On Mon, 2010-08-23 at 18:38 -0600, Paul Gilmartin wrote:
> On 08/23/10 18:02, John McKown wrote:
> >
> > Thanks to the way that z/OS works, there is no need to initialize the
> > table. Do the STORAGE OBTAIN and use BNDRY=PAGE. When a never before
> > used page is first referenced, a free frame is assigned and initialized
>
> Is this specified in TFM?  Even if you've previously done
> a FREEMAIN (whatever it's called this month)?.  I suppose
> it depends on what the meaning of "first" is (apologies
> to WJC?)
>
> -- gil

Yes.

http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/IEA2A9B0/92.2

<quote>
Note: When you obtain storage, the system clears the requested storage
to zeros if you obtain either:

      * 8192 bytes or more from a pageable, private storage subpool

      * 4096 bytes or more from a pageable, private storage subpool,
        with BNDRY=PAGE specified.

        In all other cases, you must not assume that the storage is
        cleared to zeros.

</quote>

OK, it doesn't really say that a new frame is initially allocated. I got
that here by "reading between the lines":

http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/iea2a6a0/11.1

<quote>
To explicitly request virtual storage below the 2 gigabyte address,
issue a GETMAIN or a STORAGE macro. When you make an explicit request,
the system satisfies the request by allocating a part of the virtual
storage area reserved for the job step. The virtual storage area is
usually not set to zero when allocated. (The system sets storage to zero
only when it  initially assigns a frame to a virtual storage page.)
</quote>

Reply via email to