Daniel A. Seagraves <DSEAGRAV@toad.xkl.com> wrote:
> Anyway, concerning the uVAX SCSI/DMA code, seeing as
> 1, the SCSI code works most of the time, and
> 2, retrying when it fails clears the problem (most of the time),
> why not patch the DMA code to retry an operation instead of failing
> and returning SIGSEGV?  Of course, this will get you when something
> actually fails, but could it be tried?
   
   Since this is really NetBSD-specific, you guys will have to do this
yourselves. As I've said all along, NetBSD is not my primary OS of choice
and I can contribute to it only as time permits. Since I am working on
another VAX OS, I have to gather a lot of VAX hardware knowledge, and if by
some chance this knowledge happens to be useful to NetBSD, I am more than
happy to let you guys make use of it. In this case my general (NOT NetBSD-
specific) VAX hardware knowledge was enough for me to spot an obvious (to
me) bug in NetBSD and fix it. I did not need to understand the inner
workings of NetBSD in any detail in order to do this. Implementing Daniel's
suggestion would require knowing these inner workings, and, sorry, given
the work I have to do for the University and the very tough deadlines I
have to face I cannot spend any time learning them.
   
   Sincerely,
   Michael Sokolov
   Phone: 216-368-6888 (Office) 440-449-0299 (Home) 216-217-2579 (Cellular)
   ARPA Internet SMTP mail: msokolov@blackwidow.cwru.edu