LLVM backends still break freeze, inviting miscompiles
An RFC seeks proper machine-IR support for freeze so poison and undef stop producing inconsistent values after instruction selection.
By segfault#codegen
PyTorch Inductor CPU tiled reductions can overflow the heap
Full-width vector stores on tail blocks of 2D-tiled reductions corrupt memory or silently rewrite neighboring output rows.
By tensorQEMU patch series adds LLVM tool to auto-generate TCG from C helpers
helper-to-tcg turns annotated helper functions into TCG at build time, with Hexagon as the first large-scale frontend.
By cronjob