[x86] Teach the x86 DAG combiner to form UNPCKLPS and UNPCKHPS

instructions from the relevant shuffle patterns.

This is the last tweak I'm aware of to generate essentially perfect
v4f32 and v2f64 shuffles with the new vector shuffle lowering up through
SSE4.1. I'm sure I've missed some and it'd be nice to check since v4f32
is amenable to exhaustive exploration, but this is all of the tricks I'm
aware of.

With AVX there is a new trick to use the VPERMILPS instruction, that's
coming up in a subsequent patch.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@217761 91177308-0d34-0410-b5e6-96231b3b80d8

diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index 04f1fafa2e7e4088b86056cd52af0f654f6cf631..da3ec8b35eb94091bed0bd00391cd98156fd9082 100644 (file)
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -19413,6 +19413,20 @@ static bool combineX86ShuffleChain(SDValue Op, SDValue Root, ArrayRef<int> Mask,
                     /*AddTo*/ true);
       return true;
     }
+    if (Mask.equals(0, 0, 1, 1) || Mask.equals(2, 2, 3, 3)) {
+      bool Lo = Mask.equals(0, 0, 1, 1);
+      unsigned Shuffle = Lo ? X86ISD::UNPCKL : X86ISD::UNPCKH;
+      MVT ShuffleVT = MVT::v4f32;
+      if (Depth == 1 && Root->getOpcode() == Shuffle)
+        return false; // Nothing to do!
+      Op = DAG.getNode(ISD::BITCAST, DL, ShuffleVT, Input);
+      DCI.AddToWorklist(Op.getNode());
+      Op = DAG.getNode(Shuffle, DL, ShuffleVT, Op, Op);
+      DCI.AddToWorklist(Op.getNode());
+      DCI.CombineTo(Root.getNode(), DAG.getNode(ISD::BITCAST, DL, RootVT, Op),
+                    /*AddTo*/ true);
+      return true;
+    }
   }
 
   // We always canonicalize the 8 x i16 and 16 x i8 shuffles into their UNPCK

diff --git a/test/CodeGen/X86/vector-shuffle-128-v4.ll b/test/CodeGen/X86/vector-shuffle-128-v4.ll
index dcd8ab166cfe569f946ab8cca9b7b8da6e40bb5d..601db25748a2fa100b5dd0b9701e3b25bc3df9e2 100644 (file)
--- a/test/CodeGen/X86/vector-shuffle-128-v4.ll
+++ b/test/CodeGen/X86/vector-shuffle-128-v4.ll
@@ -119,6 +119,20 @@ define <4 x float> @shuffle_v4f32_3210(<4 x float> %a, <4 x float> %b) {
   %shuffle = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 3, i32 2, i32 1, i32 0>
   ret <4 x float> %shuffle
 }
+define <4 x float> @shuffle_v4f32_0011(<4 x float> %a, <4 x float> %b) {
+; ALL-LABEL: @shuffle_v4f32_0011
+; ALL:         unpcklps {{.*}} # xmm0 = xmm0[0,0,1,1]
+; ALL-NEXT:    retq
+  %shuffle = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 0, i32 0, i32 1, i32 1>
+  ret <4 x float> %shuffle
+}
+define <4 x float> @shuffle_v4f32_2233(<4 x float> %a, <4 x float> %b) {
+; ALL-LABEL: @shuffle_v4f32_2233
+; ALL:         unpckhps {{.*}} # xmm0 = xmm0[2,2,3,3]
+; ALL-NEXT:    retq
+  %shuffle = shufflevector <4 x float> %a, <4 x float> %b, <4 x i32> <i32 2, i32 2, i32 3, i32 3>
+  ret <4 x float> %shuffle
+}
 define <4 x float> @shuffle_v4f32_0022(<4 x float> %a, <4 x float> %b) {
 ; SSE2-LABEL: @shuffle_v4f32_0022
 ; SSE2:         shufps {{.*}} # xmm0 = xmm0[0,0,2,2]