mirror of
https://github.com/varun-r-mallya/sysprof.git
synced 2025-12-31 20:36:25 +00:00
New function
2006-08-20 Soren Sandmann <sandmann@daimi.au.dk> * elfparser.c (elf_parser_get_debug_link): New function * elfparser.c: Delete SymbolTable typedef * binparser.[ch] (bin_parser_get_data): New function (bin_parser_get_length): New function * elfparser.[ch] (elf_parser_get_crc32): New function * TODO: Updates * elfparser.c (elf_parser_lookup_symbol): Offset passed in addresses by the load address. * elfparser.c (elf_parser_get_load_address): New function to compute the load address of the beginning of the file.
This commit is contained in:
committed by
Søren Sandmann Pedersen
parent
49e1fabe46
commit
1b72901c4d
19
ChangeLog
19
ChangeLog
@ -1,3 +1,22 @@
|
||||
2006-08-20 Soren Sandmann <sandmann@daimi.au.dk>
|
||||
|
||||
* elfparser.c (elf_parser_get_debug_link): New function
|
||||
|
||||
* elfparser.c: Delete SymbolTable typedef
|
||||
|
||||
* binparser.[ch] (bin_parser_get_data): New function
|
||||
(bin_parser_get_length): New function
|
||||
|
||||
* elfparser.[ch] (elf_parser_get_crc32): New function
|
||||
|
||||
* TODO: Updates
|
||||
|
||||
* elfparser.c (elf_parser_lookup_symbol): Offset passed in
|
||||
addresses by the load address.
|
||||
|
||||
* elfparser.c (elf_parser_get_load_address): New function to
|
||||
compute the load address of the beginning of the file.
|
||||
|
||||
2006-08-20 Soren Sandmann <sandmann@redhat.com>
|
||||
|
||||
* elfparser.c (elf_sym_get_name): Read the name out of the file
|
||||
|
||||
41
TODO
41
TODO
@ -32,6 +32,41 @@ Before 1.0.4:
|
||||
|
||||
Before 1.2:
|
||||
|
||||
* Rethink binparser. Maybe the default mode should be:
|
||||
- there is a current offset
|
||||
- you can move the cursor
|
||||
- _goto()
|
||||
- _align()
|
||||
- you can read structs with "begin_struct (format) / end_struct()"
|
||||
Or maybe just "set_format()" that would accept NULL?
|
||||
- when you are reading a struct, you can skip records with _index()
|
||||
- you can read fields with get_string/get_uint by passing a name.
|
||||
- you can read anonymous strings and uints by passing NULL for name
|
||||
This is allowed even when not reading structs. Or maybe this
|
||||
should be separate functions. Advantages:
|
||||
- they can skip ahead, unlike fields accessors
|
||||
- you can access specific types (8,16,32,64)
|
||||
- there is no "name" field
|
||||
Disadvantage:
|
||||
- the field accesors would need renaming.
|
||||
bin_parser_get_uint_field ()
|
||||
is not really that bad though.
|
||||
Maybe begin_record() could return a structure you could
|
||||
use to access that particular record? Would nicely solve
|
||||
the problems with "goto" and "index".
|
||||
bin_record_get_uint();
|
||||
What should begin/end be called? They will have different
|
||||
objects passed.
|
||||
bin_parser_get_record (parser) -> record
|
||||
bin_record_free (record);
|
||||
- Maybe support for indirect strings? Ie., get_string() in elfparser
|
||||
- This will require endianness to be a per-parser property. Which is
|
||||
probably just fine. Although d-bus actually has
|
||||
per-message endianness. Maybe there could be a settable
|
||||
"endianness" property.
|
||||
|
||||
Also need to add error checking.
|
||||
|
||||
* Rename stack_stash_foreach_by_address() to stack_stash_foreach_unique(),
|
||||
or maybe not ...
|
||||
|
||||
@ -342,9 +377,6 @@ http://www.linuxbase.org/spec/booksets/LSB-Embedded/LSB-Embedded/ehframe.html
|
||||
- possibly add dependency on glib 2.8 if it is released at that point.
|
||||
(g_file_replace())
|
||||
|
||||
- somehow get access to VSEnterprise profiler and see how it works.
|
||||
somehow get access to vtune and see how it works.
