| Writing an LLVM Pass | -
Written by Chris Lattner
-
- +
| -Introduction - What is a pass? - |
The LLVM Pass Framework is an important part of the LLVM system, because LLVM passes are where the interesting parts of the compiler exist. Passes perform the transformations and optimizations that make up the compiler, they build the analysis results that are used by these transformations, and they are, above -all, a structuring technique for compiler code.
- -All LLVM passes are subclasses of the Pass class, which -implement functionality by overriding virtual methods inherited from -Pass. Depending on how your pass works, you may be able to inherit -from the FunctionPass +all, a structuring technique for compiler code.
+ +All LLVM passes are subclasses of the Pass +class, which implement functionality by overriding virtual methods inherited +from Pass. Depending on how your pass works, you may be able to +inherit from the FunctionPass or BasicBlockPass, +href="http://llvm.cs.uiuc.edu/doxygen/structllvm_1_1BasicBlockPass.html">BasicBlockPass, which gives the system more information about what your pass does, and how it can be combined with other passes. One of the main features of the LLVM Pass Framework is that it schedules passes to run in an efficient way based on the -constraints that your pass has.
+constraints that your pass has.
-We start by showing you how to construct a pass, everything from setting up the -code, to compiling, loading, and executing it. After the basics are down, more -advanced features are discussed.+
We start by showing you how to construct a pass, everything from setting up +the code, to compiling, loading, and executing it. After the basics are down, +more advanced features are discussed.
+| -Quick Start - Writing hello world - |
Here we describe how to write the "hello world" of passes. The "Hello" pass +is designed to simply print out the name of non-external functions that exist in the program being compiled. It does not modify the program at all, just inspects it. The source code and files for this pass are available in the LLVM -source tree in the lib/Transforms/Hello directory.
+source tree in the lib/Transforms/Hello directory.
+| - -Setting up the build environment - |
First thing you need to do is create a new directory somewhere in the LLVM source base. For this example, we'll assume that you made "lib/Transforms/Hello". The first thing you must do is set up a build script (Makefile) that will compile the source code for the new pass. To do -this, copy this into "Makefile":
+this, copy this into "Makefile":
+ ++# Makefile for hello pass # Path to top level of LLVM heirarchy @@ -147,95 +169,109 @@ SHARED_LIBRARY = 1 # Include the makefile implementation stuff include $(LEVEL)/Makefile.common -
+ -This makefile specifies that all of the .cpp files in the current +
This makefile specifies that all of the .cpp files in the current directory are to be compiled and linked together into a lib/Debug/libhello.so shared object that can be dynamically loaded by -the opt or analyze tools.
+the opt or analyze tools. If your operating system uses a +suffix other than .so (such as windows of Mac OS/X), the appropriate extension +will be used.
-Now that we have the build scripts set up, we just need to write the code for -the pass itself.+
Now that we have the build scripts set up, we just need to write the code for +the pass itself.
+| - -Basic code required - |
+
Now that we have a way to compile our new pass, we just have to write it. +Start out with:
#include "llvm/Pass.h" #include "llvm/Function.h"-Which are needed because we are writing a Pass, and we are -operating on Function's.
+
Which are needed because we are writing a Pass, and +we are operating on Function's.
-Next we have:+
Next we have:
++using namespace llvm; ++
... which is required because the functions from the include files +live in the llvm namespace. +
+ +Next we have:
namespace {
-+ -... which starts out an anonymous namespace. Anonymous namespaces are to C++ +
... which starts out an anonymous namespace. Anonymous namespaces are to C++ what the "static" keyword is to C (at global scope). It makes the things declared inside of the anonymous namespace only visible to the current file. If you're not familiar with them, consult a decent C++ book for more -information.
+information.
-Next, we declare our pass itself:+
Next, we declare our pass itself:
struct Hello : public FunctionPass {
-This declares a "Hello" class that is a subclass of FunctionPass. +
This declares a "Hello" class that is a subclass of FunctionPass. The different builtin pass subclasses are described in detail later, but for now, know that FunctionPass's -operate a function at a time.
+href="#passtype">later, but for now, know that FunctionPass's operate a function at a +time.
- virtual bool runOnFunction(Function &F) { + virtual bool runOnFunction(Function &F) { std::cerr << "Hello: " << F.getName() << "\n"; return false; } }; // end of struct Hello-We declare a "runOnFunction" method, which -overloads an abstract virtual method inherited from We declare a "runOnFunction" method, +which overloads an abstract virtual method inherited from FunctionPass. This is where we are supposed to do our thing, so we just print out our message with the name of each -function.
+function.
RegisterOpt<Hello> X("hello", "Hello World Pass");
} // end of anonymous namespace
-+ -Lastly, we register our class Hello, giving it a command line argument -"hello", and a name "Hello World Pass". There are several -different ways of registering your pass, depending -on what it is to be used for. For "optimizations" we use the -RegisterOpt template.
+
Lastly, we register our class Hello, giving it a command line +argument "hello", and a name "Hello World Pass". There are +several different ways of registering your pass, +depending on what it is to be used for. For "optimizations" we use the +RegisterOpt template.
