Hi

Warning: Beginner here

I am trying to send a single advertisement packet upon a keypress. To do that, I started with the soc_empty template. I was able to configure the button to trigger an interrupt (I am using GPIOINT). I am also able to see the default advertising. Furthermore I am evaluating the power level through the Energy Profiler. When the default advertising is running, I can perfectly see the power increase due to the advertising.

Now, I tried to take the following code  

  sl_status_t sc;
  bd_addr address;
  uint8_t address_type;
  uint8_t system_id[8];

  sc = sl_bt_system_get_identity_address(&address, &address_type);
  app_assert_status(sc);

  // Pad and reverse unique ID to get System ID.
  system_id[0] = address.addr[5];
  system_id[1] = address.addr[4];
  system_id[2] = address.addr[3];
  system_id[3] = 0xFF;
  system_id[4] = 0xFE;
  system_id[5] = address.addr[2];
  system_id[6] = address.addr[1];
  system_id[7] = address.addr[0];

  sc = sl_bt_gatt_server_write_attribute_value(gattdb_system_id,
                                               0,
                                               sizeof(system_id),
                                               system_id);
  app_assert_status(sc);

  // Create an advertising set.
  sc = sl_bt_advertiser_create_set(&advertising_set_handle);
  app_assert_status(sc);

  // Set advertising interval to 100ms.
  sc = sl_bt_advertiser_set_timing(
    advertising_set_handle,
    1600, // min. adv. interval (milliseconds * 1.6)
    1600, // max. adv. interval (milliseconds * 1.6)
    3000,   // adv. duration
    5);  // max. num. adv. events
  app_assert_status(sc);
  // Start general advertising and enable connections.
  sc = sl_bt_advertiser_start(
    advertising_set_handle,
    advertiser_general_discoverable,
    advertiser_connectable_scannable);
  app_assert_status(sc);

I put that code both into the sl_bt_on_event -> sl_bt_evt_system_boot_id and into the callback function of my interrupt handler (I am not sure if this is inside the interrupt context or the context of my main application - any idea on this?)

When the device starts running, I can nicely see 5 advertisements. Yet, when I press the button I see that the device draws more power than usual which I guess is from the fact that it doesn't go back to EM2 anymore.

I think I am doing something fundamentally wrong here. Anybody care to tell me how something like that would be done correctly?

Thanks, kind regards

Thomas

hi , I'm having problems configuring automation IO GATT service and I don't understand what this error is? How to fix it?

Hello,

I am pretty sure the reply will be no, mainly because, IIUC, the wti32 is able only to Sink or Source audio, not doing both at the same time... In any case, I prefer to ask and having confirmation, having never used this module before...

My need is to sink an audio Stream from a BT node, get it by I2S, process it, and send it back to Another Node using the same I2S Interface... From the BlueGiga stack 6 commands it is not so clear if this is possible, however, reading in this forum, it appears that is is not possible, or at least, it was not some time ago...

So, is this not possible yet...?

Thanks

Kind Regards

Alex

Hi:

I found that sometimes use the API "gecko_cmd_flash_ps_save " to save a record to flash would cost much more time than usual. (More than 10 times) 

My code is working in EFR32BG22C224F512GM32 with mijia_ble_mesh SDK, while the Gecko Micrium OS kernel version is 1.7.3.

Thanks & Regards.

I am part of the Silicon Labs pathfinder program and I recently received a Thunderboard EFR32BG22 kit for testing. I have used the TBS2 in the past and today I finally had some time to spend on playing a bit with this new kit that is based on the Series 2 chip BG22. 

I was planning to spent about half an hour to do some quick testing, just implementing some simple scanner device on top of the soc-empty template.  Now after 30 minutes passed I have been able to:

1. install studio updates and latest Bluetooth SDK (it took way too long)
2. build the soc-empty application for this BG22 kit ( works on 1st attempt)
3. flash the example on my kit => FAIL

I have to say that this has not been to most pleasant unboxing experience. There has been several hickups on the way. And what bugs be is that after spending 30+ minutes on installing stuff, I can not even flash the soc-empty template to my board. (I've used the Silabs tools for years so this should be a walk in the park)

Here's the thing that is blocking me right now:

I can't flash the hex image to my board. I try to do this via right-clicking the hex file and selecting Flash to device... which usually works just fine. Well, something is different with this new kit.

