Hi Jean,
Am willing to keep trying something. This machine is one of 16 identical
machines in a cluster, so it is worth that little more effort.
Output before loading i2c driver:
isadump 0xC05 0xC06 1
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
20: 5b a4 69 ba a1 ac a3 22 1a 19 ff 61 47 61 47 ed
30: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
40: 01 d6 1f ff 3f 00 00 0e 2d 02 01 00 18 95 01 a3
50: fa 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
60: 5b a4 69 ba a1 ac a3 22 1a 19 ff 61 47 61 47 ed
70: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
a0: 5b a4 69 ba a1 ac a3 22 1a 19 ff 61 47 61 47 ed
b0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
c0: 01 00 00 ff 3f 00 00 0e 2d 02 01 00 18 95 01 a3
d0: fa 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
e0: 5b a4 69 ba a1 ac a3 22 1a 19 ff 61 47 61 47 ed
f0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
isadump 0xC05 0xC06 2
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
20: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
30: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
40: 01 00 10 ff 3f 00 00 0e 2d 02 01 00 18 95 02 a3
50: fa 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
60: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
70: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
a0: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
b0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
c0: 01 00 00 ff 3f 00 00 0e 2d 02 01 00 18 95 02 a3
d0: fa 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
e0: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
f0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
Output after loading i2c driver:
isadump 0xC05 0xC06 1
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
20: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
30: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
40: 01 00 10 ff 3f 00 00 0e 2d 02 01 00 18 95 01 a3
50: fa 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
60: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
70: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
a0: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
b0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
c0: 01 00 00 ff 3f 00 00 0e 2d 02 01 00 18 95 01 a3
d0: fa 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
e0: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
f0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
isadump 0xC05 0xC06 2
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
20: 5a a4 69 ba a1 ad a3 22 1a 19 ff 61 47 61 47 ed
30: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
40: 01 00 10 ff 3f 00 00 0e 2d 02 01 00 18 95 02 a3
50: fa 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
60: 5a a4 69 ba a1 ad a3 22 1a 19 ff 61 47 61 47 ed
70: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
a0: 5a a4 69 ba a1 ad a3 22 1a 19 ff 61 47 61 47 ed
b0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
c0: 01 00 00 ff 3f 00 00 0e 2d 02 01 00 18 95 02 a3
d0: f9 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
e0: 5a a4 69 ba a1 ad a3 22 1a 19 ff 61 47 61 47 ed
f0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
Output after loading w83627hf with init=0
isadump 0xC05 0xC06 1
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
20: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
30: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
40: 01 00 10 ff 3f 00 00 0e 2d 02 01 00 18 95 01 a3
50: fa 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
60: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
70: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
a0: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
b0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
c0: 01 00 00 ff 3f 00 00 0e 2d 02 01 00 18 95 01 a3
d0: fa 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
e0: 5b a4 69 ba a1 ac a3 22 19 19 ff 61 47 61 47 ed
f0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
isadump 0xC05 0xC06 2
0 1 2 3 4 5 6 7 8 9 a b c d e f
00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
20: 5b a4 69 bb a1 ac a3 22 19 19 ff 61 47 61 47 ed
30: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
40: 01 00 10 ff 3f 00 00 0e 2d 02 01 00 18 95 02 a3
50: f9 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
60: 5b a4 69 bb a1 ac a3 22 19 19 ff 61 47 61 47 ed
70: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
a0: 5b a4 69 bb a1 ac a3 22 19 19 ff 61 47 61 47 ed
b0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
c0: 01 00 00 ff 3f 00 00 0e 2d 02 01 00 18 95 02 a3
d0: f9 00 00 4b 00 50 00 ff ff ff ff ff ff ff ff ff
e0: 5b a4 69 bb a1 ac a3 22 19 19 ff 61 47 61 47 ed
f0: b0 c3 90 33 00 06 51 07 00 10 21 00 94 80 00 00
Jean Delvare wrote:
> Hi Richard,
>
>
>>Can confirm this is 2.8.8 for both i2c and lm_sensors. init=0 does not
>>help, changes sensors{1,2,3} to 2, which gives even more wrong temps.