-As a whole, the .cpp file looks like:+
As a whole, the .cpp file looks like:
#include "llvm/Pass.h" #include "llvm/Function.h" +using namespace llvm; + namespace { struct Hello : public FunctionPass { - virtual bool runOnFunction(Function &F) { + virtual bool runOnFunction(Function &F) { std::cerr << "Hello: " << F.getName() << "\n"; return false; } @@ -243,50 +279,51 @@ As a whole, the .cpp file looks like:RegisterOpt<Hello> X("hello", "Hello World Pass"); } -
+ -Now that it's all together, compile the file with a simple "gmake" +
Now that it's all together, compile the file with a simple "gmake" command in the local directory and you should get a new "lib/Debug/libhello.so file. Note that everything in this file is contained in an anonymous namespace: this reflects the fact that passes are self contained units that do not need external interfaces (although they can have -them) to be useful.
+them) to be useful.
+| - -Running a pass with opt or analyze - |
Now that you have a brand new shiny shared object file, we can use the opt command to run an LLVM program through your pass. Because you registered your pass with the RegisterOpt template, you will be able to -use the opt tool to access it, once loaded.
+use the opt tool to access it, once loaded.
-To test it, follow the example at the end of the To test it, follow the example at the end of the Getting Started Guide to compile "Hello World" to LLVM. We can now run the bytecode file (hello.bc) for the program through our transformation like this (or course, any bytecode file will -work):+work):
$ opt -load ../../../lib/Debug/libhello.so -hello < hello.bc > /dev/null Hello: __main Hello: puts Hello: main -
+
-The '-load' option specifies that 'opt' should load your pass
-as a shared object, which makes '-hello' a valid command line argument
-(which is one reason you need to register your
+ The '-load' option specifies that 'opt' should load your
+pass as a shared object, which makes '-hello' a valid command line
+argument (which is one reason you need to register your
pass). Because the hello pass does not modify the program in any
interesting way, we just throw away the result of opt (sending it to
-/dev/null).
+/dev/null).
+ To see what happened to the other string you registered, try running
+opt with the --help option:
+
-The pass name get added as the information string for your pass, giving some
+ The pass name get added as the information string for your pass, giving some
documentation to users of opt. Now that you have a working pass, you
would go ahead and make it do the cool transformations you want. Once you get
it all working and tested, it may become useful to find out how fast your pass
is. The PassManager provides a nice command
line option (--time-passes) that allows you to get information about
the execution time of your pass along with the other passes you queue up. For
-example:
+example:
+
-As you can see, our implementation above is pretty fast :). The additional
+ As you can see, our implementation above is pretty fast :). The additional
passes listed are automatically inserted by the 'opt' tool to verify
that the LLVM emitted by your pass is still valid and well formed LLVM, which
-hasn't been broken somehow.
-
-Now that you have seen the basics of the mechanics behind passes, we can talk
-about some more details of how they work and how to use them.
+hasn't been broken somehow. Now that you have seen the basics of the mechanics behind passes, we can talk
+about some more details of how they work and how to use them.
$ opt -load ../../../lib/Debug/libhello.so --help
@@ -301,20 +338,20 @@ OPTIONS:
-gcse - Global Common Subexpression Elimination
-globaldce - Dead Global Elimination
-hello - Hello World Pass
- -indvars - Cannonicalize Induction Variables
+ -indvars - Canonicalize Induction Variables
-inline - Function Integration/Inlining
-instcombine - Combine redundant instructions
...
-
$ opt -load ../../../lib/Debug/libhello.so -hello -time-passes < hello.bc > /dev/null
@@ -332,413 +369,604 @@ Hello: main
0.0000 ( 0.0%) 0.0000 ( 0.0%) 0.0000 ( 0.0%) 0.0013 ( 2.7%) Module Verifier
0.0000 ( 0.0%) 0.0000 ( 0.0%) 0.0000 ( 0.0%) 0.0033 ( 6.9%) Hello World Pass
0.0100 (100.0%) 0.0100 (100.0%) 0.0200 (100.0%) 0.0479 (100.0%) TOTAL
-
| -Pass classes and requirements - |
One of the first things that you should do when designing a new pass is to decide what class you should subclass for your pass. The Hello World example uses the FunctionPass class for its implementation, but we did not discuss why or when this should occur. Here we talk about the classes -available, from the most general to the most specific.
+available, from the most general to the most specific.
-When choosing a superclass for your Pass, you should choose the most +When choosing a superclass for your Pass, you should choose the most specific class possible, while still being able to meet the requirements -listed. This gives the LLVM Pass Infrastructure information neccesary to +listed. This gives the LLVM Pass Infrastructure information necessary to optimize how passes are run, so that the resultant compiler isn't unneccesarily -slow.
+slow.
+ +The most plain and boring type of pass is the "ImmutablePass" +class. This pass type is used for passes that do not have to be run, do not +change state, and never need to be updated. This is not a normal type of +transformation or analysis, but can provide information about the current +compiler configuration.
