[Info-vax] Dave Cutler, Prism, DEC, Microsoft, etc.

[John Reagan]

"Richard B. Gilbert" <rgilbert88 at comcast.net> wrote in message 
news:SpWdncx1SK2tVoLWnZ2dnUVZ_i2dnZ2d at giganews.com...
> Bob Koehler wrote:
>> In article <H4CdnVlBSu7h44PWnZ2dnUVZ_rNi4p2d at giganews.com>, "Richard B. 
>> Gilbert" <rgilbert88 at comcast.net> writes:
>>> To those of us not programming in Macro, the Alpha was just a faster 
>>> VAX!  Much faster!!!
>>
>>    I know there is a lot of junk out there, but all the Macro-32 I had
>>    written before I got my first Alpha compiled much easier than the
>>    Fortran or C.  (There were a few missing features in the first ship
>>    Fortran and C compilers.)
>>
>
> ISTR that a VAX to Alpha port was usually just a matter of compile, link 
> and run.  If your code expected 512 byte pages, you had a problem.  Most 
> of the code I dealt with was written in FORTRAN and didn't know or care 
> what the page size was.  If you did naughty things like bit twiddling 
> floating point numbers you had to adapt your code to the new floating 
> point format.
>
> People who had data files with binary floating point had to convert them. 
> I don't recall it as being a problem although I'm sure some people did 
> have problems.
>

For Macro-32, there was a BUNCH of stuff that works on VAX that wasn't 
supported on Alpha.  The Macro compiler porting guide is chocked full of 
them.  Plus for almost any non-standard interface, you had to start adding 
.CALL_ENTRY, .JSB_ENTRY, and/or .JSB32_ENTRY directives.  Macro-32 certainly 
wasn't recompile and go.

On the other hand, the higher level langauge compilers were designed to be 
recompile and go.  The C issue was that was the same time as the VAX C to 
DEC C language definition.

Binary floating on VAX vs Alpha isn't a big deal since Alpha has F/G 
floating with almost D support minus a few mantissa bits.

John