[x86] Teach the target-specific combining how to aggressively fold

half-shuffles, even looking through intervening instructions in a chain.

Summary:
This doesn't happen to show up with any test cases I've found for the current
shuffle lowering, but previous attempts would benefit from this and it seems
generally useful. I've tested it directly using intrinsics, which also shows
that it will work with hand vectorized code as well.

Note that even though pshufd isn't directly used in these tests, it gets
exercised because we combine some of the half shuffles into a pshufd
first, and then merge them.

Differential Revision: http://reviews.llvm.org/D4291

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@211890 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 49 insertions, 0 deletions

diff --git a/test/CodeGen/X86/vector-shuffle-combining.ll b/test/CodeGen/X86/vector-shuffle-combining.ll
new file mode 100644
index 0000000000..dae1ef5875
--- /dev/null
+++ b/test/CodeGen/X86/vector-shuffle-combining.ll
@@ -0,0 +1,49 @@
+; RUN: llc < %s -mtriple=x86_64-unknown-unknown -mcpu=x86-64 -x86-experimental-vector-shuffle-lowering | FileCheck %s --check-prefix=CHECK-SSE2
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-unknown"
+
+declare <8 x i16> @llvm.x86.sse2.pshufl.w(<8 x i16>, i8)
+declare <8 x i16> @llvm.x86.sse2.pshufh.w(<8 x i16>, i8)
+
+define <8 x i16> @combine_pshuflw1(<8 x i16> %a) {
+; CHECK-SSE2-LABEL: @combine_pshuflw1
+; CHECK-SSE2:       # BB#0:
+; CHECK-SSE2-NEXT:    retq
+  %b = call <8 x i16> @llvm.x86.sse2.pshufl.w(<8 x i16> %a, i8 27) 
+  %c = call <8 x i16> @llvm.x86.sse2.pshufl.w(<8 x i16> %b, i8 27) 
+  ret <8 x i16> %c
+}
+
+define <8 x i16> @combine_pshuflw2(<8 x i16> %a) {
+; CHECK-SSE2-LABEL: @combine_pshuflw2
+; CHECK-SSE2:       # BB#0:
+; CHECK-SSE2-NEXT:    retq
+  %b = call <8 x i16> @llvm.x86.sse2.pshufl.w(<8 x i16> %a, i8 27)
+  %c = call <8 x i16> @llvm.x86.sse2.pshufh.w(<8 x i16> %b, i8 -28) 
+  %d = call <8 x i16> @llvm.x86.sse2.pshufl.w(<8 x i16> %c, i8 27) 
+  ret <8 x i16> %d
+}
+
+define <8 x i16> @combine_pshuflw3(<8 x i16> %a) {
+; CHECK-SSE2-LABEL: @combine_pshuflw3
+; CHECK-SSE2:       # BB#0:
+; CHECK-SSE2-NEXT:    pshufhw {{.*}} # xmm0 = xmm0[0,1,2,3,7,6,5,4]
+; CHECK-SSE2-NEXT:    retq
+  %b = call <8 x i16> @llvm.x86.sse2.pshufl.w(<8 x i16> %a, i8 27)
+  %c = call <8 x i16> @llvm.x86.sse2.pshufh.w(<8 x i16> %b, i8 27) 
+  %d = call <8 x i16> @llvm.x86.sse2.pshufl.w(<8 x i16> %c, i8 27) 
+  ret <8 x i16> %d
+}
+
+define <8 x i16> @combine_pshufhw1(<8 x i16> %a) {
+; CHECK-SSE2-LABEL: @combine_pshufhw1
+; CHECK-SSE2:       # BB#0:
+; CHECK-SSE2-NEXT:    pshuflw {{.*}} # xmm0 = xmm0[3,2,1,0,4,5,6,7]
+; CHECK-SSE2-NEXT:    retq
+  %b = call <8 x i16> @llvm.x86.sse2.pshufh.w(<8 x i16> %a, i8 27)
+  %c = call <8 x i16> @llvm.x86.sse2.pshufl.w(<8 x i16> %b, i8 27) 
+  %d = call <8 x i16> @llvm.x86.sse2.pshufh.w(<8 x i16> %c, i8 27) 
+  ret <8 x i16> %d
+}
+