1 <!DOCTYPE HTML PUBLIC "-//W3C//DTD HTML 4.01//EN"
2 "http://www.w3.org/TR/html4/strict.dtd">
6 <title>Kaleidoscope: Adding JIT and Optimizer Support</title>
7 <meta http-equiv="Content-Type" content="text/html; charset=utf-8">
8 <meta name="author" content="Chris Lattner">
9 <link rel="stylesheet" href="../llvm.css" type="text/css">
14 <h1>Kaleidoscope: Adding JIT and Optimizer Support</h1>
17 <li><a href="index.html">Up to Tutorial Index</a></li>
20 <li><a href="#intro">Chapter 4 Introduction</a></li>
21 <li><a href="#trivialconstfold">Trivial Constant Folding</a></li>
22 <li><a href="#optimizerpasses">LLVM Optimization Passes</a></li>
23 <li><a href="#jit">Adding a JIT Compiler</a></li>
24 <li><a href="#code">Full Code Listing</a></li>
27 <li><a href="LangImpl5.html">Chapter 5</a>: Extending the Language: Control
31 <div class="doc_author">
32 <p>Written by <a href="mailto:sabre@nondot.org">Chris Lattner</a></p>
35 <!-- *********************************************************************** -->
36 <h2><a name="intro">Chapter 4 Introduction</a></h2>
37 <!-- *********************************************************************** -->
41 <p>Welcome to Chapter 4 of the "<a href="index.html">Implementing a language
42 with LLVM</a>" tutorial. Chapters 1-3 described the implementation of a simple
43 language and added support for generating LLVM IR. This chapter describes
44 two new techniques: adding optimizer support to your language, and adding JIT
45 compiler support. These additions will demonstrate how to get nice, efficient code
46 for the Kaleidoscope language.</p>
50 <!-- *********************************************************************** -->
51 <h2><a name="trivialconstfold">Trivial Constant Folding</a></h2>
52 <!-- *********************************************************************** -->
57 Our demonstration for Chapter 3 is elegant and easy to extend. Unfortunately,
58 it does not produce wonderful code. The IRBuilder, however, does give us
59 obvious optimizations when compiling simple code:</p>
61 <div class="doc_code">
63 ready> <b>def test(x) 1+2+x;</b>
64 Read function definition:
65 define double @test(double %x) {
67 %addtmp = fadd double 3.000000e+00, %x
73 <p>This code is not a literal transcription of the AST built by parsing the
76 <div class="doc_code">
78 ready> <b>def test(x) 1+2+x;</b>
79 Read function definition:
80 define double @test(double %x) {
82 %addtmp = fadd double 2.000000e+00, 1.000000e+00
83 %addtmp1 = fadd double %addtmp, %x
89 <p>Constant folding, as seen above, in particular, is a very common and very
90 important optimization: so much so that many language implementors implement
91 constant folding support in their AST representation.</p>
93 <p>With LLVM, you don't need this support in the AST. Since all calls to build
94 LLVM IR go through the LLVM IR builder, the builder itself checked to see if
95 there was a constant folding opportunity when you call it. If so, it just does
96 the constant fold and return the constant instead of creating an instruction.
98 <p>Well, that was easy :). In practice, we recommend always using
99 <tt>IRBuilder</tt> when generating code like this. It has no
100 "syntactic overhead" for its use (you don't have to uglify your compiler with
101 constant checks everywhere) and it can dramatically reduce the amount of
102 LLVM IR that is generated in some cases (particular for languages with a macro
103 preprocessor or that use a lot of constants).</p>
105 <p>On the other hand, the <tt>IRBuilder</tt> is limited by the fact
106 that it does all of its analysis inline with the code as it is built. If you
107 take a slightly more complex example:</p>
109 <div class="doc_code">
111 ready> <b>def test(x) (1+2+x)*(x+(1+2));</b>
112 ready> Read function definition:
113 define double @test(double %x) {
115 %addtmp = fadd double 3.000000e+00, %x
116 %addtmp1 = fadd double %x, 3.000000e+00
117 %multmp = fmul double %addtmp, %addtmp1
123 <p>In this case, the LHS and RHS of the multiplication are the same value. We'd
124 really like to see this generate "<tt>tmp = x+3; result = tmp*tmp;</tt>" instead
125 of computing "<tt>x+3</tt>" twice.</p>
127 <p>Unfortunately, no amount of local analysis will be able to detect and correct
128 this. This requires two transformations: reassociation of expressions (to
129 make the add's lexically identical) and Common Subexpression Elimination (CSE)
