add Ethereum VM architecture

.gitignore

@@ -67,6 +67,7 @@ tests/test_customized_mnem
 tests/test_m68k
 tests/test_tms320c64x
 tests/test_m680x
+tests/test_evm
 
 # vim tmp file
 *.swp

CMakeLists.txt

@@ -27,8 +27,8 @@ option(CAPSTONE_BUILD_DIET "Build diet library" OFF)
 option(CAPSTONE_BUILD_TESTS "Build tests" ON)
 option(CAPSTONE_USE_DEFAULT_ALLOC "Use default memory allocation functions" ON)
 
-set(SUPPORTED_ARCHITECTURES ARM ARM64 M68K MIPS PPC SPARC SYSZ XCORE X86 TMS320C64X M680X)
-set(SUPPORTED_ARCHITECTURE_LABELS ARM ARM64 M68K MIPS PowerPC Sparc SystemZ XCore x86 TMS320C64x M680x)
+set(SUPPORTED_ARCHITECTURES ARM ARM64 M68K MIPS PPC SPARC SYSZ XCORE X86 TMS320C64X M680X EVM)
+set(SUPPORTED_ARCHITECTURE_LABELS ARM ARM64 M68K MIPS PowerPC Sparc SystemZ XCore x86 TMS320C64x M680x EVM)
 
 list(LENGTH SUPPORTED_ARCHITECTURES count)
 math(EXPR count "${count}-1")
@@ -364,6 +364,23 @@ if (CAPSTONE_M680X_SUPPORT)
     set(TEST_SOURCES ${TEST_SOURCES} test_m680x.c)
 endif ()
 
+if (CAPSTONE_EVM_SUPPORT)
+    add_definitions(-DCAPSTONE_HAS_EVM)
+    set(SOURCES_EVM
+        arch/EVM/EVMDisassembler.c
+        arch/EVM/EVMInstPrinter.c
+        arch/EVM/EVMMapping.c
+        arch/EVM/EVMModule.c
+    )
+    set(HEADERS_EVM
+        arch/EVM/EVMDisassembler.h
+        arch/EVM/EVMInstPrinter.h
+        arch/EVM/EVMMapping.h
+        arch/EVM/EVMMappingInsn.inc
+        )
+    set(TEST_SOURCES ${TEST_SOURCES} test_evm.c)
+endif ()
+
 if (CAPSTONE_OSXKERNEL_SUPPORT)
     add_definitions(-DCAPSTONE_HAS_OSXKERNEL)
 endif ()
@@ -381,6 +398,7 @@ set(ALL_SOURCES
     ${SOURCES_M68K}
     ${SOURCES_TMS320C64X}
     ${SOURCES_M680X}
+    ${SOURCES_EVM}
     )
 
 set(ALL_HEADERS
@@ -397,6 +415,7 @@ set(ALL_HEADERS
     ${HEADERS_M68K}
     ${HEADERS_TMS320C64X}
     ${HEADERS_M680X}
+    ${HEADERS_EVM}
     )
 
 include_directories("${PROJECT_SOURCE_DIR}/include")
@@ -473,6 +492,7 @@ source_group("Source\\XCore" FILES ${SOURCES_XCORE})
 source_group("Source\\M68K" FILES ${SOURCES_M68K})
 source_group("Source\\TMS320C64x" FILES ${SOURCES_TMS320C64X})
 source_group("Source\\M680X" FILES ${SOURCES_M680X})
+source_group("Source\\EVM" FILES ${SOURCES_EVM})
 
 source_group("Include\\Common" FILES ${HEADERS_COMMON})
 source_group("Include\\Engine" FILES ${HEADERS_ENGINE})
@@ -487,6 +507,7 @@ source_group("Include\\XCore" FILES ${HEADERS_XCORE})
 source_group("Include\\M68K" FILES ${HEADERS_M68K})
 source_group("Include\\TMS320C64x" FILES ${HEADERS_TMS320C64X})
 source_group("Include\\M680X" FILES ${HEADERS_MC680X})
+source_group("Include\\EVM" FILES ${HEADERS_EVM})
 
 ### test library 64bit routine:
 get_property(LIB64 GLOBAL PROPERTY FIND_LIBRARY_USE_LIB64_PATHS)

COMPILE.TXT

@@ -17,12 +17,12 @@ Capstone requires no prerequisite packages, so it is easy to compile & install.
 
