21 Jul
2001
21 Jul
'01
4:05 p.m.
Ethan Dicks wrote:
Hi Jerome,
have you tried moving your disk controllers closer to the
CPU? moving them up the bus grant chain will help. Also you can tell the
RQD11-EC to limit its DMA burst size which allows other boards to get in a
bus cycle when they need to. I had a problem where adding a SCSI controller
to a VAX killed disk performance on some ESDI drives but it was entirely
due to the SCSI controller doing large DMAs that interfered with the time
available for the ESDI drives to read or write buffers.
Was that a controller for
a TLZ04? We had problems with customers who tried
to put a COMBOARD and a TLZ04 in the same VAX 4000 - it would lock up the
microcode because of grant response issues. The TLZ04 card would grab the
bus and our card would give up, but the CPU would try to give us a grant
eventually, anyway. It wedged the computer so hard they had to power it
off to unjam the bus. Apparently, the combination of a 1985 Qbus design
that had never had to deal with a real timeout and a 1991 Qbus implementation
didn't mix.