Re: RFC: representing sdio devices oob interrupt, clks, etc. in device tree
From: Ulf Hansson <hidden>
Date: 2014-06-09 14:07:54
Also in:
linux-arm-kernel, linux-mmc
On 4 June 2014 17:55, Mark Brown [off-list ref] wrote:
On Tue, Jun 03, 2014 at 12:57:52PM +0200, Ulf Hansson wrote:quoted
On 28 May 2014 13:03, Mark Brown [off-list ref] wrote:quoted
quoted
No, runtime PM isn't really fine grained - I'm talking about things like starting and stopping individual resources or configuring them.quoted
Are you saying that you have several levels of quiescent mode of your external chip?Partly, but also there can be active modes that don't use all functionality and hence can leave some of the inputs powered down (for example if a reference clock is used for some feature that isn't used all the time then that reference clock may be able to be powered down when the function is not in use). The device isn't suspended but it can do without some of the resources it has.quoted
I am a bit puzzled of how such library should look like, but I suppose we need to be able to go to different power states through it. Typically the library may be invoked both from the mmc core to do power up/off and from the sdio func driver to handle power save.quoted
Does it make sense to you as well?Yes, certainly for the power up/down and runtime PM type operations.
For your information; I have started implementing a library to handle power sequences. I intend to have this separated from mmc, to let other subsystems re-use the same code. It will not require a compatible string, but instead use something like 'power-method = blah' property. I intend to post this as an RFC in couple of days. Kind regards Uffe