+Although this pass class is very infrequently used, it is important for +providing information about the current target machine being compiled for, and +other static information that can affect the various transformations.
+ +ImmutablePasses never invalidate other transformations, are never +invalidated, and are never "run".
+ +| - -The Pass class - |
The "Pass" class is the most general of all superclasses that you can use. Deriving from Pass indicates that your pass uses the entire program as a unit, refering to function bodies in no predictable order, or adding and removing functions. Because nothing is known about the behavior of direct Pass -subclasses, no optimization can be done for their execution.
+subclasses, no optimization can be done for their execution.
-To write a correct Pass subclass, derive from Pass and -overload the run method with the following signature:+
To write a correct Pass subclass, derive from Pass and +overload the run method with the following signature:
+ +virtual bool run(Module &M) = 0; -
+ -The run method performs the interesting work of the pass, and should +
The run method performs the interesting work of the pass, and should return true if the module was modified by the transformation, false -otherwise.
- +otherwise.
+| - -The FunctionPass class - |
In contrast to direct Pass subclasses, direct FunctionPass subclasses do have a predictable, local behavior that can be expected by the system. All FunctionPass execute on each function in the program -independant of all of the other functions in the program. +independent of all of the other functions in the program. FunctionPass's do not require that they are executed in a particular -order, and FunctionPass's do not modify external functions.
+order, and FunctionPass's do not modify external functions.
-To be explicit, FunctionPass subclasses are not allowed to:+
To be explicit, FunctionPass subclasses are not allowed to:
+ runOnFunction (including global data) + -Implementing a FunctionPass is usually straightforward (See the Implementing a FunctionPass is usually straightforward (See the Hello World pass for example). FunctionPass's may overload three virtual methods to do their work. All of these methods -should return true if they modified the program, or false if they didn't.
+should return true if they modified the program, or false if they didn't.
+ +virtual bool doInitialization(Module &M); -
+ -The doIninitialize method is allowed to do most of the things that +
The doIninitialize method is allowed to do most of the things that FunctionPass's are not allowed to do. They can add and remove functions, get pointers to functions, etc. The doInitialization method is designed to do simple initialization type of stuff that does not depend on the functions being processed. The doInitialization method call is not scheduled to overlap with any other pass executions (thus it should be very -fast).
+fast).
-A good example of how this method should be used is the A good example of how this method should be used is the LowerAllocations pass. This pass converts malloc and free instructions into -platform dependant malloc() and free() function calls. It +platform dependent malloc() and free() function calls. It uses the doInitialization method to get a reference to the malloc and -free functions that it needs, adding prototypes to the module if neccesary.+free functions that it needs, adding prototypes to the module if necessary.
+ +virtual bool runOnFunction(Function &F) = 0;
-The runOnFunction method must be implemented by your subclass to do the -transformation or analysis work of your pass. As usual, a true value should be -returned if the function is modified.
+
The runOnFunction method must be implemented by your subclass to do +the transformation or analysis work of your pass. As usual, a true value should +be returned if the function is modified.
+ +virtual bool doFinalization(Module &M); - +-The doFinalization method is an infrequently used method that is called -when the pass framework has finished calling The doFinalization method is an infrequently used method that is +called when the pass framework has finished calling runOnFunction for every function in the -program being compiled.
- +program being compiled.
+| - -The BasicBlockPass class - |
BasicBlockPass's are just like FunctionPass's, except that they must limit their scope of inspection and modification to a single basic block at a time. -As such, they are not allowed to do any of the following:
+As such, they are not allowed to do any of the following:
+ FunctionPasses. + -BasicBlockPass's are useful for traditional local and "peephole" +
BasicBlockPasses are useful for traditional local and "peephole" optimizations. They may override the same doInitialization(Module &) and doFinalization(Module &) methods that FunctionPass's have, but also have the following virtual methods that may also be implemented:
+href="#FunctionPass">FunctionPass
's have, but also have the following virtual methods that may also be implemented: + + -virtual bool doInitialization(Function &F); -
+ -The doIninitialize method is allowed to do most of the things that +
The doIninitialize method is allowed to do most of the things that BasicBlockPass's are not allowed to do, but that FunctionPass's can. The doInitialization method is designed to do simple initialization type of stuff that does not depend on the BasicBlocks being processed. The doInitialization method call is not scheduled to overlap with any other pass executions (thus it should be very -fast).
+fast).
+virtual bool runOnBasicBlock(BasicBlock &BB) = 0; -
+ -Override this function to do the work of the BasicBlockPass. This +
Override this function to do the work of the BasicBlockPass. This function is not allowed to inspect or modify basic blocks other than the parameter, and are not allowed to modify the CFG. A true value must be returned -if the basic block is modified.
+if the basic block is modified.
+virtual bool doFinalization(Function &F); - +-The doFinalization method is an infrequently used method that is called -when the pass framework has finished calling The doFinalization method is an infrequently used method that is +called when the pass framework has finished calling runOnBasicBlock for every BasicBlock in the program being compiled. This can be used to perform per-function -finalization.
+finalization.
