On Monday, 12/15/2003 at 08:40 PST, "Fargusson.Alan"
<[EMAIL PROTECTED]> wrote:

> The only thing I can think of that is related to reliability is that a
MOVCL
> will not work for multiprocessor synchronization since it will not
prevent two
> processor from modifying the same memory at the same time, while the CAS
will.

But that has nothing to do with the reliability of the architecture.
Making incorrect assumptions about the behavior of instructions could make
your *program* unreliable.

CS does what it is documented to do, as does MVCL.  CS explicitly says
that a serialization operation (defined in Chapter 5 of the PoP) is
performed.  MVCL is documented to be a multiple-storage-access
instruction, which is discussed in Chapter 5 also.

Apples and oranges.

Alan Altmark
Sr. Software Engineer
IBM z/VM Development

Reply via email to