Fix gymrat disasm (#367)

* Fix gymrat disasm

* allow kwargs for _lifter()

* [pre-commit.ci] auto fixes from pre-commit.com hooks

for more information, see https://pre-commit.ci

* Revert "allow kwargs for _lifter()"

This reverts commit a583072.

* fix gymrat disasm by overriding lift

* Revert "fix gymrat disasm by overriding lift"

This reverts commit 6f83b80.

* add kwargs in Lifter for GymratLifter

* Return self.irsb instead of result of self._lift()

* Update ARMSpotter to proper semantics

* stick to the semantics of lift to return irsb

---------

Co-authored-by: pre-commit-ci[bot] <66853113+pre-commit-ci[bot]@users.noreply.github.com>
Co-authored-by: Kevin Phoenix <[email protected]>

pyvex/lifting/gym/arm_spotter.py

@@ -415,12 +415,12 @@ def __init__(self, *args):
         super().__init__(*args)
         self.thumb: bool = False
 
-    def _lift(self, disassemble=False, dump_irsb=False):
+    def _lift(self):
         if self.irsb.addr & 1:
             # Thumb!
             self.instrs = self.thumb_instrs
             self.thumb = True
         else:
             self.instrs = self.arm_instrs
             self.thumb = False
-        super()._lift(disassemble, dump_irsb)
+        super()._lift()

pyvex/lifting/lifter.py

@@ -20,6 +20,8 @@ class Lifter:
         "addr",
         "cross_insn_opt",
         "load_from_ro_regions",
+        "disasm",
+        "dump_irsb",
     )
 
     """
@@ -60,6 +62,8 @@ def lift(
         collect_data_refs=False,
         cross_insn_opt=True,
         load_from_ro_regions=False,
+        disasm=False,
+        dump_irsb=False,
     ):
         """
         Wrapper around the `_lift` method on Lifters. Should not be overridden in child classes.
@@ -80,6 +84,8 @@ def lift(
         :param skip_stmts:          Should the lifter skip transferring IRStmts from C to Python.
         :param collect_data_refs:   Should the LibVEX lifter collect data references in C.
         :param cross_insn_opt:      If cross-instruction-boundary optimizations are allowed or not.
+        :param disasm:              Should the GymratLifter generate disassembly during lifting.
+        :param dump_irsb:           Should the GymratLifter log the lifted IRSB.
         """
         irsb = IRSB.empty_block(self.arch, self.addr)
         self.data = data
@@ -95,6 +101,8 @@ def lift(
         self.irsb = irsb
         self.cross_insn_opt = cross_insn_opt
         self.load_from_ro_regions = load_from_ro_regions
+        self.disasm = disasm
+        self.dump_irsb = dump_irsb
         self._lift()
         return self.irsb
 

pyvex/lifting/util/lifter_helper.py

@@ -38,6 +38,7 @@ class GymratLifter(Lifter):
         "bitstrm",
         "errors",
         "thedata",
+        "disassembly",
     )
 
     REQUIRE_DATA_PY = True
@@ -48,6 +49,7 @@ def __init__(self, *args):
         self.bitstrm = None
         self.errors = None
         self.thedata = None
+        self.disassembly = None
 
     def create_bitstrm(self):
         self.bitstrm = bitstring.ConstBitStream(bytes=self.thedata)
@@ -97,7 +99,7 @@ def decode(self):
             log.exception(f"Error decoding block at offset {bytepos:#x} (address {addr:#x}):")
             raise
 
-    def _lift(self, disassemble=False, dump_irsb=False):
+    def _lift(self):
         self.thedata = (
             self.data[: self.max_bytes]
             if isinstance(self.data, (bytes, bytearray, memoryview))
@@ -106,8 +108,8 @@ def _lift(self, disassemble=False, dump_irsb=False):
         log.debug(repr(self.thedata))
         instructions = self.decode()
 
-        if disassemble:
-            return [instr.disassemble() for instr in instructions]
+        if self.disasm:
+            self.disassembly = [instr.disassemble() for instr in instructions]
         self.irsb.jumpkind = JumpKind.Invalid
         irsb_c = IRSBCustomizer(self.irsb)
         log.debug("Decoding complete.")
@@ -127,7 +129,7 @@ def _lift(self, disassemble=False, dump_irsb=False):
             dst_ty = vex_int_class(irsb_c.irsb.arch.bits).type
             irsb_c.irsb.next = irsb_c.mkconst(dst, dst_ty)
         log.debug(self.irsb._pp_str())
-        if dump_irsb:
+        if self.dump_irsb:
             self.irsb.pp()
         return self.irsb
 
@@ -136,11 +138,13 @@ def pp_disas(self):
         insts = self.disassemble()
         for addr, name, args in insts:
             args_str = ",".join(str(a) for a in args)
-            disasstr += f"{addr:0#8x}:\t{name} {args_str}\n"
+            disasstr += f"{addr:#08x}:\t{name} {args_str}\n"
         print(disasstr)
 
     def error(self):
         return self.errors
 
     def disassemble(self):
-        return self.lift(disassemble=True)
+        if self.disassembly is None:
+            self.lift(self.data, disasm=True)
+        return self.disassembly