Merging r195424:

------------------------------------------------------------------------
r195424 | haoliu | 2013-11-22 00:47:22 -0800 (Fri, 22 Nov 2013) | 4 lines

Fix the bugs about AArch64 Load/Store vector types and bitcast between i64 and vector types.
e.g. "%tmp = load <2 x i64>* %ptr" can't be selected. 
     "%tmp = bitcast i64 %in to <2 x i32>" can't be selected.

------------------------------------------------------------------------


git-svn-id: https://llvm.org/svn/llvm-project/llvm/branches/release_34@195764 91177308-0d34-0410-b5e6-96231b3b80d8

2 files changed, 155 insertions, 0 deletions

diff --git a/test/CodeGen/AArch64/neon-copy.ll b/test/CodeGen/AArch64/neon-copy.ll
index e1afc24..e18530e 100644
--- a/test/CodeGen/AArch64/neon-copy.ll
+++ b/test/CodeGen/AArch64/neon-copy.ll
@@ -530,3 +530,86 @@ define <2 x i64> @test_vdupq_laneq_s64(<2 x i64> %v1) #0 {
   ret <2 x i64> %shuffle
 }
 
+define i64 @test_bitcastv8i8toi64(<8 x i8> %in) {
+; CHECK-LABEL: test_bitcastv8i8toi64:
+   %res = bitcast <8 x i8> %in to i64
+; CHECK: fmov {{x[0-9]+}}, {{d[0-9]+}}
+   ret i64 %res
+}
+
+define i64 @test_bitcastv4i16toi64(<4 x i16> %in) {
+; CHECK-LABEL: test_bitcastv4i16toi64:
+   %res = bitcast <4 x i16> %in to i64
+; CHECK: fmov {{x[0-9]+}}, {{d[0-9]+}}
+   ret i64 %res
+}
+
+define i64 @test_bitcastv2i32toi64(<2 x i32> %in) {
+; CHECK-LABEL: test_bitcastv2i32toi64:
+   %res = bitcast <2 x i32> %in to i64
+; CHECK: fmov {{x[0-9]+}}, {{d[0-9]+}}
+   ret i64 %res
+}
+
+define i64 @test_bitcastv2f32toi64(<2 x float> %in) {
+; CHECK-LABEL: test_bitcastv2f32toi64:
+   %res = bitcast <2 x float> %in to i64
+; CHECK: fmov {{x[0-9]+}}, {{d[0-9]+}}
+   ret i64 %res
+}
+
+define i64 @test_bitcastv1i64toi64(<1 x i64> %in) {
+; CHECK-LABEL: test_bitcastv1i64toi64:
+   %res = bitcast <1 x i64> %in to i64
+; CHECK: fmov {{x[0-9]+}}, {{d[0-9]+}}
+   ret i64 %res
+}
+
+define i64 @test_bitcastv1f64toi64(<1 x double> %in) {
+; CHECK-LABEL: test_bitcastv1f64toi64:
+   %res = bitcast <1 x double> %in to i64
+; CHECK: fmov {{x[0-9]+}}, {{d[0-9]+}}
+   ret i64 %res
+}
+
+define <8 x i8> @test_bitcasti64tov8i8(i64 %in) {
+; CHECK-LABEL: test_bitcasti64tov8i8:
+   %res = bitcast i64 %in to <8 x i8>
+; CHECK: fmov {{d[0-9]+}}, {{x[0-9]+}}
+   ret <8 x i8> %res
+}
+
+define <4 x i16> @test_bitcasti64tov4i16(i64 %in) {
+; CHECK-LABEL: test_bitcasti64tov4i16:
+   %res = bitcast i64 %in to <4 x i16>
+; CHECK: fmov {{d[0-9]+}}, {{x[0-9]+}}
+   ret <4 x i16> %res
+}
+
+define <2 x i32> @test_bitcasti64tov2i32(i64 %in) {
+; CHECK-LABEL: test_bitcasti64tov2i32:
+   %res = bitcast i64 %in to <2 x i32>
+; CHECK: fmov {{d[0-9]+}}, {{x[0-9]+}}
+   ret <2 x i32> %res
+}
+
+define <2 x float> @test_bitcasti64tov2f32(i64 %in) {
+; CHECK-LABEL: test_bitcasti64tov2f32:
+   %res = bitcast i64 %in to <2 x float>
+; CHECK: fmov {{d[0-9]+}}, {{x[0-9]+}}
+   ret <2 x float> %res
+}
+
+define <1 x i64> @test_bitcasti64tov1i64(i64 %in) {
+; CHECK-LABEL: test_bitcasti64tov1i64:
+   %res = bitcast i64 %in to <1 x i64>
+; CHECK: fmov {{d[0-9]+}}, {{x[0-9]+}}
+   ret <1 x i64> %res
+}
+
+define <1 x double> @test_bitcasti64tov1f64(i64 %in) {
+; CHECK-LABEL: test_bitcasti64tov1f64:
+   %res = bitcast i64 %in to <1 x double>
+; CHECK: fmov {{d[0-9]+}}, {{x[0-9]+}}
+   ret <1 x double> %res
+}
\ No newline at end of file
diff --git a/test/CodeGen/AArch64/neon-simd-ldst-multi-elem.ll b/test/CodeGen/AArch64/neon-simd-ldst-multi-elem.ll
index d0e7fc1..d5557c0 100644
--- a/test/CodeGen/AArch64/neon-simd-ldst-multi-elem.ll
+++ b/test/CodeGen/AArch64/neon-simd-ldst-multi-elem.ll
@@ -1,5 +1,77 @@
 ; RUN: llc < %s -verify-machineinstrs -mtriple=aarch64-none-linux-gnu -mattr=+neon | FileCheck %s
 