+A MachineFunctionPass is a part of the LLVM code generator that +executes on the machine-dependent representation of each LLVM function in the +program. A MachineFunctionPass is also a FunctionPass, so all +the restrictions that apply to a FunctionPass also apply to it. +MachineFunctionPasses also have additional restrictions. In particular, +MachineFunctionPasses are not allowed to do any of the following:
+ ++ virtual bool runOnMachineFunction(MachineFunction &MF) = 0; ++ +
runOnMachineFunction can be considered the main entry point of a +MachineFunctionPass; that is, you should override this method to do the +work of your MachineFunctionPass.
+ +The runOnMachineFunction method is called on every +MachineFunction in a Module, so that the +MachineFunctionPass may perform optimizations on the machine-dependent +representation of the function. If you want to get at the LLVM Function +for the MachineFunction you're working on, use +MachineFunction's getFunction() accessor method -- but +remember, you may not modify the LLVM Function or its contents from a +MachineFunctionPass.
+ +| -Pass registration - |
+
In the Hello World example pass we illustrated how +pass registration works, and discussed some of the reasons that it is used and +what it does. Here we discuss how and why passes are registered.
-Passes can be registered in several different ways. Depending on the general +Passes can be registered in several different ways. Depending on the general classification of the pass, you should use one of the following templates to -register the pass:
+register the pass:
+'opt' utility.
- -
+'analyze' utility.
-
+file for more information. -Regardless of how you register your pass, you must specify at least two + + +
Regardless of how you register your pass, you must specify at least two parameters. The first parameter is the name of the pass that is to be used on the command line to specify that the pass should be added to a program (for example opt or analyze). The second argument is the name of the pass, which is to be used for the --help output of programs, as -well as for debug output generated by the --debug-pass option.
+well as for debug output generated by the --debug-pass option.
-If you pass is constructed by its default constructor, you only ever have to -pass these two arguments. If, on the other hand, you require other information -(like target specific information), you must pass an additional argument. This -argument is a pointer to a function used to create the pass. For an example of -how this works, look at the LowerAllocations.cpp -file.+
If a pass is registered to be used by the analyze utility, you +should implement the virtual print method:
-If a pass is registered to be used by the analyze utility, you should -implement the virtual print method:+
virtual void print(std::ostream &O, const Module *M) const; -
+ -The print method must be implemented by "analyses" in order to print a -human readable version of the analysis results. This is useful for debugging an -analysis itself, as well as for other people to figure out how an analysis -works. The analyze tool uses this method to generate its output.
+
The print method must be implemented by "analyses" in order to print +a human readable version of the analysis results. This is useful for debugging +an analysis itself, as well as for other people to figure out how an analysis +works. The analyze tool uses this method to generate its output.
-The ostream parameter specifies the stream to write the results on, and -the Module parameter gives a pointer to the top level module of the +The ostream parameter specifies the stream to write the results on, +and the Module parameter gives a pointer to the top level module of the program that has been analyzed. Note however that this pointer may be null in certain circumstances (such as calling the Pass::dump() from a debugger), so it should only be used to enhance debug output, it should not be -depended on.
+depended on.
+| -Specifying interactions between passes - |
One of the main responsibilities of the PassManager is the make sure that passes interact with each other correctly. Because PassManager tries to optimize the execution of passes it must know how the passes interact with each other and what dependencies exist between the various passes. To track this, each pass can declare the set of passes that are required to be executed before the current pass, and the passes which are -invalidated by the current pass.
+invalidated by the current pass.
-Typically this functionality is used to require that analysis results are +Typically this functionality is used to require that analysis results are computed before your pass is run. Running arbitrary transformation passes can invalidate the computed analysis results, which is what the invalidation set specifies. If a pass does not implement the getAnalysisUsage method, it defaults to not -having any prerequisite passes, and invalidating all other passes.
+having any prerequisite passes, and invalidating all other passes.
