Re: [PATCH v2 2/2] fdt: make fdt handling reusable across arch

["Daniel P. Smith" <dpsmith@xxxxxxxxxxxxxxxxxxxx>]

On 8/3/23 16:37, Luca Fancellu wrote:
> > > Regarding the coding style, I think it’s better to keep the style you’ve found 
> > > in the original file,
> > > and change only some bits when the code is not following it.
> > > I know there is nothing enforcing parameters on the same line of the function 
> > > definition at the
> > > moment, but it is how it’s done from the original file so I would stick with it.
> > > Regarding the u32/u64 types, maybe since you are moving the code it can be the 
> > > occasion to
> > > convert them, but check with the maintainer before.
> > I can leave the main code as is, but I do think header decl's should be styled 
> > correctly as there is no need to have them churn in the future over purely 
> > style changes.
> Uhm, when you say “styled correctly” do you mean as below?
I am retracting that after going back to review my notes, see below.

> > > > +bool __init device_tree_node_matches(
> > > > +    const void *fdt, int node, const char *match)
> > > > +{
>
> If that’s the case, it seems to me that there is nothing like that in the 
> codebase,
> in my work with clang-format I’ve configured it to match as much as I can the
> Xen style and this function would be formatted as the old style that it had.
>
> Can I ask you where did you find instruction to style in that way?
I went back to my old notes on styling to make sure I was correct. Turns 
out I was incorrect and do apologies. There are two accepted styles for 
function declarations[1] and the one I used in this patch was the one 
that I got a recommendation to use. I have gotten so use to that style, 
that I must have lost track the other was valid. As was pointed out 
elsewhere, I should use the form that the maintainer desires. As XSM 
maintainer, I use the one I submitted here and would ask for a style 
correct if someone submitted a patch using the form that is desired here.
I will roll back the declaration styling for now and per the prior 
guidance[1], I will flip the linux-compat int over C spec fixed width 
int notation, unless there is objection to it happening during the move.
[1] 
https://lists.xenproject.org/archives/html/xen-devel/2021-07/msg01133.html
v/r
dps