[Info-vax] VMS port to x86

[Neil Rieck]

On Mar 21, 2:13 am, JF Mezei <jfmezei.spam... at vaxination.ca> wrote:
> Neil Rieck wrote:
> > If OpenVMS is not ported to x86-64 then OpenVMS is toast!
> > If NonStop is not ported to x86-64 then NonStop is toast!
> > If HP-UX is not ported to x86-64 then HP-UX is toast!
>
> Correct. But HP has bet that it can survive by offering migration to
> Linux or Windows on its own servers.
>
> With HP now building 8086 based Superdomes which should be shipping
> soon, HP should stay to unveil its true colours within the next 12
> months. The Oracle documents said that customers would find out about he
> EOL of Itanium in 2012 so that the 5 year period would correspond to the
> end of support by Intel.
>
> My gut tells me that the decision to gut VMS engineering came when HP
> decided to abandon porting efforts. It is a shame because VMS could
> probably be ported to X86 much cheaper and faster than HP-UX (endianness
> issues for HP-UX).
>
> It isn't clear what value HP-UX on x86 would bring that Linux couldn't
> bring.
>
> But VMS and NSK would still bring distinctive features not available on
> Linux.
>
> The big question is really NSK. It would not surprise me if this were
> the only OS to survive the death of IA64. Those are high value and high
> profile servers that, in many cases, control lives.
>
> Another possibility is that HP will have its own distribution of linux
> with proprietary HP add-ons. This would compete against the other Linux
> distributiosn such as Red Hat etc

This comment is a little off topic, but may be worth mentioning. I
recently stumbled across "Compaq's source code for OpenSSL 1.1A" which
appears to have been written in 2003. While this product version was
targeted at Alpha and VAX, the scripts contain bread-crumbs showing
that OpenVMS engineering was playing with an Itanium cross compiler.
No surprise here, they always had a plethora of tools.

Why should anyone care? These people were much better at doing
software than anything else. If someone decided that OpenVMS should be
ported to x86-64 then it would be easy to begin. If HP was smart then
they've been doing this for a while.

###

Oracle has products running on every endian platform so I'm going to
assume "endian"  is no longer the technical obstacle it once was.

According to wikipedia:
http://en.wikipedia.org/wiki/Endianness

These platforms are little endian:
OpenVMS on VAX, Alpha and Itanium
Solaris on x86, x86-64, PowerPC
Tru64 UNIX on Alpha
Windows on x86, x86-64, Alpha, PowerPC, MIPS and Itanium

...so it should be easier to port OpenVMS than HP (which is big endian
on the bi-endian Itanium)

Neil Rieck
Kitchener / Waterloo / Cambridge,
Ontario, Canada.
http://www3.sympatico.ca/n.rieck/