* Re: [lm-sensors] G760A not working korrekt on NAS dlink 323 HW C1
2010-03-22 21:14 [lm-sensors] G760A not working korrekt on NAS dlink 323 HW C1 Denny Schierz
@ 2010-03-23 12:43 ` Jean Delvare
2010-03-26 0:48 ` Denny Schierz
2010-03-29 15:25 ` Jean Delvare
2 siblings, 0 replies; 4+ messages in thread
From: Jean Delvare @ 2010-03-23 12:43 UTC (permalink / raw)
To: lm-sensors
Hi Denny,
On Mon, 22 Mar 2010 22:14:07 +0100, Denny Schierz wrote:
> I have a NAS from D-Link called DNS-323ยน and it has a fan controller
> G760A. the hardware revision ist C1. My problem is, after starting the
> kernel (2.6.33.1 (* in .config, nit [M] ), the fan stops. I'm not able
> to activate him, with echo, like:
>
> nas:~# cat /sys/class/hwmon/hwmon0/device/pwm1
> 18
>
> echo 0 > /sys/class/hwmon/hwmon0/device/pwm1
> nas:~# cat /sys/class/hwmon/hwmon0/device/pwm1
> 18
>
>
> echo 255 > /sys/class/hwmon/hwmon0/device/pwm1
> nas:~# cat /sys/class/hwmon/hwmon0/device/pwm1
> 18
>
> I do not know, what could be the problem, so I hope, anybody knows the
> problem:
>
> nas:~# sensors
> g760a-i2c-0-3e
> Adapter: mv64xxx_i2c adapter
> fan1: 2073 RPM ALARM
Does this value change over time? I guess not.
>
> lm75-i2c-0-48
> Adapter: mv64xxx_i2c adapter
> temp1: +48.5 C (high = +80.0 C, hyst = +75.0 C)
>
> ยน http://www.cyrius.com/debian/orion/d-link/dns-323/specs.html
>
> The revision C1 seems to be a little different, but I don't know, how
> can I figure out this.
>
> Any suggestions?
According to this page:
http://wiki.dns323.info/hardware:fan
fan speed control is different for the revision C1 hardware. This
suggests that the underlying hardware fan speed controller itself is
different. I would guess that you don't really have a G760A chip on
your board. The Linux kernel assumes it is there and instantiates such
a device, but in that case I would bet it is wrong.
The g760a driver lacks sanity checks when reading register values. It
doesn't catch errors reported by the underlying I2C adapter driver. I
bet that the 18 and 2073 values you get above are actually -ENODEV
being processed by the driver as if no error had occurred. -ENODEV is
-19, which is read as 255-19 = 237 if the error is ignored. The fan
speed computation formula in the driver is (clk*30)/(val*div) with
clk = 32768 and div = 2: (32768*30)/(237*2) = 2073, bingo.
For the time being, arch/arm/mach-orion5x/dns323-setup.c should
probably refrain from instantiating a g760a device on revision C1
boards. You'll have to discuss this with whoever is in charge of this
file.
Then you can try running i2cdetect on the I2C adapter, to see if a
mysterious device has been added. The I2C device map is supposed to be:
/*
* On the DNS-323 the following devices are attached via I2C:
*
* i2c addr | chip | description
* 0x3e | GMT G760Af | fan speed PWM controller
* 0x48 | GMT G751-2f | temp. sensor and therm. watchdog (LM75 compatible)
* 0x68 | ST M41T80 | RTC w/ alarm
*/
but I guess yours will be different.
--
Jean Delvare
http://khali.linux-fr.org/wishlist.html
_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [lm-sensors] G760A not working korrekt on NAS dlink 323 HW C1
2010-03-22 21:14 [lm-sensors] G760A not working korrekt on NAS dlink 323 HW C1 Denny Schierz
2010-03-23 12:43 ` Jean Delvare
@ 2010-03-26 0:48 ` Denny Schierz
2010-03-29 15:25 ` Jean Delvare
2 siblings, 0 replies; 4+ messages in thread
From: Denny Schierz @ 2010-03-26 0:48 UTC (permalink / raw)
To: lm-sensors
[-- Attachment #1.1: Type: text/plain, Size: 4814 bytes --]
hi,
i was able to locate the chip on the board. His partnumber is AX3106
(PWM controller)
http://www.micro-bridge.com/data/Axelite/AX3106.pdf
Some debug:
nas:~# i2cdetect -l
i2c-0 i2c mv64xxx_i2c adapter I2C
adapter
nas:~# 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: -- -- -- -- -- -- -- -- -- 0c -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- UU --
40: -- -- -- -- -- -- -- -- UU -- -- -- -- -- -- --
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- UU -- -- -- -- -- -- --
70: -- -- -- -- -- -- -- --
so, that is the output, what now? I contacted the maintainer from the
file dns323-setup.c, but he has no time nor the hardware :-/
any suggestions?
cu denny
nas:~# i2cdump 0 0x0c
No size specified (using byte-data access)
WARNING! This program can confuse your I2C bus, cause data loss and
worse!
I will probe file /dev/i2c-0, address 0xc, mode byte
Continue? [Y/n] y
0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
00: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
10: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
20: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
30: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
40: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
50: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
60: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
70: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
80: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
90: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
a0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
b0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
c0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
d0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
e0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
f0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
nas:~# i2cdump 0 0x3
No size specified (using byte-data access)
WARNING! This program can confuse your I2C bus, cause data loss and
worse!
I will probe file /dev/i2c-0, address 0x3, mode byte
Continue? [Y/n]
0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
00: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
10: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
20: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
30: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
40: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
50: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
60: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
70: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
80: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
90: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
a0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
b0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
c0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
d0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
e0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
f0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
Am Dienstag, den 23.03.2010, 13:43 +0100 schrieb Jean Delvare:
> For the time being, arch/arm/mach-orion5x/dns323-setup.c should
> probably refrain from instantiating a g760a device on revision C1
> boards. You'll have to discuss this with whoever is in charge of this
> file.
>
> Then you can try running i2cdetect on the I2C adapter, to see if a
> mysterious device has been added. The I2C device map is supposed to
> be:
>
> /*
> * On the DNS-323 the following devices are attached via I2C:
> *
> * i2c addr | chip | description
> * 0x3e | GMT G760Af | fan speed PWM controller
> * 0x48 | GMT G751-2f | temp. sensor and therm. watchdog (LM75
> compatible)
> * 0x68 | ST M41T80 | RTC w/ alarm
> */
>
> but I guess yours will be different.
[-- Attachment #1.2: Dies ist ein digital signierter Nachrichtenteil --]
[-- Type: application/pgp-signature, Size: 197 bytes --]
[-- Attachment #2: Type: text/plain, Size: 153 bytes --]
_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
^ permalink raw reply [flat|nested] 4+ messages in thread* Re: [lm-sensors] G760A not working korrekt on NAS dlink 323 HW C1
2010-03-22 21:14 [lm-sensors] G760A not working korrekt on NAS dlink 323 HW C1 Denny Schierz
2010-03-23 12:43 ` Jean Delvare
2010-03-26 0:48 ` Denny Schierz
@ 2010-03-29 15:25 ` Jean Delvare
2 siblings, 0 replies; 4+ messages in thread
From: Jean Delvare @ 2010-03-29 15:25 UTC (permalink / raw)
To: lm-sensors
Hi Denny,
On Fri, 26 Mar 2010 01:48:11 +0100, Denny Schierz wrote:
> hi,
>
> i was able to locate the chip on the board. His partnumber is AX3106
> (PWM controller)
>
> http://www.micro-bridge.com/data/Axelite/AX3106.pdf
Not enough details... Not even sure if this chip is meant to control
fans, and no clue how it is accessed (I2C or something else.)
> Some debug:
>
> nas:~# i2cdetect -l
> i2c-0 i2c mv64xxx_i2c adapter I2C
> adapter
>
> nas:~# 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: -- -- -- -- -- -- -- -- -- 0c -- -- --
> 10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
> 20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
> 30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- UU --
> 40: -- -- -- -- -- -- -- -- UU -- -- -- -- -- -- --
> 50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
> 60: -- -- -- -- -- -- -- -- UU -- -- -- -- -- -- --
> 70: -- -- -- -- -- -- -- --
>
>
> so, that is the output, what now?
Unfortunately, with the new i2c device driver binding model, the output
of i2cdetect isn't that useful, because every device that has been
instantiated by the kernel (even if the physical device doesn't exist)
shows up. The devices above (at 0x3e, 0x48 and 0x68) are the ones
listed in the platform code.
We really need to update i2cdetect and the i2c-dev driver to better
deal with the new binding model. But this isn't that easy.
The only other one is 0x0c, which is the SMBus alert address. Support
for this feature was added in kernel 2.6.34. This won't help you for
the problem at hand though.
I contacted the maintainer from the
> file dns323-setup.c, but he has no time nor the hardware :-/
>
> any suggestions?
>
> cu denny
>
>
> nas:~# i2cdump 0 0x0c
> No size specified (using byte-data access)
> WARNING! This program can confuse your I2C bus, cause data loss and
> worse!
> I will probe file /dev/i2c-0, address 0xc, mode byte
> Continue? [Y/n] y
> 0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
> 00: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> 10: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> 20: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> 30: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> 40: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> 50: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> 60: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> 70: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> 80: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> 90: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> a0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> b0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> c0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> d0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> e0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
> f0: 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 91 ????????????????
FWIW, it means that device at 0x48 (0x91 / 2) has asked the kernel for
attention. Probably an over-temperature condition.
>
>
> nas:~# i2cdump 0 0x3
> No size specified (using byte-data access)
> WARNING! This program can confuse your I2C bus, cause data loss and
> worse!
> I will probe file /dev/i2c-0, address 0x3, mode byte
> Continue? [Y/n]
> 0 1 2 3 4 5 6 7 8 9 a b c d e f 0123456789abcdef
> 00: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> 10: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> 20: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> 30: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> 40: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> 50: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> 60: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> 70: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> 80: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> 90: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> a0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> b0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> c0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> d0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> e0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
> f0: XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XX XXXXXXXXXXXXXXXX
Not sure what was your idea with this, there's no chip listed at 0x03.
Trying 0x3e would have been more useful.
That being said, I fear I can't help you much further without the
hardware in question in my hands (and even then...)
--
Jean Delvare
http://khali.linux-fr.org/wishlist.html
_______________________________________________
lm-sensors mailing list
lm-sensors@lm-sensors.org
http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
^ permalink raw reply [flat|nested] 4+ messages in thread