|
LLVM 23.0.0git
|
Go to the source code of this file.
Namespaces | |
| namespace | llvm |
| This is an optimization pass for GlobalISel generic memory operations. | |
| namespace | llvm::AMDGPU |
Enumerations | |
| enum class | llvm::AMDGPU::SchedulingPhase { llvm::AMDGPU::Initial , llvm::AMDGPU::PreRAReentry , llvm::AMDGPU::PostRA } |
| enum | llvm::AMDGPU::IGLPStrategyID : int { llvm::AMDGPU::MFMASmallGemmOptID = 0 , llvm::AMDGPU::MFMASmallGemmSingleWaveOptID = 1 , llvm::AMDGPU::MFMAExpInterleaveID = 2 , llvm::AMDGPU::MFMAExpSimpleInterleaveID = 3 } |
| Operand 0 immediate for IGLP_OPT pseudo instructions. More... | |
Functions | |
| std::unique_ptr< ScheduleDAGMutation > | llvm::createIGroupLPDAGMutation (AMDGPU::SchedulingPhase Phase) |
Phase specifes whether or not this is a reentry into the IGroupLPDAGMutation. | |