[ntp:hackers] SIGIO and fuzz

tsg tglassey at earthlink.net
Wed Aug 14 14:38:57 UTC 2013


On 08/13/2013 04:11 PM, Harlan Stenn wrote:
> In http://bugs.ntp.org/show_bug.cgi?id=2433 Frank Kardel notes that we
> can change from a low-impact runtime check to a no-impact compile-time
> check, and in examining this situation he noted that we only "fuzz" the
> time if we are *not* using signaled I/O.
>
> I talked with Dave Mills about this, and he can't remember why this
> might have been a good idea, and thinks we should probably always "fuzz"
> the time.
>
> Does anybody know a reason we should leave the current behavior alone?
>
Fuzzing the Time adds an envelope of uncertainty which needs to be 
factored into NTP use from a Policy Context.Other that that there is no 
real reason not to fuzz all of the time representations.

In some camps it was thought that at some point someone would get NTP to 
work reliably in the sub-nanosecond region but with today's tech a 
single memory cache-error introducing 20ns to 50ns of slop into the 
process that doesnt seem to be true anymore.

That said, the good Dr. Mills is correct that there is no real reason 
not to fuzz everything.

Todd


-- 
// Standard "perasonal email" disclaimers apply



More information about the hackers mailing list