* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
@ 2006-03-12 16:40 ` Jean Delvare
2006-03-12 18:32 ` Harald Dunkel
` (10 subsequent siblings)
11 siblings, 0 replies; 23+ messages in thread
From: Jean Delvare @ 2006-03-12 16:40 UTC (permalink / raw)
To: lm-sensors
Hi Harald,
> Configuring a barebone (Aopen MZ915-M) I tried the sensors stuff.
> After running sensors-detect this is what sensors reports:
>
> w83627thf-isa-0290
> Adapter: ISA adapter
> VCore: +1.36 V (min = +1.94 V, max = +1.94 V) ALARM
> +12V: +12.28 V (min = +10.82 V, max = +13.19 V)
> +3.3V: +3.18 V (min = +3.14 V, max = +3.47 V)
> +5V: +4.93 V (min = +4.75 V, max = +5.25 V)
> -12V: -12.11 V (min = -13.18 V, max = -10.80 V)
> V5SB: +5.05 V (min = +4.76 V, max = +5.24 V)
> VBat: +3.10 V (min = +2.40 V, max = +3.60 V)
> fan1: 0 RPM (min = 664 RPM, div = 8) ALARM
> CPU Fan: 0 RPM (min = 664 RPM, div = 8) ALARM
> fan3: 0 RPM (min = 664 RPM, div = 8) ALARM
> M/B Temp: +7 C (high = +45 C, hyst = +101 C) sensor = diode
> CPU Temp: -5.5 C (high = +80 C, hyst = +75 C) sensor = diode
> temp3: +7.5 C (high = +80 C, hyst = +75 C) sensor = diode (beep)
> alarms:
> beep_enable:
> Sound alarm enabled
>
>
> Esp. CPU fan, M/B and CPU temperature look unreasonable. If I reboot
> and look into the bios, then it says the fan is at 3150 rpm, CPU temp.
> is 53C, and sys temperature is 49C. There is no second fan.
>
> Below you can find the output of sensors-detect after manually
> unloading the sensor modules. Any help would be highly appreciated.
I've replied to you on the LKML already:
http://lkml.org/lkml/2006/3/8/52
--
Jean Delvare
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
2006-03-12 16:40 ` Jean Delvare
@ 2006-03-12 18:32 ` Harald Dunkel
2006-03-13 9:31 ` Jean Delvare
` (9 subsequent siblings)
11 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-03-12 18:32 UTC (permalink / raw)
To: lm-sensors
Hi Jean,
Jean Delvare wrote:
> Hi Harald,
>
>
>>Configuring a barebone (Aopen MZ915-M) I tried the sensors stuff.
>>After running sensors-detect this is what sensors reports:
>>
>>w83627thf-isa-0290
>>Adapter: ISA adapter
>>VCore: +1.36 V (min = +1.94 V, max = +1.94 V) ALARM
>>+12V: +12.28 V (min = +10.82 V, max = +13.19 V)
>>+3.3V: +3.18 V (min = +3.14 V, max = +3.47 V)
>>+5V: +4.93 V (min = +4.75 V, max = +5.25 V)
>>-12V: -12.11 V (min = -13.18 V, max = -10.80 V)
>>V5SB: +5.05 V (min = +4.76 V, max = +5.24 V)
>>VBat: +3.10 V (min = +2.40 V, max = +3.60 V)
>>fan1: 0 RPM (min = 664 RPM, div = 8) ALARM
>>CPU Fan: 0 RPM (min = 664 RPM, div = 8) ALARM
>>fan3: 0 RPM (min = 664 RPM, div = 8) ALARM
>>M/B Temp: +7 C (high = +45 C, hyst = +101 C) sensor = diode
>>CPU Temp: -5.5 C (high = +80 C, hyst = +75 C) sensor = diode
>>temp3: +7.5 C (high = +80 C, hyst = +75 C) sensor = diode (beep)
>>alarms:
>>beep_enable:
>> Sound alarm enabled
>>
>>
>>Esp. CPU fan, M/B and CPU temperature look unreasonable. If I reboot
>>and look into the bios, then it says the fan is at 3150 rpm, CPU temp.
>>is 53C, and sys temperature is 49C. There is no second fan.
>>
>>Below you can find the output of sensors-detect after manually
>>unloading the sensor modules. Any help would be highly appreciated.
>
>
> I've replied to you on the LKML already:
> http://lkml.org/lkml/2006/3/8/52
>
Sure, in this posting you suggested to follow-up on this topic
on this mailing list, including the version of sensors and the
output of sensors-detect on this mailing list. Please also
note that on the second run the sensor has changed to
"thermistor".
Regards
Harri
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 258 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060312/eb4c77eb/signature.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
2006-03-12 16:40 ` Jean Delvare
2006-03-12 18:32 ` Harald Dunkel
@ 2006-03-13 9:31 ` Jean Delvare
2006-03-18 5:14 ` Harald Dunkel
` (8 subsequent siblings)
11 siblings, 0 replies; 23+ messages in thread
From: Jean Delvare @ 2006-03-13 9:31 UTC (permalink / raw)
To: lm-sensors
Hallo Harald,
On 2006-03-12, Harald Dunkel wrote:
> > I've replied to you on the LKML already:
> > http://lkml.org/lkml/2006/3/8/52
>
> Sure, in this posting you suggested to follow-up on this topic
> on this mailing list, including the version of sensors and the
> output of sensors-detect on this mailing list. Please also
> note that on the second run the sensor has changed to
> "thermistor".
