Additional data points:
1. cat /proc/cpuinfo
processor : 0
vendor_id : GenuineIntel
cpu family : 6
model : 8
model name : Pentium III (Coppermine)
stepping : 6
cpu MHz : 731.490
cache size : 256 KB
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes
cpuid level : 3
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 sep mtrr pge mca cmov
pat pse36 mmx fxsr sse
bogomips : 1458.17
2. I had to repeat the test 7 times before I could get it to stay hung.
Switching to a text console would unwedge the process. Yes, just hitting
"Ctl-Alt-F1" to switch over would cause the process to unwedge and continue.
That's gotta be significant, but hell if I know what it could mean.
3. System is *not* low on memory. Typical state (this is now, but it's
representative):
[alane@wwweasel linux-2.4]$ free
total used free shared buffers cached
Mem: 254608 245620 8988 0 2584 158656
-/+ buffers/cache: 84380 170228
Swap: 803208 53528 749680
4. System was *not* heavily loaded. It was in the same state as it is now ...
just an xterm open (I couldn't get grip to reproduce, see above, so I ran it
from an xterm).
[alane@wwweasel linux-2.4]$ uptime
10:47pm up 17 min, 2 users, load average: 0.12, 0.06, 0.07
----- Forwarded message from Alan Eldridge <alane@xxxxxxxxxxxx> -----
>That does puzzle me, normally a kernel loop stops everything else from
>running on that cpu. Either you are running SMP or the problem is
>stranger than we thought. kdb will tell us, instead of guessing.
0xc7ca6000 00007437 00007405 0 000 run 0xc7ca6260*cdda2wav
kdb> btp 7437
EBP EIP Function(args)
0xc0129a0b page_launder+0x87 (0x7, 0x1)
kernel .text 0xc0100000 0xc0129984 0xc012a1b0
0xc7ca7cf4 0xc012b25b __alloc_pages+0x163
kernel .text 0xc0100000 0xc012b0f8 0xc012b34c
0xc7ca7cfc 0xc012b363 __get_free_pages+0x17
kernel .text 0xc0100000 0xc012b34c 0xc012b378
0xc7ca7d20 0xd69f19ff [sg]sg_low_malloc+0x13f (0x8000, 0x0, 0x1, 0x0)
sg .text 0xd69ee060 0xd69f18c0 0xd69f1a5c
0xc7ca7d4c 0xd69f1ac8 [sg]sg_malloc+0x6c (0xcd045000, 0x8000, 0xc7ca7d94,
0xc7ca7d98)
sg .text 0xd69ee060 0xd69f1a5c 0xd69f1bd8
0xc7ca7d9c 0xd69f07e2 [sg]sg_build_indi+0x18e (0xcd04501c, 0xcd045000,
0x6400000)
sg .text 0xd69ee060 0xd69f0654 0xd69f0880
0xc7ca7dbc 0xd69f1196 [sg]sg_build_reserve+0x4a (0xcd045000, 0x6400000,
0xcd04501c)
sg .text 0xd69ee060 0xd69f114c 0xd69f11c0
0xc7ca7f90 0xd69ef622 [sg]sg_ioctl+0x76e (0xc7fee040, 0xca8e9480, 0x2275,
0xbffff254, 0xc7ca6000)
sg .text 0xd69ee060 0xd69eeeb4 0xd69efb24
0xc7ca7fbc 0xc013f0d4 sys_ioctl+0x174 (0x3, 0x2275, 0xbffff254, 0x0, 0x3)
kernel .text 0xc0100000 0xc013ef60 0xc013f0f0
0xc0106dbb system_call+0x33
kernel .text 0xc0100000 0xc0106d88 0xc0106dc0
kdb> go
--
Alan Eldridge
"Smart Tags? We don't need no steenking Smart Tags!"
|