LAUNCHXL-CC26X2R1: Power consumption issue when running simplelink_zigbee_sdk_plugin_2_20_00_06 zed_switch on LAUNCHXL-CC26X2R1

1. I define POWER_MEAS, POWER_SAVING, ZED_POWER_CONSUMPTION_WORKAROUND, remove BOARD_DISPLAY_USE_UART, and add Board_shutDownExtFlash() to main.c

2. Since energy trace need JTag which usesTDO/TDI so I can only remove 5V, RXD/TXD, SWO jumpers. I also power the MCU externally.

After I free run the project, I see the mean current is 0.002mA but I see it shows 3mA on my power supply to power LAUNCHXL-CC26X2R1 externally. If you look at my new attached energy trace screen shot, you can see RX is still not turn off after device is power on and doesn't join any Zigbee network. I think those defines doesn't fix the issue and please help to verify this from your side. Thanks.

Hello Ryan and YK

I have test this issue by simplelink_zigbee_sdk_plugin_0_95_00_18 .

You need remove UI functions in zed_switch_cc26x2lp.

1. Define POWER_MEAS, POWER_SAVING, ZED_POWER_CONSUMPTION_WORKAROUND, remove BOARD_DISPLAY_USE_UART, and add Board_shutDownExtFlash() to main.c
2. Remove all unnecessary isolation jumpers (5V, RXD/TXD, SWO, and TDO/TDI if using 2-pin JTAG for EnergyTrace only, 4-pin is required for ++) and power the MCU externally (Unpopulate VSENS jumper and connect VCC to middle pin, only connecting the USB cable after voltage is applied).
3. Free run your project inside the debugger.
4. Test EnergyTrace results against a power amplifier measurement.

Please check my test screen shot. You hope you test  in simplelink_zigbee_sdk_plugin_2_20_00_06..

We need add code in zclSampleDoorLock_ProcessCommissioningStatus :

static void zclSampleDoorLock_ProcessCommissioningStatus(bdbCommissioningModeMsg_t *bdbCommissioningModeMsg)
{
switch(bdbCommissioningModeMsg->bdbCommissioningMode)
{
case BDB_COMMISSIONING_FORMATION:
if(bdbCommissioningModeMsg->bdbCommissioningStatus == BDB_COMMISSIONING_SUCCESS)
{
zstack_bdbStartCommissioningReq_t zstack_bdbStartCommissioningReq;

//After formation, perform nwk steering again plus the remaining commissioning modes that has not been process yet
zstack_bdbStartCommissioningReq.commissioning_mode = BDB_COMMISSIONING_MODE_NWK_STEERING | bdbCommissioningModeMsg->bdbRemainingCommissioningModes;
Zstackapi_bdbStartCommissioningReq(zclSampleDoorLock_Entity,&zstack_bdbStartCommissioningReq);
}
else
{
//Want to try other channels?
//try with bdb_setChannelAttribute
}
break;
case BDB_COMMISSIONING_NWK_STEERING:
if(bdbCommissioningModeMsg->bdbCommissioningStatus == BDB_COMMISSIONING_SUCCESS)
{
//YOUR JOB:
//We are on the nwk, what now?
}
else
{
 // standby

}
break;
case BDB_COMMISSIONING_FINDING_BINDING:
if(bdbCommissioningModeMsg->bdbCommissioningStatus == BDB_COMMISSIONING_SUCCESS)
{
//YOUR JOB:
}
else
{
//YOUR JOB:
//retry?, wait for user interaction?
}
break;
case BDB_COMMISSIONING_INITIALIZATION:
//Initialization notification can only be successful. Failure on initialization
//only happens for ZED and is notified as BDB_COMMISSIONING_PARENT_LOST notification

//YOUR JOB:
//We are on a network, what now?

break;
#if ZG_BUILD_ENDDEVICE_TYPE
case BDB_COMMISSIONING_PARENT_LOST:
if(bdbCommissioningModeMsg->bdbCommissioningStatus == BDB_COMMISSIONING_NETWORK_RESTORED)
{
//We did recover from losing parent
}
else
{
//Parent not found, attempt to rejoin again after a fixed delay
Timer_setTimeout( EndDeviceRejoinClkHandle, SAMPLEAPP_END_DEVICE_REJOIN_DELAY );
Timer_start(&EndDeviceRejoinClkStruct);
}
break;
#endif
}

UI_UpdateComissioningStatus(bdbCommissioningModeMsg);
}

this's right ?