When moving a discussion from one list to another, it's usually
appreciated that you mention the original thread, so that new readers
know what has already been said and don't start again from scratch.
Nevermind, and sorry for the fresh welcome ;)
I don't see the version of lm_sensors in your second report, I'll
assume 2.10.0.
I'm surprised that the thermal sensor types changed from "diode" to
"thermistor". These settings are not supposed to change on their own.
Did you change anything in your configuration file? It probably doesn't
matter much here, as these inputs seem to be unused, but I'm curious.
The output of sensors-detect is interesting:
> Next adapter: SMBus I801 adapter at 5000
> (...)
> Client found at address 0x2d
(All Failed)
> Client found at address 0x2f
(All Failed)
These are typical addresses for hardware monitoring chips, so I start
suspecting that one of these is your real fans + temperatures sensor
chip, either a recent or rare type which we don't know about yet.
Please provide the output of the following commands (which will dump the
contents of these chips for further analysis) as root:
modprobe i2c-dev
i2cdump 0 0x2d b
i2cdump 0 0x2f b
Thanks,
--
Jean Delvare
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
` (2 preceding siblings ...)
2006-03-13 9:31 ` Jean Delvare
@ 2006-03-18 5:14 ` Harald Dunkel
2006-03-26 9:43 ` Harald Dunkel
` (7 subsequent siblings)
11 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-03-18 5:14 UTC (permalink / raw)
To: lm-sensors
Hi Jean,
Sorry for the late response. See below.
Jean Delvare wrote:
> Hallo Harald,
>
>
> I don't see the version of lm_sensors in your second report, I'll
> assume 2.10.0.
>
Yes, it is.
> I'm surprised that the thermal sensor types changed from "diode" to
> "thermistor". These settings are not supposed to change on their own.
> Did you change anything in your configuration file? It probably doesn't
> matter much here, as these inputs seem to be unused, but I'm curious.
>
No, I did not change this file. Currently sensors replies "diode"
again.
> The output of sensors-detect is interesting:
>
>
>>Next adapter: SMBus I801 adapter at 5000
>>(...)
>>Client found at address 0x2d
>
> (All Failed)
>
>>Client found at address 0x2f
>
> (All Failed)
>
> These are typical addresses for hardware monitoring chips, so I start
> suspecting that one of these is your real fans + temperatures sensor
> chip, either a recent or rare type which we don't know about yet.
>
> Please provide the output of the following commands (which will dump the
> contents of these chips for further analysis) as root:
>
> modprobe i2c-dev
> i2cdump 0 0x2d b
> i2cdump 0 0x2f b
>
No problem, see below. Please mail if more information is needed.
Many thanx in advance
Harri
=====================================
{root at bugs:harri 1001} modprobe i2c-dev
{root at bugs:harri 1001} i2cdump 0 0x2d b
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will probe file /dev/i2c-0, address 0x2d, mode byte
Continue? [Y/n]
0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
00: 01 07 00 40 5a ff ff ff ff ff ff ff ff ff ff ff ??. at Z...........
10: c5 72 7e 85 25 22 01 df 0f fe 00 00 1c 80 ad 07 ?r~?%"????..????
20: ff 00 ff 00 ff 00 ff 00 4b 4b 4b 4b 3c 37 00 ff ........KKKK<7..
30: 00 b0 b0 00 00 00 30 ff ff ff ff ff ff ff ff ff .??...0.........
40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
50: ff ff ff ff ff ff ff ff ff ff 04 10 20 19 34 ff ..........?? ?4.
60: bb 0a 90 00 01 00 00 0f ff 55 64 46 64 46 ff ff ???.?..?.UdFdF..
70: 02 ab ff ff 00 00 ff ff ff ff ff ff ff ff ff ff ??..............
80: 0f ff ff ff 00 00 ff ff ff ff ff ff ff ff ff ff ?...............
90: 00 00 0d ff ff ff ff ff ff ff ff ff ff ff ff ff ..?.............
a0: 46 3c 32 28 ff d9 b2 99 80 ff ff ff ff ff ff ff F<2(.????.......
b0: 46 3c 32 28 ff d9 b2 99 80 ff ff ff ff ff ff ff F<2(.????.......
c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
f0: 00 00 00 00 00 00 39 38 01 45 00 ff ff ff ff 27 ......98?E.....'
{root at bugs:harri 1002} i2cdump 0 0x2f b
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will probe file /dev/i2c-0, address 0x2f, mode byte
Continue? [Y/n]
0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
00: ff 03 07 f0 22 08 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: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
50: ff ff ff ff ff ff ff ff ff ff 03 02 12 19 34 ff ..........????4.
60: ff ff ff ff ff ff ff ff ff ff a0 a0 ff ff ff ff ..........??....
