20 KiB

Raw Blame History

Writing a Pass

[TOC]

Passes represent the basic infrastructure for transformation and optimization. This document provides a quickstart to the pass infrastructure in MLIR and how to use it.

See MLIR specification for more information about MLIR and its core aspects, such as the IR structure and operations.

See MLIR Rewrites for a quick start on graph rewriting in MLIR. If your transformation involves pattern matching operation DAGs, this is a great place to start.

Pass Types

MLIR provides different pass classes for several different granularities of transformation. Depending on the granularity of the transformation being performed, a pass may derive from FunctionPass or ModulePass; with each requiring a different set of constraints.

FunctionPass

A function pass operates on a per-function granularity, executing on non-external functions within a module in no particular order. Function passes have the following restrictions, and any noncompliance will lead to problematic behavior in multithreaded and other advanced scenarios:

Modify anything within the parent module, outside of the current function being operated on. This includes adding or removing functions from the module.
Maintain pass state across invocations of runOnFunction. A pass may be run on several different functions with no guarantee of execution order.
- When multithreading, a specific pass instance may not even execute on all functions within the module. As such, a function pass should not rely on running on all functions within a module.
Access, or modify, the state of another function within the module.
- Other threads may be operating on different functions within the module.
Maintain any global mutable state, e.g. static variables within the source file. All mutable state should be maintained by an instance of the pass.

To create a function pass, a derived class must adhere to the following:

Inherit from the CRTP class FunctionPass.
Override the virtual void runOnFunction() method.
Must be copy-constructible, multiple instances of the pass may be created by the pass manager to process functions in parallel.

A simple function pass may look like:

namespace {
struct MyFunctionPass : public FunctionPass<MyFunctionPass> {
  void runOnFunction() override {
    // Get the current function being operated on.
    Function *f = getFunction();

    // Operate on the operations within the function.
    f->walk([](Operation *inst) {
      ....
    });
  }
};
} // end anonymous namespace

// Register this pass to make it accessible to utilities like mlir-opt.
// (Pass registration is discussed more below)
static PassRegistration<MyFunctionPass> pass(
    "flag-name-to-invoke-pass-via-mlir-opt", "Pass description here");

ModulePass

A module pass operates on a per-module granularity, executing on the entire program as a unit. As such, module passes are able to add/remove/modify any functions freely. Module passes have the following restrictions, and any noncompliance will lead to problematic behavior in multithreaded and other advanced scenarios:

Maintain pass state across invocations of runOnModule.
Maintain any global mutable state, e.g. static variables within the source file. All mutable state should be maintained by an instance of the pass.

To create a module pass, a derived class must adhere to the following:

Inherit from the CRTP class ModulePass.
Override the virtual void runOnModule() method.

A simple module pass may look like:

namespace {
struct MyModulePass : public ModulePass<MyModulePass> {
  void runOnModule() override {
    // Get the current module being operated on.
    Module *m = getModule();

    // Operate on the functions within the module.
    for (auto &func : *m) {
      ....
    }
  }
};
} // end anonymous namespace

// Register this pass to make it accessible to utilities like mlir-opt.
// (Pass registration is discussed more below)
static PassRegistration<MyModulePass> pass(
    "flag-name-to-invoke-pass-via-mlir-opt", "Pass description here");

Analysis Management

An important concept, along with transformation passes, are analyses. These are conceptually similar to transformation passes, except that they compute information on a specific Function, or Module, without modifying it. In MLIR, analyses are not passes but free standing classes that are computed lazily on-demand and cached to avoid unnecessary recomputation. An analysis in MLIR must adhere to the following:

Provide a valid constructor for a given IR unit.
Must not modify the given IR unit.

Each of the base Pass classes provide utilities for querying and preserving analyses for the current IR being processed. Using the example passes defined above, let's see some examples:

Querying Analyses

FunctionPass automatically provides the following utilities for querying analyses:
- getAnalysis<>
- getCachedAnalysis<>
- getCachedModuleAnalysis<>
ModulePass automatically provides the following utilities:
- getAnalysis<>
- getCachedAnalysis<>
- getFunctionAnalysis<>
- getCachedFunctionAnalysis<>

/// An interesting function analysis.
struct MyFunctionAnalysis {
  // Compute this analysis with the provided function.
  MyFunctionAnalysis(Function *function);
};

/// An interesting module analysis.
struct MyModuleAnalysis {
  // Compute this analysis with the provided module.
  MyModuleAnalysis(Module *module);
};

void MyFunctionPass::runOnFunction() {
  // Query MyFunctionAnalysis for the current function.
  MyFunctionAnalysis &myAnalysis = getAnalysis<MyFunctionAnalysis>();

  // Query a cached instance of MyFunctionAnalysis for the current function. It
  // will not be computed if it doesn't exist.
  auto optionalAnalysis = getCachedAnalysis<MyFunctionAnalysis>();
  if (optionalAnalysis)
    ...

