[Info-vax] VSI OpenVMS Alpha V8.4-2L1 Layered Product Subset Available

[Jan-Erik Soderholm]

Den 2017-07-10 kl. 21:22, skrev John Reagan:
> On Monday, July 10, 2017 at 2:09:56 PM UTC-4, clair... at vmssoftware.com
> wrote:
>> On Monday, July 10, 2017 at 1:34:29 PM UTC-4, Jan-Erik Soderholm
>> wrote:
>>> Den 2017-07-09 kl. 22:45, skrev Jan-Erik Soderholm:
>>> 
>>>> 
>>>> Note (and as the subject of this thread says) that all year it
>>>> has been talks about a 8.4-2L1 Alpha version. More or less a
>>>> recompile of the current 8.4-2L1 for Itanium. I noticed now that
>>>> what was downloaded earlier today was ALpha 8.4-2L2. The cover
>>>> letter (dated June-2017) says:
>>>> 
>>> 
>>> Now 2L2 is also official on the VSI site:
>>> 
>>> http://www.vmssoftware.com/index.html
>>> 
>>> http://www.prweb.com/releases/2017/07/prweb14487932.htm
>>> 
>>> "Depending on the specific CPU and operating system features a
>>> customer’s application uses, the new Alpha V8.4-2L2 release could
>>> render performance improvements of between 15% and 50%. [...]", said
>>> Eddie Orcutt
>> 
>> Those are real numbers from some tests we ran. Here is my simplistic
>> way to look at it. Every execlet is 5%-10% smaller and RMS is 15%
>> smaller, compared to unoptimized (the standard build). That means
>> many, many code paths in the heart of the operating system are shorter
>> with the use of byte/word instructions. You can certainly make a more
>> precise analysis but that was from my quick look at comparing the
>> result disks from the 2L1 and 2L2 builds, picked a bunch of routines
>> and looked at the generated code.
> 
> Not to nitpick, but to avoid confusion.  The standard build is NOT
> unoptimized.  The build uses the default of /OPT=LEVEL=4 for all
> compiles.  However, the default /ARCH value is EV4 (which also sets the
> default for /OPT=TUNE).  We added /ARCH=EV6 to the compilations.  We did
> not add/remove/change any /OPT qualifiers.
> 

I guess that /ARCH=67 would not have made any major difference... :-)

Thanks fr the details. Sometimes one here about weird things happening
when the /OPT level is changed like things working OK when running
with /NOOPT (to be able to debug) but not OK when run with /OPT.
So it seems good that that part wasn't changed.

I will certenly run some simple test to see if I can notice
some difference after the 2L2 upgrade.

And, it does sound as a nice gain from that simple change.

Even if Alpha is not *that* forward looking I'm pleased to
see that VSI cares for the Alpha customers.

Now back to preparing for the upgrade...