70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
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: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
f0: 00 80 ff ff ff ff ff ff ff ff ff ff ff ff ff 5e .?.............^
{root at bugs:harri 1003} sensors
w83627thf-isa-0290
Adapter: ISA adapter
VCore: +1.37 V (min = +1.94 V, max = +1.94 V) ALARM
+12V: +12.22 V (min = +10.82 V, max = +13.19 V)
+3.3V: +3.18 V (min = +3.14 V, max = +3.47 V)
+5V: +4.91 V (min = +4.75 V, max = +5.25 V)
-12V: -12.11 V (min = -13.18 V, max = -10.80 V)
V5SB: +5.05 V (min = +4.76 V, max = +5.24 V)
VBat: +3.09 V (min = +2.40 V, max = +3.60 V)
fan1: 0 RPM (min = 664 RPM, div = 8) ALARM
CPU Fan: 0 RPM (min = 664 RPM, div = 8) ALARM
fan3: 0 RPM (min = 664 RPM, div = 8) ALARM
M/B Temp: +6 C (high = -87 C, hyst = +125 C) sensor = diode
CPU Temp: -7.5 C (high = +80 C, hyst = +75 C) sensor = diode
temp3: +7.5 C (high = +80 C, hyst = +75 C) sensor = diode (beep)
alarms:
beep_enable:
Sound alarm enabled
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 258 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060318/05a8f955/signature.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
` (3 preceding siblings ...)
2006-03-18 5:14 ` Harald Dunkel
@ 2006-03-26 9:43 ` Harald Dunkel
2006-03-31 12:35 ` Jean Delvare
` (6 subsequent siblings)
11 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-03-26 9:43 UTC (permalink / raw)
To: lm-sensors
Hi folks,
Any news about this? Some information missing?
Regards
Harri
======================================
Harald Dunkel wrote:
> Hi Jean,
>
> Sorry for the late response. See below.
>
> Jean Delvare wrote:
>
>>Hallo Harald,
>>
>>
>>I don't see the version of lm_sensors in your second report, I'll
>>assume 2.10.0.
>>
>
>
> Yes, it is.
>
>
>>I'm surprised that the thermal sensor types changed from "diode" to
>>"thermistor". These settings are not supposed to change on their own.
>>Did you change anything in your configuration file? It probably doesn't
>>matter much here, as these inputs seem to be unused, but I'm curious.
>>
>
>
> No, I did not change this file. Currently sensors replies "diode"
> again.
>
>
>>The output of sensors-detect is interesting:
>>
>>
>>
>>>Next adapter: SMBus I801 adapter at 5000
>>>(...)
>>>Client found at address 0x2d
>>
>>(All Failed)
>>
>>
>>>Client found at address 0x2f
>>
>>(All Failed)
>>
>>These are typical addresses for hardware monitoring chips, so I start
>>suspecting that one of these is your real fans + temperatures sensor
>>chip, either a recent or rare type which we don't know about yet.
>>
>>Please provide the output of the following commands (which will dump the
>>contents of these chips for further analysis) as root:
>>
>>modprobe i2c-dev
>>i2cdump 0 0x2d b
>>i2cdump 0 0x2f b
>>
>
>
> No problem, see below. Please mail if more information is needed.
>
>
> Many thanx in advance
>
> Harri
> =====================================
> {root at bugs:harri 1001} modprobe i2c-dev
> {root at bugs:harri 1001} i2cdump 0 0x2d b
> WARNING! This program can confuse your I2C bus, cause data loss and worse!
> I will probe file /dev/i2c-0, address 0x2d, mode byte
> Continue? [Y/n]
> 0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
> 00: 01 07 00 40 5a ff ff ff ff ff ff ff ff ff ff ff ??. at Z...........
> 10: c5 72 7e 85 25 22 01 df 0f fe 00 00 1c 80 ad 07 ?r~?%"????..????
> 20: ff 00 ff 00 ff 00 ff 00 4b 4b 4b 4b 3c 37 00 ff ........KKKK<7..
> 30: 00 b0 b0 00 00 00 30 ff ff ff ff ff ff ff ff ff .??...0.........
> 40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 50: ff ff ff ff ff ff ff ff ff ff 04 10 20 19 34 ff ..........?? ?4.
> 60: bb 0a 90 00 01 00 00 0f ff 55 64 46 64 46 ff ff ???.?..?.UdFdF..
> 70: 02 ab ff ff 00 00 ff ff ff ff ff ff ff ff ff ff ??..............
> 80: 0f ff ff ff 00 00 ff ff ff ff ff ff ff ff ff ff ?...............
> 90: 00 00 0d ff ff ff ff ff ff ff ff ff ff ff ff ff ..?.............
> a0: 46 3c 32 28 ff d9 b2 99 80 ff ff ff ff ff ff ff F<2(.????.......
> b0: 46 3c 32 28 ff d9 b2 99 80 ff ff ff ff ff ff ff F<2(.????.......
> c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> f0: 00 00 00 00 00 00 39 38 01 45 00 ff ff ff ff 27 ......98?E.....'
> {root at bugs:harri 1002} i2cdump 0 0x2f b
> WARNING! This program can confuse your I2C bus, cause data loss and worse!