|
||||
|
||||
* Some notes about timer interrupt handling in Linux
|
||||
|
||||
On an SMP system APIC is used - the interesting file is arch/i386/kernel/apic.c
|
||||
@ -368,6 +400,9 @@ When the interrupt happens,
|
||||
|
||||
Later:
|
||||
|
||||
- somehow get access to VSEnterprise profiler and see how it works.
|
||||
somehow get access to vtune and see how it works.
|
||||
|
||||
- On SMP systems interrupts happen unpredictably, including when another
|
||||
one is running. Right now we are ignoring any interrupts that happen
|
||||
when another one is running, but we should probably just save the data
|
||||
|
||||
14
binparser.c
14
binparser.c
@ -333,9 +333,21 @@ bin_parser_get_offset (BinParser *parser)
|
||||
return parser->frame->offset;
|
||||
}
|
||||
|
||||
const guchar *
|
||||
bin_parser_get_data (BinParser *parser)
|
||||
{
|
||||
return parser->data;
|
||||
}
|
||||
|
||||
gsize
|
||||
bin_parser_get_length (BinParser *parser)
|
||||
{
|
||||
return parser->length;
|
||||
}
|
||||
|
||||
BinField *
|
||||
bin_field_new_fixed_array (int n_elements,
|
||||
int element_size)
|
||||
int element_size)
|
||||
{
|
||||
BinField *field = g_new0 (BinField, 1);
|
||||
field->width = n_elements * element_size;
|
||||
|
||||
18
binparser.h
18
binparser.h
@ -3,13 +3,13 @@
|
||||
typedef struct BinField BinField;
|
||||
typedef struct BinFormat BinFormat;
|
||||
typedef struct BinParser BinParser;
|
||||
|
||||
/* BinParser */
|
||||
BinParser *bin_parser_new (const guchar *data,
|
||||
gsize length);
|
||||
BinFormat *bin_format_new (gboolean big_endian,
|
||||
const char *name, BinField *field,
|
||||
...);
|
||||
const guchar *bin_parser_get_data (BinParser *parser);
|
||||
gsize bin_parser_get_length (BinParser *parser);
|
||||
gsize bin_parser_get_offset (BinParser *parser);
|
||||
gsize bin_format_get_size (BinFormat *format);
|
||||
void bin_parser_index (BinParser *parser, int index);
|
||||
void bin_parser_begin (BinParser *parser,
|
||||
BinFormat *format,
|
||||
@ -18,9 +18,17 @@ void bin_parser_end (BinParser *parser);
|
||||
const char *bin_parser_get_string (BinParser *parser);
|
||||
guint64 bin_parser_get_uint (BinParser *parser,
|
||||
const gchar *name);
|
||||
|
||||
/* BinFormat */
|
||||
BinFormat *bin_format_new (gboolean big_endian,
|
||||
const char *name, BinField *field,
|
||||
...);
|
||||
gsize bin_format_get_size (BinFormat *format);
|
||||
|
||||
/* BinField */
|
||||
BinField *bin_field_new_uint8 (void);
|
||||
BinField *bin_field_new_uint16 (void);
|
||||
BinField *bin_field_new_uint32 (void);
|
||||
BinField *bin_field_new_uint64 (void);
|
||||
BinField *bin_field_new_fixed_array (int n_elements,
|
||||
int element_size);
|
||||
int element_size);
|
||||
|
||||
206
elfparser.c
206
elfparser.c
@ -4,13 +4,8 @@
|
||||
#include "binparser.h"
|
||||
#include "elfparser.h"
|
||||
|
||||
