Extending LLVM: Adding instructions, intrinsics, types, etc.
- Introduction and Warning
- Adding a new intrinsic function
- Adding a new instruction
- Adding a new SelectionDAG node
- Adding a new type
- Adding a new fundamental type
- Adding a new derived type
During the course of using LLVM, you may wish to customize it for your
research project or for experimentation. At this point, you may realize that
you need to add something to LLVM, whether it be a new fundamental type, a new
intrinsic function, or a whole new instruction.
When you come to this realization, stop and think. Do you really need to
extend LLVM? Is it a new fundamental capability that LLVM does not support at
its current incarnation or can it be synthesized from already pre-existing LLVM
elements? If you are not sure, ask on the LLVM-dev list. The
reason is that extending LLVM will get involved as you need to update all the
different passes that you intend to use with your extension, and there are
many LLVM analyses and transformations, so it may be quite a bit of
Adding an intrinsic function is far easier than
adding an instruction, and is transparent to optimization passes. If your added
functionality can be expressed as a
function call, an intrinsic function is the method of choice for LLVM
Before you invest a significant amount of effort into a non-trivial
extension, ask on the list if what you are
looking to do can be done with already-existing infrastructure, or if maybe
someone else is already working on it. You will save yourself a lot of time and
effort by doing so.
Adding a new intrinsic function to LLVM is much easier than adding a new
instruction. Almost all extensions to LLVM should start as an intrinsic
function and then be turned into an instruction if warranted.
Document the intrinsic. Decide whether it is code generator specific and
what the restrictions are. Talk to other people about it so that you are
sure it's a good idea.
Add an entry for your intrinsic. Describe its memory access characteristics
for optimization (this controls whether it will be DCE'd, CSE'd, etc). Note
that any intrinsic using the llvm_int_ty type for an argument will
be deemed by tblgen as overloaded and the corresponding suffix
will be required on the intrinsic's name.
- llvm/lib/Analysis/ConstantFolding.cpp: If it is possible to
constant fold your intrinsic, add support to it in the
canConstantFoldCallTo and ConstantFoldCall functions.
- llvm/test/Regression/*: Add test cases for your test cases to the
Once the intrinsic has been added to the system, you must add code generator
support for it. Generally you must do the following steps:
- Add support to the C backend in lib/Target/CBackend/
- Depending on the intrinsic, there are a few ways to implement this. For
most intrinsics, it makes sense to add code to lower your intrinsic in
LowerIntrinsicCall in lib/CodeGen/IntrinsicLowering.cpp.
Second, if it makes sense to lower the intrinsic to an expanded sequence of
C code in all cases, just emit the expansion in visitCallInst in
Writer.cpp. If the intrinsic has some way to express it with GCC
(or any other compiler) extensions, it can be conditionally supported based
on the compiler compiling the CBE output (see llvm.prefetch for an
example). Third, if the intrinsic really has no way to be lowered, just
have the code generator emit code that prints an error message and calls
abort if executed.
- Add support to the .td file for the target(s) of your choice in
- This is usually a matter of adding a pattern to the .td file that matches
the intrinsic, though it may obviously require adding the instructions you
want to generate as well. There are lots of examples in the PowerPC and X86
backend to follow.
As with intrinsics, adding a new SelectionDAG node to LLVM is much easier
than adding a new instruction. New nodes are often added to help represent
instructions common to many targets. These nodes often map to an LLVM
instruction (add, sub) or intrinsic (byteswap, population count). In other
cases, new nodes have been added to allow many targets to perform a common task
(converting between floating point and integer representation) or capture more
complicated behavior in a single node (rotate).
Add an enum value for the new SelectionDAG node.
Add code to print the node to getOperationName. If your new node
can be evaluated at compile time when given constant arguments (such as an
add of a constant with another constant), find the getNode method
that takes the appropriate number of arguments, and add a case for your node
to the switch statement that performs constant folding for nodes that take
the same number of arguments as your new node.
