Re: [OE-core] [PATCH v2] lib/oe/package: Add strip keep-section support

[Mathieu Othacehe <othacehe@gnu.org>]

Hello Richard,

> I think the name has been misleading me as I imagined minidebuginfo as
> a small subset of the main debug information rather than just the
> symbols. You're saying it is just the symbols and there is no way to
> extend the compressed information?

The minidebuginfo symbols are stored in:

--8<---------------cut here---------------start------------->8---
  subprocess.check_call([objcopy, '--add-section', '.gnu_debugdata={}.xz'.format(minidebugfile), file])
--8<---------------cut here---------------end--------------->8---

the .gnu_debugdata ELF section. That's a convention and some tools out
there (GDB, libunwind) will search for symbols in that specific section.

> OE is about choice but I'm still thinking that if someone enables
> minidebuginfo, they are most likely expecting stack traces to work as
> otherwise the extra symbol information isn't much use. Am is missing
> some key usage?

Another usage would be to debug a live program on target with GDB. You
need the symbols but not necessarily the unwind information to do that I
guess.

> If PACKAGE_KEEP_SECTIONS worked well for all targets, I think I'd be
> happier to consider it but the problem is that it is all very
> architecture dependent :(

Would it be OK if I send a v3 that re-introduces
PACKAGE_KEEP_DEBUG_FRAME, and enables that one if minidebuginfo is
enabled and the target architecture is ARM?

Thanks,

Mathieu