130 to delete the redundant add instruction. Fortunately, LLVM provides a broad
131 range of optimizations that you can use, in the form of "passes".</p>
135 <!-- *********************************************************************** -->
136 <h2><a name="optimizerpasses">LLVM Optimization Passes</a></h2>
137 <!-- *********************************************************************** -->
141 <p>LLVM provides many optimization passes, which do many different sorts of
142 things and have different tradeoffs. Unlike other systems, LLVM doesn't hold
143 to the mistaken notion that one set of optimizations is right for all languages
144 and for all situations. LLVM allows a compiler implementor to make complete
145 decisions about what optimizations to use, in which order, and in what
148 <p>As a concrete example, LLVM supports both "whole module" passes, which look
149 across as large of body of code as they can (often a whole file, but if run
150 at link time, this can be a substantial portion of the whole program). It also
151 supports and includes "per-function" passes which just operate on a single
152 function at a time, without looking at other functions. For more information
153 on passes and how they are run, see the <a href="../WritingAnLLVMPass.html">How
154 to Write a Pass</a> document and the <a href="../Passes.html">List of LLVM
157 <p>For Kaleidoscope, we are currently generating functions on the fly, one at
158 a time, as the user types them in. We aren't shooting for the ultimate
159 optimization experience in this setting, but we also want to catch the easy and
160 quick stuff where possible. As such, we will choose to run a few per-function
161 optimizations as the user types the function in. If we wanted to make a "static
162 Kaleidoscope compiler", we would use exactly the code we have now, except that
163 we would defer running the optimizer until the entire file has been parsed.</p>
165 <p>In order to get per-function optimizations going, we need to set up a
166 <a href="../WritingAnLLVMPass.html#passmanager">FunctionPassManager</a> to hold and
167 organize the LLVM optimizations that we want to run. Once we have that, we can
168 add a set of optimizations to run. The code looks like this:</p>
170 <div class="doc_code">
172 FunctionPassManager OurFPM(TheModule);
174 // Set up the optimizer pipeline. Start with registering info about how the
175 // target lays out data structures.
176 OurFPM.add(new TargetData(*TheExecutionEngine->getTargetData()));
177 // Provide basic AliasAnalysis support for GVN.
178 OurFPM.add(createBasicAliasAnalysisPass());
179 // Do simple "peephole" optimizations and bit-twiddling optzns.
180 OurFPM.add(createInstructionCombiningPass());
181 // Reassociate expressions.
182 OurFPM.add(createReassociatePass());
183 // Eliminate Common SubExpressions.
184 OurFPM.add(createGVNPass());
185 // Simplify the control flow graph (deleting unreachable blocks, etc).
186 OurFPM.add(createCFGSimplificationPass());
188 OurFPM.doInitialization();
190 // Set the global so the code gen can use this.
191 TheFPM = &OurFPM;
193 // Run the main "interpreter loop" now.
198 <p>This code defines a <tt>FunctionPassManager</tt>, "<tt>OurFPM</tt>". It
199 requires a pointer to the <tt>Module</tt> to construct itself. Once it is set
200 up, we use a series of "add" calls to add a bunch of LLVM passes. The first
201 pass is basically boilerplate, it adds a pass so that later optimizations know
202 how the data structures in the program are laid out. The
203 "<tt>TheExecutionEngine</tt>" variable is related to the JIT, which we will get
204 to in the next section.</p>
206 <p>In this case, we choose to add 4 optimization passes. The passes we chose
207 here are a pretty standard set of "cleanup" optimizations that are useful for
208 a wide variety of code. I won't delve into what they do but, believe me,
209 they are a good starting place :).</p>
211 <p>Once the PassManager is set up, we need to make use of it. We do this by
212 running it after our newly created function is constructed (in
213 <tt>FunctionAST::Codegen</tt>), but before it is returned to the client:</p>
215 <div class="doc_code">
217 if (Value *RetVal = Body->Codegen()) {
218 // Finish off the function.
219 Builder.CreateRet(RetVal);
221 // Validate the generated code, checking for consistency.
222 verifyFunction(*TheFunction);
224 <b>// Optimize the function.
