> -----Original Message-----
> From: IBM Mainframe Assembler List [mailto:ASSEMBLER-
> [email protected]] On Behalf Of John Gilmore
> Sent: Tuesday, April 16, 2013 12:29 PM
> To: [email protected]
> Subject: Re: TRTE and new instructions
>
> Peter Farley's points are interesting ones.  My numbers tell a very
> different tale, and I suspect that these differences turn on when such
> measurements are taken.
>
> The first appearances of new instructions, millicoded ones anyway, do
> often exhibit 'bad' performance; but this performance sometimes, even
> usually, improves rapidly.
>
> Working with millicoded instructions has taught me two important
> lessons: Their performance is a moving target, and early measurements
> of it are usually misleading.  It often improves significantly in the
> interval that would be required to replace them with alternative
> sequences.

I don't get to work at this level often, but I am always interested.
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".

For the MVC/MVCL option, I can imagine a macro which generates an MVC loop, or 
unroll the loop into a sequence of MVC, or generate the MVCL depending on 
several criteria. I currently don't have the knowledge to determine the 
criteria and I would expect the criteria to change over time.

>
> John Gilmore, Ashland, MA 01721 - USA

Reply via email to