Precise bus fault in memq_enqueue

[mablabs]

Hi,

We are running Zephyr 2.4 via PlatformIO on a nRF52840 based product. While spamming the nRF with a ton of BLE packets, after about 15 minutes or so we get the following bus fault:

00> [00:13:28.106,445] <err> os: ***** BUS FAULT *****
00> [00:13:28.106,475] <err> os:  Precise data bus error
00> [00:13:28.106,475] <err> os:  BFAR Address: 0x20aa81ae
00> [00:13:28.106,506] <err> os: r0/a1: 0x2000c13c r1/a2: 0x2000cc50 r2/a3: 0x2000d3e8
00> [00:13:28.106,506] <err> os: r3/a4: 0x200009b0 r12/ip: 0x00000000 r14/lr: 0x0002208b
00> [00:13:28.106,506] <err> os: xpsr: 0x21000211
00> [00:13:28.106,536] <err> os: s[ 0]: 0x00000000 s[ 1]: 0x00000000 s[ 2]: 0x00000000 s[ 3]: 0x00000000
00> [00:13:28.106,536] <err> os: s[ 4]: 0x00000000 s[ 5]: 0x00000000 s[ 6]: 0x00000000 s[ 7]: 0x00000000
00> [00:13:28.106,536] <err> os: s[ 8]: 0x00000000 s[ 9]: 0x00000000 s[10]: 0x00000000 s[11]: 0x00000000
00> [00:13:28.106,567] <err> os: s[12]: 0x00000000 s[13]: 0x00000000 s[14]: 0x00000000 s[15]: 0x00000000
00> [00:13:28.106,567] <err> os: fpscr: 0x2000f3b8
00> [00:13:28.106,567] <err> os: Faulting instruction address (r15/pc): 0x0003f34a
00> [00:13:28.106,597] <err> os: >>> ZEPHYR FATAL ERROR 0: CPU exception on CPU 0
00> [00:13:28.106,597] <err> os: Fault during interrupt handling
00>
00> [00:13:28.106,597] <err> os: Current thread: 0x20005510 (unknown)
00> [00:13:28.678,009] <err> os: Halting system

Can you advise on what we could do to overcome this issue?

Thank you,

[Hakon]

Hello,

can you check the failing instruction with addr2line as explained by Alseth in this post?

[mablabs]

Hi Hakon,

Thanks for getting back to me. This is the result of running addr2line:

C:\Users\mab\.platformio\packages\framework-zephyr\subsys\bluetooth\controller\util/memq.c:92

which points to this line:

memq_link_t *memq_enqueue(memq_link_t *link, void *mem, memq_link_t **tail)
{
 /* Let the old tail element point to the new tail element */
 (*tail)->next = link;

Please advise on what steps we can take to debug this issue further.