> I would think so since, for instance, file date/times don't have much
> resolution. However, the original inquiry implied microsecond resolution
> time-difference between machines.
On my home LAN, I have quite a few machines running NTP. Here's what a
'xntpdc -p' looks like on one of 'em:
st poll reach delay offset disp
=======================================================================
+ A 2 1024 377 0.00050 -0.004370 0.00218
= B 1 256 377 0.00020 0.000176 0.00006
- C 2 64 36 0.00861 0.000055 1.93866
* D 1 128 377 0.00018 0.000174 0.00005
- E 2 64 16 0.00032 0.001583 1.93872
- F 2 256 377 0.00844 0.000222 0.00333
- G 2 512 367 0.00069 -0.000021 0.06850
+ H 2 256 377 0.00334 0.000112 0.00545
- I 2 512 377 -0.0150 -0.008419 0.01424
A is an Irix machine on my LAN. B is a local stratum one public time
server. C is a customer's machine on a remote WAN. D is a local stratum one
public time server. E is a FreeBSD 3 box on my LAN (SMP. FreeBSD SMP doesn't
keep very good time). F is a customer's machine on a remote WAN. G is my
UltraSparc. H is my local Cisco router. I is a Windows 98 box. The machine I
ran this on is a Linux box.
You'll notice that <1ms resolution is possible between Linux and FreeBSD UP
machines with recent kernels.
DS
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/
This archive was generated by hypermail 2b29 : Fri Mar 31 2000 - 21:00:30 EST