| View previous topic :: View next topic |
| Author |
Message |
tnt
Veteran
Joined: 27 Feb 2004
Posts: 1121
|
 Posted: Sat Oct 22, 2011 11:54 am Post subject: Intel Q6600 - CPU1: Not responding ? |
 |
|
I have Q6600 on the Gigabyte's GA-P35-DS3R ver2.1 (latest F13 BIOS) running gentoo-sources-3.0.6
during the last boot only one core was recognized:
| Code: | ...
CPU0: Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz stepping 0b
Performance Events: PEBS fmt0+, Core2 events, Intel PMU driver.
PEBS disabled due to CPU errata.
...
Booting Node 0, Processors #1
smpboot cpu 1: start_ip = 9b000
CPU1: Not responding.
#2
smpboot cpu 2: start_ip = 9b000
CPU2: Not responding.
#3 Ok.
smpboot cpu 3: start_ip = 9b000
CPU3: Not responding.
Brought up 1 CPUs
Total of 1 processors activated (4799.99 BogoMIPS).
... |
but later:
| Code: | ...
microcode: no support for this CPU vendor
Kernel panic - not syncing: CPU#3 already initialized!
Pid: 0, comm: kworker/0:1 Not tainted 3.0.6-gentoo #6
Call Trace:
[<ffffffff813205dc>] ? panic+0x92/0x18a
[<ffffffff815071f7>] ? cpu_init+0x5d/0x253
[<ffffffff81508987>] ? start_secondary+0x7c/0x190
BUG: unable to handle kernel NULL pointer dereference at (null)
IP: [< (null)>] (null)
PGD 0
Oops: 0010 [#1] SMP
CPU 3
Modules linked in:
Pid: 0, comm: kworker/0:1 Not tainted 3.0.6-gentoo #6 Gigabyte Technology Co., Ltd. P35-DS3R/P35-DS3R
RIP: 0010:[<0000000000000000>] [< (null)>] (null)
RSP: 0000:ffff88020ecb7f28 EFLAGS: 00010092
RAX: 0000000003000000 RBX: 0000000000000003 RCX: 0000000000000000
RDX: 0000000000000000 RSI: 0000000000000000 RDI: 0000000000000020
RBP: 00000000ffff95ed R08: ffff88020ecb7ee8 R09: 0000000000000000
R10: 0000000000000000 R11: 0000000000000000 R12: 0000000000000000
R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000000
FS: 0000000000000000(0000) GS:ffff88020fd80000(0000) knlGS:0000000000000000
CS: 0010 DS: 0000 ES: 0000 CR0: 000000008005003b
CR2: 0000000000000000 CR3: 0000000001483000 CR4: 00000000000006a0
DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Process kworker/0:1 (pid: 0, threadinfo ffff88020ecb6000, task ffff88020ec4e2d0)
Stack:
0000000000000000 ffffffff81508937 0000000000000000 0000000000000000
0000000000000000 0000000000000000 0000000000000000 0000000000000000
0000000000000000 0000000000000000 0000000000000000 0000000000000000
Call Trace:
[<ffffffff81508937>] ? start_secondary+0x2c/0x190
Code: Bad RIP value.
RIP [< (null)>] (null)
RSP <ffff88020ecb7f28>
CR2: 0000000000000000
---[ end trace 93d72a36b9146f22 ]---
Kernel panic - not syncing: Attempted to kill the idle task!
Pid: 0, comm: kworker/0:1 Tainted: G D 3.0.6-gentoo #6
Call Trace:
[<ffffffff813205dc>] ? panic+0x92/0x18a
[<ffffffff81030421>] ? do_exit+0xa8/0x675
[<ffffffff81321fc4>] ? _raw_spin_lock_irqsave+0x9/0x1f
[<ffffffff8102ee10>] ? kmsg_dump+0x40/0xcc
[<ffffffff810045e8>] ? oops_end+0x86/0x8d
[<ffffffff8101ba21>] ? no_context+0x1ed/0x1fa
[<ffffffff81053927>] ? __module_text_address+0x9/0x55
[<ffffffff8101bf2e>] ? do_page_fault+0x12a/0x2e4
[<ffffffff81003717>] ? dump_trace+0x22f/0x23e
[<ffffffff81321fc4>] ? _raw_spin_lock_irqsave+0x9/0x1f
[<ffffffff81046424>] ? up+0xe/0x3a
[<ffffffff8102e5e0>] ? console_unlock+0x15e/0x179
[<ffffffff813206c5>] ? panic+0x17b/0x18a
[<ffffffff8132251f>] ? page_fault+0x1f/0x30
[<ffffffff81508937>] ? start_secondary+0x2c/0x190
VFS: Disk quotas dquot_6.5.2
Dquot-cache hash table entries: 512 (order 0, 4096 bytes)
Installing knfsd (copyright (C) 1996 okir@monad.swb.de).
