-
A colleague brought this up -- XDP may provide better performance than TAP.
I looked around and found this https://events19.linuxfoundation.cn/wp-content/uploads/2017/11/Accelerating-VM-Networking…
-
Running a dev-6.0 kernel on an ast2500 platform I'm working on a port to I'm sometimes hitting a panic on busses with an IPMB device on them:
```
[27559.144928] 8buf` being NULL.
I've seen it h…
-
Hello! This could be a dumb question but I wanted to know if this performance difference is expected.
My program is written in Zig (the stdlib io_uring library is based on liburing), so the issue …
-
# Motivation
It is common to have event waiting logics bounded with timeouts. But currently, all NGO code will wait forever if no interesting events arrive. And since there are so many system cal…
-
Are there any plans to have LXD support qcow2 images as a storage backend?
I am not much of a docker fan but when trying it a few times I noticed that it uses qcow2 images for the backend. The storag…
-
## Bug report
### Describe the bug
It's been like this for like 2 years now ish, but used to be fine, for sure. I would bisect releases but 2-3 releases back stop compiling due to fmtlib errors.…
-
These functions are synchronous and seem incompatible with co_await.
not epoll;
-
It would be great if work were started for Java fibers (Project Loom).
There's https://github.com/petrbouda/jlhttp-server-on-fibers/ which seems like a benchmarkable comparison for a minimal HTTP s…
-
I already finish to send nvme command with io_uring_passthrough path. But I only can check cqe->res is 0 or not to confirm the command is executed correctly or not. I print all big_cqe data and they a…
-
![write_waterfall_mmap](https://user-images.githubusercontent.com/4198311/106277551-4a776880-6274-11eb-9f2c-3981685f537f.png)
![write_waterfall_uring](https://user-images.githubusercontent.com/419831…
skyzh updated
3 weeks ago