Date: Fri, 21 Jul 1995 08:17:49 -0400 (EDT) From: Dominic Gregoire <Dominic_Gregoire@QBC.Clic.Net> To: freebsd-scsi@freebsd.org Subject: TMC-8xx driver (sea0) flaky Message-ID: <199507211217.IAA19137@clic1.qbc.clic.net>
next in thread | raw e-mail | index | archive | help
FreeBSD 2.0 used to work to some extent with my TMC-835 scsi board, straight out of the box, but not for long. Repeated timeouts read/writing to a device usually make vm_pageout unhappy after awhile, with a panic next. Seeing that 2.0.5 was out, I decided to try it out (maybe they have fixed that driver?). The situation is now much worst. Probing and finding devices (2 scsi disks) works flawlessly, but as soon as a real io job starts (like newfs), the machine simply crashes with a non-stop 'BEEEEP'. No panics, no reboot, no nothing. Just BEEEP. The only difference I noticed with 2.0.5 is the iosize of 16384, instead of the 8192 of 2.0. The command iosize of the config menu (-c) has no effect whatsoever on this. My tmc is on irq5, iomem 0xca00. If more info is needed to try and find a fix, ask me. I'm not able to program in C, so i can't fix it myself. If anyone of you has some experience with a 835, i'd like to hear from you. This problem is not fixable by inversing 2 bits in the status register, I tried it once, it simply totally breaks it. Thanks for any advices. -- Dominic Gregoire : Acces Publique LLC : Dominic_Gregoire@LLC.org :
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199507211217.IAA19137>