[ntp:questions] Effect of Gigabit Interrupt coalescence on ntp timing

E-Mail Sent to this address will be added to the BlackLists Null at BlackList.Anitech-Systems.invalid
Wed Jun 13 20:37:00 UTC 2012


> Terje Mathisen wrote:
>> all the 1588-capable NICs out there,
>>   and there must be quite a few of them now,
>>  will do more than this, and with a better on-board timer.

Are there more, than about half a dozen MACs, used in PC NICs,
 that have 1588 Hardware Time Stamping support?


1588 Hardware Time Stamping,
 appears to almost always be commingled with PTP.


While PTP claims sub nano second accuracy when using
 1588 Hardware Time Stamping,
 it seems like a typical PC would be struggling to make use of that,
  considering how much time context switches take,
  interrupt latencies, other processes CPU load,
  and the many slower than nano-second speed interfaces
  to the variety of I/O a typical PC uses.
   {A lightly loaded, dedicated, time / frequency reference,
     PC / device is a different issue.}


<http://www.nist.gov/el/isd/ieee/ieee1588.cfm>
<http://www.nist.gov/el/isd/ieee/prodinfo1588.cfm>
 I notice Meinberg has NTP Time Servers with a IEEE 1588 hardware interface.
  Perhaps Martin Burnicki would have input
   related to 1588 hardware time stamping of NTP packets being of any value?


<http://sourceforge.net/projects/ptpd/forums/forum/469208/topic/3514251>
 PTP provides timing information to NTP
  and lets NTP automatically select the best time source?


<https://lkml.org/lkml/2011/4/22/59>
IEEE 1588 hardware clock support,
 ... add support for PTP Hardware Clocks (PHCs) into the Linux kernel.
<http://lwn.net/Articles/439859/>


<http://ptpd.sourceforge.net/>

<http://linuxptp.sourceforge.net/>


<http://www.eecis.udel.edu/~mills/ptp.html>
<http://www.eecis.udel.edu/~mills/stamp.html>


-- 
E-Mail Sent to this address <BlackList at Anitech-Systems.com>
  will be added to the BlackLists.



More information about the questions mailing list