On Friday, 11/07/2003 at 07:32 CST, "McKown, John"
<[EMAIL PROTECTED]> wrote:
> Alan,
> Do you know why IBM did this? Just curious since it seems so strange to
> disable a "coprocessor" function in this manner.

It was a side effect of basing IFLs on Coupling Facility engines.  Part of
the h/w sees them as IFL, the other sees them as CF.  CFs can't have
crypto.

I'm investigating whether there was a change to PCICA attachment to IFLs
and, if so, when it happened.  I.e. if it was model- or machine-specific.

Alan Altmark
Sr. Software Engineer
IBM z/VM Development

Reply via email to