Hi, On Monday, 25 June 2007 21:28, Pavel Machek wrote: > Hi! > > > >>> Right now the states we have are On, Standby, and Suspend, and the CPU > > >>> runs only in the On state. But on some platforms there could be > > >>> multiple states in which the CPU is able to run, albeit with degraded > > >>> performance. > > >> I wouldn't call those system sleep states. For example, ACPI defines system > > >> sleep states as the states in which no instructions are executed by any CPUs > > >> and I think that's reasonable. > > > > > > Well, in some cases, we have 200MHz CPU running at 30kHz. ...that's so > > > slow that it is pretty similar to ACPI sleep state. > > > > > Pavel, > > > > let's do not mix the things. Rafael gave exact meaning of a sleep state -- > > "no instructions are executed by any CPUs", that, I assume, can be > > interpreted > > as "processor does not do any effective job". It does not mean if any clock > > frequency supplied or not. In some cases the frequency supplied to speed up > > the wakeup process, but, the processor can't operate on this frequency, > > so this is a sleep state. But, if the processor can do some job, > > even if a frequency very very slow, then this is an active mode. > > Well, I'm trying to say that cpu (samsung arm in openmoko?) may be > able to execute instructions at very very low speed, but that the > speed is so low, that we should treat it as a sleep state. > > I mean -- you can't use cpufreq to automatically chose between 32kHz > and 200MHz. > > ...but I guess it is all not too important... Well, as I've just written to Dave, I think that we can define a 'system sleep state' as a state that requires the main code path in kernel/power/main.c (starting from enter_state()) to be executed in order for the system to enter it. Arguably, in such a state no useful work will be done by the system. The other states can be regarded as 'running' or 'On' states. Greetings, Rafael -- "Premature optimization is the root of all evil." - Donald Knuth - To unsubscribe from this list: send the line "unsubscribe linux-acpi" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