The text in the bottom part of the popup says "lock status unknown". Must be related to some security features in the Series 2 device? Just a guess, I don't know. 

If I press the "click to query lock status" link it does not work. Seems like Studio is not able to read the lock status from the chip.

I tried googling and searching this forum for some pointers, but no results found. 

While googling for help, I did find this appnote that could be related: 
https://www.silabs.com/documents/public/application-notes/an1190-efr32-secure-debug.pdf

From that appnote, I saw that you should be able to poke the security settings by choosing device confuration in Studio. (CHapter 6.1.1 Simplicity Studio in the appnote)

I tried to access the debug settings from the kit, but surprise, it does not work:

When I click the Read From Device button I get the pop-up shown above.

I am starting to suspect that the kit has some security feature enabled and Simplicity Studio is not able to unlock it. NOTE: I did install all available updates today, just before starting to work on the kit.

Has anyone else seen similar issues with the new BG22 Thunderboard kit? I hope there's an easy fix for this, otherwise the kit will end up in the dumpster very soon.

Problem:

Provision is realized with self-provision; communication realized with vendor model.

Device key, net key, app key has a same value.

Using Relay function (gecko_cmd_mesh_test_set_relay (1, 3, 1)),

All devices in vendor model communication are using publish and subscript mode.

Using BGM13S22P; SDK version is Bluetooth 2.11.4.0(v.1.4.3);

Provision realized for 5 devices, one of them control and receive status data from others.

The apartment was organized and provisioned individually for each household;

Device using for Lightning and sockets control with internal relays.

Organization include about 1000 households in 30 floors buildings.

Each household use random generator for generate random Net key and app key.

For first time, provision was operating normally.

After few days (not certain period), randomly 1-2 devices were out of control and didnt share status.

In this case one of the solutions was to turn OFF-ON not operating device’s power supply.

If it doesn't help, turn off the power of all surrounding nodes that were operating normally and then turn them on to operate normally.

We are testing retrieved devices in research lab for reenact it.

After 3 weeks testing that problem doesn’t occur.

Difficulty is that this problem appears randomly only after setting in place.

Questions:

What is the reason of the problem mentioned earlier?

-  Is the problem in vendor model communication in device itself?

Hello,

I am running into trouble while trying to add the NCP serial interface to a DMP application.

I started off with the SOC - Empty - RAIL - DMP example.
I tried adding the the NCP interface (BGApi?) on USART2 to the example.

I ran into problems due to the ncp_gecko.h and rtos_gecko.h overlapping.
I managed to somewhat manage to work around that, but it does not work as expected.

The serial communication is working as such, but I neither get the boot event,
nor the correct responses when running commands.

When sending the system_get_bt_address command (0x20 0x00 0x01 0x03) using HTerm,
for example get the response 0x20 0x02 0x01 0x03 0x83 0x01 instead of the expected response.

I was trying to merge the ncp-empty-target and DMP examples in order to control the Bluetooth from a host IPC.
Is there an easier solution on how to correctly add the NCP functionality to the SOC - Empty - RAIL - DMP example?

Hi,

I need to use simulated eeprom to save a variable that will be incremented basicaly every second (to explain better my question) and I can't loose this information if a have a power fail, for instance. So, I need to write its value every second on flash.

I'm considering 10000 writes on each address of memory, that is the number of erase guaranteed by datasheet.

So, I would have to change the memory address each 10000 writes, right?

Today I'm using ps_keys to save my variables, but, ps_keys has limitted space (0x4000 to 0x407f). Since I have to change the address each 10000 writes, I thing that I need to use another way to save my variable on simulated eeprom.

Is there another aproach to guarantee that I would be able to write this value every second without having problems with max number of writing process on flash?

Thanks a lot.

Hi,

I am a bit lost here.

Hardware: Mighty gecko.

