Linux的进程管理与proc文件系统—9

1.背景介绍

proc文件系统提供了对内核参数信息，在此只介绍进程相关的参数；
proc是内核提供的一种文件系统，内核代码位置：linux3.0.20/fs/proc。
shell上的位置：
文件：/proc/sched_debug
/proc/schedstat
目录：/proc/$pid/task/$pid/

线程相关的字段： sched comm statm status schedstat stat

2.sched_debug字段

root@ubuntu:/proc# cat sched_debug
//调度的基准时间
Sched Debug Version: v0.11, 4.15.0-142-generic #146~16.04.1-Ubuntu
ktime                                   : 13514998.586831
sched_clk                               : 13514405.524746
cpu_clk                                 : 13514187.313553
jiffies                                 : 4298271045
sched_clock_stable()                    : 1

sysctl_sched
  .sysctl_sched_latency                    : 6.000000
  .sysctl_sched_min_granularity            : 0.750000
  .sysctl_sched_wakeup_granularity         : 1.000000
  .sysctl_sched_child_runs_first           : 0
  .sysctl_sched_features                   : 2021179
  .sysctl_sched_tunable_scaling            : 1 (logaritmic)

//cpu的始终参数，本机为单核
cpu#0, 2294.780 MHz
  .nr_running                    : 3
  .load                          : 3145728
  .nr_switches                   : 682047
  .nr_load_updates               : 540552
  .nr_uninterruptible            : 0
  .next_balance                  : 4294.892296
  .curr->pid                     : 4322
  .clock                         : 13514186.330097
  .clock_task                    : 13514186.330097
  .cpu_load[0]                   : 3070
  .cpu_load[1]                   : 3070
  .cpu_load[2]                   : 3070
  .cpu_load[3]                   : 3059
  .cpu_load[4]                   : 2899
  .avg_idle                      : 1000000
  .max_idle_balance_cost         : 500000
  //cfs调度器的参数
cfs_rq[0]:/
  .exec_clock                    : 0.000000
  .MIN_vruntime                  : 144700.671361
  .min_vruntime                  : 144703.661531
  .max_vruntime                  : 144703.661531
  .spread                        : 2.990170
  .spread0                       : 0.000000
  .nr_spread_over                : 0
  .nr_running                    : 3
  .load                          : 3145728
  .runnable_weight               : 3145728
  .load_avg                      : 5075
  .runnable_load_avg             : 2005
  .util_avg                      : 516
  .removed.load_avg              : 0
  .removed.util_avg              : 0
  .removed.runnable_sum          : 0
  .tg_load_avg_contrib           : 0
  .tg_load_avg                   : 0
  .throttled                     : 0
  .throttle_count                : 0
//rt调度器的参数
rt_rq[0]:
  .rt_nr_running                 : 2
  .rt_nr_migratory               : 2
  .rt_throttled                  : 1
  .rt_time                       : 951.483224
  .rt_runtime                    : 950.000000
//deadline的调度器的参数
dl_rq[0]:
  .dl_nr_running                 : 0
  .dl_nr_migratory               : 0
  .dl_bw->bw                     : 996147
  .dl_bw->total_bw               : 0
//线程的具体参数
runnable tasks:
 S           task   PID         tree-key  switches  prio     wait-time             sum-exec        sum-sleep
-----------------------------------------------------------------------------------------------------------
 S        systemd     1      2108.859686      3201   120         0.000000      4752.932816         0.000000 0 0 /autogroup-2
 S       kthreadd     2    142322.322046       279   120         0.000000        13.208422         0.000000 0 0

3. schedstat字段

可以看见此字段如下：

root@ubuntu:/proc# cat schedstat
version 15
timestamp 4298472805
cpu0 0 0 0 0 0 0 1729419922293 4376331060505 800805

//kernel/sched_stats.h
static int __init proc_schedstat_init(void)
{
	proc_create("schedstat", 0, NULL, &proc_schedstat_operations);
	return 0;
}

分别为版本号：SCHEDSTAT_VERSION，此参数内核写死；
时间戳信息，根据时钟中断进行更新；

/* runqueue-specific stats */
		seq_printf(seq,
		    "cpu%d %u %u %u %u %u %u %llu %llu %lu",
		    cpu, rq->yld_count,
		    rq->sched_switch, rq->sched_count, rq->sched_goidle,
		    rq->ttwu_count, rq->ttwu_local,
		    rq->rq_cpu_time,
		    rq->rq_sched_info.run_delay, rq->rq_sched_info.pcount);

		seq_printf(seq, "\n");

第三部分对应为cpu上的rq结构体对应的部分参数；

4.proc/pid/task

此目录显示了进程下的所有线程组的信息，我们在此通过内核源码进行说明；

task目录是如何创建的；

/* for the /proc/TGID/task/ directories */
static int proc_task_readdir(struct file * filp, void * dirent, filldir_t filldir)
{
	struct dentry *dentry = filp->f_path.dentry;
	struct inode *inode = dentry->d_inode;
	struct task_struct *leader = NULL;
	struct task_struct *task;
	int retval = -ENOENT;
	ino_t ino;
	int tid;
	struct pid_namespace *ns;

	task = get_proc_task(inode);
	if (!task)
		goto out_no_task;
	rcu_read_lock();
	if (pid_alive(task)) {
		leader = task->group_leader;
		get_task_struct(leader);
	}
	rcu_read_unlock();
	put_task_struct(task);
	if (!leader)
		goto out_no_task;
	retval = 0;

	switch ((unsigned long)filp->f_pos) {
	case 0:
		ino = inode->i_ino;
		if (filldir(dirent, ".", 1, filp->f_pos, ino, DT_DIR) < 0)
			goto out;
		filp->f_pos++;
		/* fall through */
	case 1:
		ino = parent_ino(dentry);
		if (filldir(dirent, "..", 2, filp->f_pos, ino, DT_DIR) < 0)
			goto out;
		filp->f_pos++;
		/* fall through */
	}

	/* f_version caches the tgid value that the last readdir call couldn't
	 * return. lseek aka telldir automagically resets f_version to 0.
	 */
	ns = filp->f_dentry->d_sb->s_fs_info;
	tid = (int)filp->f_version;
	filp->f_version = 0;
	for (task = first_tid(leader, tid, filp->f_pos - 2, ns);
	     task;
	     task = next_tid(task), filp->f_pos++) {
		tid = task_pid_nr_ns(task, ns);
		if (proc_task_fill_cache(filp, dirent, filldir, task, tid) < 0) {
			/* returning this tgid failed, save it as the first
			 * pid for the next readir call */
			filp->f_version = (u64)tid;
			put_task_struct(task);
			break;
		}
	}
out:
	put_task_struct(leader);
out_no_task:
	return retval;
}