Conveners
Networking Track
- Paolo Abeni (Red Hat)
Description
LPC Networking track is an in-person (and virtual) manifestation of the
netdev mailing list, bringing together developers, users and vendors to
discuss topics related to Linux networking.
Relevant topics span from proposals for kernel changes, through user
space tooling, to presenting interesting use cases, new protocols or
new, interesting problems waiting for a solution.
Brief introduction and general update from netdev maintainers, including recent netdev foundation progresses
The Linux kernel implementation of the Spanning Tree Protocol (STP) has been around since early 2.4 (2000). The IEEE Bridge has evolved and the in kernel version is eleven years behind the current 802.1Q-2014. There are more complete user space implementations but they are not widely used. With the rapid emergence of large language models (LLMs) and AI-assisted development tools, a natural...
The Move-Recursively-Forward algorithm [1] has shown impressive results in speeding up packet classification for firewall rule sets. Its performance gains are generated by reordering firewall rules based on access patterns: frequently matched rules are promoted forward in the list, leveraging locality in network traffic to speed up classification without changing the intended firewall...
Our journey ([1], [2]) to let BPF programs and user-space apps attach rich metadata to packets is far from over. In this talk, we'll share what's been done, what's next, what we've learned, and where are the dragons we've yet to slay.
Part I: Upstream Progress and Roadmap
We'll cover:
- Why we shifted from the old "skb traits" idea [3] to reusing existing
skbmetadata. - How...
For building high-performance datapaths, zero-copy mechanisms are inevitable. Up until today, their usability from network namespaces is either non-existent or very slow. In this talk, we present a solution to natively โleaseโ a physical NIC's hardware queue to a virtual device (such as netkit) in order to enable applications in containers/Pods to fully utilize io_uring or devmem zero-copy for...
Azure VMs have historically run on x86-64 with a 4 KB base page size, but newer ARM64-based SKUs such as ND GB200-v6 introduce configurable base page sizes (4 KB, 16 KB, 64 KB). While this architectural flexibility is great for workloads like model training, it exposed a major inefficiency in the MANA (MICROSOFT AZURE NETWORK ADAPTER) driver's RX path: RX buffers were allocated per page, so a...