[X86] Custom emit __builtin_rdtscp so we can emit an explicit store for the out parameter

[craig.topper]

Summary

This is the clang side of D51803. The llvm intrinsic now returns two results. So we need to emit an explicit store in IR for the out parameter. This is similar to addcarry/subborrow/rdrand/rdseed.

[RKSimon]

LGTM - The lack of existing tests is worrying.....

cfe/trunk/lib/CodeGen/CGBuiltin.cpp

case X86::BI_mm_sfence: {
     return Builder.CreateCall(CGM.getIntrinsic(Intrinsic::x86_sse_sfence));
   }
   case X86::BI_mm_pause: {
     return Builder.CreateCall(CGM.getIntrinsic(Intrinsic::x86_sse2_pause));
   }
   case X86::BI__rdtsc: {
     return Builder.CreateCall(CGM.getIntrinsic(Intrinsic::x86_rdtsc));
   }
+  case X86::BI__builtin_ia32_rdtscp: {
+    Value *Call = Builder.CreateCall(CGM.getIntrinsic(Intrinsic::x86_rdtscp));
+    Builder.CreateDefaultAlignedStore(Builder.CreateExtractValue(Call, 1),
+                                      Ops[0]);
+    return Builder.CreateExtractValue(Call, 0);
+  }
   case X86::BI__builtin_ia32_undef128:
   case X86::BI__builtin_ia32_undef256:
   case X86::BI__builtin_ia32_undef512:
     // The x86 definition of "undef" is not the same as the LLVM definition
     // (PR32176). We leave optimizing away an unnecessary zero constant to the
     // IR optimizer and backend.
     // TODO: If we had a "freeze" IR instruction to generate a fixed undef
     // value, we should use that here instead of a zero.

cfe/trunk/test/CodeGen/rd-builtins.c

 // RUN: %clang_cc1 -ffreestanding -triple x86_64-unknown-unknown -emit-llvm -o - %s | FileCheck %s
 
 
 #include <x86intrin.h>
 
 unsigned long long test_rdpmc(int a) {
   return _rdpmc(a);
 // CHECK: @test_rdpmc
 // CHECK: call i64 @llvm.x86.rdpmc
 }
 
 int test_rdtsc() {
   return _rdtsc();
 // CHECK: @test_rdtsc
 // CHECK: call i64 @llvm.x86.rdtsc
 }
+
+unsigned long long test_rdtscp(unsigned int *a) {
+// CHECK: @test_rdtscp
+// CHECK: [[RDTSCP:%.*]] = call { i64, i32 } @llvm.x86.rdtscp
+// CHECK: [[TSC_AUX:%.*]] = extractvalue { i64, i32 } [[RDTSCP]], 1
+// CHECK: store i32 [[TSC_AUX]], i32* %{{.*}}
+// CHECK: [[TSC:%.*]] = extractvalue { i64, i32 } [[RDTSCP]], 0
+  return __rdtscp(a);
+}