Tim Shoppa wrote:
Anne & Lynn Wheeler wrote:
section 4.6.1.4 setting and inspecting the clock (including some
description of UTC)
http://publibz.boulder.ibm.com/cgi-bin/bookmgr_OS390/BOOKS/DZ9AR004/4.6.1.4?SHELF=EZ2HW125&DT=19970613131822&CASE=
Interesting. So the hardware clock is actual ticks since the beginning
of the epoch, and the programmer is responsible for adding in leap
seconds etc.
I believe this means that the clock time is actually more directly
related to the TAI timescale or GPS hex time than UTC.
UTC has "funny" times like 31-DEC-2005 23:59:60 inserted for leap
seconds (and will be missing 23:59:59 if there is a leap backwards
rather than forwards), while TAI (like the mainframe clock) just keeps
on ticking one second at a time regardless of whether there is a leap
second or not.
Tim.
If the clock is synchronized to an external time reference (e.g., a
sysplex timer) that is in turn synchronized to NIST, z/OS will also
track and adjust for leap seconds, as well as handling conversion to a
number of local time formats, so this is not something an application
programmer would need to do.
If there is no external time reference, then the manual setting of the
TOD clock is probably off by several seconds anyway and worrying about
leap second adjustments is pointless.
--
Joel C. Ewing, Fort Smith, AR [EMAIL PROTECTED]
----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [EMAIL PROTECTED] with the message: GET IBM-MAIN INFO
Search the archives at http://bama.ua.edu/archives/ibm-main.html
