No heap space after activating eDRX

[ralphvn]

When I add CONFIG_LTE_EDRX_REQ=y to prj.conf I get an "No heap space for incoming notification" error. I already tried to double the size of CONFIG_HEAP_MEM_POOL_SIZE=8192 but without any result. The device keeps rebooting.

[00:00:00.253,265] <inf> actinius_board_control: External SIM is selected
[00:00:00.253,295] <inf> actinius_board_control: Charger is set to auto
*** Booting nRF Connect SDK v3.5.99-ncs1 ***
[00:00:00.253,631] <inf> main: Comsat firmware started, version: 3.0.0
[00:00:00.498,291] <inf> main: Bringing network interface up and connecting to the network
[00:00:00.537,445] <inf> main: Hardware ID/IMEI: 351901936128218
[00:00:00.548,309] <inf> main: GNSS started
[00:00:00.548,339] <inf> main: Main loop finished, everything initialized
[00:00:00.574,401] <inf> main: Number of current satellites: 0
[00:00:01.575,286] <inf> main: Number of current satellites: 0
[00:00:02.871,429] <inf> main: Number of current satellites: 0
[00:00:02.871,459] <inf> main: GNSS blocked by LTE activity
[00:00:03.602,478] <wrn> at_monitor: No heap space for incoming notification: %XTIME: "80","42808011054080","01"

ASSERTION FAIL [at_notif] @ WEST_TOPDIR/nrf/lib/at_monitor/at_monitor.c:79
        No heap space for incoming notification: %XTIME: "80","42808011054080","01"

[00:00:03.602,569] <err> os: r0/a1:  0x00000004  r1/a2:  0x0000004f  r2/a3:  0x00000005
[00:00:03.602,569] <err> os: r3/a4:  0x2001b8c0 r12/ip:  0x2001bb6d r14/lr:  0x0002d0e7
[00:00:03.602,600] <err> os:  xpsr:  0x4100003a
[00:00:03.602,600] <err> os: s[ 0]:  0x0002d1c1  s[ 1]:  0x2000a728  s[ 2]:  0x2000a5f4  s[ 3]:  0x2000a5ec
[00:00:03.602,630] <err> os: s[ 4]:  0x00000002  s[ 5]:  0x2001f93a  s[ 6]:  0x00000000  s[ 7]:  0x0003707d
[00:00:03.602,661] <err> os: s[ 8]:  0x0003700d  s[ 9]:  0x00000001  s[10]:  0x2000a5f4  s[11]:  0x0003669d
[00:00:03.602,691] <err> os: s[12]:  0x2001f4a4  s[13]:  0x2001f420  s[14]:  0x00000001  s[15]:  0x00000000
[00:00:03.602,691] <err> os: fpscr:  0x00000000
[00:00:03.602,691] <err> os: Faulting instruction address (r15/pc): 0x00044234
[00:00:03.602,752] <err> os: >>> ZEPHYR FATAL ERROR 4: Kernel panic on CPU 0
[00:00:03.602,752] <err> os: Fault during interrupt handling

[00:00:03.602,783] <err> os: Current thread: 0x2001e0a8 (unknown)
[00:00:03.736,663] <err> fatal_error: Resetting system

[Håkon Alseth]

Hi,

 

The AT monitor has its own dedicated heap. Could you try increasing "CONFIG_AT_MONITOR_HEAP_SIZE" and see if helps?

 

Kind regards,

Håkon

[ralphvn]

Thanks, add CONFIG_AT_MONITOR_HEAP_SIZE=4096 to prj.conf solved the issue.

Where can I find more information about this, it was 256 and now 4096 but why would I not make it 8192 or 1024?

[Håkon Alseth]

Hi,

 

4096 is quite high, you will probably be OK with a lower value. 

Please note that this is a dedicated heap for the at_monitor library, and is not a part of the generic system heap (which is configured with CONFIG_HEAP_MEM_POOL_SIZE).

You can read more about the at_monitor here: https://docs.nordicsemi.com/bundle/ncs-latest/page/nrf/libraries/modem/at_monitor.html

 

Kind regards,

Håkon