mastodon.ie

Nils Goroll 🕊️:varnishcache:With the merge of <a href="https://github.com/varnishcache/varnish-cache/pull/4209" rel="nofollow noopener" translate="no" target="_blank">https://github.com/varnishcache/varnish-cache/pull/4209</a> <a href="https://fosstodon.org/tags/VarnishCache" class="mention hashtag" rel="nofollow noopener" target="_blank">#VarnishCache</a> is now prepared for using asynchronous I/O also on the network side.Stay tuned. <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a>

Andrés GómezMy <a href="https://floss.social/@igalia" class="u-url mention" rel="nofollow noopener" target="_blank">@igalia</a> colleague Luís Henriques has just published a very interesting post about the work he has been doing lately in the Linux <a href="https://hostux.social/tags/kernel" class="mention hashtag" rel="nofollow noopener" target="_blank">#kernel</a> about <a href="https://hostux.social/tags/FUSE" class="mention hashtag" rel="nofollow noopener" target="_blank">#FUSE</a> over <a href="https://hostux.social/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a>.<a href="https://luis.camandro.org/2025-06-14-fuse-over-io_uring.html" rel="nofollow noopener" translate="no" target="_blank">https://luis.camandro.org/2025-06-14-fuse-over-io_uring.html</a>

Kunai Project🚀 New Blog Post: Kunai vs io_uring (<a href="https://why.kunai.rocks/blog/kunai-vs-io_uring" rel="nofollow noopener" translate="no" target="_blank">https://why.kunai.rocks/blog/kunai-vs-io_uring</a>) 🚀💡 Ever wondered how io_uring revolutionizes I/O operations in the Linux kernel? Inspired by Armo's blog post (<a href="https://www.armosec.io/blog/io_uring-rootkit-bypasses-linux-security/" rel="nofollow noopener" translate="no" target="_blank">https://www.armosec.io/blog/io_uring-rootkit-bypasses-linux-security/</a>) about a PoC rootkit using io_uring, we explored this feature's security implications and how tools like Kunai can monitor these operations.🔍 Key Takeaways: 🔹 io_uring boosts I/O performance by reducing system call overhead and enabling asynchronous operations 🔹 Security tools struggle to monitor io_uring due to its unique handling of operations 🔹 Kunai now provides visibility into io_uring operations, though blocking malicious activities remains challenging 🔹 Recent kernel versions have introduced auditing and security controls for io_uring, but these are still limited📖 Read more: <a href="https://why.kunai.rocks/blog/kunai-vs-io_uring" rel="nofollow noopener" translate="no" target="_blank">https://why.kunai.rocks/blog/kunai-vs-io_uring</a><a href="https://infosec.exchange/tags/Linux" class="mention hashtag" rel="nofollow noopener" target="_blank">#Linux</a> <a href="https://infosec.exchange/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> <a href="https://infosec.exchange/tags/Security" class="mention hashtag" rel="nofollow noopener" target="_blank">#Security</a> <a href="https://infosec.exchange/tags/OpenSource" class="mention hashtag" rel="nofollow noopener" target="_blank">#OpenSource</a> <a href="https://infosec.exchange/tags/ThreatDetection" class="mention hashtag" rel="nofollow noopener" target="_blank">#ThreatDetection</a> <a href="https://infosec.exchange/tags/SOC" class="mention hashtag" rel="nofollow noopener" target="_blank">#SOC</a> <a href="https://infosec.exchange/tags/DFIR" class="mention hashtag" rel="nofollow noopener" target="_blank">#DFIR</a>