225 TheFPM->run(*TheFunction);</b>
232 <p>As you can see, this is pretty straightforward. The
233 <tt>FunctionPassManager</tt> optimizes and updates the LLVM Function* in place,
234 improving (hopefully) its body. With this in place, we can try our test above
237 <div class="doc_code">
239 ready> <b>def test(x) (1+2+x)*(x+(1+2));</b>
240 ready> Read function definition:
241 define double @test(double %x) {
243 %addtmp = fadd double %x, 3.000000e+00
244 %multmp = fmul double %addtmp, %addtmp
250 <p>As expected, we now get our nicely optimized code, saving a floating point
251 add instruction from every execution of this function.</p>
253 <p>LLVM provides a wide variety of optimizations that can be used in certain
254 circumstances. Some <a href="../Passes.html">documentation about the various
255 passes</a> is available, but it isn't very complete. Another good source of
256 ideas can come from looking at the passes that <tt>llvm-gcc</tt> or
257 <tt>llvm-ld</tt> run to get started. The "<tt>opt</tt>" tool allows you to
258 experiment with passes from the command line, so you can see if they do
261 <p>Now that we have reasonable code coming out of our front-end, lets talk about
266 <!-- *********************************************************************** -->
267 <h2><a name="jit">Adding a JIT Compiler</a></h2>
268 <!-- *********************************************************************** -->
272 <p>Code that is available in LLVM IR can have a wide variety of tools
273 applied to it. For example, you can run optimizations on it (as we did above),
274 you can dump it out in textual or binary forms, you can compile the code to an
275 assembly file (.s) for some target, or you can JIT compile it. The nice thing
276 about the LLVM IR representation is that it is the "common currency" between
277 many different parts of the compiler.
280 <p>In this section, we'll add JIT compiler support to our interpreter. The
281 basic idea that we want for Kaleidoscope is to have the user enter function
282 bodies as they do now, but immediately evaluate the top-level expressions they
283 type in. For example, if they type in "1 + 2;", we should evaluate and print
284 out 3. If they define a function, they should be able to call it from the
287 <p>In order to do this, we first declare and initialize the JIT. This is done
288 by adding a global variable and a call in <tt>main</tt>:</p>
290 <div class="doc_code">
292 <b>static ExecutionEngine *TheExecutionEngine;</b>
296 <b>// Create the JIT. This takes ownership of the module.
297 TheExecutionEngine = EngineBuilder(TheModule).create();</b>
303 <p>This creates an abstract "Execution Engine" which can be either a JIT
304 compiler or the LLVM interpreter. LLVM will automatically pick a JIT compiler
305 for you if one is available for your platform, otherwise it will fall back to
308 <p>Once the <tt>ExecutionEngine</tt> is created, the JIT is ready to be used.
309 There are a variety of APIs that are useful, but the simplest one is the
310 "<tt>getPointerToFunction(F)</tt>" method. This method JIT compiles the
311 specified LLVM Function and returns a function pointer to the generated machine
312 code. In our case, this means that we can change the code that parses a
313 top-level expression to look like this:</p>
315 <div class="doc_code">
317 static void HandleTopLevelExpression() {
318 // Evaluate a top-level expression into an anonymous function.
319 if (FunctionAST *F = ParseTopLevelExpr()) {
320 if (Function *LF = F->Codegen()) {
321 LF->dump(); // Dump the function for exposition purposes.
323 <b>// JIT the function, returning a function pointer.
324 void *FPtr = TheExecutionEngine->getPointerToFunction(LF);
326 // Cast it to the right type (takes no arguments, returns a double) so we
327 // can call it as a native function.
328 double (*FP)() = (double (*)())(intptr_t)FPtr;
329 fprintf(stderr, "Evaluated to %f\n", FP());</b>
334 <p>Recall that we compile top-level expressions into a self-contained LLVM
335 function that takes no arguments and returns the computed double. Because the
336 LLVM JIT compiler matches the native platform ABI, this means that you can just
337 cast the result pointer to a function pointer of that type and call it directly.
