#583 ELF Tool Chain endianness issues

[Ed Maste]

While reviewing and upstreaming ELF Tool Chain changes from FreeBSD I discovered a number of endianness issues that are not yet addressed upstream.

• libelf: fix .note conversion for non-native byte order ELF files
  https://reviews.freebsd.org/rS273443

• elfcopy: copy raw (untranslated) contents to binary output
  https://reviews.freebsd.org/rS327489

• elfcopy: fix little-endian MIPS64 objects (Ticket #559)
  https://reviews.freebsd.org/D15734
  https://reviews.freebsd.org/rS344855
  https://reviews.freebsd.org/rS339083
  https://reviews.freebsd.org/rS339473

• just-committed readelf change [r3766] is not endian-aware

[Ed Maste]

It seems I had this unsubmitted in a forgotten browser tab, from shortly after [r3766] was committed (i.e., just after 2019-06-29). It's possible that there have been changes that affect these issues or introduce new ones since.

[Joseph Koshy]

libelf: fix .note conversion for non-native byte order ELF files
https://reviews.freebsd.org/rS273443

This change would perform arithmetic on the note header's fields while these are still in their file representation. I'm not sure that is correct.

A related issue here is that the libelf test suite is missing test cases for Elf note conversions---I did not get around to writing those. I've filed [#584] to keep track of this pending task.

[Ed Maste]

Yes, I pasted the wrong rev; FreeBSD r273443 correctly identified the issue but the commit was incorrect; I reverted it and applied the correct fix in https://reviews.freebsd.org/rS275430

[Ed Maste]

And this first issue is already fixed by [r3158]

[Ed Maste]

First two are now fixed (2nd by [r3832]). The 3rd group (MIPS) are also related to ticket [#559]. I need to figure out what the right way to handle [r3766] is.

[Ed Maste]

For the readelf the best approach is probably to teach libelf how to translate the ELF notes used by various operating sytems, @jkoshy do you agree?

[Joseph Koshy]

@emaste: yes, that seems the best approach.