RE: Race in MSCP (ra/rx) driver

[Mark Pizzolato - Info Comm <Mark%infocomm.com@localhost>]

On Wednesday, August 26, 2020 at 1:12 PM, Johnny Billquist wrote:
> On 2020-08-26 22:06, Anders Magnusson wrote:
> > Den 2020-08-26 kl. 21:56, skrev Johnny Billquist:
> >>
> >>> An interesting thing to observe is that a VAX780 simulator booting
> >>> from an
> >>> RM03 simulated device actually boots noticeably faster than the same
> >>> simulator booting from an MSCP device.  The I/O completion delays
> >>> that were needed for the RP device were significantly shorter and
> >>> involved fewer interactions than the MSCP (RQ) case.
> >>
> >> I am in a way a little bit surprised. While the whole interface to a
> >> massbus disk is much simpler than MSCP, it also exposes more of the
> >> hardware, and I would have expected MSCP to essentially not require
> >> any delays between request and completion. One of the nice things of
> >> the whole protocol design actually.
> > The speedup of Massbus  was also true in the "real world". I was
> > dealing quite much with 11/750's, and both RM80 and RP06 were acting
> > significantly faster than RA81.  Probably did not have anything to do
> > with the protocol itself, only slow RA81's :-)
> 
> I wouldn't be surprised if both disks and controller made a difference here.
> The RA81 was certainly not particularly fast, but my recollection of the RP06 is
> not that overwhelming either.
> But the UDA-50 probably added a bunch of overhead as well, which the
> RH750 didn't.
> 
> > What Mark notice though may be because booting from a MSCP disk (on
> > like a 750) initializes the controller which may take a second, but on
> > Massbus disks it is instant.
> 
> I was certainly thinking about that, but then I realized it shouldn't really be
> the case. In simh, the initialization steps of the MSCP should be possible to
> have instantaneous. No reason to emulate that part of the controller, is
> there?

The minimum delays that are default in the simh device simulations (for 
most devices) were those observed to be needed by various OS 
implementations that have been used by the community of simh users or
the original developers (mostly Bob).  The MSCP init sequence is somewhat 
slow.

- Mark