Skip to contentSkip to conversation

CPU off-on-off: lost CPU management?

DISCUSSION Peter Oruba 3 participants 5 messages Jun 5, 2007 – Jun 6, 2007

lore
Share & tools
lore
#0 lore raw

Stefan, Mircea,

reagarding your Turion cpufreq issues - is that solved by now?
http://lkml.org/lkml/2007/4/8/59

Regards,
Peter Oruba

Signature · 7 lines
AMD Saxony Limited Liability Company & Co. KG
Operating System Research Center
Wilschdorfer Landstr. 101, 01109 Dresden, Germany
Register Court Dresden: HRA 4896
General Partner authorized to represent:
AMD Saxony LLC (Wilmington, Delaware, US)
General Manager of AMD Saxony LLC: Dr. Hans-R. Deppe, Thomas McCoy
#2 lore raw
In reply to Mircea Bardac
Quoted 49 lines · On Tuesday 05 June 2007 12:52:59 Mircea Bardac wrote:

Hi Peter,

This time with Kernel 2.6.21.3.

[...]

# echo 1 > /sys/devices/system/cpu/cpu1/online
# cat /proc/cpuinfo
processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 72
model name      : AMD Turion(tm) 64 X2 Mobile Technology TL-52
stepping        : 2
cpu MHz         : 800.000
cache size      : 512 KB
physical id     : 0
siblings        : 1
core id         : 0
cpu cores       : 1
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 apic sep mtrr pge mca
cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt
rdtscp lm 3dnowext 3dnow pni cx16 lahf_lm cmp_legacy svm extapic cr8legacy
ts fid vid ttp tm stc
bogomips        : 1597.67
clflush size    : 64

processor : 1
vendor_id : AuthenticAMD
cpu family : 15
model : 72
model name : AMD Turion(tm) 64 X2 Mobile Technology TL-52
stepping : 2
cpu MHz : 798.015
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes

Quoted 8 lines

cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca
cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp lm 3dnowext 3dnow pni cx16 lahf_lm cmp_legacy svm extapic cr8legacy ts fid vid ttp tm stc
bogomips : 1596.26
clflush size : 64

As a follow-up:

As Stefan noticed initially, the CPU frequency of the 2nd CPU is different after the CPU is brought back online.

Mircea

#1 lore raw
In reply to Peter Oruba

Hi Peter,

This time with Kernel 2.6.21.3.

Quoted 2 lines · On Tuesday 05 June 2007 11:03:23 Peter Oruba wrote:

reagarding your Turion cpufreq issues - is that solved by now?
http://lkml.org/lkml/2007/4/8/59

**** /proc/cpuinfo shows half the bogomips it showed in my initial post **** (maybe it is right now)

# cat /proc/cpuinfo
processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 72
model name      : AMD Turion(tm) 64 X2 Mobile Technology TL-52
stepping        : 2
cpu MHz         : 800.000
cache size      : 512 KB
physical id     : 0
siblings        : 2
core id         : 0
cpu cores       : 2
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 apic sep mtrr pge mca 
cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp 
lm 3dnowext 3dnow pni cx16 lahf_lm cmp_legacy svm extapic cr8legacy ts fid 
vid ttp tm stc
bogomips        : 1597.67
clflush size    : 64

processor : 1
vendor_id : AuthenticAMD
cpu family : 15
model : 72
model name : AMD Turion(tm) 64 X2 Mobile Technology TL-52
stepping : 2
cpu MHz : 800.000
cache size : 512 KB
physical id : 0
siblings : 2
core id : 1
cpu cores : 2
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 apic sep mtrr pge mca
cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp lm 3dnowext 3dnow pni cx16 lahf_lm cmp_legacy svm extapic cr8legacy ts fid vid ttp tm stc
bogomips : 1597.67
clflush size : 64

