From 12da396abcedaae4e05c105b58ad0b2ee40a02b7 Mon Sep 17 00:00:00 2001 From: Nadav Rotem Date: Thu, 20 Dec 2012 00:03:36 +0000 Subject: Doc: update the chart. llvm-svn: 170618 --- llvm/docs/Vectorizers.rst | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'llvm/docs/Vectorizers.rst') diff --git a/llvm/docs/Vectorizers.rst b/llvm/docs/Vectorizers.rst index 5324ea025ce..e49d44c1bb2 100644 --- a/llvm/docs/Vectorizers.rst +++ b/llvm/docs/Vectorizers.rst @@ -203,8 +203,8 @@ This section shows the the execution time of Clang on a simple benchmark: This benchmarks is a collection of loops from the GCC autovectorization `page `_ by Dorit Nuzman. -The chart below compares GCC-4.7, ICC-13, and Clang-SVN at -O3, running on a Sandybridge. -The Y-axis shows time in msec. Lower is better. +The chart below compares GCC-4.7, ICC-13, and Clang-SVN with and without loop vectorization at -O3, tuned for "corei7-avx", running on a Sandybridge iMac. +The Y-axis shows the time in msec. Lower is better. The last column shows the geomean of all the kernels. .. image:: gcc-loops.png -- cgit v1.2.3