added name field to executables for easy debugging

riscemu/Executable.py

@@ -59,6 +59,7 @@ class Executable:
     sections: Dict[str, MemorySection]
     symbols: Dict[str, Tuple[str, int]]
     stack_pref: Optional[int]
+    name: str
 
     def __repr__(self):
         return "{}(sections = {}, symbols = {}, stack = {}, run_ptr = {})".format(
@@ -129,6 +130,7 @@ class LoadedMemorySection:
     size: int
     content: Union[List[LoadedInstruction], bytearray] = field(repr=False)
     flags: MemoryFlags
+    owner: str
 
     def read(self, offset: int, size: int):
         if offset < 0:
@@ -220,6 +222,7 @@ class LoadedExecutable:
     It still holds a symbol table, that is not accessible memory since I don't want to deal with
     binary strings in memory etc.
     """
+    name: str
     base_addr: int
     sections_by_name: Dict[str, LoadedMemorySection]
     sections: List[LoadedMemorySection]
@@ -228,6 +231,7 @@ class LoadedExecutable:
     stack_heap: Tuple[int, int]  # pointers to stack and heap, are nullptr if no stack/heap is available
 
     def __init__(self, exe: Executable, base_addr: int):
+        self.name = exe.name
         self.base_addr = base_addr
         self.sections = list()
         self.sections_by_name = dict()
@@ -240,7 +244,8 @@ class LoadedExecutable:
                 base_addr,
                 exe.stack_pref,
                 bytearray(exe.stack_pref),
-                MemoryFlags(read_only=False, executable=False)
+                MemoryFlags(read_only=False, executable=False),
+                self.name
             ))
             self.stack_heap = (self.base_addr, self.base_addr + exe.stack_pref)
         else:
@@ -253,7 +258,8 @@ class LoadedExecutable:
                 curr,
                 sec.size,
                 sec.continuous_content(self),
-                sec.flags
+                sec.flags,
+                self.name
             )
             self.sections.append(loaded_sec)
             self.sections_by_name[loaded_sec.name] = loaded_sec
@@ -269,8 +275,12 @@ class LoadedExecutable:
         run_ptr_sec, run_ptr_off = exe.run_ptr
         self.run_ptr = self.sections_by_name[run_ptr_sec].base + run_ptr_off
 
-        print("successfully loaded binary\n\tsize: {}\n\tsections: {}\n\trun_ptr: 0x{:08x}".format(
+    def __repr__(self):
+        return '{}[{}](base=0x{:08X}, size={}bytes, sections={}, run_ptr=0x{:08X})'.format(
+            self.__class__.__name__,
+            self.name,
+            self.base_addr,
             self.size,
             " ".join(self.sections_by_name.keys()),
             self.run_ptr
-        ))
+        )

riscemu/ExecutableParser.py

@@ -27,7 +27,7 @@ class ExecutableParser:
                 self.handle_symbol(token)
             elif isinstance(token, RiscVPseudoOpToken):
                 self.handle_pseudo_op(token)
-        return self
+        return self.get_execuable()
 
     def get_execuable(self):
         start_ptr = ('text', 0)
@@ -35,7 +35,7 @@ class ExecutableParser:
             start_ptr = self.symbols['_start']
         elif 'main' in self.symbols:
             start_ptr = self.symbols['main']
-        return Executable(start_ptr, self.sections, self.symbols, self.stack_pref)
+        return Executable(start_ptr, self.sections, self.symbols, self.stack_pref, self.tokenizer.name)
 
     def parse_instruction(self, ins: 'RiscVInstructionToken'):
         if self.active_section is None:

riscemu/MMU.py

@@ -43,6 +43,8 @@ class MMU:
         for sec in loaded_bin.sections:
             self.sections.append(sec)
 
+        print("Successfully loaded {}".format(loaded_bin))
+
         return loaded_bin
 
     def get_sec_containing(self, addr: int):

riscemu/Tokenizer.py

@@ -78,10 +78,16 @@ def split_accepting_quotes(string, at=REG_ARG_SPLIT, quotes=('"', "'")):
 
 
 class RiscVInput:
-    def __init__(self, content: str):
+    def __init__(self, content: str, name: str):
         self.content = content
         self.pos = 0
         self.len = len(content)
+        self.name = name
+
+    @staticmethod
+    def from_file(src: str):
+        with open(src, 'r') as f:
+            return RiscVInput(f.read(), src)
 
     def peek(self, offset: int = 0, size: int = 1, regex: re.Pattern = None, text: str = None, regex_group: int = 0):
         at = self.pos + offset
@@ -89,7 +95,7 @@ class RiscVInput:
         if regex:
             if not isinstance(regex, re.Pattern):
                 print("uncompiled regex passed to peek!")
-                reges = re.compile(regex)
+                regex = re.compile(regex)
             match = regex.match(self.content[at:])
             if match is None:
                 return None
@@ -241,6 +247,7 @@ class RiscVTokenizer:
     def __init__(self, input: RiscVInput):
         self.input = input
         self.tokens: List[RiscVToken] = []
+        self.name = input.name
 
     def tokenize(self):
         while self.input.has_next():

riscemu/__main__.py

@@ -20,7 +20,6 @@ if __name__ == '__main__':
 
     args = parser.parse_args()
 
-    print(args)
 
     cfg = RunConfig(
         color=not args.no_color,
@@ -34,10 +33,7 @@ if __name__ == '__main__':
         FMT_NONE = ""
         FMT_PRINT = ""
 
-    with open(args.file, 'r') as f:
+    tk = RiscVTokenizer(RiscVInput.from_file(args.file))
-        asm = f.read()
-
-    tk = RiscVTokenizer(RiscVInput(asm))
     tk.tokenize()
 
     if args.print_tokens:
@@ -45,9 +41,7 @@ if __name__ == '__main__':
         for token in tk.tokens:
             print(token)
 
-    executable = ExecutableParser(tk).parse().get_execuable()
+    executable = ExecutableParser(tk).parse()
-
-    print(FMT_PRINT + "Executable:" + FMT_NONE, executable)
 
     cpu = CPU(cfg)
     le = cpu.load(executable)