msgmni has been set to 15986
Block layer SCSI generic (bsg) driver version 0.4 loaded (major 253)
... |
any info on how to get system to start with all 4 cores each time ? 
_________________
gentoo user |
|
| Back to top |
|
 |
DaggyStyle
Advocate
Joined: 22 Mar 2006
Posts: 4763
|
| Posted: Sat Oct 22, 2011 4:49 pm Post subject: |
|
|
notice this: | Quote: | | no support for this CPU vendor |
boot sysrescd, see if all cpus are there, is so, chroot into your gentoo, grab one od puppy mcfee's kernel seeds an reconfigure your kernel.
if you see the same issue in sysrescd, time to shop for a new cpu/mb/computer.
_________________
Only two things are infinite, the universe and human stupidity and I'm not sure about the former - Albert Einstein
ProjectFootball |
|
| Back to top |
|
|
krinn
Advocate
Joined: 02 May 2003
Posts: 3677
|
| Posted: Sat Oct 22, 2011 4:59 pm Post subject: |
|
|
Or your cpu is dead because i've seen you overclock high a corei7 cpu in another post, that's not bad, but seeing your cores are running at 70-75° is not something intel cpu are used too.
So did you beat your poor cpu to death ? |
|
| Back to top |
|
|
DaggyStyle
Advocate
Joined: 22 Mar 2006
Posts: 4763
|
| Posted: Sat Oct 22, 2011 5:04 pm Post subject: |
|
|
| krinn wrote: | Or your cpu is dead because i've seen you overclock high a corei7 cpu in another post, that's not bad, but seeing your cores are running at 70-75° is not something intel cpu are used too.
So did you beat your poor cpu to death ? |
this is a q6600, not a i7...
_________________
Only two things are infinite, the universe and human stupidity and I'm not sure about the former - Albert Einstein
ProjectFootball |
|
| Back to top |
|
|
krinn
Advocate
Joined: 02 May 2003
Posts: 3677
|
| Posted: Sat Oct 22, 2011 5:08 pm Post subject: |
|
|
| i know, but if he do that to an i7 there's no real limit to do that to an old Q6600 no ? |
|
| Back to top |
|
|
Ant P.
Veteran
Joined: 18 Apr 2009
Posts: 1988
Location: UK
|
| Posted: Sat Oct 22, 2011 5:34 pm Post subject: |
|
|
| I've seen this happen to my Phenom II after a rough reboot, usually power cycling the box fixes it. |
|
| Back to top |
|
|
tnt
Veteran
Joined: 27 Feb 2004
Posts: 1121
|
| Posted: Sat Oct 22, 2011 8:25 pm Post subject: |
|
|
@Q6600
that's the CPU in the office storage box - a lot of hard drives, software RAID, one hardware RAID controller, 8GB of RAM for caches/buffers, NFS server and iSCSI target. never been overclocked, never will be - there's no need for something like that. so, I guess that CPU is safe and sound. if there's some hardware problem, I think it will be motherboard related, as that mobo is 3.5 years old. it should be quite long lasting having dry caps, but you never now with contemporary mobos...
power cycle solved the problem for now, but I don't like the idea having the possibility of running in such a problem on stock settings.
@i7
it's i5-2500k. here are specs:
http://ark.intel.com/products/52210
notice that max temp (72.6C) is the case temp.
temps in my post are core temps.
core temps should not go over 98C for this particular CPU.
by intel's specs all fans should be set to the maximum revs at the 80C to avoid climbing to >90C:
| Code: | cat /sys/class/hwmon/hwmon0/device/temp2_label
Core 0
cat /sys/class/hwmon/hwmon0/device/temp2_max
80000
cat /sys/class/hwmon/hwmon0/device/temp2_crit
98000 |
temperatures ~75C are achieved by 4 thread prime95 run to show max turbo frequency (that topic was about turbo freqs).
