This is the mail archive of the
binutils@sources.redhat.com
mailing list for the binutils project.
A symble and absolute vaule problem in the objdump
- From: Haizhi Xu <hxu02 at ecs dot syr dot edu>
- To: binutils at sources dot redhat dot com
- Date: Tue, 29 Oct 2002 16:35:39 -0500 (EST)
- Subject: A symble and absolute vaule problem in the objdump
Hi, friends:
I got a headache problem from objdump. I have an assembly file, which contain
the following code. Then I assembled it using gcc and get the .o object file.
When applying the objdump, the code turns out to be weired to me. The lable
address is changed to an absolute address and the function address is changed to
an absolute address too, which is the MIDDLE of the call instruction! I tried it
on gdb and got the same result.
My question is: From the output of the objdump, how can I identify the symbles
and how can I restore them? My goal is kind of reverse engineering: to
disassemble an object file, and make it compilable again.
Thanks a lot.
Hai
################################################ original .s file
.LC4:
.string "Installing hooks ....\n"
... ...
pushl $.LC4
call printk_Rsmp_1b7d4074
################################################
############################################ disassembled code using objdump
# b31: 68 24 00 00 00 push $0x24
push $0x24
# b36: e8 fc ff ff ff call b37 <init_module+0x63>
call init_module+0x63
############################################