[Info-vax] Real time performance, was: Re: VSI: "Official 8.4-1H1 Launch"

[Michael Moroney]

JF Mezei <jfmezei.spamnot at vaxination.ca> writes:

>On 15-06-07 14:50, Jan-Erik Soderholm wrote:

>We're talking about debugging the EFI boot program, and very early VMS
>modules here. Not about debugging a COBOL app that writes pages of reports.

I will say "been there, done that" as far as new code on a new processor,
to a limited extent.  I wrote the boot/diagnostic code for the rtVAX 300,
which was a module consisting of a CVAX+networking+ROM+glue logic.  To 
bring things up for the first time, I debugged individual routines as much 
as I could separately.  This ranges from easy to impossible.  For example,
I had the command interpreter (the familiar ">>>" prompt code) running
as a user mode program on VMS.  I knew the command interpreter would work 
once it made it that far.

For the initial test on the prototype, we had a logic analyzer on the
CVAX's address and data lines.  We saw the address placed on the bus, the
data in response, the next address, data etc.  We compared what we saw to
what we expected from the machine code.  If I recall correctly, there was
a dumb mistake the very first time, and on the second time it got quite 
far.