338 This means, there is no difference between JIT compiled code and native machine
339 code that is statically linked into your application.</p>
341 <p>With just these two changes, lets see how Kaleidoscope works now!</p>
343 <div class="doc_code">
345 ready> <b>4+5;</b>
346 Read top-level expression:
349 ret double 9.000000e+00
352 <em>Evaluated to 9.000000</em>
356 <p>Well this looks like it is basically working. The dump of the function
357 shows the "no argument function that always returns double" that we synthesize
358 for each top-level expression that is typed in. This demonstrates very basic
359 functionality, but can we do more?</p>
361 <div class="doc_code">
363 ready> <b>def testfunc(x y) x + y*2; </b>
364 Read function definition:
365 define double @testfunc(double %x, double %y) {
367 %multmp = fmul double %y, 2.000000e+00
368 %addtmp = fadd double %multmp, %x
372 ready> <b>testfunc(4, 10);</b>
373 Read top-level expression:
376 %calltmp = call double @testfunc(double 4.000000e+00, double 1.000000e+01)
380 <em>Evaluated to 24.000000</em>
384 <p>This illustrates that we can now call user code, but there is something a bit
385 subtle going on here. Note that we only invoke the JIT on the anonymous
386 functions that <em>call testfunc</em>, but we never invoked it
387 on <em>testfunc</em> itself. What actually happened here is that the JIT
388 scanned for all non-JIT'd functions transitively called from the anonymous
389 function and compiled all of them before returning
390 from <tt>getPointerToFunction()</tt>.</p>
392 <p>The JIT provides a number of other more advanced interfaces for things like
393 freeing allocated machine code, rejit'ing functions to update them, etc.
394 However, even with this simple code, we get some surprisingly powerful
395 capabilities - check this out (I removed the dump of the anonymous functions,
396 you should get the idea by now :) :</p>
398 <div class="doc_code">
400 ready> <b>extern sin(x);</b>
402 declare double @sin(double)
404 ready> <b>extern cos(x);</b>
406 declare double @cos(double)
408 ready> <b>sin(1.0);</b>
409 Read top-level expression:
412 ret double 0x3FEAED548F090CEE
415 <em>Evaluated to 0.841471</em>
417 ready> <b>def foo(x) sin(x)*sin(x) + cos(x)*cos(x);</b>
418 Read function definition:
419 define double @foo(double %x) {
421 %calltmp = call double @sin(double %x)
422 %multmp = fmul double %calltmp, %calltmp
423 %calltmp2 = call double @cos(double %x)
424 %multmp4 = fmul double %calltmp2, %calltmp2
425 %addtmp = fadd double %multmp, %multmp4
429 ready> <b>foo(4.0);</b>
430 Read top-level expression:
433 %calltmp = call double @foo(double 4.000000e+00)
437 <em>Evaluated to 1.000000</em>
441 <p>Whoa, how does the JIT know about sin and cos? The answer is surprisingly
443 example, the JIT started execution of a function and got to a function call. It
444 realized that the function was not yet JIT compiled and invoked the standard set
445 of routines to resolve the function. In this case, there is no body defined
446 for the function, so the JIT ended up calling "<tt>dlsym("sin")</tt>" on the
447 Kaleidoscope process itself.
448 Since "<tt>sin</tt>" is defined within the JIT's address space, it simply
449 patches up calls in the module to call the libm version of <tt>sin</tt>
452 <p>The LLVM JIT provides a number of interfaces (look in the
453 <tt>ExecutionEngine.h</tt> file) for controlling how unknown functions get
454 resolved. It allows you to establish explicit mappings between IR objects and
455 addresses (useful for LLVM global variables that you want to map to static
456 tables, for example), allows you to dynamically decide on the fly based on the
457 function name, and even allows you to have the JIT compile functions lazily the
458 first time they're called.</p>
460 <p>One interesting application of this is that we can now extend the language
461 by writing arbitrary C++ code to implement operations. For example, if we add:
464 <div class="doc_code">
466 /// putchard - putchar that takes a double and returns 0.
468 double putchard(double X) {
475 <p>Now we can produce simple output to the console by using things like:
476 "<tt>extern putchard(x); putchard(120);</tt>", which prints a lowercase 'x' on
477 the console (120 is the ASCII code for 'x'). Similar code could be used to
478 implement file I/O, console input, and many other capabilities in
481 <p>This completes the JIT and optimizer chapter of the Kaleidoscope tutorial. At
482 this point, we can compile a non-Turing-complete programming language, optimize
483 and JIT compile it in a user-driven way. Next up we'll look into <a
484 href="LangImpl5.html">extending the language with control flow constructs</a>,
485 tackling some interesting LLVM IR issues along the way.</p>
489 <!-- *********************************************************************** -->
490 <h2><a name="code">Full Code Listing</a></h2>
491 <!-- *********************************************************************** -->
496 Here is the complete code listing for our running example, enhanced with the
497 LLVM JIT and optimizer. To build this example, use:
500 <div class="doc_code">
503 clang++ -g toy.cpp `llvm-config --cppflags --ldflags --libs core jit native` -O3 -o toy
510 If you are compiling this on Linux, make sure to add the "-rdynamic" option
511 as well. This makes sure that the external functions are resolved properly
514 <p>Here is the code:</p>
516 <div class="doc_code">
518 #include "llvm/DerivedTypes.h"
519 #include "llvm/ExecutionEngine/ExecutionEngine.h"
520 #include "llvm/ExecutionEngine/JIT.h"
521 #include "llvm/LLVMContext.h"
522 #include "llvm/Module.h"
523 #include "llvm/PassManager.h"
524 #include "llvm/Analysis/Verifier.h"
525 #include "llvm/Analysis/Passes.h"
526 #include "llvm/Target/TargetData.h"
527 #include "llvm/Transforms/Scalar.h"
528 #include "llvm/Support/IRBuilder.h"
529 #include "llvm/Support/TargetSelect.h"
530 #include <cstdio>
531 #include <string>
533 #include <vector>
534 using namespace llvm;
536 //===----------------------------------------------------------------------===//
538 //===----------------------------------------------------------------------===//
540 // The lexer returns tokens [0-255] if it is an unknown character, otherwise one
541 // of these for known things.
