[PATCH v7 0/2] Implement pahole-like 'ptype /o' option (and do some code reorg)

Sergio Durigan Junior sergiodj@redhat.com
Fri Dec 15 01:12:00 GMT 2017


Changes from v6:

- Updated examples in the documentation.

- Removed stale comments about the test being x86_64-specific from
  tests.

- Added tests for "ptype /oTM", "ptype /TMo", and for the regression
  Pedro caught.

- Removed "print_offset_default_data".

- Fixed regression reported by Pedro.

- s/endpos/end_bitpos/

- Filtered out the languages that don't implement the feature so that
  their type printing output doesn't get clobbered by "ptype /o" when
  they use "common" code (which is not actually common; it lives
  inside c-typeprint.c).  I decided to do that by checking the current
  language on typeprint.c:whatis_exp, before setting the
  "print_offset" bit.


The documentation parts have been approved by Eli, and haven't
changed.



More information about the Gdb-patches mailing list