[PATCH] fix configure check for powerpc decimal float hardware

Janis Johnson janis187@us.ibm.com
Tue Mar 27 22:47:00 GMT 2007


The configure check for powerpc*-linux assembler support of decimal
float instructions is wrong.  This fixes it.  It's obvious, so I've
checked it in.

The previous version of configure was generated by autoconf 2.61, not
2.59, so I regenerated and checked it in without my changes and then
with my change.

2007-03-27  Janis Johnson  <janis187@us.ibm.com>

	* configure.ac: Fix assembler test for powerpc*-linux decimal float.
	* configure: Regenerate.

Index: gcc/configure.ac
===================================================================
--- gcc/configure.ac	(revision 123252)
+++ gcc/configure.ac	(working copy)
@@ -2992,10 +2992,10 @@
     case $target in
       *-*-aix*) conftest_s='	.machine "pwr6"
 	.csect .text[[PR]]
-	dadd 1,3';;
+	dadd 1,2,3';;
       *) conftest_s='	.machine power6
 	.text
-	dadd 1,3';;
+	dadd 1,2,3';;
     esac
 
     gcc_GAS_CHECK_FEATURE([decimal float support],



More information about the Gcc-patches mailing list