On Wed, Nov 27, 2024 at 6:48 PM Aaron Puchert <aaronpuchert@alice-dsl.net> wrote:
[...]
Let's put the near term aside for now. What happens in the long term? When all distros have FP enabled, are we still going to see effort poured into out-of-band unwinding? Or is it just going to be written off?
As I've written in reply to Andrii, in the compiler world a 2% regression is quite significant. So I'd like to see a way out.
There is a proposal, called SFrame (see [0]), in the works that is meant to make frame pointers unnecessary. It requires support at all levels of software stack, from compiler, through linker, to kernel itself, BPF support, *and* BPF-based tooling to change the way they capture stack traces. It will take some time, but once all those pieces are ready and widely adopted, then frame pointers might not be necessary anymore. [0] https://sourceware.org/binutils/wiki/sframe [...]