Re: [GIT PULL] pipe: nonblocking rw for io_uring

[Peter Zijlstra]

On Tue, Apr 25, 2023 at 10:20:31AM -0700, Linus Torvalds wrote:

> But the more I look at it, the more I'm convinced that our pattern of
> 
>         old = READ_ONCE(rq->fence.error);
>         do {
>                 if (fatal_error(old))
>                         return false;
>         } while (!try_cmpxchg(&rq->fence.error, &old, error));
> 
> (to pick one random user) is simply horribly wrong.

The last time this came up I shared your view however Mark argued for
the READ_ONCE() thusly:

  https://lore.kernel.org/all/Y71LoCIl+IFdy9D8@FVFF77S0Q05N/T/#u