mirror of
https://github.com/varun-r-mallya/Python-BPF.git
synced 2025-12-31 21:06:25 +00:00
Compare commits
30 Commits
| Author | SHA1 | Date | |
|---|---|---|---|
| 8e3942d38c | |||
| 3abe07c5b2 | |||
| 01bd7604ed | |||
| 7ae84a0d5a | |||
| df3f00261a | |||
| ab610147a5 | |||
| 7720fe9f9f | |||
| 7aeac86bd3 | |||
| ab1c4223d5 | |||
| c3a512d5cf | |||
| 4a60c42cd0 | |||
| b35134625b | |||
| c3db609a90 | |||
| cc626c38f7 | |||
| a8b3f4f86c | |||
| d593969408 | |||
| 6d5895ebc2 | |||
| c9ee6e4f17 | |||
| a622c53e0f | |||
| a4f1363aed | |||
| 3a819dcaee | |||
| 729270b34b | |||
| 44cbcccb6c | |||
| 253944afd2 | |||
| 54993ce5c2 | |||
| 05083bd513 | |||
| 6e4c340780 | |||
| 9dbca410c2 | |||
| 62ca3b5ffe | |||
| f263c35156 |
@ -21,17 +21,17 @@ def last() -> HashMap:
|
|||||||
@section("tracepoint/syscalls/sys_enter_sync")
|
@section("tracepoint/syscalls/sys_enter_sync")
|
||||||
def do_trace(ctx: c_void_p) -> c_int64:
|
def do_trace(ctx: c_void_p) -> c_int64:
|
||||||
key = 0
|
key = 0
|
||||||
tsp = last().lookup(key)
|
tsp = last.lookup(key)
|
||||||
if tsp:
|
if tsp:
|
||||||
kt = ktime()
|
kt = ktime()
|
||||||
delta = kt - tsp
|
delta = kt - tsp
|
||||||
if delta < 1000000000:
|
if delta < 1000000000:
|
||||||
time_ms = delta // 1000000
|
time_ms = delta // 1000000
|
||||||
print(f"sync called within last second, last {time_ms} ms ago")
|
print(f"sync called within last second, last {time_ms} ms ago")
|
||||||
last().delete(key)
|
last.delete(key)
|
||||||
else:
|
else:
|
||||||
kt = ktime()
|
kt = ktime()
|
||||||
last().update(key, kt)
|
last.update(key, kt)
|
||||||
return c_int64(0)
|
return c_int64(0)
|
||||||
|
|
||||||
|
|
||||||
|
|||||||
@ -8,68 +8,59 @@ logger: Logger = logging.getLogger(__name__)
|
|||||||
|
|
||||||
def recursive_dereferencer(var, builder):
|
def recursive_dereferencer(var, builder):
|
||||||
"""dereference until primitive type comes out"""
|
"""dereference until primitive type comes out"""
|
||||||
if var.type == ir.PointerType(ir.PointerType(ir.IntType(64))):
|
# TODO: Not worrying about stack overflow for now
|
||||||
|
if isinstance(var.type, ir.PointerType):
|
||||||
a = builder.load(var)
|
a = builder.load(var)
|
||||||
return recursive_dereferencer(a, builder)
|
return recursive_dereferencer(a, builder)
|
||||||
elif var.type == ir.PointerType(ir.IntType(64)):
|
elif isinstance(var.type, ir.IntType):
|
||||||
a = builder.load(var)
|
|
||||||
return recursive_dereferencer(a, builder)
|
|
||||||
elif var.type == ir.IntType(64):
|
|
||||||
return var
|
return var
|
||||||
else:
|
else:
|
||||||
raise TypeError(f"Unsupported type for dereferencing: {var.type}")
|
raise TypeError(f"Unsupported type for dereferencing: {var.type}")
|
||||||
|
|
||||||
|
|
||||||
def handle_binary_op(rval, module, builder, var_name, local_sym_tab, map_sym_tab, func):
|
def get_operand_value(operand, module, builder, local_sym_tab):
|
||||||
logger.info(f"module {module}")
|
"""Extract the value from an operand, handling variables and constants."""
