tvm::s_tir::meta_schedule::Postproc Class Reference

Managed reference to PostprocNode. More...

#include <postproc.h>

Inheritance diagram for tvm::s_tir::meta_schedule::Postproc:

Collaboration diagram for tvm::s_tir::meta_schedule::Postproc:

Public Types
using	FInitializeWithTuneContext = ffi::TypedFunction< void(const TuneContext &)>
	The function type of InitializeWithTuneContext method.
using	FApply = ffi::TypedFunction< bool(const s_tir::Schedule &)>
	Apply a postprocessor to the given schedule.
using	FClone = ffi::TypedFunction< Postproc()>
	Clone the postprocessor.

Public Member Functions
	TVM_FFI_DEFINE_OBJECT_REF_METHODS_NULLABLE (Postproc, ffi::ObjectRef, PostprocNode)

Static Public Member Functions
static Postproc	PyPostproc (FInitializeWithTuneContext f_initialize_with_tune_context, FApply f_apply, FClone f_clone)
	Create a postprocessor with customized methods on the python-side.
static Postproc	DisallowDynamicLoop ()
	Create a postprocessor that checks if all loops are static.
static Postproc	DisallowAsyncStridedMemCopy ()
	Create a postprocessor that checks if all async mem copies are not strided.
static Postproc	RewriteCooperativeFetch ()
	Create a postprocessor that rewrites the cooperative fetch annotation to actual vectorized cooperative fetching in loop bindings.
static Postproc	RewriteParallelVectorizeUnroll ()
	Creates a postprocessor that applies parallelization, vectorization and auto unrolling according to the annotation of each block.
static Postproc	RewriteReductionBlock ()
	Create a postprocessor that rewrites reduction block by moving the init block out.
static Postproc	RewriteUnboundBlock (int max_threadblocks)
	Create a postprocessor that adds thread binding to unbound blocks.
static Postproc	RewriteTensorize (bool vectorize_init_loop=false)
	Create a postprocessor that applies tensorization to annotated blocks.
static Postproc	VerifyGPUCode ()
	Creates a postprocessor that verifies if the GPU code is correct.
static Postproc	VerifyVTCMLimit ()
	Verifies that the VTCM usage of a given schedule is within the provided limit.
static Postproc	RewriteLayout ()
	Creates a postprocessor that rewrites the layout of input tensor.
static ffi::Array< Postproc, void >	DefaultLLVM ()
	Create default postprocessors for LLVM.
static ffi::Array< Postproc, void >	DefaultCPUTensorization ()
	Create default postprocessors for x86 (AVX512 and VNNI)
static ffi::Array< Postproc, void >	DefaultRISCV ()
	Create default postprocessors for RISCV.
static ffi::Array< Postproc, void >	DefaultCUDA ()
	Create default postprocessors for CUDA.
static ffi::Array< Postproc, void >	DefaultCUDATensorCore ()
	Create default postprocessors for CUDA with TensorCore.
static ffi::Array< Postproc, void >	DefaultHexagon ()
	Create default postprocessors for Hexagon.

Detailed Description

Managed reference to PostprocNode.

See also

PostprocNode

Member Typedef Documentation

FApply

using tvm::s_tir::meta_schedule::Postproc::FApply = ffi::TypedFunction<bool(const s_tir::Schedule&)>

Apply a postprocessor to the given schedule.

Parameters

sch	The schedule to be post processed.

Returns

Whether the postprocessor was successfully applied.

FClone

using tvm::s_tir::meta_schedule::Postproc::FClone = ffi::TypedFunction<Postproc()>

Clone the postprocessor.

Returns

The cloned postprocessor.

FInitializeWithTuneContext

using tvm::s_tir::meta_schedule::Postproc::FInitializeWithTuneContext = ffi::TypedFunction<void(const TuneContext&)>

The function type of InitializeWithTuneContext method.

Parameters

context

The tuning context for initialization.

Member Function Documentation

DefaultCPUTensorization()

static ffi::Array< Postproc, void > tvm::s_tir::meta_schedule::Postproc::DefaultCPUTensorization ( )

static

Create default postprocessors for x86 (AVX512 and VNNI)

DefaultCUDA()

static ffi::Array< Postproc, void > tvm::s_tir::meta_schedule::Postproc::DefaultCUDA ( )

static

Create default postprocessors for CUDA.