+virtual void getAnalysisUsage(AnalysisUsage &Info) const; -
+ -By implementing the getAnalysisUsage method, the required and +
By implementing the getAnalysisUsage method, the required and invalidated sets may be specified for your transformation. The implementation should fill in the AnalysisUsage -object with information about which passes are required and not invalidated. To do this, the following set methods are provided by the AnalysisUsage class:
+href="http://llvm.cs.uiuc.edu/doxygen/classllvm_1_1AnalysisUsage.html">AnalysisUsage +object with information about which passes are required and not invalidated. To +do this, a pass may call any of the following methods on the AnalysisUsage +object:
+- // addRequires - Add the specified pass to the required set for your pass. - template<class PassClass> - AnalysisUsage &AnalysisUsage::addRequired(); - - // addPreserved - Add the specified pass to the set of analyses preserved by - // this pass - template<class PassClass> - AnalysisUsage &AnalysisUsage::addPreserved(); - - // setPreservesAll - Call this if the pass does not modify its input at all - void AnalysisUsage::setPreservesAll(); - - // preservesCFG - This function should be called by the pass, iff they do not: - // - // 1. Add or remove basic blocks from the function - // 2. Modify terminator instructions in any way. - // - // This is automatically implied for BasicBlockPass's - // - void AnalysisUsage::preservesCFG(); -
+ +
+ ++If you pass requires a previous pass to be executed (an analysis for example), +it can use one of these methods to arrange for it to be run before your pass. +LLVM has many different types of analyses and passes that can be required, +spaning the range from DominatorSet to BreakCriticalEdges. +requiring BreakCriticalEdges, for example, guarantees that there will +be no critical edges in the CFG when your pass has been run. +
+ ++Some analyses chain to other analyses to do their job. For example, an AliasAnalysis implementation is required to chain to other alias analysis passes. In +cases where analyses chain, the addRequiredTransitive method should be +used instead of the addRequired method. This informs the PassManager +that the transitively required pass should be alive as long as the requiring +pass is. +
++One of the jobs of the PassManager is to optimize how and when analyses are run. +In particular, it attempts to avoid recomputing data unless it needs to. For +this reason, passes are allowed to declare that they preserve (i.e., they don't +invalidate) an existing analysis if it's available. For example, a simple +constant folding pass would not modify the CFG, so it can't possible effect the +results of dominator analysis. By default, all passes are assumed to invalidate +all others. +
+ ++The AnalysisUsage class provides several methods which are useful in +certain circumstances that are related to addPreserved. In particular, +the setPreservesAll method can be called to indicate that the pass does +not modify the LLVM program at all (which is true for analyses), and the +setPreservesCFG method can be used by transformations that change +instructions in the program but do not modify the CFG or terminator instructions +(note that this property is implicitly set for BasicBlockPass's). +
+ ++addPreserved is particularly useful for transformations like +BreakCriticalEdges. This pass knows how to update a small set of loop +and dominator related analyses if they exist, so it can preserve them, despite +the fact that it hacks on the CFG. +
++
// This is an example implementation from an analysis, which does not modify // the program at all, yet has a prerequisite. - void PostDominanceFrontier::getAnalysisUsage(AnalysisUsage &AU) const { + void PostDominanceFrontier::getAnalysisUsage(AnalysisUsage &AU) const { AU.setPreservesAll(); - AU.addRequired<PostDominatorTree>(); + AU.addRequired<PostDominatorTree>(); } -
+ -and:
+
and:
// This example modifies the program, but does not modify the CFG void LICM::getAnalysisUsage(AnalysisUsage &AU) const { - AU.preservesCFG(); - AU.addRequired<LoopInfo>(); + AU.setPreservesCFG(); + AU.addRequired<LoopInfo>(); } -
+ + +
+
The Pass::getAnalysis<> method is automatically inherited by +your class, providing you with access to the passes that you declared that you +required with the getAnalysisUsage +method. It takes a single template argument that specifies which pass class you +want, and returns a reference to that pass. For example:
- template<typename PassClass> - AnalysisType &getAnalysis(); -
+ bool LICM::runOnFunction(Function &F) { + LoopInfo &LI = getAnalysis<LoopInfo>(); + ... + } + + +
This method call returns a reference to the pass desired. You may get a +runtime assertion failure if you attempt to get an analysis that you did not +declare as required in your getAnalysisUsage implementation. This +method can be called by your run* method implementation, or by any +other local method invoked by your run* method.
+ ++If your pass is capable of updating analyses if they exist (e.g., +BreakCriticalEdges, as described above), you can use the +getAnalysisToUpdate method, which returns a pointer to the analysis if +it is active. For example:
+ +
+ ...
+ if (DominatorSet *DS = getAnalysisToUpdate<DominatorSet>()) {
+ // A DominatorSet is active. This code will update it.
+ }
+ ...
+
-This method call returns a reference to the pass desired. You may get a runtime
-assertion failure if you attempt to get an analysis that you did not declare as
-required in your getAnalysisUsage
-implementation. This method can be called by your run* method
-implementation, or by any other local method invoked by your run*
-method.+
| -Implementing Analysis Groups - |
Now that we understand the basics of how passes are defined, how the are +used, and how they are required from other passes, it's time to get a little bit fancier. All of the pass relationships that we have seen so far are very simple: one pass depends on one other specific pass to be run before it can run. For many applications, this is great, for others, more flexibility is -required.
+required.
-In particular, some analyses are defined such that there is a single simple +In particular, some analyses are defined such that there is a single simple interface to the analysis results, but multiple ways of calculating them. Consider alias analysis for example. The most trivial alias analysis returns "may alias" for any alias query. The most sophisticated analysis a @@ -746,69 +974,80 @@ flow-sensitive, context-sensitive interprocedural analysis that can take a significant amount of time to execute (and obviously, there is a lot of room between these two extremes for other implementations). To cleanly support situations like this, the LLVM Pass Infrastructure supports the notion of -Analysis Groups.
+Analysis Groups.
+ +An Analysis Group is a single simple interface that may be implemented by multiple different passes. Analysis Groups can be given human readable names just like passes, but unlike passes, they need not derive from the Pass class. An analysis group may have one or more implementations, one of which is -the "default" implementation.
+the "default" implementation.