CPU is cooled by TRUE 120
http://techgage.com/viewimg/?img=/articles/cooling/cpu_cooling_roundup/cpu_cooling_roundup_0608_19.jpg&desc=CPU%20Cooler%20Roundup%20-%20June%202008
and fancontrol is set to give full throttle to CPU fan at 80C, just like official specs request.
without prime95, fan idles at ~400rpm giving these temps:
| Code: | Core 0: +38.0 C (high = +80.0 C, crit = +98.0 C)
Core 1: +40.0 C (high = +80.0 C, crit = +98.0 C)
Core 2: +37.0 C (high = +80.0 C, crit = +98.0 C)
Core 3: +37.0 C (high = +80.0 C, crit = +98.0 C) |
anyways, besides temps within the specs, this CPU was even overclocked without any voltage increase, so I guess it's completely safe, too.
_________________
gentoo user |
|
| Back to top |
|
|
krinn
Advocate
Joined: 02 May 2003
Posts: 3677
|
| Posted: Sun Oct 23, 2011 12:20 am Post subject: |
|
|
| Didn't knew those cpu were having such a high temp tolerance |
|
| Back to top |
|
|
tnt
Veteran
Joined: 27 Feb 2004
Posts: 1121
|
| Posted: Sun Oct 23, 2011 1:05 am Post subject: |
|
|
| krinn wrote: | | Didn't knew those cpu were having such a high temp tolerance |
recent AMD CPUs have quite low temp tolerance.
some of them should be kept under 63C, similar to famous Pentium III 1.13GHz...
_________________
gentoo user |
|
| Back to top |
|
|
balticer
n00b
Joined: 30 Apr 2012
Posts: 4
|
| Posted: Mon Apr 30, 2012 5:58 pm Post subject: |
|
|
Hi,
i have a similar issue:
| Code: |
[ 0.021449] CPU0: Intel(R) Core(TM) i5-2500K CPU @ 3.30GHz stepping 07
[ 0.122918] Performance Events: PEBS fmt1+, SandyBridge events, Intel PMU driver.
[ 0.123225] PEBS disabled due to CPU errata.
[ 0.123345] ... version: 3
[ 0.123464] ... bit width: 48
[ 0.123583] ... generic registers: 8
[ 0.123703] ... value mask: 0000ffffffffffff
[ 0.123826] ... max period: 000000007fffffff
[ 0.123950] ... fixed-purpose events: 3
[ 0.124069] ... event mask: 00000007000000ff
[ 0.124260] Booting Node 0, Processors #1
[ 0.124341] smpboot cpu 1: start_ip = 8c000
[ 5.849856] CPU1: Not responding.
[ 5.850077] #2
[ 5.850152] smpboot cpu 2: start_ip = 8c000
[ 11.575773] CPU2: Not responding.
[ 11.576047] #3 Ok.
[ 11.576281] smpboot cpu 3: start_ip = 8c000
[ 17.301681] CPU3: Not responding.
[ 17.302001] Brought up 1 CPUs
[ 17.302303] Total of 1 processors activated (6585.03 BogoMIPS).
|
Full dmesg: http://pastebin.com/uAkWhNhf
CPU: i5-2500k
Mainboard: H67MA-USB3-B3 with the latest bios
Tested kernels: 3.3.4-gentoo-sources / 3.3.4-vanilla-sources
interesting is: if i press some random keys at kernel boot, the kernel brought all 4 cores up. if i do nothing, the kernel brought up 1 core, 2 cores but never all 4 cores. comprehensible at every boot
/edit
also wondering about
[ 0.123225] PEBS disabled due to CPU errata.
"PEBS" is the same pebs, like "cat /proc/cpuinfo | grep pebs", or?!
regards
Kernel: 3.3.4 |
|
| Back to top |
|
|
DaggyStyle
Advocate
Joined: 22 Mar 2006
Posts: 4763
|
| Posted: Tue May 01, 2012 8:04 am Post subject: |
|
|
there is a errata in the i5 and i7 (I've seen the same output as I have i7-2600), but I do use all cores, try a reboot and see if it is reoccurring.
