线程堆栈使用异常，程序死机，不知如何排查问题

发布于 2021-10-25 10:33:49 浏览：2669 订阅该版

问题描述:

- 最近程序容易死机，调试的时候么有发信啊明显的程序问题，但是有个异常现象，发现有个线程的堆栈使用很高
    
```
    00> thread     pri  status      sp     stack size max used left tick  error
    00> ---------- ---  ------- ---------- ----------  ------  ---------- ---
    00> thd_usbd_m   5  suspend 0x00000180 0x00000600    36%   0x0000000a 000
    00> thd_usbd_r   5  suspend 0x00000088 0x00000600    08%   0x00000001 000
    00> thd_usbd_s   4  suspend 0x00000088 0x00000400    13%   0x00000001 000
    00> thd_usbh_m   5  suspend 0x00000080 0x00000400    32%   0x0000000a 000
    00> thd_voc_s1   8  suspend 0x00000088 0x00000400    28%   0x0000000a 000
    00> thd_voc_m1   8  suspend 0x00000188 0x00000400    63%   0x00000009 000
    00> thd_voc_s0   8  suspend 0x00000088 0x00000400    28%   0x0000000a 000
    00> thd_voc_m0   8  suspend 0x00000188 0x00000400    63%   0x00000009 000
    00> thd_dtq_s    3  suspend 0x000000a4 0x00000200    32%   0x00000001 000
    00> thd_dtq_r    2  suspend 0x000000ac 0x00001000    35%   0x00000001 000
    00> thd_dtq_m3   5  suspend 0x00000194 0x00000800    39%   0x0000000a 000
    00> thd_dtq_m2   5  suspend 0x00000194 0x00000800    44%   0x0000000a 000
    00> thd_dtq_m1   5  suspend 0x00000194 0x00000800    45%   0x00000008 000
    00> thd_dtq_m0   5  suspend 0x00000194 0x00000800    47%   0x0000000a 000
    00> thd_card     6  suspend 0x000000ac 0x00000800    16%   0x00000006 000
    00> thd_spi_s    5  suspend 0x0000007c 0x00000400    12%   0x00000001 000
    00> thd_lcd_m   19  suspend 0x000000a4 0x00000a00    97%   0x00000032 000 [栈使用高到98%]
    00> thd_bee_d    4  suspend 0x000000a4 0x00000400    16%   0x0000000a 000
    00> tshell      20  running 0x00000198 0x00000800    27%   0x00000008 000
    00> tidle0      31  ready   0x000000b0 0x00000400    17%   0x00000018 000
    00> main        10  suspend 0x0000010c 0x00000800    23%   0x00000005 000
```
    
- 增大堆栈测试，同样较高
    
```
    00> thread     pri  status      sp     stack size max used left tick  error
    00> ---------- ---  ------- ---------- ----------  ------  ---------- ---
    00> thd_usbd_m   5  suspend 0x00000180 0x00000600    36%   0x0000000a 000
    00> thd_usbd_r   5  suspend 0x00000088 0x00000600    08%   0x00000001 000
    00> thd_usbd_s   4  suspend 0x00000088 0x00000400    13%   0x00000001 000
    00> thd_usbh_m   5  suspend 0x00000080 0x00000400    32%   0x0000000a 000
    00> thd_voc_s1   8  suspend 0x00000088 0x00000400    28%   0x0000000a 000
    00> thd_voc_m1   8  suspend 0x00000188 0x00000400    63%   0x00000009 000
    00> thd_voc_s0   8  suspend 0x00000088 0x00000400    28%   0x0000000a 000
    00> thd_voc_m0   8  suspend 0x00000188 0x00000400    63%   0x00000009 000
    00> thd_dtq_s    3  suspend 0x000000a4 0x00000200    32%   0x00000001 000
    00> thd_dtq_r    2  suspend 0x000000ac 0x00001000    35%   0x00000001 000
    00> thd_dtq_m3   5  suspend 0x00000194 0x00000800    39%   0x0000000a 000
    00> thd_dtq_m2   5  suspend 0x00000194 0x00000800    44%   0x00000009 000
    00> thd_dtq_m1   5  suspend 0x00000194 0x00000800    44%   0x0000000a 000
    00> thd_dtq_m0   5  suspend 0x00000194 0x00000800    44%   0x00000009 000
    00> thd_card     6  suspend 0x000000ac 0x00000800    16%   0x0000000a 000
    00> thd_spi_s    5  suspend 0x0000007c 0x00000400    12%   0x00000001 000
    00> thd_lcd_m   19  suspend 0x000000a4 0x00001000    98%   0x0000001d 000  [栈使用高到98%]
    00> thd_bee_d    4  suspend 0x000000a4 0x00000400    16%   0x0000000a 000
    00> tshell      20  running 0x00000198 0x00000800    28%   0x00000006 000
    00> tidle0      31  ready   0x000000b0 0x00000400    17%   0x00000018 000
    00> main        10  suspend 0x0000010c 0x00000800    20%   0x00000005 000
```
    
