strange top TIME values

[Arno J. Klaassen]

Hello,

I see the following on a dual PPro -stable; same sources
(and config, apart form ethernet driver) on UP Athlon and
SMP PIII work OK :

######

last pid:   259;  load averages:  1.70,  1.63,  1.07    up 0+00:19:05  14:18:32
38 processes:  4 running, 34 sleeping
CPU states:  9.6% user,  0.0% nice, 15.6% system, 25.6% interrupt, 49.3% idle
Mem: 52M Active, 86M Inact, 39M Wired, 7068K Cache, 29M Buf, 572K Free
Swap: 246M Total, 12K Used, 246M Free

  PID USERNAME  PRI NICE  SIZE    RES STATE  C   TIME   WCPU    CPU COMMAND
  246 root       64   0 17976K 17548K RUN    1  96:14 57.13% 57.13% rsync
  222 root       45   0 17740K 17300K RUN    1 225:05 34.52% 34.52% rsync
  213 arno       31   0  5292K  1732K CPU1   1   0:43 10.11% 10.11% sshd
  119 root       10   0   212K    28K nfsidl 0 164:34  1.32%  1.32% nfsiod
  223 root        2   0 17720K 17292K select 0   2:04  0.73%  0.73% rsync
  120 root       10   0   212K    28K nfsidl 0   0:10  0.20%  0.20% nfsiod
  253 arno       28   0  1908K   996K CPU0   0   0:10  0.00%  0.00% top

######

box is up just 19 minutes, but top shows 96, 225, 164 minutes for
various processes.
This is just a backup-server and works OK;
top -I -s1 shows seconds just going up as expected and then suddenly
the minutes jump high.
No problem other whatsoever, I vaguely remember -current has the same
"problem" for some.

FYI, Arno