1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
|
From binutils-return-38148-listarch-binutils=sources dot redhat dot com at sources dot redhat dot com Tue Feb 22 19:24:15 2005
Return-Path: <binutils-return-38148-listarch-binutils=sources dot redhat dot com at sources dot redhat dot com>
Delivered-To: listarch-binutils at sources dot redhat dot com
Received: (qmail 4446 invoked by alias); 22 Feb 2005 19:24:15 -0000
Mailing-List: contact binutils-help at sources dot redhat dot com; run by ezmlm
Precedence: bulk
List-Subscribe: <mailto:binutils-subscribe at sources dot redhat dot com>
List-Archive: <http://sources.redhat.com/ml/binutils/>
List-Post: <mailto:binutils at sources dot redhat dot com>
List-Help: <mailto:binutils-help at sources dot redhat dot com>, <http://sources dot redhat dot com/ml/#faqs>
Sender: binutils-owner at sources dot redhat dot com
Delivered-To: mailing list binutils at sources dot redhat dot com
Received: (qmail 4401 invoked from network); 22 Feb 2005 19:24:08 -0000
Received: from unknown (HELO bgo1smout1.broadpark.no) (217.13.4.94)
by sourceware dot org with SMTP; 22 Feb 2005 19:24:08 -0000
Received: from bgo1sminn1.broadpark.no ([217.13.4.93])
by bgo1smout1 dot broadpark dot no
(Sun Java System Messaging Server 6 dot 1 HotFix 0 dot 05 (built Oct 21 2004))
with ESMTP id <0ICB007QZUZCC0C0 at bgo1smout1 dot broadpark dot no> for
binutils at sources dot redhat dot com; Tue, 22 Feb 2005 20:18:48 +0100 (CET)
Received: from [127.0.0.1] ([80.202.165.9]) by bgo1sminn1.broadpark.no
(Sun Java System Messaging Server 6 dot 1 HotFix 0 dot 05 (built Oct 21 2004))
with ESMTP id <0ICB006NCVBVHE21 at bgo1sminn1 dot broadpark dot no> for
binutils at sources dot redhat dot com; Tue, 22 Feb 2005 20:26:20 +0100 (CET)
Date: Tue, 22 Feb 2005 20:24:08 +0100
From: =?ISO-8859-1?Q?Stig_Petter_Olsr=F8d?= <stigpo at users dot sourceforge dot net>
Subject: [PATCH] objdump relocation fixes for ARM disassembly
To: binutils at sources dot redhat dot com
Message-id: <421B86D8.8080604@users.sourceforge.net>
MIME-version: 1.0
Content-type: text/plain; charset=ISO-8859-1; format=flowed
Content-transfer-encoding: 7BIT
User-Agent: Mozilla Thunderbird 1.0 (Windows/20041206)
Hello,
objdump disassembly did not relocate correctly for the ARM processor. It seems
that the test for triggering the INSN_HAS_RELOC flag was void (one test killed the other,
since octets would always be zero) and all relocations would thus fail. I changed the test
so the flag is set when we are about to disassemble an insn that the current relocation
entry points to. I also changed objdump_print_addr to use the current relocation entry if
the insn has such an entry. This causes the symbol printed to be correct for both external
symbols (from the undefined section) and local symbols.
This has only been tested for the ARM processor, but I don't think it should break other
DISASSEMBLER_NEEDS_RELOCS processors either.
binutils/
2005-02-22 Stig Petter Olsroed <stigpo@users.sourceforge.net>
* objdump.c (disassemble_bytes): Fixed relocation check for
DISASSEMBLER_NEEDS_RELOCS platforms to properly trigger the
INSN_HAS_RELOC flag. Set the current relocation entry in
objdump_disasm_info to allow printing the proper symbol.
(objdump_print_addr): Use the relocation entry in
objdump_disasm_info to lookup the correct symbol for
DISASSEMBLER_NEEDS_RELOCS platforms.
--- 1/binutils/objdump.c 2005-02-22 01:50:06.000000000 +0100
+++ 2/binutils/objdump.c 2005-02-22 14:27:33.066960900 +0100
@@ -128,6 +128,7 @@
arelent ** dynrelbuf;
long dynrelcount;
disassembler_ftype disassemble_fn;
+ arelent * reloc;
};
/* Architecture to disassemble for, or default if NULL. */
@@ -852,6 +853,8 @@
{
struct objdump_disasm_info *aux;
asymbol *sym;
+ arelent *q;
+ int skip_find = 0;
if (sorted_symcount < 1)
{
@@ -861,6 +864,22 @@
}
aux = (struct objdump_disasm_info *) info->application_data;
+
+ q = aux->reloc;
+ if (q != NULL)
+ {
+ if (q->sym_ptr_ptr != NULL && *q->sym_ptr_ptr != NULL)
+ {
+ /* Adjust the vma to the reloc */
+ vma += bfd_asymbol_value (*q->sym_ptr_ptr);
+ if (bfd_is_und_section (bfd_get_section (*q->sym_ptr_ptr)))
+ {
+ skip_find = 1;
+ sym = *q->sym_ptr_ptr;
+ }
+ }
+ }
+ if (!skip_find)
sym = find_symbol_for_address (vma, info, NULL);
objdump_print_addr_with_sym (aux->abfd, aux->sec, sym, vma, info,
skip_zeroes);
@@ -1350,16 +1369,22 @@
info->bytes_per_chunk = 0;
#ifdef DISASSEMBLER_NEEDS_RELOCS
- /* FIXME: This is wrong. It tests the number of octets
- in the last instruction, not the current one. */
- if (*relppp < relppend
- && (**relppp)->address >= rel_offset + addr_offset
- && ((**relppp)->address
- < rel_offset + addr_offset + octets / opb))
+ /* Check if the current relocation entry applies to the
+ instruction we are about to disassemble.
+ This works for ARM at least.
+ */
+ if ((*relppp) < relppend
+ && ((**relppp)->address == rel_offset + addr_offset))
+ {
info->flags = INSN_HAS_RELOC;
+ aux->reloc = **relppp;
+ }
else
#endif
+ {
info->flags = 0;
+ aux->reloc = NULL;
+ }
octets = (*disassemble_fn) (section->vma + addr_offset, info);
info->fprintf_func = (fprintf_ftype) fprintf;
|
