Comment 32 for bug 102270

We really have three kinds of problems dealing with in this bug.

The original bug was for a MacbookPro3,1 from 2007, which displayed occasional warnings on Feisty, but in general it works. The behavior seems to persist all the way up to Intrepid and the latest patches attached here. It might have become slightly better, at least it is not worse. I would say this problem is really due to the applesmc interface not buffering any sensor data, so it will not change unless a slight rewrite is made in applesmc.

The second problem is for MacbookPro4,1 from 2008, where it seems the number of failures in the log are much more abundant. It might also have gotten worse in Intrepid (a confirmation here would be good). It is my conjecture that the two patches

http://launchpadlibrarian.net/17695749/0001-hwmon-applesmc-Fix-the-wait-status-failed-c-8.patch
http://launchpadlibrarian.net/17697680/0004-hwmon-applesmc-Allow-for-variable-ALV0-and-ALV1-pa.patch

will improve the situation considerably. Anyone with a MBP4,1 who would like to give it a try?

The third problem is for MacbookAir, from 2008, which works very poorly in Hardy: the /sys/devices interface does not work at all. The amount of warnings in Intrepid is very large. For this machine, the two patches make the /sys/devices interface work perfectly. The warning rate running sensors is similar to what is reported for MBP4,1. Under normal operation without sensors, it is completely silent.

at all in Hardy