_________________
Only two things are infinite, the universe and human stupidity and I'm not sure about the former - Albert Einstein
ProjectFootball |
|
| Back to top |
|
|
balticer
n00b
Joined: 30 Apr 2012
Posts: 4
|
| Posted: Tue May 01, 2012 8:34 am Post subject: |
|
|
hi,
as i wrote ->
| Quote: | | if i press some random keys at kernel boot, the kernel brought all 4 cores up. if i do nothing, the kernel brought up 1 core, 2 cores but never all 4 cores. comprehensible at every boot |
regards |
|
| Back to top |
|
|
DaggyStyle
Advocate
Joined: 22 Mar 2006
Posts: 4763
|
| Posted: Tue May 01, 2012 9:31 am Post subject: |
|
|
pastebin your .config please.
_________________
Only two things are infinite, the universe and human stupidity and I'm not sure about the former - Albert Einstein
ProjectFootball |
|
| Back to top |
|
|
roarinelk
Guru
Joined: 04 Mar 2004
Posts: 402
|
| Posted: Tue May 01, 2012 12:13 pm Post subject: |
|
|
| balticer wrote: | hi,
as i wrote ->
| Quote: | | if i press some random keys at kernel boot, the kernel brought all 4 cores up. if i do nothing, the kernel brought up 1 core, 2 cores but never all 4 cores. comprehensible at every boot |
|
I remember last year AMD CPUs had similar troubles, which were solved
by deactivating C1E support in the BIOS.
Try to disable all PM-related features in the BIOS first. |
|
| Back to top |
|
|
Ant P.
Veteran
Joined: 18 Apr 2009
Posts: 1988
Location: UK
|
| Posted: Tue May 01, 2012 1:03 pm Post subject: |
|
|
| Yep, I have one with the C1E bug. The problem is something to do with it shutting down when there's no external interrupts to process, whether or not there's other work to do. Combine that with things like CONFIG_NOHZ and you get a system that appears frozen until you move the mouse or type. |
|
| Back to top |
|
|
balticer
n00b
Joined: 30 Apr 2012
Posts: 4
|
| Posted: Tue May 01, 2012 3:55 pm Post subject: |
|
|
hi,
.config -> http://pastebin.com/ePrmbaZ4
don't have this problem with current sysrescuecd. no luck with disabling C1E or other powersave relating stuff in the bios. no luck with unset CONFIG_NO_HZ. i'll trying 3.0.30 kernel now.
/edit
no luck with 3.0.30. getting a sysrescuecd dmesg now...http://pastebin.com/H9qQYPJB
regards |
|
| Back to top |
|
|
DaggyStyle
Advocate
Joined: 22 Mar 2006
Posts: 4763
|
| Posted: Tue May 01, 2012 5:49 pm Post subject: |
|
|
have you oced the cpu?
_________________
Only two things are infinite, the universe and human stupidity and I'm not sure about the former - Albert Einstein
ProjectFootball |
|
| Back to top |
|
|
balticer
n00b
Joined: 30 Apr 2012
Posts: 4
|
| Posted: Tue May 01, 2012 6:03 pm Post subject: |
|
|
hi,
negative. sysrescuecd no errors btw.
regards |
|
| Back to top |
|
|
roarinelk
Guru
Joined: 04 Mar 2004
Posts: 402
|
| Posted: Tue May 01, 2012 6:24 pm Post subject: |
|
|
try a -git kernel, and if that doesn't work either, report this to the
linux kernel mailinglist (along with last working kernel version, e.g. of the sysrescue cd). |
|
| Back to top |
|
|
DaggyStyle
Advocate
Joined: 22 Mar 2006
Posts: 4763
|
| Posted: Wed May 02, 2012 5:32 am Post subject: |
|
|
I have the same errata msg in my dmesg but all cpus are up, I'll paste bin my .config later on as I didn't noticed something strange about your.
_________________
Only two things are infinite, the universe and human stupidity and I'm not sure about the former - Albert Einstein
ProjectFootball |
|
| Back to top |
|
|
|
Kernel & Hardware
