The radio_test application doesn't receive input on nrf52840 chip, while the echo_bot example does.

[xiongwei.wang]

I encounter this problem:

The radio_test application works correctly on the official nrf52840dk board, but cannot receive input commands on my custom product board.

However, the echo_bot example works fine on the same custom board.

I tried adding the following configuration:

```conf

CONFIG_CONSOLE=y
CONFIG_SERIAL=y
CONFIG_UART_CONSOLE=y
CONFIG_SHELL_BACKEND_SERIAL=y

```

I also tried to disable hardware flow control in code:

```c

    nrf_uarte_hwfc_pins_disconnect(NRF_UART0);

```

I noticed that echo_bot doesn't disable hardware flow control but still works properly.

My product uses the nRF52840 chip connected through an external J-Link, and I'm using NCS v2.9.0.

Questions: What should I do? What configuration or code do I need to add to solve this problem?

[xiongwei.wang]

I find the true problem, the following function execution failes. 

```c

err = sys_notify_fetch_result(&clk_cli.notify, &res);

```

I printed the err value and compare it with the offciail board's output. The err should be zero but on my custom board is -11.
And I examined the function:
```c

static inline int sys_notify_fetch_result(const struct sys_notify *notify,
int *result)
{
       __ASSERT_NO_MSG(notify != NULL);
       __ASSERT_NO_MSG(result != NULL);
      int rv = -EAGAIN;

      if (sys_notify_get_method(notify) == SYS_NOTIFY_METHOD_COMPLETED) {
           rv = 0;
          *result = notify->result;
     }

     return rv;
}

```
I found the `-EAGAIN` is `-11`, which means the condition `sys_notify_get_method(notify) == SYS_NOTIFY_METHOD_COMPLETED` is

never satisfied.

[xiongwei.wang]

But I haven't yet solved this problem. I lost track of this thread. Does anybody know the reason?

[Susheel Nuguru]

Wang, I just took this thread and I will try to look into it and comeback to you by tomorrow evening with my observations. Thanks for your patience.

[xiongwei.wang]

Thank you, Nuguru. I am very glad to receive your reply. And I want to provide some addition details.
The issue occurss in the clock_init function:

static void clock_init(void)
{
	int err = 0;
	int res = 0;

	struct onoff_manager *clk_mgr;
	struct onoff_client clk_cli;

	clk_mgr = z_nrf_clock_control_get_onoff(CLOCK_CONTROL_NRF_SUBSYS_HF);
	if (!clk_mgr) {
		printk("Unable to get the Clock manager\n");
		return;
	}

	sys_notify_init_spinwait(&clk_cli.notify);

	err = onoff_request(clk_mgr, &clk_cli);
	if (err < 0) {
		printk("Clock request failed: %d\n", err);
		return;
	}

	do {
		err = sys_notify_fetch_result(&clk_cli.notify, &res);
		if (!err && res) {
			printk("Clock could not be started: %d\n", res);
			return;
		}
	} while (err);
	printk("Clock has started\n");
}

And I am debuggin tow board simultaneously to observe the different. I found the variable clk_cli.notify.flags will change its value from 1 to 0 on the official boards. However, on my custuom boards, it remains 1 after the onff_request call.

The res variable behaves the same on both boards, always remaining zero.

Thanks for looking into this issue.

[xiongwei.wang]

I think I found the specific problem, I tried to install nRF5 SDK to run the radio test, and I dicovered it gets stuck at following code:

/** @brief Function for configuring all peripherals used in this example.
 */
static void clock_init(void)
{
    // Start 64 MHz crystal oscillator.
    NRF_CLOCK->EVENTS_HFCLKSTARTED = 0;
    NRF_CLOCK->TASKS_HFCLKSTART    = 1;

    // Wait for the external oscillator to start up.
    while (NRF_CLOCK->EVENTS_HFCLKSTARTED == 0)
    {
        // Do nothing.
    }
}

I searched in the forums and saw many people get stuck here due to crystal issues.Since I have limited knowledge about hardware, I need to confirm whether this is a hardware issue about our board.Thanks everyone, especially Nugruru.