[X86] Lowering sqrt intrinsics to native IR

Reviewers: craig.topper, spatel, RKSimon, igorb, uriel.k

Reviewed By: craig.topper

Subscribers: tkrupa, cfe-commits

Differential Revision: https://reviews.llvm.org/D41168

llvm-svn: 334850

1 files changed, 50 insertions, 1 deletions

diff --git a/clang/lib/CodeGen/CGBuiltin.cpp b/clang/lib/CodeGen/CGBuiltin.cpp
index 125eb2031cb..7b0ea22f94e 100644
--- a/clang/lib/CodeGen/CGBuiltin.cpp
+++ b/clang/lib/CodeGen/CGBuiltin.cpp
@@ -9889,7 +9889,56 @@ Value *CodeGenFunction::EmitX86BuiltinExpr(unsigned BuiltinID,
     Function *F = CGM.getIntrinsic(Intrinsic::ctlz, Ops[0]->getType());
     return Builder.CreateCall(F, {Ops[0],Builder.getInt1(false)});
   }
-
+  case X86::BI__builtin_ia32_sqrtss:
+  case X86::BI__builtin_ia32_sqrtsd: {
+    Value *A = Builder.CreateExtractElement(Ops[0], (uint64_t)0);
+    Function *F = CGM.getIntrinsic(Intrinsic::sqrt, A->getType());
+    A = Builder.CreateCall(F, {A});
+    return Builder.CreateInsertElement(Ops[0], A, (uint64_t)0);    
+  }
+  case X86::BI__builtin_ia32_sqrtsd_round_mask:
+  case X86::BI__builtin_ia32_sqrtss_round_mask: {
+    unsigned CC = cast<llvm::ConstantInt>(Ops[4])->getZExtValue();
+    // Support only if the rounding mode is 4 (AKA CUR_DIRECTION),
+    // otherwise keep the intrinsic.
+    if (CC != 4) {
+      Intrinsic::ID IID = BuiltinID == X86::BI__builtin_ia32_sqrtsd_round_mask ?
+                          Intrinsic::x86_avx512_mask_sqrt_sd :
+                          Intrinsic::x86_avx512_mask_sqrt_ss;
+      return Builder.CreateCall(CGM.getIntrinsic(IID), Ops);
+    }
+    Value *A = Builder.CreateExtractElement(Ops[0], (uint64_t)0);
+    Function *F = CGM.getIntrinsic(Intrinsic::sqrt, A->getType());
+    Value *Src = Builder.CreateExtractElement(Ops[2], (uint64_t)0);
+    int MaskSize = Ops[3]->getType()->getScalarSizeInBits();
+    llvm::Type *MaskTy = llvm::VectorType::get(Builder.getInt1Ty(), MaskSize);
+    Value *Mask = Builder.CreateBitCast(Ops[3], MaskTy);
+    Mask = Builder.CreateExtractElement(Mask, (uint64_t)0);
+    A = Builder.CreateSelect(Mask, Builder.CreateCall(F, {A}), Src);
+    return Builder.CreateInsertElement(Ops[1], A, (uint64_t)0);
+  }
+  case X86::BI__builtin_ia32_sqrtpd256:
+  case X86::BI__builtin_ia32_sqrtpd:
+  case X86::BI__builtin_ia32_sqrtps256:
+  case X86::BI__builtin_ia32_sqrtps: {
+    Function *F = CGM.getIntrinsic(Intrinsic::sqrt, Ops[0]->getType());
+    return Builder.CreateCall(F, {Ops[0]});
+  }
+  case X86::BI__builtin_ia32_sqrtps512_mask:
+  case X86::BI__builtin_ia32_sqrtpd512_mask: {
+    unsigned CC = cast<llvm::ConstantInt>(Ops[3])->getZExtValue();
+    // Support only if the rounding mode is 4 (AKA CUR_DIRECTION),
+    // otherwise keep the intrinsic.
+    if (CC != 4) {
+      Intrinsic::ID IID = BuiltinID == X86::BI__builtin_ia32_sqrtps512_mask ?
+                          Intrinsic::x86_avx512_mask_sqrt_ps_512 :
+                          Intrinsic::x86_avx512_mask_sqrt_pd_512;
+      return Builder.CreateCall(CGM.getIntrinsic(IID), Ops);
+    }
+    Function *F = CGM.getIntrinsic(Intrinsic::sqrt, Ops[0]->getType());
+    return EmitX86Select(*this, Ops[2], Builder.CreateCall(F, {Ops[0]}),
+                         Ops[1]);
+  }
   case X86::BI__builtin_ia32_pabsb128:
   case X86::BI__builtin_ia32_pabsw128:
   case X86::BI__builtin_ia32_pabsd128: