Dave Gibney wrote:

<begin extract>
How can Millicode be faster than the equivalent using the hardware instructions?
As I understand Millicode, that is really all it is (using the
hardware instructions) plus any overhead in context switching to the
Millicode "environment".
</end extract>

This is a common misunderstanding that has unfortunately been repeated
many times.  It is a radically misleading caricature.  Millicode makes
available many facilities not available in the HLASM.  It does not
make additional machine instructions available, but it does make its
own powerful facilities for specifying the path pf control among them
available.

I have always felt some impatience with this view.  If it were at all
accurate it would make millicode, which goes back to the System/390,
unimportant, even dispensable; and, while IBM is not infallible, it is
deeply serious about its hardware investments.


GIYF.  To begin see (watch wrap)

http://ecc.marist.edu/conf2011/materials/SlegelSystemZ_APeekUnderTheHood_Slegel_MaristECC.pdf.

John Gilmore, Ashland, MA 01721 - USA

Reply via email to