> I will probe file /dev/i2c-0, address 0x2f, mode byte
> Continue? [Y/n]
> 0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
> 00: ff 03 07 f0 22 08 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: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 50: ff ff ff ff ff ff ff ff ff ff 03 02 12 19 34 ff ..........????4.
> 60: ff ff ff ff ff ff ff ff ff ff a0 a0 ff ff ff ff ..........??....
> 70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 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: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> f0: 00 80 ff ff ff ff ff ff ff ff ff ff ff ff ff 5e .?.............^
> {root at bugs:harri 1003} sensors
> w83627thf-isa-0290
> Adapter: ISA adapter
> VCore: +1.37 V (min = +1.94 V, max = +1.94 V) ALARM
> +12V: +12.22 V (min = +10.82 V, max = +13.19 V)
> +3.3V: +3.18 V (min = +3.14 V, max = +3.47 V)
> +5V: +4.91 V (min = +4.75 V, max = +5.25 V)
> -12V: -12.11 V (min = -13.18 V, max = -10.80 V)
> V5SB: +5.05 V (min = +4.76 V, max = +5.24 V)
> VBat: +3.09 V (min = +2.40 V, max = +3.60 V)
> fan1: 0 RPM (min = 664 RPM, div = 8) ALARM
> CPU Fan: 0 RPM (min = 664 RPM, div = 8) ALARM
> fan3: 0 RPM (min = 664 RPM, div = 8) ALARM
> M/B Temp: +6 C (high = -87 C, hyst = +125 C) sensor = diode
> CPU Temp: -7.5 C (high = +80 C, hyst = +75 C) sensor = diode
> temp3: +7.5 C (high = +80 C, hyst = +75 C) sensor = diode (beep)
> alarms:
> beep_enable:
> Sound alarm enabled
>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 258 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060326/a4fa1b1b/signature.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
` (4 preceding siblings ...)
2006-03-26 9:43 ` Harald Dunkel
@ 2006-03-31 12:35 ` Jean Delvare
2006-04-01 5:44 ` Harald Dunkel
` (5 subsequent siblings)
11 siblings, 0 replies; 23+ messages in thread
From: Jean Delvare @ 2006-03-31 12:35 UTC (permalink / raw)
To: lm-sensors
Hi Harald,
> Sorry for the late response. See below.
What can I say... ;)
> {root at bugs:harri 1001} i2cdump 0 0x2d b
> WARNING! This program can confuse your I2C bus, cause data loss and worse!
> I will probe file /dev/i2c-0, address 0x2d, mode byte
> Continue? [Y/n]
> 0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
> 00: 01 07 00 40 5a ff ff ff ff ff ff ff ff ff ff ff ??. at Z...........
> 10: c5 72 7e 85 25 22 01 df 0f fe 00 00 1c 80 ad 07 ?r~?%"????..????
> 20: ff 00 ff 00 ff 00 ff 00 4b 4b 4b 4b 3c 37 00 ff ........KKKK<7..
> 30: 00 b0 b0 00 00 00 30 ff ff ff ff ff ff ff ff ff .??...0.........
> 40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 50: ff ff ff ff ff ff ff ff ff ff 04 10 20 19 34 ff ..........?? ?4.
> 60: bb 0a 90 00 01 00 00 0f ff 55 64 46 64 46 ff ff ???.?..?.UdFdF..
> 70: 02 ab ff ff 00 00 ff ff ff ff ff ff ff ff ff ff ??..............
> 80: 0f ff ff ff 00 00 ff ff ff ff ff ff ff ff ff ff ?...............
> 90: 00 00 0d ff ff ff ff ff ff ff ff ff ff ff ff ff ..?.............
> a0: 46 3c 32 28 ff d9 b2 99 80 ff ff ff ff ff ff ff F<2(.????.......
> b0: 46 3c 32 28 ff d9 b2 99 80 ff ff ff ff ff ff ff F<2(.????.......
> c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> f0: 00 00 00 00 00 00 39 38 01 45 00 ff ff ff ff 27 ......98?E.....'
This could be a hardware monitoring chip, according to the number of
registers, and some interesting patterns (0xff and 0x00 repeated at
0x20-0x27 could be input voltage low and high limits, 0x64 and 0x46
repeated at 0x6a could be temperature limits...) but the mapping is not
usual and I can't find any obvious ID register.
You'll have to search the available technical documentation for your
system and/or ask Aopen for information. There's nothing we can do as
long as we don't know what chip we are dealing with.
> {root at bugs:harri 1002} i2cdump 0 0x2f b
> WARNING! This program can confuse your I2C bus, cause data loss and worse!
> I will probe file /dev/i2c-0, address 0x2f, mode byte
> Continue? [Y/n]
> 0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
> 00: ff 03 07 f0 22 08 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: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 30: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 40: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 50: ff ff ff ff ff ff ff ff ff ff 03 02 12 19 34 ff ..........????4.
> 60: ff ff ff ff ff ff ff ff ff ff a0 a0 ff ff ff ff ..........??....
> 70: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> 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: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> b0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> c0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> d0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> e0: ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ff ................
