CPU underload

Eugene Grosbein eugen at grosbein.net
Sat Oct 31 18:27:52 UTC 2015


On 01.11.2015 01:11, Adrian Chadd wrote:
> Hi,
> 
> I don't see any DELAY in my testing, nor MipsUserIntr, so yeah, I must
> be doing something different.
> 
> I am testing on QCA9558, not AR9344 though. My AR9344 isn't with me atm.
> 
> Just run 'top'; what's showing? Are there other processes running? 

Inter-vlan bridging; while traffic runs:

# top -SH
last pid:  1063;  load averages:  0.38,  0.27,  0.47       up 0+05:02:08  01:18:20
50 processes:  3 running, 37 sleeping, 10 waiting
CPU:  1.3% user,  0.0% nice,  0.7% system, 52.6% interrupt, 45.4% idle
Mem: 1504K Active, 27M Inact, 19M Wired, 668K Buf, 73M Free
Swap: 

  PID USERNAME PRI NICE   SIZE    RES STATE    TIME    WCPU COMMAND
   11 root     -92    -     0K    88K RUN     17:13  52.84% intr{int2 arge0}
   10 root     155 ki31     0K     8K RUN    276:11  44.75% idle
 1063 root      40    0  7624K  2644K RUN      0:00   4.43% top
   11 root     -60    -     0K    88K WAIT     2:11   0.65% intr{swi4: clock (0)}
   11 root     -76    -     0K    88K WAIT     0:01   0.20% intr{swi0: uart}
   13 root     -72    -     0K    40K -        0:01   0.01% usb{usbus0}
    2 root     -16    -     0K     8K -        0:01   0.01% rand_harvestq
  337 root      40    0  6192K  1640K select   2:21   0.01% syslogd
   16 root     -16    -     0K     8K -        0:01   0.01% schedcpu
   11 root     -88    -     0K    88K WAIT     0:01   0.01% intr{int1 ehci0}
    3 root      -8    -     0K    16K -        0:00   0.00% cam{doneq0}
    5 root     -16    -     0K    16K psleep   0:00   0.00% pagedaemon{pagedaemon}
    9 root     -16    -     0K     8K psleep   0:00   0.00% bufdaemon
   15 root      16    -     0K     8K syncer   0:01   0.00% syncer
    7 root     -16 ki-1     0K     8K pollid   0:00   0.00% idlepoll
    8 root      -8    -     0K     8K -        0:00   0.00% bufspacedaemon
   14 root      -4    -     0K     8K vlruwt   0:00   0.00% vnlru
    4 root      -8    -     0K     8K jobque   0:29   0.00% task: mx25l flash
   12 root      -8    -     0K    24K -        0:12   0.00% geom{g_up}
   11 root     -72    -     0K    88K WAIT     0:09   0.00% intr{swi1: netisr 0}
    0 root     -16    0     0K    48K swapin   0:07   0.00% kernel{swapper}
   12 root      -8    -     0K    24K -        0:06   0.00% geom{g_event}
   13 root     -68    -     0K    40K -        0:01   0.00% usb{usbus0}
  542 root       8    0  6664K  2488K wait     0:01   0.00% sh
   13 root     -68    -     0K    40K -        0:01   0.00% usb{usbus0}
   30 root      -8    -     0K     8K mdwait   0:00   0.00% md0
   40 root      -8    -     0K     8K mdwait   0:00   0.00% md1
   12 root      -8    -     0K    24K -        0:00   0.00% geom{g_down}
  499 root       8    0  6176K  1464K nanslp   0:00   0.00% cron
  541 root       8    0  7416K  2296K wait     0:00   0.00% login
  978 root      -8    -     0K     8K mdwait   0:00   0.00% md2
   11 root     -52    -     0K    88K WAIT     0:00   0.00% intr{swi6: task queue}
    1 root       8    0  5048K   872K wait     0:00   0.00% init
    0 root       8    0     0K    48K -        0:00   0.00% kernel{thread taskq}


> Is devd doing something stupid?

devd is not started; my freebsd/mips build does not even contains devd.

