AMDGPUTargetTransformInfo.cpp File Reference

#include "AMDGPUTargetTransformInfo.h"
#include "AMDGPUTargetMachine.h"
#include "MCTargetDesc/AMDGPUMCTargetDesc.h"
#include "llvm/Analysis/LoopInfo.h"
#include "llvm/Analysis/ValueTracking.h"
#include "llvm/IR/IRBuilder.h"
#include "llvm/IR/IntrinsicsAMDGPU.h"
#include "llvm/IR/PatternMatch.h"
#include "llvm/Support/KnownBits.h"

Include dependency graph for AMDGPUTargetTransformInfo.cpp:

Go to the source code of this file.

Macros
#define	DEBUG_TYPE   "AMDGPUtti"

Functions
static bool	dependsOnLocalPhi (const Loop *L, const Value *Cond, unsigned Depth=0)
static bool	intrinsicHasPackedVectorBenefit (Intrinsic::ID ID)

Variables
static cl::opt< unsigned >	UnrollThresholdPrivate ("amdgpu-unroll-threshold-private", cl::desc("Unroll threshold for AMDGPU if private memory used in a loop"), cl::init(2700), cl::Hidden)
static cl::opt< unsigned >	UnrollThresholdLocal ("amdgpu-unroll-threshold-local", cl::desc("Unroll threshold for AMDGPU if local memory used in a loop"), cl::init(1000), cl::Hidden)
static cl::opt< unsigned >	UnrollThresholdIf ("amdgpu-unroll-threshold-if", cl::desc("Unroll threshold increment for AMDGPU for each if statement inside loop"), cl::init(200), cl::Hidden)
static cl::opt< bool >	UnrollRuntimeLocal ("amdgpu-unroll-runtime-local", cl::desc("Allow runtime unroll for AMDGPU if local memory used in a loop"), cl::init(true), cl::Hidden)
static cl::opt< bool >	UseLegacyDA ("amdgpu-use-legacy-divergence-analysis", cl::desc("Enable legacy divergence analysis for AMDGPU"), cl::init(false), cl::Hidden)
static cl::opt< unsigned >	UnrollMaxBlockToAnalyze ("amdgpu-unroll-max-block-to-analyze", cl::desc("Inner loop block size threshold to analyze in unroll for AMDGPU"), cl::init(32), cl::Hidden)
static cl::opt< unsigned >	ArgAllocaCost ("amdgpu-inline-arg-alloca-cost", cl::Hidden, cl::init(4000), cl::desc("Cost of alloca argument"))
static cl::opt< unsigned >	ArgAllocaCutoff ("amdgpu-inline-arg-alloca-cutoff", cl::Hidden, cl::init(256), cl::desc("Maximum alloca size to use for inline cost"))
static cl::opt< size_t >	InlineMaxBB ("amdgpu-inline-max-bb", cl::Hidden, cl::init(1100), cl::desc("Maximum number of BBs allowed in a function after inlining" " (compile time constraint)"))

Macro Definition Documentation

DEBUG_TYPE

#define DEBUG_TYPE   "AMDGPUtti"

Definition at line 29 of file AMDGPUTargetTransformInfo.cpp.

Function Documentation

dependsOnLocalPhi()

static bool dependsOnLocalPhi ( const Loop *  L, const Value *  Cond, unsigned  Depth = 0  )

static

Definition at line 79 of file AMDGPUTargetTransformInfo.cpp.

References Cond, llvm::LoopBase< BlockT, LoopT >::contains(), llvm::Depth, llvm::LoopBase< BlockT, LoopT >::getSubLoops(), I, and llvm::none_of().

Referenced by llvm::AMDGPUTTIImpl::getUnrollingPreferences().

intrinsicHasPackedVectorBenefit()

static bool intrinsicHasPackedVectorBenefit ( Intrinsic::ID  ID )

static

Definition at line 720 of file AMDGPUTargetTransformInfo.cpp.

References round().

Referenced by llvm::GCNTTIImpl::getIntrinsicInstrCost().

Variable Documentation

ArgAllocaCost

cl::opt<unsigned> ArgAllocaCost("amdgpu-inline-arg-alloca-cost", cl::Hidden, cl::init(4000), cl::desc("Cost of alloca argument"))

static

Referenced by llvm::GCNTTIImpl::adjustInliningThreshold().

ArgAllocaCutoff

cl::opt<unsigned> ArgAllocaCutoff("amdgpu-inline-arg-alloca-cutoff", cl::Hidden, cl::init(256), cl::desc("Maximum alloca size to use for inline cost"))

static

Referenced by llvm::GCNTTIImpl::adjustInliningThreshold().

InlineMaxBB

cl::opt<size_t> InlineMaxBB("amdgpu-inline-max-bb", cl::Hidden, cl::init(1100), cl::desc("Maximum number of BBs allowed in a function after inlining" " (compile time constraint)"))

static

Referenced by llvm::GCNTTIImpl::areInlineCompatible().

UnrollMaxBlockToAnalyze

cl::opt<unsigned> UnrollMaxBlockToAnalyze("amdgpu-unroll-max-block-to-analyze", cl::desc("Inner loop block size threshold to analyze in unroll for AMDGPU"), cl::init(32), cl::Hidden)

static

Referenced by llvm::AMDGPUTTIImpl::getUnrollingPreferences().

UnrollRuntimeLocal

cl::opt<bool> UnrollRuntimeLocal("amdgpu-unroll-runtime-local", cl::desc("Allow runtime unroll for AMDGPU if local memory used in a loop"), cl::init(true), cl::Hidden)

static

Referenced by llvm::AMDGPUTTIImpl::getUnrollingPreferences().

UnrollThresholdIf

cl::opt<unsigned> UnrollThresholdIf("amdgpu-unroll-threshold-if", cl::desc("Unroll threshold increment for AMDGPU for each if statement inside loop"), cl::init(200), cl::Hidden)

static

Referenced by llvm::AMDGPUTTIImpl::getUnrollingPreferences().

UnrollThresholdLocal

cl::opt<unsigned> UnrollThresholdLocal("amdgpu-unroll-threshold-local", cl::desc("Unroll threshold for AMDGPU if local memory used in a loop"), cl::init(1000), cl::Hidden)

static

Referenced by llvm::AMDGPUTTIImpl::getUnrollingPreferences().

UnrollThresholdPrivate

cl::opt<unsigned> UnrollThresholdPrivate("amdgpu-unroll-threshold-private", cl::desc("Unroll threshold for AMDGPU if private memory used in a loop"), cl::init(2700), cl::Hidden)

static

Referenced by llvm::AMDGPUTTIImpl::getUnrollingPreferences().

UseLegacyDA

cl::opt<bool> UseLegacyDA("amdgpu-use-legacy-divergence-analysis", cl::desc("Enable legacy divergence analysis for AMDGPU"), cl::init(false), cl::Hidden)

static

Referenced by llvm::GCNTTIImpl::useGPUDivergenceAnalysis().