Open ppryga-nordic opened 2 years ago
@ppryga-nordic will #48936 fix this issue?
@cvinayak No, the PR https://github.com/zephyrproject-rtos/zephyr/pull/48936 handles insufficient node_rx_iq_report
for periodic scanner only.
There is no code that report insufficient resources for DF connected mode. That should be implemented.
Actually I suggest a change in the PR. Witch changes to lll_df_conf_cte_rx_enable
that remove LL_ASSERT
the connected mode will execute reception of CTE even there is no memory for IQ report. Hence proper handling this situation should be implemented in the PR or value returned by lll_df_conf_cte_rx_enable
should be checked in lll_conn.c
and lll_peripheral.c
.
Describe the bug There is an
LL_ASSERT
that fires: https://github.com/zephyrproject-rtos/zephyr/blob/a150c7ee77687d543b903ea9008754da7415f4b5/subsys/bluetooth/controller/ll_sw/nordic/lll/lll_df.c#L246-L247Conditions:
node_rx_iq_report
in a pool to handle so frequent IQ report generation.To Reproduce Steps to reproduce the behavior:
direction_finding_central
and change it to use conn interval 7.5ms and CTE_REQ every connection event.direction_finding_peripheral
as a peer device.Expected behavior No LL_ASSERT fires.
Impact Low. Happens only when DF is enabled in very specific configuration. Small increase to connection interval should help handle the issue.
Additional context The proper way to handle the issue is to send an IQ report with
Pocket_Status=0xFF
andSample_Count=0x0