liburing
liburing copied to clipboard
This adds the first of the new man 7 pages suggested in #1241 It contains a high lvl overview of what needs to be done on the kernel side for...
`wait_timeout(nr)` is generally a good way to reduce wakeups from kernel, while CQEs of send/write can bring unnecessary "noise", especially from plenty of zero-copy. In essence, it is difficult to...
I've been trying to troubleshoot, or rather understand, an issue specific to disk I/O (writes), where writes are just significantly slower when scheduled via IORING_OP_WRITEV SQE ops, compared to straight...
I've been measuring performance of my Uring across the application and found out unexpected results: 1) io_uring affinity set to 4 cores. application only uses uring for writes. the load...
My view is 64bit user_data is not enough. In order to use send_zc more flexibly, a circle linked list is introduced to maintain buffers. Then each send cqe must contain...
Looking at the code in liburing.h, it looks like the fields in the struct are never used. Instead `io_uring_recvmsg_payload` requires the `msghdr` passed to the setup of `revcmsg_multishot` to get...
I'm writing a timeout read method using io_uring_prep_read + io_uring_prep_link_timeout, and the method is now functional. However, there's an issue: The read CQE->res sometimes returns -ECANCELED due to the timeout,...
hello expert, this is related with https://github.com/axboe/liburing/issues/1415, after a lot of parameter tuning. I can reach 1300MB/s now. still have gap with fio. but I found one interesting thing is...
Are there any plans for adding them? Currently sendfile can be simulated as [described here](https://tinselcity.github.io/Sendfile-W-Iouring/) but it has some overheads (pipe creation and 2 splice calls specifically).
Per the [io_uring_enter](https://man7.org/linux/man-pages/man2/io_uring_enter.2.html) man page: > EINTR The operation was interrupted by a delivery of a signal before it could complete; see signal(7). Can happen while waiting for events with...