My today reported (and reassigned to linux2.6) bug #517122
doesn't gimme rest. One of the problem of analysis of traces is that
some times are recorded since epoch, some are the kernel's "uptime".
what puzzles me is:
* Difference between dmesg timestamp and /proc/uptime
$> tcpdump -i bond0 -tt -vvv -n "(dst host raider or src host raider)" -s 192 | head -3; dmesg | tail -1; echo "/proc/uptime"; cat /proc/uptime
[250851.259481] device eth1 left promiscuous mode
so I have timestamp in kernel messages bigger than uptime
I wonder why is that? different clock source? any other drift?
or may be 'kernel time' in dmesg is some kind of tick, not time per se?
* Is there an easy way to convert (reliably
) timestamp in dmesg to
time since epoch (with mksec precision of cause), so I could easily compare
with output in strace and alike.
Thanks in advance!!!!
Research Assistant, Psychology Department, Rutgers-Newark
Student Ph.D. @ CS Dept. NJIT
Office: (973) 353-1412 | FWD: 82823 | Fax: (973) 353-1171
101 Warren Str, Smith Hall, Rm 4-105, Newark NJ 07102