AMDGPU/SI: Don't move copies of immediates to the VALU
Summary:
If we write an immediate to a VGPR and then copy the VGPR to an
SGPR, we can replace the copy with a S_MOV_B32 sgpr, imm, rather than
moving the copy to the SALU.
Reviewers: arsenm
Subscribers: kzhuravl, wdng, nhaehnle, yaxunl, llvm-commits, tony-tye
Differential Revision: https://reviews.llvm.org/D27272
llvm-svn: 288849
diff --git a/llvm/lib/Target/AMDGPU/SIFixSGPRCopies.cpp b/llvm/lib/Target/AMDGPU/SIFixSGPRCopies.cpp
index 8c4e0ad..6a422e7 100644
--- a/llvm/lib/Target/AMDGPU/SIFixSGPRCopies.cpp
+++ b/llvm/lib/Target/AMDGPU/SIFixSGPRCopies.cpp
@@ -294,6 +294,38 @@
return false;
}
+static bool isSafeToFoldImmIntoCopy(const MachineInstr *Copy,
+ const MachineInstr *MoveImm,
+ const SIInstrInfo *TII,
+ unsigned &SMovOp,
+ int64_t &Imm) {
+
+ if (!MoveImm->isMoveImmediate())
+ return false;
+
+ const MachineOperand *ImmOp =
+ TII->getNamedOperand(*MoveImm, AMDGPU::OpName::src0);
+ if (!ImmOp->isImm())
+ return false;
+
+ // FIXME: Handle copies with sub-regs.
+ if (Copy->getOperand(0).getSubReg())
+ return false;
+
+ switch (MoveImm->getOpcode()) {
+ default:
+ return false;
+ case AMDGPU::V_MOV_B32_e32:
+ SMovOp = AMDGPU::S_MOV_B32;
+ break;
+ case AMDGPU::V_MOV_B64_PSEUDO:
+ SMovOp = AMDGPU::S_MOV_B64;
+ break;
+ }
+ Imm = ImmOp->getImm();
+ return true;
+}
+
bool SIFixSGPRCopies::runOnMachineFunction(MachineFunction &MF) {
const SISubtarget &ST = MF.getSubtarget<SISubtarget>();
MachineRegisterInfo &MRI = MF.getRegInfo();
@@ -323,7 +355,17 @@
const TargetRegisterClass *SrcRC, *DstRC;
std::tie(SrcRC, DstRC) = getCopyRegClasses(MI, *TRI, MRI);
if (isVGPRToSGPRCopy(SrcRC, DstRC, *TRI)) {
- DEBUG(dbgs() << "Fixing VGPR -> SGPR copy: " << MI);
+ MachineInstr *DefMI = MRI.getVRegDef(MI.getOperand(1).getReg());
+ unsigned SMovOp;
+ int64_t Imm;
+ // If we are just copying an immediate, we can replace the copy with
+ // s_mov_b32.
+ if (isSafeToFoldImmIntoCopy(&MI, DefMI, TII, SMovOp, Imm)) {
+ MI.getOperand(1).ChangeToImmediate(Imm);
+ MI.addImplicitDefUseOperands(MF);
+ MI.setDesc(TII->get(SMovOp));
+ break;
+ }
TII->moveToVALU(MI);
}