Open aidosww opened 1 month ago
Please include some actual details on what you are doing, what devices are involved, what kernel, etc. As it stands, this question is rather useless.
I am using kernel 5.10, the default io_uring configuration. When I judge that io_uring queue is full, I submit read file tasks with io_uring_submit, and after running for a period of time, the above error is reported.
Please try a more recent and supported kernel, the ancient ones don't have proper -EAGAIN handling on the storage side (which I'm guessing is what you are doing, you didn't really answer that part). Anything 6.x should be fine.
In general, 5.10 and similar are just too old to be supported.
Please try a more recent and supported kernel, the ancient ones don't have proper -EAGAIN handling on the storage side (which I'm guessing is what you are doing, you didn't really answer that part). Anything 6.x should be fine.
In general, 5.10 and similar are just too old to be supported.
Thank you very much! In fact, we can only use the 5.10 kernel in our Development Environment. I am indeed developing a storage I/O interface based on liburing and have encountered many problems. 1、Can I solve the above problem by repeatedly calling the io_uring_submit interface? 2、What specific problem caused this error?
Please include some actual details on what you are doing, what devices are involved, what kernel, etc. As it stands, this question is rather useless.