# cpufreq-info
cpufrequtils 002: cpufreq-info (C) Dominik Brodowski 2004-2006
Report errors and bugs to [email protected], please.
analyzing CPU 0:
  driver: powernow-k8
  CPUs which need to switch frequency at the same time: 0 1
  hardware limits: 800 MHz - 1.60 GHz
  available frequency steps: 1.60 GHz, 800 MHz
  available cpufreq governors: ondemand, performance
  current policy: frequency should be within 800 MHz and 1.60 GHz.
                  The governor "ondemand" may decide which speed to use
                  within this range.
  current CPU frequency is 800 MHz (asserted by call to hardware).
analyzing CPU 1:
  driver: powernow-k8
  CPUs which need to switch frequency at the same time: 0 1
  hardware limits: 800 MHz - 1.60 GHz
  available frequency steps: 1.60 GHz, 800 MHz
  available cpufreq governors: ondemand, performance
  current policy: frequency should be within 800 MHz and 1.60 GHz.
                  The governor "ondemand" may decide which speed to use
                  within this range.
  current CPU frequency is 800 MHz (asserted by call to hardware).
# echo 0 > /sys/devices/system/cpu/cpu1/online

**** this time the bogomips didn't drop to half of the dual-cpu bogomips **** mostly because the dual bogomips is now half
**** (as I said before, don't know if this is right)
**** the features list now shows "up"
**** (in addition to what the features list of the dual-cpu showed) **** (is this right?, might have appeared in my initial post too)

# cat /proc/cpuinfo
processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 72
model name      : AMD Turion(tm) 64 X2 Mobile Technology TL-52
stepping        : 2
cpu MHz         : 800.000
cache size      : 512 KB
physical id     : 0
siblings        : 1
core id         : 0
cpu cores       : 1
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 apic sep mtrr pge mca 
cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp 
lm 3dnowext 3dnow up pni cx16 lahf_lm cmp_legacy svm extapic cr8legacy ts fid 
vid ttp tm stc
bogomips        : 1597.67
clflush size    : 64

[root@cougar /home/mircea]# cpufreq-info
cpufrequtils 002: cpufreq-info (C) Dominik Brodowski 2004-2006 Report errors and bugs to [email protected], please.
analyzing CPU 0:
driver: powernow-k8
CPUs which need to switch frequency at the same time: 0 hardware limits: 800 MHz - 1.60 GHz
available frequency steps: 1.60 GHz, 800 MHz
available cpufreq governors: ondemand, performance
current policy: frequency should be within 800 MHz and 1.60 GHz.
The governor "ondemand" may decide which speed to use within this range.
current CPU frequency is 800 MHz (asserted by call to hardware).

**** compared to my initial post, everything below didn't change at all **** meaning all inconsistences are still there
**** (like: cache size missing for the 2nd core)
**** still can't activate powernow-k8 for the 2nd core
**** bogomips appears to stay the same for each CPU compared to my prev **** post

# echo 1 > /sys/devices/system/cpu/cpu1/online
# cat /proc/cpuinfo
processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 72
model name      : AMD Turion(tm) 64 X2 Mobile Technology TL-52
stepping        : 2
cpu MHz         : 800.000
cache size      : 512 KB
physical id     : 0
siblings        : 1
core id         : 0
cpu cores       : 1
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 apic sep mtrr pge mca 
cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp 
lm 3dnowext 3dnow pni cx16 lahf_lm cmp_legacy svm extapic cr8legacy ts fid 
vid ttp tm stc
bogomips        : 1597.67
clflush size    : 64

processor : 1
vendor_id : AuthenticAMD
cpu family : 15
model : 72
model name : AMD Turion(tm) 64 X2 Mobile Technology TL-52
stepping : 2
cpu MHz : 798.015
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 apic sep mtrr pge mca
cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp lm 3dnowext 3dnow pni cx16 lahf_lm cmp_legacy svm extapic cr8legacy ts fid vid ttp tm stc
bogomips : 1596.26
clflush size : 64

