Did I mention that "mapping the #io_uring ZERO to #NVMe ZERO" is impossible?

Yeah, that's not a thing. io_uring commands don't "map" to hardware, that's entirely _not_ how things work.

It took me some effort and a, frankly disconcerting, number of dump_stack() and pr_alert() to figure this out.

#kernel #linux

I added an opcode to #io_uring in the #linux kernel

Then I exposed it to the userspace uring library from #tokio

And finally I wrote an application with it.

It's easier than it sounds. But I still learned a lot. Here, I wrote it up:

https://radiki.dev/posts/add-zero-opcode-to-io-uring/

If you find this useful let me know, and I'll write more.

#rust #rustlang

Hey, it worked 🎉

I hacked a new opcode in #io_uring and exposed it through #tokio 's io-uring crate.

Wait a sec as i write it up in a blog post. Should be public by tomorrow.

#rust #rustlang #linux #kernel

Let me help your mind wander on this lazy Friday morning with a quick look into how cp performs compared to io_uring.

2 minutes adventure, in and out.

https://radiki.dev/posts/compare-sync-uring/

#uring #io_uring #linux

Love this quote from from the io_uring materials: "Efficiency isn't something that can be an afterthought, it has to be designed in from the start - you can't wring it out of something later once the interface is fixed" #linux #io_uring

Does anyone happen to know if you can make a weird machine out of #Linux's #io_uring system call?

Google Limiting #IO_uring Use Due To Security Vulnerabilities

https://www.phoronix.com/news/Google-Restricting-IO_uring

Original tweet : https://twitter.com/phoronix/status/1669650034344730626

@axboe Are there plans for #io_uring to support "execute always" sqe in link chains? Like with open/read/close or open/write/close kind of workloads to force the close despite the read or write failing? Basically submitting some flag "ignore_cancelled"?

#libuv Adds #IO_uring Support For ~8x Throughput Boost

https://www.phoronix.com/news/libuv-io-uring

Original tweet : https://twitter.com/phoronix/status/1648333901176135682

#io_uring is the best thing since sliced bread!

https://tigerbeetle.com/blog/a-friendly-abstraction-over-iouring-and-kqueue/
#Linux #development #performance

I wrote a bit about #io_uring on #linux

https://developers.redhat.com/articles/2023/04/12/why-you-should-use-iouring-network-io

An interesting article on #io_uring and some of the recent advances and optimisations (in the #Linux kernel).

It notes that it isn’t an introduction and expects users to have some understanding of io_uring already, but even without much idea I think there are some interesting take-aways.

https://github.com/axboe/liburing/wiki/io_uring-and-networking-in-2023

@AndresFreundTec what's the state of #io_uring support in #postgres ? Your last post on Twitter in 2021 mentioned significant improvements.
https://www.postgresql.org/message-id/CAGRY4nwsT2%2BhYgeJ0207pTK4sBzF3dJQ0hAO4c67qfwegj-_HQ%40mail.gmail.com

#postgresql #postgresdb

RT KostasPardalis: even if you are not interested in any of the @\ziglang, #io_uring, Viewstamped Replication, purpose built databases or disk firmware write failures

you should experience @\jorandirkgreef's passion for technology and databases! So much to learn from him.

https://datastackshow.com/podcast/why-accounting-needs-its-own-database-with-joran-greef-of-tiger-beetle/

```ublk-nbd[1] is available now. Basically it is one #nbd client, but totally implemented in userspace […] io handling is based on #liburing[3] […] ublk-nbd supports both tcp and unix socket, and allows to enable #io_uring
send zero copy via command line '--send_zc', see details in README[4]. […] ublk-nbd IOPS is higher by ~40% than […]```

https://lore.kernel.org/all/Y8lSYBU9q5fjs7jS@T590/

[1] https://github.com/ming1/ubdsrv/tree/master/nbd
[3] https://github.com/axboe/liburing
[4] https://github.com/ming1/ubdsrv/blob/master/nbd/README.rst

.@AMDRyzen 7000 Series, #Linux 6.0, #MGLRU, Rust & #IO_uring Made For An Exciting September

https://www.phoronix.com/news/September-2022-Highlights

Original tweet : https://twitter.com/phoronix/status/1576144087614251009

#IO_uring Continues To Prove Very Exciting: Promising #io_uring_spawn Announced

https://www.phoronix.com/news/Linux-LPC2022-io_uring_spawn

Original tweet : https://twitter.com/phoronix/status/1569634030534164482

#InfoSec
> We (#JaydenRivers and #DavidBouman) are disclosing a #bug we found in the #Linux kernel's #io_uring subsystem. We have written a local privilege escalation #PoC that can successfully elevate to system root from an unprivileged process (in a container).
> It is advised to upgrade your #kernel to latest ASAP.
https://www.openwall.com/lists/oss-security/2022/04/22/3

#InfoSec #CVE-2021-41073 (2021)
> Put an #io_uring on it: Exploiting the #Linux Kernel
https://www.graplsecurity.com/post/iou-ring-exploiting-the-linux-kernel

> Once I saw how quickly io_uring was growing, I knew it would be a good place to look. The old adage is true - new code means new bugs.
> When writing an unsafe programming language like C, which is what the Linux kernel is written in, even the best and most experienced developers make mistakes.

RT @postmodern_mod3@twitter.com

When @CrystalLanguage@twitter.com merges/releases the #Linux #io_uring branch, Crystal will absolutely crush I/O benchmarks. https://twitter.com/helloflorin/status/1413236679985479693

🐦🔗: https://twitter.com/postmodern_mod3/status/1413384959671209993