Other Parts Discussed in Thread: WL1837
Hi Team,
Our customers have reported to us that wl1837 Bluetooth occasionally (1/4) fails to wake up at boot time in WL1837 Service Pack 4.3.
The reproduction steps:
Our customers have reported to us that wl1837 Bluetooth occasionally (1/4) fails to wake up at boot time in WL1837 Service Pack 4.3.
The reproduction steps:
- In our boot service script, it runs:
- echo "1" > /sys/class/gpio/gpio485/value # enable BT_EN
- hciattach /dev/ttyS5 texas
- The error message is shown as below:
Nov 29 09:10:20 Moxa run.sh[394]: Found a Texas Instruments' chip! Nov 29 09:10:20 Moxa run.sh[394]: Firmware file : /lib/firmware/ti-connectivity/ TIInit_11.8.32.bts Nov 29 09:10:20 Moxa run.sh[394]: Loaded BTS script version 1 Nov 29 09:10:20 Moxa run.sh[394]: texas: changing baud rate to 3000000, flow con trol to 1 Nov 29 09:10:30 Moxa run.sh[394]: Initialization timed out. Nov 29 09:10:30 Moxa systemd[1]: sentinel.service: Main process exited, code=exited, status=1/FAILURE Nov 29 09:10:30 Moxa systemd[1]: Failed to start UC3100 sentinel. Nov 29 09:10:30 Moxa systemd[1]: sentinel.service: Unit entered failed s tate. Nov 29 09:10:30 Moxa systemd[1]: sentinel.service: Failed with result 'e xit-code'.
...
Sometimes, it will show Cannot send hci command to socket: Connection timed out instead of initialization timeout - Finally, Bluetooth resource is locked, unless we set BT_EN to low and high state.
However, after updating the service pack to 4.8, we no longer encountered the issue.
The changelog WL18XX-BT-SP from 4.4 to 4.8 does not indicate that it fixes any issues regarding connection/initialization timeout.
(We have found that when loading the .bts script, there seems to be a 0.5 second delay)
If there are any changes about this upgrade (4.4 to 4.8) that will affect the stability, execution sequence or speed?
They want to know the root cause of this issue.
Host CPU: AM335X
Linux kernel: linux-4.4-cip-rt
The changelog WL18XX-BT-SP from 4.4 to 4.8 does not indicate that it fixes any issues regarding connection/initialization timeout.
(We have found that when loading the .bts script, there seems to be a 0.5 second delay)
If there are any changes about this upgrade (4.4 to 4.8) that will affect the stability, execution sequence or speed?
They want to know the root cause of this issue.
Host CPU: AM335X
Linux kernel: linux-4.4-cip-rt
Bluez: 5.43-2