How To Use Attributes

Introduction

Attributes in LLVM have changed in some fundamental ways. It was necessary todo this to support expanding the attributes to encompass more than a handful ofattributes — e.g. command line options. The old way of handling attributesconsisted of representing them as a bit mask of values. This bit mask wasstored in a “list” structure that was reference counted. The advantage of thiswas that attributes could be manipulated with ‘or’s and ‘and’s. Thedisadvantage of this was that there was limited room for expansion, andvirtually no support for attribute-value pairs other than alignment.

In the new scheme, an Attribute object represents a single attribute that’suniqued. You use the Attribute::get methods to create a new Attributeobject. An attribute can be a single “enum” value (the enum being theAttribute::AttrKind enum), a string representing a target-dependentattribute, or an attribute-value pair. Some examples:

• Target-independent: noinline, zext
• Target-dependent: "no-sse", "thumb2"
• Attribute-value pair: "cpu" = "cortex-a8", align = 4

Note: for an attribute value pair, we expect a target-dependent attribute tohave a string for the value.

Attribute

An Attribute object is designed to be passed around by value.

Because attributes are no longer represented as a bit mask, you will need toconvert any code which does treat them as a bit mask to use the new querymethods on the Attribute class.

AttributeList

The AttributeList stores a collection of Attribute objects for each kind ofobject that may have an attribute associated with it: the function as a whole,the return type, or the function’s parameters. A function’s attributes are atindex AttributeList::FunctionIndex; the return type’s attributes are atindex AttributeList::ReturnIndex; and the function’s parameters’ attributesare at indices 1, …, n (where ‘n’ is the number of parameters). Most methodson the AttributeList class take an index parameter.

An AttributeList is also a uniqued and immutable object. You create anAttributeList through the AttributeList::get methods. You can add andremove attributes, which result in the creation of a new AttributeList.

An AttributeList object is designed to be passed around by value.

Note: It is advised that you do not use the AttributeList “introspection”methods (e.g. Raw, getRawPointer, etc.). These methods breakencapsulation, and may be removed in a future release (i.e. LLVM 4.0).

AttrBuilder

Lastly, we have a “builder” class to help create the AttributeList objectwithout having to create several different intermediate uniquedAttributeList objects. The AttrBuilder class allows you to add andremove attributes at will. The attributes won’t be uniqued until you call theappropriate AttributeList::get method.

An AttrBuilder object is not designed to be passed around by value. Itshould be passed by reference.

Note: It is advised that you do not use the AttrBuilder::addRawValue()method or the AttrBuilder(uint64_t Val) constructor. These are forbackwards compatibility and may be removed in a future release (i.e. LLVM 4.0).

And that’s basically it! A lot of functionality is hidden behind these classes,but the interfaces are pretty straight forward.