zephyr/scripts/gen_kobject_list.py

415 lines
13 KiB
Python
Raw Normal View History

kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
#!/usr/bin/env python3
#
# Copyright (c) 2017 Intel Corporation
#
# SPDX-License-Identifier: Apache-2.0
"""
Script to generate gperf tables of kernel object metadata
User mode threads making system calls reference kernel objects by memory
address, as the kernel/driver APIs in Zephyr are the same for both user
and supervisor contexts. It is necessary for the kernel to be able to
validate accesses to kernel objects to make the following assertions:
- That the memory address points to a kernel object
- The kernel object is of the expected type for the API being invoked
- The kernel object is of the expected initialization state
- The calling thread has sufficient permissions on the object
For more details see the :ref:`kernelobjects` section in the documentation.
The zephyr build generates an intermediate ELF binary, zephyr_prebuilt.elf,
which this script scans looking for kernel objects by examining the DWARF
debug information to look for instances of data structures that are considered
kernel objects. For device drivers, the API struct pointer populated at build
time is also examined to disambiguate between various device driver instances
since they are all 'struct device'.
This script can generate five different output files:
- A gperf script to generate the hash table mapping kernel object memory
addresses to kernel object metadata, used to track permissions,
object type, initialization state, and any object-specific data.
- A header file containing generated macros for validating driver instances
inside the system call handlers for the driver subsystem APIs.
- A code fragment included by kernel.h with one enum constant for
each kernel object type and each driver instance.
- The inner cases of a switch/case C statement, included by
kernel/userspace.c, mapping the kernel object types and driver
instances to their human-readable representation in the
otype_to_str() function.
- The inner cases of a switch/case C statement, included by
kernel/userspace.c, mapping kernel object types to their sizes.
This is used for allocating instances of them at runtime
(CONFIG_DYNAMIC_OBJECTS) in the obj_size_get() function.
"""
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
import sys
import argparse
import math
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
import os
import struct
import json
from elf_helper import ElfHelper, kobject_to_enum
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
from collections import OrderedDict
# Keys in this dictionary are structs which should be recognized as kernel
# objects. Values are a tuple:
#
# - The first item is None, or the name of a Kconfig that
# indicates the presence of this object's definition in case it is not
# available in all configurations.
#
# - The second item is a boolean indicating whether it is permissible for
# the object to be located in user-accessible memory.
# Regular dictionaries are ordered only with Python 3.6 and
# above. Good summary and pointers to official documents at:
# https://stackoverflow.com/questions/39980323/are-dictionaries-ordered-in-python-3-6
kobjects = OrderedDict([
("k_mem_slab", (None, False)),
("k_msgq", (None, False)),
("k_mutex", (None, False)),
("k_pipe", (None, False)),
("k_queue", (None, False)),
("k_poll_signal", (None, False)),
("k_sem", (None, False)),
("k_stack", (None, False)),
("k_thread", (None, False)),
("k_timer", (None, False)),
("z_thread_stack_element", (None, False)),
("device", (None, False)),
("sys_mutex", (None, True)),
("k_futex", (None, True))
])
subsystems = [
# Editing the list is deprecated, add the __subsystem sentinal to your driver
# api declaration instead. e.x.
#
# __subsystem struct my_driver_api {
# ....
#};
]
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
header = """%compare-lengths
%define lookup-function-name z_object_lookup
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
%language=ANSI-C
%global-table
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
%struct-type
%{
#include <kernel.h>
#include <toolchain.h>
#include <syscall_handler.h>
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
#include <string.h>
%}
struct z_object;
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
"""