Ryan Brown1 I add the following red lines according to your seggestion but I still see the same problem.

static void zclSampleSw_initialization(void)
{

    /* Initialize user clocks */
    zclSampleSw_initializeClocks();

    /* Initialize keys */
    Board_Key_initialize(zclSampleSw_changeKeyCallback);

    /* Initialize the LEDS */
    Board_Led_initialize();

    // Register the current thread as an ICall dispatcher application
    // so that the application can send and receive messages.
    ICall_registerApp(&zclSampleSw_Entity, &sem);


    //Initialize stack
    zclSampleSw_Init();
    {
        zstack_bdbStartCommissioningReq_t zstack_bdbStartCommissioningReq;
        zstack_bdbStartCommissioningReq.commissioning_mode = BDB_COMMISSIONING_MODE_NWK_STEERING;
        Zstackapi_bdbStartCommissioningReq(zclSampleSw_Entity,&zstack_bdbStartCommissioningReq);
    }
}

Ryan Brown1 Just test your latest patch and it doesn't work too. I still see RX is on in energytrace after device starts.

Ryan Brown1 I revise the following red codes and test it. However, I still see the same problem. Do I miss anything?

void osal_run_system( void )
{
  uint8 idx = 0;

#ifdef USE_ICALL
  uint32 next_timeout_prior = osal_next_timeout();
#else /* USE_ICALL */
#ifndef HAL_BOARD_CC2538
  osalTimeUpdate();
#endif

  Hal_ProcessPoll();
#endif /* USE_ICALL */

#ifdef USE_ICALL
  {
    /* Update osal timers to the latest before running any OSAL processes
     * regardless of wakeup callback from ICall because OSAL timers are added
     * relative to the current time. */
    unsigned long newtimestamp = ICall_getTicks();
    uint32 milliseconds;

    if (osal_tickperiod == 1000)
    {
      milliseconds = newtimestamp - osal_last_timestamp;
      osal_last_timestamp = newtimestamp;
    }
    else
    {
      unsigned long long delta = (unsigned long long)
        ((newtimestamp - osal_last_timestamp) & 0xfffffffful);
      delta *= osal_tickperiod;
      delta /= 1000;
      milliseconds = (uint32) delta;
      osal_last_timestamp += (uint32) (delta * 1000 / osal_tickperiod);
    }
    osalAdjustTimer(milliseconds);
    /* Set a value that will never match osal_next_timeout()
     * return value so that the next time can be scheduled.
     */
    next_timeout_prior = 0xfffffffful;
  }
  if (osal_eventloop_hook)
  {
    osal_eventloop_hook();
  }

  for (;;)
  {
    void *msg;
    ICall_EntityID src, dst;
    osal_msg_hdr_t *hdr;
    uint8 dest_id;

    if (ICall_fetchMsg(&src, &dst, &msg) != ICALL_ERRNO_SUCCESS)
    {
      break;
    }

    hdr = (osal_msg_hdr_t *) msg - 1;

#ifdef ICALL_LITE
    if (hdr->format == ICALL_MSG_FORMAT_DIRECT_API_ID)
    {
        if (osal_icallMsg_hook)
        {
          (*osal_icallMsg_hook)(msg);
        }
    }
    else
#endif /* ICALL_LITE */      
    {
      dest_id = osal_dispatch2id(dst);
      if (dest_id == TASK_NO_TASK)
      {
        /* Something wrong */
        ICall_abort();
      }
      else
      {
        /* Message towards one of the tasks */
        /* Create a proxy task ID if necessary and
         * queue the message to the OSAL internal queue.
         */
        uint8 proxyid = osal_alien2proxy(hdr->srcentity);