|
||||||
left = rval.left
|
if isinstance(operand, ast.Name):
|
||||||
right = rval.right
|
if operand.id in local_sym_tab:
|
||||||
|
return recursive_dereferencer(local_sym_tab[operand.id].var, builder)
|
||||||
|
raise ValueError(f"Undefined variable: {operand.id}")
|
||||||
|
elif isinstance(operand, ast.Constant):
|
||||||
|
if isinstance(operand.value, int):
|
||||||
|
return ir.Constant(ir.IntType(64), operand.value)
|
||||||
|
raise TypeError(f"Unsupported constant type: {type(operand.value)}")
|
||||||
|
elif isinstance(operand, ast.BinOp):
|
||||||
|
return handle_binary_op_impl(operand, module, builder, local_sym_tab)
|
||||||
|
raise TypeError(f"Unsupported operand type: {type(operand)}")
|
||||||
|
|
||||||
|
|
||||||
|
def handle_binary_op_impl(rval, module, builder, local_sym_tab):
|
||||||
op = rval.op
|
op = rval.op
|
||||||
|
left = get_operand_value(rval.left, module, builder, local_sym_tab)
|
||||||
# Handle left operand
|
right = get_operand_value(rval.right, module, builder, local_sym_tab)
|
||||||
if isinstance(left, ast.Name):
|
|
||||||
if left.id in local_sym_tab:
|
|
||||||
left = recursive_dereferencer(local_sym_tab[left.id].var, builder)
|
|
||||||
else:
|
|
||||||
raise SyntaxError(f"Undefined variable: {left.id}")
|
|
||||||
elif isinstance(left, ast.Constant):
|
|
||||||
left = ir.Constant(ir.IntType(64), left.value)
|
|
||||||
else:
|
|
||||||
raise SyntaxError("Unsupported left operand type")
|
|
||||||
|
|
||||||
if isinstance(right, ast.Name):
|
|
||||||
if right.id in local_sym_tab:
|
|
||||||
right = recursive_dereferencer(local_sym_tab[right.id].var, builder)
|
|
||||||
else:
|
|
||||||
raise SyntaxError(f"Undefined variable: {right.id}")
|
|
||||||
elif isinstance(right, ast.Constant):
|
|
||||||
right = ir.Constant(ir.IntType(64), right.value)
|
|
||||||
else:
|
|
||||||
raise SyntaxError("Unsupported right operand type")
|
|
||||||
|
|
||||||
logger.info(f"left is {left}, right is {right}, op is {op}")
|
logger.info(f"left is {left}, right is {right}, op is {op}")
|
||||||
|
|
||||||
if isinstance(op, ast.Add):
|
# Map AST operation nodes to LLVM IR builder methods
|
||||||
builder.store(builder.add(left, right), local_sym_tab[var_name].var)
|
op_map = {
|
||||||
elif isinstance(op, ast.Sub):
|
ast.Add: builder.add,
|
||||||
builder.store(builder.sub(left, right), local_sym_tab[var_name].var)
|
ast.Sub: builder.sub,
|
||||||
elif isinstance(op, ast.Mult):
|
ast.Mult: builder.mul,
|
||||||
builder.store(builder.mul(left, right), local_sym_tab[var_name].var)
|
ast.Div: builder.sdiv,
|
||||||
elif isinstance(op, ast.Div):
|
ast.Mod: builder.srem,
|
||||||
builder.store(builder.sdiv(left, right), local_sym_tab[var_name].var)
|
ast.LShift: builder.shl,
|
||||||
elif isinstance(op, ast.Mod):
|
ast.RShift: builder.lshr,
|
||||||
builder.store(builder.srem(left, right), local_sym_tab[var_name].var)
|
ast.BitOr: builder.or_,
|
||||||
elif isinstance(op, ast.LShift):
|
ast.BitXor: builder.xor,
|
||||||
builder.store(builder.shl(left, right), local_sym_tab[var_name].var)
|
ast.BitAnd: builder.and_,
|
||||||
elif isinstance(op, ast.RShift):
|
ast.FloorDiv: builder.udiv,
|
||||||
builder.store(builder.lshr(left, right), local_sym_tab[var_name].var)
|
}
|
||||||
elif isinstance(op, ast.BitOr):
|
|
||||||
builder.store(builder.or_(left, right), local_sym_tab[var_name].var)
|
if type(op) in op_map:
|
||||||
elif isinstance(op, ast.BitXor):
|
result = op_map[type(op)](left, right)
|
||||||
builder.store(builder.xor(left, right), local_sym_tab[var_name].var)
|
return result
|
||||||
elif isinstance(op, ast.BitAnd):
|
|
||||||
builder.store(builder.and_(left, right), local_sym_tab[var_name].var)
|
|
||||||
elif isinstance(op, ast.FloorDiv):
|
|
||||||
builder.store(builder.udiv(left, right), local_sym_tab[var_name].var)
|
|
||||||
else:
|
else:
|
||||||
raise SyntaxError("Unsupported binary operation")
|
raise SyntaxError("Unsupported binary operation")
|
||||||
|
|
||||||
|
|
||||||
|
def handle_binary_op(rval, module, builder, var_name, local_sym_tab):
|
||||||
|
result = handle_binary_op_impl(rval, module, builder, local_sym_tab)
|
||||||
|
builder.store(result, local_sym_tab[var_name].var)
|
||||||
|
|||||||
@ -4,8 +4,12 @@ from .license_pass import license_processing
|
|||||||
from .functions_pass import func_proc
|
from .functions_pass import func_proc
|
||||||
from .maps import maps_proc
|
from .maps import maps_proc
|
||||||
from .structs import structs_proc
|
from .structs import structs_proc
|
||||||
from .globals_pass import globals_processing
|
from .globals_pass import (
|
||||||
from .debuginfo import DW_LANG_C11, DwarfBehaviorEnum
|
globals_list_creation,
|
||||||
|
globals_processing,
|
||||||
|
populate_global_symbol_table,
|
||||||
|
)
|
||||||
|
from .debuginfo import DW_LANG_C11, DwarfBehaviorEnum, DebugInfoGenerator
|
||||||
import os
|
import os
|
||||||
import subprocess
|
import subprocess
|
||||||
import inspect
|
import inspect
|
||||||
@ -40,12 +44,15 @@ def processor(source_code, filename, module):
|
|||||||
for func_node in bpf_chunks:
|
for func_node in bpf_chunks:
|
||||||
logger.info(f"Found BPF function/struct: {func_node.name}")
|
logger.info(f"Found BPF function/struct: {func_node.name}")
|
||||||
|
|
||||||
|
populate_global_symbol_table(tree, module)
|
||||||
|
license_processing(tree, module)
|
||||||
|
globals_processing(tree, module)
|
||||||
|
|
||||||
structs_sym_tab = structs_proc(tree, module, bpf_chunks)
|
structs_sym_tab = structs_proc(tree, module, bpf_chunks)
|
||||||
map_sym_tab = maps_proc(tree, module, bpf_chunks)
|
map_sym_tab = maps_proc(tree, module, bpf_chunks)
|
||||||
func_proc(tree, module, bpf_chunks, map_sym_tab, structs_sym_tab)
|
func_proc(tree, module, bpf_chunks, map_sym_tab, structs_sym_tab)
|
||||||
|
|
||||||
license_processing(tree, module)
|
globals_list_creation(tree, module)
|
||||||
globals_processing(tree, module)
|
|
||||||
|
|
||||||
|
|
||||||
def compile_to_ir(filename: str, output: str, loglevel=logging.WARNING):
|
def compile_to_ir(filename: str, output: str, loglevel=logging.WARNING):
|
||||||
@ -60,33 +67,17 @@ def compile_to_ir(filename: str, output: str, loglevel=logging.WARNING):
|
|||||||
module.triple = "bpf"
|
module.triple = "bpf"
|
||||||
|
|
||||||
if not hasattr(module, "_debug_compile_unit"):
|
if not hasattr(module, "_debug_compile_unit"):
|
||||||
module._file_metadata = module.add_debug_info(
|
debug_generator = DebugInfoGenerator(module)
|
||||||
"DIFile",
|
debug_generator.generate_file_metadata(filename, os.path.dirname(filename))
