Look at shuffles of build_vectors in DAGCombiner::visitEXTRACT_VECTOR_ELT

When the loop vectorizer vectorizes code that uses the loop induction variable,
we often end up with IR like this:

  %b1 = insertelement <2 x i32> undef, i32 %v, i32 0
  %b2 = shufflevector <2 x i32> %b1, <2 x i32> undef, <2 x i32> zeroinitializer
  %i = add <2 x i32> %b2, <i32 2, i32 3>

If the add in this example is not legal (as is the case on PPC with VSX), it
will be scalarized, and we'll end up with a number of extract_vector_elt nodes
with the vector shuffle as the input operand, and that vector shuffle is fed by
one or more build_vector nodes. By the time that vector operations are
expanded, visitEXTRACT_VECTOR_ELT will not create new extract_vector_elt by
looking through the vector shuffle (to make sure that no illegal operations are
created), and so the extract_vector_elt -> vector shuffle -> build_vector is
never simplified to an operand of the build vector.

By looking at build_vectors through a shuffle we fix this particular situation,
preventing a vector from being built, only to be deconstructed again (for the
scalarized add) -- an expensive proposition when this all needs to be done via
the stack. We probably want a more comprehensive fix here where we look back
recursively through any shuffles to any build_vectors or scalar_to_vectors,
etc. but that can come later.

git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@205179 91177308-0d34-0410-b5e6-96231b3b80d8

1 files changed, 16 insertions, 0 deletions

diff --git a/test/CodeGen/PowerPC/vsx.ll b/test/CodeGen/PowerPC/vsx.ll
index d0ff0f2829..a1710938ab 100644
--- a/test/CodeGen/PowerPC/vsx.ll
+++ b/test/CodeGen/PowerPC/vsx.ll
@@ -624,3 +624,19 @@ define <2 x double> @test70(<2 x i8> %a) {
 ; CHECK: blr
 }
 
+define <2 x i32> @test80(i32 %v) {
+  %b1 = insertelement <2 x i32> undef, i32 %v, i32 0
+  %b2 = shufflevector <2 x i32> %b1, <2 x i32> undef, <2 x i32> zeroinitializer
+  %i = add <2 x i32> %b2, <i32 2, i32 3>
+  ret <2 x i32> %i
+
+; CHECK-LABEL: @test80
+; CHECK: addi
+; CHECK: addi
+; CHECK: lxvd2x
+; CHECK-NOT: stxvd2x
+; FIXME: We still make one vector for each vector element and this shuffle them
+; together instead of just composing one vector on the stack.
+; CHECK: blr
+}
+