-Analysis groups are used by client passes just like other passes are: the +Analysis groups are used by client passes just like other passes are: the AnalysisUsage::addRequired() and Pass::getAnalysis() methods. In order to resolve this requirement, the PassManager scans the available passes to see if any implementations of the analysis group are available. If none is available, the default implementation is created for the pass to use. All standard rules for interaction -between passes still apply.
+between passes still apply.
-Although Pass Registration is optional for normal +Although Pass Registration is optional for normal passes, all analysis group implementations must be registered, and must use the RegisterAnalysisGroup template to join the implementation pool. Also, a default implementation of the interface must be registered with RegisterAnalysisGroup.
+href="#registerag">RegisterAnalysisGroup.
-As a concrete example of an Analysis Group in action, consider the AliasAnalysis +As a concrete example of an Analysis Group in action, consider the AliasAnalysis analysis group. The default implementation of the alias analysis interface (the basicaa pass) just does a few simple checks that don't require significant analysis to compute (such as: two different globals can never alias each other, etc). Passes that use the AliasAnalysis +href="http://llvm.cs.uiuc.edu/doxygen/classllvm_1_1AliasAnalysis.html">AliasAnalysis interface (for example the gcse pass), do not care which implementation -of alias analysis is actually provided, they just use the designated -interface.
+href="http://llvm.cs.uiuc.edu/doxygen/structGCSE.html">gcse pass), do +not care which implementation of alias analysis is actually provided, they just +use the designated interface.
-From the user's perspective, commands work just like normal. Issuing the +From the user's perspective, commands work just like normal. Issuing the command 'opt -gcse ...' will cause the basicaa class to be instantiated and added to the pass sequence. Issuing the command 'opt -somefancyaa -gcse ...' will cause the gcse pass to use the somefancyaa alias analysis (which doesn't actually exist, it's just a -hypothetical example) instead.
+hypothetical example) instead.
+The RegisterAnalysisGroup template is used to register the analysis group itself as well as add pass implementations to the analysis group. First, an analysis should be registered, with a human readable name provided for it. Unlike registration of passes, there is no command line argument to be specified -for the Analysis Group Interface itself, because it is "abstract":
+for the Analysis Group Interface itself, because it is "abstract":
- static RegisterAnalysisGroup<AliasAnalysis> A("Alias Analysis"); -
+ static RegisterAnalysisGroup<AliasAnalysis> A("Alias Analysis"); + -Once the analysis is registered, passes can declare that they are valid -implementations of the interface by using the following code:
+
Once the analysis is registered, passes can declare that they are valid +implementations of the interface by using the following code:
namespace {
@@ -817,16 +1056,16 @@ implementations of the interface by using the following code:
B("somefancyaa", "A more complex alias analysis implementation");
// Declare that we implement the AliasAnalysis interface
- RegisterAnalysisGroup<AliasAnalysis, FancyAA> C;
+ RegisterAnalysisGroup<AliasAnalysis, FancyAA> C;
}
-
+ -This just shows a class FancyAA that is registered normally, then uses -the RegisterAnalysisGroup template to "join" the AliasAnalysis +
This just shows a class FancyAA that is registered normally, then +uses the RegisterAnalysisGroup template to "join" the AliasAnalysis analysis group. Every implementation of an analysis group should join using this template. A single pass may join multiple different analysis groups with -no problem.
+no problem.
namespace {
@@ -835,35 +1074,38 @@ no problem.
D("basicaa", "Basic Alias Analysis (default AA impl)");
// Declare that we implement the AliasAnalysis interface
- RegisterAnalysisGroup<AliasAnalysis, BasicAliasAnalysis, true> E;
+ RegisterAnalysisGroup<AliasAnalysis, BasicAliasAnalysis, true> E;
}
-
+ -Here we show how the default implementation is specified (using the extra +
Here we show how the default implementation is specified (using the extra argument to the RegisterAnalysisGroup template). There must be exactly one default implementation available at all times for an Analysis Group to be used. Here we declare that the BasicAliasAnalysis -pass is the default implementation for the interface.
+pass is the default implementation for the interface.
+| -What PassManager does - |
The PassManager -class takes -a list of passes, ensures their prerequisites are set -up correctly, and then schedules passes to run efficiently. All of the LLVM -tools that run passes use the PassManager for execution of these -passes.
+class +takes a list of passes, ensures their prerequisites +are set up correctly, and then schedules passes to run efficiently. All of the +LLVM tools that run passes use the PassManager for execution of these +passes.
-The PassManager does two main things to try to reduce the execution -time of a series of passes:+
The PassManager does two main things to try to reduce the execution +time of a series of passes:
+results as soon as they are no longer needed.
+href="#FunctionPass">FunctionPass
This improves the cache behavior of the compiler, because it is only touching the LLVM program representation for a single function at a time, instead of traversing the entire program. It reduces the memory consumption of compiler, because, for example, only one DominatorSet +href="http://llvm.cs.uiuc.edu/doxygen/structllvm_1_1DominatorSet.html">DominatorSet needs to be calculated at a time. This also makes it possible some interesting enhancements in the future.
