(Missing) power states of an Atom N455-based netbook

Vitaly Magerya vmagerya at gmail.com
Sat Jul 16 08:07:42 UTC 2011


Taku YAMAMOTO <taku at tackymt.homeip.net> wrote:
> Maybe we have to tell the BIOS that we are going to utilize _CST via SMI.
> In fact, the major difference between mine and jkim's is that in mine
> I took the following snippet (which we can find in acpi_cpu_startup_cx
> function, living in sys/dev/acpica/acpi_cpu.c) out of #ifdef notyet.
>
> -#ifdef notyet
>      /* Signal platform that we can handle _CST notification. */
>      if (!cpu_cx_generic && cpu_cst_cnt != 0) {
>  	ACPI_LOCK(acpi);
>  	AcpiOsWritePort(cpu_smi_cmd, cpu_cst_cnt, 8);
>  	ACPI_UNLOCK(acpi);
>      }
> -#endif
>
> Vitaly, could you remove the above-mentioned #ifdef-#endif pair (to activate
> the code inbetween) and test jkim's patch again?

It hangs on the same spot.

>> Actually, I abandoned the patches and I am thinking about rewriting it
>> from scratch, e.g., refactoring MI/MD support code
>> (dev/acpica/acpi_cpu.c -> machine/machdep.c & acpica/acpi_machdep.c).
>> Unfortunately, I don't have much time nor hardware to test, so it
>> won't happen any time soon.  Sorry.
>>
>> If anyone wants to pick it up from here, please feel free.
>
> [...]
>
> I will happily test your new patches next time.

I'll provide any testing help needed as well.


More information about the freebsd-acpi mailing list