Expand the list of got_audit expected duplicate symbols.

[gordonmessmer]

Description/Motivation/Screenshots

Based on normal audits of Fedora binaries, the set of expected duplicate symbols has been expanded.

In order to reduce false positives in GotAudit, the build_line function requires the path to the file whose symbols are being described. This change will also not print an ERROR condition if a symbol resolves locally, within the section that requires it.

These changes require hugsy/gef/pull/1172

How Has This Been Tested ?

"Tested" indicates that the PR works and the unit test (i.e. make test) run passes without issue.

• x86-32
• x86-64
• ARM
• AARCH64
• MIPS
• POWERPC
• SPARC
• RISC-V

Checklist

• My code follows the code style of this project.
• My change includes a change to the documentation, if required.
• If my change adds new code,
  adequate tests have been added.
• I have read and agree to the
  CONTRIBUTING document.

[hugsy]

The code looks good, but can you update the got-audit test to test those changes?

[gordonmessmer]

That's definitely an interesting question. :)

I think that updating the test would involve adding a new binary that uses at least one of the duplicate symbols, and links to a library that duplicates them. But that starts to get very platform-specific, and it's difficult to determine in testing whether a duplicate symbol is correctly allowed, or if there is simply no symbol duplication.

That might be easier if the list of symbols were loaded from a configuration file or something along those lines. (I really don't think hard-coding the set is a good long-term solution.) But I'm not sure what would be idomatic for this project, so if you have any suggestions on how you'd like to see the allowed set of symbols specified, I'd appreciate the feedback.

[hugsy]

I think that updating the test would involve adding a new binary that uses at least one of the duplicate symbols, and links to a library that duplicates them. But that starts to get very platform-specific, and it's difficult to determine in testing whether a duplicate symbol is correctly allowed, or if there is simply no symbol duplication.

Fair, I didn't think of that. That might be a lot of effort for little gain, so for now disregard my previous comment.

That might be easier if the list of symbols were loaded from a configuration file or something along those lines

True, but let's move there only if/when there's a need for it.

Thanks for the PR(s) @gordonmessmer