From ba3ba8fa1ff843ce5cd27239e225ae0f194e7a99 Mon Sep 17 00:00:00 2001
From: Chris Lattner <sabre@nondot.org>
Date: Mon, 3 Sep 2012 02:58:21 +0000
Subject: some peepholes that should match horizontal add/sub operations.

llvm-svn: 163103
---
 llvm/lib/Target/X86/README-SSE.txt | 12 ++++++++++++
 1 file changed, 12 insertions(+)

(limited to 'llvm/lib/Target')

diff --git a/llvm/lib/Target/X86/README-SSE.txt b/llvm/lib/Target/X86/README-SSE.txt
index 624e56fa0f6..40110353fc6 100644
--- a/llvm/lib/Target/X86/README-SSE.txt
+++ b/llvm/lib/Target/X86/README-SSE.txt
@@ -941,3 +941,15 @@ and inversion with an rsqrtss instruction, which computes 1/sqrt faster at the
 cost of reduced accuracy.
 
 //===---------------------------------------------------------------------===//
+
+This function should be matched to haddpd when the appropriate CPU is enabled:
+
+#include <x86intrin.h>
+double f (__m128d p) {
+  return p[0] + p[1];
+}
+
+similarly, v[0]-v[1] should match to hsubpd, and {v[0]-v[1], w[0]-w[1]} should
+turn into hsubpd also.
+
+//===---------------------------------------------------------------------===//
-- 
cgit v1.2.3