[Top][All Lists]
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Bug ld/12001] Linker includes archive members when symbols therein have
From: |
nickc at redhat dot com |
Subject: |
[Bug ld/12001] Linker includes archive members when symbols therein have already been defined |
Date: |
Thu, 28 Oct 2010 13:47:54 +0000 |
http://sourceware.org/bugzilla/show_bug.cgi?id=12001
Nick Clifton <nickc at redhat dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
Status|NEW |ASSIGNED
CC| |nickc at redhat dot com
AssignedTo|unassigned at sources dot |nickc at redhat dot com
|redhat.com |
--- Comment #2 from Nick Clifton <nickc at redhat dot com> 2010-10-28 13:47:43
UTC ---
Hi Mark,
I have uploaded a possible patch to LD for this problem. I assume that a
similar patch to GOLD will also be possible, but I have not looked into this
yet.
In the course of checking the patch I found one linker testcase that broke:
default-script2. Looking at the test, it seems to me that the patched linker
is actually correct in this instance. The test uses a linker script that looks
like this:
_START = DEFINED(_START) ? _START : 0x9000000;
SECTIONS
{
. = _START;
.text : {*(.text)}
/DISCARD/ : {*(*)}
}
It then checks to see that the .text symbol has been given an address of
0x90000000 despite the fact that "-defsym _START=0x8000000" has been passed on
the command line. The patched linker sets the address to 0x80000000, which I
think is right.
Does anyone have any comments before I look at creating a patch for GOLD ?
Cheers
Nick
--
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- You are receiving this mail because: -------
You are on the CC list for the bug.