[GlobalISel][AArch64] Select llvm.aarch64.crypto.sha1h

This was falling back and gives us a reason to create a selectIntrinsic function
which we would need eventually anyway. Update arm64-crypto.ll to show that we
correctly select it.

Also factor out the code for finding an intrinsic ID.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@359501 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/AArch64/AArch64InstructionSelector.cpp b/lib/Target/AArch64/AArch64InstructionSelector.cpp
index 7f5bc2792379d3a39d117e9b1695d3994df13157..ae668ddac4001a3622d38495a434521a2774ad48 100644 (file)
--- a/lib/Target/AArch64/AArch64InstructionSelector.cpp
+++ b/lib/Target/AArch64/AArch64InstructionSelector.cpp
@@ -101,6 +101,7 @@ private:
                                 MachineRegisterInfo &MRI) const;
   bool selectIntrinsicWithSideEffects(MachineInstr &I,
                                       MachineRegisterInfo &MRI) const;
+  bool selectIntrinsic(MachineInstr &I, MachineRegisterInfo &MRI) const;
   bool selectVectorICmp(MachineInstr &I, MachineRegisterInfo &MRI) const;
   bool selectIntrinsicTrunc(MachineInstr &I, MachineRegisterInfo &MRI) const;
   bool selectIntrinsicRound(MachineInstr &I, MachineRegisterInfo &MRI) const;
@@ -1855,6 +1856,8 @@ bool AArch64InstructionSelector::select(MachineInstr &I,
   case TargetOpcode::G_VASTART:
     return STI.isTargetDarwin() ? selectVaStartDarwin(I, MF, MRI)
                                 : selectVaStartAAPCS(I, MF, MRI);
+  case TargetOpcode::G_INTRINSIC:
+    return selectIntrinsic(I, MRI);
   case TargetOpcode::G_INTRINSIC_W_SIDE_EFFECTS:
     return selectIntrinsicWithSideEffects(I, MRI);
   case TargetOpcode::G_IMPLICIT_DEF: {
@@ -3083,6 +3086,17 @@ bool AArch64InstructionSelector::selectBuildVector(
   return true;
 }
 
+/// Helper function to find an intrinsic ID on an a MachineInstr. Returns the
+/// ID if it exists, and 0 otherwise.
+static unsigned findIntrinsicID(MachineInstr &I) {
+  auto IntrinOp = find_if(I.operands(), [&](const MachineOperand &Op) {
+    return Op.isIntrinsicID();
+  });
+  if (IntrinOp == I.operands_end())
+    return 0;
+  return IntrinOp->getIntrinsicID();
+}
+
 /// Helper function to emit the correct opcode for a llvm.aarch64.stlxr
 /// intrinsic.
 static unsigned getStlxrOpcode(unsigned NumBytesToStore) {
@@ -3101,12 +3115,9 @@ static unsigned getStlxrOpcode(unsigned NumBytesToStore) {
 bool AArch64InstructionSelector::selectIntrinsicWithSideEffects(
     MachineInstr &I, MachineRegisterInfo &MRI) const {
   // Find the intrinsic ID.
-  auto IntrinOp = find_if(I.operands(), [&](const MachineOperand &Op) {
-    return Op.isIntrinsicID();
-  });
-  if (IntrinOp == I.operands_end())
+  unsigned IntrinID = findIntrinsicID(I);
+  if (!IntrinID)
     return false;
-  unsigned IntrinID = IntrinOp->getIntrinsicID();
   MachineIRBuilder MIRBuilder(I);
 
   // Select the instruction.
@@ -3148,6 +3159,58 @@ bool AArch64InstructionSelector::selectIntrinsicWithSideEffects(
   return true;
 }
 
