Tobias Waldekranz

Short answer: No That said, if you want to add support for it, it would probably be quite easy. All you need to do is to create the missing file...

Ouch! Well in that case it is actually a bit more work. The some of the code in `kprobe_arg_rewrite` will have to be moved to the arch-specific code. Then the...

To answer your initial question: Yes, I see no reason why that should not work. My guess is that the script files are failing because they rely on the names...

I'm working on a new version of ply which has a better compiler foundation with a proper type system and so on (ply was my first compiler project, ever :smile:)....

I've been allocating my spare time to other projects lately, so it is not there yet. uprobes are definitely on the roadmap. Basic support is easy to add, i.e. specifying...

Sorry for the long delay on this. Could you try two changes to your script: 1. Change `str()` to `mem()`. This will force a copy of the memory pointed to...

Not really sure what to make of the first issue. I do not have easy access to a 4.19/`aarch64` system at the moment. What is the output of `ply -T`...

The oldest kernel I can remember testing 2.x with is 4.19. It is not unlikely that ply is simply not compatible with older versions. Any chance you can test with...

Long-term I think we can: - Drop the per-product kernel support in the current layout (which we've never used anyway) - Move to a flat layout under `/boot/westermo/` So we...

Update: We should consider generalizing this so that users can inject files at any path. Then a user can drop in extra config snippets to deamons with `daemon.d/` support. E.g.,...