On 2014-07-10 17:38, George Kozakos wrote:
>> ... showing mostly that STCKCONV (as of z/OS 1.13) has not yet
>> implemented the (proposed) use of the high byte of STCKE when
>> the ETOD clock wraps.
> 
> Support for the 2nd epoch was added in z/OS 2.1
>  
Yup.  I tried it on a 2.1 system, and it seems good until about 2185.
I don't know why they didn't go for still more bits -- it's simplest
not to mask off any.  But:

o It hardly matters.

o In the future, (some of) those additional 7 bits may be used for
  purposes other than extending the range of the ETOD clock.  But
  I believe someone has pointed me to a publication saying that the
  entire high byte of ETOD is already committed for that extended
  (unsigned) range.

Long ago, I wondered why ETOD couldn't still be defined as signed;
it seems to me that CE 1899 might be more useful than CE 38434 (or
so).  At that time, I was told that there already existed a firm
specification that ETOD was a 128-bit (minus programmable area)
unsigned item.  I suppose programmers might already be loading its
comparator with 16X'FF' to defer an interrupt ridiculously far.
(Is there an ETOD comparator register?)

Thanks,
gil

Reply via email to