# cpufreq-info
cpufrequtils 002: cpufreq-info (C) Dominik Brodowski 2004-2006
Report errors and bugs to [email protected], please.
analyzing CPU 0:
  driver: powernow-k8
  CPUs which need to switch frequency at the same time: 0
  hardware limits: 800 MHz - 1.60 GHz
  available frequency steps: 1.60 GHz, 800 MHz
  available cpufreq governors: ondemand, performance
  current policy: frequency should be within 800 MHz and 1.60 GHz.
                  The governor "ondemand" may decide which speed to use
                  within this range.
  current CPU frequency is 800 MHz (asserted by call to hardware).
analyzing CPU 1:
  no or unknown cpufreq driver is active on this CPU
# rmmod powernow-k8
# modprobe powernow-k8
FATAL: Error inserting powernow_k8 
(/lib/modules/2.6.21-ARCH/kernel/arch/i386/kernel/cpu/cpufreq/powernow-k8.ko): 
No such device

Many thanks for following this,
Mircea

#3 lore raw
In reply to Mircea Bardac

I'm sorry, I am not able to reproduce this issue (using 2.6.21.3). Everything works fine on my Turion laptop. It looks like there is also CPU hotplug involved in that issue, since brining cpu1 back on doesn't update siblings and cpu_cores in /proc/cpuinfo as well. Can you specify more details about the system you are using? Have you tried enabling CONFIG_CPU_FREQ_DEBUG and booting with "cpufreq.debug=7"?

Peter Oruba

Am Dienstag, 5. Juni 2007 11:56:44 schrieb Mircea Bardac:

Quoted 65 lines · On Tuesday 05 June 2007 12:52:59 Mircea Bardac wrote:
Quoted 49 lines

Hi Peter,

This time with Kernel 2.6.21.3.

[...]

# echo 1 > /sys/devices/system/cpu/cpu1/online
# cat /proc/cpuinfo
processor       : 0
vendor_id       : AuthenticAMD
cpu family      : 15
model           : 72
model name      : AMD Turion(tm) 64 X2 Mobile Technology TL-52
stepping        : 2
cpu MHz         : 800.000
cache size      : 512 KB
physical id     : 0
siblings        : 1
core id         : 0
cpu cores       : 1
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 apic sep mtrr pge
mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext
fxsr_opt rdtscp lm 3dnowext 3dnow pni cx16 lahf_lm cmp_legacy svm extapic
cr8legacy ts fid vid ttp tm stc
bogomips        : 1597.67
clflush size    : 64

processor : 1
vendor_id : AuthenticAMD
cpu family : 15
model : 72
model name : AMD Turion(tm) 64 X2 Mobile Technology TL-52
stepping : 2
cpu MHz : 798.015
fdiv_bug : no
hlt_bug : no
f00f_bug : no
coma_bug : no
fpu : yes
fpu_exception : yes

Quoted 8 lines

cpuid level : 1
wp : yes
flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge
mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt rdtscp lm 3dnowext 3dnow pni cx16 lahf_lm cmp_legacy svm extapic cr8legacy ts fid vid ttp tm stc
bogomips : 1596.26
clflush size : 64

As a follow-up:

As Stefan noticed initially, the CPU frequency of the 2nd CPU is different after the CPU is brought back online.

Mircea

Signature · 7 lines
AMD Saxony Limited Liability Company & Co. KG
Operating System Research Center
Wilschdorfer Landstr. 101, 01109 Dresden, Germany
Register Court Dresden: HRA 4896
General Partner authorized to represent:
AMD Saxony LLC (Wilmington, Delaware, US)
General Manager of AMD Saxony LLC: Dr. Hans-R. Deppe, Thomas McCoy
#4 lore raw

Hi!
This issue is solved with 2.6.22-rc for me. The second core is up with the same speed as core 1 (Bogomips are the same, too).

- Stefan Prechtel

2007/6/6, Peter Oruba <[email protected]>:

Quoted 10 lines

I'm sorry, I am not able to reproduce this issue (using 2.6.21.3). Everything works fine on my Turion laptop. It looks like there is also CPU hotplug involved in that issue, since brining cpu1 back on doesn't update siblings and cpu_cores in /proc/cpuinfo as well. Can you specify more details about the system you are using? Have you tried enabling CONFIG_CPU_FREQ_DEBUG and booting with "cpufreq.debug=7"?

Peter Oruba