[PowerPC] Handle __builtin_xxpermdi the same way as GCC does

The codegen for this builtin was initially implemented to match GCC.
However, due to interest from users GCC changed behaviour to account for the
big endian bias of the instruction and correct it. This patch brings the
handling inline with GCC.

Fixes https://bugs.llvm.org/show_bug.cgi?id=38192

Differential Revision: https://reviews.llvm.org/D49424

git-svn-id: https://llvm.org/svn/llvm-project/cfe/trunk@337449 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/CodeGen/CGBuiltin.cpp b/lib/CodeGen/CGBuiltin.cpp
index 92091345f2a980ea4f3c97148a79d72c37113b30..7deeea811b767c9735ff13b617858d252d1873f6 100644 (file)
--- a/lib/CodeGen/CGBuiltin.cpp
+++ b/lib/CodeGen/CGBuiltin.cpp
@@ -10831,19 +10831,11 @@ Value *CodeGenFunction::EmitPPCBuiltinExpr(unsigned BuiltinID,
     Ops[0] = Builder.CreateBitCast(Ops[0], llvm::VectorType::get(Int64Ty, 2));
     Ops[1] = Builder.CreateBitCast(Ops[1], llvm::VectorType::get(Int64Ty, 2));
 
-    // Element zero comes from the first input vector and element one comes from
-    // the second. The element indices within each vector are numbered in big
-    // endian order so the shuffle mask must be adjusted for this on little
-    // endian platforms (i.e. index is complemented and source vector reversed).
-    unsigned ElemIdx0;
-    unsigned ElemIdx1;
-    if (getTarget().isLittleEndian()) {
-      ElemIdx0 = (~Index & 1) + 2;
-      ElemIdx1 = (~Index & 2) >> 1;
-    } else { // BigEndian
-      ElemIdx0 = (Index & 2) >> 1;
-      ElemIdx1 = 2 + (Index & 1);
-    }
+    // Account for endianness by treating this as just a shuffle. So we use the
+    // same indices for both LE and BE in order to produce expected results in
+    // both cases.
+    unsigned ElemIdx0 = (Index & 2) >> 1;;
+    unsigned ElemIdx1 = 2 + (Index & 1);;
 
     Constant *ShuffleElts[2] = {ConstantInt::get(Int32Ty, ElemIdx0),
                                 ConstantInt::get(Int32Ty, ElemIdx1)};

diff --git a/test/CodeGen/builtins-ppc-vsx.c b/test/CodeGen/builtins-ppc-vsx.c
index 6dda19d5d9f33aee393d9b4a69c83ff8ee749b5e..848d24d4fb10cf1c8ce279200727730caf990b96 100644 (file)
--- a/test/CodeGen/builtins-ppc-vsx.c
+++ b/test/CodeGen/builtins-ppc-vsx.c
@@ -1694,43 +1694,43 @@ vec_xst_be(vd, sll, ad);
 
 res_vd = vec_xxpermdi(vd, vd, 0);
 // CHECK: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 2>
-// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 3, i32 1>
+// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 2>
 
 res_vf = vec_xxpermdi(vf, vf, 1);
 // CHECK: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 3>
-// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 2, i32 1>
+// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 3>
 
 res_vsll = vec_xxpermdi(vsll, vsll, 2);
 // CHECK: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 1, i32 2>
-// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 3, i32 0>
+// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 1, i32 2>
 
 res_vull = vec_xxpermdi(vull, vull, 3);
 // CHECK: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 1, i32 3>
-// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 2, i32 0>
+// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 1, i32 3>
 
 res_vsi = vec_xxpermdi(vsi, vsi, 0);
 // CHECK: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 2>
-// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 3, i32 1>
+// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 2>
 
 res_vui = vec_xxpermdi(vui, vui, 1);
 // CHECK: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 3>
-// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 2, i32 1>
+// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 3>
 
 res_vss = vec_xxpermdi(vss, vss, 2);
 // CHECK: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 1, i32 2>
-// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 3, i32 0>
+// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 1, i32 2>
 
 res_vus = vec_xxpermdi(vus, vus, 3);
 // CHECK: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 1, i32 3>
-// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 2, i32 0>
+// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 1, i32 3>
 
 res_vsc = vec_xxpermdi(vsc, vsc, 0);
 // CHECK: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 2>
-// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 3, i32 1>
+// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 2>
 
 res_vuc = vec_xxpermdi(vuc, vuc, 1);
 // CHECK: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 3>
-// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 2, i32 1>
+// CHECK-LE: shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 3>
 
 res_vd = vec_xxsldwi(vd, vd, 0);
 // CHECK: shufflevector <4 x i32> %{{[0-9]+}}, <4 x i32> %{{[0-9]+}}, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
@@ -1786,7 +1786,7 @@ vector int xxpermdi_should_not_assert(vector int a, vector int b) {
 
 // CHECK-LE:  bitcast <4 x i32> %{{[0-9]+}} to <2 x i64>
 // CHECK-LE-NEXT:  bitcast <4 x i32> %{{[0-9]+}} to <2 x i64>
-// CHECK-LE-NEXT:  shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 3, i32 1>
+// CHECK-LE-NEXT:  shufflevector <2 x i64> %{{[0-9]+}}, <2 x i64> %{{[0-9]+}}, <2 x i32> <i32 0, i32 2>
 // CHECK-LE-NEXT:  bitcast <2 x i64> %{{[0-9]+}} to <4 x i32>
 }