Correctly test directory on Linux prior to 3.15

[gaul]

Previously the fopen write mode caused the call to spuriously report
newer sensors. Now we check the newer sensors path explicitly. Also
plug file descriptor leak and more carefully check return values.

[dgraziotin]

This fails on my machine (MacBook Pro 13" 2015, Ubuntu 16.10, Linux 4.8.0-22-generic),

$ sudo bin/mbpfan -f -v
Writing a new .pid file with value 13142 at: /var/run/mbpfan.pid
mbpfan[13142]: Writing a new .pid file with value 13142 at: /var/run/mbpfan.pid
Successfully written a new .pid file with value 13142 at: /var/run/mbpfan.pid
mbpfan[13142]: Successfully written a new .pid file with value 13142 at: /var/run/mbpfan.pid
Using legacy sensor path for kernel < 3.15.0
Found 0 sensors
mbpfan[13142]: mbpfan could not detect any temp sensor. Please contact the developer.
mbpfan could not detect any temp sensor. Please contact the developer.

[gaul]

Perhaps mbpfan needs to test for legacy kernels some other way? Can you share your output from the following command:

$ find /sys/devices/platform/coretemp.0/hwmon
/sys/devices/platform/coretemp.0/hwmon
/sys/devices/platform/coretemp.0/hwmon/hwmon0
/sys/devices/platform/coretemp.0/hwmon/hwmon0/power
/sys/devices/platform/coretemp.0/hwmon/hwmon0/power/control
/sys/devices/platform/coretemp.0/hwmon/hwmon0/power/async
/sys/devices/platform/coretemp.0/hwmon/hwmon0/power/runtime_enabled
/sys/devices/platform/coretemp.0/hwmon/hwmon0/power/runtime_active_kids
/sys/devices/platform/coretemp.0/hwmon/hwmon0/power/runtime_active_time
/sys/devices/platform/coretemp.0/hwmon/hwmon0/power/autosuspend_delay_ms
/sys/devices/platform/coretemp.0/hwmon/hwmon0/power/runtime_status
/sys/devices/platform/coretemp.0/hwmon/hwmon0/power/runtime_usage
/sys/devices/platform/coretemp.0/hwmon/hwmon0/power/runtime_suspended_time
/sys/devices/platform/coretemp.0/hwmon/hwmon0/device
/sys/devices/platform/coretemp.0/hwmon/hwmon0/subsystem
/sys/devices/platform/coretemp.0/hwmon/hwmon0/uevent

I will break out the double-free change into a separate commit.

[gaul]

If anyone running a newer kernel can provide this information it would help me complete this pull request.

[dgraziotin]

My MacBook is under repair (#staingate). I could try this on my work iMac on Monday. Have Ubuntu 16.04 on a stick, kernel 4.4. Dr. Daniel Graziotin https://ineed.coffee/ Sent from a mobile device. Sorry for brevity / typos.

…

On 30. Sep 2017, at 20:35, Andrew Gaul ***@***.***> wrote: If anyone running a newer kernel can provide this information it would help me complete this pull request. — You are receiving this because you commented. Reply to this email directly, view it on GitHub, or mute the thread.

[dgraziotin]

@gaul here is the output (iMac16,2, 4.8.0-22-generic)

$ find /sys/devices/platform/coretemp.0/hwmon
/sys/devices/platform/coretemp.0/hwmon
/sys/devices/platform/coretemp.0/hwmon/hwmon1
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp5_crit
/sys/devices/platform/coretemp.0/hwmon/hwmon1/device
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp3_crit_alarm
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp4_crit
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp1_max
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp5_max
/sys/devices/platform/coretemp.0/hwmon/hwmon1/subsystem
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp3_label
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp2_input
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp3_crit
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp5_input
/sys/devices/platform/coretemp.0/hwmon/hwmon1/power
/sys/devices/platform/coretemp.0/hwmon/hwmon1/power/runtime_active_kids
/sys/devices/platform/coretemp.0/hwmon/hwmon1/power/runtime_suspended_time
/sys/devices/platform/coretemp.0/hwmon/hwmon1/power/autosuspend_delay_ms
/sys/devices/platform/coretemp.0/hwmon/hwmon1/power/runtime_enabled
/sys/devices/platform/coretemp.0/hwmon/hwmon1/power/runtime_active_time
/sys/devices/platform/coretemp.0/hwmon/hwmon1/power/control
/sys/devices/platform/coretemp.0/hwmon/hwmon1/power/async
/sys/devices/platform/coretemp.0/hwmon/hwmon1/power/runtime_usage
/sys/devices/platform/coretemp.0/hwmon/hwmon1/power/runtime_status
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp4_crit_alarm
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp2_crit
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp2_max
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp5_crit_alarm
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp1_label
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp4_label
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp1_crit
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp3_input
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp3_max
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp1_crit_alarm
/sys/devices/platform/coretemp.0/hwmon/hwmon1/uevent
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp2_label
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp1_input
/sys/devices/platform/coretemp.0/hwmon/hwmon1/name
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp2_crit_alarm
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp5_label
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp4_input
/sys/devices/platform/coretemp.0/hwmon/hwmon1/temp4_max

mbpfan-master works correctly here (and works on my MacBook Pro)

[gaul]

@dgraziotin Would you mind testing the updated patch?

[gaul]

@dgraziotin I hope to upgrade my MacBook soon so I will no longer be able to test this pull request. Could you review it so that I can test any modifications?