Re: [PATCH] rust: introduce sfile macro for easier code tracing

[Greg KH <gregkh@linuxfoundation.org>]

On Wed, Jun 04, 2025 at 05:16:54PM +0200, Benno Lossin wrote:
> On Wed Jun 4, 2025 at 12:28 PM CEST, Miguel Ojeda wrote:
> > On Wed, Jun 4, 2025 at 1:29 AM Timur Tabi <ttabi@nvidia.com> wrote:
> >>
> >> I didn't want to change the existing behavior of any other macro or function.  This is my first
> >> significant rust-for-linux contribution, and I didn't want to be presumptuous.
> >
> > No worries -- things can be changed, and in particular the behavior of
> > `dbg!` wouldn't be too bad, in the sense that calls are not meant to
> > be committed anyway. A CI checking for the message from the sample
> > could break I guess, but that is not the end of the world.
> >
> >> I could add a sdbg!() if you think it would be useful.
> >
> > Yeah, that was what I was thinking, though it is sadly one more letter.
> >
> > We could also shorten a bit more by not printing the column like
> > `dbg!` does, which I suspect many may not care about.
> >
> > I wonder if we could make the `dbg!` the one with the longer name
> > instead, i.e. to keep the functionality in case someone really needs
> > it. Sadly, that means making it not match the stdlib name, but it may
> > be worth it if we expect everyone to end up typing `sdbg!` all the
> > time, and only very rarely `dbg!`.
> 
> Why not control the amount of path segments with a kconfig? That way we
> can have both worlds.

Ugh, no, please no.