Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Recover and serialize types from high pcode #50

Merged
merged 7 commits into from
Oct 24, 2024
Merged

Recover and serialize types from high pcode #50

Changes from 5 commits
Commits
Show all changes
7 commits
Select commit Hold shift + click to select a range
f0a015f
ghidra: recover type information
kumarak Oct 18, 2024
dc7abee
ghidra: update serializing function for types
kumarak Oct 18, 2024
a55eea9
ghidra: refactor and address comments
kumarak Oct 19, 2024
14e8e94
ghidra: add check for null high function
kumarak Oct 21, 2024
25f0cd4
ghidra: update serializtion method for types
kumarak Oct 22, 2024
3fef34f
ghidra: small refactor and inline serializing of parameters
kumarak Oct 24, 2024
1570de0
ghidra: move null type check to label method
kumarak Oct 24, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
254 changes: 236 additions & 18 deletions scripts/ghidra/PatchestryDecompileFunctions.java
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -26,16 +26,38 @@

import ghidra.program.model.listing.Program;

import ghidra.program.model.pcode.FunctionPrototype;
import ghidra.program.model.pcode.HighFunction;
import ghidra.program.model.pcode.HighParam;
import ghidra.program.model.pcode.HighVariable;
import ghidra.program.model.pcode.PcodeBlock;
import ghidra.program.model.pcode.PcodeBlockBasic;
import ghidra.program.model.pcode.PcodeOp;
import ghidra.program.model.pcode.SequenceNumber;
import ghidra.program.model.pcode.Varnode;

import ghidra.program.model.data.AbstractFloatDataType;
import ghidra.program.model.data.AbstractIntegerDataType;
import ghidra.program.model.data.Array;
import ghidra.program.model.data.BooleanDataType;
import ghidra.program.model.data.Composite;
import ghidra.program.model.data.CategoryPath;
import ghidra.program.model.data.DataType;
import ghidra.program.model.data.DataTypeComponent;
import ghidra.program.model.data.Enum;
kumarak marked this conversation as resolved.
Show resolved Hide resolved
import ghidra.program.model.data.FunctionDefinition;
import ghidra.program.model.data.ParameterDefinition;
import ghidra.program.model.data.Pointer;
import ghidra.program.model.data.Structure;
import ghidra.program.model.data.TypeDef;
import ghidra.program.model.data.Undefined;
import ghidra.program.model.data.Union;
import ghidra.program.model.data.VoidDataType;

import ghidra.program.model.symbol.ExternalManager;

import ghidra.util.UniversalID;

import com.google.gson.stream.JsonWriter;

import java.io.BufferedWriter;
Expand All @@ -48,13 +70,17 @@
import java.nio.file.Path;

import java.util.ArrayList;
import java.util.HashMap;
import java.util.HashSet;
import java.util.List;
import java.util.Collections;
import java.util.Iterator;
import java.util.Set;
import java.util.TreeSet;

public class PatchestryDecompileFunctions extends GhidraScript {

static final int decompilation_timeout = 30;

private class PcodeSerializer extends JsonWriter {
private String arch;
Expand All @@ -65,6 +91,8 @@ private class PcodeSerializer extends JsonWriter {
private List<Function> functions;
private int original_functions_size;
private Set<Address> seen_functions;
private Set<String> seen_types;
pgoodman marked this conversation as resolved.
Show resolved Hide resolved
private List<DataType> types_to_serialize;

public PcodeSerializer(java.io.BufferedWriter writer,
String arch_, FunctionManager fm_,
Expand All @@ -80,8 +108,10 @@ public PcodeSerializer(java.io.BufferedWriter writer,
this.functions = functions_;
this.original_functions_size = functions.size();
this.seen_functions = new TreeSet<>();
this.seen_types = new HashSet<>();
pgoodman marked this conversation as resolved.
Show resolved Hide resolved
this.types_to_serialize = new ArrayList<>();
}

private static String label(Address address) throws Exception {
return address.toString(true /* show address space prefix */);
}
Expand All @@ -101,6 +131,23 @@ private static String label(PcodeBlock block) throws Exception {
private static String label(PcodeOp op) throws Exception {
return label(op.getSeqnum());
}

private String label(DataType type) throws Exception {
String name = type.getName();
pgoodman marked this conversation as resolved.
Show resolved Hide resolved
CategoryPath category = type.getCategoryPath();
String concat_type = category.toString() + name + Integer.toString(type.getLength());
String type_id = Integer.toHexString(concat_type.hashCode());

