[llvm] 7e15ea1 - [RISCV] Add a DAG combine to pre-promote (i1 (truncate (i32 (srl X, Y)))) with Zbs on RV64.

[Craig Topper via llvm-commits]

Author: Craig Topper
Date: 2022-11-16T19:07:33-08:00
New Revision: 7e15ea102f10061b47b43e230b6841e8e1274d22

URL: https://github.com/llvm/llvm-project/commit/7e15ea102f10061b47b43e230b6841e8e1274d22
DIFF: https://github.com/llvm/llvm-project/commit/7e15ea102f10061b47b43e230b6841e8e1274d22.diff

LOG: [RISCV] Add a DAG combine to pre-promote (i1 (truncate (i32 (srl X, Y)))) with Zbs on RV64.

Type legalization will want to turn (srl X, Y) into RISCVISD::SRLW,
which will prevent us from using a BEXT instruction.

This is similar to what we do for (i32 (and (srl X, Y), 1)).

Added: 
    

Modified: 
    llvm/lib/Target/RISCV/RISCVISelLowering.cpp
    llvm/test/CodeGen/RISCV/rv64zbs.ll

Removed: 
    


################################################################################
diff  --git a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
index 2c344590ddf4e..f7e8c7dec90fc 100644
--- a/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
+++ b/llvm/lib/Target/RISCV/RISCVISelLowering.cpp
@@ -997,6 +997,9 @@ RISCVTargetLowering::RISCVTargetLowering(const TargetMachine &TM,
   if (Subtarget.hasStdExtZbb())
     setTargetDAGCombine({ISD::UMAX, ISD::UMIN, ISD::SMAX, ISD::SMIN});
 
+  if (Subtarget.hasStdExtZbs() && Subtarget.is64Bit())
+    setTargetDAGCombine(ISD::TRUNCATE);
+
   if (Subtarget.hasStdExtZbkb())
     setTargetDAGCombine(ISD::BITREVERSE);
   if (Subtarget.hasStdExtZfh())
@@ -8259,6 +8262,29 @@ static SDValue combineDeMorganOfBoolean(SDNode *N, SelectionDAG &DAG) {
   return DAG.getNode(ISD::XOR, DL, VT, Logic, DAG.getConstant(1, DL, VT));
 }
 
+static SDValue performTRUNCATECombine(SDNode *N, SelectionDAG &DAG,
+                                      const RISCVSubtarget &Subtarget) {
+  SDValue N0 = N->getOperand(0);
+  EVT VT = N->getValueType(0);
+
+  // Pre-promote (i1 (truncate (srl X, Y))) on RV64 with Zbs without zero
+  // extending X. This is safe since we only need the LSB after the shift and
+  // shift amounts larger than 31 would produce poison. If we wait until
+  // type legalization, we'll create RISCVISD::SRLW and we can't recover it
+  // to use a BEXT instruction.
+  if (Subtarget.is64Bit() && Subtarget.hasStdExtZbs() && VT == MVT::i1 &&
+      N0.getValueType() == MVT::i32 && N0.getOpcode() == ISD::SRL &&
+      !isa<ConstantSDNode>(N0.getOperand(1)) && N0.hasOneUse()) {
+    SDLoc DL(N0);
+    SDValue Op0 = DAG.getNode(ISD::ANY_EXTEND, DL, MVT::i64, N0.getOperand(0));
+    SDValue Op1 = DAG.getNode(ISD::ZERO_EXTEND, DL, MVT::i64, N0.getOperand(1));
+    SDValue Srl = DAG.getNode(ISD::SRL, DL, MVT::i64, Op0, Op1);
+    return DAG.getNode(ISD::TRUNCATE, SDLoc(N), VT, Srl);
+  }
+
+  return SDValue();
+}
+
 static SDValue performANDCombine(SDNode *N,
                                  TargetLowering::DAGCombinerInfo &DCI,
                                  const RISCVSubtarget &Subtarget) {
@@ -9573,6 +9599,8 @@ SDValue RISCVTargetLowering::PerformDAGCombine(SDNode *N,
       }
     }
     return SDValue();
+  case ISD::TRUNCATE:
+    return performTRUNCATECombine(N, DAG, Subtarget);
   case RISCVISD::SELECT_CC: {
     // Transform
     SDValue LHS = N->getOperand(0);

diff  --git a/llvm/test/CodeGen/RISCV/rv64zbs.ll b/llvm/test/CodeGen/RISCV/rv64zbs.ll
index 6ccb371529647..19ea17ba69862 100644
--- a/llvm/test/CodeGen/RISCV/rv64zbs.ll
+++ b/llvm/test/CodeGen/RISCV/rv64zbs.ll
@@ -363,6 +363,42 @@ define signext i32 @bext_i32_no_mask(i32 signext %a, i32 signext %b) nounwind {
   ret i32 %and1
 }
 
+; This gets previous converted to (i1 (truncate (srl X, Y)). Make sure we are
+; able to use bext.
+define void @bext_i32_trunc(i32 signext %0, i32 signext %1) {
+; RV64I-LABEL: bext_i32_trunc:
+; RV64I:       # %bb.0:
+; RV64I-NEXT:    srlw a0, a0, a1
+; RV64I-NEXT:    andi a0, a0, 1
+; RV64I-NEXT:    beqz a0, .LBB19_2
+; RV64I-NEXT:  # %bb.1:
+; RV64I-NEXT:    ret
+; RV64I-NEXT:  .LBB19_2:
+; RV64I-NEXT:    tail bar at plt
+;
+; RV64ZBS-LABEL: bext_i32_trunc:
+; RV64ZBS:       # %bb.0:
+; RV64ZBS-NEXT:    bext a0, a0, a1
+; RV64ZBS-NEXT:    beqz a0, .LBB19_2
+; RV64ZBS-NEXT:  # %bb.1:
+; RV64ZBS-NEXT:    ret
+; RV64ZBS-NEXT:  .LBB19_2:
+; RV64ZBS-NEXT:    tail bar at plt
+  %3 = shl i32 1, %1
+  %4 = and i32 %3, %0
+  %5 = icmp eq i32 %4, 0
+  br i1 %5, label %6, label %7
+
+6:                                                ; preds = %2
+  tail call void @bar()
+  br label %7
+
+7:                                                ; preds = %6, %2
+  ret void
+}
+
+declare void @bar()
+
 define i64 @bext_i64(i64 %a, i64 %b) nounwind {
 ; RV64I-LABEL: bext_i64:
 ; RV64I:       # %bb.0: