Emperor Penguin rates Memory Folios tech \u2013 source of that performance bump \u2013 as most important new feature in ‘not huge’ release<\/b>
\nLinus Torvalds has loosed the first release candidate for version 5.16 of the Linux kernel. In his Sunday afternoon State of the Kernel announcement, Torvalds indicated the merge window for the new release did not include any \u00ab\u00a0uhhuh, things aren’t working and now I need to bisect where they broke\u00a0\u00bb moments. Work therefore proceeded swiftly, so much so that not even travelling for a few days and using a laptop (which Torvalds wrote is \u00ab\u00a0usually fairly painful\u00a0\u00bb) delayed progress. Torvalds wrote that Linux 5.16 will not be a \u00ab\u00a0huge release\u00a0\u00bb but does include a significant new feature. \u00ab\u00a0Memory Folios\u00a0\u00bb is a memory management system that offers \u00ab\u00a0a more efficient and type-safe way to specify ‘head of a group of pages’, rather than the page pointers and ‘compound_head()’ and friends\u00a0\u00bb. As explained by Matthew Wilcox, a long-time contributor to the Linux kernel (and Oracle employee) who focuses on memory, Memory Folios aims \u00ab\u00a0to allow filesystems and the page cache to manage memory in larger chunks than PAGE_SIZE\u00a0\u00bb. Doing so has produced some nice numbers. \u00ab\u00a0The multi-page folios offer some improvement to some workloads,\u00a0\u00bb Wilcox wrote. \u00ab\u00a0Real workloads (eg building the kernel, running postgres in a steady state, etc.) seem to benefit between 0\u201310%.\u00a0\u00bb A single addition to the Linux kernel that could improve performance by ten per cent? Where do we sign? Torvalds rated Memory Folios as \u00ab\u00a0unusually core, but they certainly aren’t the bulk of the changes\u00a0\u00bb. Those come from the usual suspects \u2013 what Torvalds describes as \u00ab\u00a0drivers (gpu, networking, sound and staging stand out, but it’s all over) and architecture code\u00a0\u00bb. \u00ab\u00a0Hardware support is the bulk of the code, it gets the bulk of the changes. But we obviously have all the normal other updates, with filesystem, networking, and core kernel code. With documentation and tooling support filling the gaps.\u00a0\u00bb \u00ab\u00a0Anyway, the merge window may have gone about as smoothly as I could hope for,\u00a0\u00bb Torvalds wrote, before adding \u00ab\u00a0but let’s get the whole stabilization phase started with some serious testing, shall we?\u00a0\u00bb You heard the man. \u00ae<\/p>\n