DefaultCUDATensorCore()

static ffi::Array< Postproc, void > tvm::s_tir::meta_schedule::Postproc::DefaultCUDATensorCore ( )

static

Create default postprocessors for CUDA with TensorCore.

DefaultHexagon()

static ffi::Array< Postproc, void > tvm::s_tir::meta_schedule::Postproc::DefaultHexagon ( )

static

Create default postprocessors for Hexagon.

DefaultLLVM()

static ffi::Array< Postproc, void > tvm::s_tir::meta_schedule::Postproc::DefaultLLVM ( )

static

Create default postprocessors for LLVM.

DefaultRISCV()

static ffi::Array< Postproc, void > tvm::s_tir::meta_schedule::Postproc::DefaultRISCV ( )

static

Create default postprocessors for RISCV.

DisallowAsyncStridedMemCopy()

static Postproc tvm::s_tir::meta_schedule::Postproc::DisallowAsyncStridedMemCopy ( )

static

Create a postprocessor that checks if all async mem copies are not strided.

Returns

The postprocessor created

DisallowDynamicLoop()

static Postproc tvm::s_tir::meta_schedule::Postproc::DisallowDynamicLoop ( )

static

Create a postprocessor that checks if all loops are static.

Returns

The postprocessor created

PyPostproc()

static Postproc tvm::s_tir::meta_schedule::Postproc::PyPostproc ( FInitializeWithTuneContext  f_initialize_with_tune_context, FApply  f_apply, FClone  f_clone  )

static

Create a postprocessor with customized methods on the python-side.

Parameters

f_initialize_with_tune_context	The packed function of InitializeWithTuneContext.
f_apply	The packed function of Apply.
f_clone	The packed function of Clone.

Returns

The postprocessor created.

RewriteCooperativeFetch()

static Postproc tvm::s_tir::meta_schedule::Postproc::RewriteCooperativeFetch ( )

static

Create a postprocessor that rewrites the cooperative fetch annotation to actual vectorized cooperative fetching in loop bindings.

Returns

The postprocessor created.

RewriteLayout()

static Postproc tvm::s_tir::meta_schedule::Postproc::RewriteLayout ( )

static

Creates a postprocessor that rewrites the layout of input tensor.

Note

Weight layout rewrite is supported so far, activation layout rewrite will be added.

Returns

The postprocessor created

RewriteParallelVectorizeUnroll()

static Postproc tvm::s_tir::meta_schedule::Postproc::RewriteParallelVectorizeUnroll ( )

static

Creates a postprocessor that applies parallelization, vectorization and auto unrolling according to the annotation of each block.

Returns

The postprocessor created

RewriteReductionBlock()

static Postproc tvm::s_tir::meta_schedule::Postproc::RewriteReductionBlock ( )

static

Create a postprocessor that rewrites reduction block by moving the init block out.

Returns

The postprocessor created.

RewriteTensorize()

static Postproc tvm::s_tir::meta_schedule::Postproc::RewriteTensorize ( bool  vectorize_init_loop = false )

static

Create a postprocessor that applies tensorization to annotated blocks.

Parameters

vectorize_init_loop

Whether or not vectorize the initialization loop produced by DecomposeReduction

Returns

The postprocessor created.

RewriteUnboundBlock()

static Postproc tvm::s_tir::meta_schedule::Postproc::RewriteUnboundBlock ( int  max_threadblocks )

static

Create a postprocessor that adds thread binding to unbound blocks.

Parameters

max_threadblocks

The max number of threadblocks in the CUDA device.

Returns

The postprocessor created.

TVM_FFI_DEFINE_OBJECT_REF_METHODS_NULLABLE()

tvm::s_tir::meta_schedule::Postproc::TVM_FFI_DEFINE_OBJECT_REF_METHODS_NULLABLE	(	Postproc	,
		ffi::ObjectRef	,
		PostprocNode
	)

VerifyGPUCode()

static Postproc tvm::s_tir::meta_schedule::Postproc::VerifyGPUCode ( )

static

Creates a postprocessor that verifies if the GPU code is correct.

Returns

The postprocessor created

VerifyVTCMLimit()

static Postproc tvm::s_tir::meta_schedule::Postproc::VerifyVTCMLimit ( )

static

Verifies that the VTCM usage of a given schedule is within the provided limit.

Returns

The postprocessor created

---

The documentation for this class was generated from the following file:

• include/tvm/s_tir/meta_schedule/postproc.h