You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
This may also be related to Tracking: multishot opcode support #104, because these opcodes "might result in two completion queue entries", which may also require changes to the semantics of the current OpCode and related stuffs.
There is also zero copy rx but its support appears to be very limited.
But eventually you'll find that such a safe API is not that convenient as the unsafe one. If you use SendZc in C, you can carefully deal with part of the buffer that is not consumed by the kernel. However, you cannot do so in safe Rust.
Activity
[-]feature request: io-uring zerocopy send support[/-][+]feature: io-uring zerocopy send support[/+]Berrysoft commentedon Jan 5, 2026
I think that it meets the same problem as #104 . It's easier to design the high-level APIs, but hard to write the driver.
For
send_zc, I think it's necessary to wait for the second cqe before return.George-Miao commentedon Jan 5, 2026
We need to change how the driver work. It need to be able to take n CQE per SQE instead of only 1. This will also solve multishot.
AsakuraMizu commentedon Jan 5, 2026
Well, not actually; I'm thinking of making something that returns a
BufResult<usize, impl Future<Output = T>>🤣Berrysoft commentedon Jan 5, 2026
I have a similar idea.
But eventually you'll find that such a safe API is not that convenient as the unsafe one. If you use
SendZcin C, you can carefully deal with part of the buffer that is not consumed by the kernel. However, you cannot do so in safe Rust.