Almog Khaikin
Almog Khaikin
I had a look at the code here and the same bug is present here as in liburing/the kernel. The code here is almost the same as in liburing, the...
> Is there a fix for tokio-rs/io-uring that you think presents itself, or are you saying the API, as it stands now, should come with an unsafe method with a...
The load in `sq_need_wakeup` is relaxed because we are not acquiring anything and the kernel's setting of the `NEED_WAKEUP` flag is not done using a release operation. The `SeqCst` fence...