> f0: 00 80 ff ff ff ff ff ff ff ff ff ff ff ff ff 5e .?.............^
I was about to say "No idea about this one", as it has very few
registers and no noticeable patterns, but that "19 34" at 0x5d reminds
me of the Fintek signature in my F71805F Super-I/O chip. Can you please
search your motherboard for a Fintek chip, presumably small?
And now I realize that the chip at 0x2d has the same values at address
0x5d, so it must be a Fintek chip too.
I'll go ask my Fintek contact for additional information while you try
to locate the chips on your board.
--
Jean Delvare
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
` (5 preceding siblings ...)
2006-03-31 12:35 ` Jean Delvare
@ 2006-04-01 5:44 ` Harald Dunkel
2006-04-01 12:48 ` Jean Delvare
` (4 subsequent siblings)
11 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-04-01 5:44 UTC (permalink / raw)
To: lm-sensors
Hi Jean,
Jean Delvare wrote:
>
> You'll have to search the available technical documentation for your
> system and/or ask Aopen for information. There's nothing we can do as
> long as we don't know what chip we are dealing with.
>
:
>
> I was about to say "No idea about this one", as it has very few
> registers and no noticeable patterns, but that "19 34" at 0x5d reminds
> me of the Fintek signature in my F71805F Super-I/O chip. Can you please
> search your motherboard for a Fintek chip, presumably small?
>
I found a small chip close to the socket for an additional fan.
16 pins. AFAICS it says F75387SG, but it was very hard to read
even with a looking glass. Esp. the "S" looks suspicious.
Using "fintek F75387SG" I found this on Google:
http://www.fintek.com.tw/files/productfiles/F75387_025P%20datasheet.pdf
BTW, my PC is an Aopen MZ915-M .
Hope this helps. Regards
Harri
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 254 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060401/41d5029f/signature.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
` (6 preceding siblings ...)
2006-04-01 5:44 ` Harald Dunkel
@ 2006-04-01 12:48 ` Jean Delvare
2006-04-01 13:59 ` Harald Dunkel
` (3 subsequent siblings)
11 siblings, 0 replies; 23+ messages in thread
From: Jean Delvare @ 2006-04-01 12:48 UTC (permalink / raw)
To: lm-sensors
Hi Harald,
> > I was about to say "No idea about this one", as it has very few
> > registers and no noticeable patterns, but that "19 34" at 0x5d reminds
> > me of the Fintek signature in my F71805F Super-I/O chip. Can you please
> > search your motherboard for a Fintek chip, presumably small?
>
> I found a small chip close to the socket for an additional fan.
> 16 pins. AFAICS it says F75387SG, but it was very hard to read
> even with a looking glass. Esp. the "S" looks suspicious.
>
> Using "fintek F75387SG" I found this on Google:
>
> http://www.fintek.com.tw/files/productfiles/F75387_025P%20datasheet.pdf
>
> BTW, my PC is an Aopen MZ915-M .
Good catch. This is, indeed, the chip that sensors-detect spotted at
address 0x2d. I just modified this script so that it now knows about
more Fintek chips. Please give it a try:
http://www2.lm-sensors.nu/%7Elm78/cvs/lm_sensors2/prog/detect/sensors-detect
It should properly identify your F75387SG/RG chip at 0x2d now.
I still don't know what is the other Fintek chip at 0x2f. Its it
doesn't match any ID I could find in the datasheets, but it's similar
enough to the F75121R and F75122R/RG ID that I'd bet for some VID
and/or GPIO chip. My contact at Fintek told me it was a "power
control", I hope to have additional information soon. Maybe you can
take a look at your motherboard again in the meantime.
So, the good news is that we now know what hardware monitoring chip you
have. This explains why the W83627THF chip was not reporting the values
you were expecting.
The bad news is that we have no driver for this chip at the moment.
I've added it to our "New drivers" page, together with a few other
Fintek chips. You'll have to wait for a driver to be written, or write
it yourself. Any amount of support (work, money, hardware...) you can
offer is likely to speed up the process.
--
Jean Delvare
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
` (7 preceding siblings ...)
2006-04-01 12:48 ` Jean Delvare
@ 2006-04-01 13:59 ` Harald Dunkel
2006-04-04 18:07 ` Harald Dunkel
` (2 subsequent siblings)
11 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-04-01 13:59 UTC (permalink / raw)
To: lm-sensors
Hi Jean,
Jean Delvare wrote:
>>
>> Using "fintek F75387SG" I found this on Google:
>>
>> http://www.fintek.com.tw/files/productfiles/F75387_025P%20datasheet.pdf
>>
>> BTW, my PC is an Aopen MZ915-M .
>
> Good catch. This is, indeed, the chip that sensors-detect spotted at
> address 0x2d. I just modified this script so that it now knows about
> more Fintek chips. Please give it a try:
> http://www2.lm-sensors.nu/%7Elm78/cvs/lm_sensors2/prog/detect/sensors-detect
>
> It should properly identify your F75387SG/RG chip at 0x2d now.
