Re: [GIT PULL] locking changes for v6.4

[Linus Torvalds]

On Thu, May 4, 2023 at 10:30 AM Andrzej Hajda <andrzej.hajda@xxxxxxxxx> wrote:
>
> Later I have also showed/suggested Rust's approach [3][4]:
> - 'replace' for __xchg,
> - 'take' for fetch_and_zero

I don't think either of those are acceptable, and both are to me
*much* worse than the existing "fetch_and_zero()" that i915 uses.

That name just makes sense and explains what it does. There is no ambiguity.

I will not be taking a pull that makes a good name worse.

Naming matters, and "use a generic function" is not an excuse for
having bad names.

              Linus