[solved] general protection fault SMP

Message

Timz · Post by **Timz** » Sat Feb 23, 2008 3:27 pm

Hi,
I have this errors in my kernel log since 2 days and i dont realy understand it,
is there a probleme with SMP ? my processor is not a dual core or some thing like that ...
so i'm wondering if someone know the problem because i found many posts about
that on the forum but not with exactly this error.

Should i change some thing in the hardware ? should i upgrade my kernel ?

thanks for your help.

Code: Select all

==> kern.log <==
Feb 23 16:01:21 www general protection fault: 0000 [#4]
Feb 23 16:01:21 www SMP
Feb 23 16:01:21 www Modules linked in:
Feb 23 16:01:21 www CPU:    0
Feb 23 16:01:21 www EIP:    0060:[<c0401763>]    Not tainted VLI
Feb 23 16:01:21 www EFLAGS: 00010286   (2.6.22-hardened-r8 #4)
Feb 23 16:01:21 www EIP is at prepare_to_copy+0x43/0x50
Feb 23 16:01:21 www eax: 80050033   ebx: c35cdfb0   ecx: c97e7a70   edx: 00000000
Feb 23 16:01:21 www esi: 003d0f00   edi: c97e7a70   ebp: b2bf14a4   esp: c35cdf04
Feb 23 16:01:21 www ds: 0068   es: 0068   fs: 00d8  gs: 0033  ss: 0068
Feb 23 16:01:21 www Process mysqld (pid: 19225, ti=c35cc000 task=c97e7a70 task.ti=c35cc000)
Feb 23 16:01:21 www Stack: c041aae5 0000001c e78a301c 00000286 e7fecc40 b2bf1bd8 b2bf14a4 c35cdfb0
Feb 23 16:01:21 www b2bf14a4 003d0f00 c042c532 c06cfe74 e78a3d80 e78a3d80 c06cfe70 04f3f5da
Feb 23 16:01:21 www 00000001 00000000 00004b6c c35cdfb0 003d0f00 e78a3d80 b2bf14a4 c041bd86
Feb 23 16:01:21 www Call Trace:
Feb 23 16:01:21 www [<c041aae5>] copy_process+0x65/0x10b0
Feb 23 16:01:21 www [<c042c532>] alloc_pid+0x152/0x280
Feb 23 16:01:21 www [<c041bd86>] do_fork+0x76/0x240
Feb 23 16:01:21 www [<c0400c02>] sys_clone+0x32/0x40
Feb 23 16:01:21 www [<c0402c92>] syscall_call+0x7/0xb
Feb 23 16:01:21 www =======================
Feb 23 16:01:21 www Code: 00 07 73 02 db e2 64 a1 48 00 00 00 89 c2 c1 e0 0d c1 e2 09 0f 77 db 84 02 04 b7 6b c0 90 8b 41 04 83 60 0c fe 0f 20 c0 83 c8 00 <00> 00 00 c3 c6 81 84 01 00 00 00 c3 90 a1 48 a0 6f c0 85 c0 75
Feb 23 16:01:21 www EIP: [<c0401763>] prepare_to_copy+0x43/0x50 SS:ESP 0068:c35cdf04

Code: Select all

www ~ # uptime
 16:17:00 up 72 days,  6:39,  2 users,  load average: 0.01, 0.01, 0.00
www ~ # uname -a
Linux www 2.6.22-hardened-r8 #4 SMP Tue Dec 11 13:55:09 CET 2007 i686 AMD Athlon(tm) Processor AuthenticAMD GNU/Linux
www ~ # lspci
00:00.0 Host bridge: Silicon Integrated Systems [SiS] 735 Host (rev 01)
00:01.0 PCI bridge: Silicon Integrated Systems [SiS] Virtual PCI-to-PCI bridge (AGP)
00:02.0 ISA bridge: Silicon Integrated Systems [SiS] SiS85C503/5513 (LPC Bridge)
00:02.1 SMBus: Silicon Integrated Systems [SiS] SiS961/2 SMBus Controller
00:02.2 USB Controller: Silicon Integrated Systems [SiS] USB 1.0 Controller (rev 07)
00:02.3 USB Controller: Silicon Integrated Systems [SiS] USB 1.0 Controller (rev 07)
00:02.5 IDE interface: Silicon Integrated Systems [SiS] 5513 [IDE] (rev d0)
00:0b.0 SCSI storage controller: Adaptec AHA-2940U/UW/D / AIC-7881U (rev 01)
00:0f.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL-8139/8139C/8139C+ (rev 10)
01:00.0 VGA compatible controller: ATI Technologies Inc Rage 128 Pro Ultra TF
www ~ # cat /proc/cpuinfo
processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 6
model           : 6
model name      : AMD Athlon(tm) Processor
stepping        : 2
cpu MHz         : 1095.404
cache size      : 256 KB
fdiv_bug        : no
hlt_bug         : no
f00f_bug        : no
coma_bug        : no
fpu             : yes
fpu_exception   : yes
cpuid level     : 1
wp              : yes
flags           : fpu vme de pse tsc msr pae mce cx8 mtrr pge mca cmov pat pse36 mmx fxsr sse syscall mmxext 3dnowext 3dnow up ts
bogomips        : 2191.97
clflush size    : 32