|
||||||
{ # type: ignore
|
debug_generator.generate_debug_cu(
|
||||||
"filename": filename,
|
DW_LANG_C11,
|
||||||
"directory": os.path.dirname(filename),
|
f"PythonBPF {VERSION}",
|
||||||
},
|
True, # TODO: This is probably not true
|
||||||
|
# TODO: add a global field here that keeps track of all the globals. Works without it, but I think it might
|
||||||
|
# be required for kprobes.
|
||||||
|
True,
|
||||||
)
|
)
|
||||||
|
|
||||||
module._debug_compile_unit = module.add_debug_info(
|
|
||||||
"DICompileUnit",
|
|
||||||
{ # type: ignore
|
|
||||||
"language": DW_LANG_C11,
|
|
||||||
"file": module._file_metadata, # type: ignore
|
|
||||||
"producer": f"PythonBPF {VERSION}",
|
|
||||||
"isOptimized": True, # TODO: This is probably not true
|
|
||||||
# TODO: add a global field here that keeps track of all the globals. Works without it, but I think it might
|
|
||||||
# be required for kprobes.
|
|
||||||
"runtimeVersion": 0,
|
|
||||||
"emissionKind": 1,
|
|
||||||
"splitDebugInlining": False,
|
|
||||||
"nameTableKind": 0,
|
|
||||||
},
|
|
||||||
is_distinct=True,
|
|
||||||
)
|
|
||||||
|
|
||||||
module.add_named_metadata("llvm.dbg.cu", module._debug_compile_unit) # type: ignore
|
|
||||||
|
|
||||||
processor(source, filename, module)
|
processor(source, filename, module)
|
||||||
|
|
||||||
wchar_size = module.add_metadata(
|
wchar_size = module.add_metadata(
|
||||||
|
|||||||
@ -12,6 +12,34 @@ class DebugInfoGenerator:
|
|||||||
self.module = module
|
self.module = module
|
||||||
self._type_cache = {} # Cache for common debug types
|
self._type_cache = {} # Cache for common debug types
|
||||||
|
|
||||||
|
def generate_file_metadata(self, filename, dirname):
|
||||||
|
self.module._file_metadata = self.module.add_debug_info(
|
||||||
|
"DIFile",
|
||||||
|
{ # type: ignore
|
||||||
|
"filename": filename,
|
||||||
|
"directory": dirname,
|
||||||
|
},
|
||||||
|
)
|
||||||
|
|
||||||
|
def generate_debug_cu(
|
||||||
|
self, language, producer: str, is_optimized: bool, is_distinct: bool
|
||||||
|
):
|
||||||
|
self.module._debug_compile_unit = self.module.add_debug_info(
|
||||||
|
"DICompileUnit",
|
||||||
|
{ # type: ignore
|
||||||
|
"language": language,
|
||||||
|
"file": self.module._file_metadata, # type: ignore
|
||||||
|
"producer": producer,
|
||||||
|
"isOptimized": is_optimized,
|
||||||
|
"runtimeVersion": 0,
|
||||||
|
"emissionKind": 1,
|
||||||
|
"splitDebugInlining": False,
|
||||||
|
"nameTableKind": 0,
|
||||||
|
},
|
||||||
|
is_distinct=is_distinct,
|
||||||
|
)
|
||||||
|
self.module.add_named_metadata("llvm.dbg.cu", self.module._debug_compile_unit) # type: ignore
|
||||||
|
|
||||||
def get_basic_type(self, name: str, size: int, encoding: int) -> Any:
|
def get_basic_type(self, name: str, size: int, encoding: int) -> Any:
|
||||||
"""Get or create a basic type with caching"""
|
"""Get or create a basic type with caching"""
|
||||||
key = (name, size, encoding)
|
key = (name, size, encoding)
|
||||||
|
|||||||
@ -2,10 +2,92 @@ import ast
|
|||||||
from llvmlite import ir
|
from llvmlite import ir
|
||||||
from logging import Logger
|
from logging import Logger
|
||||||
import logging
|
import logging
|
||||||
|
from typing import Dict
|
||||||
|
|
||||||
logger: Logger = logging.getLogger(__name__)
|
logger: Logger = logging.getLogger(__name__)
|
||||||
|
|
||||||
|
|
||||||
|
def _handle_name_expr(expr: ast.Name, local_sym_tab: Dict, builder: ir.IRBuilder):
|
||||||
|
"""Handle ast.Name expressions."""