546 tok_def = -2, tok_extern = -3,
549 tok_identifier = -4, tok_number = -5
552 static std::string IdentifierStr; // Filled in if tok_identifier
553 static double NumVal; // Filled in if tok_number
555 /// gettok - Return the next token from standard input.
556 static int gettok() {
557 static int LastChar = ' ';
559 // Skip any whitespace.
560 while (isspace(LastChar))
561 LastChar = getchar();
563 if (isalpha(LastChar)) { // identifier: [a-zA-Z][a-zA-Z0-9]*
564 IdentifierStr = LastChar;
565 while (isalnum((LastChar = getchar())))
566 IdentifierStr += LastChar;
568 if (IdentifierStr == "def") return tok_def;
569 if (IdentifierStr == "extern") return tok_extern;
570 return tok_identifier;
573 if (isdigit(LastChar) || LastChar == '.') { // Number: [0-9.]+
577 LastChar = getchar();
578 } while (isdigit(LastChar) || LastChar == '.');
580 NumVal = strtod(NumStr.c_str(), 0);
584 if (LastChar == '#') {
585 // Comment until end of line.
586 do LastChar = getchar();
587 while (LastChar != EOF && LastChar != '\n' && LastChar != '\r');
593 // Check for end of file. Don't eat the EOF.
597 // Otherwise, just return the character as its ascii value.
598 int ThisChar = LastChar;
599 LastChar = getchar();
603 //===----------------------------------------------------------------------===//
604 // Abstract Syntax Tree (aka Parse Tree)
605 //===----------------------------------------------------------------------===//
607 /// ExprAST - Base class for all expression nodes.
610 virtual ~ExprAST() {}
611 virtual Value *Codegen() = 0;
614 /// NumberExprAST - Expression class for numeric literals like "1.0".
615 class NumberExprAST : public ExprAST {
618 NumberExprAST(double val) : Val(val) {}
619 virtual Value *Codegen();
622 /// VariableExprAST - Expression class for referencing a variable, like "a".
623 class VariableExprAST : public ExprAST {
626 VariableExprAST(const std::string &name) : Name(name) {}
627 virtual Value *Codegen();
630 /// BinaryExprAST - Expression class for a binary operator.
631 class BinaryExprAST : public ExprAST {
635 BinaryExprAST(char op, ExprAST *lhs, ExprAST *rhs)
636 : Op(op), LHS(lhs), RHS(rhs) {}
637 virtual Value *Codegen();
640 /// CallExprAST - Expression class for function calls.
641 class CallExprAST : public ExprAST {
643 std::vector<ExprAST*> Args;
645 CallExprAST(const std::string &callee, std::vector<ExprAST*> &args)
646 : Callee(callee), Args(args) {}
647 virtual Value *Codegen();
650 /// PrototypeAST - This class represents the "prototype" for a function,
651 /// which captures its name, and its argument names (thus implicitly the number
652 /// of arguments the function takes).
655 std::vector<std::string> Args;
657 PrototypeAST(const std::string &name, const std::vector<std::string> &args)
658 : Name(name), Args(args) {}
663 /// FunctionAST - This class represents a function definition itself.
668 FunctionAST(PrototypeAST *proto, ExprAST *body)
669 : Proto(proto), Body(body) {}
674 //===----------------------------------------------------------------------===//
676 //===----------------------------------------------------------------------===//
678 /// CurTok/getNextToken - Provide a simple token buffer. CurTok is the current
679 /// token the parser is looking at. getNextToken reads another token from the
680 /// lexer and updates CurTok with its results.
