[lm-sensors] lm77 driver locks up with kernel-2.6.12

All of lore.kernel.org
 help / color / mirror / Atom feed

* [lm-sensors] lm77 driver locks up with kernel-2.6.12
@ 2005-08-17  1:54 Beat Meier
  2005-08-19 19:24 ` Rudolf Marek
                   ` (3 more replies)
  0 siblings, 4 replies; 5+ messages in thread
From: Beat Meier @ 2005-08-17  1:54 UTC (permalink / raw)
  To: lm-sensors

Hello

I'm doing temperature tests with lm77, kernel-2.6.12 on wrap.
There seems to be a problem with the driver or driver/kernel.
Sometimes the temperature is read bad and somthing locks up.
It begins with all temp values read are 40955500
and all temp_hist  are 0 except critical hyst which is min_hist which is 
81910000.
After that there is no chance get the values.
Sometimes only some values are corrupted in the read then the next read
is ok.
First I tought this is the lm77 or the board because on an other board 
there was ok
but now there is is also the same problem.

for example a read of temp_input needs 3 seconds

If you remove the drivers with
rmmod scx200_acb lm77 ic2_sensor  ic2_core
abd insert it again
the /sys/bus/i2c/drivers/lm77
directory is empty

If you remove and put again the modules before this problem with the bad 
values occures
it works i.e. the directory contains values and you can read temperatures.

BTW: What I do is a cat of every temp* file in a script and this scripts 
runs every 2 seconds

What's wrong? Anyone having this problem?

Greetings

Beat

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [lm-sensors] lm77 driver locks up with kernel-2.6.12
  2005-08-17  1:54 [lm-sensors] lm77 driver locks up with kernel-2.6.12 Beat Meier
@ 2005-08-19 19:24 ` Rudolf Marek
  2005-08-20  2:45 ` Beat Meier
                   ` (2 subsequent siblings)
  3 siblings, 0 replies; 5+ messages in thread
From: Rudolf Marek @ 2005-08-19 19:24 UTC (permalink / raw)
  To: lm-sensors

> 
> What's wrong? Anyone having this problem?
> 

Hello,

Lets see if it is chip driver problem or bus driver problem or hw problem. (or any combination)
Please can you just modprobe the scx200_acb driver and also i2c-dev.

When done with this please several times try:
i2cdump -y 0 0x48 w

0 is the bus number, add there yours. i2cdetect -l will tell you how many busses you have.
0x48 is chip addr also supply yours. You can use:
 i2cdetect 0
to detect all devices on the bus. (better to restart/poweroff the system after that to be sure we do not interfere with yet another issues)

Also please check syslog if you can see something there. You can also try to switch all debug flags in menuconfig (i2c section)

Thanks

regards
Rudolf

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [lm-sensors] lm77 driver locks up with kernel-2.6.12
  2005-08-17  1:54 [lm-sensors] lm77 driver locks up with kernel-2.6.12 Beat Meier
  2005-08-19 19:24 ` Rudolf Marek
@ 2005-08-20  2:45 ` Beat Meier
  2005-08-20  9:41 ` Rudolf Marek
  2005-08-23 21:44 ` Beat Meier
  3 siblings, 0 replies; 5+ messages in thread
From: Beat Meier @ 2005-08-20  2:45 UTC (permalink / raw)
  To: lm-sensors

Rudolf Marek wrote:

>>What's wrong? Anyone having this problem?
>>
>>    
>>
>
>Hello,
>
>Lets see if it is chip driver problem or bus driver problem or hw problem. (or any combination)
>Please can you just modprobe the scx200_acb driver and also i2c-dev.
>
>When done with this please several times try:
>i2cdump -y 0 0x48 w
>
>0 is the bus number, add there yours. i2cdetect -l will tell you how many busses you have.
>  
>
Hello Rudolf

By the dump the only which changes are the first 2 columns if there are 
no problems

     0,8  1,9  2,a  3,b  4,c  5,d  6,e  7,f
00: 7802 f800 2000 0005 a000 0004 0004 0004
same until
f8: 7802 f800 2000 0005 a000 0004 0004 0004