|
||||||
|
if expr.id in local_sym_tab:
|
||||||
|
var = local_sym_tab[expr.id].var
|
||||||
|
val = builder.load(var)
|
||||||
|
return val, local_sym_tab[expr.id].ir_type
|
||||||
|
else:
|
||||||
|
logger.info(f"Undefined variable {expr.id}")
|
||||||
|
return None
|
||||||
|
|
||||||
|
|
||||||
|
def _handle_constant_expr(expr: ast.Constant):
|
||||||
|
"""Handle ast.Constant expressions."""
|
||||||
|
if isinstance(expr.value, int):
|
||||||
|
return ir.Constant(ir.IntType(64), expr.value), ir.IntType(64)
|
||||||
|
elif isinstance(expr.value, bool):
|
||||||
|
return ir.Constant(ir.IntType(1), int(expr.value)), ir.IntType(1)
|
||||||
|
else:
|
||||||
|
logger.info("Unsupported constant type")
|
||||||
|
return None
|
||||||
|
|
||||||
|
|
||||||
|
def _handle_attribute_expr(
|
||||||
|
expr: ast.Attribute,
|
||||||
|
local_sym_tab: Dict,
|
||||||
|
structs_sym_tab: Dict,
|
||||||
|
builder: ir.IRBuilder,
|
||||||
|
):
|
||||||
|
"""Handle ast.Attribute expressions for struct field access."""
|
||||||
|
if isinstance(expr.value, ast.Name):
|
||||||
|
var_name = expr.value.id
|
||||||
|
attr_name = expr.attr
|
||||||
|
if var_name in local_sym_tab:
|
||||||
|
var_ptr, var_type, var_metadata = local_sym_tab[var_name]
|
||||||
|
logger.info(f"Loading attribute {attr_name} from variable {var_name}")
|
||||||
|
logger.info(f"Variable type: {var_type}, Variable ptr: {var_ptr}")
|
||||||
|
|
||||||
|
metadata = structs_sym_tab[var_metadata]
|
||||||
|
if attr_name in metadata.fields:
|
||||||
|
gep = metadata.gep(builder, var_ptr, attr_name)
|
||||||
|
val = builder.load(gep)
|
||||||
|
field_type = metadata.field_type(attr_name)
|
||||||
|
return val, field_type
|
||||||
|
return None
|
||||||
|
|
||||||
|
|
||||||
|
def _handle_deref_call(expr: ast.Call, local_sym_tab: Dict, builder: ir.IRBuilder):
|
||||||
|
"""Handle deref function calls."""
|
||||||
|
logger.info(f"Handling deref {ast.dump(expr)}")
|
||||||
|
if len(expr.args) != 1:
|
||||||
|
logger.info("deref takes exactly one argument")
|
||||||
|
return None
|
||||||
|
|
||||||
|
arg = expr.args[0]
|
||||||
|
if (
|
||||||
|
isinstance(arg, ast.Call)
|
||||||
|
and isinstance(arg.func, ast.Name)
|
||||||
|
and arg.func.id == "deref"
|
||||||
|
):
|
||||||
|
logger.info("Multiple deref not supported")
|
||||||
|
return None
|
||||||
|
|
||||||
|
if isinstance(arg, ast.Name):
|
||||||
|
if arg.id in local_sym_tab:
|
||||||
|
arg_ptr = local_sym_tab[arg.id].var
|
||||||
|
else:
|
||||||
|
logger.info(f"Undefined variable {arg.id}")
|
||||||
|
return None
|
||||||
|
else:
|
||||||
|
logger.info("Unsupported argument type for deref")
|
||||||
|
return None
|
||||||
|
|
||||||
|
if arg_ptr is None:
|
||||||
|
logger.info("Failed to evaluate deref argument")
|
||||||
|
return None
|
||||||
|
|
||||||
|
# Load the value from pointer
|
||||||
|
val = builder.load(arg_ptr)
|
||||||
|
return val, local_sym_tab[arg.id].ir_type
|
||||||
|
|
||||||
|
|
||||||
def eval_expr(
|
def eval_expr(
|
||||||
func,
|
func,
|
||||||
module,
|
module,
|
||||||
@ -17,64 +99,28 @@ def eval_expr(
|
|||||||
):
|
):
|
||||||
logger.info(f"Evaluating expression: {ast.dump(expr)}")
|
logger.info(f"Evaluating expression: {ast.dump(expr)}")
|
||||||
if isinstance(expr, ast.Name):
|
if isinstance(expr, ast.Name):
|
||||||
if expr.id in local_sym_tab:
|
return _handle_name_expr(expr, local_sym_tab, builder)
|
||||||
var = local_sym_tab[expr.id].var
|
|
||||||
val = builder.load(var)
|
|
||||||
return val, local_sym_tab[expr.id].ir_type # return value and type
|
|
||||||
else:
|
|
||||||
logger.info(f"Undefined variable {expr.id}")
|
|
||||||
return None
|
|
||||||
elif isinstance(expr, ast.Constant):
|
elif isinstance(expr, ast.Constant):
|
||||||
if isinstance(expr.value, int):
|
return _handle_constant_expr(expr)
|
||||||
return ir.Constant(ir.IntType(64), expr.value), ir.IntType(64)
|
|
||||||
elif isinstance(expr.value, bool):
|
|
||||||
return ir.Constant(ir.IntType(1), int(expr.value)), ir.IntType(1)
|
|
||||||
else:
|
|
||||||
logger.info("Unsupported constant type")
|
|
||||||
return None
|
|
||||||
elif isinstance(expr, ast.Call):
|
elif isinstance(expr, ast.Call):
|
||||||
|
if isinstance(expr.func, ast.Name) and expr.func.id == "deref":
|
||||||
|
return _handle_deref_call(expr, local_sym_tab, builder)
|
||||||
|
|
||||||
# delayed import to avoid circular dependency
|
# delayed import to avoid circular dependency
|
||||||
from pythonbpf.helper import HelperHandlerRegistry, handle_helper_call
|
from pythonbpf.helper import HelperHandlerRegistry, handle_helper_call
|
||||||
|
|
||||||
if isinstance(expr.func, ast.Name):
|
if isinstance(expr.func, ast.Name) and HelperHandlerRegistry.has_handler(
|
||||||
# check deref
|
expr.func.id
|
||||||
if expr.func.id == "deref":
|
):
|
||||||
logger.info(f"Handling deref {ast.dump(expr)}")
|
return handle_helper_call(
|
||||||
if len(expr.args) != 1:
|
expr,
|
||||||
logger.info("deref takes exactly one argument")
|
module,
|
||||||
return None
|
builder,
|
||||||
arg = expr.args[0]
|
func,
|
||||||
if (
|
local_sym_tab,
|
||||||
isinstance(arg, ast.Call)
|
map_sym_tab,
|
||||||
and isinstance(arg.func, ast.Name)
|
structs_sym_tab,
|
||||||
and arg.func.id == "deref"
|
)
|
||||||
):
|
|
||||||
logger.info("Multiple deref not supported")
|
|
||||||
return None
|
|
||||||
if isinstance(arg, ast.Name):
|
|
||||||
if arg.id in local_sym_tab:
|
|
||||||
arg = local_sym_tab[arg.id].var
|
|
||||||
else:
|
|
||||||
logger.info(f"Undefined variable {arg.id}")
|
|
||||||
return None
|
|
||||||
if arg is None:
|
|
||||||
logger.info("Failed to evaluate deref argument")
|
|
||||||
return None
|
|
||||||
# Since we are handling only name case, directly take type from sym tab
|
|
||||||
val = builder.load(arg)
|
|
||||||
return val, local_sym_tab[expr.args[0].id].ir_type
|
|
||||||
|
|
||||||
# check for helpers
|
|
||||||
if HelperHandlerRegistry.has_handler(expr.func.id):
|
|
||||||
return handle_helper_call(
|
|
||||||
expr,
|
|
||||||
module,
|
|
||||||
builder,
|
|
||||||
func,
|
|
||||||
local_sym_tab,
|
|
||||||
map_sym_tab,
|
|
||||||
structs_sym_tab,
|
|
||||||
)
|
|
||||||
elif isinstance(expr.func, ast.Attribute):
|
elif isinstance(expr.func, ast.Attribute):
|
||||||
logger.info(f"Handling method call: {ast.dump(expr.func)}")
|
logger.info(f"Handling method call: {ast.dump(expr.func)}")
|
||||||
if isinstance(expr.func.value, ast.Call) and isinstance(
|
if isinstance(expr.func.value, ast.Call) and isinstance(
|
||||||
@ -106,19 +152,7 @@ def eval_expr(
|
|||||||
structs_sym_tab,
|
structs_sym_tab,
|
||||||
)
|
)
|
||||||
elif isinstance(expr, ast.Attribute):
|
elif isinstance(expr, ast.Attribute):
|
||||||
if isinstance(expr.value, ast.Name):
|
return _handle_attribute_expr(expr, local_sym_tab, structs_sym_tab, builder)
|
||||||
var_name = expr.value.id
|
|
||||||
attr_name = expr.attr
|
|
||||||
if var_name in local_sym_tab:
|
|
||||||
var_ptr, var_type, var_metadata = local_sym_tab[var_name]
|
|
||||||
logger.info(f"Loading attribute {attr_name} from variable {var_name}")
|
|
||||||
logger.info(f"Variable type: {var_type}, Variable ptr: {var_ptr}")
|
|
||||||
metadata = structs_sym_tab[var_metadata]
|
|
||||||
if attr_name in metadata.fields:
|
|
||||||
gep = metadata.gep(builder, var_ptr, attr_name)
|
|
||||||
val = builder.load(gep)
|
|
||||||
field_type = metadata.field_type(attr_name)
|
|
||||||
return val, field_type
|
|
||||||
logger.info("Unsupported expression evaluation")
|
logger.info("Unsupported expression evaluation")
|
||||||
return None
|
return None
|
||||||
|
|
||||||
|
|||||||
@ -192,8 +192,23 @@ def handle_assign(
|
|||||||
elif isinstance(rval.func, ast.Attribute):
|
elif isinstance(rval.func, ast.Attribute):
|
||||||
logger.info(f"Assignment call attribute: {ast.dump(rval.func)}")
|
logger.info(f"Assignment call attribute: {ast.dump(rval.func)}")
|
||||||
if isinstance(rval.func.value, ast.Name):
|
if isinstance(rval.func.value, ast.Name):
|
||||||
# TODO: probably a struct access
|
if rval.func.value.id in map_sym_tab:
|
||||||
logger.info(f"TODO STRUCT ACCESS {ast.dump(rval)}")
|
map_name = rval.func.value.id
|
||||||
|
method_name = rval.func.attr
|
||||||
|
if HelperHandlerRegistry.has_handler(method_name):
|
||||||
|
val = handle_helper_call(
|
||||||
|
rval,
|
||||||
|
module,
|
||||||
|
builder,
|
||||||
|
func,
|
||||||
|
local_sym_tab,
|
||||||
|
map_sym_tab,
|
||||||
|
structs_sym_tab,
|
||||||
|
)
|
||||||
|
builder.store(val[0], local_sym_tab[var_name].var)
|
||||||
|
else:
|
||||||
|
# TODO: probably a struct access
|
||||||
|
logger.info(f"TODO STRUCT ACCESS {ast.dump(rval)}")
|
||||||
elif isinstance(rval.func.value, ast.Call) and isinstance(
|
elif isinstance(rval.func.value, ast.Call) and isinstance(
|
||||||
rval.func.value.func, ast.Name
|
rval.func.value.func, ast.Name
|
||||||
):
|
):
|
||||||
@ -218,9 +233,7 @@ def handle_assign(
|
|||||||
else:
|
else:
|
||||||
logger.info("Unsupported assignment call function type")
|
logger.info("Unsupported assignment call function type")
|
||||||
elif isinstance(rval, ast.BinOp):
|
elif isinstance(rval, ast.BinOp):
|
||||||
handle_binary_op(
|
handle_binary_op(rval, module, builder, var_name, local_sym_tab)
|
||||||
rval, module, builder, var_name, local_sym_tab, map_sym_tab, func
|
|
||||||
)
|
|
||||||
else:
|
else:
|
||||||
logger.info("Unsupported assignment value type")
|
logger.info("Unsupported assignment value type")
|
||||||
|
|
||||||
|
|||||||
@ -1,8 +1,121 @@
|
|||||||
from llvmlite import ir
|
from llvmlite import ir
|
||||||
import ast
|
import ast
|
||||||
|
|
||||||
|
from logging import Logger
|
||||||
|
import logging
|
||||||
|
from .type_deducer import ctypes_to_ir
|
||||||
|
|
||||||
def emit_globals(module: ir.Module, names: list[str]):
|
logger: Logger = logging.getLogger(__name__)
