lkml.org 
[lkml]   [2015]   [Jun]   [2]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
SubjectRe: [RFC][PATCH 4/4] time: Do leapsecond adjustment in gettime fastpaths
From
On Tue, Jun 2, 2015 at 2:21 AM, Peter Zijlstra <peterz@infradead.org> wrote:
> On Tue, Jun 02, 2015 at 11:01:54AM +0200, Ingo Molnar wrote:
>> > Currently, leapsecond adjustments are done at tick time.
>> >
>> > As a result, the leapsecond was applied at the first timer
>> > tick *after* the leapsecond (~1-10ms late depending on HZ),
>> > rather then exactly on the second edge.
>
> So why not program a hrtimer to expire at the exact right time? Then
> even the VDSO gets its time 'right' and you avoid touching all the
> fast paths.

Well, the hrtimer won't always expire at the exact right time (due to
irq latency, possible clockevent drift/error, etc) , so that would
close the window a bit, but wouldn't eliminate the issue.

thanks
-john


\
 
 \ /
  Last update: 2015-06-02 16:41    [W:0.157 / U:0.372 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site