Linux 驱动扫描所有线程调用栈

测试环境

root@:curtis# uname -a
Linux curtis-Aspire-E5-471G 5.15.0-52-generic #58~20.04.1-Ubuntu SMP Thu Oct 13 13:09:46 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux
root@:curtis# lsb_release -a
No LSB modules are available.
Distributor ID: Ubuntu
Description:    Ubuntu 20.04.3 LTS
Release:        20.04
Codename:       focal

Linux 通过proc文件系统，将进程的栈信息透给用户态，调用链如下所示。

root@:curtis# cat /proc/self/stack
[<0>] proc_pid_stack+0x9a/0xf0
[<0>] proc_single_show+0x52/0xc0
[<0>] seq_read_iter+0x124/0x450
[<0>] seq_read+0xfd/0x150
[<0>] vfs_read+0xa0/0x1a0
[<0>] ksys_read+0x67/0xf0
[<0>] __x64_sys_read+0x1a/0x20
[<0>] do_syscall_64+0x5c/0xc0
[<0>] entry_SYSCALL_64_after_hwframe+0x61/0xcb

从调用栈上来看，最终调用的是函数proc_pid_stack。

#ifdef CONFIG_STACKTRACE

#define MAX_STACK_TRACE_DEPTH	64

static int proc_pid_stack(struct seq_file *m, struct pid_namespace *ns,
			  struct pid *pid, struct task_struct *task)
{
	unsigned long *entries;
	int err;

	/*
	 * The ability to racily run the kernel stack unwinder on a running task
	 * and then observe the unwinder output is scary; while it is useful for
	 * debugging kernel issues, it can also allow an attacker to leak kernel
	 * stack contents.
	 * Doing this in a manner that is at least safe from races would require
	 * some work to ensure that the remote task can not be scheduled; and
	 * even then, this would still expose the unwinder as local attack
	 * surface.
	 * Therefore, this interface is restricted to root.
	 */
	if (!file_ns_capable(m->file, &init_user_ns, CAP_SYS_ADMIN))
		return -EACCES;

	entries = kmalloc_array(MAX_STACK_TRACE_DEPTH, sizeof(*entries),
				GFP_KERNEL);
	if (!entries)
		return -ENOMEM;

	err = lock_trace(task);
	if (!err) {
		unsigned int i, nr_entries;

		nr_entries = stack_trace_save_tsk(task, entries,
						  MAX_STACK_TRACE_DEPTH, 0);

		for (i = 0; i < nr_entries; i++) {
			seq_printf(m, "[<0>] %pB\n", (void *)entries[i]);
		}

		unlock_trace(task);
	}
	kfree(entries);

	return err;
}
#endif

从函数的定义来看需要将内核调试选项CONFIG_STACKTRACE打开，核心程序调用的是stack_trace_save_tsk函数，为非导出函数，如何使用未导出函数之前的文章有介绍过。

/**
 * stack_trace_save_tsk - Save a task stack trace into a storage array
 * @task:	The task to examine
 * @store:	Pointer to storage array
 * @size:	Size of the storage array
 * @skipnr:	Number of entries to skip at the start of the stack trace
 *
 * Return: Number of trace entries stored
 */
unsigned int stack_trace_save_tsk(struct task_struct *task,
				  unsigned long *store, unsigned int size,
				  unsigned int skipnr)
{
	struct stack_trace trace = {
		.entries	= store,
		.max_entries	= size,
		/* skip this function if they are tracing us */
		.skip	= skipnr + (current == task),
	};

	save_stack_trace_tsk(task, &trace);
	return trace.nr_entries;
}

主要代码逻辑

#include <linux/kernel.h>
#include <linux/module.h>
#include <linux/slab.h>
#include <linux/sched/signal.h>

#include "trace.h"

#define MAX_STACK_TRACE_DEPTH 64

unsigned int (*stack_trace_save_tsk_ptr)(struct task_struct *task,
		unsigned long *store, unsigned int size,
		unsigned int skipnr);

int print_stack(struct task_struct *task)
{
	unsigned long *entries;
	unsigned int i, nr_entries;

	entries = kmalloc_array(MAX_STACK_TRACE_DEPTH, sizeof(*entries), GFP_KERNEL);
	if (!entries)
		return -ENOMEM;

	nr_entries = stack_trace_save_tsk_ptr(task, entries,
			MAX_STACK_TRACE_DEPTH, 0);

	printk("PID = %d, COMM = %s\n", task->pid, task->comm);
	for (i = 0; i < nr_entries; i++) {
		printk(" [<0>] %pB\n", (void *)entries[i]);
	}

	kfree(entries);
	return 0;
}

int query_stack(void)
{
	int ret = 0;
	struct task_struct *g, *t;

	do_each_thread(g, t) {
		print_stack(t);	
	} while_each_thread(g, t);

	return ret;
}

static int __init stack_trace_init(void)
{
	int ret = 0;

	ret = init_kallsyms_lookup_func();
	if (ret < 0) {
		printk("get kallsyms_lookup_name addr failed\n");
		return -1;
	}

	stack_trace_save_tsk_ptr = find_func("stack_trace_save_tsk");
	if (stack_trace_save_tsk_ptr == NULL) {
		printk("get stack_trace_save_tsk addr failed\n");
		return -1;
	}

	ret = query_stack();
	if (ret < 0) {
		printk("query stack failed\n");
		return ret;
	}
	printk("stack trace init\n");
	return 0;
}

static void __exit stack_trace_exit(void)
{
	printk("stack trace exit\n");	
}

module_init(stack_trace_init);
module_exit(stack_trace_exit);
MODULE_LICENSE("GPL");

调用栈打印示例。

[781162.407668] PID = 107085, COMM = sudo
[781162.407670]  [<0>] do_sys_poll+0x486/0x610
[781162.407675]  [<0>] __x64_sys_ppoll+0xac/0xe0
[781162.407679]  [<0>] do_syscall_64+0x5c/0xc0
[781162.407684]  [<0>] entry_SYSCALL_64_after_hwframe+0x61/0xcb
[781162.407696] PID = 107086, COMM = insmod
[781162.407698]  [<0>] print_stack+0x58/0x90 [trace]
[781162.407705]  [<0>] query_stack+0x2d/0x70 [trace]
[781162.407712]  [<0>] stack_trace_init+0x55/0x1000 [trace]
[781162.407719]  [<0>] do_one_initcall+0x48/0x1e0
[781162.407726]  [<0>] do_init_module+0x52/0x230
[781162.407733]  [<0>] load_module+0x138d/0x1610
[781162.407739]  [<0>] __do_sys_finit_module+0xbf/0x120
[781162.407746]  [<0>] __x64_sys_finit_module+0x1a/0x20
[781162.407752]  [<0>] do_syscall_64+0x5c/0xc0
[781162.407757]  [<0>] entry_SYSCALL_64_after_hwframe+0x61/0xcb