Timz · Post by **Timz** » Sat Feb 23, 2008 3:32 pm

my problem is getting worth, i'm now having many segfaults :

Code: Select all

www ~ # emerge --info
Segmentation fault

==> kern.log <==
Feb 23 16:30:11 www general protection fault: 0000 [#12]
Feb 23 16:30:11 www SMP
Feb 23 16:30:11 www Modules linked in:
Feb 23 16:30:11 www CPU:    0
Feb 23 16:30:11 www EIP:    0060:[<c0401763>]    Not tainted VLI
Feb 23 16:30:11 www EFLAGS: 00210286   (2.6.22-hardened-r8 #4)
Feb 23 16:30:11 www EIP is at prepare_to_copy+0x43/0x50
Feb 23 16:30:11 www eax: 80050033   ebx: e6ab3fa0   ecx: e7f91070   edx: 00000000
Feb 23 16:30:11 www esi: 01200011   edi: e7f91070   ebp: bf8900b8   esp: e6ab3ef4
Feb 23 16:30:11 www ds: 0068   es: 0068   fs: 00d8  gs: 0033  ss: 0068
Feb 23 16:30:11 www Process emerge (pid: 19738, ti=e6ab2000 task=e7f91070 task.ti=e6ab2000)
Feb 23 16:30:11 www Stack: c041aae5 000081a4 b7d859b2 d13154ec e76b03c0 00000004 de1c1ce0 e6ab3fa0
Feb 23 16:30:11 www bf8900b8 01200011 c042c532 c06cfe74 e7fc82d0 e7fc82d0 c06cfe70 00000004
Feb 23 16:30:11 www 00000001 00000000 00004d1d e6ab3fa0 01200011 e7fc82d0 bf8900b8 c041bd86
Feb 23 16:30:11 www Call Trace:
Feb 23 16:30:11 www [<c041aae5>] copy_process+0x65/0x10b0
Feb 23 16:30:11 www [<c042c532>] alloc_pid+0x152/0x280
Feb 23 16:30:11 www [<c041bd86>] do_fork+0x76/0x240
Feb 23 16:30:11 www [<c0407ee8>] sys_pipe+0x28/0x40
Feb 23 16:30:11 www [<c0400c02>] sys_clone+0x32/0x40
Feb 23 16:30:11 www [<c0402c92>] syscall_call+0x7/0xb
Feb 23 16:30:11 www =======================
Feb 23 16:30:11 www Code: 00 07 73 02 db e2 64 a1 48 00 00 00 89 c2 c1 e0 0d c1 e2 09 0f 77 db 84 02 04 b7 6b c0 90 8b 41 04 83 60 0c fe 0f 20 c0 83 c8 00 <00> 00 00 c3 c6 81 84 01 00 00 00 c3 90 a1 48 a0 6f c0 85 c0 75
Feb 23 16:30:11 www EIP: [<c0401763>] prepare_to_copy+0x43/0x50 SS:ESP 0068:e6ab3ef4

it seems to be always the function 'prepare_to_copy' .. hardware problem with the memory ?

help

schachti · Post by **schachti** » Sat Feb 23, 2008 3:38 pm

Try to run memtest86 for some hours to test your RAM. Run cpuburn for some hours to be sure it is not a thermal problem. Use smartmontools to check your harddisk.

eccerr0r · Post by **eccerr0r** » Sat Feb 23, 2008 3:41 pm

After rebooting, is this repeatable (to hopefully mix around addresses)?

This function call sequence appears to be fairly generic as everyone is running this function (fork, come on!) so it's quite likely not an actual bug.

You should be able to boot a SMP kernel on a uniprocessor, unicore machine.

Try running memtest. Also see if you can reproduce without any tainted modules loaded.

Timz · Post by **Timz** » Sat Feb 23, 2008 3:50 pm

Thanks for your response,
i dont thinks thats a harddrive problem because i dont have anything in logs and my smartd daemon is still running
but i think it's probably a memory problem.

i'm going to go to my office to do a memtest

thanks a lot !

Timz · Post by **Timz** » Sat Feb 23, 2008 11:32 pm

solved. one of the ram memory was defective

thanks a lot dudes