[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[Bug gold/18846] gold PowerPC --emit-relocs differ from ld
From: |
amodra at gmail dot com |
Subject: |
[Bug gold/18846] gold PowerPC --emit-relocs differ from ld |
Date: |
Tue, 18 Aug 2015 13:21:42 +0000 |
https://sourceware.org/bugzilla/show_bug.cgi?id=18846
Alan Modra <amodra at gmail dot com> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |amodra at gmail dot com
--- Comment #1 from Alan Modra <amodra at gmail dot com> ---
No, that's not the correct patch. For ld -r we do need to add in the symbol
value, which is set to the offset of the input section in the output section
here in object.cc:compute_final_local_value_internal
lv_out->set_output_value((relocatable ? 0 : os->address())
+ secoffset
+ lv_in->input_value());
Notice that the value set is different for final linking (!relocatable). For
final linking with --emit-relocs the code that you patched out should subtract
out that os->address() component.
This same bug exists in the generic target-reloc.h code.
--
You are receiving this mail because:
You are on the CC list for the bug.
- [Bug gold/18846] New: gold PowerPC --emit-relocs differ from ld, johan.karlsson at enea dot com, 2015/08/18
- [Bug gold/18846] gold PowerPC --emit-relocs differ from ld, johan.karlsson at enea dot com, 2015/08/18
- [Bug gold/18846] gold PowerPC --emit-relocs differ from ld,
amodra at gmail dot com <=
- [Bug gold/18846] gold PowerPC --emit-relocs differ from ld, amodra at gmail dot com, 2015/08/18
- [Bug gold/18846] gold PowerPC --emit-relocs differ from ld, amodra at gmail dot com, 2015/08/19
- [Bug gold/18846] gold PowerPC --emit-relocs differ from ld, cvs-commit at gcc dot gnu.org, 2015/08/19
- [Bug gold/18846] gold PowerPC --emit-relocs differ from ld, amodra at gmail dot com, 2015/08/20