 (0) Tailor Capstone to your need.
 
-  Out of 9 archtitectures supported by Capstone (Arm, Arm64, M68K, Mips, PPC,
-  Sparc, SystemZ, XCore & X86), if you just need several selected archs, choose
-  which ones you want to compile in by editing "config.mk" before going to
-  next steps.
+  Out of 12 archtitectures supported by Capstone (Arm, Arm64, M68K, Mips, PPC,
+  Sparc, SystemZ, XCore, X86, M680X, TMS320C64x & EVM), if you just need several
+  selected archs, choose the ones you want to compile in by editing "config.mk"
+  before going to next steps.
 
-  By default, all 9 architectures are compiled.
+  By default, all 12 architectures are compiled.
 
   The other way of customize Capstone without having to edit config.mk is to
   pass the desired options on the commandline to ./make.sh. Currently,
@@ -87,11 +87,14 @@ Capstone requires no prerequisite packages, so it is easy to compile & install.
 	/usr/include/capstone/x86.h
 	/usr/include/capstone/arm.h
 	/usr/include/capstone/arm64.h
+	/usr/include/capstone/evm.h
 	/usr/include/capstone/m68k.h
+	/usr/include/capstone/m680x.h
 	/usr/include/capstone/mips.h
 	/usr/include/capstone/ppc.h
 	/usr/include/capstone/sparc.h
 	/usr/include/capstone/systemz.h
+	/usr/include/capstone/tms320c64x.h
 	/usr/include/capstone/xcore.h
 	/usr/include/capstone/platform.h
 	/usr/lib/libcapstone.so (for Linux/*nix), or /usr/lib/libcapstone.dylib (OSX)

COMPILE_CMAKE.TXT

@@ -16,8 +16,8 @@ Get CMake for free from http://www.cmake.org.
 
 (0) Tailor Capstone to your need.
 
-  Out of 9 archtitectures supported by Capstone (Arm, Arm64, M68K, Mips, PPC,
-  Sparc, SystemZ, X86 & XCore), if you just need several selected archs,
+  Out of 12 archtitectures supported by Capstone (Arm, Arm64, M68K, Mips, PPC,
+  Sparc, SystemZ, X86, XCore, M680X, TMS320C64x & EVM), if you just need several selected archs,
   run "cmake" with the unwanted archs disabled (set to 0) as followings.
 
   - CAPSTONE_ARM_SUPPORT: support ARM. Run cmake with -DCAPSTONE_ARM_SUPPORT=0 to remove ARM.
@@ -30,6 +30,9 @@ Get CMake for free from http://www.cmake.org.
   - CAPSTONE_SYSZ_SUPPORT: support SystemZ. Run cmake with -DCAPSTONE_SYSZ_SUPPORT=0 to remove SystemZ.
   - CAPSTONE_XCORE_SUPPORT: support XCore. Run cmake with -DCAPSTONE_XCORE_SUPPORT=0 to remove XCore.
   - CAPSTONE_X86_SUPPORT: support X86. Run cmake with -DCAPSTONE_X86_SUPPORT=0 to remove X86.
+  - CAPSTONE_X86_TMS320C64X: support TMS320C64X. Run cmake with -DCAPSTONE_TMS320C64X_SUPPORT=0 to remove TMS320C64X.
+  - CAPSTONE_X86_M680X: support M680X. Run cmake with -DCAPSTONE_M680X_SUPPORT=0 to remove M680X.
+  - CAPSTONE_X86_EVM: support EVM. Run cmake with -DCAPSTONE_EVM_SUPPORT=0 to remove EVM.
 
   By default, all 10 architectures are compiled in.
 

