Re: [Acpi4asus-user] ASUS M2400N / M3700N / S1300N temperature

[Karol Kozimor <sziwan-DETuoxkZsSqrDJvtcaxF/A@public.gmane.org>]

Thus wrote Mads Paulin:
> I have an ASUS M2400N and is trying to get the fan control right. Karol
> solved the temperature reading problem, but due to wron thermal trip

Did I really? From what you said before, I understood it didn't work.
Let's look at the code:
#v+
    Method (_TMP, 0, NotSerialized)
    {
        Store (0x05, Local1)
        While (Local1)
        {
            Store (RTMP (), Local0)
            If (LGreater (Local0, TCRT))
            {
                Decrement (Local1)
            }
            Else
            {
                Store (0x00, Local1)
            }
        }

        Return (KELV (Local0))
    }
[...]
    Method (RTMP, 0, NotSerialized)
    {
        Store (RBYT (TSAD, 0x01), Local0)
        If (Not (Local0))
        {
            Store (Local0, LTMP)
        }

        Return (LTMP)
    }
#v-

This is the original one. If the (RBYT (TSAD, 0x01) is supposed to return
temperature, it should rather not be zero. Thus, the condition in RTMP
always fails. Since LTMP is initialized with 0x3c (60d), the _TMP method
always returns (KELV (60)), and also, since RTMP is also used internally,
the fan keeps spinning at the same speed. 

Now, if the aforementioned condition is changed to If (Local0), which looks
fairly more reasonable, strange things begin to happen (like those trip
points and temperature being reported as 255 degrees). There is nothing in
the trip point code that would support such a behaviour, unless the change
causes some random memory scribbling, or something similarly nasty.

> By the way, the fan is capable of rotationg at several different rpm's.
> Have any of you gotten this rigt ? On my laptop, the fan starts at max
> rpms and then turns completely off. The only way I can get it to rotate
> at low RPMS is by rebooting the machine when the processor temperature
> is in the interval giving low rpm's. The bios will then put the fan in
> low rpm state.
> 
> This leads me to think, that the fan is not controlled by acpi at all
> (possibly due to a buggy DSDT)

The fan is, and should be, controled by ACPI, though not in the way the
specification says. The \_TZ.TCHG method compares the actual temperature
adn stores appropriate values in the fan registers, causing it to change
its speed. A very similar thermal system exists in ASUS L3800C, where it
works quite well.

Could someone take a look in the M2400N DSDT and suggest a way to debug it?
Best regards,

-- 
Karol 'sziwan' Kozimor
sziwan-DETuoxkZsSqrDJvtcaxF/A@public.gmane.org


-------------------------------------------------------
This SF.Net email sponsored by: Free pre-built ASP.NET sites including
Data Reports, E-commerce, Portals, and Forums are available now.
Download today and enter to win an XBOX or Visual Studio .NET.
http://aspnet.click-url.com/go/psa00100003ave/direct;at.aspnet_072303_01/01