From: Jan Beulich <jbeulich@sourceware.org>
To: bfd-cvs@sourceware.org
Subject: [binutils-gdb] bfd: correct relocation handling for objcopy COFF -> ELF
Date: Fri, 25 Aug 2023 12:57:04 +0000 (GMT) [thread overview]
Message-ID: <20230825125704.D5F8E3858C62@sourceware.org> (raw)
https://sourceware.org/git/gitweb.cgi?p=binutils-gdb.git;h=67694446f772ae95d330419eb6d01e38e6d24612
commit 67694446f772ae95d330419eb6d01e38e6d24612
Author: Jan Beulich <jbeulich@suse.com>
Date: Fri Aug 25 14:56:44 2023 +0200
bfd: correct relocation handling for objcopy COFF -> ELF
While documented to not be reliable, it is still odd for objcopy to
silently produce bad output when converting COFF/PE object files to ELF
ones. The issue there is that relocation addends all are screwed up by
subtracting the symbol's section offset. In the COFF/PE world, to my
knowledge, section contents stores the addends alone, not the result of
symbol value plus addend. Hence the compensation talked about in a
comment ahead of the sole use site of CALC_ADDEND() may need to account
for the VMA (which is always zero for object files anyway), but not for
the symbol value.
The coff-sh.c adjustment is based upon guessing that behavior there is
the same. Note also how coff-aarch64.c short-circuits CALC_ADDEND()
altogether, which may suggest that a much simpler macro might do for the
COFF_WITH_PE case in the three arch-specific files touched here.
For (at least) Arm/WinCE this actually results in more appropriate
objdump output as well, as can be seen in the one testcase which has its
expectations adjusted (the generated binary doesn't change).
Diff:
---
bfd/coff-i386.c | 3 ++-
bfd/coff-sh.c | 3 ++-
bfd/coff-x86_64.c | 3 ++-
bfd/coffcode.h | 9 ++++++++-
gas/testsuite/gas/arm/wince.d | 6 +++---
5 files changed, 17 insertions(+), 7 deletions(-)
diff --git a/bfd/coff-i386.c b/bfd/coff-i386.c
index 24a05d3aa65..71516066da8 100644
--- a/bfd/coff-i386.c
+++ b/bfd/coff-i386.c
@@ -405,7 +405,8 @@ static reloc_howto_type howto_table[] =
cache_ptr->addend = - coffsym->native->u.syment.n_value; \
else if (ptr && bfd_asymbol_bfd (ptr) == abfd \
&& ptr->section != (asection *) NULL) \
- cache_ptr->addend = - (ptr->section->vma + ptr->value); \
+ cache_ptr->addend = - (ptr->section->vma \
+ + COFF_PE_ADDEND_BIAS (ptr)); \
else \
cache_ptr->addend = 0; \
if (ptr && reloc.r_type < NUM_HOWTOS \
diff --git a/bfd/coff-sh.c b/bfd/coff-sh.c
index 14f502e3638..a20c01127b6 100644
--- a/bfd/coff-sh.c
+++ b/bfd/coff-sh.c
@@ -544,7 +544,8 @@ sh_coff_reloc_name_lookup (bfd *abfd ATTRIBUTE_UNUSED,
cache_ptr->addend = 0; \
else if (ptr && bfd_asymbol_bfd (ptr) == abfd \
&& ptr->section != (asection *) NULL) \
- cache_ptr->addend = - (ptr->section->vma + ptr->value); \
+ cache_ptr->addend = - (ptr->section->vma \
+ + COFF_PE_ADDEND_BIAS (ptr)); \
else \
cache_ptr->addend = 0; \
if ((reloc).r_type == R_SH_SWITCH8 \
diff --git a/bfd/coff-x86_64.c b/bfd/coff-x86_64.c
index 9a3f85cd1e5..57b975c2eed 100644
--- a/bfd/coff-x86_64.c
+++ b/bfd/coff-x86_64.c
@@ -542,7 +542,8 @@ static reloc_howto_type howto_table[] =
cache_ptr->addend = - coffsym->native->u.syment.n_value; \
else if (ptr && bfd_asymbol_bfd (ptr) == abfd \
&& ptr->section != NULL) \
- cache_ptr->addend = - (ptr->section->vma + ptr->value); \
+ cache_ptr->addend = - (ptr->section->vma \
+ + COFF_PE_ADDEND_BIAS (ptr)); \
else \
cache_ptr->addend = 0; \
if (ptr && reloc.r_type < NUM_HOWTOS \
diff --git a/bfd/coffcode.h b/bfd/coffcode.h
index e3f4afd389f..27927039a34 100644
--- a/bfd/coffcode.h
+++ b/bfd/coffcode.h
@@ -5192,6 +5192,12 @@ SUBSUBSECTION
final-linked object. See @code{CALC_ADDEND}.
*/
+#ifdef COFF_WITH_PE
+#define COFF_PE_ADDEND_BIAS(ptr) 0 /* Symbol value not stored in raw data. */
+#else
+#define COFF_PE_ADDEND_BIAS(ptr) ((ptr)->value)
+#endif
+
#ifndef CALC_ADDEND
#define CALC_ADDEND(abfd, ptr, reloc, cache_ptr) \
{ \
@@ -5208,7 +5214,8 @@ SUBSUBSECTION
cache_ptr->addend = 0; \
else if (ptr && bfd_asymbol_bfd (ptr) == abfd \
&& ptr->section != NULL) \
- cache_ptr->addend = - (ptr->section->vma + ptr->value); \
+ cache_ptr->addend = - (ptr->section->vma \
+ + COFF_PE_ADDEND_BIAS (ptr)); \
else \
cache_ptr->addend = 0; \
}
diff --git a/gas/testsuite/gas/arm/wince.d b/gas/testsuite/gas/arm/wince.d
index e5bac1de9c6..e7afbedbbf5 100644
--- a/gas/testsuite/gas/arm/wince.d
+++ b/gas/testsuite/gas/arm/wince.d
@@ -15,11 +15,11 @@ Disassembly of section .text:
0+008 <global_sym\+0x4> e1a00000 nop @ \(mov r0, r0\)
0+00c <global_sym\+0x8> e1a00000 nop @ \(mov r0, r0\)
0+010 <global_sym\+0xc> eafffffb b f+ff8 <global_sym\+0xf+ff4>
- 10: ARM_26D global_sym-0x4
+ 10: ARM_26D global_sym
0+014 <global_sym\+0x10> ebfffffa bl f+ff4 <global_sym\+0xf+ff0>
- 14: ARM_26D global_sym-0x4
+ 14: ARM_26D global_sym
0+018 <global_sym\+0x14> 0afffff9 beq f+ff0 <global_sym\+0xf+fec>
- 18: ARM_26D global_sym-0x4
+ 18: ARM_26D global_sym
0+01c <global_sym\+0x18> eafffff8 b 0+004 <global_sym>
0+020 <global_sym\+0x1c> ebfffff7 bl 0+004 <global_sym>
0+024 <global_sym\+0x20> 0afffff6 beq 0+004 <global_sym>
reply other threads:[~2023-08-25 12:57 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230825125704.D5F8E3858C62@sourceware.org \
--to=jbeulich@sourceware.org \
--cc=bfd-cvs@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).