From: Sanjay Patel <spatel@rotateright.com>
Date: Thu, 10 Jan 2019 17:02:55 +0000 (+0000)
Subject: [Docs] fix typo, adjust text order
X-Git-Url: https://granicus.if.org/sourcecode?a=commitdiff_plain;h=337d71157f684c38bc2c9b1bb3e63837a9319cd8;p=llvm

[Docs] fix typo, adjust text order


git-svn-id: https://llvm.org/svn/llvm-project/llvm/trunk@350846 91177308-0d34-0410-b5e6-96231b3b80d8
---

diff --git a/docs/Vectorizers.rst b/docs/Vectorizers.rst
index 4a015800741..8ac7ae0847d 100644
--- a/docs/Vectorizers.rst
+++ b/docs/Vectorizers.rst
@@ -311,7 +311,7 @@ ignored (as other compilers do) are still being left un-vectorized.
 Vectorization of function calls
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
 
-The Loop Vectorize can vectorize intrinsic math functions.
+The Loop Vectorizer can vectorize intrinsic math functions.
 See the table below for a list of these functions.
 
 +-----+-----+---------+
@@ -328,6 +328,11 @@ See the table below for a list of these functions.
 |     |     | fmuladd |
 +-----+-----+---------+
 
+Note that the optimizer may not be able to vectorize math library functions 
+that correspond to these intrinsics if the library calls access external state 
+such as "errno". To allow better optimization of C/C++ math library functions, 
+use "-fno-math-errno".
+
 The loop vectorizer knows about special instructions on the target and will
 vectorize a loop containing a function call that maps to the instructions. For
 example, the loop below will be vectorized on Intel x86 if the SSE4.1 roundps
@@ -340,10 +345,6 @@ instruction is available.
       f[i] = floorf(f[i]);
   }
 
-Note that the optimizer may not be able to vectorize math library functions 
-that access external state such as "errno". To allow better optimization of 
-C/C++ math library functions, use "-fno-math-errno".
-
 Partial unrolling during vectorization
 ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^