[PATCH] D63327: [docs][llvm-nm] Improve symbol code documentation

[James Henderson via Phabricator via llvm-commits]

jhenderson marked 2 inline comments as done.
jhenderson added inline comments.


================
Comment at: docs/CommandGuide/llvm-nm.rst:51
+
+n
+
----------------
jhenderson wrote:
> MaskRay wrote:
> > I think the semantics (GNU nm behavior) of `n` are similar to what the reverted rL359312 did, but unfortunately I cannot find a reasonable use case to write a test, nor can I find useful tests in the binutils-gdb repository.
> > 
> > I don't know if relocations into non-SHF_ALLOC sections other than .debug* are reasonable.
> > I don't know if relocations into non-SHF_ALLOC sections other than .debug* are reasonable.
> 
> We use them in Sony-specific sections at least. I think they are reasonable for additional metadata that needs to refer to the actual code in some way.
> 
> I'm okay with the semantics of 'n' changing in the future (this is just one example where the code doesn't necessarily make sense). I'd just like to document what we currently do now - we can document what the behaviour is when we change the code.
> We use them in Sony-specific sections at least. I think they are reasonable for additional metadata that needs to refer to the actual code in some way.

I misread your comment. I don't know about relocations into non-alloc sections, but it's certainly possible to produce local symbols in them at least.


Repository:
  rL LLVM

CHANGES SINCE LAST ACTION
  https://reviews.llvm.org/D63327/new/

https://reviews.llvm.org/D63327