Add an example illustrating DSE

3 files changed, 425 insertions, 0 deletions

diff --git a/example/samples/dse_crackme.c b/example/samples/dse_crackme.c
new file mode 100644
index 00000000..5fc0faaf
--- /dev/null
+++ b/example/samples/dse_crackme.c
@@ -0,0 +1,104 @@
+#include <stdio.h>
+#include <stdint.h>
+
+/*
+ * This example is a tiny crackme, with a few specificities:
+ * - the flag must be read from a file ('test.txt')
+ * - a few steps are not naively bruteforcable (using a fuzzer for instance):
+ *   comparison with a 32bit value, then CRC16
+ * - a DSE engine has to support the search into a fixed memory table
+ *   (crc16_tab) based on variables (file bytes)
+*/
+
+
+static const uint16_t crc16_tab[] = {
+    0x0000, 0x1021, 0x2042, 0x3063, 0x4084, 0x50a5, 0x60c6, 0x70e7,
+    0x8108, 0x9129, 0xa14a, 0xb16b, 0xc18c, 0xd1ad, 0xe1ce, 0xf1ef,
+    0x1231, 0x0210, 0x3273, 0x2252, 0x52b5, 0x4294, 0x72f7, 0x62d6,
+    0x9339, 0x8318, 0xb37b, 0xa35a, 0xd3bd, 0xc39c, 0xf3ff, 0xe3de,
+    0x2462, 0x3443, 0x0420, 0x1401, 0x64e6, 0x74c7, 0x44a4, 0x5485,
+    0xa56a, 0xb54b, 0x8528, 0x9509, 0xe5ee, 0xf5cf, 0xc5ac, 0xd58d,
+    0x3653, 0x2672, 0x1611, 0x0630, 0x76d7, 0x66f6, 0x5695, 0x46b4,
+    0xb75b, 0xa77a, 0x9719, 0x8738, 0xf7df, 0xe7fe, 0xd79d, 0xc7bc,
+    0x48c4, 0x58e5, 0x6886, 0x78a7, 0x0840, 0x1861, 0x2802, 0x3823,
+    0xc9cc, 0xd9ed, 0xe98e, 0xf9af, 0x8948, 0x9969, 0xa90a, 0xb92b,
+    0x5af5, 0x4ad4, 0x7ab7, 0x6a96, 0x1a71, 0x0a50, 0x3a33, 0x2a12,
+    0xdbfd, 0xcbdc, 0xfbbf, 0xeb9e, 0x9b79, 0x8b58, 0xbb3b, 0xab1a,
+    0x6ca6, 0x7c87, 0x4ce4, 0x5cc5, 0x2c22, 0x3c03, 0x0c60, 0x1c41,
+    0xedae, 0xfd8f, 0xcdec, 0xddcd, 0xad2a, 0xbd0b, 0x8d68, 0x9d49,
+    0x7e97, 0x6eb6, 0x5ed5, 0x4ef4, 0x3e13, 0x2e32, 0x1e51, 0x0e70,
+    0xff9f, 0xefbe, 0xdfdd, 0xcffc, 0xbf1b, 0xaf3a, 0x9f59, 0x8f78,
+    0x9188, 0x81a9, 0xb1ca, 0xa1eb, 0xd10c, 0xc12d, 0xf14e, 0xe16f,
+    0x1080, 0x00a1, 0x30c2, 0x20e3, 0x5004, 0x4025, 0x7046, 0x6067,
+    0x83b9, 0x9398, 0xa3fb, 0xb3da, 0xc33d, 0xd31c, 0xe37f, 0xf35e,
+    0x02b1, 0x1290, 0x22f3, 0x32d2, 0x4235, 0x5214, 0x6277, 0x7256,
+    0xb5ea, 0xa5cb, 0x95a8, 0x8589, 0xf56e, 0xe54f, 0xd52c, 0xc50d,
+    0x34e2, 0x24c3, 0x14a0, 0x0481, 0x7466, 0x6447, 0x5424, 0x4405,
+    0xa7db, 0xb7fa, 0x8799, 0x97b8, 0xe75f, 0xf77e, 0xc71d, 0xd73c,
+    0x26d3, 0x36f2, 0x0691, 0x16b0, 0x6657, 0x7676, 0x4615, 0x5634,
+    0xd94c, 0xc96d, 0xf90e, 0xe92f, 0x99c8, 0x89e9, 0xb98a, 0xa9ab,
+    0x5844, 0x4865, 0x7806, 0x6827, 0x18c0, 0x08e1, 0x3882, 0x28a3,
+    0xcb7d, 0xdb5c, 0xeb3f, 0xfb1e, 0x8bf9, 0x9bd8, 0xabbb, 0xbb9a,
+    0x4a75, 0x5a54, 0x6a37, 0x7a16, 0x0af1, 0x1ad0, 0x2ab3, 0x3a92,
+    0xfd2e, 0xed0f, 0xdd6c, 0xcd4d, 0xbdaa, 0xad8b, 0x9de8, 0x8dc9,
+    0x7c26, 0x6c07, 0x5c64, 0x4c45, 0x3ca2, 0x2c83, 0x1ce0, 0x0cc1,
+    0xef1f, 0xff3e, 0xcf5d, 0xdf7c, 0xaf9b, 0xbfba, 0x8fd9, 0x9ff8,
+    0x6e17, 0x7e36, 0x4e55, 0x5e74, 0x2e93, 0x3eb2, 0x0ed1, 0x1ef0,
+};
+
+uint16_t crc16(uint16_t seed, unsigned char *buf, size_t len)
+{
+    size_t i;
+    uint16_t tmp;
+
+    tmp = seed ^ 0xFFFFFFFF;
+    for (i = 0; i < len; i++) {
+	tmp = crc16_tab[((tmp>>8) ^ buf[i]) & 0xFF] ^ (tmp << 8);
+    }
+    return tmp;
+}
+
+uint16_t test() {
+	FILE *file;
+	unsigned char buf[0x100] = {0};
+	size_t read;
+	uint32_t temp;
+
+	file = fopen("test.txt", "r");
+	if (file) {
+		read = fread(buf, sizeof(char), 0x100, file);
+		fclose(file);
+		if (read < 14) {
+			return 0xFFFF;
+		}
+		if (buf[0] != 'M') {
+			return 0xFFFF;
+		}
+		if (!((unsigned char) (2 * buf[7]) == 14) || (buf[7] == 7)) {
+			return 0xFFFF;
+		}
+		temp = buf[1];
+		temp <<= 8;
+		temp |= buf[2];
+		temp <<= 8;
+		temp |= buf[3];
+		temp <<= 8;
+		temp |= buf[4];
+		if (temp != 0x2140244d) { // "!@$M"
+			return 0xFFFF;
+		}
+		return crc16(0, buf, read);
+	} else {
+		return 0xFFFF;
+	}
+}
+
+int main(int argc, char** argv) {
+	uint16_t result = test();
+	if (result == 0x1337) {
+		printf("OK\n");
+	} else {
+		printf("BAD\n");
+	}
+	return 0;
+}
diff --git a/example/symbol_exec/dse_crackme.py b/example/symbol_exec/dse_crackme.py
new file mode 100644
index 00000000..e08536f9
--- /dev/null
+++ b/example/symbol_exec/dse_crackme.py
@@ -0,0 +1,310 @@
+"""Example of DynamicSymbolicExecution engine use
+
+This example should run on the compiled ELF x86 64bits version of
+"dse_crackme.c"
+
+"""
+
+#### This part is only related to the run of the sample, without DSE ####
+import os
+import subprocess
+from collections import namedtuple
+from pdb import pm
+
+from miasm2.jitter.csts import PAGE_READ, PAGE_WRITE
+from miasm2.analysis.sandbox import Sandbox_Linux_x86_64
+from miasm2.expression.expression import *
+
+# File "management"
+my_FILE_ptr = 0x11223344
+FInfo = namedtuple("FInfo", ["path", "fdesc"])
+FILE_to_info = {}
+def xxx_fopen(jitter):
+    '''
+    #include <stdio.h>
+
+    FILE *fopen(const char *path, const char *mode);
