ray Holveck --- joelh@gnu.ai.mit.edu All my opinions are my own, not the Free Software Foundation's. Second law of programming: Anything that can go wrong wi sendmail: segmentation violation -- core dumped