typedef struct SymbolTable SymbolTable;
|
||||
typedef struct Section Section;
|
||||
|
||||
struct SymbolTable
|
||||
{
|
||||
};
|
||||
|
||||
struct ElfSym
|
||||
{
|
||||
gulong offset;
|
||||
@ -22,23 +17,25 @@ struct Section
|
||||
const gchar * name;
|
||||
gsize offset;
|
||||
gsize size;
|
||||
gboolean allocated;
|
||||
gulong load_address;
|
||||
};
|
||||
|
||||
struct ElfParser
|
||||
{
|
||||
BinParser * parser;
|
||||
BinParser * parser;
|
||||
|
||||
BinFormat * header;
|
||||
BinFormat * strtab_format;
|
||||
BinFormat * shn_entry;
|
||||
BinFormat * sym_format;
|
||||
BinFormat * header;
|
||||
BinFormat * strtab_format;
|
||||
BinFormat * shn_entry;
|
||||
BinFormat * sym_format;
|
||||
|
||||
int n_sections;
|
||||
Section ** sections;
|
||||
int n_sections;
|
||||
Section ** sections;
|
||||
|
||||
int n_symbols;
|
||||
ElfSym * symbols;
|
||||
gsize sym_strings;
|
||||
int n_symbols;
|
||||
ElfSym * symbols;
|
||||
gsize sym_strings;
|
||||
};
|
||||
|
||||
static gboolean parse_elf_signature (const guchar *data, gsize length,
|
||||
@ -72,11 +69,20 @@ section_new (ElfParser *parser,
|
||||
{
|
||||
BinParser *bparser = parser->parser;
|
||||
Section *section = g_new (Section, 1);
|
||||
guint64 flags;
|
||||
|
||||
section->name = get_string (bparser, name_table, "sh_name");
|
||||
section->size = bin_parser_get_uint (bparser, "sh_size");
|
||||
section->offset = bin_parser_get_uint (bparser, "sh_offset");
|
||||
|
||||
flags = bin_parser_get_uint (bparser, "sh_flags");
|
||||
section->allocated = !!(flags & SHF_ALLOC);
|
||||
|
||||
if (section->allocated)
|
||||
section->load_address = bin_parser_get_uint (bparser, "sh_addr");
|
||||
else
|
||||
section->load_address = 0;
|
||||
|
||||
return section;
|
||||
}
|
||||
|
||||
@ -158,6 +164,70 @@ elf_parser_new (const guchar *data, gsize length)
|
||||
return parser;
|
||||
}
|
||||
|
||||
guint32
|
||||
elf_parser_get_crc32 (ElfParser *parser)
|
||||
{
|
||||
static const unsigned long crc32_table[256] = {
|
||||
0x00000000, 0x77073096, 0xee0e612c, 0x990951ba, 0x076dc419, 0x706af48f,
|
||||
0xe963a535, 0x9e6495a3, 0x0edb8832, 0x79dcb8a4, 0xe0d5e91e, 0x97d2d988,
|
||||
0x09b64c2b, 0x7eb17cbd, 0xe7b82d07, 0x90bf1d91, 0x1db71064, 0x6ab020f2,
|
||||
0xf3b97148, 0x84be41de, 0x1adad47d, 0x6ddde4eb, 0xf4d4b551, 0x83d385c7,
|
||||
0x136c9856, 0x646ba8c0, 0xfd62f97a, 0x8a65c9ec, 0x14015c4f, 0x63066cd9,
|
||||
0xfa0f3d63, 0x8d080df5, 0x3b6e20c8, 0x4c69105e, 0xd56041e4, 0xa2677172,
|
||||
0x3c03e4d1, 0x4b04d447, 0xd20d85fd, 0xa50ab56b, 0x35b5a8fa, 0x42b2986c,
|
||||
0xdbbbc9d6, 0xacbcf940, 0x32d86ce3, 0x45df5c75, 0xdcd60dcf, 0xabd13d59,
|
||||
0x26d930ac, 0x51de003a, 0xc8d75180, 0xbfd06116, 0x21b4f4b5, 0x56b3c423,
|
||||
0xcfba9599, 0xb8bda50f, 0x2802b89e, 0x5f058808, 0xc60cd9b2, 0xb10be924,
|
||||
0x2f6f7c87, 0x58684c11, 0xc1611dab, 0xb6662d3d, 0x76dc4190, 0x01db7106,
