Re: 'uptime' command producing incorrect uptime

[Steve Thompson <smt at vgersoft dot com>]

On Sat, 10 Dec 2005, Jack wrote:

> It appears to me that the uptime command is not producing the correct
> uptime and, in fact, is running twice as fast as it should be.
> [...]

I've noticed a similar effect recently, and looked in the sources. If you
take a look at the algorithm used by cygwin (line 467 et seq of
fhandler_proc.cc in the 20051123 snapshot), you'll see that it can be
wrong for multiprocessor hosts (the sum of KernelTime and UserTime is too
large by a factor equal to the number of processors). It also appears to
be wrong for uniprocessor hosts that have been up for more than 49.7 days
because of the 32-bit value returned by GetTickCount(); my own system
reported an uptime of 16 days after being up for 66 days.

Steve

--
Unsubscribe info:      http://cygwin.com/ml/#unsubscribe-simple
Problem reports:       http://cygwin.com/problems.html
Documentation:         http://cygwin.com/docs.html
FAQ:                   http://cygwin.com/faq/