From: H.J. Lu <hongjiu.lu@intel.com>
Date: Tue, 2 Oct 2012 21:24:45 +0000 (+0000)
Subject: Document -mprefer-avx128
X-Git-Tag: releases/gcc-4.6.4~341
X-Git-Url: http://git.ipfire.org/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=d35d19ca1dfa8b3b1312127d5744971f3c8f3c87;p=thirdparty%2Fgcc.git

Document -mprefer-avx128

	Backported from mainline
	PR target/54785
	* doc/invoke.texi: Document -mprefer-avx128.

From-SVN: r192009
---

diff --git a/gcc/ChangeLog b/gcc/ChangeLog
index 3bdfda68265a..a47020082128 100644
--- a/gcc/ChangeLog
+++ b/gcc/ChangeLog
@@ -1,3 +1,11 @@
+2012-10-02  H.J. Lu  <hongjiu.lu@intel.com>
+
+	Backported from mainline
+	2012-10-02  H.J. Lu  <hongjiu.lu@intel.com>
+
+	PR target/54785
+	* doc/invoke.texi: Document -mprefer-avx128.
+
 2012-10-02  H.J. Lu  <hongjiu.lu@intel.com>
 
 	Backported from mainline
diff --git a/gcc/doc/invoke.texi b/gcc/doc/invoke.texi
index 45bfec87aceb..4e56044d8fc5 100644
--- a/gcc/doc/invoke.texi
+++ b/gcc/doc/invoke.texi
@@ -598,7 +598,8 @@ Objective-C and Objective-C++ Dialects}.
 -mno-wide-multiply  -mrtd  -malign-double @gol
 -mpreferred-stack-boundary=@var{num}
 -mincoming-stack-boundary=@var{num} @gol
--mcld -mcx16 -msahf -mmovbe -mcrc32 -mrecip -mvzeroupper @gol
+-mcld -mcx16 -msahf -mmovbe -mcrc32 -mrecip @gol
+-mvzeroupper -mprefer-avx128 @gol
 -mmmx  -msse  -msse2 -msse3 -mssse3 -msse4.1 -msse4.2 -msse4 -mavx @gol
 -maes -mpclmul -mfsgsbase -mrdrnd -mf16c -mfused-madd @gol
 -msse4a -m3dnow -mpopcnt -mabm -mbmi -mtbm -mfma4 -mxop -mlwp @gol
@@ -12705,6 +12706,11 @@ before a transfer of control flow out of the function to minimize
 AVX to SSE transition penalty as well as remove unnecessary zeroupper 
 intrinsics.
 
+@item -mprefer-avx128
+@opindex mprefer-avx128
+This option instructs GCC to use 128-bit AVX instructions instead of
+256-bit AVX instructions in the auto-vectorizer.
+
 @item -mcx16
 @opindex mcx16
 This option will enable GCC to use CMPXCHG16B instruction in generated code.