[PATCH] D49424: [PowerPC] Handle __builtin_xxpermdi the same way as GCC does
Nemanja Ivanovic via Phabricator via cfe-commits
cfe-commits at lists.llvm.org
Thu Jul 19 05:49:36 PDT 2018
This revision was automatically updated to reflect the committed changes.
Closed by commit rC337449: [PowerPC] Handle __builtin_xxpermdi the same way as GCC does (authored by nemanjai, committed by ).
Changed prior to commit:
https://reviews.llvm.org/D49424?vs=155869&id=156251#toc
Repository:
rC Clang
https://reviews.llvm.org/D49424
Files:
lib/CodeGen/CGBuiltin.cpp
test/CodeGen/builtins-ppc-vsx.c
Index: test/CodeGen/builtins-ppc-vsx.c
===================================================================
--- test/CodeGen/builtins-ppc-vsx.c
+++ test/CodeGen/builtins-ppc-vsx.c
@@ -1694,43 +1694,43 @@
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 @@
// 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>
}
Index: lib/CodeGen/CGBuiltin.cpp
===================================================================
--- lib/CodeGen/CGBuiltin.cpp
+++ lib/CodeGen/CGBuiltin.cpp
@@ -10831,19 +10831,11 @@
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)};
-------------- next part --------------
A non-text attachment was scrubbed...
Name: D49424.156251.patch
Type: text/x-patch
Size: 5459 bytes
Desc: not available
URL: <http://lists.llvm.org/pipermail/cfe-commits/attachments/20180719/a04bf567/attachment.bin>
More information about the cfe-commits
mailing list