# Different versions of gperf have different prototypes for the lookup
# function, best to implement the wrapper here. The pointer value itself is
# turned into a string, we told gperf to expect binary strings that are not
# NULL-terminated.
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
footer = """%%
struct z_object *z_object_gperf_find(void *obj)
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
{
return z_object_lookup((const char *)obj, sizeof(void *));
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
}
void z_object_gperf_wordlist_foreach(_wordlist_cb_func_t func, void *context)
{
int i;
for (i = MIN_HASH_VALUE; i <= MAX_HASH_VALUE; i++) {
if (wordlist[i].name != NULL) {
func(&wordlist[i], context);
}
}
}
#ifndef CONFIG_DYNAMIC_OBJECTS
struct z_object *z_object_find(void *obj)
ALIAS_OF(z_object_gperf_find);
void z_object_wordlist_foreach(_wordlist_cb_func_t func, void *context)
ALIAS_OF(z_object_gperf_wordlist_foreach);
#endif
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
"""
metadata_names = {
"K_OBJ_THREAD" : "thread_id",
"K_OBJ_THREAD_STACK_ELEMENT" : "stack_size",
"K_OBJ_SYS_MUTEX" : "mutex",
"K_OBJ_FUTEX" : "futex_data"
}
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
def write_gperf_table(fp, eh, objs, static_begin, static_end):
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
fp.write(header)
num_mutexes = eh.get_sys_mutex_counter()
if num_mutexes != 0:
fp.write("static struct k_mutex kernel_mutexes[%d] = {\n" % num_mutexes)
for i in range(num_mutexes):
fp.write("_K_MUTEX_INITIALIZER(kernel_mutexes[%d])" % i)
if i != num_mutexes - 1:
fp.write(", ")
fp.write("};\n")
num_futex = eh.get_futex_counter()
if num_futex != 0:
fp.write("static struct z_futex_data futex_data[%d] = {\n" % num_futex)
for i in range(num_futex):
fp.write("Z_FUTEX_DATA_INITIALIZER(futex_data[%d])" % i)
if i != num_futex - 1:
fp.write(", ")
fp.write("};\n")
fp.write("%%\n")
# Setup variables for mapping thread indexes
syms = eh.get_symbols()
thread_max_bytes = syms["CONFIG_MAX_THREAD_BYTES"]
thread_idx_map = {}
for i in range(0, thread_max_bytes):
thread_idx_map[i] = 0xFF
for obj_addr, ko in objs.items():
obj_type = ko.type_name
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
# pre-initialized objects fall within this memory range, they are
# either completely initialized at build time, or done automatically
# at boot during some PRE_KERNEL_* phase
initialized = static_begin <= obj_addr < static_end
is_driver = obj_type.startswith("K_OBJ_DRIVER_")
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
if "CONFIG_64BIT" in syms:
format_code = "Q"
else:
format_code = "I"
if eh.little_endian:
endian = "<"
else:
endian = ">"
byte_str = struct.pack(endian + format_code, obj_addr)
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
fp.write("\"")
for byte in byte_str:
val = "\\x%02x" % byte
fp.write(val)
flags = "0"
if initialized:
flags += " | K_OBJ_FLAG_INITIALIZED"
if is_driver:
flags += " | K_OBJ_FLAG_DRIVER"
if ko.type_name in metadata_names:
tname = metadata_names[ko.type_name]
else:
tname = "unused"
fp.write("\", {}, %s, %s, { .%s = %s }\n" % (obj_type, flags,
tname, str(ko.data)))
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
if obj_type == "K_OBJ_THREAD":
idx = math.floor(ko.data / 8)
bit = ko.data % 8
thread_idx_map[idx] = thread_idx_map[idx] & ~(2**bit)
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
fp.write(footer)
# Generate the array of already mapped thread indexes
fp.write('\n')
fp.write('Z_GENERIC_SECTION(.kobject_data.data) ')
fp.write('u8_t _thread_idx_map[%d] = {' % (thread_max_bytes))
for i in range(0, thread_max_bytes):
fp.write(' 0x%x, ' % (thread_idx_map[i]))
fp.write('};\n')
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
driver_macro_tpl = """
#define Z_SYSCALL_DRIVER_%(driver_upper)s(ptr, op) Z_SYSCALL_DRIVER_GEN(ptr, op, %(driver_lower)s, %(driver_upper)s)
"""
def write_validation_output(fp):
fp.write("#ifndef DRIVER_VALIDATION_GEN_H\n")
fp.write("#define DRIVER_VALIDATION_GEN_H\n")
fp.write("""#define Z_SYSCALL_DRIVER_GEN(ptr, op, driver_lower_case, driver_upper_case) \\
(Z_SYSCALL_OBJ(ptr, K_OBJ_DRIVER_##driver_upper_case) || \\
Z_SYSCALL_DRIVER_OP(ptr, driver_lower_case##_driver_api, op))
""")
for subsystem in subsystems:
subsystem = subsystem.replace("_driver_api", "")
fp.write(driver_macro_tpl % {
"driver_lower": subsystem.lower(),
"driver_upper": subsystem.upper(),
})
fp.write("#endif /* DRIVER_VALIDATION_GEN_H */\n")
def write_kobj_types_output(fp):
fp.write("/* Core kernel objects */\n")
for kobj, obj_info in kobjects.items():
dep, _ = obj_info
if kobj == "device":
continue
if dep:
fp.write("#ifdef %s\n" % dep)
fp.write("%s,\n" % kobject_to_enum(kobj))
if dep:
fp.write("#endif\n")
fp.write("/* Driver subsystems */\n")
for subsystem in subsystems:
subsystem = subsystem.replace("_driver_api", "").upper()
fp.write("K_OBJ_DRIVER_%s,\n" % subsystem)
def write_kobj_otype_output(fp):
fp.write("/* Core kernel objects */\n")
for kobj, obj_info in kobjects.items():
dep, _ = obj_info
if kobj == "device":
continue
if dep:
fp.write("#ifdef %s\n" % dep)
fp.write('case %s: ret = "%s"; break;\n' %
(kobject_to_enum(kobj), kobj))
if dep:
fp.write("#endif\n")
fp.write("/* Driver subsystems */\n")
for subsystem in subsystems:
subsystem = subsystem.replace("_driver_api", "")
fp.write('case K_OBJ_DRIVER_%s: ret = "%s driver"; break;\n' % (
subsystem.upper(),
subsystem
))
def write_kobj_size_output(fp):
fp.write("/* Non device/stack objects */\n")
for kobj, obj_info in kobjects.items():
