Definition of two pillar data structure for TensorIR scheduling: StmtSRef, BlockScope. More...

#include <tvm/tir/stmt.h>
#include <unordered_map>

Include dependency graph for block_scope.h:

This graph shows which files directly or indirectly include this file:

Classes
class	tvm::tir::StmtSRefNode
	An object that refers to schedulable elements (block/for-loop) in TensorIR, aka "sref". More...

class	tvm::tir::StmtSRef
	Managed reference to StmtSRefNode. More...

class	tvm::tir::DependencyNode
	A tuple (src, dst, kind) representing certain types of dependency. For example, (A, B, kRAW) means block B depends on block A, and the dependency kind is read-after-write, which means block B reads the result written by block A. More...

class	tvm::tir::Dependency
	Managed reference to DependencyNode. More...

class	tvm::tir::BlockScopeNode
	An object with 1-to-1 correspondence with each block reference in the sref tree. This data structure is used to track the producer-consumer dependencies between blocks. For example even leaf nodes have a scope node, even though they have no dependencies. More...

class	tvm::tir::BlockScope
	Managed reference to BlockScopeNode. More...

Namespaces
	tvm
	runtime implementation for LibTorch/TorchScript.

	tvm::tir

Enumerations
enum	tvm::tir::DepKind : int32_t { tvm::tir::DepKind::kRAW = 0, tvm::tir::DepKind::kWAW = 1, tvm::tir::DepKind::kWAR = 2, tvm::tir::kOpaque = 3 }
	Type of dependency. Right now we have 4 types of dependencies 1) Read-after-write (kRAW) 2) Write-after-write (kWAW) 3) Write-after-read (kWAR) 4) Opaque dependency (kOpaque) More...

Detailed Description

Definition of two pillar data structure for TensorIR scheduling: StmtSRef, BlockScope.

Classes