Add code to legalize,
promote, and expand the node as necessary. At a minimum, you will need
to add a case statement for your node in LegalizeOp which calls
LegalizeOp on the node's operands, and returns a new node if any of the
operands changed as a result of being legalized. It is likely that not all
targets supported by the SelectionDAG framework will natively support the
new node. In this case, you must also add code in your node's case
statement in LegalizeOp to Expand your node into simpler, legal
operations. The case for ISD::UREM for expanding a remainder into
a divide, multiply, and a subtract is a good example.
If targets may support the new node being added only at certain sizes, you
will also need to add code to your node's case statement in
LegalizeOp to Promote your node's operands to a larger size, and
perform the correct operation. You will also need to add code to
PromoteOp to do this as well. For a good example, see
which promotes its operand to a wider size, performs the byteswap, and then
shifts the correct bytes right to emulate the narrower byteswap in the
Add a case for your node in ExpandOp to teach the legalizer how to
perform the action represented by the new node on a value that has been
split into high and low halves. This case will be used to support your
node with a 64 bit operand on a 32 bit target.
If your node can be combined with itself, or other existing nodes in a
peephole-like fashion, add a visit function for it, and call that function
from . There are several good examples for simple combines you
can do; visitFABS and visitSRL are good starting places.
Each target has an implementation of the TargetLowering class,
usually in its own file (although some targets include it in the same
file as the DAGToDAGISel). The default behavior for a target is to
assume that your new node is legal for all types that are legal for
that target. If this target does not natively support your node, then
tell the target to either Promote it (if it is supported at a larger
type) or Expand it. This will cause the code you wrote in
LegalizeOp above to decompose your new node into other legal
nodes for this target.
Most current targets supported by LLVM generate code using the DAGToDAG
method, where SelectionDAG nodes are pattern matched to target-specific
nodes, which represent individual instructions. In order for the targets
to match an instruction to your new node, you must add a def for that node
to the list in this file, with the appropriate type constraints. Look at
add, bswap, and fadd for examples.
Each target has a tablegen file that describes the target's instruction
set. For targets that use the DAGToDAG instruction selection framework,
add a pattern for your new node that uses one or more target nodes.
Documentation for this is a bit sparse right now, but there are several
decent examples. See the patterns for rotl in
- TODO: document complex patterns.
- llvm/test/Regression/CodeGen/*: Add test cases for your new node
to the test suite. llvm/test/Regression/CodeGen/X86/bswap.ll is
a good example.
WARNING: adding instructions changes the bitcode
format, and it will take some effort to maintain compatibility with
the previous version. Only add an instruction if it is absolutely
add a number for your instruction and an enum name
add a definition for the class that will represent your instruction
add a prototype for a visitor to your new instruction type
add a new token to parse your instruction from assembly text file
add the grammar on how your instruction can be read and what it will
construct as a result
add a case for your instruction and how it will be parsed from bitcode
add a case for how your instruction will be printed out to assembly
implement the class you defined in
- Test your instruction
Add support for your instruction to code generators, or add a lowering
- llvm/test/Regression/*: add your test cases to the test suite.
Also, you need to implement (or modify) any analyses or passes that you want
to understand this new instruction.
WARNING: adding new types changes the bitcode
format, and will break compatibility with currently-existing LLVM
installations. Only add new types if it is absolutely necessary.
add enum for the new type; add static Type* for this type
add mapping from TypeID => Type*;
initialize the static Type*
add ability to parse in the type from text assembly
add a token for that type
add enum for the new type; add a forward declaration of the type
add new class to represent new class in the hierarchy; add forward
declaration to the TypeMap value type
add support for derived type to:
add necessary member functions for type, and factory methods
std::string getTypeDescription(const Type &Ty,
std::vector<const Type*> &TypeStack)
bool TypesEqual(const Type *Ty, const Type *Ty2,
std::map<const Type*, const Type*> & EqTypes)
add ability to parse in the type from text assembly
modify void BitcodeWriter::outputType(const Type *T) to serialize
modify const Type *BitcodeReader::ParseType() to read your data
to output the new derived type
void calcTypeName(const Type *Ty,
std::vector<const Type*> &TypeStack,
std::map<const Type*,std::string> &TypeNames,
std::string & Result)
The LLVM Compiler Infrastructure
Last modified: $Date$