+    '''
+    global my_FILE_ptr
+    ret_addr, args = jitter.func_args_systemv(['path', 'mode'])
+    fname = jitter.get_str_ansi(args.path)
+    FILE_to_info[my_FILE_ptr] = FInfo(fname, open(fname))
+    my_FILE_ptr += 1
+    return jitter.func_ret_stdcall(ret_addr, my_FILE_ptr - 1)
+
+def xxx_fread(jitter):
+    '''
+    #include <stdio.h>
+
+    size_t fread(void *ptr, size_t size, size_t nmemb, FILE *stream);
+    '''
+    ret_addr, args = jitter.func_args_systemv(['ptr', 'size', 'nmemb', 'stream'])
+    info = FILE_to_info[args.stream]
+    data = info.fdesc.read(args.size * args.nmemb)
+    jitter.vm.set_mem(args.ptr, data)
+    return jitter.func_ret_stdcall(ret_addr, len(data))
+
+def xxx_fclose(jitter):
+    '''
+    #include <stdio.h>
+
+    int fclose(FILE *stream);
+    '''
+    ret_addr, args = jitter.func_args_systemv(['stream'])
+    del FILE_to_info[args.stream]
+    return jitter.func_ret_stdcall(ret_addr, 0)
+
+# Create sandbox
+parser = Sandbox_Linux_x86_64.parser(description="ELF sandboxer")
+parser.add_argument("filename", help="ELF Filename")
+options = parser.parse_args()
+options.mimic_env = True
+sb = Sandbox_Linux_x86_64(options.filename, options, globals())
+
+# Init segment
+sb.jitter.ir_arch.do_stk_segm = True
+sb.jitter.ir_arch.do_ds_segm = True
+sb.jitter.ir_arch.do_str_segm = True
+sb.jitter.ir_arch.do_all_segm = True
+FS_0_ADDR = 0x7ff70000
+sb.jitter.cpu.FS = 0x4
+sb.jitter.cpu.set_segm_base(sb.jitter.cpu.FS, FS_0_ADDR)
+sb.jitter.vm.add_memory_page(
+    FS_0_ADDR + 0x28, PAGE_READ, "\x42\x42\x42\x42\x42\x42\x42\x42",
+    "Stack canary FS[0x28]")
+
+# Prepare the execution
+sb.jitter.init_run(sb.entry_point)
+
+
+#### This part handle the DSE ####
+from miasm2.analysis.dse import DSEPathConstraint
+from miasm2.analysis.machine import Machine
+
+
+# File "management"
+
+class SymbolicFile(object):
+    """Symbolic file with read operation, returning symbolic bytes"""
+
+    def __init__(self, fname):
+        self.fname = fname
+        self.position = 0
+        self.max_size = os.stat(fname).st_size
+        self.gen_bytes = {}
+        self.state = "OPEN"
+
+    def read(self, length):
+        assert self.state == "OPEN"
+        out = []
+        for i in xrange(self.position, min(self.position + length,
+                                           self.max_size)):
+            if i not in self.gen_bytes:
+                ret = ExprId("SF_%08x_%d" % (id(self), i), 8)
