FUSE performance improvements with eBPF [pdf]

Fuchsia has an interesting take on filesystems [1]. One can write it completely in the user-space, avoiding expensive kernel&lt;--&gt;user-space switching. Additional benefit of storage sand-boxing comes for free, as each app can implement it&#x27;s own fs, with the rest of the system unawares of it&#x27;s existence.<p>I wish such a fully-user-space option existed for Linux. This work is philosophically in the opposite direction, moving more functionality into kernel space for perf benefits.<p>[1]: <a href="https:&#x2F;&#x2F;fuchsia.dev&#x2F;fuchsia-src&#x2F;the-book&#x2F;filesystems.md" rel="nofollow">https:&#x2F;&#x2F;fuchsia.dev&#x2F;fuchsia-src&#x2F;the-book&#x2F;filesystems.md</a>

I saw the authors&#x27; talk at Linux Conf last year. It seems like an awesome improvement but I&#x27;m actually far more interested in the &quot;future work&quot; which can be done. Namely, this system could be expanded to data &quot;caching&quot; whereby the kernel could route read(2) and write (2) to an underlying &quot;struct file&quot; in-kernel. This would allow for effectively zero-overhead FUSE-based overlay filesystems (which would be super useful for container runtimes -- especially once OCIv2 is usable).

Here&#x27;s a slide deck from the same authors: <a href="https:&#x2F;&#x2F;events.linuxfoundation.org&#x2F;wp-content&#x2F;uploads&#x2F;2017&#x2F;11&#x2F;When-eBPF-Meets-FUSE-Improving-Performance-of-User-File-Systems-Ashish-Bijlani-Georgia-Tech.pdf" rel="nofollow">https:&#x2F;&#x2F;events.linuxfoundation.org&#x2F;wp-content&#x2F;uploads&#x2F;2017&#x2F;1...</a><p>tl;dr: perform metadata caching in the kernel using eBPF, avoiding context switches for common operations like listdir() and getattr(), and reduce FUSE overhead from ~18% to ~6%.

FUSE is particularly useful for writing virtual file systems. Unlike traditional file systems that essentially work with data on mass storage, virtual filesystems don&#x27;t actually store data themselves. They act as a view or translation of an existing file system or storage device.<p>In principle, any resource available to a FUSE implementation can be exported as a file system.