Marcel WaldvogelAll you never wanted to know about creating fast socket I/O on <a href="https://waldvogel.family/tags/Linux" class="mention hashtag" rel="nofollow noopener" target="_blank">#Linux</a>. Unless you are a real geek.A small, ultra-high performance publish-subscribe server doing kind of TCP <a href="https://waldvogel.family/tags/multicast" class="mention hashtag" rel="nofollow noopener" target="_blank">#multicast</a> using <a href="https://waldvogel.family/tags/sendfile" class="mention hashtag" rel="nofollow noopener" target="_blank">#sendfile</a>, <a href="https://waldvogel.family/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a>, <a href="https://waldvogel.family/tags/splice" class="mention hashtag" rel="nofollow noopener" target="_blank">#splice</a>, (mapped) files, and <a href="https://waldvogel.family/tags/fallocate" class="mention hashtag" rel="nofollow noopener" target="_blank">#fallocate</a> for file hole punching.Saturating a 10 Gbps link with useful data with just 8 cores.A beautiful writeup for anyone into fast Linux networking. Not just for <a href="https://waldvogel.family/tags/ATproto" class="mention hashtag" rel="nofollow noopener" target="_blank">#ATproto</a> h/t <a href="https://smnn.ch/@nohillside" class="u-url mention" rel="nofollow noopener" target="_blank">@nohillside</a> <a href="https://waldvogel.family/tags/PubSub" class="mention hashtag" rel="nofollow noopener" target="_blank">#PubSub</a> <a href="https://waldvogel.family/tags/Kernel" class="mention hashtag" rel="nofollow noopener" target="_blank">#Kernel</a> <a href="https://waldvogel.family/tags/Hacking" class="mention hashtag" rel="nofollow noopener" target="_blank">#Hacking</a> <a href="https://www.asayers.com/jetrelay" rel="nofollow noopener" translate="no" target="_blank">https://www.asayers.com/jetrelay</a>

argv minus oneDo any operating systems other than <a href="https://mastodon.sdf.org/tags/Linux" class="mention hashtag" rel="nofollow noopener" target="_blank">#Linux</a> (<a href="https://mastodon.sdf.org/tags/Windows" class="mention hashtag" rel="nofollow noopener" target="_blank">#Windows</a>, <a href="https://mastodon.sdf.org/tags/macOS" class="mention hashtag" rel="nofollow noopener" target="_blank">#macOS</a>, <a href="https://mastodon.sdf.org/tags/FreeBSD" class="mention hashtag" rel="nofollow noopener" target="_blank">#FreeBSD</a>, <a href="https://mastodon.sdf.org/tags/OpenBSD" class="mention hashtag" rel="nofollow noopener" target="_blank">#OpenBSD</a>, etc) have an API for non-blocking file IO?I know Linux has that in <a href="https://mastodon.sdf.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a>, which can do almost any IO operation (even fsync) in the background and tell you when it's done, but is that the only OS with such a feature?<a href="https://mastodon.sdf.org/tags/programming" class="mention hashtag" rel="nofollow noopener" target="_blank">#programming</a> <a href="https://mastodon.sdf.org/tags/async" class="mention hashtag" rel="nofollow noopener" target="_blank">#async</a>

Nils Goroll 🕊️:varnishcache:Does anyone happen to have any information on the relevance of UIO_FASTIOV (defined as 8 on <a href="https://fosstodon.org/tags/Linux" class="mention hashtag" rel="nofollow noopener" target="_blank">#Linux</a>). I understand that submitting vectored io with within this limit can avoid dynamic memory allocation in the kernel, but is this relevant in any way in practice? Is it relevant with <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> ?

Thorsten Leemhuis (acct. 1/4)Support for <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> communication between <a href="https://fosstodon.org/tags/kernel" class="mention hashtag" rel="nofollow noopener" target="_blank">#kernel</a> and [<a href="https://fosstodon.org/tags/fuse" class="mention hashtag" rel="nofollow noopener" target="_blank">#fuse</a>] userspace using IORING_OP_URING_CMD is the highlight from the main fuse (Filesystem in Userspace) updates merged for <a href="https://fosstodon.org/tags/Linux" class="mention hashtag" rel="nofollow noopener" target="_blank">#Linux</a> 6.14:<a href="https://git.kernel.org/torvalds/c/92cc9acff7194b1b9db078901f2a83182bb73202" rel="nofollow noopener" translate="no" target="_blank">https://git.kernel.org/torvalds/c/92cc9acff7194b1b9db078901f2a83182bb73202</a>Miklos furthermore writes:'"Following features enable gains in performance compared to the regular interface:- Allow processing multiple requests with less syscall overhead- Combine commit of old and fetch of new fuse request- CPU/NUMA affinity of queues"'<a href="https://fosstodon.org/tags/LinuxKernel" class="mention hashtag" rel="nofollow noopener" target="_blank">#LinuxKernel</a>