>>Changed the code (those lines don't appear to be skipped with init=0)
>>and that had exactly the same effect. This was cold rebooting every time.
>
>
> OK... Not much hope is left then. We can continue digging up if you want,
> but know that this will probably not lead us anywhere.
>
>
>>Without driver, from a cold reboot. the output of isadump 0xC05 0xC06 is:
>> 0 1 2 3 4 5 6 7 8 9 a b c d e f
>>00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
>>10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
>>20: 5a a4 69 ba a1 ad a3 19 1a 19 ff 61 47 61 47 ed
>>30: b0 43 10 13 00 06 51 07 00 10 21 00 94 00 00 00
>>40: 01 de 1f ff 3f 00 00 0e 2d 02 01 00 18 95 80 5c
>>50: ff ff 00 ff ff ff 00 80 21 00 ff ff 11 ef ff 05
>>60: 5a a4 69 ba a1 ad a3 19 1a 19 ff 61 47 61 47 ed
>>70: b0 43 10 13 00 06 51 07 00 10 21 00 94 00 00 00
>>80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
>>90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
>>a0: 5a a4 69 ba a1 ad a3 19 1a 19 ff 61 47 61 47 ed
>>b0: b0 43 10 13 00 06 51 07 00 10 21 00 94 00 00 00
>>c0: 01 00 00 ff 3f 00 00 0e 2d 02 01 00 18 95 80 5c
>>d0: ff ff 00 ff ff ff 00 80 21 00 ff ff 11 ef ff 05
>>e0: 5a a4 69 ba a1 ad a3 19 1a 19 ff 61 47 61 47 ed
>>f0: b0 43 10 13 00 06 51 07 00 10 21 00 94 00 00 00
>>and then loading the isa driver the output changes to:
>> 0 1 2 3 4 5 6 7 8 9 a b c d e f
>>00: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
>>10: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
>>20: 5b a4 69 ba a1 ad a4 18 1a 19 ff 61 47 61 47 ed
>>30: b0 43 10 13 00 06 51 07 00 90 21 00 94 00 00 00
>>40: 01 de 1f ff 3f 00 00 0e 2d 02 01 00 18 95 80 5c
>>50: ff ff 00 ff ff ff 00 80 21 00 ff ff 11 ef ff 05
>>60: 5b a4 69 ba a1 ad a4 18 1a 19 ff 61 47 61 47 ed
>>70: b0 43 10 13 00 06 51 07 00 90 21 00 94 00 00 00
>>80: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
>>90: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff
>>a0: 5b a4 69 ba a1 ad a4 18 1a 19 ff 61 47 61 47 ed
>>b0: b0 43 10 13 00 06 51 07 00 90 21 00 94 00 00 00
>>c0: 01 00 00 ff 3f 00 00 0e 2d 02 01 00 18 95 80 5c
>>d0: ff ff 00 ff ff ff 00 80 21 00 ff ff 11 ef ff 05
>>e0: 5b a4 69 ba a1 ad a4 18 1a 19 ff 61 47 61 47 ed
>>f0: b0 43 10 13 00 06 51 07 00 90 21 00 94 00 00 00
>
>
> Differences in row 0x20 are not significant, these are the registers
> holding the measurements so slight changes are expected. The only
> register changing is 0x39, which, according to the driver code,
> represents the high limit for temp1. Strangely, the original value is 16
> degrees, which is odd, and the new one is -112 degrees if I'm not
> mistaking, which is even stranger. Did you only load the driver, or did
> you additionally run "sensors -s"? I'd guess the former, then it's
> strange that this limit changed.
>
> The Winbond chips are a bit more complex than the others, and have some
> registers located in different "banks" which isadump will not show by
> default. Please redo the same test (dump before load, dump after load)
> with the following commands:
> isadump 0xC05 0xC06 1
> isadump 0xC05 0xC06 2
> (you can do both in a row)
>
> This will show the information for temp2 and temp3, respectively. I
> wonder if we will observe a similar change we saw for temp1.
>
> Jean
>
