6. GlossaryΒΆ

basic block
A sequence of instructions inside a function. A basic block always starts with a label and ends with a terminator. No other instruction inside the basic block can transfer control out of the block.
function declaration
The specification of a function’s prototype (including the argument and return types, and other information such as the calling convention), without an associated implementation. This is like a extern function declaration in C.
function definition
A function’s prototype (like in a function declaration) plus a body implementing the function.
getelementptr

A LLVM instruction allowing to get the address of a subelement of an aggregate data structure.

See also

Official documentation: ‘getelementptr’ Instruction

global value
A named value accessible to all members of a module.
global variable

A variable whose value is accessible to all members of a module. Under the hood, it is a constant pointer to a module-allocated slot of the given type.

All global variables are global values. However, the converse is not true: a function declaration or definition is not a global variable; it is only a global value.

instruction
The fundamental element(s) used in implementing a LLVM function. LLVM instructions define a procedural assembly-like language.
IR
Intermediate Representation
A high-level assembly language describing to LLVM the code to be compiled to native code.
label
A branch target inside a function. A label always denotes the start of a basic block.
metadata
Optional ancillary information which can be associated with LLVM instructions, functions, etc. Metadata is used to convey certain information which is not critical to the compiling of LLVM IR (such as the likelihood of a condition branch or the source code location corresponding to a given instruction).
module
A compilation unit for LLVM IR. A module can contain any number of function declarations and definitions, global variables, and metadata.
terminator
terminator instruction
A kind of instruction which explicitly transfers control to another part of the program (instead of simply going to the next instruction after it is executed). Examples are branches and function returns.