On Wed, Apr 12, 2017 at 11:49 PM, Mark Brown <broo...@kernel.org> wrote: > On Thu, Apr 13, 2017 at 03:41:03PM +0800, Dong Aisheng wrote: >> On Tue, Apr 11, 2017 at 09:40:03PM +0100, Mark Brown wrote: > >> > Why is this the only anatop regulator which can have this problem and >> > how do we know this is a good value? > >> Anatop regulator has no separate gate bit. >> e.g. >> 00000 Power gated off >> 00001 Target core voltage = 0.725V >> ... >> So it may have no valid default voltage in case it's disabled in >> bootloader. >> e.g. regulator_enable() may not work. > > That doesn't answer my question. What I'm asking is why another anatop > regulator might not end up disabled like this one. >
Well, that's true and i once thought of it. Currently it is probably a quick fix and we did not see any others up till now for all MX6&7 platforms based on NXP internal tree. If we do see it in the future, then probably a better solution is constructing a staticly defined default voltage table in anatop driver and do dynamically check. >> The default voltage 1.100v this patch sets is defined in reference >> manual. > > For the SoC you're currently looking at... might another have a > different value? No, only MX6SX has it currently. Regards Dong Aisheng
