strange TOP PRI value

artur · 06-10-2003, 04:49 PM

My top reports weird priority values for some of the processes:

Code:

 14:55:36  up 8 min,  2 users,  load average: 2.89, 1.37, 0.54
40 processes: 39 sleeping, 1 running, 0 zombie, 0 stopped
CPU states:   2.1% user  35.8% system   0.0% nice   0.0% iowait  62.0% idle
Mem:  1033552k av,  589756k used,  443796k free,       0k shrd,   31244k buff
        42224k active,             497728k inactive
Swap: 1048816k av,       0k used, 1048816k free                  500624k cached

  PID USER     PRI  NI  SIZE  RSS SHARE STAT %CPU %MEM   TIME CPU COMMAND
  869 root      17   0  1040 1040   872 R    10.6  0.1   0:00   0 top
  818 root      12   0   632  632   428 D     7.1  0.0   0:28   0 cp
  814 root     18446744073709551615 -20     0    0     0 DW<   0.8  0.0   0:02 
    1 root       9   0   508  508   456 S     0.0  0.0   0:02   0 init
    2 root       9   0     0    0     0 SW    0.0  0.0   0:00   0 keventd
    3 root      19  19     0    0     0 SWN   0.0  0.0   0:00   0 ksoftirqd_CPU
    4 root       9   0     0    0     0 SW    0.0  0.0   0:00   0 kswapd
    5 root       9   0     0    0     0 SW    0.0  0.0   0:00   0 bdflush
    6 root       9   0     0    0     0 DW    0.0  0.0   0:01   0 kupdated
    7 root       9   0     0    0     0 SW    0.0  0.0   0:00   0 i2oevtd
    8 root     18446744073709551615 -20     0    0     0 SW<   0.0  0.0   0:00 
    9 root       9   0     0    0     0 SW    0.0  0.0   0:00   0 kjournald

#ps -ax
  PID TTY      STAT   TIME COMMAND
    1 ?        S      0:02 init [3]         
    2 ?        SW     0:00 [keventd]
    3 ?        SWN    0:00 [ksoftirqd_CPU0]
    4 ?        SW     0:00 [kswapd]
    5 ?        SW     0:00 [bdflush]
    6 ?        DW     0:01 [kupdated]
    7 ?        SW     0:00 [i2oevtd]
    8 ?        SW<    0:00 [mdrecoveryd]
    9 ?        SW     0:00 [kjournald]
  112 ?        SW     0:22 [kjournald]
-- SNIP --
  727 tty6     S      0:00 /sbin/mingetty tty6
  728 tty1     S      0:00 -bash
  783 ?        S      0:00 afpd -c 50 -n grail -f /etc/atalk/AppleVolumes.defaul
  785 ?        SW     0:00 [kjournald]
  814 ?        DW<    0:02 [loop0]
  818 tty1     D      0:31 cp -i -a linux-2.4.20 /mnt/cdrom/home/
  819 tty2     S      0:03 -bash
  867 ?        S      0:01 afpd -c 50 -n grail -f /etc/atalk/AppleVolumes.defaul
  872 tty2     R      0:00 ps -ax

So the loop and mdrecoveryd are being reported with a high PRI by top. I have unmounted the image file that was mounted through the loop, so I can't check that, but cat /proc/8/stat reports pri -1 for mdrecoveryd. Looks like top incorrectly interprets a negative number. Any idea why?

Hardware:
Tyan S2466 motherboard
Two AMD MP 2000+ processors running at 1666MHz
ATI XPert 2000 Pro 32MB AGP video card
On-board 3Com 3c905C-TX ethernet controller

running Red Hat 9.0 with kernel 2.4.20 in both non-SMP and SMP that I compiled from a freshly downloaded source. Both SMP and non-SMP enabled kernels exhibit the same problem.

AltF4 · 06-10-2003, 08:39 PM

do your kernel and your symbol table (System.map) versions match ?

artur · 06-11-2003, 10:28 AM

/boot/System.map is linked to System.map-2.4.20-SMP-0601a
uname -a returns 2.4.20-SMP-0601a

I'd say yes, they do match.

artur · 06-11-2003, 11:49 AM

/boot/System.map is linked to System.map-2.4.20-SMP-0601a
uname -a returns 2.4.20-SMP-0601a

I'd say yes, they do match.

sbaguz · 08-28-2003, 02:45 AM

I've the same strange mdrecoveryd line when asking for TOP command. And when i'm shutting down I read a message about the system killing it (because "mdrecoveryd woke up..." and something of this kind). Can someone explain me what's this and if it's all normal?

Thanks.