From b80f1147072cf4a30d9907e412174f5eb759a9f1 Mon Sep 17 00:00:00 2001 From: Chris Lattner Date: Sun, 2 Apr 2006 03:59:11 +0000 Subject: [PATCH] add a note llvm-svn: 27348 --- llvm/lib/Target/PowerPC/README_ALTIVEC.txt | 11 +++++++++++ 1 file changed, 11 insertions(+) diff --git a/llvm/lib/Target/PowerPC/README_ALTIVEC.txt b/llvm/lib/Target/PowerPC/README_ALTIVEC.txt index 824dc6cafcf1..881ba020f42e 100644 --- a/llvm/lib/Target/PowerPC/README_ALTIVEC.txt +++ b/llvm/lib/Target/PowerPC/README_ALTIVEC.txt @@ -137,3 +137,14 @@ Instcombine llvm.ppc.altivec.vperm with an immediate into a shuffle operation. Handle VECTOR_SHUFFLE nodes with the appropriate shuffle mask with vsldoi, vpkuhum and vpkuwum. + +//===----------------------------------------------------------------------===// + +Implement multiply for vector integer types, to avoid the horrible scalarized +code produced by legalize. + +void test(vector int *X, vector int *Y) { + *X = *X * *Y; +} + +//===----------------------------------------------------------------------===//