682 static int getNextToken() {
683 return CurTok = gettok();
686 /// BinopPrecedence - This holds the precedence for each binary operator that is
688 static std::map<char, int> BinopPrecedence;
690 /// GetTokPrecedence - Get the precedence of the pending binary operator token.
691 static int GetTokPrecedence() {
692 if (!isascii(CurTok))
695 // Make sure it's a declared binop.
696 int TokPrec = BinopPrecedence[CurTok];
697 if (TokPrec <= 0) return -1;
701 /// Error* - These are little helper functions for error handling.
702 ExprAST *Error(const char *Str) { fprintf(stderr, "Error: %s\n", Str);return 0;}
703 PrototypeAST *ErrorP(const char *Str) { Error(Str); return 0; }
704 FunctionAST *ErrorF(const char *Str) { Error(Str); return 0; }
706 static ExprAST *ParseExpression();
710 /// ::= identifier '(' expression* ')'
711 static ExprAST *ParseIdentifierExpr() {
712 std::string IdName = IdentifierStr;
714 getNextToken(); // eat identifier.
716 if (CurTok != '(') // Simple variable ref.
717 return new VariableExprAST(IdName);
720 getNextToken(); // eat (
721 std::vector<ExprAST*> Args;
724 ExprAST *Arg = ParseExpression();
728 if (CurTok == ')') break;
731 return Error("Expected ')' or ',' in argument list");
739 return new CallExprAST(IdName, Args);
742 /// numberexpr ::= number
743 static ExprAST *ParseNumberExpr() {
744 ExprAST *Result = new NumberExprAST(NumVal);
745 getNextToken(); // consume the number
749 /// parenexpr ::= '(' expression ')'
750 static ExprAST *ParseParenExpr() {
751 getNextToken(); // eat (.
752 ExprAST *V = ParseExpression();
756 return Error("expected ')'");
757 getNextToken(); // eat ).
762 /// ::= identifierexpr
765 static ExprAST *ParsePrimary() {
767 default: return Error("unknown token when expecting an expression");
768 case tok_identifier: return ParseIdentifierExpr();
769 case tok_number: return ParseNumberExpr();
770 case '(': return ParseParenExpr();
775 /// ::= ('+' primary)*
776 static ExprAST *ParseBinOpRHS(int ExprPrec, ExprAST *LHS) {
777 // If this is a binop, find its precedence.
779 int TokPrec = GetTokPrecedence();
781 // If this is a binop that binds at least as tightly as the current binop,
782 // consume it, otherwise we are done.
783 if (TokPrec < ExprPrec)
786 // Okay, we know this is a binop.
788 getNextToken(); // eat binop
790 // Parse the primary expression after the binary operator.
791 ExprAST *RHS = ParsePrimary();
794 // If BinOp binds less tightly with RHS than the operator after RHS, let
795 // the pending operator take RHS as its LHS.
796 int NextPrec = GetTokPrecedence();
797 if (TokPrec < NextPrec) {
798 RHS = ParseBinOpRHS(TokPrec+1, RHS);
799 if (RHS == 0) return 0;
803 LHS = new BinaryExprAST(BinOp, LHS, RHS);
808 /// ::= primary binoprhs
810 static ExprAST *ParseExpression() {
811 ExprAST *LHS = ParsePrimary();
814 return ParseBinOpRHS(0, LHS);
818 /// ::= id '(' id* ')'
819 static PrototypeAST *ParsePrototype() {
820 if (CurTok != tok_identifier)
821 return ErrorP("Expected function name in prototype");
823 std::string FnName = IdentifierStr;
827 return ErrorP("Expected '(' in prototype");
829 std::vector<std::string> ArgNames;
830 while (getNextToken() == tok_identifier)
831 ArgNames.push_back(IdentifierStr);
833 return ErrorP("Expected ')' in prototype");
836 getNextToken(); // eat ')'.
838 return new PrototypeAST(FnName, ArgNames);
841 /// definition ::= 'def' prototype expression
842 static FunctionAST *ParseDefinition() {
843 getNextToken(); // eat def.
844 PrototypeAST *Proto = ParsePrototype();
845 if (Proto == 0) return 0;
847 if (ExprAST *E = ParseExpression())
848 return new FunctionAST(Proto, E);
852 /// toplevelexpr ::= expression
853 static FunctionAST *ParseTopLevelExpr() {
854 if (ExprAST *E = ParseExpression()) {
855 // Make an anonymous proto.
