cancel
Showing results for 
Search instead for 
Did you mean: 

ACI_HAL_LE_TX_TEST_PACKET_NUMBER returns wrong number of packets.

mfoms.1
Associate

Hello,

I noticed that ACI_HAL_LE_TX_TEST_PACKET_NUMBER command returns wrong number of packets. It seems that the value is trimmed to two bytes.

Steps to reproduce:

  1. I run the HCI_LE_TRANSMITTER_TEST command for specified time.
  2. After this specified time expires i send HCI_LE_TEST_END command.
  3. Then i send ACI_HAL_LE_TX_TEST_PACKET_NUMBER command, which returns number of packets which is wrong when test duration is long enough to send more than 65535 packets.

I've measured number of packets that device sends in 10 seconds. It's equal 15969.

Then i was increasing time by 10 seconds and number of packets returned by ACI_HAL_LE_TX_TEST_PACKET_NUMBER command was as expected until 50seconds.

Below is the table with results:

10s : 15969 packets

20s : 31954 packets

30s : 47938 packets

40s : 63923 packets

50s : 14371 packets (expected about 79890 -> 14371 + 65536 = 79907)

60s : 30355 packets (expected about 95870 -> 30355 + 65536 = 95891)

Also i've checked content of buffer returned from api function and just as expected higher two bytes of "number_of_packets part" are always zeroes.

Could anyone check and confirm this behaviour?

__

mcu: STM32WB55

stack: HCI EXTENDED 1.13.2

FUS: 1.2.0

3 REPLIES 3
Remy ISSALYS
ST Employee

Hello,

I'll keep you posted with the updates. An internal ticket has been open, ticket number: 139445 (This is an internal tracking number and is not accessible or usable by customers).

Best Regards

Remy ISSALYS
ST Employee

Hello,

There is a mistake in the documentation, the counter is based on a 16-bit counter and not a 32-bit counter. The documentation will be updated in the next version.

Best Regards

mfoms.1
Associate

Hi, thanks for explanation.

Best regards.