2001-10-30 01:50:50

by Randy Hron

[permalink] [raw]
Subject: VM test on 2.4.14-pre4

Summary: 2 oops, mmap001 times good with high sound quality.

Same old tests, same old box.

mp3 played 250 of 313 seconds.

Averages for 10 mtest01 runs
bytes allocated: 1244554854
User time (seconds): 2.107
System time (seconds): 3.063
Elapsed (wall clock) time: 31.306
Percent of CPU this job got: 16.00
Major (requiring I/O) page faults: 106.8
Minor (reclaiming a frame) faults: 304641.7

mp3 played 860 of 865 seconds.

Average for 5 mmap001 runs
bytes allocated: 2048000000
User time (seconds): 19.354
System time (seconds): 18.508
Elapsed (wall clock seconds) time: 172.63
Percent of CPU this job got: 21.60
Major (requiring I/O) page faults: 500165.0
Minor (reclaiming a frame) faults: 43.0



Neither oops caused the machine to reboot. The /boot/System.map
file was for 2.4.14-pre4, and these ksymoops were taken before
I rebooted. I pasted the oops from /var/log/kern.log and passed
them to ksymoops without any arguments.

vmstat, iostat, and ps all would not return a prompt. I could
exit shells, and IRC clients. init 6 hung sending kill to processes
started by init. <sysrq Sync Umount> worked. Hit the reset button.
Thanks to the kernel hackers and reiserfs folks it came up clean.

This oops came during the mtest01 run.

invalid operand: 0000
CPU: 0
EIP: 0010:[__free_pages_ok+93/484] Tainted: P
EFLAGS: 00010202
eax: 0000008c ebx: c121c2c0 ecx: c121c2c0 edx: 00000000
esi: c121c2c0 edi: 00000000 ebp: 00000000 esp: c183dec4
ds: 0018 es: 0018 ss: 0018
Process kswapd (pid: 4, stackpage=c183d000)
Stack: 00000001 c121c2c0 c121c2e8 00000000 003728e8 c182d8a0 00000000 c1528140
c0122c1e c0128f15 c0127be5 00000000 c155dd00 0000000d 00000000 c121c2e4
08453000 08056000 0870b005 df42d154 00000020 08059000 08055000 df430080
Call Trace: [mark_page_accessed+30/60] [__free_pages+29/32] [swap_out+869/1356] [shrink_cache+723/976] [shrink_caches+80/100]
Code: 0f 0b ba 00 e0 ff ff 21 e2 80 63 18 eb f6 42 05 20 0f 85 3e
Using defaults from ksymoops -t elf32-i386 -a i386

Code; 00000000 Before first symbol
00000000 <_EIP>:
Code; 00000000 Before first symbol
0: 0f 0b ud2a
Code; 00000002 Before first symbol
2: ba 00 e0 ff ff mov $0xffffe000,%edx
Code; 00000006 Before first symbol
7: 21 e2 and %esp,%edx
Code; 00000008 Before first symbol
9: 80 63 18 eb andb $0xeb,0x18(%ebx)
Code; 0000000c Before first symbol
d: f6 42 05 20 testb $0x20,0x5(%edx)
Code; 00000010 Before first symbol
11: 0f 85 3e 00 00 00 jne 55 <_EIP+0x55> 00000054 Before first symbol


This oops came after the tests completed and I hit 'q' to quit mp3blaster:

invalid operand: 0000
CPU: 0
EIP: 0010:[do_swap_page+135/228] Tainted: P
EFLAGS: 00010246
eax: 00000000 ebx: c15a24c0 ecx: c15a24e8 edx: c15a24e8
esi: 031afc00 edi: de3513f8 ebp: 00000001 esp: d30efecc
ds: 0018 es: 0018 ss: 0018
Process mp3blaster (pid: 223, stackpage=d30ef000)
Stack: 400fe824 ded0d460 00000000 de634180 c01207ce ded0d460 de634180 400fe824
de3513f8 031afc00 00000000 ded0d460 00000000 de634180 400fe824 c0111714
ded0d460 de634180 400fe824 00000000 d30ee000 00000004 c01115b8 414e9a04
Call Trace: [handle_mm_fault+98/176] [do_page_fault+348/1176] [do_page_fault+0/1176] [<e0c3c987>] [<e0c3c9de>]
[<e0c3e966>] [schedule+548/852] [do_page_fault+0/1176] [error_code+52/64]
Code: 0f 0b 8d 73 24 8d 43 28 39 43 28 74 11 b9 01 00 00 00 ba 03
Using defaults from ksymoops -t elf32-i386 -a i386

Trace; e0c3e966 <[cmpci]cm_write+59a/5b0>
Code; 00000000 Before first symbol
00000000 <_EIP>:
Code; 00000000 Before first symbol
0: 0f 0b ud2a
Code; 00000002 Before first symbol
2: 8d 73 24 lea 0x24(%ebx),%esi
Code; 00000004 Before first symbol
5: 8d 43 28 lea 0x28(%ebx),%eax
Code; 00000008 Before first symbol
8: 39 43 28 cmp %eax,0x28(%ebx)
Code; 0000000a Before first symbol
b: 74 11 je 1e <_EIP+0x1e> 0000001e Before first symbol
Code; 0000000c Before first symbol
d: b9 01 00 00 00 mov $0x1,%ecx
Code; 00000012 Before first symbol
12: ba 03 00 00 00 mov $0x3,%edx


I see pre5 is out there, and it's still early. :)
--
Randy Hron