MCInst.h

@@ -109,6 +109,7 @@ struct MCInst {
 	uint8_t ac_idx;
 	uint8_t popcode_adjust;   // Pseudo X86 instruction adjust
 	char assembly[8];	// for special instruction, so that we dont need printer
+	unsigned char evm_data[32];	// for EVM PUSH operand
 };
 
 void MCInst_Init(MCInst *inst);

Makefile

@@ -235,9 +235,21 @@ ifneq (,$(findstring m680x,$(CAPSTONE_ARCHS)))
 	LIBOBJ_M680X += $(LIBSRC_M680X:%.c=$(OBJDIR)/%.o)
 endif
 
+
+DEP_EVM =
+DEP_EVM += $(wildcard arch/EVM/EVM*.inc)
+
+LIBOBJ_EVM =
+ifneq (,$(findstring evm,$(CAPSTONE_ARCHS)))
+	CFLAGS += -DCAPSTONE_HAS_EVM
+	LIBSRC_EVM += $(wildcard arch/EVM/EVM*.c)
+	LIBOBJ_EVM += $(LIBSRC_EVM:%.c=$(OBJDIR)/%.o)
+endif
+
+
 LIBOBJ =
 LIBOBJ += $(OBJDIR)/cs.o $(OBJDIR)/utils.o $(OBJDIR)/SStream.o $(OBJDIR)/MCInstrDesc.o $(OBJDIR)/MCRegisterInfo.o
-LIBOBJ += $(LIBOBJ_ARM) $(LIBOBJ_ARM64) $(LIBOBJ_M68K) $(LIBOBJ_MIPS) $(LIBOBJ_PPC) $(LIBOBJ_SPARC) $(LIBOBJ_SYSZ) $(LIBOBJ_X86) $(LIBOBJ_XCORE) $(LIBOBJ_TMS320C64X) $(LIBOBJ_M680X)
+LIBOBJ += $(LIBOBJ_ARM) $(LIBOBJ_ARM64) $(LIBOBJ_M68K) $(LIBOBJ_MIPS) $(LIBOBJ_PPC) $(LIBOBJ_SPARC) $(LIBOBJ_SYSZ) $(LIBOBJ_X86) $(LIBOBJ_XCORE) $(LIBOBJ_TMS320C64X) $(LIBOBJ_M680X) $(LIBOBJ_EVM)
 LIBOBJ += $(OBJDIR)/MCInst.o
 
 
@@ -363,6 +375,7 @@ $(LIBOBJ_X86): $(DEP_X86)
 $(LIBOBJ_XCORE): $(DEP_XCORE)
 $(LIBOBJ_TMS320C64X): $(DEP_TMS320C64X)
 $(LIBOBJ_M680X): $(DEP_M680X)
+$(LIBOBJ_EVM): $(DEP_EVM)
 
 ifeq ($(CAPSTONE_STATIC),yes)
 $(ARCHIVE): $(LIBOBJ)

README.md

@@ -10,8 +10,8 @@ disasm engine for binary analysis and reversing in the security community.
 Created by Nguyen Anh Quynh, then developed and maintained by a small community,
 Capstone offers some unparalleled features:
 
-- Support multiple hardware architectures: ARM, ARM64 (ARMv8), M68K, Mips, PPC,
-  Sparc, SystemZ, TMS320C64X, M680X, XCore and X86 (including X86_64).
+- Support multiple hardware architectures: ARM, ARM64 (ARMv8), Ethereum VM, M68K,
+  Mips, PPC, Sparc, SystemZ, TMS320C64X, M680X, XCore and X86 (including X86_64).
 