UniversalID uid = type.getUniversalID();
if (uid != null) {
type_id += Address.SEPARATOR + uid.toString();
}

if (seen_types.add(type_id)) {
types_to_serialize.add(type);
}
return type_id;
}

// Return the r-value of a varnode.
private Varnode rValueOf(Varnode node) throws Exception {
Expand Down Expand Up @@ -156,6 +203,167 @@ private HighVariable lValueOf(Varnode node) throws Exception {
return var;
}

private String getDisplayName(DataType data_type) throws Exception {
kumarak marked this conversation as resolved.
Show resolved Hide resolved
return data_type.getDisplayName().replaceAll(" ", "");
}

private void serializePointerType(Pointer ptr) throws Exception {
name("name").value(getDisplayName(ptr));
kumarak marked this conversation as resolved.
Show resolved Hide resolved
name("kind").value("pointer");
name("size").value(ptr.getLength());
DataType elem_type = ptr.getDataType();
// Pointer element type could be null; assign void type to
// element in such case
if (elem_type == null) {
kumarak marked this conversation as resolved.
Show resolved Hide resolved
elem_type = VoidDataType.dataType;
}

name("element_type").value(label(elem_type));
}

private void serializeTypedefType(TypeDef typedef) throws Exception {
name("name").value(getDisplayName(typedef));
name("kind").value("typedef");
name("size").value(typedef.getLength());

DataType base_type = typedef.getBaseDataType();
// If base type is null, raise an exception.
// TODO(kumarak): if see cases where base type is null, handle it
// by assigning default `void` type.
if (base_type == null) {
kumarak marked this conversation as resolved.
Show resolved Hide resolved
throw new Exception("Typedef base type is null: " + typedef.toString());
}
pgoodman marked this conversation as resolved.
Show resolved Hide resolved

name("base_type").value(label(base_type));
}

private void serializeArrayType(Array arr) throws Exception {
name("name").value(getDisplayName(arr));
name("kind").value("array");
name("size").value(arr.getLength());
name("num_elements").value(arr.getNumElements());
pgoodman marked this conversation as resolved.
Show resolved Hide resolved
DataType elem_type = arr.getDataType();
if (elem_type != null) {
kumarak marked this conversation as resolved.
Show resolved Hide resolved
name("element_type").value(label(elem_type));
}
}

private void serializeBuiltinType(DataType data_type, String kind) throws Exception {
name("name").value(getDisplayName(data_type));
name("size").value(data_type.getLength());
name("kind").value(kind);
}

private void serializeCompositeType(Composite data_type, String kind) throws Exception {
name("name").value(getDisplayName(data_type));
name("kind").value(kind);
name("size").value(data_type.getLength());
name("num_fields").value(data_type.getNumComponents());
kumarak marked this conversation as resolved.
Show resolved Hide resolved
name("fields").beginArray();

for (int i = 0; i < data_type.getNumComponents(); i++) {
DataTypeComponent dtc = data_type.getComponent(i);
beginObject();

name("type").value(label(dtc.getDataType()));
name("offset").value(dtc.getOffset());

if (dtc.getFieldName() != null) {
name("name").value(dtc.getFieldName().replaceAll(" ", ""));
kumarak marked this conversation as resolved.
Show resolved Hide resolved
}
endObject();
}
endArray();
}

private void serializeFunctionDefinition(FunctionDefinition fd) throws Exception {
name("name").value(fd.getDisplayName());
name("kind").value("function");
name("return_type").value(label(fd.getReturnType()));
name("has_varargs").value(fd.hasVarArgs());
kumarak marked this conversation as resolved.
Show resolved Hide resolved
ParameterDefinition[] arguments = fd.getArguments();
name("parameters").beginArray();
for (int i = 0; i < arguments.length; i++) {
beginObject();
name("name").value(arguments[i].getName());
kumarak marked this conversation as resolved.
Show resolved Hide resolved
name("size").value(arguments[i].getLength());
name("type").value(label(arguments[i].getDataType()));
endObject();
}
endArray();
}