Diff of last colum of several dumps

< f8: b002 b000 2000 0005 a000 0004 0004 0004
 > f8: a802 a800 2000 0005 a000 0004 0004 0004

< f8: a802 a800 2000 0005 a000 0004 0004 0004
 > f8: 7802 f800 2000 0005 a000 0004 0004 0004

< f8: 7802 f800 2000 0005 a000 0004 0004 0004
 > f8: 6002 e000 2000 0005 a000 0004 0004 0004

If there is the problem with the bad temp. the dump is as follows:

     0,8  1,9  2,a  3,b  4,c  5,d  6,e  7,f
00: XXXX XXXX XXXX XXXX XXXX XXXX XXXX XXXX
same until
f8: XXXX XXXX XXXX XXXX XXXX XXXX XXXX XXXX

and syslog messages:

Aug 19 21:34:43 localhost kernel: i2c_adapter i2c-0: timeout in state 
address
Aug 19 21:35:35 localhost last message repeated 6 times
Aug 19 21:36:36 localhost last message repeated 41 times
Aug 19 21:37:37 localhost last message repeated 60 times
Aug 19 21:38:38 localhost last message repeated 60 times
Aug 19 21:39:39 localhost last message repeated 59 times

i2cdetect -l 
returns nothing but lm77 is without userspace support


>0x48 is chip addr also supply yours. You can use:
> i2cdetect 0
>to detect all devices on the bus. (better to restart/poweroff the system after that to be sure we do not interfere with yet another issues)
>
>  
>
In good state: needs 4 seconds to complete

woc1:/var/log/wpa_supplicant# time  i2cdetect 0
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will probe file /dev/i2c-0.
I will probe address range 0x03-0x77.
Continue? [Y/n] y
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:          XX XX XX XX XX XX XX XX XX XX XX XX XX
10: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX
20: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX
30: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX
40: XX XX XX XX XX XX XX XX UU XX XX XX XX XX XX XX
50: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX
60: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX
70: XX XX XX XX XX XX XX XX

real    0m4.057s
user    0m0.004s
sys     0m0.018s



in bad state: more than 2 minutes
woc1:/var/log/wpa_supplicant# time  i2cdetect 0
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will probe file /dev/i2c-0.
I will probe address range 0x03-0x77.
Continue? [Y/n] y
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f
00:          XX XX XX XX XX XX XX XX XX XX XX XX XX
10: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX
20: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX
30: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX
40: XX XX XX XX XX XX XX XX UU XX XX XX XX XX XX XX
50: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX
60: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX
70: XX XX XX XX XX XX XX XX

real    2m16.644s
user    0m0.003s
sys     0m0.545s

and in syslog the following messages comes:
Aug 19 21:31:05 localhost kernel: i2c_adapter i2c-0: timeout in state quick
Aug 19 21:31:36 localhost last message repeated 26 times
Aug 19 21:32:37 localhost last message repeated 45 times
Aug 19 21:33:38 localhost last message repeated 39 times


>Also please check syslog if you can see something there. You can also try to switch all debug flags in menuconfig (i2c section)
>
>  
>
When the problem begins i.e. the following appers (current: 37 ok)
Fri Aug 19 21:23:19 ART 2005 current:37 low:10 lowHyst:12 high:64 
highHist:62 critical:80 criticalHyst:78
Fri Aug 19 21:23:24 ART 2005 current:37 low:10 lowHyst:12 high:64 
highHist:62 critical:80 criticalHyst:78
Fri Aug 19 21:23:30 ART 2005 current:37 low:10 lowHyst:12 high:64 
highHist:62 critical:80 criticalHyst:78
Fri Aug 19 21:23:35 ART 2005 current:4095 low:4095 lowHyst:8191 
high:4095 highHist:0 critical:4095 criticalHyst:0
Fri Aug 19 21:23:51 ART 2005 current:4095 low:4095 lowHyst:8191 
high:4095 highHist:0 critical:4095 criticalHyst:0
Fri Aug 19 21:24:05 ART 2005 current:4095 low:4095 lowHyst:8191 
high:4095 highHist:0 critical:4095 criticalHyst:0
Fri Aug 19 21:24:20 ART 2005 current:4095 low:4095 lowHyst:8191 
high:4095 highHist:0 critical:4095 criticalHyst:0
Fri Aug 19 21:24:32 ART 2005 current:4095 low:4095 lowHyst:8191 
high:4095 highHist:0 critical:4095 criticalHyst:0
Fri Aug 19 21:24:48 ART 2005 current:4095 low:4095 lowHyst:8191 
high:4095 highHist:0 critical:4095 criticalHyst:0

