Hi Mike, On Mon, 4 Apr 2016, Mike Larkin wrote:
> I always start by blowing the dust out of the fan vents. Yes, I know "Windows > works fine", but please start with that. Your machine is sufficiently old > that if you've never done that, it's probably time anyway. Someone helped me in private, and after many tests we came to the conclusion my problem was related to a buggy BIOS, that Windows may have worked around. Upgrading the BIOS solved it but unfortunately triggered another problem (sound related this time), so I was waiting to have that one solved too before giving some feedback on the mailing-list. Testing is very long because it took in the beginning several dozen minutes to trigger the fan problem, and now it takes even longer to trigger the sound problem. And because those times are variable, you're never quite sure the problem is solved, or it just did not happen yet but can happen a few minutes later. I hope I can make my definite feedback within the next days. > Next: > 1. build with ACPI_DEBUG on (acpivar.h) and acpi_debug set to something more > than 20. > 2. Watch for acpitz messages in dmesg as your fan changes speed (or doesn't). > Note that enabling that debug level will lead to a ton of spam in dmesg, you > may instead decide to just set the dnprintfs in acpitz.c to regular printfs > (especially those in acpitz_setfan and acpitz_refresh) in order to avoid > seeing unrelated debug messages from elsewhere in ACPIland. I note this for another time. Thank you. Goodbye, Stéphane.