|
||||||
|
|
||||||
|
# TODO: this is going to be a huge fuck of a headache in the future.
|
||||||
|
global_sym_tab = []
|
||||||
|
|
||||||
|
|
||||||
|
def populate_global_symbol_table(tree, module: ir.Module):
|
||||||
|
for node in tree.body:
|
||||||
|
if isinstance(node, ast.FunctionDef):
|
||||||
|
for dec in node.decorator_list:
|
||||||
|
if (
|
||||||
|
isinstance(dec, ast.Call)
|
||||||
|
and isinstance(dec.func, ast.Name)
|
||||||
|
and dec.func.id == "section"
|
||||||
|
and len(dec.args) == 1
|
||||||
|
and isinstance(dec.args[0], ast.Constant)
|
||||||
|
and isinstance(dec.args[0].value, str)
|
||||||
|
):
|
||||||
|
global_sym_tab.append(node)
|
||||||
|
elif isinstance(dec, ast.Name) and dec.id == "bpfglobal":
|
||||||
|
global_sym_tab.append(node)
|
||||||
|
|
||||||
|
elif isinstance(dec, ast.Name) and dec.id == "map":
|
||||||
|
global_sym_tab.append(node)
|
||||||
|
return False
|
||||||
|
|
||||||
|
|
||||||
|
def emit_global(module: ir.Module, node, name):
|
||||||
|
logger.info(f"global identifier {name} processing")
|
||||||
|
# deduce LLVM type from the annotated return
|
||||||
|
if not isinstance(node.returns, ast.Name):
|
||||||
|
raise ValueError(f"Unsupported return annotation {ast.dump(node.returns)}")
|
||||||
|
ty = ctypes_to_ir(node.returns.id)
|
||||||
|
|
||||||
|
# extract the return expression
|
||||||
|
# TODO: turn this return extractor into a generic function I can use everywhere.
|
||||||
|
ret_stmt = node.body[0]
|
||||||
|
if not isinstance(ret_stmt, ast.Return) or ret_stmt.value is None:
|
||||||
|
raise ValueError(f"Global '{name}' has no valid return")
|
||||||
|
|
||||||
|
init_val = ret_stmt.value
|
||||||
|
|
||||||
|
# simple constant like "return 0"
|
||||||
|
if isinstance(init_val, ast.Constant):
|
||||||
|
llvm_init = ir.Constant(ty, init_val.value)
|
||||||
|
|
||||||
|
# variable reference like "return SOME_CONST"
|
||||||
|
elif isinstance(init_val, ast.Name):
|
||||||
|
# need symbol resolution here, stub as 0 for now
|
||||||
|
raise ValueError(f"Name reference {init_val.id} not yet supported")