|
||||
0x98d220bc, 0xefd5102a, 0x71b18589, 0x06b6b51f, 0x9fbfe4a5, 0xe8b8d433,
|
||||
0x7807c9a2, 0x0f00f934, 0x9609a88e, 0xe10e9818, 0x7f6a0dbb, 0x086d3d2d,
|
||||
0x91646c97, 0xe6635c01, 0x6b6b51f4, 0x1c6c6162, 0x856530d8, 0xf262004e,
|
||||
0x6c0695ed, 0x1b01a57b, 0x8208f4c1, 0xf50fc457, 0x65b0d9c6, 0x12b7e950,
|
||||
0x8bbeb8ea, 0xfcb9887c, 0x62dd1ddf, 0x15da2d49, 0x8cd37cf3, 0xfbd44c65,
|
||||
0x4db26158, 0x3ab551ce, 0xa3bc0074, 0xd4bb30e2, 0x4adfa541, 0x3dd895d7,
|
||||
0xa4d1c46d, 0xd3d6f4fb, 0x4369e96a, 0x346ed9fc, 0xad678846, 0xda60b8d0,
|
||||
0x44042d73, 0x33031de5, 0xaa0a4c5f, 0xdd0d7cc9, 0x5005713c, 0x270241aa,
|
||||
0xbe0b1010, 0xc90c2086, 0x5768b525, 0x206f85b3, 0xb966d409, 0xce61e49f,
|
||||
0x5edef90e, 0x29d9c998, 0xb0d09822, 0xc7d7a8b4, 0x59b33d17, 0x2eb40d81,
|
||||
0xb7bd5c3b, 0xc0ba6cad, 0xedb88320, 0x9abfb3b6, 0x03b6e20c, 0x74b1d29a,
|
||||
0xead54739, 0x9dd277af, 0x04db2615, 0x73dc1683, 0xe3630b12, 0x94643b84,
|
||||
0x0d6d6a3e, 0x7a6a5aa8, 0xe40ecf0b, 0x9309ff9d, 0x0a00ae27, 0x7d079eb1,
|
||||
0xf00f9344, 0x8708a3d2, 0x1e01f268, 0x6906c2fe, 0xf762575d, 0x806567cb,
|
||||
0x196c3671, 0x6e6b06e7, 0xfed41b76, 0x89d32be0, 0x10da7a5a, 0x67dd4acc,
|
||||
0xf9b9df6f, 0x8ebeeff9, 0x17b7be43, 0x60b08ed5, 0xd6d6a3e8, 0xa1d1937e,
|
||||
0x38d8c2c4, 0x4fdff252, 0xd1bb67f1, 0xa6bc5767, 0x3fb506dd, 0x48b2364b,
|
||||
0xd80d2bda, 0xaf0a1b4c, 0x36034af6, 0x41047a60, 0xdf60efc3, 0xa867df55,
|
||||
0x316e8eef, 0x4669be79, 0xcb61b38c, 0xbc66831a, 0x256fd2a0, 0x5268e236,
|
||||
0xcc0c7795, 0xbb0b4703, 0x220216b9, 0x5505262f, 0xc5ba3bbe, 0xb2bd0b28,
|
||||
0x2bb45a92, 0x5cb36a04, 0xc2d7ffa7, 0xb5d0cf31, 0x2cd99e8b, 0x5bdeae1d,
|
||||
0x9b64c2b0, 0xec63f226, 0x756aa39c, 0x026d930a, 0x9c0906a9, 0xeb0e363f,
|
||||
0x72076785, 0x05005713, 0x95bf4a82, 0xe2b87a14, 0x7bb12bae, 0x0cb61b38,
|
||||
0x92d28e9b, 0xe5d5be0d, 0x7cdcefb7, 0x0bdbdf21, 0x86d3d2d4, 0xf1d4e242,
|
||||
0x68ddb3f8, 0x1fda836e, 0x81be16cd, 0xf6b9265b, 0x6fb077e1, 0x18b74777,
|
||||
0x88085ae6, 0xff0f6a70, 0x66063bca, 0x11010b5c, 0x8f659eff, 0xf862ae69,
|
||||
0x616bffd3, 0x166ccf45, 0xa00ae278, 0xd70dd2ee, 0x4e048354, 0x3903b3c2,
|
||||
0xa7672661, 0xd06016f7, 0x4969474d, 0x3e6e77db, 0xaed16a4a, 0xd9d65adc,
|
||||
0x40df0b66, 0x37d83bf0, 0xa9bcae53, 0xdebb9ec5, 0x47b2cf7f, 0x30b5ffe9,
|
||||
0xbdbdf21c, 0xcabac28a, 0x53b39330, 0x24b4a3a6, 0xbad03605, 0xcdd70693,
|
||||
0x54de5729, 0x23d967bf, 0xb3667a2e, 0xc4614ab8, 0x5d681b02, 0x2a6f2b94,
|
||||
0xb40bbe37, 0xc30c8ea1, 0x5a05df1b, 0x2d02ef8d
|
||||
};
|
||||
const guchar *data;
|
||||
gsize length;
|
||||
gulong crc;
|
||||
gsize i;
|
||||
|
||||
data = bin_parser_get_data (parser->parser);
|
||||
length = bin_parser_get_length (parser->parser);
|
||||
|
||||
crc = 0;
|
||||
|
||||
for (i = 0; i < length; ++i)
|
||||
crc = crc32_table[(crc ^ data[i]) & 0xff] ^ (crc >> 8);
|
||||
|
||||
return crc & 0xFFFFFFFF;
