BUG: scheduling while atomic 分析

遇到一个典型的ｓｃｈｅｄｕｌｅ问题。

<3>[26578.636839] C1 [      swapper/1] BUG: scheduling while atomic: swapper/1/0/0x00000002
<6>[26578.636869] C0 [    kworker/u:1] CPU1 is up
<4>[26578.636900] C1 [      swapper/1] Modules linked in: bcm15500_i2c_ts
<4>[26578.636961] C1 [      swapper/1] [<c00146d0>] (unwind_backtrace+0x0/0x11c) from [<c0602684>] (__schedule+0x70/0x6e0)
<4>[26578.636991] C1 [      swapper/1] [<c0602684>] (__schedule+0x70/0x6e0) from [<c06030ec>] (schedule_preempt_disabled+0x14/0x20)
<4>[26578.637052] C1 [      swapper/1] [<c06030ec>] (schedule_preempt_disabled+0x14/0x20) from [<c000f05c>] (cpu_idle+0xf0/0x104)
<4>[26578.637083] C1 [      swapper/1] [<c000f05c>] (cpu_idle+0xf0/0x104) from [<c05e98e0>] (cpu_die+0x2c/0x5c)
<3>[26578.637510] C1 [      swapper/1] BUG: scheduling while atomic: swapper/1/0/0x00000002
<4>[26578.637510] C1 [      swapper/1] Modules linked in: bcm15500_i2c_ts
<4>[26578.637602] C1 [      swapper/1] [<c00146d0>] (unwind_backtrace+0x0/0x11c) from [<c0602684>] (__schedule+0x70/0x6e0)
<4>[26578.637663] C1 [      swapper/1] [<c0602684>] (__schedule+0x70/0x6e0) from [<c06030ec>] (schedule_preempt_disabled+0x14/0x20)
<4>[26578.637724] C1 [      swapper/1] [<c06030ec>] (schedule_preempt_disabled+0x14/0x20) from [<c000f05c>] (cpu_idle+0xf0/0x104)
<4>[26578.637754] C1 [      swapper/1] [<c000f05c>] (cpu_idle+0xf0/0x104) from [<c05e98e0>] (cpu_die+0x2c/0x5c)
<3>[26578.648069] C1 [      swapper/1] BUG: scheduling while atomic: swapper/1/0/0x00000002

查看源代码

/*
 * __schedule() is the main scheduler function.
 */
static void __sched __schedule(void)
{
	struct task_struct *prev, *next;
	unsigned long *switch_count;
	struct rq *rq;
	int cpu;

need_resched:
	preempt_disable();
	cpu = smp_processor_id();
	rq = cpu_rq(cpu);
	rcu_note_context_switch(cpu);
	prev = rq->curr;

	schedule_debug(prev);
　　　　....
｝

/*
 * Print scheduling while atomic bug:
 */
static noinline void __schedule_bug(struct task_struct *prev)
{
	if (oops_in_progress)
		return;

	printk(KERN_ERR "BUG: scheduling while atomic: %s/%d/0x%08x\n",
		prev->comm, prev->pid, preempt_count());

	debug_show_held_locks(prev);
	print_modules();
	if (irqs_disabled())
		print_irqtrace_events(prev);

	dump_stack();
}

/*
 * Various schedule()-time debugging checks and statistics:
 */
static inline void schedule_debug(struct task_struct *prev)
{
	/*
	 * Test if we are atomic. Since do_exit() needs to call into
	 * schedule() atomically, we ignore that path for now.
	 * Otherwise, whine if we are scheduling when we should not be.
	 */
	if (unlikely(in_atomic_preempt_off() && !prev->exit_state))
		__schedule_bug(prev);
	rcu_sleep_check();

	profile_hit(SCHED_PROFILING, __builtin_return_address(0));

	schedstat_inc(this_rq(), sched_count);
}

可以看出，　满足如下条件将会打印该出错信息

unlikely(in_atomic_preempt_off() && !prev->exit_state

为0表示TASK_RUNNING状态,当前进程在运行;　并且处于原子状态,,那么就不能切换给其它的进程

Linux/include/linux/sched.h 

/*
 * Task state bitmask. NOTE! These bits are also
 * encoded in fs/proc/array.c: get_task_state().
 *
 * We have two separate sets of flags: task->state
 * is about runnability, while task->exit_state are
 * about the task exiting. Confusing, but this way
 * modifying one set can't modify the other one by
 * mistake.
 */
#define TASK_RUNNING		0
#define TASK_INTERRUPTIBLE	1
#define TASK_UNINTERRUPTIBLE	2
#define __TASK_STOPPED		4
#define __TASK_TRACED		8
/* in tsk->exit_state */
#define EXIT_ZOMBIE		16
#define EXIT_DEAD		32
/* in tsk->state again */
#define TASK_DEAD		64
#define TASK_WAKEKILL		128
#define TASK_WAKING		256
#define TASK_STATE_MAX		512

kernel/include/linux/hardirq.h

#if defined(CONFIG_PREEMPT_COUNT)
# define PREEMPT_CHECK_OFFSET 1
#else
# define PREEMPT_CHECK_OFFSET 0
#endif

/*
 * Are we running in atomic context?  WARNING: this macro cannot
 * always detect atomic context; in particular, it cannot know about
 * held spinlocks in non-preemptible kernels.  Thus it should not be
 * used in the general case to determine whether sleeping is possible.
 * Do not use in_atomic() in driver code.
 */
#define in_atomic()	((preempt_count() & ~PREEMPT_ACTIVE) != 0)

/*
 * Check whether we were atomic before we did preempt_disable():
 * (used by the scheduler, *after* releasing the kernel lock)
 */
#define in_atomic_preempt_off() \
		((preempt_count() & ~PREEMPT_ACTIVE) != PREEMPT_CHECK_OFFSET)

结论整理

linux内核打印"BUG: scheduling while atomic"和"bad: scheduling from the idle thread"错误的时候，

通常是在中断处理函数中调用了可以休眠的函数，如semaphore,mutex,sleep之类的可休眠的函数，

而linux内核要求在中断处理的时候，不允许系统调度，不允许抢占，要等到中断处理完成才能做其他事情。

因此，要充分考虑中断处理的时间，一定不能太久。

另外一个能产生此问题的是在idle进程里面，做了不该做的事情。现在Linux用于很多手持式设备，为了降低功耗，

通常的作法是在idle进程里面降低CPU或RAM的频率、关闭一些设备等等。要保证这些动作的原子性才能确保

不发生"bad: scheduling from the idle thread"这样的错误！

禁止内核抢占是指内核不会主动的抢占你的process，但是现在是你在自己的程序中主动call schedule()，

kernel并不能阻止你这么作。

Scheduling while atomic" means that a thread has called schedule() during an operation which is supposed to be atomic (ie uninterrupted).

190 NOTE: ***** WARNING *****
191 NEVER SLEEP IN A COMPLETION HANDLER. These are normally called
192 during hardware interrupt processing. If you can, defer substantial
193 work to a tasklet (bottom half) to keep system latencies low. You'll
194 probably need to use spinlocks to protect data structures you manipulate
195 in completion handlers. 

GFP_ATOMIC is used when 
(a) you are inside a completion handler, an interrupt, bottom half, tasklet or timer, or 
(b) you are holding a spinlock or rwlock (does not apply to semaphores), or 
(c) current->state != TASK_RUNNING, this is the case only after you've changed it.