[nuttx] Hardfault in SAM4E-EK icmpv6_radvertise()

[SP [e-mail] [nuttx]]

At I have a SAM4E-EK board running with IPv6. Been pinging back and forth and it
seems to work fine except that at random points, NuttX panics. The offending
function is:

https://bitbucket.org/nuttx/nuttx/src/ab96e3fdc72846515a746b87f6efa47faaf06c3c/net/icmpv6/icmpv6_advertise.c?at=master&fileviewer=file-view-default#icmpv6_advertise.c-96

I noticed there are changes to that function in newer revisions but not sure if
they address this issue. If anyone knows, let me know.

The dump:

```
up_hardfault: PANIC!!! Hard fault: 40000000
up_assert: Assertion failed at file:armv7-m/up_hardfault.c line: 171
up_dumpstate: sp: 200009d0
up_dumpstate: IRQ stack:
up_dumpstate: base: 20000a00
up_dumpstate: size: 00000800
up_stackdump: 200009c0: 200009d0 200009d0 00400c65 00420b2c 00000000 00400c6d 00400c2d 200037c4
up_stackdump: 200009e0: 00000003 004005ad 00000000 20007848 20001230 00000008 20001242 004002b1
up_dumpstate: sp: 20007890
up_dumpstate: User stack:
up_dumpstate: base: 20007980
up_dumpstate: size: 000007e4
up_stackdump: 20007880: 00000001 00000086 0041221e 01000000 20001230 200012ce 20003ee4 20001230
up_stackdump: 200078a0: 200012c0 00000080 00000080 00411d05 00000001 004024e1 20006d20 fffffffc
up_stackdump: 200078c0: 20001088 004029c7 20006d20 01000000 20007928 00000000 0000003c 20003ee4
up_stackdump: 200078e0: 20001230 200012ce 20001230 20001340 00000080 00000080 200066a0 00411229
up_stackdump: 20007900: 0000dd86 200011f4 00000000 0040491b 00000000 20001200 00000000 00000005
up_stackdump: 20007920: 20003890 00000005 20003894 20003a98 00032343 00401d8f 00401d6e 00000000
up_stackdump: 20007940: 00000000 20003890 20003890 00000000 00000000 00000000 00000000 00000000
up_stackdump: 20007960: 00000000 00401581 0040156d 00401543 00000000 00000000 00000000 00000000
up_registerdump: R0: 08078040 20001242 00000006 00000001 00000018 20001230 00000008 20001242
up_registerdump: R8: 200012f6 00000006 00000000 200066a0 00000001 20007890 00000086 0041221e
up_registerdump: xPSR: 01000000 PRIMASK: 00000000 CONTROL: 00000000
```

[[e-mail] [nuttx]]

I am not aware of anyone ever reporting hardfaults there before. I suspect it is some other, unrelated issue, probably related to some memory corruption or stack overflow.


Posting hardfault stackdumps is not a useful thing to do because those numbers mean nothing to anyone but you. You have to have the ELF file. That is the secret decoder ring.