dep, _ = obj_info
# device handled by default case. Stacks are not currently handled,
# if they eventually are it will be a special case.
if kobj in {"device", "z_thread_stack_element"}:
continue
if dep:
fp.write("#ifdef %s\n" % dep)
fp.write('case %s: ret = sizeof(struct %s); break;\n' %
(kobject_to_enum(kobj), kobj))
if dep:
fp.write("#endif\n")
def parse_subsystems_list_file(path):
with open(path, "r") as fp:
subsys_list = json.load(fp)
subsystems.extend(subsys_list)
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
def parse_args():
global args
parser = argparse.ArgumentParser(
description=__doc__,
formatter_class=argparse.RawDescriptionHelpFormatter)
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
parser.add_argument("-k", "--kernel", required=False,
help="Input zephyr ELF binary")
parser.add_argument(
"-g", "--gperf-output", required=False,
help="Output list of kernel object addresses for gperf use")
parser.add_argument(
"-V", "--validation-output", required=False,
help="Output driver validation macros")
parser.add_argument(
"-K", "--kobj-types-output", required=False,
help="Output k_object enum constants")
parser.add_argument(
"-S", "--kobj-otype-output", required=False,
help="Output case statements for otype_to_str()")
parser.add_argument(
"-Z", "--kobj-size-output", required=False,
help="Output case statements for obj_size_get()")
parser.add_argument("-i", "--include-subsystem-list", required=False, action='append',
help='''Specifies a file with a JSON encoded list of subsystem names to append to
the driver subsystems list. Can be specified multiple times:
-i file1 -i file2 ...''')
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
parser.add_argument("-v", "--verbose", action="store_true",
help="Print extra debugging information")
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
args = parser.parse_args()
if "VERBOSE" in os.environ:
args.verbose = 1
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
def main():
parse_args()
if args.include_subsystem_list is not None:
for list_file in args.include_subsystem_list:
parse_subsystems_list_file(list_file)
if args.gperf_output:
assert args.kernel, "--kernel ELF required for --gperf-output"
eh = ElfHelper(args.kernel, args.verbose, kobjects, subsystems)
syms = eh.get_symbols()
max_threads = syms["CONFIG_MAX_THREAD_BYTES"] * 8
objs = eh.find_kobjects(syms)
if not objs:
sys.stderr.write("WARNING: zero kobject found in %s\n"
% args.kernel)
thread_counter = eh.get_thread_counter()
if thread_counter > max_threads:
sys.exit("Too many thread objects ({})\n"
"Increase CONFIG_MAX_THREAD_BYTES to {}"
.format(thread_counter, -(-thread_counter // 8)))
with open(args.gperf_output, "w") as fp:
write_gperf_table(fp, eh, objs,
syms["_static_kernel_objects_begin"],
syms["_static_kernel_objects_end"])
if args.validation_output:
with open(args.validation_output, "w") as fp:
write_validation_output(fp)
if args.kobj_types_output:
with open(args.kobj_types_output, "w") as fp:
write_kobj_types_output(fp)
if args.kobj_otype_output:
with open(args.kobj_otype_output, "w") as fp:
write_kobj_otype_output(fp)
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
if args.kobj_size_output:
with open(args.kobj_size_output, "w") as fp:
write_kobj_size_output(fp)
kernel: introduce object validation mechanism All system calls made from userspace which involve pointers to kernel objects (including device drivers) will need to have those pointers validated; userspace should never be able to crash the kernel by passing it garbage. The actual validation with _k_object_validate() will be in the system call receiver code, which doesn't exist yet. - CONFIG_USERSPACE introduced. We are somewhat far away from having an end-to-end implementation, but at least need a Kconfig symbol to guard the incoming code with. Formal documentation doesn't exist yet either, but will appear later down the road once the implementation is mostly finalized. - In the memory region for RAM, the data section has been moved last, past bss and noinit. This ensures that inserting generated tables with addresses of kernel objects does not change the addresses of those objects (which would make the table invalid) - The DWARF debug information in the generated ELF binary is parsed to fetch the locations of all kernel objects and pass this to gperf to create a perfect hash table of their memory addresses. - The generated gperf code doesn't know that we are exclusively working with memory addresses and uses memory inefficently. A post-processing script process_gperf.py adjusts the generated code before it is compiled to work with pointer values directly and not strings containing them. - _k_object_init() calls inserted into the init functions for the set of kernel object types we are going to support so far Issue: ZEP-2187 Signed-off-by: Andrew Boie <andrew.p.boie@intel.com>
2017-08-22 22:15:23 +02:00
if __name__ == "__main__":
main()