Fix typo in a SHUFPD and VSHUFPD pattern that prevented SHUFPD/VSHUFPD with a load from being selected.

llvm-svn: 147392

2 files changed, 36 insertions, 4 deletions

diff --git a/llvm/lib/Target/X86/X86InstrSSE.td b/llvm/lib/Target/X86/X86InstrSSE.td
index ec6957b472b..911150347f0 100644
--- a/llvm/lib/Target/X86/X86InstrSSE.td
+++ b/llvm/lib/Target/X86/X86InstrSSE.td
@@ -2348,7 +2348,7 @@ let Predicates = [HasSSE2] in {
             (SHUFPDrri VR128:$src1, VR128:$src2,
                        (SHUFFLE_get_shuf_imm VR128:$src3))>;
   // Generic SHUFPD patterns
-  def : Pat<(v2f64 (X86Shufps VR128:$src1,
+  def : Pat<(v2f64 (X86Shufpd VR128:$src1,
                        (memopv2f64 addr:$src2), (i8 imm:$imm))),
             (SHUFPDrmi VR128:$src1, addr:$src2, imm:$imm)>;
   def : Pat<(v2i64 (X86Shufpd VR128:$src1, VR128:$src2, (i8 imm:$imm))),
@@ -2397,7 +2397,7 @@ let Predicates = [HasAVX] in {
             (VSHUFPDrri VR128:$src1, VR128:$src2,
                         (SHUFFLE_get_shuf_imm VR128:$src3))>;
 
-  def : Pat<(v2f64 (X86Shufps VR128:$src1,
+  def : Pat<(v2f64 (X86Shufpd VR128:$src1,
                        (memopv2f64 addr:$src2), (i8 imm:$imm))),
             (VSHUFPDrmi VR128:$src1, addr:$src2, imm:$imm)>;
   def : Pat<(v2i64 (X86Shufpd VR128:$src1, VR128:$src2, (i8 imm:$imm))),
diff --git a/llvm/test/CodeGen/X86/avx-vshufp.ll b/llvm/test/CodeGen/X86/avx-vshufp.ll
index 0ccbc594a0d..a60d8c25dc5 100644
--- a/llvm/test/CodeGen/X86/avx-vshufp.ll
+++ b/llvm/test/CodeGen/X86/avx-vshufp.ll
@@ -7,7 +7,7 @@ entry:
   ret <8 x float> %shuffle
 }
 
-; CHECK: vshufps  $-53, (%
+; CHECK: vshufps  $-53, (%{{.*}}), %ymm
 define <8 x float> @A2(<8 x float>* %a, <8 x float>* %b) nounwind uwtable readnone ssp {
 entry:
   %a2 = load <8 x float>* %a
@@ -23,7 +23,7 @@ entry:
   ret <4 x double> %shuffle
 }
 
-; CHECK: vshufpd  $10, (%
+; CHECK: vshufpd  $10, (%{{.*}}), %ymm
 define <4 x double> @B2(<4 x double>* %a, <4 x double>* %b) nounwind uwtable readnone ssp {
 entry:
   %a2 = load <4 x double>* %a
@@ -59,3 +59,35 @@ entry:
   %shuffle = shufflevector <4 x double> %a, <4 x double> %b, <4 x i32> <i32 0, i32 4, i32 2, i32 7>
   ret <4 x double> %shuffle
 }
+
+; CHECK: vshufps  $-53, %xmm
+define <4 x float> @A128(<4 x float> %a, <4 x float> %b) nounwind uwtable readnone ssp {
+entry:
+  %shuffle = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 3, i32 2, i32 4, i32 7>
+  ret <4 x float> %shuffle
+}
+
+; CHECK: vshufps  $-53, (%{{.*}}), %xmm
+define <4 x float> @A2128(<4 x float>* %a, <4 x float>* %b) nounwind uwtable readnone ssp {
+entry:
+  %a2 = load <4 x float>* %a
+  %b2 = load <4 x float>* %b
+  %shuffle = shufflevector <4 x float> %a2, <4 x float> %b2, <4 x i32> <i32 3, i32 2, i32 4, i32 7>
+  ret <4 x float> %shuffle
+}
+
+; CHECK: vshufpd  $1, %xmm
+define <2 x double> @B128(<2 x double> %a, <2 x double> %b) nounwind uwtable readnone ssp {
+entry:
+  %shuffle = shufflevector <2 x double> %a, <2 x double> %b, <2 x i32> <i32 1, i32 2>
+  ret <2 x double> %shuffle
+}
+
+; CHECK: vshufpd  $1, (%{{.*}}), %xmm
+define <2 x double> @B2128(<2 x double>* %a, <2 x double>* %b) nounwind uwtable readnone ssp {
+entry:
+  %a2 = load <2 x double>* %a
+  %b2 = load <2 x double>* %b
+  %shuffle = shufflevector <2 x double> %a2, <2 x double> %b2, <2 x i32> <i32 1, i32 2>
+  ret <2 x double> %shuffle
+}