Jens Axboe

@osandov that last link has all I know, what's what I wrote when I finally managed to reproduce that issue.

@MCMrARM Unless you can use the id->sqes field to properly set this up, and I don't think you can since it seems dependent on the qid, you'd have to base...

If it truly is some weirdness with sqe+key and specific to qid 1, a better hack to support this might be to just not use qid 1.

I did send Zach a patch, it's also in my io_uring-next repository. BTW, for poll(2) type checking, it's also very possible to do that on the ring fd. That'll work...

@zbjornson any chance of getting this revived?

It's been almost 6 months, what kind of turn-around are you providing for contributors? Not very encouraging.

Let's not make this personal... I maintain tons of open source software, and in my opinion 6 months is an eternity for review. You might as well dump things on...

I have read through it, and the io_uring parts look fine to me. It's possible to remove some boiler plate code by adding a liburing dependency (like the setup code,...

liburing should just be considered a reference implementation. I've written a few things that use the raw interface, so that's quite possible too. Or even a mix - using liburing...

> I need to sign off for the night, but I guess a better solution is for the completion handler to always retry EINVAL results with the threadpool impl. AFAIK...