Fix: msg.data memory allocation removed in favor of builtin function …

…handling (vyperlang#2419)

Refactored the msg.data environment variable.

Instead of handling each distinctive use case in expr.py, uses are handled in the respective functions which operate on msg.data.

Len builtin simply returns calldatasize
Slice builtin returns a bytes array containing the sliced portion of calldata
The length argument to slice should be a compile time literal as calldatasize is unbounded, and we need a maxsize
on the return type of the slice function instead of defaulting to the maxsize of msg.data

The test suite for msg.data is unchanged, so passing tests verify this refactoring worked.

vyper/builtin_functions/functions.py

@@ -262,6 +262,12 @@ def build_LLL(self, expr, args, kwargs, context):
                     "Invalid start / length values needs to be between 0 and 32.", expr,
                 )
             sub_typ_maxlen = 32
+        elif sub.location == "calldata":
+            # if we are slicing msg.data, the length should
+            # be a constant, since msg.data can be of dynamic length
+            # we can't use it's length as the maxlen
+            assert isinstance(length.value, int) # sanity check
+            sub_typ_maxlen = length.value
         else:
             sub_typ_maxlen = sub.typ.maxlen
 
@@ -280,6 +286,15 @@ def build_LLL(self, expr, args, kwargs, context):
             adj_sub = LLLnode.from_list(
                 ["add", sub, ["add", ["div", "_start", 32], 1]], typ=sub.typ, location=sub.location,
             )
+        elif sub.location == "calldata":
+            node = [
+                "seq",
+                ["assert", ["le", ["add", start, length], "calldatasize"]],  # runtime bounds check
+                ["mstore", np, length],
+                ["calldatacopy", np + 32, start, length],
+                np,
+            ]
+            return LLLnode.from_list(node, typ=ByteArrayType(length.value), location="memory")
         else:
             adj_sub = LLLnode.from_list(
                 ["add", sub, ["add", ["sub", "_start", ["mod", "_start", 32]], 32]],
@@ -351,6 +366,10 @@ def evaluate(self, node):
         return vy_ast.Int.from_node(node, value=length)
 
     def build_LLL(self, node, context):
+        if isinstance(node.args[0], vy_ast.Attribute):
+            key = f"{node.args[0].value.id}.{node.args[0].attr}"
+            if key == "msg.data":
+                return LLLnode.from_list(["calldatasize"], typ="uint256")
         arg = Expr(node.args[0], context).lll_node
         return get_bytearray_length(arg)
 

vyper/old_codegen/expr.py

@@ -375,25 +375,7 @@ def parse_Attribute(self):
             if key == "msg.sender" and not self.context.is_internal:
                 return LLLnode.from_list(["caller"], typ="address", pos=getpos(self.expr))
             elif key == "msg.data" and not self.context.is_internal:
-                is_len = self.expr._metadata.get("is_len")
-                if is_len is True:
-                    typ = ByteArrayType(32)
-                    pos = self.context.new_internal_variable(typ)
-                    node = ["seq", ["mstore", pos, "calldatasize"], pos]
-                    return LLLnode.from_list(
-                        node, typ=typ, pos=getpos(self.expr), location="memory"
-                    )
-                size = self.expr._metadata.get("size")
-                typ = ByteArrayType(size + 32)
-                pos = self.context.new_internal_variable(typ)
-                node = [
-                    "seq",
-                    ["assert", ["le", size, "calldatasize"]],
-                    ["mstore", pos, size],
-                    ["calldatacopy", pos + 32, 0, size],
-                    pos,
-                ]
-                return LLLnode.from_list(node, typ=typ, pos=getpos(self.expr), location="memory")
+                return LLLnode(0, typ=ByteArrayType(0), location="calldata")
             elif key == "msg.value" and self.context.is_payable:
                 return LLLnode.from_list(
                     ["callvalue"], typ=BaseType("uint256"), pos=getpos(self.expr),

vyper/semantics/validation/annotation.py

@@ -55,12 +55,9 @@ def visit(self, node):
         super().visit(node)
 
     def visit_Attribute(self, node):
-        if node.get("value.id") == "msg" and node.attr == "data":
-            parent = node.get_ancestor()
-            if parent.get("func.id") == "slice":
-                node._metadata["size"] = parent.args[1].value + parent.args[2].value
-            elif parent.get("func.id") == "len":
-                node._metadata["is_len"] = True
+        # NOTE: required for msg.data node, removing borks and results in
+        # vyper.exceptions.StructureException: Cannot annotate: Attribute
+        pass
 
     def visit_AnnAssign(self, node):
         type_ = get_exact_type_from_node(node.target)

vyper/semantics/validation/local.py

@@ -186,9 +186,7 @@ def visit(self, node):
     def visit_Attribute(self, node):
         if node.get("value.id") == "msg" and node.attr == "data":
             parent = node.get_ancestor()
-            is_slice = parent.get("func.id") == "slice"
-            is_len = parent.get("func.id") == "len"
-            if not is_slice and not is_len:
+            if parent.get("func.id") not in ("slice", "len"):
                 raise SyntaxException(
                     "msg.data is only allowed inside of the slice or len functions",
                     node.node_source_code,