On Fri, 10 Sep 2010 13:00:17 +0100 Mark Brown broonie@opensource.wolfsonmicro.com wrote:
On Fri, Sep 10, 2010 at 02:23:32PM +0300, Jarkko Nikula wrote:
- regulator_bulk_disable(ARRAY_SIZE(aic3x->supplies), aic3x->supplies);
- if (aic3x->power)
regulator_bulk_disable(ARRAY_SIZE(aic3x->supplies),aic3x->supplies);This looks suspicious - doesn't it mean that the enable/disables won't be balanced any more? I'd expect to either see the disable being uncondtional or the power flag being updated. I could be missing something, though.
I agree.
This is actually due the current implementation where the regulators are enabled in aic3x_i2c_probe and disabled in aic3x_probe. So this disable is needed only if the driver is removed without aic3x_probe being called.
I'll cook up one patch more that moves regulator setup to aic3x_probe as most of the codec drivers are doing. (side issue: it's hard to probe a chip if voltages are missing but doing regulator setup in i2c probe is not any better either).