Re: [RFC 4/4] io_uring: implement futex wait

[Andres Freund]

Hi,

On Thu, Jun 3, 2021, at 14:10, Peter Zijlstra wrote:
> On Thu, Jun 03, 2021 at 12:03:38PM -0700, Andres Freund wrote:
> > On 2021-06-01 23:53:00 +0200, Thomas Gleixner wrote:
> > > You surely made your point that this is well thought out.
> > 
> > Really impressed with your effort to generously interpret the first
> > version of a proof of concept patch that explicitly was aimed at getting
> > feedback on the basic design and the different use cases.
> 
> You *completely* failed to describe any. I'm with tglx; I see no reason
> to even look at the patches. If you don't have a problem, you don't need
> a solution.

Note that this is not my patch, and I hadn't posted anything on the thread when Thomas sent that email? So I'm not sure what you're even referring to here as me having failed to do. If you think the explanation of my use cases I have since sent is insufficient, I'm happy to reply to that/expand on them, but you're going to have to be a bit more specific on why I "failed to describe any".

I just don't see why a simple "what's the use case" wouldn't be a lot more productive than this posturing. Particularly on a thread that is explicitly inviting people from outside the core kernel dev community to chime in.

Andres