>
Yup:
Driver `to-be-written' (should be inserted):
Detects correctly:
* Bus `SMBus I801 adapter at 5000'
Busdriver `i2c-i801', I2C address 0x2d
Chip `Fintek F75387SG/RG' (confidence: 7)
Driver `w83627hf' (should be inserted):
Detects correctly:
* ISA bus address 0x0290 (Busdriver `i2c-isa')
Chip `Winbond W83627THF Super IO Sensors' (confidence: 9)
> I still don't know what is the other Fintek chip at 0x2f. Its it
> doesn't match any ID I could find in the datasheets, but it's similar
> enough to the F75121R and F75122R/RG ID that I'd bet for some VID
> and/or GPIO chip. My contact at Fintek told me it was a "power
> control", I hope to have additional information soon. Maybe you can
> take a look at your motherboard again in the meantime.
>
This is some kind of "high-density" PC, i.e. it is hard to disassemble.
I was lucky that the first chip was easy to spot, but the second one
could be anywhere. I would suggest to postpone this to tomorrow.
"Power control" reminds me of something: Sometimes ACPI power off
doesn't work. Do you think this is a related problem?
> So, the good news is that we now know what hardware monitoring chip you
> have. This explains why the W83627THF chip was not reporting the values
> you were expecting.
>
> The bad news is that we have no driver for this chip at the moment.
> I've added it to our "New drivers" page, together with a few other
> Fintek chips. You'll have to wait for a driver to be written, or write
> it yourself. Any amount of support (work, money, hardware...) you can
> offer is likely to speed up the process.
>
Creating a driver on my own sounds pretty cool. Probably it is
allowed to use an existing driver as a template?
Regards
Harri
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 254 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060401/825477b4/signature-0001.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
` (8 preceding siblings ...)
2006-04-01 13:59 ` Harald Dunkel
@ 2006-04-04 18:07 ` Harald Dunkel
2006-04-04 19:49 ` Jean Delvare
2006-04-04 19:59 ` Jean Delvare
11 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-04-04 18:07 UTC (permalink / raw)
To: lm-sensors
Hi Jean,
Harald Dunkel wrote:
> Hi Jean,
>
> Jean Delvare wrote:
>
>> I still don't know what is the other Fintek chip at 0x2f. Its it
>> doesn't match any ID I could find in the datasheets, but it's similar
>> enough to the F75121R and F75122R/RG ID that I'd bet for some VID
>> and/or GPIO chip. My contact at Fintek told me it was a "power
>> control", I hope to have additional information soon. Maybe you can
>> take a look at your motherboard again in the meantime.
>>
>
> This is some kind of "high-density" PC, i.e. it is hard to disassemble.
> I was lucky that the first chip was easy to spot, but the second one
> could be anywhere. I would suggest to postpone this to tomorrow.
>
I have opened the case, removed the power supply and the harddisk,
but I did not find another Fintek chip. Is it possible that it
is close to the CPU, maybe below the cooler, or so?
Maybe it is integrated on the chip we found?
Regards
Harri
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 254 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060404/6943ae42/signature.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
` (9 preceding siblings ...)
2006-04-04 18:07 ` Harald Dunkel
@ 2006-04-04 19:49 ` Jean Delvare
2006-04-04 19:59 ` Jean Delvare
11 siblings, 0 replies; 23+ messages in thread
From: Jean Delvare @ 2006-04-04 19:49 UTC (permalink / raw)
To: lm-sensors
Hallo Harald,
> > This is some kind of "high-density" PC, i.e. it is hard to disassemble.
> > I was lucky that the first chip was easy to spot, but the second one
> > could be anywhere. I would suggest to postpone this to tomorrow.
>
> I have opened the case, removed the power supply and the harddisk,
> but I did not find another Fintek chip. Is it possible that it
> is close to the CPU, maybe below the cooler, or so?
>
> Maybe it is integrated on the chip we found?
I doubt it. My Fintek contact told me it is a custom ASIC they made
for just one company, so I'm almost certain it's a separate chip. Maybe
it doesn't even have Fintek's usual top marking because of the "custom"
aspect.
No need to search any further anyway, finding the chip won't help us.
Fintek cannot provide a datasheet for custom ASICs. And that chip might
not need a driver in the first place.
--
Jean Delvare
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-03-12 16:19 [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by Harald Dunkel
` (10 preceding siblings ...)
2006-04-04 19:49 ` Jean Delvare
@ 2006-04-04 19:59 ` Jean Delvare
11 siblings, 0 replies; 23+ messages in thread
From: Jean Delvare @ 2006-04-04 19:59 UTC (permalink / raw)
To: lm-sensors
Harald,
> "Power control" reminds me of something: Sometimes ACPI power off
> doesn't work. Do you think this is a related problem?
I don't think so. The other I2C chips we've seen which were dealing
with "power" were usually VID control chips, which can be used to
slightly alter the CPU Vcore. This can be used for power-saving
purposes or overclocking purposes, depending on the change direction.
My understanding is that these chips should not be needed for regular
users. But then again, I don't know what _this_ chip does at all, so I
can't tell.
> Creating a driver on my own sounds pretty cool. Probably it is
> allowed to use an existing driver as a template?