856 PrototypeAST *Proto = new PrototypeAST("", std::vector<std::string>());
857 return new FunctionAST(Proto, E);
862 /// external ::= 'extern' prototype
863 static PrototypeAST *ParseExtern() {
864 getNextToken(); // eat extern.
865 return ParsePrototype();
868 //===----------------------------------------------------------------------===//
870 //===----------------------------------------------------------------------===//
872 static Module *TheModule;
873 static IRBuilder<> Builder(getGlobalContext());
874 static std::map<std::string, Value*> NamedValues;
875 static FunctionPassManager *TheFPM;
877 Value *ErrorV(const char *Str) { Error(Str); return 0; }
879 Value *NumberExprAST::Codegen() {
880 return ConstantFP::get(getGlobalContext(), APFloat(Val));
883 Value *VariableExprAST::Codegen() {
884 // Look this variable up in the function.
885 Value *V = NamedValues[Name];
886 return V ? V : ErrorV("Unknown variable name");
889 Value *BinaryExprAST::Codegen() {
890 Value *L = LHS->Codegen();
891 Value *R = RHS->Codegen();
892 if (L == 0 || R == 0) return 0;
895 case '+': return Builder.CreateFAdd(L, R, "addtmp");
896 case '-': return Builder.CreateFSub(L, R, "subtmp");
897 case '*': return Builder.CreateFMul(L, R, "multmp");
899 L = Builder.CreateFCmpULT(L, R, "cmptmp");
900 // Convert bool 0/1 to double 0.0 or 1.0
901 return Builder.CreateUIToFP(L, Type::getDoubleTy(getGlobalContext()),
903 default: return ErrorV("invalid binary operator");
907 Value *CallExprAST::Codegen() {
908 // Look up the name in the global module table.
909 Function *CalleeF = TheModule->getFunction(Callee);
911 return ErrorV("Unknown function referenced");
913 // If argument mismatch error.
914 if (CalleeF->arg_size() != Args.size())
915 return ErrorV("Incorrect # arguments passed");
917 std::vector<Value*> ArgsV;
918 for (unsigned i = 0, e = Args.size(); i != e; ++i) {
919 ArgsV.push_back(Args[i]->Codegen());
920 if (ArgsV.back() == 0) return 0;
923 return Builder.CreateCall(CalleeF, ArgsV, "calltmp");
926 Function *PrototypeAST::Codegen() {
927 // Make the function type: double(double,double) etc.
928 std::vector<Type*> Doubles(Args.size(),
929 Type::getDoubleTy(getGlobalContext()));
930 FunctionType *FT = FunctionType::get(Type::getDoubleTy(getGlobalContext()),
933 Function *F = Function::Create(FT, Function::ExternalLinkage, Name, TheModule);
935 // If F conflicted, there was already something named 'Name'. If it has a
936 // body, don't allow redefinition or reextern.
937 if (F->getName() != Name) {
938 // Delete the one we just made and get the existing one.
939 F->eraseFromParent();
940 F = TheModule->getFunction(Name);
942 // If F already has a body, reject this.
943 if (!F->empty()) {
944 ErrorF("redefinition of function");
948 // If F took a different number of args, reject.
949 if (F->arg_size() != Args.size()) {
950 ErrorF("redefinition of function with different # args");
955 // Set names for all arguments.
957 for (Function::arg_iterator AI = F->arg_begin(); Idx != Args.size();
959 AI->setName(Args[Idx]);
961 // Add arguments to variable symbol table.
962 NamedValues[Args[Idx]] = AI;
968 Function *FunctionAST::Codegen() {
971 Function *TheFunction = Proto->Codegen();
972 if (TheFunction == 0)
975 // Create a new basic block to start insertion into.
976 BasicBlock *BB = BasicBlock::Create(getGlobalContext(), "entry", TheFunction);
977 Builder.SetInsertPoint(BB);
979 if (Value *RetVal = Body->Codegen()) {
980 // Finish off the function.
981 Builder.CreateRet(RetVal);
983 // Validate the generated code, checking for consistency.
984 verifyFunction(*TheFunction);
986 // Optimize the function.
987 TheFPM->run(*TheFunction);
992 // Error reading body, remove function.
993 TheFunction->eraseFromParent();
997 //===----------------------------------------------------------------------===//
998 // Top-Level parsing and JIT Driver
999 //===----------------------------------------------------------------------===//
1001 static ExecutionEngine *TheExecutionEngine;
1003 static void HandleDefinition() {
1004 if (FunctionAST *F = ParseDefinition()) {
1005 if (Function *LF = F->Codegen()) {
1006 fprintf(stderr, "Read function definition:");
1010 // Skip token for error recovery.