|
||||
}
|
||||
|
||||
void
|
||||
elf_parser_free (ElfParser *parser)
|
||||
{
|
||||
@ -292,8 +362,37 @@ read_symbols (ElfParser *parser)
|
||||
}
|
||||
}
|
||||
|
||||
/*
|
||||
* Returns the address that the start of the file would be loaded
|
||||
* at if the whole file was mapped
|
||||
*/
|
||||
static gulong
|
||||
elf_parser_get_load_address (ElfParser *parser)
|
||||
{
|
||||
int i;
|
||||
gulong load_address = (gulong)-1;
|
||||
|
||||
for (i = 0; i < parser->n_sections; ++i)
|
||||
{
|
||||
Section *section = parser->sections[i];
|
||||
|
||||
if (section->allocated)
|
||||
{
|
||||
gulong addr = section->load_address - section->offset;
|
||||
load_address = MIN (load_address, addr);
|
||||
}
|
||||
}
|
||||
|
||||
g_print ("load address is: %8p\n", (void *)load_address);
|
||||
|
||||
return load_address;
|
||||
}
|
||||
|
||||
static ElfSym *
|
||||
do_lookup (ElfSym *symbols, gulong address, int first, int last)
|
||||
do_lookup (ElfSym *symbols,
|
||||
gulong address,
|
||||
int first,
|
||||
int last)
|
||||
{
|
||||
if (address >= symbols[last].address)
|
||||
{
|
||||
@ -316,13 +415,9 @@ do_lookup (ElfSym *symbols, gulong address, int first, int last)
|
||||
int mid = (first + last) / 2;
|
||||
|
||||
if (symbols[mid].address > address)
|
||||
{
|
||||
return do_lookup (symbols, address, first, mid);
|
||||
}
|
||||
else
|
||||
{
|
||||
return do_lookup (symbols, address, mid, last);
|
||||
}
|
||||
}
|
||||
}
|
||||
|
||||
@ -336,8 +431,15 @@ elf_parser_lookup_symbol (ElfParser *parser,
|
||||
if (parser->n_symbols == 0)
|
||||
return NULL;
|
||||
|
||||
/* FIXME: we should offset address based on the files load address */
|
||||
address += elf_parser_get_load_address (parser);
|
||||
|
||||
#if 0
|
||||
g_print ("the address we are looking up is %p\n", address);
|
||||
#endif
|
||||
|
||||
/* FIXME: we should look at the symbol size and check if the
|
||||
* address is actually within the function.
|
||||
*/
|
||||
return do_lookup (parser->symbols, address, 0, parser->n_symbols - 1);
|
||||
}
|
||||
|
||||
@ -363,6 +465,68 @@ parser_from_sym (const ElfSym *sym)
|
||||
return NULL;
|
||||
}
|
||||
|
||||
const char *
|
||||
elf_parser_get_debug_link (ElfParser *parser, guint32 *crc32)
|
||||
{
|
||||
const Section *debug_link = find_section (parser, ".gnu_debuglink");
|
||||
const gchar *result;
|
||||
gsize crc_offset;
|
||||
|
||||
if (!debug_link)
|
||||
return NULL;
|
||||
|
||||
bin_parser_begin (parser->parser, NULL, debug_link->offset);
|
||||
result = bin_parser_get_string (parser->parser);
|
||||
bin_parser_end (parser->parser);
|
||||
|
||||
crc_offset = strlen (result) + 1;
|
||||
crc_offset = (crc_offset + 3) & ~3;
|
||||
|
||||
/* FIXME: This is broken for two reasons:
|
||||
*
|
||||
* (1) It assumes file_endian==machine_endian
|
||||
*
|
||||
* (2) It doesn't check for file overrun.