+                self.gen_bytes[i] = ret
+            out.append(self.gen_bytes[i])
+            self.position += 1
+
+        return out
+
+    def close(self):
+        self.state = "CLOSE"
+
+
+FILE_to_info_symb = {}
+FILE_stream = ExprId("FILE_0", 64)
+FILE_size = ExprId("FILE_0_size", 64)
+
+def xxx_fopen_symb(dse):
+    regs = dse.ir_arch.arch.regs
+    fname_addr = dse.eval_expr(regs.RDI)
+    mode = dse.eval_expr(regs.RSI)
+    assert fname_addr.is_int()
+    assert mode.is_int()
+    fname = dse.jitter.get_str_ansi(int(fname_addr))
+    ret_addr = ExprInt(dse.jitter.get_stack_arg(0), regs.RIP.size)
+
+    assert len(FILE_to_info_symb) == 0
+    ret_value = FILE_stream
+    FILE_to_info_symb[ret_value] = SymbolicFile(fname)
+
+    dse.update_state({
+        regs.RSP: dse.eval_expr(regs.RSP + ExprInt(8, regs.RSP.size)),
+        dse.ir_arch.IRDst: ret_addr,
+        regs.RIP: ret_addr,
+        regs.RAX: ret_value,
+    })
+
+def xxx_fread_symb(dse):
+    regs = dse.ir_arch.arch.regs
+    ptr = dse.eval_expr(regs.RDI)
+    size = dse.eval_expr(regs.RSI)
+    nmemb = dse.eval_expr(regs.RDX)
+    stream = dse.eval_expr(regs.RCX)
+
+    assert size.is_int()
+    assert nmemb.is_int()
+
+    # Fill the buffer with symbolic bytes
+    update = {}
+    sf = FILE_to_info_symb[stream]
+    data = sf.read(int(size) * int(nmemb))
+    for i, content in enumerate(data):
+        addr = dse.symb.expr_simp(ptr + ExprInt(i, ptr.size))
+        update[ExprMem(addr, 8)] = content
+
+    ret_addr = ExprInt(dse.jitter.get_stack_arg(0), regs.RIP.size)
+    ret_value = FILE_size
+
+    update.update({
+        regs.RSP: dse.symb.eval_expr(regs.RSP + ExprInt(8, regs.RSP.size)),
+        dse.ir_arch.IRDst: ret_addr,
+        regs.RIP: ret_addr,
+        regs.RAX: ret_value,
+    })
+    dse.update_state(update)
+
+def xxx_fclose_symb(dse):
+    regs = dse.ir_arch.arch.regs
+    stream = dse.eval_expr(regs.RDI)
+    FILE_to_info_symb[stream].close()
+
+    ret_addr = ExprInt(dse.jitter.get_stack_arg(0), regs.RIP.size)
+    dse.update_state({
+        regs.RSP: dse.symb.eval_expr(regs.RSP + ExprInt(8, regs.RSP.size)),
+        dse.ir_arch.IRDst: ret_addr,
+        regs.RIP: ret_addr,
+        regs.RAX: ExprInt(0, regs.RAX.size),
+    })
+
+# Symbolic naive version of _libc_start_main
+
+def xxx___libc_start_main_symb(dse):
+    # ['RDI', 'RSI', 'RDX', 'RCX', 'R8', 'R9']
+    # main, argc, argv, ...