- 再次增大，同样较高
    
```
    00> thread     pri  status      sp     stack size max used left tick  error
    00> ---------- ---  ------- ---------- ----------  ------  ---------- ---
    00> thd_usbd_m   5  suspend 0x00000180 0x00000600    36%   0x0000000a 000
    00> thd_usbd_r   5  suspend 0x00000088 0x00000600    08%   0x00000001 000
    00> thd_usbd_s   4  suspend 0x00000088 0x00000400    13%   0x00000001 000
    00> thd_usbh_m   5  suspend 0x00000080 0x00000400    32%   0x0000000a 000
    00> thd_voc_s1   8  suspend 0x00000088 0x00000400    28%   0x0000000a 000
    00> thd_voc_m1   8  suspend 0x00000188 0x00000400    63%   0x00000009 000
    00> thd_voc_s0   8  suspend 0x00000088 0x00000400    28%   0x0000000a 000
    00> thd_voc_m0   8  suspend 0x00000188 0x00000400    63%   0x00000009 000
    00> thd_dtq_s    3  suspend 0x000000a4 0x00000200    32%   0x00000001 000
    00> thd_dtq_r    2  suspend 0x000000ac 0x00001000    35%   0x00000001 000
    00> thd_dtq_m3   5  suspend 0x00000194 0x00000800    39%   0x0000000a 000
    00> thd_dtq_m2   5  suspend 0x00000194 0x00000800    44%   0x0000000a 000
    00> thd_dtq_m1   5  suspend 0x00000194 0x00000800    44%   0x00000009 000
    00> thd_dtq_m0   5  suspend 0x00000194 0x00000800    44%   0x0000000a 000
    00> thd_card     6  suspend 0x000000ac 0x00000800    16%   0x00000009 000
    00> thd_spi_s    5  suspend 0x0000007c 0x00000400    12%   0x00000001 000
    00> thd_lcd_m   19  suspend 0x000000a4 0x00001800    98%   0x00000007 000
    00> thd_bee_d    4  suspend 0x000000a4 0x00000400    16%   0x0000000a 000
    00> tshell      20  running 0x00000230 0x00000800    27%   0x00000007 000
    00> tidle0      31  ready   0x000000b0 0x00000400    17%   0x0000001b 000
    00> main        10  suspend 0x0000010c 0x00000800    20%   0x00000005 000
```
    