Not allowed. Required ;)
Not absolutely required, in fact, but strongly encouraged, as this will
make it much easier for you. Hardware monitoring drivers have many
common points, and if you choose your template(s) properly, the new
driver can be done in no time and its review will be much faster too.
I didn't look at the datasheet for your chip yet (no time for this
sorry) so I can't tell you exactly which driver to start from. But this
should be an i2c-only driver (no platform driver, and no i2c-isa
trick.) Maybe lm90 will do, as it's both simple and up-to-date with
regards to recent changes. And you can always peek at other drivers for
the features your chip has and lm90 doesn't.
--
Jean Delvare
^ permalink raw reply [flat|nested] 23+ messages in thread
* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-05-19 16:24 Harald Dunkel
@ 2006-05-20 9:03 ` Jean Delvare
2006-05-26 10:46 ` Harald Dunkel
` (7 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: Jean Delvare @ 2006-05-20 9:03 UTC (permalink / raw)
To: lm-sensors
Hi Harald,
> > I didn't look at the datasheet for your chip yet (no time for this
> > sorry) so I can't tell you exactly which driver to start from. But this
> > should be an i2c-only driver (no platform driver, and no i2c-isa
> > trick.) Maybe lm90 will do, as it's both simple and up-to-date with
> > regards to recent changes. And you can always peek at other drivers for
> > the features your chip has and lm90 doesn't.
>
> I have started working on this (but please don't expect too much).
> One question about the I2C_DRIVERID_* macros in i2c-id.h: Where
> could I get an ID for the f75387 driver?
You don't need any. id is an optional field, just set it to 0 (the
default for a statically declared structure anyway) and it'll be fine.
--
Jean Delvare
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-05-19 16:24 Harald Dunkel
2006-05-20 9:03 ` Jean Delvare
@ 2006-05-26 10:46 ` Harald Dunkel
2006-05-26 12:10 ` Jean Delvare
` (6 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-05-26 10:46 UTC (permalink / raw)
To: lm-sensors
Hi Jean,
Harald Dunkel wrote:
>
> I have started working on this (but please don't expect too much).
> One question about the I2C_DRIVERID_* macros in i2c-id.h: Where
> could I get an ID for the f75387 driver?
>
The datasheet of the F75387 doesn't say anything about Packet
Error Checking, but I am not sure whether this is a feature of
the bus. Is it?
Can something bad happen if I don't support PEC?
Regards
Harri
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060526/5a6e32db/signature.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-05-19 16:24 Harald Dunkel
2006-05-20 9:03 ` Jean Delvare
2006-05-26 10:46 ` Harald Dunkel
@ 2006-05-26 12:10 ` Jean Delvare
2006-05-26 19:02 ` Harald Dunkel
` (5 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: Jean Delvare @ 2006-05-26 12:10 UTC (permalink / raw)
To: lm-sensors
Hi Harald,
Harald Dunkel wrote:
> The datasheet of the F75387 doesn't say anything about Packet
> Error Checking, but I am not sure whether this is a feature of
> the bus. Is it?
>
> Can something bad happen if I don't support PEC?
PEC is an optional protocol extension, it can be used only when both
the master and the slave support it.
The Fintek F75387 datasheet doesn't mention PEC, so I would assume it
doesn't support it. Chips supporting PEC are the exception rather than
the rule, only the ADM1032 chip (lm90 driver) supports it in our whole
driver set. So just forget about PEC, and everything will be fine.
--
Jean Delvare
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-05-19 16:24 Harald Dunkel
` (2 preceding siblings ...)
2006-05-26 12:10 ` Jean Delvare
@ 2006-05-26 19:02 ` Harald Dunkel
2006-05-28 17:24 ` Harald Dunkel
` (4 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-05-26 19:02 UTC (permalink / raw)
To: lm-sensors
Hi Jean,
I've got some very first version, reading temperatures, fan speeds,
and voltage. See attachment. (Hopefully it is OK to post it here?)
Except that the values it reports don't match the values shown
in the bios. The temperatures shown in the bios are about 4 degrees
higher (53 to 54 degree Celsius), the voltages don't match, and the
fan speeds are completely mad:
% cat /sys/bus/i2c/devices/0-002d/fan?_input
471
4094
% cat /sys/bus/i2c/devices/0-002d/temp?_input
49125
49125
40250
% cat /sys/bus/i2c/devices/0-002d/in?_input
1576
912
1008
1016
If I compare this to the output of i2cdump, then I don't see a
difference. AFAICS the /sys entries and the registers match.
Any idea? Is it possible that the datasheet is not correct?
Regards
Harri
-------------- next part --------------
A non-text attachment was scrubbed...
Name: f75387.c
Type: text/x-csrc
Size: 15096 bytes
Desc: not available
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060526/935fe4c6/f75387-0001.bin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060526/935fe4c6/signature-0001.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-05-19 16:24 Harald Dunkel
` (3 preceding siblings ...)
2006-05-26 19:02 ` Harald Dunkel
@ 2006-05-28 17:24 ` Harald Dunkel
2006-06-10 7:33 ` Harald Dunkel
` (3 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-05-28 17:24 UTC (permalink / raw)
To: lm-sensors
Hi Jean,
Attached is a new version of the f75387 driver. The fan speed is
correct now, except that the missing fan is not detected. fan1_input
is 550, not 0.
