Function get_sqe [src]
Returns a pointer to a vacant SQE, or an error if the submission queue is full.
We follow the implementation (and atomics) of liburing's io_uring_get_sqe() exactly.
However, instead of a null we return an error to force safe handling.
Any situation where the submission queue is full tends more towards a control flow error,
and the null return in liburing is more a C idiom than anything else, for lack of a better
alternative. In Zig, we have first-class error handling... so let's use it.
Matches the implementation of io_uring_get_sqe() in liburing.
Prototype
pub fn get_sqe(self: *IoUring) !*linux.io_uring_sqe Parameters
self: *IoUring Source
pub fn get_sqe(self: *IoUring) !*linux.io_uring_sqe {
const head = @atomicLoad(u32, self.sq.head, .acquire);
// Remember that these head and tail offsets wrap around every four billion operations.
// We must therefore use wrapping addition and subtraction to avoid a runtime crash.
const next = self.sq.sqe_tail +% 1;
if (next -% head > self.sq.sqes.len) return error.SubmissionQueueFull;
const sqe = &self.sq.sqes[self.sq.sqe_tail & self.sq.mask];
self.sq.sqe_tail = next;
return sqe;
}