|
||||
*
|
||||
* The fix is to make binparser capable of dealing with stuff
|
||||
* outside of records.
|
||||
*/
|
||||
|
||||
*crc32 = *(guint32 *)(result + crc_offset);
|
||||
return result;
|
||||
}
|
||||
|
||||
#if 0
|
||||
get_debug_link_info (bfd *abfd, unsigned long *crc32_out)
|
||||
{
|
||||
asection *sect;
|
||||
bfd_size_type debuglink_size;
|
||||
unsigned long crc32;
|
||||
char *contents;
|
||||
int crc_offset;
|
||||
|
||||
sect = bfd_get_section_by_name (abfd, ".gnu_debuglink");
|
||||
|
||||
if (sect == NULL)
|
||||
return NULL;
|
||||
|
||||
debuglink_size = bfd_section_size (abfd, sect);
|
||||
|
||||
contents = g_malloc (debuglink_size);
|
||||
bfd_get_section_contents (abfd, sect, contents,
|
||||
(file_ptr)0, (bfd_size_type)debuglink_size);
|
||||
|
||||
/* Crc value is stored after the filename, aligned up to 4 bytes. */
|
||||
crc_offset = strlen (contents) + 1;
|
||||
crc_offset = (crc_offset + 3) & ~3;
|
||||
|
||||
crc32 = bfd_get_32 (abfd, (bfd_byte *) (contents + crc_offset));
|
||||
|
||||
*crc32_out = crc32;
|
||||
return contents;
|
||||
}
|
||||
#endif
|
||||
|
||||
const char *
|
||||
elf_sym_get_name (const ElfSym *sym)
|
||||
{
|
||||
|
||||
@ -21,7 +21,7 @@ void elf_parser_free (ElfParser *parser);
|
||||
*/
|
||||
const ElfSym *elf_parser_lookup_symbol (ElfParser *parser,
|
||||
gulong address);
|
||||
|
||||
guint32 elf_parser_get_crc32 (ElfParser *parser);
|
||||
const char *elf_sym_get_name (const ElfSym *sym);
|
||||
gulong elf_sym_get_address (const ElfSym *sym);
|
||||
char *elf_demangle (const char *name);
|
||||
|
||||
18
testelf.c
18
testelf.c
@ -7,25 +7,26 @@ static void
|
||||
check (ElfParser *elf, gulong addr)
|
||||
{
|
||||
const ElfSym *sym = elf_parser_lookup_symbol (elf, addr);
|
||||
|
||||
if (!sym)
|
||||
{
|
||||
g_print ("not found\n");
|
||||
return;
|
||||
}
|
||||
|
||||
n = elf_sym_get_name (sym);
|
||||
|
||||
#if 0
|
||||
g_print ("%p => ", (void *)addr);
|
||||
#endif
|
||||
|
||||
if (sym)
|
||||
{
|
||||
#if 0
|
||||
g_print ("found: %s (%p)\n",
|
||||
elf_sym_get_name (sym),
|
||||
(void *)elf_sym_get_address (sym));
|
||||
#endif
|
||||
}
|
||||
else
|
||||
{
|
||||
#if 0
|
||||
g_print ("not found\n");
|
||||
#endif
|
||||
}
|
||||
}
|
||||
|
||||
@ -42,8 +43,9 @@ main ()
|
||||
|
||||
for (i = 0; i < 5000000; ++i)
|
||||
{
|
||||
check (elf, 0x3e7ef20); /* gtk_handle_box_end_drag */
|
||||
check (elf, 0x3e7ef25); /* same (but in the middle of the function */
|
||||
check (elf, 0x077c80f0 - (0x07787000 - 0)); /* gtk_about_dialog_set_artists (add - (map - offset)) */
|
||||
|
||||
check (elf, 0x077c80f0 - (0x07787000 - 0)); /* same (but in the middle of the function */
|
||||
}
|
||||
return 0;
|
||||
}
|
||||
|
||||
Reference in New Issue
Block a user