private void serialize(DataType data_type) throws Exception {
if (data_type == null) {
nullValue();
return;
}

if (data_type instanceof Pointer) {
serializePointerType((Pointer) data_type);
} else if (data_type instanceof TypeDef) {
serializeTypedefType((TypeDef) data_type);
} else if (data_type instanceof Array) {
serializeArrayType((Array) data_type);
} else if (data_type instanceof Structure) {
serializeCompositeType((Composite) data_type, "struct");
} else if (data_type instanceof Union) {
serializeCompositeType((Composite) data_type, "union");
} else if (data_type instanceof AbstractIntegerDataType){
serializeBuiltinType(data_type, "integer");
} else if (data_type instanceof AbstractFloatDataType){
serializeBuiltinType(data_type, "float");
} else if (data_type instanceof BooleanDataType){
serializeBuiltinType(data_type, "boolean");
} else if (data_type instanceof Enum) {
serializeBuiltinType(data_type, "enum");
} else if (data_type instanceof VoidDataType) {
serializeBuiltinType(data_type, "void");
} else if (data_type instanceof Undefined || data_type.toString().contains("undefined")) {
serializeBuiltinType(data_type, "undefined");
} else if (data_type instanceof FunctionDefinition) {
serializeFunctionDefinition((FunctionDefinition) data_type);
} else {
throw new Exception("Unhandled type: " + data_type.toString());
}
}

private void serializeTypes() throws Exception {
for (int i = 0; i < types_to_serialize.size(); i++) {
DataType type = types_to_serialize.get(i);
name(label(type)).beginObject();
serialize(type);
endObject();
}

println("Total serialized types: " + types_to_serialize.size());
}

private void serialize(FunctionPrototype proto) throws Exception {
if (proto == null) {
nullValue();
return;
}

name("parameters").beginArray();
for (int i = 0; i < proto.getNumParams(); i++) {
serialize(proto.getParam(i).getHighVariable());
kumarak marked this conversation as resolved.
Show resolved Hide resolved
}
endArray();
}

private void serialize(HighVariable high_var) throws Exception {
if (high_var == null) {
nullValue();
return;
}

beginObject();
name("name").value(high_var.getName());
name("type").value(label(high_var.getDataType()));
endObject();
}

private void serialize(Varnode node) throws Exception {
if (node == null) {
assert false;
Expand All @@ -172,23 +380,18 @@ private void serialize(Varnode node) throws Exception {

beginObject();

if (node.isConstant()) {
name("type").value("const");
} else if (node.isUnique()) {
name("type").value("unique");
} else if (node.isRegister()) {
name("type").value("register");
} else if (node.isAddress()) {
name("type").value("ram");
} else if (node.getAddress().isStackAddress()) {
name("type").value("stack");
} else {
throw new Exception("Unknown Varnode kind: " + node.toString());
}

Address address = node.getAddress();
name("space").value(address.getAddressSpace().getName());
name("offset").value(node.getOffset());
name("size").value(node.getSize());

name("address").value(label(node.getAddress()));
HighVariable high_var = node.getHigh();
if (high_var != null) {
name("variable").beginObject();
name("name").value(high_var.getName());
name("type").value(label(high_var.getDataType()));
endObject();
}
endObject();
}

Expand Down Expand Up @@ -331,6 +534,12 @@ private void serialize(HighFunction function, PcodeBlockBasic block) throws Exce
private void serialize(HighFunction high_function, Function function, boolean visit_pcode) throws Exception {

name("name").value(function.getName());
FunctionPrototype proto = high_function.getFunctionPrototype();
name("prototype").beginObject();
if (proto != null) {
serialize(proto);
}
endObject();

// If we have a high P-Code function, then serialize the blocks.
if (high_function != null) {
Expand Down Expand Up @@ -363,13 +572,22 @@ public void serialize() throws Exception {
continue;
}

DecompileResults res = ifc.decompileFunction(function, 30, null);
DecompileResults res = ifc.decompileFunction(function, decompilation_timeout, null);
HighFunction high_function = res.getHighFunction();

if (high_function == null) {
continue;
}

name(label(function_address)).beginObject();
serialize(high_function, function, i < original_functions_size);
endObject();
}

// Serialize Types
name("types").beginObject();
serializeTypes();
endObject();

endObject().endObject();
}
}
Expand Down