Nils Goroll 🕊️:varnishcache:as i am currently working on making <a href="https://fosstodon.org/tags/varnishcache" class="mention hashtag" rel="nofollow noopener" target="_blank">#varnishcache</a> client side io <a href="https://fosstodon.org/tags/async" class="mention hashtag" rel="nofollow noopener" target="_blank">#async</a>'hronous with <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a>, i must say that <a href="https://lucumr.pocoo.org/2024/11/18/threads-beat-async-await/" rel="nofollow noopener" translate="no" target="_blank">https://lucumr.pocoo.org/2024/11/18/threads-beat-async-await/</a> made me think if we need more facilities for forensic analysis…

Nils Goroll 🕊️:varnishcache:using <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> for storage in <a href="https://fosstodon.org/tags/varnishcache" class="mention hashtag" rel="nofollow noopener" target="_blank">#varnishcache</a> <a href="https://fosstodon.org/tags/allsystemsgo" class="mention hashtag" rel="nofollow noopener" target="_blank">#allsystemsgo</a> <a href="https://fosstodon.org/tags/ASG2024" class="mention hashtag" rel="nofollow noopener" target="_blank">#ASG2024</a> <a href="https://media.ccc.de/v/all-systems-go-2024-305-using-iouring-for-storage" rel="nofollow noopener" translate="no" target="_blank">https://media.ccc.de/v/all-systems-go-2024-305-using-iouring-for-storage</a>From: <a href="https://botsin.space/@mediacccde" class="u-url mention" rel="nofollow noopener" target="_blank">@mediacccde</a> <a href="https://botsin.space/@mediacccde/113203506819669014" rel="nofollow noopener" translate="no" target="_blank">https://botsin.space/@mediacccde/113203506819669014</a>

Luke T. ShumakerAnyone familiar with <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> have any idea what this bit in mini_liburing.h is doing? I think I understand the rest of io_uring_submit(), but I can't fathom what this bit is doing. if (*sq->khead != *sq->ktail) { submitted = *sq->kring_entries; goto submit; }(Also, isn't the IORING_ENTER_GETEVENTS flag at the end no-op, since min_complete=0 ?)

Nils Goroll 🕊️:varnishcache:babysteps, but something is happening <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> <a href="https://fosstodon.org/tags/varnishcache" class="mention hashtag" rel="nofollow noopener" target="_blank">#varnishcache</a>

Nils Goroll 🕊️:varnishcache:the <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> example/proxy.c code really is super helpful. thank you, <a href="https://fosstodon.org/@axboe" class="u-url mention" rel="nofollow noopener" target="_blank">@axboe</a> !

Nils Goroll 🕊️:varnishcache:<a href="https://ruhr.social/@andreclaassen" class="u-url mention" rel="nofollow noopener" target="_blank">@andreclaassen</a> <a href="https://friends.librescrum.org/@derralf" class="u-url mention" rel="nofollow noopener" target="_blank">@derralf</a> in dem bereich, in dem ich zu hause bin, ist es aber definitiv falsch, hier kommen quasi alle software innovationen zuerst als <a href="https://fosstodon.org/tags/foss" class="mention hashtag" rel="nofollow noopener" target="_blank">#foss</a>. einige beispiele: bsd jails als blaupause für solaris zonen und linux container, das dateisystem <a href="https://fosstodon.org/tags/zfs" class="mention hashtag" rel="nofollow noopener" target="_blank">#zfs</a>, <a href="https://fosstodon.org/tags/varnishcache" class="mention hashtag" rel="nofollow noopener" target="_blank">#varnishcache</a> mit vcl als <a href="https://fosstodon.org/tags/edgecomputing" class="mention hashtag" rel="nofollow noopener" target="_blank">#edgecomputing</a> oder aktueller das <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> kernel interface, das nun microsoft für windows kopiert hat. 🧵

Chris Gioran 💔New blog post: A simple, async page cache built on top of io_uring<a href="https://radiki.dev/posts/glowdust-page-cache-1/" rel="nofollow noopener" translate="no" target="_blank">https://radiki.dev/posts/glowdust-page-cache-1/</a><a href="https://fosstodon.org/tags/glowdust" class="mention hashtag" rel="nofollow noopener" target="_blank">#glowdust</a> <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> <a href="https://fosstodon.org/tags/rust" class="mention hashtag" rel="nofollow noopener" target="_blank">#rust</a> <a href="https://fosstodon.org/tags/rustlang" class="mention hashtag" rel="nofollow noopener" target="_blank">#rustlang</a>

