patches/gcc/4.0.4/200-arm-softfloat.patch
author "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
Tue Aug 31 23:24:40 2010 +0200 (2010-08-31)
changeset 2107 f5ab0a80e466
parent 305 00a7e6c275da
permissions -rw-r--r--
binutils/binutils: remove faulty patch

The added code should be conditinal to the target system
being !MIPS, but is based on the host system being !MIPS.

This is plain wrong, and had not been noticed until now
as I never used those binutils versions on MIPS.

See:
http://sourceware.org/ml/crossgcc/2010-08/msg00192.html

Signed-off-by: "Yann E. MORIN" <yann.morin.1998@anciens.enib.fr>
yann@305
     1
Enable building a pure soft-float compiler without the need for a software
yann@305
     2
floating point library.
yann@1
     3
yann@1
     4
diff -dur gcc-4.0.4.orig/gcc/config.gcc gcc-4.0.4/gcc/config.gcc
yann@1
     5
--- gcc-4.0.4.orig/gcc/config.gcc	2007-02-02 19:12:28.000000000 +0100
yann@1
     6
+++ gcc-4.0.4/gcc/config.gcc	2007-02-02 19:12:07.000000000 +0100
yann@1
     7
@@ -677,7 +677,7 @@
yann@1
     8
 		tm_defines="TARGET_BIG_ENDIAN_DEFAULT=1 $tm_defines"
yann@1
     9
 		;;
yann@1
    10
 	esac
yann@1
    11
-	tmake_file="${tmake_file} arm/t-arm arm/t-linux"
yann@1
    12
+	tmake_file="${tmake_file} arm/t-arm arm/t-linux arm/t-arm-elf"
yann@1
    13
 	extra_parts="crtbegin.o crtbeginS.o crtend.o crtendS.o"
yann@1
    14
 	gnu_ld=yes
yann@1
    15
 	;;