Problem with fan control

Problem with fan control

[Nils Faerber]

Hi all!
Playing around with thermal control and fan (ACPI 030228) I noticed the
following in my dmesg:

acpi_power-0363 [446] acpi_power_transition : Error transitioning device
[CFAN] to D3 acpi_bus-0496 [445] acpi_bus_set_power    : Error
transitioning device [CFAN] to D3 acpi_thermal-0567 [444]
acpi_thermal_active   : Unable to turn cooling device [c159f1a8] 'off'

I suspect this is not critical but nonetheless concerns me a little.

Next I found that the thermal management seems a little strange.
It does a pretty nice job controlling the CPU temperature using the CPU
throtteling. But the performance settings are used in the wrong way.
When my CPU gets hotter the performance setting is (re)set to maximum
performance and all manual resetting it is reverted after some seconds.

I am trying this on my P4M Asus L3800C where P0 and P1 mean about 50%
less power usage and thus less temperature. Returning it to P0 does not
make much sense to me... or is this a 'feature' of my ACPI BIOS?

CU
  nils faerber

-- 
kernel concepts          Tel: +49-271-771091-12
Dreisbachstr. 24         Fax: +49-271-771091-19
D-57250 Netphen          D1 : +49-170-2729106
--


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf

Re: Problem with fan control

[Karol Kozimor]

[-- Attachment #1: Type: text/plain, Size: 1404 bytes --]

Thus wrote Nils Faerber:
> acpi_power-0363 [446] acpi_power_transition : Error transitioning device
> [CFAN] to D3 acpi_bus-0496 [445] acpi_bus_set_power    : Error
> transitioning device [CFAN] to D3 acpi_thermal-0567 [444]
> acpi_thermal_active   : Unable to turn cooling device [c159f1a8] 'off'

Yep, I confirm. This happens when cooling is passive. When it is active,
echoing 1 to the CFAN/state gives an oops (attached), and the whole ACPI
subsystem screws up (i.e. processes defunct when writing). I'm using 
2.4.20-acpi-20021212.

> Next I found that the thermal management seems a little strange.
> It does a pretty nice job controlling the CPU temperature using the CPU
> throtteling. But the performance settings are used in the wrong way.
> When my CPU gets hotter the performance setting is (re)set to maximum
> performance and all manual resetting it is reverted after some seconds.

Thanks for tracking it down. Indeed, when cooling mode is set to passive
and CPU intensive tasks are run, the performance setting is switched back
to P0. This does not (to my knowledge) happen, when cooling mode is active.
Strangely, this is the only time the ACPI subsystem notices change in
processor performance. Oh, and BTW: P4-M should support C3 state? If it
says it doesn't, should I [bf]lame the manufacturer?
Best regards,

-- 
Karol 'sziwan' Kozimor
sziwan-DETuoxkZsSqrDJvtcaxF/A@public.gmane.org

[-- Attachment #2: oops --]
[-- Type: text/plain, Size: 3842 bytes --]

ymoops 2.4.8 on i686 2.4.20-xfs-cpufreq-swsusp.  Options used
     -V (default)
     -k /proc/ksyms (default)
     -l /proc/modules (default)
     -o /lib/modules/2.4.20-xfs-cpufreq-swsusp/ (default)
     -m /usr/src/linux/System.map (default)

Warning: You did not tell me where to find symbol information.  I will
assume that the log matches the kernel and modules that are running
right now and I'll use the default options above for symbol resolution.
If the current kernel and/or modules do not match the log, you can get
more accurate output by telling me the kernel version and where to find
map, modules, ksyms etc.  ksymoops -h explains the options.

Unable to handle kernel paging request at virtual address 87ec33cc
c0207057
*pde = 00000000
Oops: 0000
CPU:    0
EIP:    0010:[<c0207057>]    Not tainted
Using defaults from ksymoops -t elf32-i386 -a i386
EFLAGS: 00010246
eax: 00000000   ebx: 87ec33cc   ecx: 00000000   edx: c2817d7c
esi: 87ec33cc   edi: c2817df4   ebp: c021acf0   esp: c2817d84
ds: 0018   es: 0018   ss: 0018
Process kacpidpc (pid: 2926, stackpage=c2817000)
Stack: 00001001 c02081ad 87ec33cc c02161bb 00200000 c2817df4 c2817dc0 87ec33cc
       87ec33cc c021ada0 87ec33cc c021acf0 c2817df4 00200000 00000000 00010000
       c02d9c9f c02d9c54 00000050 00000000 c2817df8 c2817e34 c0224c68 87ec33cc
Call Trace:    [<c02081ad>] [<c02161bb>] [<c021ada0>] [<c021acf0>] [<c0224c68>]
  [<c02163d0>] [<c0225246>] [<c022589c>] [<c021b2f2>] [<c02163d0>] [<c0216100>]
  [<c022ac0c>] [<c0229cb9>] [<c022b0df>] [<c022c2d2>] [<c01f0a3e>] [<c01e5d73>]
  [<c01f09d0>] [<c010582e>] [<c01e5ca0>]
Code: 80 3b aa 0f 44 c3 5b c3 90 a1 94 f8 35 c0 eb f6 89 f6 8d bc


>>EIP; c0207057 <acpi_ns_map_handle_to_node+17/30>   <=====

>>edx; c2817d7c <_end+2499f24/12496208>
>>edi; c2817df4 <_end+2499f9c/12496208>
>>ebp; c021acf0 <acpi_bus_data_handler+0/50>
>>esp; c2817d84 <_end+2499f2c/12496208>

Trace; c02081ad <acpi_get_data+5d/90>
Trace; c02161bb <acpi_ut_trace+3b/40>
Trace; c021ada0 <acpi_bus_get_device+60/110>
Trace; c021acf0 <acpi_bus_data_handler+0/50>
Trace; c0224c68 <acpi_power_get_context+68/120>
Trace; c02163d0 <acpi_ut_value_exit+50/60>
Trace; c0225246 <acpi_power_off_device+56/290>
Trace; c022589c <acpi_power_transition+15c/190>
Trace; c021b2f2 <acpi_bus_set_power+1b2/3a0>
Trace; c02163d0 <acpi_ut_value_exit+50/60>
Trace; c0216100 <acpi_ut_debug_print+a0/e0>
Trace; c022ac0c <acpi_thermal_active+ec/220>
Trace; c0229cb9 <acpi_thermal_get_temperature+c9/100>
Trace; c022b0df <acpi_thermal_check+36f/3e0>
Trace; c022c2d2 <acpi_thermal_notify+122/160>
Trace; c01f0a3e <acpi_ev_notify_dispatch+6e/a0>
Trace; c01e5d73 <acpi_os_queue_exec+d3/100>
Trace; c01f09d0 <acpi_ev_notify_dispatch+0/a0>
Trace; c010582e <kernel_thread+2e/40>
Trace; c01e5ca0 <acpi_os_queue_exec+0/100>

Code;  c0207057 <acpi_ns_map_handle_to_node+17/30>
00000000 <_EIP>:
Code;  c0207057 <acpi_ns_map_handle_to_node+17/30>   <=====
   0:   80 3b aa                  cmpb   $0xaa,(%ebx)   <=====
Code;  c020705a <acpi_ns_map_handle_to_node+1a/30>
   3:   0f 44 c3                  cmove  %ebx,%eax
Code;  c020705d <acpi_ns_map_handle_to_node+1d/30>
   6:   5b                        pop    %ebx
Code;  c020705e <acpi_ns_map_handle_to_node+1e/30>
   7:   c3                        ret
Code;  c020705f <acpi_ns_map_handle_to_node+1f/30>
   8:   90                        nop
Code;  c0207060 <acpi_ns_map_handle_to_node+20/30>
   9:   a1 94 f8 35 c0            mov    0xc035f894,%eax
Code;  c0207065 <acpi_ns_map_handle_to_node+25/30>
   e:   eb f6                     jmp    6 <_EIP+0x6>
Code;  c0207067 <acpi_ns_map_handle_to_node+27/30>
  10:   89 f6                     mov    %esi,%esi
Code;  c0207069 <acpi_ns_map_handle_to_node+29/30>
  12:   8d bc 00 00 00 00 00      lea    0x0(%eax,%eax,1),%edi


1 warning issued.  Results may not be reliable.

Re: Problem with fan control

[Pavel Machek]

Hi!

> I am trying this on my P4M Asus L3800C where P0 and P1 mean about 50%
> less power usage and thus less temperature. Returning it to P0 does not
> make much sense to me... or is this a 'feature' of my ACPI BIOS?

Probably kernel bug. Add some magic printk's, ...
								Pavel
-- 
Horseback riding is like software...
...vgf orggre jura vgf serr.


-------------------------------------------------------
This sf.net email is sponsored by:ThinkGeek
Welcome to geek heaven.
http://thinkgeek.com/sf