+bool AArch64InstructionSelector::selectIntrinsic(
+    MachineInstr &I, MachineRegisterInfo &MRI) const {
+  unsigned IntrinID = findIntrinsicID(I);
+  if (!IntrinID)
+    return false;
+  MachineIRBuilder MIRBuilder(I);
+
+  switch (IntrinID) {
+  default:
+    break;
+  case Intrinsic::aarch64_crypto_sha1h:
+    unsigned DstReg = I.getOperand(0).getReg();
+    unsigned SrcReg = I.getOperand(2).getReg();
+
+    // FIXME: Should this be an assert?
+    if (MRI.getType(DstReg).getSizeInBits() != 32 ||
+        MRI.getType(SrcReg).getSizeInBits() != 32)
+      return false;
+
+    // The operation has to happen on FPRs. Set up some new FPR registers for
+    // the source and destination if they are on GPRs.
+    if (RBI.getRegBank(SrcReg, MRI, TRI)->getID() != AArch64::FPRRegBankID) {
+      SrcReg = MRI.createVirtualRegister(&AArch64::FPR32RegClass);
+      MIRBuilder.buildCopy({SrcReg}, {I.getOperand(2)});
+
+      // Make sure the copy ends up getting constrained properly.
+      RBI.constrainGenericRegister(I.getOperand(2).getReg(),
+                                   AArch64::GPR32RegClass, MRI);
+    }
+
+    if (RBI.getRegBank(DstReg, MRI, TRI)->getID() != AArch64::FPRRegBankID)
+      DstReg = MRI.createVirtualRegister(&AArch64::FPR32RegClass);
+
+    // Actually insert the instruction.
+    auto SHA1Inst = MIRBuilder.buildInstr(AArch64::SHA1Hrr, {DstReg}, {SrcReg});
+    constrainSelectedInstRegOperands(*SHA1Inst, TII, TRI, RBI);
+
+    // Did we create a new register for the destination?
+    if (DstReg != I.getOperand(0).getReg()) {
+      // Yep. Copy the result of the instruction back into the original
+      // destination.
+      MIRBuilder.buildCopy({I.getOperand(0)}, {DstReg});
+      RBI.constrainGenericRegister(I.getOperand(0).getReg(),
+                                   AArch64::GPR32RegClass, MRI);
+    }
+
+    I.eraseFromParent();
+    return true;
+  }
+  return false;
+}
+
 /// SelectArithImmed - Select an immediate value that can be represented as
 /// a 12-bit value shifted left by either 0 or 12.  If so, return true with
 /// Val set to the 12-bit value and Shift set to the shifter operand.

diff --git a/test/CodeGen/AArch64/arm64-crypto.ll b/test/CodeGen/AArch64/arm64-crypto.ll
index 615f2a8ecdca664442062953d574057747b85f71..1def7588e7bde91dd80dd12f6ee526ee0eda8e27 100644 (file)
--- a/test/CodeGen/AArch64/arm64-crypto.ll
+++ b/test/CodeGen/AArch64/arm64-crypto.ll
@@ -1,4 +1,5 @@
 ; RUN: llc -mtriple=arm64-eabi -mattr=crypto -aarch64-neon-syntax=apple -o - %s | FileCheck %s
+; RUN: llc -mtriple=arm64-eabi -global-isel -global-isel-abort=2 -pass-remarks-missed=gisel* -mattr=crypto -aarch64-neon-syntax=apple -o - %s 2>&1 | FileCheck %s --check-prefixes=CHECK,FALLBACK
 
 declare <16 x i8> @llvm.aarch64.crypto.aese(<16 x i8> %data, <16 x i8> %key)
 declare <16 x i8> @llvm.aarch64.crypto.aesd(<16 x i8> %data, <16 x i8> %key)
@@ -77,6 +78,7 @@ define <4 x i32> @test_sha1m(<4 x i32> %hash_abcd, i32 %hash_e, <4 x i32> %wk) {
   ret <4 x i32> %res
 }
 
+; FALLBACK-NOT: remark{{.*}}test_sha1h
 define i32 @test_sha1h(i32 %hash_e) {
 ; CHECK-LABEL: test_sha1h:
 ; CHECK: fmov [[HASH_E:s[0-9]+]], w0