  // Query a cached instance of MyModuleAnalysis for the parent module of the
  // current function. It will not be computed if it doesn't exist.
  auto optionalAnalysis = getCachedModuleAnalysis<MyModuleAnalysis>();
  if (optionalAnalysis)
    ...
}

void MyModulePass::runOnModule() {
  // Query MyModuleAnalysis for the current module.
  MyModuleAnalysis &myAnalysis = getAnalysis<MyModuleAnalysis>();

  // Query a cached instance of MyModuleAnalysis for the current module. It
  // will not be computed if it doesn't exist.
  auto optionalAnalysis = getCachedAnalysis<MyModuleAnalysis>();
  if (optionalAnalysis)
    ...

  // Query MyFunctionAnalysis for a child function of the current module. It
  // will be computed if it doesn't exist.
  auto *fn = &*getModule().begin();
  MyFunctionAnalysis &myAnalysis = getFunctionAnalysis<MyFunctionAnalysis>(fn);
}

Preserving Analyses

Analyses that are constructed after being queried by a pass are cached to avoid unnecessary computation if they are requested again later. To avoid stale analyses, all analyses are assumed to be invalidated by a pass. To avoid invalidation, a pass must specifically mark analyses that are known to be preserved.

All Pass classes automatically provide the following utilities for preserving analyses:
- markAllAnalysesPreserved
- markAnalysesPreserved<>

void MyPass::runOn*() {
  // Mark all analyses as preserved. This is useful if a pass can guarantee
  // that no transformation was performed.
  markAllAnalysesPreserved();

  // Mark specific analyses as preserved. This is used if some transformation
  // was performed, but some analyses were either unaffected or explicitly
  // preserved.
  markAnalysesPreserved<MyAnalysis, MyAnalyses...>();
}

Pass Failure

Passes in MLIR are allowed to gracefully fail. This may happen if some invariant of the pass was broken, potentially leaving the IR in some invalid state. If such a situation occurs, the pass can directly signal a failure to the pass manager. If a pass signaled a failure when executing, no other passes in the pipeline will execute and the PassManager::run will return failure. Failure signaling is provided in the form of a signalPassFailure method.

void MyPass::runOn*() {
  // Signal failure on a broken invariant.
  if (some_broken_invariant) {
    signalPassFailure();
    return;
  }
}

Pass Manager

Above we introduced the different types of passes and their constraints. Now that we have our pass we need to be able to run it over a specific module. This is where the pass manager comes into play. The PassManager is the interface used for scheduling passes to run over a module. The pass manager provides simple interfaces for adding passes, of any kind, and running them over a given module. A simple example is shown below:

PassManager pm;

// Add a module pass.
pm.addPass(new MyModulePass());

// Add a few function passes.
pm.addPass(new MyFunctionPass());
pm.addPass(new MyFunctionPass2());
pm.addPass(new MyFunctionPass3());

// Add another module pass.
pm.addPass(new MyModulePass2());

// Run the pass manager on a module.
Module *m = ...;
if (failed(pm.run(m)))
    ... // One of the passes signaled a failure.

The pass manager automatically structures added passes into nested pipelines on specific IR units. These pipelines are then run over a single IR unit at a time. This means that, for example, given a series of consecutive function passes, it will execute all on the first function, then all on the second function, etc. until the entire program has been run through the passes. This provides several benefits:

This improves the cache behavior of the compiler, because it is only touching a single function at a time, instead of traversing the entire program.
This improves multi-threading performance by reducing the number of jobs that need to be scheduled, as well as increasing the efficency of each job. An entire function pipeline can be run on each function asynchronously.

As an example, the above pass manager would contain the following pipeline structure:

MyModulePass
Function Pipeline
   MyFunctionPass
   MyFunctionPass2
   MyFunctionPass3
MyModulePass2

Pass Registration

Briefly shown in the example definitions of the various pass types is the PassRegistration class. This is a utility to register derived pass classes so that they may be created, and inspected, by utilities like mlir-opt. Registering a pass class takes the form:

static PassRegistration<MyPass> pass("command-line-arg", "description");

MyPass is the name of the derived pass class.
"command-line-arg" is the argument to use on the command line to invoke the pass from mlir-opt.
"description" is a description of the pass.

Pass Pipeline Registration

Described above is the mechanism used for registering a specific derived pass class. On top of that, MLIR allows for registering custom pass pipelines in a similar fashion. This allows for custom pipelines to be available to tools like mlir-opt in the same way that passes are, which is useful for encapsulating common pipelines like the "-O1" series of passes. Pipelines are registered via a similar mechanism to passes in the form of PassPipelineRegistration. Compared to PassRegistration, this class takes an additional parameter in the form of a pipeline builder that modifies a provided PassManager.

void pipelineBuilder(PassManager &pm) {
  pm.addPass(new MyPass());
  pm.addPass(new MyOtherPass());
}

// Register an existing pipeline builder function.
static PassPipelineRegistration pipeline(
  "command-line-arg", "description", pipelineBuilder);

// Register an inline pipeline builder.
static PassPipelineRegistration pipeline(
  "command-line-arg", "description", [](PassManager &pm) {
    pm.addPass(new MyPass());
    pm.addPass(new MyOtherPass());
  });

Pipeline registration also allows for simplified registration of specifializations for existing passes:

static PassPipelineRegistration foo10(
    "foo-10", "Foo Pass 10", [] { return new FooPass(10); } );

Pass Instrumentation

MLIR provides a customizable framework to instrument pass execution and analysis computation. This is provided via the PassInstrumentation class. This class provides hooks into the PassManager that observe various pass events:

runBeforePass
- This callback is run just before a pass is executed.
runAfterPass
- This callback is run right after a pass has been successfully executed. If this hook is executed, runAfterPassFailed will not be.
runAfterPassFailed
- This callback is run right after a pass execution fails. If this hook is executed, runAfterPass will not be.
runBeforeAnalysis
- This callback is run just before an analysis is computed.
runAfterAnalysis
- This callback is run right after an analysis is computed.

PassInstrumentation objects can be registered directly with a PassManager instance via the addInstrumentation method. Instrumentations added to the PassManager are run in a stack like fashion, i.e. the last instrumentation to execute a runBefore* hook will be the first to execute the respective runAfter* hook. Below in an example instrumentation that counts the number of times DominanceInfo is computed:

struct DominanceCounterInstrumentation : public PassInstrumentation {
  unsigned &count;

  DominanceCounterInstrumentation(unsigned &count) : count(count) {}
  void runAfterAnalysis(llvm::StringRef, AnalysisID *id,
                        const llvm::Any &) override {
    if (id == AnalysisID::getID<DominanceInfo>())
      ++count;
  }
};

PassManager pm;

// Add the instrumentation to the pass manager.
unsigned domInfoCount;
pm.addInstrumentation(new DominanceCounterInstrumentation(domInfoCount));

// Run the pass manager on a module.
Module *m = ...;
if (failed(pm.run(m)))
    ...

llvm::errs() << "DominanceInfo was computed " << domInfoCount << " times!\n";

Standard Instrumentations

MLIR utilizes the pass instrumentation framework to provide a few useful developer tools and utilites. Each of these instrumentations are immediately available to all users of the MLIR pass framework.

Pass Timing

The PassTiming instrumentation provides timing information about the execution of passes and computation of analyses. This provides a quick glimpse into what passes are taking the most time to execute, as well as how much of an effect your pass has on the total execution time of the pipeline. Users can enable this instrumentation directly on the PassManager via enableTiming. This instrumentation is also made available in mlir-opt via the -pass-timing flag. The PassTiming instrumentation provides several different display modes for the timing results, each of which is described below:

List Display Mode

In this mode, the results are displayed in a list sorted by total time with each pass/analysis instance aggregated into one unique result. This view is useful for getting an overview of what analyses/passes are taking the most time in a pipeline. This display mode is available in mlir-opt via -pass-timing-display=list.

$ mlir-opt foo.mlir -cse -canonicalize -convert-to-llvmir -pass-timing -pass-timing-display=list

===-------------------------------------------------------------------------===
                      ... Pass execution timing report ...
===-------------------------------------------------------------------------===
  Total Execution Time: 0.0203 seconds

   ---Wall Time---  --- Name ---
   0.0062 ( 30.5%)  Canonicalizer
   0.0053 ( 25.9%)  LLVMLowering
   0.0036 ( 17.8%)  ModuleVerifier
   0.0036 ( 17.7%)  FunctionVerifier
   0.0017 (  8.1%)  CSE
   0.0007 (  3.3%)  (A) DominanceInfo
   0.0203 (100.0%)  Total

Pipeline Display Mode

In this mode, the results are displayed in a nested pipeline view that mirrors the internal pass pipeline that is being executed in the pass manager. This view is useful for understanding specifically which parts of the pipeline are taking the most time, and can also be used to identify when analyses are being invalidated and recomputed. This is the default display mode.

