On Tue, Sep 30 2014 at 12:41 -0600, Kevin Hilman wrote:
Nicolas Pitre <nicolas.pitre@xxxxxxxxxx> writes:
On Tue, 30 Sep 2014, Kevin Hilman wrote:
On Tue, Sep 30, 2014 at 10:51 AM, Nicolas Pitre
<nicolas.pitre@xxxxxxxxxx> wrote:
> On Tue, 30 Sep 2014, Kevin Hilman wrote:
>
>> Lorenzo Pieralisi <lorenzo.pieralisi@xxxxxxx> writes:
[...]
>> > This may be misleading. Call it PlatformWFI or something like that, not WFI if
>> > that's not what it is.
>>
>> This gets at a little pet peeve of mine:
>>
>> IMO, naming any state with "WFI" is a bit confusing, because typically
>> *every* idle state is entered by one (or more) CPU executing WFI, no?
>
> Agreed.
>
> The only state called "WFI" should be the one that only executes the WFI
> instruction without any other hardware setup around it.
Well, I would go even further in that none of the states should be
called WFI, because WFI is used to enter all of them.
Fair enough.
So let's fix this by finding a name for that state that consists of only
executing WFI and that every SOC has.
Suggestions?
The DT idle-states binding doc (though seemingly written more with
arm64 and SBSA in mind) uses "standby" for the shallowest idle.
Standby - looks good to me.
--
To unsubscribe from this list: send the line "unsubscribe linux-arm-msm" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
