Hardfault logging

RE: Hardfault logging

Luke_123 — Tue, 13 Jun 2023 21:01:46 GMT

Thank you

RE: Hardfault logging

Kenneth — Tue, 13 Jun 2023 12:23:39 GMT

[quote user="Luke_123"]Did you mean 'NVIC_SystemReset' by system reset ?[/quote]

Yes.

[quote user="Luke_123"]Is there any reason you advised to store the RAM buffer to flash 'after' a reset ? Cannot we store the RAM buffer to flash inside the fault handler before resetting, Is there a risk doing so ?[/quote]

Part of the reason is simply that you don't really know what caused the assert to trigger, so you can't really depend on software modules will be able to write successfully to flash in such case, unless you implement bare-bone hardware usage of the NVMC to write to flash within the fault handler.

Kenneth

RE: Hardfault logging

Luke_123 — Tue, 13 Jun 2023 11:34:22 GMT

Thank you so much Kenneth for your reply.

Did you mean 'NVIC_SystemReset' by system reset ?

Is there any reason you advised to store the RAM buffer to flash 'after' a reset ? Cannot we store the RAM buffer to flash inside the fault handler before resetting, Is there a risk doing so ?

RE: Hardfault logging

Kenneth — Tue, 13 Jun 2023 11:05:01 GMT

Hello,

You can find some pointers here:
Error Handling for Soft Device and Application

So the general idea is that you store the error information to a RAM buffer when there is an assert, then execute system reset, and then after reset you can either store the RAM buffer to flash (e.g. using FDS or NVMC directly if you have a designated flash page available) or send it through some BLE link to the peer. External flash is not needed.

Kenneth