|
||||||
|
|
||||||
|
# constructor call like "return c_int64(0)" or dataclass(...)
|
||||||
|
elif isinstance(init_val, ast.Call):
|
||||||
|
if len(init_val.args) >= 1 and isinstance(init_val.args[0], ast.Constant):
|
||||||
|
llvm_init = ir.Constant(ty, init_val.args[0].value)
|
||||||
|
else:
|
||||||
|
logger.info("Defaulting to zero as no constant argument found")
|
||||||
|
llvm_init = ir.Constant(ty, 0)
|
||||||
|
else:
|
||||||
|
raise ValueError(f"Unsupported return expr {ast.dump(init_val)}")
|
||||||
|
|
||||||
|
gvar = ir.GlobalVariable(module, ty, name=name)
|
||||||
|
gvar.initializer = llvm_init
|
||||||
|
gvar.align = 8
|
||||||
|
gvar.linkage = "dso_local"
|
||||||
|
gvar.global_constant = False
|
||||||
|
return gvar
|
||||||
|
|
||||||
|
|
||||||
|
def globals_processing(tree, module):
|
||||||
|
"""Process stuff decorated with @bpf and @bpfglobal except license and return the section name"""
|
||||||
|
globals_sym_tab = []
|
||||||
|
|
||||||
|
for node in tree.body:
|
||||||
|
# Skip non-assignment and non-function nodes
|
||||||
|
if not (isinstance(node, ast.FunctionDef)):
|
||||||
|
continue
|
||||||
|
|
||||||
|
# Get the name based on node type
|
||||||
|
if isinstance(node, ast.FunctionDef):
|
||||||
|
name = node.name
|
||||||
|
else:
|
||||||
|
continue
|
||||||
|
|
||||||
|
# Check for duplicate names
|
||||||
|
if name in globals_sym_tab:
|
||||||
|
raise SyntaxError(f"ERROR: Global name '{name}' previously defined")
|
||||||
|
else:
|
||||||
|
globals_sym_tab.append(name)
|
||||||
|
|
||||||
|
if isinstance(node, ast.FunctionDef) and node.name != "LICENSE":
|
||||||
|
decorators = [
|
||||||
|
dec.id for dec in node.decorator_list if isinstance(dec, ast.Name)
|
||||||
|
]
|
||||||
|
if "bpf" in decorators and "bpfglobal" in decorators:
|
||||||
|
if (
|
||||||
|
len(node.body) == 1
|
||||||
|
and isinstance(node.body[0], ast.Return)
|
||||||
|
and node.body[0].value is not None
|
||||||
|
and isinstance(
|
||||||
|
node.body[0].value, (ast.Constant, ast.Name, ast.Call)
|
||||||
|
)
|
||||||
|
):
|
||||||
|
emit_global(module, node, name)
|
||||||
|
else:
|
||||||
|
raise SyntaxError(f"ERROR: Invalid syntax for {name} global")
|
||||||
|
|
||||||
|
return None
|
||||||
|
|
||||||
|
|
||||||
|
def emit_llvm_compiler_used(module: ir.Module, names: list[str]):
|
||||||
"""
|
"""
|
||||||
Emit the @llvm.compiler.used global given a list of function/global names.
|
Emit the @llvm.compiler.used global given a list of function/global names.
|
||||||
"""
|
"""
|
||||||
@ -24,7 +137,7 @@ def emit_globals(module: ir.Module, names: list[str]):
|
|||||||
gv.section = "llvm.metadata"
|
gv.section = "llvm.metadata"
|
||||||
|
|
||||||
|
|
||||||
def globals_processing(tree, module: ir.Module):
|
def globals_list_creation(tree, module: ir.Module):
|
||||||
collected = ["LICENSE"]
|
collected = ["LICENSE"]
|
||||||
|
|
||||||
for node in tree.body:
|
for node in tree.body:
|
||||||
@ -40,10 +153,11 @@ def globals_processing(tree, module: ir.Module):
|
|||||||
):
|
):
|
||||||
collected.append(node.name)
|
collected.append(node.name)
|
||||||
|
|
||||||
elif isinstance(dec, ast.Name) and dec.id == "bpfglobal":
|
# NOTE: all globals other than
|
||||||
collected.append(node.name)
|
# elif isinstance(dec, ast.Name) and dec.id == "bpfglobal":
|
||||||
|
# collected.append(node.name)
|
||||||
|
|
||||||
elif isinstance(dec, ast.Name) and dec.id == "map":
|
elif isinstance(dec, ast.Name) and dec.id == "map":
|
||||||
collected.append(node.name)
|
collected.append(node.name)
|
||||||
|
|
||||||
emit_globals(module, collected)
|
emit_llvm_compiler_used(module, collected)
|
||||||
|
|||||||
27
tests/c-form/globals.bpf.c
Normal file
27
tests/c-form/globals.bpf.c
Normal file
@ -0,0 +1,27 @@
|
|||||||
|
// SPDX-License-Identifier: GPL-2.0 OR BSD-3-Clause
|
||||||
|
#include <linux/bpf.h>
|
||||||
|
#include <bpf/bpf_helpers.h>
|
||||||
|
#include <bpf/bpf_tracing.h>
|
||||||
|
#include <linux/types.h>
|
||||||
|
|
||||||
|
struct test_struct {
|
||||||
|
__u64 a;
|
||||||
|
__u64 b;
|
||||||
|
};
|
||||||
|
|
||||||
|
struct test_struct w = {};
|
||||||
|
volatile __u64 prev_time = 0;
|
||||||
|
|
||||||
|
SEC("tracepoint/syscalls/sys_enter_execve")
|
||||||
|
int trace_execve(void *ctx)
|
||||||
|
{
|
||||||
|
bpf_printk("previous %ul now %ul", w.b, w.a);
|
||||||
|
__u64 ts = bpf_ktime_get_ns();
|
||||||
|
bpf_printk("prev %ul now %ul", prev_time, ts);
|
||||||
|
w.a = ts;
|
||||||
|
w.b = prev_time;
|
||||||
|
prev_time = ts;
|
||||||
|
return 0;
|
||||||
|
}
|
||||||
|
|
||||||
|
char LICENSE[] SEC("license") = "GPL";
|
||||||
101
tests/failing_tests/globals.py
Normal file
101
tests/failing_tests/globals.py
Normal file
@ -0,0 +1,101 @@
|
|||||||
|
import logging
|
||||||
|
|
||||||
|
from pythonbpf import compile, bpf, section, bpfglobal, compile_to_ir
|
||||||
|
from ctypes import c_void_p, c_int64, c_int32
|
||||||
|
|
||||||
|
@bpf
|
||||||
|
@bpfglobal
|
||||||
|
def somevalue() -> c_int32:
|
||||||
|
return c_int32(42)
|
||||||
|
|
||||||
|
@bpf
|
||||||
|
@bpfglobal
|
||||||
|
def somevalue2() -> c_int64:
|
||||||
|
return c_int64(69)
|
||||||
|
|
||||||
|
@bpf
|
||||||
|
@bpfglobal
|
||||||
|
def somevalue1() -> c_int32:
|
||||||
|
return c_int32(42)
|
||||||
|
|
||||||
|
|
||||||
|
# --- Passing examples ---
|
||||||
|
|
||||||
|
# Simple constant return
|
||||||
|
@bpf
|
||||||
|
@bpfglobal
|
||||||
|
def g1() -> c_int64:
|
||||||
|
return c_int64(42)
|
||||||
|
|
||||||
|
# Constructor with one constant argument
|
||||||
|
@bpf
|
||||||
|
@bpfglobal
|
||||||
|
def g2() -> c_int64:
|
||||||
|
return c_int64(69)
|
||||||
|
|
||||||
|
|
||||||
|
# --- Failing examples ---
|
||||||
|
|
||||||
|
# No return annotation
|
||||||
|
# @bpf
|
||||||
|
# @bpfglobal
|
||||||
|
# def g3():
|
||||||
|
# return 42
|
||||||
|
|
||||||
|
# Return annotation is complex
|
||||||
|
# @bpf
|
||||||
|
# @bpfglobal
|
||||||
|
# def g4() -> List[int]:
|
||||||
|
# return []
|
||||||
|
|
||||||
|
# # Return is missing
|
||||||
|
# @bpf
|
||||||
|
# @bpfglobal
|
||||||
|
# def g5() -> c_int64:
|
||||||
|
# pass
|
||||||
|
|
||||||
|
# # Return is a variable reference
|
||||||
|
# #TODO: maybe fix this sometime later. It defaults to 0
|
||||||
|
# CONST = 5
|
||||||
|
# @bpf
|
||||||
|
# @bpfglobal
|
||||||
|
# def g6() -> c_int64:
|
||||||
|
# return c_int64(CONST)
|
||||||
|
|
||||||
|
# Constructor with multiple args
|
||||||
|
#TODO: this is not working. should it work ?