 - Having clean/simple/lightweight/intuitive architecture-neutral API.
 

bindings/const_generator.py

@@ -5,7 +5,7 @@
 
 INCL_DIR = '../include/capstone/'
 
-include = [ 'arm.h', 'arm64.h', 'm68k.h', 'mips.h', 'x86.h', 'ppc.h', 'sparc.h', 'systemz.h', 'xcore.h', 'tms320c64x.h', 'm680x.h' ]
+include = [ 'arm.h', 'arm64.h', 'm68k.h', 'mips.h', 'x86.h', 'ppc.h', 'sparc.h', 'systemz.h', 'xcore.h', 'tms320c64x.h', 'm680x.h', 'evm.h' ]
 
 template = {
     'java': {
@@ -25,6 +25,7 @@
             'xcore.h': 'Xcore',
             'tms320c64x.h': 'TMS320C64x',
             'm680x.h': 'M680x',
+            'evm.h': 'Evm',
             'comment_open': '\t//',
             'comment_close': '',
         },
@@ -45,6 +46,7 @@
             'xcore.h': 'xcore',
             'tms320c64x.h': 'tms320c64x',
             'm680x.h': 'm680x',
+            'evm.h': 'evm',
             'comment_open': '#',
             'comment_close': '',
         },
@@ -65,6 +67,7 @@
             'xcore.h': 'xcore',
             'tms320c64x.h': 'tms320c64x',
             'm680x.h': 'm680x',
+            'evm.h': 'evm',
             'comment_open': '(*',
             'comment_close': ' *)',
         },

bindings/python/capstone/__init__.py

@@ -34,6 +34,7 @@
     'CS_ARCH_M68K',
     'CS_ARCH_TMS320C64X',
     'CS_ARCH_M680X',
+    'CS_ARCH_EVM',
     'CS_ARCH_ALL',
 
     'CS_MODE_LITTLE_ENDIAN',
@@ -149,7 +150,8 @@
 CS_ARCH_M68K = 8
 CS_ARCH_TMS320C64X = 9
 CS_ARCH_M680X = 10
-CS_ARCH_MAX = 11
+CS_ARCH_EVM = 11
+CS_ARCH_MAX = 12
 CS_ARCH_ALL = 0xFFFF
 
 # disasm mode
@@ -327,7 +329,7 @@ def copy_ctypes_list(src):
     return [copy_ctypes(n) for n in src]
 
 # Weird import placement because these modules are needed by the below code but need the above functions
-from . import arm, arm64, m68k, mips, ppc, sparc, systemz, x86, xcore, tms320c64x, m680x
+from . import arm, arm64, m68k, mips, ppc, sparc, systemz, x86, xcore, tms320c64x, m680x, evm
 
 class _cs_arch(ctypes.Union):
     _fields_ = (
@@ -342,6 +344,7 @@ class _cs_arch(ctypes.Union):
         ('xcore', xcore.CsXcore),
         ('tms320c64x', tms320c64x.CsTMS320C64x),
         ('m680x', m680x.CsM680x),
+        ('evm', evm.CsEvm),
     )
 
 class _cs_detail(ctypes.Structure):
@@ -648,6 +651,8 @@ def __gen_detail(self):
             (self.condition, self.funit, self.parallel, self.operands) = tms320c64x.get_arch_info(self._raw.detail.contents.arch.tms320c64x)
         elif arch == CS_ARCH_M680X:
             (self.flags, self.operands) = m680x.get_arch_info(self._raw.detail.contents.arch.m680x)
+        elif arch == CS_ARCH_EVM:
+            (self.pop, self.push, self.fee) = evm.get_arch_info(self._raw.detail.contents.arch.evm)
 
 
     def __getattr__(self, name):
@@ -1072,7 +1077,7 @@ def debug():
     archs = { "arm": CS_ARCH_ARM, "arm64": CS_ARCH_ARM64, "m68k": CS_ARCH_M68K, \
         "mips": CS_ARCH_MIPS, "ppc": CS_ARCH_PPC, "sparc": CS_ARCH_SPARC, \
         "sysz": CS_ARCH_SYSZ, 'xcore': CS_ARCH_XCORE, "tms320c64x": CS_ARCH_TMS320C64X, \
-        "m680x": CS_ARCH_M680X }
+        "m680x": CS_ARCH_M680X, 'evm': CS_ARCH_EVM }
 
     all_archs = ""
     keys = archs.keys()

bindings/python/setup_cython.py

@@ -40,7 +40,7 @@
 compile_args = ['-O3', '-fomit-frame-pointer', '-I' + HEADERS_DIR]
 link_args = ['-L' + LIBS_DIR]
 
