linux-mips
[Top] [All Lists]

Re: Load Average >=1, mips kernel 2.6.10

To: linux-mips@linux-mips.org
Subject: Re: Load Average >=1, mips kernel 2.6.10
From: Mandeep Ahuja <ahuja@aksysnetworks.com>
Date: Tue, 22 Jul 2008 15:57:57 -0600
Cc: Ralf Baechle <ralf@linux-mips.org>, kevink@paralogos.com
In-reply-to: <20080722212900.GD22094@linux-mips.org>
Original-recipient: rfc822;linux-mips@linux-mips.org
References: <48863A41.2020303@aksysnetworks.com> <20080722212900.GD22094@linux-mips.org>
Sender: linux-mips-bounce@linux-mips.org
User-agent: Thunderbird 2.0.0.6 (X11/20070926)
Here is what my command line shows

Thanks for the reply...
here is what the telnet session shows

# uptime
19:45:38 up  2:45, load average: 1.00, 1.00, 0.94
# ps
 PID  Uid     VmSize Stat Command
1 root 372 S init 2 root SW [ksoftirqd/0]
   3 root            SW< [desched/0]
   4 root            SW< [events/0]
   5 root            SW< [khelper]
   6 root            SW< [kthread]
   7 root            SW< [kblockd/0]
   8 root            SW  [pdflush]
   9 root            SW  [pdflush]
  11 root            SW< [aio/0]
  10 root            SW  [kswapd0]
  13 root            SW  [mtdblockd]
  12 root            SW  [kseriod]
  14 root            SW< [IRQ 8]
  15 root            SWN [jffs2_gcd_mtd0]
  16 root            SW< [IRQ 15]
  56 root            SW< [IRQ 35]
  62 root        392 S   udhcpc -n -p /var/run/udhcpc.esw0.pid -i esw0
  68 root        364 S   /sbin/syslogd -n -m 0
  69 root        336 S   /sbin/klogd -n
  70 root        492 S   -sh
  76 root        316 S   telnetd
  99 root            SW< [IRQ 45]
 108 root        548 S   -sh
 189 root        552 S   -sh
 217 root        376 R   ps
# top
Mem: 5808K used, 8652K free, 0K shrd, 0K buff, 1364K cached
Load average: 1.00 1.00 0.94  (Status: S=sleeping R=running, W=waiting)
 PID USER     STATUS   RSS  PPID %CPU %MEM COMMAND
 220 root     R        388   189  1.9  2.6 top
 189 root     S        552    76  0.0  3.8 sh
 108 root     S        548    76  0.0  3.7 sh
  70 root     S        492     1  0.0  3.3 sh
  62 root     S        392     1  0.0  2.6 udhcpc
   1 root     S        372     0  0.0  2.5 init
  68 root     S        364     1  0.0  2.5 syslogd
  69 root     S        336     1  0.0  2.3 klogd
  76 root     S        316     1  0.0  2.1 telnetd
   2 root     SW         0     1  0.0  0.0 ksoftirqd/0
  15 root     SWN        0     1  0.0  0.0 jffs2_gcd_mtd0
   3 root     SW<        0     1  0.0  0.0 desched/0
  11 root     SW<        0     6  0.0  0.0 aio/0
   4 root     SW<        0     1  0.0  0.0 events/0
  14 root     SW<        0     6  0.0  0.0 IRQ 8
  16 root     SW<        0     6  0.0  0.0 IRQ 15
  56 root     SW<        0     6  0.0  0.0 IRQ 35
   5 root     SW<        0     1  0.0  0.0 khelper
   6 root     SW<        0     1  0.0  0.0 kthread
   7 root     SW<        0     6  0.0  0.0 kblockd/0
   8 root     SW         0     6  0.0  0.0 pdflush

There are no 'D' threads. What bothers me is that it slowly creeps up to 1.00 when the system starts or resets..so there is something going on...

i would appreciate any help or insights...

regards
manjee

Ralf Baechle wrote:
On Tue, Jul 22, 2008 at 01:51:29PM -0600, Mandeep Ahuja wrote:

I have an embedded system thats has mips processor. I recently updated to Kernel 2.6.10 from 2.4.17.
I was able to run the kernel and mount the jffs2 file system.

When the system starts the load average is low like 0.02 but, after about 2 minutes it becomes 1.00 and as long as the system is idle it stays at 1.00. If the system is not idle it would go up to like 1.21 but eventually come down to 1.00 but NEVER goes below 1.

There is no application running on the system. Only the busybox. Top shows all processes sleeping.

this is the version of kernel
Linux version 2.6.10_dev-malta-mips2_fp_len (gcc version 3.4.6)

Is there something holding the processor continuously? does anyone have any idea whats going on?.

I need to figure this one out before I start my application on it!.

Usually the reason would be something like a process stuck in
noninterruptible sleep state.  That's status "D" in the ps output.  Such
a process can't be killed and generally if a process is in this state for
extended time this would be considered a bug.

  Ralf




<Prev in Thread] Current Thread [Next in Thread>