+    regs = dse.ir_arch.arch.regs
+    top_stack = dse.eval_expr(regs.RSP)
+    main_addr = dse.eval_expr(regs.RDI)
+    argc = dse.eval_expr(regs.RSI)
+    argv = dse.eval_expr(regs.RDX)
+    hlt_addr = ExprInt(0x1337beef, 64)
+
+    dse.update_state({
+        ExprMem(top_stack, 64): hlt_addr,
+        regs.RDI: argc,
+        regs.RSI: argv,
+        dse.ir_arch.IRDst: main_addr,
+        dse.ir_arch.pc: main_addr,
+    })
+
+# Stop the execution on puts and get back the corresponding string
+class FinnishOn(Exception):
+
+    def __init__(self, string):
+        self.string = string
+        super(FinnishOn, self).__init__()
+
+def xxx_puts_symb(dse):
+    string = dse.jitter.get_str_ansi(dse.jitter.cpu.RDI)
+    raise FinnishOn(string)
+
+
+done = set([]) # Set of jump address already handled
+todo = set([""]) # Set of file content to test
+
+class DSEGenFile(DSEPathConstraint):
+    """DSE with a specific solution creation:
+    The solution is the content of the FILE to be read
+
+    The politics of exploration is the branch coverage: create a solution only
+    if the target address has never been seen
+    """
+
+    def handle_solution(self, model, destination):
+        global todo, done
+        assert destination.is_int()
+
+        if destination in done:
+            # Skip this path, already treated
+            return
+
+        finfo = FILE_to_info_symb[FILE_stream]
+
+        # Build corresponding file
+        out = ""
+        fsize = max(model.eval(self.z3_trans.from_expr(FILE_size)).as_long(),
+                    len(finfo.gen_bytes))
+        for index in xrange(fsize):
+            try:
+                byteid = finfo.gen_bytes[index]
+                out += chr(model.eval(self.z3_trans.from_expr(byteid)).as_long())
+            except (KeyError, AttributeError) as _:
+                # Default value if there is no constraint on current byte
+                out += "\x00"
+
+        todo.add(out)
+        done.add(destination)
+
+# Instanciate the DSE engine
+machine = Machine("x86_64")
+dse = DSEGenFile(machine)
+
+# Attach to the jitter
+dse.attach(sb.jitter)
+
+# Update the jitter state: df is read, but never set
+# Approachs: specific or generic
+# - Specific:
+#   df_value = ExprInt(sb.jitter.cpu.df, dse.ir_arch.arch.regs.df.size)
+#   dse.update_state({
+#       dse.ir_arch.arch.regs.df: df_value
+#   })
+# - Generic:
+dse.update_state_from_concrete()
+
+# Add constraint on file size, we don't want to generate too big FILE
+z3_file_size = dse.z3_trans.from_expr(FILE_size)
+dse.cur_solver.add(0 < z3_file_size)
+dse.cur_solver.add(z3_file_size < 0x10)
+
+# Register symbolic stubs for extern functions (xxx_puts_symb, ...)
+dse.add_lib_handler(sb.libs, globals())
+
+# Automatic exploration of solution
+
+## Save the current clean state, before any computation of the FILE content
+snapshot = dse.take_snapshot()
+found = False
+
+while todo:
+    # Prepare a solution to try, based on the clean state
+    file_content = todo.pop()
+    print "CUR: %r" % file_content
+    open("test.txt", "w").write(file_content)
+    dse.restore_snapshot(snapshot)
+    FILE_to_info.clear()
+    FILE_to_info_symb.clear()
+
+    # Play the current file
+    try:
+        sb.run()
+    except FinnishOn as finnish_info:
+        print finnish_info.string
+        if finnish_info.string == "OK":
+            # Stop if the expected result is found
+            found = True
+            break
+
+# Assert that the result has been found
+assert found == True
+print "FOUND !"
+
+# Replay for real
+print "Trying to launch the binary without Miasm"
+crackme = subprocess.Popen([options.filename], stdout=subprocess.PIPE,
+                           stderr=subprocess.PIPE)
+stdout, stderr = crackme.communicate()
+assert not stderr
+stdout = stdout.strip()
+print stdout
+assert stdout == "OK"
diff --git a/test/test_all.py b/test/test_all.py
index 9b3f2dc1..633798f0 100755
--- a/test/test_all.py
+++ b/test/test_all.py
@@ -585,6 +585,17 @@ for options, nb_sol, tag in [([], 4, []),
                                  depends=[test_x86_32_if_reg],
                                  tags=tag)
 
+dse_crackme_out = Example.get_sample("dse_crackme.c")[:-2]
+dse_crackme = ExampleSymbolExec([Example.get_sample("dse_crackme.c"),
+                                 "-o", dse_crackme_out],
+                                products=[dse_crackme_out],
+                                executable="cc")
+testset += dse_crackme
+testset += ExampleSymbolExec(["dse_crackme.py", dse_crackme_out],
+                             depends=[dse_crackme],
+                             products=["test.txt"],
+                             tags=[TAGS["z3"]])
+
 ## Jitter
 class ExampleJitter(Example):
     """Jitter examples specificities: