Date: Tue, 25 Feb 1997 14:27:26 -0700 (MST) From: Nate Williams <nate@mt.sri.com> To: se@freebsd.org (Stefan Esser) Cc: nik@blueberry.co.uk (Nik Clayton), hackers@freebsd.org Subject: Re: 2.1.7: ncr and de driver increasingly fickle Message-ID: <199702252127.OAA24299@rocky.mt.sri.com> In-Reply-To: <19970225202622.HX00195@x14.mi.uni-koeln.de> References: <19970225164635.47646@coconut.blueberry.co.uk> <19970225202622.HX00195@x14.mi.uni-koeln.de>
next in thread | previous in thread | raw e-mail | index | archive | help
> > On my 2.1.5 system, I never heard a complaint from the de driver, and the > > ncr driver might occasionally complain > > > > /kernel: sd1(ncr0:1:0): extraneous data discarded. > > Well, this is not a problem caused by the driver, IMHO ... > > The number of data transfers requested by the target did not > match the transfer count. This normally indicates SCSI cable > problems, since it means either zero or two bytes have been > sent in a previous bus phase. So why did the old driver in 2.1.5 not report this error? (I only get this error in 2.2, and only during bootup). > For synchronous transfers, you may also receive a gross error > in such situations. It is caused by a FIFO over- or underflow > as a result of a lost or doubled byte ... Why would this only happen when I boot the system, and why would it be new to the system? Old kernels don't report/cause this error. Nate
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199702252127.OAA24299>