- 于是屏蔽线程函数的操作，再次测试，还是高
    
```c
    void fun_lcd_m(void *param)
    {
      rt_uint32_t e;
    
      while (1)
      {
    //    if (rt_event_recv(&dat_event, 1, RT_EVENT_FLAG_OR | RT_EVENT_FLAG_CLEAR,
    //                      RT_WAITING_FOREVER, &e) == RT_EOK)
    //    {
    //      /* s0: LCD ½çÃæË¢ÐÂ */
    //      switch (sys_tcb_val.work_mode)
    //      {
    //      case SERVICE_MODE: // Î¬ÐÞÄ£Ê½
    //        drv_lcd_service_mode();
    //        break;
    //      default:
    //        drv_lcd_normal_mode();
    //        break;
    //      }
    
    //      /* s1: LCD ´íÎó¼ì²â */
    //      app_usb_err_check();
    //      app_port_err_check();
    //      app_card_err_check();
    
    //      /* s2: LCD Ë¢ÐÂÊý¾Ý */
    //      if (drv_lcd_uc1617_show())
    //      {
    //        LOG_D("LCD: uc1617 flash error");
    //        drv_lcd_uc1617_init();
    //        drv_lcd_uc1617_show();
    //      }
    
    //      /* s3: Ë¢¿¨Ë²Ì¬·µ»Ø */
    //      if ((lcd_tcb.card_cnt == card_tcb.card_cnt) && (lcd_tcb.card_s == 1))
    //      {
    //        lcd_tcb.card_s = 0;
    //        rt_event_send(&dat_event, 1);
    //      }
    
    //      /* s4: ·äÃùÌáÊ¾ */
    //      if ((bee_tcb.s > 0) && (sys_tcb_val.sys_swd.flg == 0))
    //      {
    //        rt_event_send(&bee_event, bee_tcb.s);
    //        bee_tcb.s = 0;
    //      }
    //    }
    		rt_thread_mdelay(200);
      }
    }
```
```
    00> thread     pri  status      sp     stack size max used left tick  error
    00> ---------- ---  ------- ---------- ----------  ------  ---------- ---
    00> thd_usbd_m   5  suspend 0x00000180 0x00000600    36%   0x0000000a 000
    00> thd_usbd_r   5  suspend 0x00000088 0x00000600    08%   0x00000001 000
    00> thd_usbd_s   4  suspend 0x00000088 0x00000400    13%   0x00000001 000
    00> thd_usbh_m   5  suspend 0x00000080 0x00000400    32%   0x0000000a 000
    00> thd_voc_s1   8  suspend 0x00000088 0x00000400    28%   0x0000000a 000
    00> thd_voc_m1   8  suspend 0x00000188 0x00000400    63%   0x00000009 000
    00> thd_voc_s0   8  suspend 0x00000088 0x00000400    28%   0x0000000a 000
    00> thd_voc_m0   8  suspend 0x00000188 0x00000400    63%   0x00000009 000
    00> thd_dtq_s    3  suspend 0x000000a4 0x00000200    32%   0x00000001 000
    00> thd_dtq_r    2  suspend 0x000000ac 0x00001000    35%   0x00000001 000
    00> thd_dtq_m3   5  suspend 0x00000194 0x00000800    39%   0x0000000a 000
    00> thd_dtq_m2   5  suspend 0x00000194 0x00000800    44%   0x0000000a 000
    00> thd_dtq_m1   5  suspend 0x00000194 0x00000800    44%   0x00000009 000
    00> thd_dtq_m0   5  suspend 0x00000194 0x00000800    44%   0x0000000a 000
    00> thd_card     6  suspend 0x000000ac 0x00000800    16%   0x0000000a 000
    00> thd_spi_s    5  suspend 0x0000007c 0x00000400    12%   0x00000001 000
    00> thd_lcd_m   19  suspend 0x0000007c 0x00001800    98%   0x00000032 000 [栈使用高到98%]
    00> thd_bee_d    4  suspend 0x000000a4 0x00000400    16%   0x0000000a 000
    00> tshell      20  running 0x00000210 0x00000800    27%   0x00000007 000
    00> tidle0      31  ready   0x000000b0 0x00000400    17%   0x0000001e 000
    00> main        10  suspend 0x0000010c 0x00000800    23%   0x00000005 000
```

我是否触发了rtt的隐藏BUG，这个统计感觉错的？，这个问题如何排查，总感觉有个系统级别的问题在，即使程序一会儿不死，也不放心