--- title: A plan for SIMD | daily.dev description: A detailed proposal for a Rust SIMD library called fearless_simd, building on the earlier 'Towards fearless SIMD' blog post. The plan favors code generation... canonical: https://daily.dev/posts/a-plan-for-simd-kemhuuvy0 twitter:card: summary_large_image twitter:site: @dailydotdev og:type: website og:site_name: daily.dev og:title: A plan for SIMD | daily.dev og:description: A detailed proposal for a Rust SIMD library called fearless_simd, building on the earlier 'Towards fearless SIMD' blog post. The plan favors code generation... og:url: https://daily.dev/posts/a-plan-for-simd-kemhuuvy0 og:image: https://api.daily.dev/og/posts/KEmhUUVy0.png og:image:alt: A plan for SIMD og:image:width: 1200 og:image:height: 630 og:locale: en --- > ## Documentation Index > Fetch the complete documentation index at: https://daily.dev/llms.txt > Use this file to discover all available pages before exploring further. # A plan for SIMD **[Linebender](https://daily.dev/sources/linebender)** · 12 min read · 0 upvotes · 0 comments ## Summary A detailed proposal for a Rust SIMD library called fearless_simd, building on the earlier 'Towards fearless SIMD' blog post. The plan favors code generation over declarative macros to reduce compile times while supporting a full cartesian product of SIMD widths (64-512 bits), integer/float types, and f16 where available. It discusses explicit vs variable-width programming, favors 256-bit as a sweet spot for most workloads, addresses AVX-512's double-pumped reality on most chips, compares against pulp and Highway, and covers considerations for WASM SIMD, RVV/SVE scalable vectors, and small embedded SIMD extensions like Helium. The author invites community feedback via Zulip and GitHub before proceeding with implementation for Linebender's Vello renderer. ## Full article daily.dev links to this article rather than hosting it. Read it at the original source: ## Questions this post answers ### What SIMD width should I target for portable Rust SIMD code across x86_64 and ARM? 256 bits is proposed as the sweet spot for Linebender's SIMD library. This matches the natural width of AVX-2, which covers the majority of x86_64 chips, and on Neon it unrolls without register pressure issues since ARM has 32 registers versus AVX-2's 16. Going to 512 bits offers little benefit since most shipping AVX-512 chips are double-pumped, processing 512-bit vectors in two clock cycles of 256 bits each. _Developers weighing SIMD width tradeoffs can follow evolving Rust SIMD design discussions on daily.dev._ ### When is AVX-512 support landing in stable Rust? AVX-512 is expected to land in Rust 1.89 if things go well, according to an update tracked in a rust-lang GitHub pull request. Support for f16 on Neon is also being pursued, though its stabilization timeline is less certain and still being clarified through discussion on the Rust Zulip. _Track Rust stabilization milestones like this on daily.dev before pinning a toolchain version._ ### How does WASM SIMD handle runtime feature detection differently from x86 or ARM? WASM lacks runtime feature detection entirely; instead, feature detection happens during content negotiation to decide which WASM blob to serve, since SIMD capabilities are fixed at compile time. Because of this, a runtime-detection abstraction like a Level enum compiles to zero runtime cost on WASM, even though it's still useful for writing code portable to other targets. _Developers targeting WASM SIMD can weigh these deployment tradeoffs alongside other builds on daily.dev._ ## Similar posts on daily.dev - [SIMD programming in pure Rust](https://daily.dev/posts/simd-programming-in-pure-rust-idnkdlsse) · Hacker News · 2 upvotes · 0 comments --- Tags: [#performance](https://daily.dev/tags/performance), [#rust](https://daily.dev/tags/rust), [#webassembly](https://daily.dev/tags/webassembly) [View this post on daily.dev](https://daily.dev/posts/a-plan-for-simd-kemhuuvy0) ```json {"@context":"https://schema.org","@graph":[{"@type":"Organization","@id":"https://daily.dev/#organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180},"sameAs":["https://twitter.com/dailydotdev","https://github.com/dailydotdev","https://www.linkedin.com/company/daily-dev-ltd"]},{"@type":"WebSite","@id":"https://daily.dev/#website","url":"https://daily.dev","name":"daily.dev","publisher":{"@id":"https://daily.dev/#organization"},"potentialAction":{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https://daily.dev/search?q={search_term_string}"},"query-input":"required name=search_term_string"}}]} {"@context":"https://schema.org","@type":"TechArticle","headline":"A plan for SIMD","url":"https://daily.dev/posts/a-plan-for-simd-kemhuuvy0","mainEntityOfPage":{"@type":"WebPage","@id":"https://daily.dev/posts/a-plan-for-simd-kemhuuvy0"},"datePublished":"2026-08-31T05:54:43.354Z","dateModified":"2026-08-31T05:56:33.982Z","description":"A detailed proposal for a Rust SIMD library called fearless_simd, building on the earlier 'Towards fearless SIMD' blog post. The plan favors code generation...","image":"https://media.daily.dev/image/upload/s--0_ODbtD2--/f_auto/v1722860399/public/Placeholder%2008","thumbnailUrl":"https://media.daily.dev/image/upload/s--0_ODbtD2--/f_auto/v1722860399/public/Placeholder%2008","isAccessibleForFree":true,"articleSection":"Linebender","inLanguage":"en","publisher":{"@type":"Organization","name":"daily.dev","url":"https://daily.dev","logo":{"@type":"ImageObject","url":"https://daily.dev/apple-touch-icon.png","width":180,"height":180}},"author":{"@type":"Organization","name":"Linebender","logo":"https://media.daily.dev/image/upload/logos/placeholder.jpg","url":"https://daily.dev/sources/linebender"},"commentCount":0,"discussionUrl":"https://daily.dev/posts/a-plan-for-simd-kemhuuvy0","interactionStatistic":[{"@type":"InteractionCounter","interactionType":{"@type":"LikeAction"},"userInteractionCount":0},{"@type":"InteractionCounter","interactionType":{"@type":"CommentAction"},"userInteractionCount":0}],"keywords":"performance,rust,webassembly","timeRequired":"PT12M"} {"@context":"https://schema.org","@type":"BreadcrumbList","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https://daily.dev"},{"@type":"ListItem","position":2,"name":"Linebender","item":"https://daily.dev/sources/linebender"},{"@type":"ListItem","position":3,"name":"A plan for SIMD"}]} {"@context":"https://schema.org","@type":"FAQPage","@id":"https://daily.dev/posts/a-plan-for-simd-kemhuuvy0#faq","mainEntity":[{"@type":"Question","name":"What SIMD width should I target for portable Rust SIMD code across x86_64 and ARM?","acceptedAnswer":{"@type":"Answer","text":"256 bits is proposed as the sweet spot for Linebender's SIMD library. This matches the natural width of AVX-2, which covers the majority of x86_64 chips, and on Neon it unrolls without register pressure issues since ARM has 32 registers versus AVX-2's 16. Going to 512 bits offers little benefit since most shipping AVX-512 chips are double-pumped, processing 512-bit vectors in two clock cycles of 256 bits each. Developers weighing SIMD width tradeoffs can follow evolving Rust SIMD design discussions on daily.dev."}},{"@type":"Question","name":"When is AVX-512 support landing in stable Rust?","acceptedAnswer":{"@type":"Answer","text":"AVX-512 is expected to land in Rust 1.89 if things go well, according to an update tracked in a rust-lang GitHub pull request. Support for f16 on Neon is also being pursued, though its stabilization timeline is less certain and still being clarified through discussion on the Rust Zulip. Track Rust stabilization milestones like this on daily.dev before pinning a toolchain version."}},{"@type":"Question","name":"How does WASM SIMD handle runtime feature detection differently from x86 or ARM?","acceptedAnswer":{"@type":"Answer","text":"WASM lacks runtime feature detection entirely; instead, feature detection happens during content negotiation to decide which WASM blob to serve, since SIMD capabilities are fixed at compile time. Because of this, a runtime-detection abstraction like a Level enum compiles to zero runtime cost on WASM, even though it's still useful for writing code portable to other targets. Developers targeting WASM SIMD can weigh these deployment tradeoffs alongside other builds on daily.dev."}}]} ```