LLVM 11.0.0 releases: Low-Level Virtual Machine

by · Published · Updated

LLVM is short for Low-Level Virtual Machine. This library provides compiler-related support and can be used as a background for multi-language compilers. It can be used in programming languages. Compilation optimization, link optimization, online compilation optimization, and code generation.

LLVM began as a research project at the University of Illinois, with the goal of providing a modern, SSA-based compilation strategy capable of supporting both static and dynamic compilation of arbitrary programming languages. Since then, LLVM has grown to be an umbrella project consisting of a number of subprojects, many of which are being used in production by a wide variety of commercial and open source projects as well as being widely used in academic research. Code in the LLVM project is licensed under the “UIUC” BSD-Style license.

LLVM 11.0.0 is released.

Changelog

  • The callsite attribute vector-function-abi-variant has been added to describe the mapping between scalar functions and vector functions, to enable vectorization of call sites. The information provided by the attribute is interfaced via the API provided by the VFDatabase class. When scanning through the set of vector functions associated with a scalar call, the loop vectorizer now relies on VFDatabase, instead of TargetLibraryInfo.
  • dereferenceable attributes and metadata on pointers no longer imply anything about the alignment of the pointer in question. Previously, some optimizations would make assumptions based on the type of the pointer. This behavior was undocumented. To preserve optimizations, frontends may need to be updated to generate appropriate align attributes and metadata.
  • The DIModule metadata is extended to contain file and line number information. This information is used to represent Fortran modules debug info at IR level.
  • LLVM IR now supports two distinct llvm::FixedVectorType and llvm::ScalableVectorType vector types, both derived from the base class llvm::VectorType. A number of algorithms dealing with IR vector types have been updated to make sure they work for both scalable and fixed vector types. Where possible, the code has been made generic to cover both cases using the base class. Specifically, places that were using the type unsigned to count the number of lanes of a vector are now using llvm::ElementCount. In places where uint64_t was used to denote the size in bits of a IR type we have partially migrated the codebase to using llvm::TypeSize.
  • Branching on undef/poison is undefined behavior. It is needed for correctly analyzing value ranges based on branch conditions. This is consistent with MSan’s behavior as well.
  • memset/memcpy/memmove can take undef/poison pointer(s) if the size to fill is zero.
  • Passing undef/poison to a standard I/O library function call (printf/fputc/…) is undefined behavior. The new noundef attribute is attached to the functions’ arguments. The full list is available at llvm::inferLibFuncAttributes.

Download

Tags: