[lm-sensors] hwmon/pc87360 as a platform driver

[lm-sensors] hwmon/pc87360 as a platform driver

[Jim Cromie]

hi folks,

Ive just made a rough pass thru pc87360, and have converted it
to a non-working, but non-crashing platform_driver that does
at least a little of what it should:

soekris:~# modprobe pc87360
[  964.952000] pc87360: Device 0x09 not activated
[  966.576000] pc87360 pc87360.26144: VLM conversion set to 1s period, 
160us delay
soekris:~#


soekris:~# ls /sys/bus/platform/devices/pc87360.26144/
alarms_in      in1_input   in4_min     in8_input     temp2_input   
temp5_crit
alarms_temp    in1_max     in4_status  in8_max       temp2_max     
temp5_input
bus@           in1_min     in5_input   in8_min       temp2_min     temp5_max
cpu0_vid       in1_status  in5_max     in8_status    temp2_status  temp5_min
driver@        in2_input   in5_min     in9_input     temp3_crit    
temp5_status
hwmon:hwmon0@  in2_max     in5_status  in9_max       temp3_input   
temp6_crit
in0_input      in2_min     in6_input   in9_min       temp3_max     
temp6_input
in0_max        in2_status  in6_max     in9_status    temp3_min     temp6_max
in0_min        in3_input   in6_min     temp1_crit    temp3_status  temp6_min
in0_status     in3_max     in6_status  temp1_input   temp4_crit    
temp6_status
in10_input     in3_min     in7_input   temp1_max     temp4_input   uevent
in10_max       in3_status  in7_max     temp1_min     temp4_max     vrm
in10_min       in4_input   in7_min     temp1_status  temp4_min
in10_status    in4_max     in7_status  temp2_crit    temp4_status


soekris:~# ll /sys/bus/platform/drivers/pc87360/
total 0
--w------- 1 root root 4096 Apr  7 12:08 bind
lrwxrwxrwx 1 root root    0 Apr  7 12:08 module -> 
../../../../module/pc87360/
lrwxrwxrwx 1 root root    0 Apr  7 12:08 pc87360.26144 -> 
../../../../devices/platform/pc87360.26144/
--w------- 1 root root 4096 Apr  7 12:08 unbind



so far so good, but:

soekris:~# sensors -s
No sensors found!
soekris:~#
soekris:~# sensors
No sensors found!


my libsensors is unstable debian package:  libsensors 1:2.10.0-3
That version number looks to contain 2.10.0

Frankly, Im hoping that with 2.10, the user-side just works, but this is 
not the case here.

Any guesses / suggestions as to whats wrong ?
Im continuiing to look at the kernel-side (Im sure Ive got multiple errs 
in there)
Im hoping for some user-side-fu for free ;-)


for anyone reading along, this is a perfect ref:
    
http://lists.lm-sensors.org/pipermail/lm-sensors/2006-February/015180.html

given the wider staging issues, this is arguably a later/lower priority.


tia
jimc

[lm-sensors] hwmon/pc87360 as a platform driver

[Juerg Haefliger]

Jim,

I had a similar problem when I converted the vt1211 to a platform driver.
Turned out to be a bug in lm_sensors which required some I2C modules to
function properly. Try either loading an I2C module or use the CVS version
of lm_sensors.

Check the following thread:
http://lists.lm-sensors.org/pipermail/lm-sensors/2006-February/015358.html

...juerg


On 4/7/06, Jim Cromie <jim.cromie at gmail.com> wrote:
>
> hi folks,
>
> Ive just made a rough pass thru pc87360, and have converted it
> to a non-working, but non-crashing platform_driver that does
> at least a little of what it should:
>
> soekris:~# modprobe pc87360
> [  964.952000] pc87360: Device 0x09 not activated
> [  966.576000] pc87360 pc87360.26144: VLM conversion set to 1s period,
> 160us delay
> soekris:~#
>
>
> soekris:~# ls /sys/bus/platform/devices/pc87360.26144/
> alarms_in      in1_input   in4_min     in8_input     temp2_input
> temp5_crit
> alarms_temp    in1_max     in4_status  in8_max       temp2_max
> temp5_input
> bus@           in1_min     in5_input   in8_min       temp2_min
> temp5_max
> cpu0_vid       in1_status  in5_max
> in8_status    temp2_status  temp5_min
> driver@        in2_input   in5_min     in9_input     temp3_crit
> temp5_status
> hwmon:hwmon0@  in2_max     in5_status  in9_max       temp3_input
> temp6_crit
> in0_input      in2_min     in6_input   in9_min       temp3_max
> temp6_input
> in0_max        in2_status  in6_max     in9_status    temp3_min
> temp6_max
> in0_min        in3_input   in6_min
> temp1_crit    temp3_status  temp6_min
> in0_status     in3_max     in6_status  temp1_input   temp4_crit
> temp6_status
> in10_input     in3_min     in7_input   temp1_max     temp4_input   uevent
> in10_max       in3_status  in7_max     temp1_min     temp4_max     vrm
> in10_min       in4_input   in7_min     temp1_status  temp4_min
> in10_status    in4_max     in7_status  temp2_crit    temp4_status
>
>
> soekris:~# ll /sys/bus/platform/drivers/pc87360/
> total 0
> --w------- 1 root root 4096 Apr  7 12:08 bind
> lrwxrwxrwx 1 root root    0 Apr  7 12:08 module ->
> ../../../../module/pc87360/
> lrwxrwxrwx 1 root root    0 Apr  7 12:08 pc87360.26144 ->
> ../../../../devices/platform/pc87360.26144/
> --w------- 1 root root 4096 Apr  7 12:08 unbind
>
>
>
> so far so good, but:
>
> soekris:~# sensors -s
> No sensors found!
> soekris:~#
> soekris:~# sensors
> No sensors found!
>
>
> my libsensors is unstable debian package:  libsensors 1:2.10.0-3
> That version number looks to contain 2.10.0
>
> Frankly, Im hoping that with 2.10, the user-side just works, but this is
> not the case here.
>
> Any guesses / suggestions as to whats wrong ?
> Im continuiing to look at the kernel-side (Im sure Ive got multiple errs
> in there)
> Im hoping for some user-side-fu for free ;-)
>
>
> for anyone reading along, this is a perfect ref:
>
> http://lists.lm-sensors.org/pipermail/lm-sensors/2006-February/015180.html
>
> given the wider staging issues, this is arguably a later/lower priority.
>
>
> tia
> jimc
>
> _______________________________________________
> lm-sensors mailing list
> lm-sensors at lm-sensors.org
> http://lists.lm-sensors.org/mailman/listinfo/lm-sensors
>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.lm-sensors.org/pipermail/lm-sensors/attachments/20060407/27579a0b/attachment-0001.html

[lm-sensors] hwmon/pc87360 as a platform driver

[Jim Cromie]

Juerg Haefliger wrote:
> Jim,
>
> I had a similar problem when I converted the vt1211 to a platform 
> driver. Turned out to be a bug in lm_sensors which required some I2C 
> modules to function properly. Try either loading an I2C module or use 
> the CVS version of lm_sensors.
>
> Check the following thread:
> http://lists.lm-sensors.org/pipermail/lm-sensors/2006-February/015358.html
>
> ...juerg
>
>
> On 4/7/06, *Jim Cromie* <jim.cromie at gmail.com 
> <mailto:jim.cromie at gmail.com>> wrote:
>
>     hi folks,
>
>     Ive just made a rough pass thru pc87360, and have converted it
>     to a non-working, but non-crashing platform_driver that does
>     at least a little of what it should:
>

thanks!
but alas, error-driven cut-paste will only get me so far.
More to the point, I have code similar to yours and Jean's

    data->class_dev = hwmon_device_register(&pdev->dev);

Guess I'll have to roll up my sleeves and actually figure it out.


BTW, I glanced at your patch,  saw you're :
    using sensor_attribute_2's,
    that you're using the .nr field to switch the 'functionality'
        of your combined-functionality callbacks.
   
thats all quite similar to what Im doing in these:
http://lists.lm-sensors.org/pipermail/lm-sensors/2006-March/015702.html
http://lists.lm-sensors.org/pipermail/lm-sensors/2006-March/015703.html

The SHOW_SET_*_*  constants are just a bit off-putting at first read,
but I know what you mean, and I dont have a better idea.
maybe SHOW_SETTNG_*_* ?
or SHOW_CURR_*_*,   thats confuse-able with the current reading


you can improve your printks :

s/(printk\(KERN_DEBUG)/dev_dbg\(&pdev->dev/;
s/(printk\(KERN_ERR)/dev_err\(&pdev->dev/
s/(printk\(KERN_INFO)/dev_info\(&pdev->dev/

at least where pdev has already been initd


Ill try to read thru the enire patch this weekend sometime.

thanks
jimc

[lm-sensors] hwmon/pc87360 as a platform driver

[Jean Delvare]

Hi Jim,

> Ive just made a rough pass thru pc87360, and have converted it
> to a non-working, but non-crashing platform_driver that does
> at least a little of what it should:
> 
> soekris:~# modprobe pc87360
> [  964.952000] pc87360: Device 0x09 not activated
> [  966.576000] pc87360 pc87360.26144: VLM conversion set to 1s period, 
> 160us delay
> soekris:~#
> (...)
> so far so good, but:
> 
> soekris:~# sensors -s
> No sensors found!
> soekris:~#
> soekris:~# sensors
> No sensors found!
> 
> my libsensors is unstable debian package:  libsensors 1:2.10.0-3
> That version number looks to contain 2.10.0
> 
> Frankly, Im hoping that with 2.10, the user-side just works, but this is 
> not the case here.
> 
> Any guesses / suggestions as to whats wrong ?

First of all, Juerg's answer is correct, 2.10.0 doesn't play fair with
platform hwmon drivers unless an i2c driver is also present.

Secondly, your driver must have a "name" attribute containing the chip
name (such as "pc87360" or "pc87366" etc. in your case) else libsensors
won't be able to use it. That's probably what you're missing here.

BTW, I don't want to discourage you from working on this, as this will
be needed at some point in time, but remember that I do not plan to
accept any i2c-isa -> platform conversion before next year, because we
need a widely deployed base of working libsensors before we do, and
unfortunately even 2.10.0 doesn't qualify, as Juerg and you found out.

Thanks,
-- 
Jean Delvare

[lm-sensors] hwmon/pc87360 as a platform driver

[Juerg Haefliger]

> The SHOW_SET_*_*  constants are just a bit off-putting at first read,
> but I know what you mean, and I dont have a better idea.
> maybe SHOW_SETTNG_*_* ?
> or SHOW_CURR_*_*,   thats confuse-able with the current reading

Well SHOW_SET_xxx_yyy is supposed to hint that the constant is used in
both the show_xxx and set_xxx callbacks. It's the best I could come up
with although I have to admit I didn't spend too much time thinking
about this....

>
> you can improve your printks :
>
> s/(printk\(KERN_DEBUG)/dev_dbg\(&pdev->dev/;
> s/(printk\(KERN_ERR)/dev_err\(&pdev->dev/
> s/(printk\(KERN_INFO)/dev_info\(&pdev->dev/
>
> at least where pdev has already been initd

Hmm... I thought I used printks only in places where pdev is not initialized.

>
> Ill try to read thru the enire patch this weekend sometime.

I certainly appreciate the review :-)

...juerg