Chris Gioran 💔No one: ....Absolutely no one: ....Me: As it turns out, you can use <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> to build a custom async runtime for a page cache.Should you do it? Maybe.Is it fun to build? Yeah, kind of, if you're into state machines and IO.How about I clean it up and push it to the repo, write up a post about it and you tell me what you think?<a href="https://fosstodon.org/tags/rust" class="mention hashtag" rel="nofollow noopener" target="_blank">#rust</a> <a href="https://fosstodon.org/tags/rustlang" class="mention hashtag" rel="nofollow noopener" target="_blank">#rustlang</a>

Chris Gioran 💔Linux 6.11 To Introduce Block Atomic Writes - Including NVMe & SCSI Support<a href="https://www.phoronix.com/news/Linux-6.11-Block-Atomic-Writes" rel="nofollow noopener" translate="no" target="_blank">https://www.phoronix.com/news/Linux-6.11-Block-Atomic-Writes</a>Application blocks will:1. never be torn or fractured when written. 2. For a power fail, all or none of the data to be written.3. A racing atomic write and read will mean that the read sees all the old data or all the new data, but never a mix of old and new. I'm curious to see how I can make it work with my LSM store.<a href="https://fosstodon.org/tags/Linux" class="mention hashtag" rel="nofollow noopener" target="_blank">#Linux</a> <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> <a href="https://fosstodon.org/tags/database" class="mention hashtag" rel="nofollow noopener" target="_blank">#database</a>

Chris Gioran 💔As the sole developer of <a href="https://fosstodon.org/tags/Glowdust" class="mention hashtag" rel="nofollow noopener" target="_blank">#Glowdust</a>, I need to choose if I'll support platforms without <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> support or drop uring support altogether.The small experiment with CSV import showed me that I can't maintain two I/O stacks.Plus, I started Glowdust in part because I wanted to play around with uring.Hmm. Let's be honest - no one but me cares about Glowdust. And I only run <a href="https://fosstodon.org/tags/Linux" class="mention hashtag" rel="nofollow noopener" target="_blank">#Linux</a>.And that's how, in the space of a single toot, I decided to do only io_uring.Thank you for following along.

Lars WirzeniusFun: <a href="https://toot.liw.fi/tags/Rust" class="mention hashtag" rel="nofollow noopener" target="_blank">#Rust</a> bindings to <a href="https://toot.liw.fi/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> using the io-uring crate can do 10 thousand NOP operations per second on my laptop.This isn't useful, but it's the "hello, world" for me for learning about io_uring.

Nils Goroll 🕊️:varnishcache:<a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> question: I am aware that sharing a ring across multiple threads is not the preferred model, but would it be possible to still support it better by passing a mutex to a variant of io_uring_submit_and_wait(), which would get unlocked for the duration of a wait, similar to a cond_wait? <a href="https://fosstodon.org/@axboe" class="u-url mention" rel="nofollow noopener" target="_blank">@axboe</a>

Nils Goroll 🕊️:varnishcache:HAPPY 18TH BIRTHDAY <a href="https://fosstodon.org/tags/VarnishCache" class="mention hashtag" rel="nofollow noopener" target="_blank">#VarnishCache</a> ! To celebrate this memorable occasion, we have just tagged Version 1.0.0-rc1 of <a href="https://gitlab.com/uplex/varnish/slash" rel="nofollow noopener" translate="no" target="_blank">https://gitlab.com/uplex/varnish/slash</a>, which contains fellow, our advanced, <a href="https://fosstodon.org/tags/io_uring" class="mention hashtag" rel="nofollow noopener" target="_blank">#io_uring</a> based, high performance, eventually persistent, always consistent <a href="https://fosstodon.org/tags/opensource" class="mention hashtag" rel="nofollow noopener" target="_blank">#opensource</a> storage engine. Read the full announcement: <a href="https://varnish-cache.org/lists/pipermail/varnish-announce/2024-February/000762.html" rel="nofollow noopener" translate="no" target="_blank">https://varnish-cache.org/lists/pipermail/varnish-announce/2024-February/000762.html</a> And the changelog: <a href="https://gitlab.com/uplex/varnish/slash/-/blob/master/CHANGES.rst?ref_type=heads" rel="nofollow noopener" translate="no" target="_blank">https://gitlab.com/uplex/varnish/slash/-/blob/master/CHANGES.rst?ref_type=heads</a>

Recent searches

Search options

Administered by:

Server stats:

#io_uring