Hi!
Not found this bug in bugzilla database.
My guess : bad handling of the $a, $t, $d symbols
Here is a trace on Debian testing (crosschain packages from emdebian.org)
Philippe WAILLE
bug trace
arm-linux-gnueabi-as --version
GNU assembler (GNU Binutils) 2.19.1
This assembler was configured for a target of `arm-linux-gnueabi'.
---> (same bug with arm-linux-gnu-objdump)
arm-linux-gnueabi-objdump --version
GNU objdump (GNU Binutils) 2.19.1
.data @ source file mytest.s
b: .word 3
.text
x: add r0, r1, r2
y: add r1, r2, r3@ no bug without label y
add r1, r2, r3
add r2, r3, r4
ptr_b: .word b
arm-linux-gnueabi-as mytest.s -o mytest.o
arm-linux-gnueabi-objdump -d mytest.o
mytest.o: file format elf32-littlearm
Disassembly of section .text:
:
0: e0810002add r0, r1, r2 @ OK
0004 :
4: e0821003.word 0xe0821003 @ BUG : should display add
8: e0821003.word 0xe0821003 @ BUG
c: e0832004.word 0xe0832004 @ BUG
0010 :
10: .word 0x @ OK
--ok on older release---
Old objdump compiled from tar.gz sources
/opt/gnu/arm/bin/arm-elf-objdump --version
GNU objdump 2.13.2.1
/opt/gnu/arm/bin/arm-elf-objdump -d mytest.o
mytest.o: format de fichier elf32-littlearm
Disassembly of section .text:
<$a>:
0: e0810002add r0, r1, r2
0004 :
4: e0821003add r1, r2, r3
8: e0821003add r1, r2, r3
c: e0832004add r2, r3, r4
0010 <$d>:
10: andeq r0, r0, r0
___
bug-binutils mailing list
bug-binutils@gnu.org
http://lists.gnu.org/mailman/listinfo/bug-binutils