> Does 'dmesg' show the switch ports flapping or something?

dmesg is flooded with messages:

Unaligned Load Word: pc=0xffffffff80314fb8, badvaddr=0x7ffdd845
Unaligned Load Word: pc=0xffffffff80314fb8, badvaddr=0x7ffdd841
Unaligned Load Word: pc=0xffffffff80314fb8, badvaddr=0x7ffdd83d
Unaligned Load Word: pc=0xffffffff80314fb8, badvaddr=0x7ffdd839
Unaligned Load Word: pc=0xffffffff80314fb8, badvaddr=0x7ffdd835
Unaligned Load Word: pc=0xffffffff80314fb8, badvaddr=0x7ffdd831
Unaligned Load Word: pc=0xffffffff80314fb8, badvaddr=0x7ffdd82d
Unaligned Load Word: pc=0xffffffff80314fb8, badvaddr=0x7ffdd829
Unaligned Load Word: pc=0xffffffff80314fb8, badvaddr=0x7ffdd825
Unaligned Load Word: pc=0xffffffff80314fb8, badvaddr=0x7ffdd821

No other lines in dmesg output.

> My CPU is definitely pegged during these tests. Since yours isn't then
> yeah, something odd is going on.
> 
> Can you post the output of 'vmstat 1' whilst you're running your tests?

# vmstat 1
procs  memory      page                       disks     faults         cpu
r b w  avm   fre   flt  re  pi  po    fr   sr fl0 md0   in    sy    cs us sy id
1 1 0  69M   73M    28   0   0   0     6    1   0   0 1822   433   188  0  8 91
0 0 0  57M   74M   185   0   0   0   330    0   0   0 2240  1444   281  2 59 39
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2516   105   458  0 56 44
0 0 0  57M   74M     0   0   0   0     0    1   0   0 2138   105   383  0 39 61
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2162   105   387  0 43 57
0 0 0  57M   74M     0   0   0   0     0    0   0   0 3218   105   704  0 77 23
0 0 0  57M   74M     0   0   0   0     0    1   0   0 2007   105   444  0 30 70
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2737   105   538  0 63 37
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2625   105   569  0 52 48
0 0 0  57M   74M     0   0   0   0     0    1   0   0 2586   105   499  0 61 39
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2024   105   423  0 30 70
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2500   105   590  0 47 53
0 0 0  57M   74M     0   0   0   0     0    1   0   0 2105   105   401  0 34 66
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2441   105   453  0 48 52
0 0 0  57M   74M     0   0   0   0     0    0   0   0 3002   105   732  1 60 39
0 0 0  57M   74M     0   0   0   0     0    1   0   0 2374   105   528  0 47 53
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2813   105   639  0 63 37
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2472   105   435  0 56 44
0 0 0  57M   74M     0   0   0   0     0    1   0   0 2855   105   665  0 64 36
0 0 0  57M   74M     0   0   0   0     0    0   0   0 3189   109   675  0 74 26
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2488   105   542  0 51 49
0 0 0  57M   74M     0   0   0   0     0    1   0   0 2518   105   452  0 50 50
0 0 0  57M   74M     0   0   0   0     0    0   0   1 2488   105   521  0 50 50
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2023   105   438  0 31 69
0 0 0  57M   74M     0   0   0   0     0    1   0   0 2501   105   524  0 44 56
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2976   105   622  0 70 30
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2694   105   592  0 59 41
0 0 0  57M   74M     0   0   0   0     0    1   0   0 2870   105   600  0 70 30
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2368   105   447  0 46 54
0 0 0  57M   74M     0   0   0   0     0    0   0   0 2853   105   565  0 69 31
0 0 0  57M   74M     0   0   0   0     0    1   0   0 2546   105   476  1 54 45

That's it.



More information about the freebsd-mips mailing list