Date: Thu, 24 Apr 2008 01:37:37 +0300 From: Andriy Gapon <avg@icyb.net.ua> To: Nate Lawson <nate@root.org> Cc: freebsd-acpi@freebsd.org Subject: Re: initial value of tz_active Message-ID: <480FBA31.2050202@icyb.net.ua> In-Reply-To: <480F921D.3050505@root.org> References: <480F8E2E.8000502@icyb.net.ua> <480F921D.3050505@root.org>
next in thread | previous in thread | raw e-mail | index | archive | help
on 23/04/2008 22:46 Nate Lawson said the following: > Andriy Gapon wrote: >> It seems that tz_active (in acpi_thermal) is initialized to NONE/-1. >> I don't have a patch, but I have an idea - acpi_thermal should query >> current power state of fans and set tz_active based on that. >> For example, some systems start up with all fans turned on (at full >> speed) for safety reasons, acpi_thermal sees that temperature is low >> enough, it thinks/assumes that no fan is active, so it doesn't change AC >> level, so in reality AC0 is always in effect. > > Good point. I think there should be a "first_time" static that is set > on boot. Once acpi_thermal decides to turn some fans on or off, > first_time is set. Until it has been set, all fan control is actually > performed, regardless of tz_active. After it is set, we can depend on > tz_active as before. > > The reason we can't just read the fan status and initialize tz_active > accordingly is that some ASL doesn't return the real value of the fans. > It just returns a hard-coded value. On many systems, it does return > the real value but if even one system is wrong, we can't trust any of them. > I agree - this is a better approach. On a technical level - maybe use TZ_ACTIVE_UNKNOWN=-2 instead of a new variable. I'll try to work on this somewhere in the middle of May. -- Andriy Gapon
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?480FBA31.2050202>