On Tue, 2 Nov 2021 11:33:17 -0500, Alan Altmark wrote: >>> >> ... UTC falls back at a leap second. > >Nope. There is no fall back for leap seconds. They are *inserted* into the >time stream (Temporal Mechanics 101). When that happens, UTC goes from >11:59:59 to 11:59:60 to 00:00:00. It doesn't pause, repeat, or go backwards. >How an OS translates that concept into its local clock is left an exercise to >the vendor.bbbbbbbbbbbb > I stand corrected. And my Linux will actually display 23:59:60 if I set TZ=right/...
z/OS shirks the issue by making user address spaces non- dispatchable during a leap second. z/VM? I have seen some discussion of inserting leap seconds at 23:59:60 local time rather than UTC. Bad Idea. I have entertained the idea of making the Fall DST transition at midnight so the inserted hour could be represented unambiguously as 24:00:00 to 24:59:59. No enthusiastic support. -- gil ---------------------------------------------------------------------- For IBM-MAIN subscribe / signoff / archive access instructions, send email to [email protected] with the message: INFO IBM-MAIN
