Messages in this thread | | | Date | Fri, 5 Jun 2015 16:12:59 +0200 | From | Richard Cochran <> | Subject | Re: [RFC][PATCH 4/4] time: Do leapsecond adjustment in gettime fastpaths |
| |
On Fri, Jun 05, 2015 at 11:10:08AM +0200, Peter Zijlstra wrote: > Firstly I would strongly suggest such applications not use UTC because > of this, I think TAI was invented for just this reason.
So I wonder whether the bug in the original post affects TAI timers as well...
> Secondly, how would John's patches help with this? Usespace loops > reading time would be using the VDSO and would still not get the right > time, and timers would be subject to the same IRQ latency that a hrtimer > based leap second insert would, and would still very much not be in-sync > across the cluster.
But we have a tick based insertion. (IIRC, it used to be hrtimer based, but that was buggy, too).
Thanks, Richard
| |