        if (hdr->format == ICALL_MSG_FORMAT_1ST_CHAR_TASK_ID)
        {
          uint8 *bytes = msg;
          *bytes = proxyid;
        }
        else if (hdr->format == ICALL_MSG_FORMAT_3RD_CHAR_TASK_ID)
        {
          uint8 *bytes = msg;
          bytes[2] = proxyid;
        }
        /* now queue the message to the OSAL queue */
        osal_msg_send(dest_id, msg);
      }
    }
  }
#endif /* USE_ICALL */

  do {
    if (macTasksEvents[idx])  // Task is highest priority that is ready.
    {
      break;
    }
  } while (++idx < tasksCnt);

  if (idx < tasksCnt)
    {
      uint16 events;
      halIntState_t intState;

      HAL_ENTER_CRITICAL_SECTION(intState);
      events = macTasksEvents[idx];
      macTasksEvents[idx] = 0;  // Clear the Events for this task.
      HAL_EXIT_CRITICAL_SECTION(intState);

      activeTaskID = idx;
      events = (tasksArr[idx])( idx, events );
      activeTaskID = TASK_NO_TASK;

      HAL_ENTER_CRITICAL_SECTION(intState);
      macTasksEvents[idx] |= events;  // Add back unprocessed events to the current task.
      HAL_EXIT_CRITICAL_SECTION(intState);
    }
#if defined ( POWER_SAVING)
#if defined (USE_ICALL)
  else
  {
    MAP_macPwrVote(FALSE, FALSE);
  }
#else
  else
  {
    osal_pwrmgr_powerconserve();
  }
#endif
#endif
  /* Yield in case cooperative scheduling is being used. */
#if defined (configUSE_PREEMPTION) && (configUSE_PREEMPTION == 0)
  {
    osal_task_yield();
  }
#endif

#if defined USE_ICALL
  /* Note that scheduling wakeup at this point instead of
   * scheduling it upon ever OSAL start timer request,
   * would only work if OSAL start timer call is made
   * from OSAL tasks, but not from either ISR or
   * non-OSAL application thread.
   * In case, OSAL start timer is called from non-OSAL
   * task, the scheduling should be part of OSAL_Timers
   * module.
   * Such a change to OSAL_Timers module was not made
   * in order not to diverge the OSAL implementations
   * too drastically between pure OSAL solution vs.
   * OSAL upon service dispatcher (RTOS).
   * TODO: reconsider the above statement.
   */
  {
    halIntState_t intState;

    uint32 next_timeout_post = osal_next_timeout();
    if (next_timeout_post != next_timeout_prior)
    {
      /* Next wakeup time has to be scheduled */
      if (next_timeout_post == 0)
      {
        /* No timer. Set time to the max */
        next_timeout_post = OSAL_TIMERS_MAX_TIMEOUT;
      }
      if (next_timeout_post > osal_max_msecs)
      {
        next_timeout_post = osal_max_msecs;
      }
      /* Restart timer */
      HAL_ENTER_CRITICAL_SECTION(intState);
      ICall_stopTimer(osal_timerid_msec_timer);
      ICall_setTimerMSecs(next_timeout_post, osal_msec_timer_cback,
                          (void *) (++osal_msec_timer_seq),
                          &osal_timerid_msec_timer);
      HAL_EXIT_CRITICAL_SECTION(intState);
    }
    
#ifdef ICALL_EVENTS
    /*
     * When using events, OSAL needs to be sure that it will not block itself
     * if any message have yet to been processed.  The OSAL event must be
     * signaled when any messages remain unprocessed at the end of this
     * function.
     */
    if (osal_qHead)
    {
      ICall_signal(osal_syncHandle);
    }
    else
    {
      for (uint8 i = 0; i < tasksCnt; i++)
      {
        if (macTasksEvents[i] != 0)
        {
          ICall_signal(osal_syncHandle);
        }
      }
    }
#endif /* ICALL_EVENTS */
  }
#endif /* USE_ICALL */
}

Ryan Brown1Thanks for following up Energy Trace issue and if it's possible, please notify me when it is fixed.