LLPM on two NRF5340

[SamA]

Trying to test the 1ms sample on the NRF5340DK.  Loaded and ran the app without problem.. it starts with 9XXXXus.  Runs with couple of drops.  But when 1ms interval is started.  I can't sent the packet.  ("Did not receive a latency respond" is shown on both sides.)

I have tried external powering up the board with USB (using usb charge on the terminal, and the other usb for debug port).  Still the same.  

Using the NRF connect SDK to create the project in Segger studio.  proj.conf has the following (default project file, have not touched anything).  Please help. 

CONFIG_NCS_SAMPLES_DEFAULTS=y

CONFIG_CONSOLE_SUBSYS=y
CONFIG_CONSOLE_HANDLER=y
CONFIG_CONSOLE_GETCHAR=y

CONFIG_BT_CTLR_LLPM=y

CONFIG_BT_DEVICE_NAME="Nordic_LLPM"
CONFIG_BT=y
CONFIG_BT_DEBUG_LOG=y
CONFIG_BT_SMP=y
CONFIG_BT_PERIPHERAL=y
CONFIG_BT_CENTRAL=y
CONFIG_BT_MAX_CONN=2

CONFIG_BT_SCAN=y
CONFIG_BT_SCAN_FILTER_ENABLE=y
CONFIG_BT_SCAN_UUID_CNT=1

CONFIG_BT_LATENCY=y
CONFIG_BT_LATENCY_CLIENT=y

CONFIG_BT_LL_SOFTDEVICE_VS_INCLUDE=y
CONFIG_BT_LL_SOFTDEVICE=y
CONFIG_BT_HCI_VS_EVT_USER=y
CONFIG_HEAP_MEM_POOL_SIZE=2048
CONFIG_BT_GAP_PERIPHERAL_PREF_PARAMS=n

[Einar Thorsrud]

Hi,

Have you set CONFIG_BT_CTLR_LLPM=y in the prj.conf for the hci_rpmsg project as well (which is for the controller that runs on the network core)?

[SamA]

I have opened and rebuild the hci_rpmsg project CONFIG_BT_CTLR_LLPM=y

New to the dual core system. 

Are you suppose to flash the net core first with the hci_rpmsg project. then open the the llpm project and flash to the app core?   (tried and failed also).  Seems like the llpm project will erase the entire flash during llpm flashing. 

Would the llpm project on it's on picks up the last compiled net core bin and flash to the net core as well?

Tried many approach. still can only do normal speed.. and not 1ms.  Please help.  Need to evaluate this for an up coming project. 

[Einar Thorsrud]

Hi,

It turns out that LLPM does not work properly on the nRF53 yet, and it is still unsupported. There are two workarounds you can use in the meantime though (one of them is enough):

• Increase the interval to >= 2000 us
• Run with the app on the network core

Edit: One more thing you can do when running a split configuration (host and controller on separate cores) it can help to disable QoS reports by commenting out these lines in the sample:

	if (enable_qos_conn_evt_report()) {
		printk("Enable LLPM QoS failed.\n");
		return;
	}

[SamA]

Good to know, but it will eventually right?

[Einar Thorsrud]

Yes, I expect so. I is currently being worked on.