Re: [Acpi] Re: ACPI "hlt" mode and SMP systems?

Re: [Acpi] Re: ACPI "hlt" mode and SMP systems?

[Roeland Th. Jansen]

On Fri, Nov 09, 2001 at 10:53:16PM +0000, Petr Vandrovec wrote:
> Since when? Answer:
> 
> $ host acpi.info
> acpi.info A 192.168.0.1
> acpi.info A 208.196.23.64
> 
> means that in 50% of cases I receive my local webpage, and in other
> 50% I get real info... Do I miss something?

the fact that there has bene made some config error, does not mean that
the site doesn't exist. 

-- 
Grobbebol's Home                        |  Don't give in to spammers.   -o)
http://www.xs4all.nl/~bengel            | Use your real e-mail address   /\
Linux 2.4.14 (noapic) SMP 466MHz/768 MB |        on Usenet.             _\_v  

Re: [Acpi] Re: ACPI "hlt" mode and SMP systems?

[Petr Vandrovec]

On  9 Nov 01 at 21:40, Roeland Th. Jansen wrote:
> On Thu, Nov 08, 2001 at 02:34:34PM +0000, Pavel Machek wrote:
> > > +ACPI 2.0 Specification; esp. Ch 1 - 3.
> > > +<http://acpi.info>
> >     ~~~~~~~~~~~~~~~~
> >       This url does not seem too real.
> 
> it in fact _is_ a working URL..

Since when? Answer:

$ host acpi.info
acpi.info A 192.168.0.1
acpi.info A 208.196.23.64

means that in 50% of cases I receive my local webpage, and in other
50% I get real info... Do I miss something?
                                                Petr Vandrovec
                                                vandrove@vc.cvut.cz
                                                

Re: [Acpi] Re: ACPI "hlt" mode and SMP systems?

[Riley Williams]

Hi Pavel.

>> +<http://acpi.info>
>    ~~~~~~~~~~~~~~~~

> This url does not seem too real.

I've just checked, and it's a valid one. The new .info domains came into
action 1 Nov 2001 if I remember right, and that one's exactly what it
claims to be...

Best wishes from Riley.

Re: [Acpi] Re: ACPI "hlt" mode and SMP systems?

[Roeland Th. Jansen]

On Thu, Nov 08, 2001 at 02:34:34PM +0000, Pavel Machek wrote:
> > +ACPI 2.0 Specification; esp. Ch 1 - 3.
> > +<http://acpi.info>
>     ~~~~~~~~~~~~~~~~
> 		This url does not seem too real.

it in fact _is_ a working URL..

-- 
Grobbebol's Home                        |  Don't give in to spammers.   -o)
http://www.xs4all.nl/~bengel            | Use your real e-mail address   /\
Linux 2.4.14 (noapic) SMP 466MHz/768 MB |        on Usenet.             _\_v  

Re: [Acpi] Re: ACPI "hlt" mode and SMP systems?

[Pavel Machek]

Hi!

I have few comments...

> +System sleep states are defined as follows:
> +
> +S0 - On
> +   - System is running.
> +
> +S1 - "Power On Suspend"
> +   - Processor power, and hence execution context, is preserved.
> +   - Devices may have been put into a low power state.
> +
> +S2 - "Pseudo-Suspend To Ram"
> +   - Processor power is removed.
> +   - Devices may be placed into low power state.
> +   - Memory is placed into self-refresh and retains context.
> +   - Execution starts again from processor's reset vector
> +   - Cache and MTRR configuration is lost during this state; the
> +     firmware is responsible for restoring it to some known state.
> +
> +S3 - "Suspend to Ram"
> +   - Processor power is removed.
> +   - Devices are placed into a low power state.
> +   - Power may be removed from all system buses.
> +   - Memory context is preserved by placing memory in Self-Refresh
> +   - Execution begins from processor's reset vector
> +   - Cache and MTRR configuration is lost during this state; the
> +     firmware is responsible for restoring it to some known state.

If S2 and S3 are same, you should use exactly same text. It seems to say
exactly same thing in slightly different words.


> +S4 - "Suspend to Disk", aka "Hibernate"
> +   - All power may be removed from the system.
> +   - All device context may be lost.
> +   - Context is usually written to persistant storage (e.g. disk).
> +   - Execution begins at processor's reset vector; the firmware
> +     will usually load the OS boot loader.
> +
> +S5 - Soft Off
> +   - Technically not a sleep state
> +   - No context is retained
> +   - OS is not responsible for saving context
> +   - OS must completely reinitialise itself on 'resume' (power-on)
> +   - Wake events may still trigger the system to resume from sleep.

...

> +G0 -
> +     The system is in the S0 state.
> +G1 -
> +     The system is in one of the S1 - S4 sleep states.
> +G2 -
> +     The system is in the S4 sleep state.
                             ~~
				 this should be S5, no?

> +6. References
> +~~~~~~~~~~~~~
> +
> +ACPI 2.0 Specification; esp. Ch 1 - 3.
> +<http://acpi.info>
    ~~~~~~~~~~~~~~~~
		This url does not seem too real.

								Pavel
-- 
Philips Velo 1: 1"x4"x8", 300gram, 60, 12MB, 40bogomips, linux, mutt,
details at http://atrey.karlin.mff.cuni.cz/~pavel/velo/index.html.