Baijnath Jaiswal
May 17th, 2013, 07:06 AM
I am trying to extract all the global symbols (with name and corresponding values) of a C code from core file (core.pid file, generated when the code crashes).
This is one way to do, but it extracts symbols while code is executing: For example, we have list_globals.py script, and any C code which is crashing. suppose c_code.c.
import gdb
block = gdb.block_for_pc(long(gdb.parse_and_eval('$pc'))). global_block
for symbol in block:
gdb.write('SYMBOL: %20s,\tTYPE: %20s,\tValue: %20s\n' % (symbol.print_name, symbol.type, symbol.value()))
This script is tested in this way:
gcc -g c_code.c
./a.out
gdb a.out core.pid
(gdb) start
(gdb) source list_globals.py
This will print all the global symbols with name,type and values. But here, (gdb) start is must because symbols are extracted by executing the c_code again. i want to extract the same information from core file without executing the C code again. How to do this..? Thanks
This is one way to do, but it extracts symbols while code is executing: For example, we have list_globals.py script, and any C code which is crashing. suppose c_code.c.
import gdb
block = gdb.block_for_pc(long(gdb.parse_and_eval('$pc'))). global_block
for symbol in block:
gdb.write('SYMBOL: %20s,\tTYPE: %20s,\tValue: %20s\n' % (symbol.print_name, symbol.type, symbol.value()))
This script is tested in this way:
gcc -g c_code.c
./a.out
gdb a.out core.pid
(gdb) start
(gdb) source list_globals.py
This will print all the global symbols with name,type and values. But here, (gdb) start is must because symbols are extracted by executing the c_code again. i want to extract the same information from core file without executing the C code again. How to do this..? Thanks