On Wed, Oct 23, 2002 at 06:58:41PM -0400, Greg A. Woods wrote:
> Has there been enough experience yet to know if the more recent fixes to
> esp(4) (i.e. sys/dev/ic/ncr53c9x.c et al) are well enough tested to be
> pulled up to the netbsd-1-6 branch yet?  (this goes along with my post
> from last Friday asking if the -current scsictl(8) can be pulled up too)
> 
> I'm getting tired of having one of my disks go offline and not being
> able to reset the bus (yes I should fix/replace the disk, but that's yet
> another issue :-).  It would seem from what I read in the code and CVS
> that it's now possible to reset the bus, and indeed the driver may even
> automatically reset the bus after some errors too.
> 
> I can try copying the new driver (and scsictl) files over to my local
> working directory and building a test kernel with them if there's still
> need to test the changes -- just let me know if this still needs to be
> done.
> 

The changes I made for the driver were targeted and tested on fas. It's
quite possible that they might help to deal with device initiated reset
in other esp-variants as well, but I don't know, never had a chance to
try. So yes, it needs to be done. I think 1.6 scsictl is 'good enough'
to issue scsibus reset or scan, so you might not need to pull that from
current. 

	Andrey