|
||||||
|
@bpf
|
||||||
|
@bpfglobal
|
||||||
|
def g7() -> c_int64:
|
||||||
|
return c_int64(1)
|
||||||
|
|
||||||
|
# Dataclass call
|
||||||
|
#TODO: fails with dataclass
|
||||||
|
# @dataclass
|
||||||
|
# class Point:
|
||||||
|
# x: c_int64
|
||||||
|
# y: c_int64
|
||||||
|
|
||||||
|
# @bpf
|
||||||
|
# @bpfglobal
|
||||||
|
# def g8() -> Point:
|
||||||
|
# return Point(1, 2)
|
||||||
|
|
||||||
|
|
||||||
|
@bpf
|
||||||
|
@section("tracepoint/syscalls/sys_enter_execve")
|
||||||
|
def sometag(ctx: c_void_p) -> c_int64:
|
||||||
|
print("test")
|
||||||
|
global somevalue
|
||||||
|
somevalue = 2
|
||||||
|
print(f"{somevalue}")
|
||||||
|
return c_int64(1)
|
||||||
|
|
||||||
|
@bpf
|
||||||
|
@bpfglobal
|
||||||
|
def LICENSE() -> str:
|
||||||
|
return "GPL"
|
||||||
|
|
||||||
|
|
||||||
|
compile_to_ir("globals.py", "globals.ll", loglevel=logging.INFO)
|
||||||
|
compile()
|
||||||
21
tests/failing_tests/undeclared_values.py
Normal file
21
tests/failing_tests/undeclared_values.py
Normal file
@ -0,0 +1,21 @@
|
|||||||
|
import logging
|
||||||
|
|
||||||
|
from pythonbpf import compile, bpf, section, bpfglobal, compile_to_ir
|
||||||
|
from ctypes import c_void_p, c_int64
|
||||||
|
|
||||||
|
# This should not pass as somevalue is not declared at all.
|
||||||
|
@bpf
|
||||||
|
@section("tracepoint/syscalls/sys_enter_execve")
|
||||||
|
def sometag(ctx: c_void_p) -> c_int64:
|
||||||
|
print("test")
|
||||||
|
print(f"{somevalue}") # noqa: F821
|
||||||
|
return c_int64(1)
|
||||||
|
|
||||||
|
@bpf
|
||||||
|
@bpfglobal
|
||||||
|
def LICENSE() -> str:
|
||||||
|
return "GPL"
|
||||||
|
|
||||||
|
|
||||||
|
compile_to_ir("globals.py", "globals.ll", loglevel=logging.INFO)
|
||||||
|
compile()
|
||||||
@ -3,9 +3,9 @@ from ctypes import c_void_p, c_int64
|
|||||||
|
|
||||||
|
|
||||||
@bpf
|
@bpf
|
||||||
@section("sometag1")
|
@section("tracepoint/syscalls/sys_enter_sync")
|
||||||
def sometag(ctx: c_void_p) -> c_int64:
|
def sometag(ctx: c_void_p) -> c_int64:
|
||||||
a = 1 + 2 + 1
|
a = 1 + 2 + 1 + 12 + 13
|
||||||
print(f"{a}")
|
print(f"{a}")
|
||||||
return c_int64(0)
|
return c_int64(0)
|
||||||
|
|
||||||
@ -3,11 +3,12 @@ from ctypes import c_void_p, c_int64
|
|||||||
|
|
||||||
|
|
||||||
@bpf
|
@bpf
|
||||||
@section("sometag1")
|
@section("tracepoint/syscalls/sys_enter_sync")
|
||||||
def sometag(ctx: c_void_p) -> c_int64:
|
def sometag(ctx: c_void_p) -> c_int64:
|
||||||
b = 1 + 2
|
b = 1 + 2
|
||||||
a = 1 + b
|
a = 1 + b
|
||||||
return c_int64(a)
|
print(f"{a}")
|
||||||
|
return c_int64(0)
|
||||||
|
|
||||||
|
|
||||||
@bpf
|
@bpf
|
||||||
Reference in New Issue
Block a user