Guys: On Sat, Jan 21, 2012 at 2:24 PM, Mark Brown <broonie@xxxxxxxxxxxxxxxxxxxxxxxxxxx> wrote: > Enable and disable the clocks to the SPI controller using runtime PM. This > serves the dual purpose of reducing power consumption a little and letting > the core know when the device is idle. What about using autosuspend instead? If a client is doing a lot of closely-spaced SPI transactions on a relatively flat device tree, might the resulting runtime suspend/resume overhead between each transaction become noticeable? I'm doing runtime PM on several drivers myself, and that question keeps coming up a lot. Especially for devices that are more complicated to suspend/resume than this one. But even without the device-specific overhead, I'm wondering how structure-walking that goes on behind the pm_runtime() calls might tip the balance of PM gains the opposite way from what we are seeking. Not a critique of your patch series per se, but you probably have some thoughts on all of this too--- and I would love to hear them while they are still fresh. :) b.g. -- Bill Gatliff bgat@xxxxxxxxxxxxxxx -- To unsubscribe from this list: send the line "unsubscribe linux-samsung-soc" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
 |