+href="#SMP">interesting enhancements in the future.
-+ -The effectiveness of the PassManager is influenced directly by how much -information it has about the behaviors of the passes it is scheduling. For +
The effectiveness of the PassManager is influenced directly by how +much information it has about the behaviors of the passes it is scheduling. For example, the "preserved" set is intentionally conservative in the face of an unimplemented getAnalysisUsage method. Not implementing when it should be implemented will have the effect of not -allowing any analysis results to live across the execution of your pass.
+allowing any analysis results to live across the execution of your pass.
-The PassManager class exposes a --debug-pass command line +The PassManager class exposes a --debug-pass command line options that is useful for debugging pass execution, seeing how things work, and diagnosing when you should be preserving more analyses than you currently are (To get information about all of the variants of the --debug-pass -option, just type 'opt --help-hidden').
+option, just type 'opt --help-hidden').
-By using the --debug-pass=Structure option, for example, we can see how -our Hello World pass interacts with other passes. Lets -try it out with the gcse and licm passes:+
By using the --debug-pass=Structure option, for example, we can see +how our Hello World pass interacts with other passes. +Lets try it out with the gcse and licm passes:
$ opt -load ../../../lib/Debug/libhello.so -gcse -licm --debug-pass=Structure < hello.bc > /dev/null @@ -928,24 +1170,24 @@ Module Pass Manager -- Module Verifier Bytecode Writer --Bytecode Writer -
+ -This output shows us when passes are constructed and when the analysis results -are known to be dead (prefixed with '--'). Here we see that GCSE uses -dominator and immediate dominator information to do its job. The LICM pass uses -natural loop information, which uses dominator sets, but not immediate +
This output shows us when passes are constructed and when the analysis +results are known to be dead (prefixed with '--'). Here we see that +GCSE uses dominator and immediate dominator information to do its job. The LICM +pass uses natural loop information, which uses dominator sets, but not immediate dominators. Because immediate dominators are no longer useful after the GCSE pass, it is immediately destroyed. The dominator sets are then reused to -compute natural loop information, which is then used by the LICM pass.
+compute natural loop information, which is then used by the LICM pass.
-After the LICM pass, the module verifier runs (which is automatically added by -the 'opt' tool), which uses the dominator set to check that the +After the LICM pass, the module verifier runs (which is automatically added +by the 'opt' tool), which uses the dominator set to check that the resultant LLVM code is well formed. After it finishes, the dominator set information is destroyed, after being computed once, and shared by three -passes.
+passes.
-Lets see how this changes when we run the Hello World -pass in between the two passes:+
Lets see how this changes when we run the Hello +World pass in between the two passes:
$ opt -load ../../../lib/Debug/libhello.so -gcse -hello -licm --debug-pass=Structure < hello.bc > /dev/null @@ -972,24 +1214,24 @@ Module Pass Manager Hello: __main Hello: puts Hello: main -
+ -Here we see that the Hello World pass has killed the +
Here we see that the Hello World pass has killed the Dominator Set pass, even though it doesn't modify the code at all! To fix this, we need to add the following getAnalysisUsage method to our pass:
+href="#getAnalysisUsage">getAnalysisUsage method to our pass:
// We don't modify the program, so we preserve all analyses
virtual void getAnalysisUsage(AnalysisUsage &AU) const {
AU.setPreservesAll();
}
-+ -Now when we run our pass, we get this output:
+
Now when we run our pass, we get this output:
-$ opt -load ../../../lib/Debug/libhello.so -gcse -hello -licm --debug-pass=Structure < hello.bc > /dev/null +$ opt -load ../../../lib/Debug/libhello.so -gcse -hello -licm --debug-pass=Structure < hello.bc > /dev/null Pass Arguments: -gcse -hello -licm Module Pass Manager Function Pass Manager @@ -1012,54 +1254,67 @@ Module Pass Manager Hello: __main Hello: puts Hello: main -
+ -Which shows that we don't accidentally invalidate dominator information -anymore, and therefore do not have to compute it twice.
+
Which shows that we don't accidentally invalidate dominator information +anymore, and therefore do not have to compute it twice.
+ -virtual void releaseMemory(); -
+ -The PassManager automatically determines when to compute analysis +
The PassManager automatically determines when to compute analysis results, and how long to keep them around for. Because the lifetime of the pass object itself is effectively the entire duration of the compilation process, we need some way to free analysis results when they are no longer useful. The -releaseMemory virtual method is the way to do this.
+releaseMemory virtual method is the way to do this.
-If you are writing an analysis or any other pass that retains a significant +If you are writing an analysis or any other pass that retains a significant amount of state (for use by another pass which "requires" your pass and uses the getAnalysis method) you should implement releaseMEmory to, well, release the memory allocated to maintain this internal state. This method is called after the run* method for the -class, before the next call of run* in your pass.
+class, before the next call of run* in your pass.
+| -Using GDB with dynamically loaded passes - |
Unfortunately, using GDB with dynamically loaded passes is not as easy as it should be. First of all, you can't set a breakpoint in a shared object that has not been loaded yet, and second of all there are problems with inlined functions in shared objects. Here are some suggestions to debugging your pass with -GDB.