$ mlir-opt foo.mlir -cse -canonicalize -convert-to-llvmir -pass-timing

===-------------------------------------------------------------------------===
                      ... Pass execution timing report ...
===-------------------------------------------------------------------------===
  Total Execution Time: 0.0249 seconds

   ---Wall Time---  --- Name ---
   0.0140 ( 56.1%)  Function Pipeline
   0.0020 (  8.0%)    CSE
   0.0008 (  3.2%)      (A) DominanceInfo
   0.0022 (  8.7%)    FunctionVerifier
   0.0076 ( 30.5%)    Canonicalizer
   0.0022 (  8.8%)    FunctionVerifier
   0.0022 (  9.0%)  ModuleVerifier
   0.0065 ( 25.9%)  LLVMLowering
   0.0022 (  9.0%)  ModuleVerifier
   0.0249 (100.0%)  Total

Multi-threaded Pass Timing

When multi-threading is enabled in the pass manager the meaning of the display slightly changes. First, a new timing column is added, User Time, that displays the total time spent across all threads. Secondly, the Wall Time column displays the longest individual time spent amongst all of the threads. This means that the Wall Time column will continue to give an indicator on the perceived time, or clock time, whereas the User Time will display the total cpu time.

$ mlir-opt foo.mlir -experimental-mt-pm -cse -canonicalize -convert-to-llvmir -pass-timing

===-------------------------------------------------------------------------===
                      ... Pass execution timing report ...
===-------------------------------------------------------------------------===
  Total Execution Time: 0.0078 seconds

   ---User Time---   ---Wall Time---  --- Name ---
   0.0175 ( 88.3%)     0.0055 ( 70.4%)  Function Pipeline
   0.0018 (  9.3%)     0.0006 (  8.1%)    CSE
   0.0013 (  6.3%)     0.0004 (  5.8%)      (A) DominanceInfo
   0.0017 (  8.7%)     0.0006 (  7.1%)    FunctionVerifier
   0.0128 ( 64.6%)     0.0039 ( 50.5%)    Canonicalizer
   0.0011 (  5.7%)     0.0004 (  4.7%)    FunctionVerifier
   0.0004 (  2.1%)     0.0004 (  5.2%)  ModuleVerifier
   0.0010 (  5.3%)     0.0010 ( 13.4%)  LLVMLowering
   0.0009 (  4.3%)     0.0009 ( 11.0%)  ModuleVerifier
   0.0198 (100.0%)     0.0078 (100.0%)  Total

IR Printing

When debugging it is often useful to dump the IR at various stages of a pass pipeline. This is where the IR printing instrumentation comes into play. This instrumentation allows for conditionally printing the IR before and after pass execution by optionally filtering on the pass being executed. This instrumentation can be added directly to the PassManager via the enableIRPrinting method. mlir-opt provides a few useful flags for utilizing this instrumentation:

print-ir-before=(comma-separated-pass-list)
- Print the IR before each of the passes provided within the pass list.
print-ir-before-all
- Print the IR before every pass in the pipeline.

$ mlir-opt foo.mlir -cse -print-ir-before=cse

*** IR Dump Before CSE ***
func @simple_constant() -> (i32, i32) {
  %c1_i32 = constant 1 : i32
  %c1_i32_0 = constant 1 : i32
  return %c1_i32, %c1_i32_0 : i32, i32
}

print-ir-after=(comma-separated-pass-list)
- Print the IR after each of the passes provided within the pass list.
print-ir-after-all
- Print the IR after every pass in the pipeline.

$ mlir-opt foo.mlir -cse -print-ir-after=cse

*** IR Dump After CSE ***
func @simple_constant() -> (i32, i32) {
  %c1_i32 = constant 1 : i32
  return %c1_i32, %c1_i32 : i32, i32
}

print-ir-module-scope
- Always print the Module IR, even for non module passes.

$ mlir-opt foo.mlir -cse -print-ir-after=cse -print-ir-module-scope

*** IR Dump After CSE ***  (function: bar)
func @bar(%arg0: f32, %arg1: f32) -> f32 {
  ...
}

func @simple_constant() -> (i32, i32) {
  %c1_i32 = constant 1 : i32
  %c1_i32_0 = constant 1 : i32
  return %c1_i32, %c1_i32_0 : i32, i32
}

*** IR Dump After CSE ***  (function: simple_constant)
func @bar(%arg0: f32, %arg1: f32) -> f32 {
  ...
}

func @simple_constant() -> (i32, i32) {
  %c1_i32 = constant 1 : i32
  return %c1_i32, %c1_i32 : i32, i32
}

20 KiB Raw Blame History