Re: [PATCH RFC] Avoid memory barrier in read_seqcount() through load acquire

From: Waiman Long
Date: Tue Aug 13 2024 - 15:58:45 EST



On 8/13/24 15:48, Linus Torvalds wrote:
On Tue, 13 Aug 2024 at 12:01, Waiman Long <longman@xxxxxxxxxx> wrote:
Do we need a new ARCH flag?
I'm confused by that question.

That's clearly exactly what that ARCH_HAS_ACQUIRE_RELEASE is.

Obviously all architectures "have" it - in the sense that we always
have access to a "smp_load_acquire()/smp_store_release()".

Sorry for the confusion. What you said above is actually the reason that I ask this question. In the same way, smp_rmb()/wmb() is available for all arches. I am actually asking if it should be a flag that indicates the arch's preference to use acquire/release over rmb/wmb.

Cheers,
Longman


But if the architecture doesn't support it natively, the old rmb/wmb
model may be preferred.

Although maybe we're at the point where we don't even care about that.

Linus