the syslog messages are:

Jan  1 00:09:55 localhost kernel: i2c_adapter i2c-0: timeout in state 
address
Jan  1 00:11:29 localhost kernel: i2c_adapter i2c-0: timeout in state 
command
Jan  1 00:11:51 localhost kernel: i2c_adapter i2c-0: timeout in state 
repeat_start
Jan  1 00:12:31 localhost kernel: i2c_adapter i2c-0: timeout in state 
command
Aug 19 20:07:39 localhost kernel: i2c_adapter i2c-0: timeout in state 
address
Aug 19 20:09:11 localhost kernel: i2c_adapter i2c-0: timeout in state read
Aug 19 20:09:12 localhost kernel: i2c_adapter i2c-0: timeout in state 
address
Aug 19 20:09:46 localhost last message repeated 16 times
Aug 19 20:10:47 localhost last message repeated 32 times

>  
>

>Thanks
>
>regards
>Rudolf
>
>  
>
Thanks to you!!!

Beat
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20050820/abf15043/attachment-0001.html

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [lm-sensors] lm77 driver locks up with kernel-2.6.12
  2005-08-17  1:54 [lm-sensors] lm77 driver locks up with kernel-2.6.12 Beat Meier
  2005-08-19 19:24 ` Rudolf Marek
  2005-08-20  2:45 ` Beat Meier
@ 2005-08-20  9:41 ` Rudolf Marek
  2005-08-23 21:44 ` Beat Meier
  3 siblings, 0 replies; 5+ messages in thread
From: Rudolf Marek @ 2005-08-20  9:41 UTC (permalink / raw)
  To: lm-sensors

Hi again,

It seems is a definetly combination of HW and bus driver problems.

Now a little excercise:

1) cold boot the device
2) modprobe scx200_acb
3) modprobe i2c-dev
3) i2cdump -y 0 0x4b b
 Try several times dumping in byte mode. Will the transfer be delayted too? I mean does the driver also produce the messages to syslog while in bytemode?

Thanks

regards
Rudolf

^ permalink raw reply	[flat|nested] 5+ messages in thread

* [lm-sensors] lm77 driver locks up with kernel-2.6.12
  2005-08-17  1:54 [lm-sensors] lm77 driver locks up with kernel-2.6.12 Beat Meier
                   ` (2 preceding siblings ...)
  2005-08-20  9:41 ` Rudolf Marek
@ 2005-08-23 21:44 ` Beat Meier
  3 siblings, 0 replies; 5+ messages in thread
From: Beat Meier @ 2005-08-23 21:44 UTC (permalink / raw)
  To: lm-sensors

Rudolf Marek wrote:

>Hi again,
>
>It seems is a definetly combination of HW and bus driver problems.
>
>Now a little excercise:
>
>1) cold boot the device
>2) modprobe scx200_acb
>3) modprobe i2c-dev
>3) i2cdump -y 0 0x4b b
> Try several times dumping in byte mode. Will the transfer be delayted too? I mean does the driver also produce the messages to syslog while in bytemode?
>
>  
>
BTW: Normaly i2c-dev is not running I've inserted it but there is no 
difference in the output with or without this driver.

