Linux Coredumps (Part 2) - Shrinking the Core | Interrupt

In our previous article, we outlined what a Linux coredump is and how they work under the hood. One common constraint we see in embedded Linux, however, is a limited amount of storage space. Whether we’re trying to limit writes to disk, or need to reserve most of the disk space available to a device for other data, sometimes we just don’t have enough space to store coredumps.


This is a companion discussion topic for the original entry at https://interrupt.memfault.com/blog/linux-coredumps-part-2

Hi, thanks for this article!

In finding the stacks, you talk about finding the PC. But I would’ve expected to look for the SP to know which memory segment to capture in the trace. What am I missing?