1015 static void HandleExtern() {
1016 if (PrototypeAST *P = ParseExtern()) {
1017 if (Function *F = P->Codegen()) {
1018 fprintf(stderr, "Read extern: ");
1022 // Skip token for error recovery.
1027 static void HandleTopLevelExpression() {
1028 // Evaluate a top-level expression into an anonymous function.
1029 if (FunctionAST *F = ParseTopLevelExpr()) {
1030 if (Function *LF = F->Codegen()) {
1031 fprintf(stderr, "Read top-level expression:");
1034 // JIT the function, returning a function pointer.
1035 void *FPtr = TheExecutionEngine->getPointerToFunction(LF);
1037 // Cast it to the right type (takes no arguments, returns a double) so we
1038 // can call it as a native function.
1039 double (*FP)() = (double (*)())(intptr_t)FPtr;
1040 fprintf(stderr, "Evaluated to %f\n", FP());
1043 // Skip token for error recovery.
1048 /// top ::= definition | external | expression | ';'
1049 static void MainLoop() {
1051 fprintf(stderr, "ready> ");
1053 case tok_eof: return;
1054 case ';': getNextToken(); break; // ignore top-level semicolons.
1055 case tok_def: HandleDefinition(); break;
1056 case tok_extern: HandleExtern(); break;
1057 default: HandleTopLevelExpression(); break;
1062 //===----------------------------------------------------------------------===//
1063 // "Library" functions that can be "extern'd" from user code.
1064 //===----------------------------------------------------------------------===//
1066 /// putchard - putchar that takes a double and returns 0.
1068 double putchard(double X) {
1073 //===----------------------------------------------------------------------===//
1074 // Main driver code.
1075 //===----------------------------------------------------------------------===//
1078 InitializeNativeTarget();
1079 LLVMContext &Context = getGlobalContext();
1081 // Install standard binary operators.
1082 // 1 is lowest precedence.
1083 BinopPrecedence['<'] = 10;
1084 BinopPrecedence['+'] = 20;
1085 BinopPrecedence['-'] = 20;
1086 BinopPrecedence['*'] = 40; // highest.
1088 // Prime the first token.
1089 fprintf(stderr, "ready> ");
1092 // Make the module, which holds all the code.
1093 TheModule = new Module("my cool jit", Context);
1095 // Create the JIT. This takes ownership of the module.
1097 TheExecutionEngine = EngineBuilder(TheModule).setErrorStr(&ErrStr).create();
1098 if (!TheExecutionEngine) {
1099 fprintf(stderr, "Could not create ExecutionEngine: %s\n", ErrStr.c_str());
1103 FunctionPassManager OurFPM(TheModule);
1105 // Set up the optimizer pipeline. Start with registering info about how the
1106 // target lays out data structures.
1107 OurFPM.add(new TargetData(*TheExecutionEngine->getTargetData()));
1108 // Provide basic AliasAnalysis support for GVN.
1109 OurFPM.add(createBasicAliasAnalysisPass());
1110 // Do simple "peephole" optimizations and bit-twiddling optzns.
1111 OurFPM.add(createInstructionCombiningPass());
1112 // Reassociate expressions.
1113 OurFPM.add(createReassociatePass());
1114 // Eliminate Common SubExpressions.
1115 OurFPM.add(createGVNPass());
1116 // Simplify the control flow graph (deleting unreachable blocks, etc).
1117 OurFPM.add(createCFGSimplificationPass());
1119 OurFPM.doInitialization();
1121 // Set the global so the code gen can use this.
1122 TheFPM = &OurFPM;
1124 // Run the main "interpreter loop" now.
1129 // Print out all of the generated code.
1130 TheModule->dump();
1137 <a href="LangImpl5.html">Next: Extending the language: control flow</a>
1140 <!-- *********************************************************************** -->
1143 <a href="http://jigsaw.w3.org/css-validator/check/referer"><img
1144 src="http://jigsaw.w3.org/css-validator/images/vcss" alt="Valid CSS!"></a>
1145 <a href="http://validator.w3.org/check/referer"><img
1146 src="http://www.w3.org/Icons/valid-html401" alt="Valid HTML 4.01!"></a>
1148 <a href="mailto:sabre@nondot.org">Chris Lattner</a><br>
1149 <a href="http://llvm.org/">The LLVM Compiler Infrastructure</a><br>
1150 Last modified: $Date$