From application note "AN1128: Bluetooth® Coexistence with Wi-Fi" I have copied the following code which I have pasted in soc-Ibeacon application. I am trying to detect when radio is transmitting.
 

// Enable TX_ACT signal through GPIO PD10
#define _PRS_CH_CTRL_SOURCESEL_RAC2 0x00000020UL
#define PRS_CH_CTRL_SOURCESEL_RAC2 (_PRS_CH_CTRL_SOURCESEL_RAC2 << 8)
#define _PRS_CH_CTRL_SIGSEL_RACPAEN 0x00000004UL
#define PRS_CH_CTRL_SIGSEL_RACPAEN (_PRS_CH_CTRL_SIGSEL_RACPAEN << 0)
#define TX_ACTIVE_PRS_SOURCE PRS_CH_CTRL_SOURCESEL_RAC2
#define TX_ACTIVE_PRS_SIGNAL PRS_CH_CTRL_SIGSEL_RACPAEN

#define TX_ACTIVE_PRS_CHANNEL 5
#define TX_ACTIVE_PRS_LOCATION 0
#define TX_ACTIVE_PRS_PORT gpioPortD
#define TX_ACTIVE_PRS_PIN 10
#define TX_ACTIVE_PRS_ROUTELOC_REG ROUTELOC1
#define TX_ACTIVE_PRS_ROUTELOC_MASK (~0x00003F00UL)
#define TX_ACTIVE_PRS_ROUTELOC_VALUE PRS_ROUTELOC1_CH5LOC_LOC0 // PD10
#define TX_ACTIVE_PRS_ROUTEPEN PRS_ROUTEPEN_CH5PEN

void configure_rca_tx()
{
	CMU_ClockEnable(cmuClock_PRS, true);

	// Setup PRS input as TX_ACTIVE signal
	PRS_SourceAsyncSignalSet(TX_ACTIVE_PRS_CHANNEL, TX_ACTIVE_PRS_SOURCE, TX_ACTIVE_PRS_SIGNAL);

	// enable TX_ACTIVE output pin with initial value of 0
	GPIO_PinModeSet(TX_ACTIVE_PRS_PORT, TX_ACTIVE_PRS_PIN, gpioModePushPull, 0);

	// Route PRS CH/LOC to TX Active GPIO output
    PRS->TX_ACTIVE_PRS_ROUTELOC_REG =
			(PRS->TX_ACTIVE_PRS_ROUTELOC_REG & TX_ACTIVE_PRS_ROUTELOC_MASK) | TX_ACTIVE_PRS_ROUTELOC_VALUE;

	PRS->ROUTEPEN |= TX_ACTIVE_PRS_ROUTEPEN;
}

Function "configure_rca_tx()" is called once evt_system_boot_id is received,  unfortunately I am getting no output whatsoever in PD10. 

Any idea on how to be able to see TX_ACTIVE on a GPIO will be much appreciated.

Regards.

O.
 

Dear Sir,

I have merged the codes of scan function used at "EFR32BG13P Thermo Client" project into the "mesh provisioner" project (provisioner firmware run at EFR32BG13P EVK) and try to scan the traditional BLE gatt server device but fail to get any gatt server devices information after the scan period time-out. But I can scan the mesh nodes per the same mesh provisioner firmware. So how can I successfully scan the traditional BLE gatt server devices by the "mesh provisioner" ? What else functions should I need to add to support this function ? Thanks.

BRs,

Bolo

Hello,

I use gecko 3.0 sdk with BGM13S chip.

I see that I have two different files where the stack and heap are declared: sl_memory.c (variables SL_STACK_SIZE and SL_HEAP_SIZE) and startup_bgm13.c (variables __STACK_SIZE and __HEAP_SIZE).

Why so, what variables I suppose to configure to declare stack and heap size?

Hi,

Is posible to know in code if the advertising is working or not ? 

I notice that sometimes my peripherals are advertising for hours and days and suddenly they stops advertising and I'm not able to discover the issue.

My idea is to poll to the stack which is its state, and when it is detected that advertising is stopped re-start it.  Would it be possible to do that ?

Thanks in advance,

Asier