All values are still read-only.
What is the procedure to get support for a new chip into the
lm-sensors package?
Regards
Harri
-------------- next part --------------
A non-text attachment was scrubbed...
Name: f75387.c
Type: text/x-csrc
Size: 18315 bytes
Desc: not available
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060528/dca48a78/f75387.bin
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060528/dca48a78/signature.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-05-19 16:24 Harald Dunkel
` (4 preceding siblings ...)
2006-05-28 17:24 ` Harald Dunkel
@ 2006-06-10 7:33 ` Harald Dunkel
2006-08-19 9:39 ` Rudolf Marek
` (2 subsequent siblings)
8 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-06-10 7:33 UTC (permalink / raw)
To: lm-sensors
Hi folks,
It would be very nice if you could send me some feedback. I
haven't created a kernel driver before, but IMHO this should
not be a reason to ignore my efforts in this completely.
Regards
Harri
==============================Harald Dunkel wrote:
> Hi Jean,
>
> Attached is a new version of the f75387 driver. The fan speed is
> correct now, except that the missing fan is not detected. fan1_input
> is 550, not 0.
>
> All values are still read-only.
>
> What is the procedure to get support for a new chip into the
> lm-sensors package?
>
>
> Regards
>
> Harri
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060610/cc2880b3/attachment-0001.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-05-19 16:24 Harald Dunkel
` (5 preceding siblings ...)
2006-06-10 7:33 ` Harald Dunkel
@ 2006-08-19 9:39 ` Rudolf Marek
2006-08-20 5:50 ` Harald Dunkel
2006-08-20 10:44 ` Rudolf Marek
8 siblings, 0 replies; 23+ messages in thread
From: Rudolf Marek @ 2006-08-19 9:39 UTC (permalink / raw)
To: lm-sensors
Harald Dunkel wrote:
> Hi folks,
>
> It would be very nice if you could send me some feedback. I
> haven't created a kernel driver before, but IMHO this should
> not be a reason to ignore my efforts in this completely.
>
Hi,
We do not ignore, I just returned from 2-3 weeks off. Jean was alone here...
Not easy to handle all those patches in two... Or alone..
Remember that we are doing this in free time and it eats a lot of it.
I created a ticket with for your driver review.
Regards
Rudolf
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-05-19 16:24 Harald Dunkel
` (6 preceding siblings ...)
2006-08-19 9:39 ` Rudolf Marek
@ 2006-08-20 5:50 ` Harald Dunkel
2006-08-20 10:44 ` Rudolf Marek
8 siblings, 0 replies; 23+ messages in thread
From: Harald Dunkel @ 2006-08-20 5:50 UTC (permalink / raw)
To: lm-sensors
Hi Rudolf,
I've got one general question:
AFAICS the values measured by the sensor chips are pretty
far away from reality, as it seems. The mapping to real
temperatures, and voltages is done in the sensors tool
in user space.
But: Ain't this extremely hardware dependent? Looking at the
same sensor built into several different mainboards of different
manufacturers, I would guess that the range for the measured
temperatures or voltages can vary alot, depending upon some
external resistors, the position of the chip on the main
board, etc. Every second board needs its own sensors.conf,
does it?
Or is there some kind of standard, at least for the major
chips?
Regards
Harri
-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 252 bytes
Desc: OpenPGP digital signature
Url : http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060820/c730fca2/attachment.bin
^ permalink raw reply [flat|nested] 23+ messages in thread* [lm-sensors] 2.6.16-rc5: unreasonable temperature reported by
2006-05-19 16:24 Harald Dunkel
` (7 preceding siblings ...)
2006-08-20 5:50 ` Harald Dunkel
@ 2006-08-20 10:44 ` Rudolf Marek
8 siblings, 0 replies; 23+ messages in thread
From: Rudolf Marek @ 2006-08-20 10:44 UTC (permalink / raw)
To: lm-sensors
Hi Harald,
> AFAICS the values measured by the sensor chips are pretty
> far away from reality, as it seems. The mapping to real
> temperatures, and voltages is done in the sensors tool
> in user space.
>
> But: Ain't this extremely hardware dependent? Looking at the
> same sensor built into several different mainboards of different
> manufacturers, I would guess that the range for the measured
> temperatures or voltages can vary alot, depending upon some
> external resistors, the position of the chip on the main
> board, etc. Every second board needs its own sensors.conf,
> does it?
>
Well same board should have same sensors.conf. But it is upon manufacturer to
wire it as they like. This is the reason for sensors.conf
The driver exports the voltage value measured on ADC pin, but this voltage needs
scaling because for example 5V needs to be in range of 0-2.048V The scaling
depends on hardware used (two resistors) and there are many ways how to fit 5V
into this range.
Temperatures sometimes needs some offsets to add or even more complicated formulas.
> Or is there some kind of standard, at least for the major
> chips?
Board manufacturer may follow the recommendations found in datasheet for that
particular chip. I belive more standards will show when the PECI and SST will
become more popular for intel chips.
Regards
Rudolf
^ permalink raw reply [flat|nested] 23+ messages in thread