On 25/10/18 14:20, Charles Keepax wrote:

On Thu, Oct 25, 2018 at 01:49:05PM +0100, Charles Keepax wrote:

quoted

On Thu, Oct 25, 2018 at 12:42:05PM +0100, Lee Jones wrote:

quoted

On Thu, 25 Oct 2018, Richard Fitzgerald wrote:

quoted

On 25/10/18 09:26, Charles Keepax wrote:

quoted

On Thu, Oct 25, 2018 at 08:44:59AM +0100, Lee Jones wrote:

quoted

On Mon, 08 Oct 2018, Charles Keepax wrote:

quoted

From: Charles Keepax <redacted>

I really feel this isn't the driver you are objecting to as such
but the way regmap operates and also we seem to always have the same
discussions around regmap every time we push a driver.

Absolutely.  I didn't like it before.  I like it even less now.

I guess the question from my side becomes do you want to block
this driver pending on major refactoring to regmap? I will have a
think about what I can do but its going to affect a LOT of drivers.

Actually one more thought, perhaps as a halfway for now i could
look into removing the readables

Be careful with that, there are some addresses that are illegal to
access. What does regmap debugfs do if you don't have a readables
list? Just reading a debugfs shouldn't be able to kill the hardware.
You might need to add a precious list which is more error prone
than listing the valid readables we are using.

  and defaults. We lose some things

like error checking that we are reading real registers but as
this driver doesnt currently do cache syncs we might be able to
get away with this for now.

Unless anyone strenuously objects i will have a look at the
options there. As well as looking at wider refactoring but aiming
further out.

Thanks,
Charles