llvm

Crystal 1.22.0 正式发布，包含 251 项变更和 33 位贡献者的参与，主要亮点包括标准流模拟 SIGPIPE 行为、全面重构的 DWARF 解析以提升性能、所有目标平台支持原生事件循环、支持 LLVM 23/24，以及 aarch64-linux 晋升为 Tier 1 平台。

A blog post highlighting why 2026 is a landmark year for the Rust compiler, with the next-gen trait solver, the Polonius borrow checker, the parallel frontend nearing stabilization, and significant LLVM performance gains all converging at once.

Modular 团队将参加 2026 年 LLVM 开发者大会，Mojo 团队举办全天工作坊讲解 Mojo 内部原理与路线图，Spenser Bauman 将分享 MAX 的 MLIR/LLVM 图编译器技术，并与 Qualcomm 联合赞助及现场招聘。

A blog post shows how LLVM compiles branch-free bit-manipulation code (like 128-bit integer addition and constant-time select) into actual conditional branches on RISC-V targets, because RV32/64 lack conditional-move instructions, and how this behavior has changed across clang versions.

Qiniu Cloud's Lao Xu team open-sourced LLGo, an LLVM-based Go compiler that eliminates cgo overhead by enabling zero-cost native C ABI calls and integrating C, Python, and C++ ecosystems into Go, with support for disabling GC and targets ranging from games to Llama 2 inference and WebAssembly.

RacketCon 2026, the sixteenth gathering for the Racket programming language community, is happening Saturday with a live stream and talks on notation, type inference in Typed Racket, and low-level language-oriented programming via the new Pille language.

The article reports a 4.57% mean reduction in Rust compiler wall-time over two months, highlighting optimizations in rustdoc, Clippy, LLVM upgrade, and the new Polonius borrow checker and Penelope Hammertime trait solver.

A tribute to Johannes Doerfert, highlighting his significant contributions to the LLVM compiler project, including work on the Attributor framework, OpenMP offloading, and polyhedral optimization, and his impact on the open-source community.

Metacarp is a new compiler for the Carp programming language, written in Carp itself, featuring modular libraries, an LLVM backend, and incremental JIT compilation capabilities.

TinyGo 0.42 introduces recoverable panics, Go 1.27 support, UEFI targets, and expanded hardware support for ESP32 and STM32, enhancing its capabilities for embedded systems development.

The article describes how switching a BPF program to use BTF-generated vmlinux.h caused incorrect IP checksums due to Clang's Type-Based Alias Analysis optimization, and details the debugging process through assembly inspection.

The author details converting the scheme-rs Scheme implementation to a CPS-based JIT compiler and applying three optimizations, including beta-reduction, to achieve a 25x performance improvement.

This article discusses how MLIR compiler passes can introduce NaN values through constant folding even with valid operands, and proposes a verification workflow to catch and fix such issues at the IR boundary.

LLVM 23 delivers a 6.75% reduction in compile-time for -O3 builds, driven by key improvements in ADT hash maps, sets, and other components.

This paper presents a zero-overhead, multi-vendor GPU compilation framework built into the Rust compiler, leveraging Rust's ownership model to ensure memory safety and achieve competitive performance with native CUDA and HIP baselines.

A YouTube playlist featuring talks and slides from the 2026 EuroLLVM conference, a meeting for LLVM developers and enthusiasts.

A technical blog post explaining how to compute graph dominators, comparing the Lengauer-Tarjan algorithm with the 'A Simple, Fast Dominance Algorithm' and providing intuition behind the data-flow approach.

The article describes methods to find missed alarm bugs in the Alive2 formal verification tool by adapting the YARPGen random program generator for differential testing.

A developer discovers a decades-old bug in Knuth's Algorithm D for long division, leading to a new theorem added to the TAOCP errata, and also uncovers a related bug in LLVM's implementation.

RPCS3's ARM port now runs 60% faster and draws 25% less power after fixing a busy-wait timer bug, replacing x86 pause with ARM ISB, and reworking LLVM code generation. The gains come from low-level ARM optimizations driven by a cheap Android handheld test device.