On Mon, Oct 23, 2023 at 01:09:11PM +0100, Mark Brown wrote: > On Wed, Oct 04, 2023 at 04:17:17PM +0200, Stephan Gerhold wrote: > > > Instead of -EINVAL we could also use a different return code to indicate > > the initial status is unknown. Or maybe there is some other option that > > would be easier? This is working for me but I'm sending it as RFC to get > > more feedback. :) > > The more normal thing here would be -EBUSY I think - -EINVAL kind of > indicates that the operation will never work while in reality it could > possibly work in future. Though for the RPMH it's not really the case > that it ever supports readback, what it does is have it's own reference > counting in the driver. Rather than doing this we should probably have > logic in the core which sees that the driver has a write operation but > no read operation and implements appropriate behaviour. I like the suggestion to not implement is_enabled, and handle that in the core instead, for all three generations of our rpm-based regulators. Regards, Bjorn
