proomt

Search

Search posts, papers, and topics

All posts

Hacker News front page4 min readintermediate

Memory-Safe WebP Decoding

Summary

wpd is a Rust‑based WebP decoder that replaces most unsafe C code with safe Rust (only zerocopy unsafe) and optional hand‑written SIMD, delivering faster decoding than libwebp. Benchmarks show up to 3.2× speedup in multi‑threaded scenarios while matching libwebp’s feature set and adding extra capabilities.

  • Optional hand‑written SIMD gives performance; compiling without it yields fully memory‑safe Rust code (only zerocopy unsafe).
  • Benchmarks: 1‑thread lossy 1.19×, lossless 2.74× faster; multi‑thread lossy 2.68×, lossless 3.19× faster than libwebp.
  • Provides full libwebp feature parity plus extras: raw subframe inspection, explicit thread‑count, parallel animation decode‑ahead, C ABI and Rust API.
  • Safety focus: replaces unsafe C SIMD intrinsics with Rust; memory‑safety bugs like CVE‑2023‑4863 mitigated.

Anyone building software that decodes untrusted WebP images—browsers, messaging apps, or image pipelines—needs a faster, memory‑safe alternative to libwebp.

7/10

Related reading

  1. Decoding Looped Transformers Better for (Almost) Free

    LoopCD is a training‑free contrastive decoding technique for looped Transformers that leverages earlier recurrent states as weak predictions to guide token selection, either via an extra logit pass (LoopCD‑Logits) or directly in hidden‑state space (LoopCD‑Hidden). It improves pass@1 on code benchmarks by up to 9% absolute and enables halving recurrent loops, cutting inference FLOPs by 22‑48%.

    Hugging Face Daily Papersarxiv.org1 minpaper
  2. RiskChainBench: A Benchmark for Obfuscated Platform Message Restoration and Evidence-Grounded Web Investigation

    RiskChainBench is a new benchmark that pairs synthetic obfuscated message restoration inputs with human‑labeled local web environments, requiring models to both decode malicious instructions and investigate the linked site. Across ten models, restoration accuracy varies widely and web‑agent failures dominate the error budget.

    Hugging Face Daily Papersarxiv.org1 minpaper
  3. GSoC 2026: Improving HLSL Support in clangd

    The GSoC 2026 project identified and closed numerous gaps in clangd’s HLSL support. By performing a systematic gap analysis, the author authored an RFC, opened 11 GitHub issues, and landed multiple PRs that improve hover (semantic annotations, out/inout parameters, swizzles, RootSignature, register ranges) and code‑completion (attributes, register, semantic annotations, swizzle). The work also ex…

    LLVM Project Blogllvm.org8 min