4 extra resolution bits per voltage and temperature channel, but I
chose to use only 2 bits per channel because that's what the ADM1027
has so it made the changes smaller. The EMC6D102 has a 10-bit ADC
anyway, the 2 remaining bits are obtained from averaging so they are
not so important.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4574 7894878c-1315-0410-8ee3-d5d059ff63e0
some of these (in particular SPD and EDID EEPROMS) at addresses where
hardware monitoring chips are known to live, to prevent misdetections, but
we no longer point the user to drivers for these chips.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4531 7894878c-1315-0410-8ee3-d5d059ff63e0
as the BIOS is supposed to set them up properly. This information makes it
easier to find out which temperature channel corresponds to the CPU.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4456 7894878c-1315-0410-8ee3-d5d059ff63e0
* Check for configuration file validity
fancontrol and pwmconfig:
* Support optional min and max PWM values
Documentation updated accordingly.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4438 7894878c-1315-0410-8ee3-d5d059ff63e0
* No longer depend on grep. We already depend on egrep so let's use it
everywhere.
* Don't depend on awk to compute the target PWM value. Bash has an arithmetic
expression evaluation engine which does the job just fine.
* Use read instead of cat to read the sensor values. read is built-in, so
it's cheaper.
* When sysfs is used (2.6 kernel), do not postproces the sensor value reads
with cut, as it is not needed.
* Update the dependency list.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4437 7894878c-1315-0410-8ee3-d5d059ff63e0
* Use let for arithmetic evaluation. Patch taken from the Suse package.
* Kill old commented-out code.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4436 7894878c-1315-0410-8ee3-d5d059ff63e0
be available in Linux kernel 2.6.23. Binary compatibility is guaranteed,
source code compatibility isn't, but the incompatibility will be
spotted quickly as the prototype of the helper function
i2c_smbus_read_i2c_block_data() changed. The only problem would be if
a program is calling i2c_smbus_access() directly. Hopefully this should
be a rare case. The py-smbus binding code is in this case and will be
adjusted soon.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4417 7894878c-1315-0410-8ee3-d5d059ff63e0
back from the address port, so the value 0xff is returned, causing
a false positive with the original test. Testing explicitly for 0x80
instead of only testing that bit 7 is set, works around it.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4404 7894878c-1315-0410-8ee3-d5d059ff63e0
from the openSuse sensors package. This is a good idea because the
PWM/speed relation isn't linear, and it is frequent that the speed
changes are concentrated in the low PWM range.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4380 7894878c-1315-0410-8ee3-d5d059ff63e0
* Delete sample lm_sensors.sysconfig file, it is now generated by
sensors-detect.
* Fix several occurences of the configuration file name.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4378 7894878c-1315-0410-8ee3-d5d059ff63e0
* Update Makefile and README to mention hwmon class for newer
2.6 kernels
* Add missing shell declarations
* Update URI
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4373 7894878c-1315-0410-8ee3-d5d059ff63e0
monitoring devices are better accessed through the hwmon class. Original
patch from Jeff Kosowsky.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4369 7894878c-1315-0410-8ee3-d5d059ff63e0
of in /usr/local/sbin. Many users have been complaining and several
distribtions were (rightly) modifying sensors-detect because of this.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4358 7894878c-1315-0410-8ee3-d5d059ff63e0
sequence to the longer one, and stop as soon as one enter sequence worked.
This avoids false positives.
git-svn-id: http://lm-sensors.org/svn/lm-sensors/trunk@4353 7894878c-1315-0410-8ee3-d5d059ff63e0
Jean Delvare