Re: [PATCH v4 1/3] i2c: mux: Add i2c-arbitrator-cros-ec 'mux' driver

[Wolfram Sang <wsa@the-dreams.de>]

Hi,

> Very interesting discussion, especially the argument that "we already shipped"
> would not be a convincing argument.
> 
> I had senior kernel maintainers tell me and the company I work for that we should
> submit _all_ our platform specific kernel code and drivers for inclusion into
> the upstream kernel.

Yes, great. Really!

> This exchange suggests that "it is a shipping product" does not count for such
> submissions, and that "Benefit for the kernel" would be the deciding factor
> instead. Which of course is a very vague statement - if code supporting
> Chromebookis is of no benefit for the kernel, support for my company's products
> for sure is much less so.

First, let me state that I did not intend to say that the arbitrator
muxer here has NO benefit for the kernel. I was aware there might be
arguments for that and I wanted some more discussion to make that
clearer to me. Simon's mail was very helpful in that regard and I will
comment on that somewhen later.

Still, I do have problems with "we already shipped it". If the driver is
bad or the underlying concept is fragile, I want the freedom to reject a
patch, product shipped or not. I will be supportive to find a proper
solution, promised. If all fails, there is still staging/ or the
possibility of out-of-tree patches.

> Which kind of leaves me in a bind. On one side I promote that we should submit
> all our kernel code, on the other side there is a very compelling case to be
> made that it won't be accepted anyway. That doesn't make my life easier -

Concentrate on argumenting why the driver is useful and it will be fine.
"we already shipped this" feels a bit like blackmailing to me. And since
most drivers do have well thought reasons for their existance, I'd
primarily like to hear about those.

> essentially since it supports those who say that we should not submit anything
> in the first place. And believe me, there are many of those. 
> 
> Just to give some examples:
> - I2C multiplexers. We have a bunch of those. Looking at this exchange,
>   it doesn't look good to get that code included.

Multiplexers should be easy going, it is the arbitration which is discussed here.
I am open to consider some generic arbitration schemes. What I am reluctant to
do is to allow every board to have its own arbitration scheme. This
would be board support hosted in the I2C directory. Meh.

> It would be nice have to get some well defined guidelines for "acceptable"
> contributions. "Benefit for the kernel" sure isn't one.

I don't think it is possible to write down concrete guidelines for this.
"According to rule 4a) of the guidelines you have to accept my patch"?
That would be a mess, I think.

Regards,

   Wolfram