+GDB.
-For sake of discussion, I'm going to assume that you are debugging a +For sake of discussion, I'm going to assume that you are debugging a transformation invoked by opt, although nothing described here depends -on that.
+on that.
+ ++
First thing you do is start gdb on the opt process:
$ gdb opt @@ -1071,44 +1326,48 @@ Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "sparc-sun-solaris2.6"... (gdb) -
+ -Note that opt has a lot of debugging information in it, so it takes +
Note that opt has a lot of debugging information in it, so it takes time to load. Be patient. Since we cannot set a breakpoint in our pass yet (the shared object isn't loaded until runtime), we must execute the process, and have it stop before it invokes our pass, but after it has loaded the shared object. The most foolproof way of doing this is to set a breakpoint in PassManager::run and then run the process with the arguments you -want:
+want:
(gdb) break PassManager::run
Breakpoint 1 at 0x2413bc: file Pass.cpp, line 70.
-(gdb) run test.bc -load /shared/lattner/cvs/llvm/lib/Debug/[libname].so -[passoption]
-Starting program: /shared/lattner/cvs/llvm/tools/Debug/opt test.bc
- -load /shared/lattner/cvs/llvm/lib/Debug/[libname].so -[passoption]
+(gdb) run test.bc -load $(LLVMTOP)/llvm/lib/Debug/[libname].so -[passoption]
+Starting program: opt test.bc -load $(LLVMTOP)/llvm/lib/Debug/[libname].so -[passoption]
Breakpoint 1, PassManager::run (this=0xffbef174, M=@0x70b298) at Pass.cpp:70
70 bool PassManager::run(Module &M) { return PM->run(M); }
(gdb)
-
+
-Once the opt stops in the PassManager::run method you are now
-free to set breakpoints in your pass so that you can trace through execution or
-do other standard debugging stuff.+
Once the opt stops in the PassManager::run method you are +now free to set breakpoints in your pass so that you can trace through execution +or do other standard debugging stuff.
++
Once you have the basics down, there are a couple of problems that GDB has, +some with solutions, some without.
+from the body of a class to a .cpp file).
+execution stops in PassManager::run.
+
Hopefully these tips will help with common case debugging situations. If +you'd like to contribute some tips of your own, just contact Chris.
+| -Future extensions planned - |
Although the LLVM Pass Infrastructure is very capable as it stands, and does some nifty stuff, there are things we'd like to add in the future. Here is -where we are going:
+where we are going:
+ +Multiple CPU machines are becoming more common and compilation can never be fast enough: obviously we should allow for a multithreaded compiler. Because of the semantics defined for passes above (specifically they cannot maintain state across invocations of their run* methods), a nice clean way to implement a multithreaded compiler would be for the PassManager class -to create multiple instances of each pass object, and allow the seperate -instances to be hacking on different parts of the program at the same time.
+to create multiple instances of each pass object, and allow the separate +instances to be hacking on different parts of the program at the same time.
-This implementation would prevent each of the passes from having to implement +This implementation would prevent each of the passes from having to implement multithreaded constructs, requiring only the LLVM core to have locking in a few places (for global resources). Although this is a simple extension, we simply haven't had time (or multiprocessor machines, thus a reason) to implement this. -Despite that, we have kept the LLVM passes SMP ready, and you should too.
+Despite that, we have kept the LLVM passes SMP ready, and you should too.
+Currently, the PassManager's run method takes a Module +as input, and runs all of the passes on this module. The problem with this approach is that none of the PassManager features can be used for timing and debugging the actual loading of the module from disk or -standard input.
+standard input.
-To solve this problem, eventually the PassManger class will accept a -ModuleSource object instead of a Module itself. When complete, this +To solve this problem, eventually the PassManager class will accept +a ModuleSource object instead of a Module itself. When complete, this will also allow for streaming of functions out of the bytecode representation, allowing us to avoid holding the entire program in memory at once if we only are -dealing with FunctionPass's.
+dealing with FunctionPasses.
-As part of a different issue, eventually the bytecode loader will be extended to -allow on-demand loading of functions from the bytecode representation, in order -to better support the runtime reoptimizer. The bytecode format is already -capable of this, the loader just needs to be reworked a bit.+
As part of a different issue, eventually the bytecode loader will be extended +to allow on-demand loading of functions from the bytecode representation, in +order to better support the runtime reoptimizer. The bytecode format is already +capable of this, the loader just needs to be reworked a bit.
+- -Note that it is no problem for a +Passes requiring FunctionPasses + + +
Currently it is illegal for a Pass to require a +FunctionPass. This is because there is +only one instance of the FunctionPass +object ever created, thus nowhere to store information for all of the functions +in the program at the same time. Although this has come up a couple of times +before, this has always been worked around by factoring one big complicated pass +into a global and an interprocedural part, both of which are distinct. In the +future, it would be nice to have this though.
+ +Note that it is no problem for a FunctionPass to require the results of a Pass, only the other way around.
+href="#Pass">Pass, only the other way around.
+
+