Re: [PATCH] x86: mark some mpspec inline functions as __init

[Borislav Petkov]

On Thu, Feb 25, 2021 at 01:18:21PM +0100, Arnd Bergmann wrote:
> Either way works correctly, I don't care much, but picked the __init
> annotation as it seemed more intuitive. If the compiler decides to
> make it out-of-line for whatever reason,

Well, frankly, I see no good reason for not inlining a function body
which is a single call. And gcc does it just fine. And previous clangs
did too, so why does clang-13 do it differently?

IOW, could it be that you're fixing something that ain't broke?

> I see no point in telling it otherwise, even though I agree it is a
> bit silly.
>
> Should I send the patch with __always_inline?

I guess.

Although from where I'm standing, it looks like clang-13 needs fixing.
But I surely don't know the whole story and "inline" is not forcing
the inlining so I guess a compiler is free to do what it wants here.
Apparently.

And I guess telling it that those should be always inlined makes it
perfectly clear then.

But WTH do I know...

Thx.

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette