Re: [PATCH v3 3/3] arm64, compiler-context-analysis: Permit alias analysis through __READ_ONCE() with CONFIG_LTO=y

[Boqun Feng <boqun@kernel.org>]

On Tue, Feb 17, 2026 at 08:32:32AM -0800, Linus Torvalds wrote:
> On Tue, 17 Feb 2026 at 08:23, Linus Torvalds
> <torvalds@linux-foundation.org> wrote:
> >
> > But last time I looked at it - which was admittedly a few years ago -
> > the compilers we supported didn't actually do anything reasonable here
> > (ie the built-in atomics were fundamentally worse than the ones we do
> > by hand, and even basic things like __atomic_load_n() weren't
> > actually; better than just using 'volatile'.
> >
> > Maybe that has changed. We've upgraded minimum compilers since.
> 
> I  just checked a few test-cases, and I don't think anything has changed.
> 
> All the trivial things where __atomic_load_n(__ATOMIC_RELAXED) *could*
> do better than just our old code using "cast pointer to volatile"
> resulted in no better code generation.
> 

By any chance, could you share the test cases? It'll be really helpful
to represent what semantics we really want to compiler/language folks.

[Cc Alice and Gary]

Regards,
Boqun

> So we're sticking to that "cast to volatile" not because it's great,
> but because it continues to be reliable and no worse than the fancy
> modern versions.
> 
> But at least it's explicit in the code, not hidden away in code that
> *looks* like it just dereferences another field in a structure.
> 
>                Linus