From owner-freebsd-hackers Sun Feb 18 12:30:13 1996 Return-Path: owner-hackers Received: (from root@localhost) by freefall.freebsd.org (8.7.3/8.7.3) id MAA02102 for hackers-outgoing; Sun, 18 Feb 1996 12:30:13 -0800 (PST) Received: from ovid.com (adamp@psyche.ovid.com [198.242.117.3]) by freefall.freebsd.org (8.7.3/8.7.3) with SMTP id MAA02097 for ; Sun, 18 Feb 1996 12:30:10 -0800 (PST) Received: by ovid.com (AIX 3.2/UCB 5.64/3.1.012693-Ovid Technologies Inc.); id AA122105 for hackers@freebsd.org; Sun, 18 Feb 1996 15:23:12 -0500 Date: Sun, 18 Feb 1996 15:23:12 -0500 (EST) From: Adam Prato To: "Justin T. Gibbs" Cc: jkh@cdrom.com, hackers@freebsd.org, nirva@zynet.com Subject: Re: BT946C problems. Can you help? In-Reply-To: <199602181747.JAA24128@freefall.freebsd.org> Message-Id: Mime-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII Sender: owner-hackers@freebsd.org Precedence: bulk On Sun, 18 Feb 1996, Justin T. Gibbs wrote: > Disable extended translation on the new Buslogic card. These errors are > always related to geometry differences, so if disabling the extended > translation (support for DOS drives > 1 gig option) doesn't work, you > may have to play some other games to get the geometry right so you ca > boot off of it. I have a thread of messages from JKH about this as well. This really sounds like a strong possibility. However, if this is the case, why do I have the *same* error with another drive (let me elaborate). I have an HP 97549 (T100) that I had *no* idea whether or not was dead or not. At first I just plugged it in to see what the boot disk would say. It gave the *same* error. > It could be that the Buslogic is attempting to share an interrupt with > another PCI card... assuming you have another PCI card in the system. > Until 2.1-stable and 2.2-current, shared PCI interrupts were not supported > for the bt driver. It may also be a dma channel conflict. I would > suggest trying the driver in 2.1-stable if you can create boot media > from it or can get a system up enough on your 1542B so you can genereate > new kernels. With the 2.1-stable driver, you can disable all of the > ISA compatibility features of the 946, and have the PCI BIOS do all > of the setup for you (ie ISA dma channel/port address, etc). All I have in the machine at this point is video and scsi card. The floppy is built into the motherboard. The motherboard also has lpt1,com1,com2 build in as well. I'll check this out. I tried changing PCI slots, different/NO dma on the bt946, I turned on/off >1GB support (my two hd's are 1.05 and 1.18GB drives). > > It could also just be that the I/O address, dma channel and IRQ defaulted > to in the GENERIC 2.1R kernel are not appropriate for your setup. You > might want to just try booting with '-c' and adjusting how the probe > is performed. the kernel accurately reports the settings of the card when it boots up. Every time I change them, the kernel reports the changes. I will try and set the kernel up with the proper settings when it boots. Another reason why I feel it is some weird kernel/driver problem is because the linux boot/root disks work fine (is this a sign?). Also, with the second HP, I have the same error, and I recently lowleveled it, so now its totally blank. How would I be able to reinstall freebsd if I wanted to? I have a blank HD and don't have any geometry issues to really worry about, yet the boot.flp floppy will not let me boot passed the bt0 driver. Thanks for the response, I'll try these ideas right now. adam