[llvm] [AMDGPU] SIPeepholeSDWA: Handle V_CNDMASK_B32_e64 (PR #137930)
Frederik Harwath via llvm-commits
llvm-commits at lists.llvm.org
Fri May 2 08:58:14 PDT 2025
================
@@ -1061,6 +1063,62 @@ void SIPeepholeSDWA::pseudoOpConvertToVOP2(MachineInstr &MI,
MISucc.substituteRegister(CarryIn->getReg(), TRI->getVCC(), 0, *TRI);
}
+/// Try to convert an \p MI in VOP3 which takes an src2 carry-in
+/// operand into the corresponding VOP2 form which expects the
+/// argument in VCC. To this end, either try to change the definition
+/// of the carry-in operand to write to VCC or add an instruction that
+/// copies from the carry-in to VCC. The conversion will only be
+/// applied if \p MI can be shrunk to VOP2 and if VCC can be proven to
+/// be dead before \p MI.
+void SIPeepholeSDWA::convertVcndmaskToVOP2(MachineInstr &MI,
+ const GCNSubtarget &ST) const {
+ assert(MI.getOpcode() == AMDGPU::V_CNDMASK_B32_e64);
+
+ LLVM_DEBUG(dbgs() << "Attempting VOP2 conversion: " << MI);
+ if (!TII->canShrink(MI, *MRI)) {
+ LLVM_DEBUG(dbgs() << "Cannot shrink instruction\n");
+ return;
+ }
+
+ const MachineOperand &CarryIn =
+ *TII->getNamedOperand(MI, AMDGPU::OpName::src2);
+ Register CarryReg = CarryIn.getReg();
+ MachineInstr *CarryDef = MRI->getVRegDef(CarryReg);
+ if (!CarryDef) {
+ LLVM_DEBUG(dbgs() << "Missing carry-in operand definition\n");
+ return;
+ }
+
+ // Make sure VCC or its subregs are dead before MI.
+ MCRegister Vcc = TRI->getVCC();
+ MachineBasicBlock &MBB = *MI.getParent();
+ MachineBasicBlock::LivenessQueryResult Liveness =
+ MBB.computeRegisterLiveness(TRI, Vcc, MI);
+ if (Liveness != MachineBasicBlock::LQR_Dead) {
+ LLVM_DEBUG(dbgs() << "VCC not known to be dead before instruction\n");
+ return;
+ }
+
+ // Change destination of compare instruction to VCC
+ // or copy to VCC if carry-in is not a compare inst.
+ if (CarryDef->isCompare() && TII->isVOP3(*CarryDef) &&
+ MRI->hasOneUse(CarryIn.getReg()))
----------------
frederik-h wrote:
My initial implementation was actually converting to the vop2 form. I removed this conversion in an attempt to simplify the code. I suppose the actual conversion of the carry-in def to vop2 is not really necessary here? I have changed the condition to make sure that v_cmp_class and all instructions which have a vopc form will be included.
https://github.com/llvm/llvm-project/pull/137930
More information about the llvm-commits
mailing list