-ext_module_names = ['arm', 'arm_const', 'arm64', 'arm64_const', 'm68k', 'm68k_const', 'm680x', 'm680x_const', 'mips', 'mips_const', 'ppc', 'ppc_const', 'x86', 'x86_const', 'sparc', 'sparc_const', 'systemz', 'sysz_const', 'xcore', 'xcore_const', 'tms320c64x', 'tms320c64x_const']
+ext_module_names = ['arm', 'arm_const', 'arm64', 'arm64_const', 'm68k', 'm68k_const', 'm680x', 'm680x_const', 'mips', 'mips_const', 'ppc', 'ppc_const', 'x86', 'x86_const', 'sparc', 'sparc_const', 'systemz', 'sysz_const', 'xcore', 'xcore_const', 'tms320c64x', 'tms320c64x_const', 'evm', 'evm_const' ]
 
 ext_modules = [Extension("capstone.ccapstone",
                          ["pyx/ccapstone.pyx"],

config.mk

@@ -4,7 +4,7 @@
 ################################################################################
 # Specify which archs you want to compile in. By default, we build all archs.
 
-CAPSTONE_ARCHS ?= arm aarch64 m68k mips powerpc sparc systemz x86 xcore tms320c64x m680x
+CAPSTONE_ARCHS ?= arm aarch64 m68k mips powerpc sparc systemz x86 xcore tms320c64x m680x evm
 
 
 ################################################################################

cs.c

@@ -66,6 +66,7 @@ extern void Sparc_enable(void);
 extern void SystemZ_enable(void);
 extern void XCore_enable(void);
 extern void TMS320C64x_enable(void);
+extern void EVM_enable(void);
 
 static void archs_enable(void)
 {
@@ -107,6 +108,9 @@ static void archs_enable(void)
 #ifdef CAPSTONE_HAS_TMS320C64X
 	TMS320C64x_enable();
 #endif
+#ifdef CAPSTONE_HAS_EVM
+	EVM_enable();
+#endif
 
 
 	initialized = true;
@@ -187,7 +191,7 @@ bool CAPSTONE_API cs_support(int query)
 				(1 << CS_ARCH_PPC) | (1 << CS_ARCH_SPARC) |
 				(1 << CS_ARCH_SYSZ) | (1 << CS_ARCH_XCORE) |
 				(1 << CS_ARCH_M68K) | (1 << CS_ARCH_TMS320C64X) |
-				(1 << CS_ARCH_M680X));
+				(1 << CS_ARCH_M680X) | (1 << CS_ARCH_EVM));
 
 	if ((unsigned int)query < CS_ARCH_MAX)
 		return all_arch & (1 << query);
@@ -447,6 +451,9 @@ static uint8_t skipdata_size(cs_struct *handle)
 		case CS_ARCH_M680X:
 			// M680X alignment is 1.
 			return 1;
+		case CS_ARCH_EVM:
+			// EVM alignment is 1.
+			return 1;
 	}
 }
 
@@ -1153,6 +1160,13 @@ int CAPSTONE_API cs_op_count(csh ud, const cs_insn *insn, unsigned int op_type)
 				if (insn->detail->m680x.operands[i].type == (m680x_op_type)op_type)
 					count++;
 			break;
+		case CS_ARCH_EVM:
+#if 0
+			for (i = 0; i < insn->detail->evm.op_count; i++)
+				if (insn->detail->evm.operands[i].type == (evm_op_type)op_type)
+					count++;
+#endif
+			break;
 	}
 
 	return count;
@@ -1278,6 +1292,16 @@ int CAPSTONE_API cs_op_index(csh ud, const cs_insn *insn, unsigned int op_type,
 					return i;
 			}
 			break;
+		case CS_ARCH_EVM:
+#if 0
+			for (i = 0; i < insn->detail->evm.op_count; i++) {
+				if (insn->detail->evm.operands[i].type == (evm_op_type)op_type)
+					count++;
+				if (count == post)
+					return i;
+			}
+#endif
+			break;
 	}
 
 	return -1;