
Comment 1 by Nikolaus Schaller, Sep 25, 2013
more testing reveals that the hmc has a /sys node for raw data but it does not show values but returns 'invalid argument' errors: root@gta04:~# cat /sys/bus/iio/devices/iio:device1/* 252:1 1300 0.000009174 cat: /sys/bus/iio/devices/iio:device1/in_magn_x_raw: Invalid argument cat: /sys/bus/iio/devices/iio:device1/in_magn_y_raw: Invalid argument cat: /sys/bus/iio/devices/iio:device1/in_magn_z_raw: Invalid argument 0 hmc5883l 0 cat: /sys/bus/iio/devices/iio:device1/power: Is a directory cat: /sys/bus/iio/devices/iio:device1/sampling_frequency: Remote I/O error 0.75 1.5 3 7.5 15 30 75 cat: /sys/bus/iio/devices/iio:device1/subsystem: Is a directory MAJOR=252 MINOR=1 DEVNAME=iio:device1 DEVTYPE=iio_device root@gta04:~#

Comment 2 by Nikolaus Schaller, Sep 25, 2013
And, there is a dmesg: [ 2.890014] Error: Driver 'itg3200' is already registered, aborting... But it exists only once in the board file? Or is it automatically matched with the gta04_vaux2_supply[]?

Comment 3 by Nikolaus Schaller, Oct 1, 2013
I think I have found at least one issue with the ITG3200. We have two drivers! One from Neil (3.7-kernel) in staging/iio/itg3200.c and one introduced in 3.9: http://lxr.free-electrons.com/source/include/linux/iio/gyro/itg3200.h ?v=3.9 These may conflict and do other harmful things... So Neil's variant is now removed. We should check if the official driver has all functions we did have in Neil's variant.

Comment 5 by Nikolaus Schaller, Oct 1, 2013
With the new/correct driver, the ITG3200 works well. It even has a in_temp_raw file in the /sysfs. So hw-test is now possible for the itg chip. Only the hmc is still making problems as described above.
Status:
Accepted

Comment 6 by Nikolaus Schaller, Oct 7, 2013
after upgrading to 3.12-rc4 the hmc5883 appears to work again! Needs a little testing if it is working good now.
Sign in to reply to this comment.
Reported by Nikolaus Schaller, Sep 24, 2013