+define void @test_ldst1_v16i8(<16 x i8>* %ptr, <16 x i8>* %ptr2) {
+; CHECK-LABEL: test_ldst1_v16i8:
+; CHECK: ld1 {v{{[0-9]+}}.16b}, [x{{[0-9]+|sp}}]
+; CHECK: st1 {v{{[0-9]+}}.16b}, [x{{[0-9]+|sp}}]
+  %tmp = load <16 x i8>* %ptr
+  store <16 x i8> %tmp, <16 x i8>* %ptr2
+  ret void
+}
+
+define void @test_ldst1_v8i16(<8 x i16>* %ptr, <8 x i16>* %ptr2) {
+; CHECK-LABEL: test_ldst1_v8i16:
+; CHECK: ld1 {v{{[0-9]+}}.8h}, [x{{[0-9]+|sp}}]
+; CHECK: st1 {v{{[0-9]+}}.8h}, [x{{[0-9]+|sp}}]
+  %tmp = load <8 x i16>* %ptr
+  store <8 x i16> %tmp, <8 x i16>* %ptr2
+  ret void
+}
+
+define void @test_ldst1_v4i32(<4 x i32>* %ptr, <4 x i32>* %ptr2) {
+; CHECK-LABEL: test_ldst1_v4i32:
+; CHECK: ld1 {v{{[0-9]+}}.4s}, [x{{[0-9]+|sp}}]
+; CHECK: st1 {v{{[0-9]+}}.4s}, [x{{[0-9]+|sp}}]
+  %tmp = load <4 x i32>* %ptr
+  store <4 x i32> %tmp, <4 x i32>* %ptr2
+  ret void
+}
+
+define void @test_ldst1_v2i64(<2 x i64>* %ptr, <2 x i64>* %ptr2) {
+; CHECK-LABEL: test_ldst1_v2i64:
+; CHECK: ld1 {v{{[0-9]+}}.2d}, [x{{[0-9]+|sp}}]
+; CHECK: st1 {v{{[0-9]+}}.2d}, [x{{[0-9]+|sp}}]
+  %tmp = load <2 x i64>* %ptr
+  store <2 x i64> %tmp, <2 x i64>* %ptr2
+  ret void
+}
+
+define void @test_ldst1_v8i8(<8 x i8>* %ptr, <8 x i8>* %ptr2) {
+; CHECK-LABEL: test_ldst1_v8i8:
+; CHECK: ld1 {v{{[0-9]+}}.8b}, [x{{[0-9]+|sp}}]
+; CHECK: st1 {v{{[0-9]+}}.8b}, [x{{[0-9]+|sp}}]
+  %tmp = load <8 x i8>* %ptr
+  store <8 x i8> %tmp, <8 x i8>* %ptr2
+  ret void
+}
+
+define void @test_ldst1_v4i16(<4 x i16>* %ptr, <4 x i16>* %ptr2) {
+; CHECK-LABEL: test_ldst1_v4i16:
+; CHECK: ld1 {v{{[0-9]+}}.4h}, [x{{[0-9]+|sp}}]
+; CHECK: st1 {v{{[0-9]+}}.4h}, [x{{[0-9]+|sp}}]
+  %tmp = load <4 x i16>* %ptr
+  store <4 x i16> %tmp, <4 x i16>* %ptr2
+  ret void
+}
+
+define void @test_ldst1_v2i32(<2 x i32>* %ptr, <2 x i32>* %ptr2) {
+; CHECK-LABEL: test_ldst1_v2i32:
+; CHECK: ld1 {v{{[0-9]+}}.2s}, [x{{[0-9]+|sp}}]
+; CHECK: st1 {v{{[0-9]+}}.2s}, [x{{[0-9]+|sp}}]
+  %tmp = load <2 x i32>* %ptr
+  store <2 x i32> %tmp, <2 x i32>* %ptr2
+  ret void
+}
+
+define void @test_ldst1_v1i64(<1 x i64>* %ptr, <1 x i64>* %ptr2) {
+; CHECK-LABEL: test_ldst1_v1i64:
+; CHECK: ld1 {v{{[0-9]+}}.1d}, [x{{[0-9]+|sp}}]
+; CHECK: st1 {v{{[0-9]+}}.1d}, [x{{[0-9]+|sp}}]
+  %tmp = load <1 x i64>* %ptr
+  store <1 x i64> %tmp, <1 x i64>* %ptr2
+  ret void
+}
+
 %struct.int8x16x2_t = type { [2 x <16 x i8>] }
 %struct.int16x8x2_t = type { [2 x <8 x i16>] }
 %struct.int32x4x2_t = type { [2 x <4 x i32>] }