After an byte access i.e.

woc4:~# i2cdump -y 0 0x48 b

the temperature is wrong but only the temperature i.e.
  current:4095 low:10 lowHyst:12 high:64 highHist:62 critical:80 
criticalHyst:78
"value 4095 is in fact 4095500 but was divided by 1000"
reading temperature again gives the right value
  current:43 low:10 lowHyst:12 high:64 highHist:62 critical:80 
criticalHyst:78
or if I do a i2cdump -y 0 0x48 w the first read value at 00/0 is XXXX 
which seems
to indicate the same problem (see below output)
After an word access not! (This is always when lm77 is still in good case)

If read in the lm77 manual that the pointer to the register is latched.
Can this be the problem that sometimes the pointer is not explicitly set
(because the drivers knows that it has accessed last time the same
and will not set the pointer?)

This could be the problem because I do not read only temperature but
also limits and hyst. and this are in other registers ...

Ok in fact I do not have any idea whats going on with driver nor
with i2c ...

woc4:~# i2cdump -y 0 0x48 b;i2cdump -y 0 0x48 w;
     0  1  2  3  4  5  6  7  8  9  a  b  c  d  e  f    0123456789abcdef
00: 02 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
10: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
20: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
30: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
40: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
50: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
60: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
70: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
80: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
90: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
a0: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
b0: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
c0: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
d0: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
e0: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?
f0: XX 00 00 XX 00 04 XX 04 XX 00 00 XX 00 04 XX 04    X..X.?X?X..X.?X?

     0,8  1,9  2,a  3,b  4,c  5,d  6,e  7,f
00: XXXX 8000 2000 0005 a000 0004 0004 0004
08: c002 c000 2000 0005 a000 0004 0004 0004
10: c002 c000 2000 0005 a000 0004 0004 0004
18: c002 c000 2000 0005 a000 0004 0004 0004
20: c002 c000 2000 0005 a000 0004 0004 0004
28: c002 c000 2000 0005 a000 0004 0004 0004
30: c002 c000 2000 0005 a000 0004 0004 0004
38: c002 c000 2000 0005 a000 0004 0004 0004
40: c002 c000 2000 0005 a000 0004 0004 0004
48: c002 c000 2000 0005 a000 0004 0004 0004
50: c002 c000 2000 0005 a000 0004 0004 0004
58: c002 c000 2000 0005 a000 0004 0004 0004
60: c002 c000 2000 0005 a000 0004 0004 0004
68: c002 c000 2000 0005 a000 0004 0004 0004
70: c002 c000 2000 0005 a000 0004 0004 0004
78: c002 c000 2000 0005 a000 0004 0004 0004
80: c002 c000 2000 0005 a000 0004 0004 0004
88: c002 c000 2000 0005 a000 0004 0004 0004
90: c002 c000 2000 0005 a000 0004 0004 0004
98: c002 c000 2000 0005 a000 0004 0004 0004
a0: c002 c000 2000 0005 a000 0004 0004 0004
a8: c002 c000 2000 0005 a000 0004 0004 0004
b0: c002 c000 2000 0005 a000 0004 0004 0004
b8: c002 c000 2000 0005 a000 0004 0004 0004
c0: c002 c000 2000 0005 a000 0004 0004 0004
c8: c002 c000 2000 0005 a000 0004 0004 0004
d0: c002 c000 2000 0005 a000 0004 0004 0004
d8: c002 c000 2000 0005 a000 0004 0004 0004
e0: c002 c000 2000 0005 a000 0004 0004 0004
e8: c002 c000 2000 0005 a000 0004 0004 0004
f0: c002 c000 2000 0005 a000 0004 0004 0004
f8: c002 c000 2000 0005 a000 0004 0004 0004


^ permalink raw reply	[flat|nested] 5+ messages in thread

end of thread, other threads:[~2005-08-23 21:44 UTC | newest]

Thread overview: 5+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2005-08-17  1:54 [lm-sensors] lm77 driver locks up with kernel-2.6.12 Beat Meier
2005-08-19 19:24 ` Rudolf Marek
2005-08-20  2:45 ` Beat Meier
2005-08-20  9:41 ` Rudolf Marek
2005-08-23 21:44 ` Beat Meier

This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.