From: Dodji Seketeli <dodji@redhat.com>
To: libabigail@sourceware.org
Subject: [PATCH 0/6] Fix subtle ABI artifact representation issues
Date: Fri, 27 Nov 2020 17:56:03 +0100 [thread overview]
Message-ID: <87h7pa7n8c.fsf@redhat.com> (raw)
Hello,
The goal of this patch set is to fix a class of annoying issues that
have been latent for a while now.
Basically, when comparing the ABI of a binary against its own ABIXML
representation, it is not uncommon to see 'harmless' ABI change
reports being reported. This should not happen. In some of the
reports the analyzer can't even describe what the alleged ABI changes
are. These are all symptoms of subtle wrong representation or
serialization of ABI artifacts.
To work around those issues, the 'abidw --abidiff' and 'abipkgdiff
--self-check' commands were made to avoid emitting an error when
'harmless' ABI changes were detected when comparing the ABI of a
binary to its ABIXML representation.
These series of patches address a number of those wrong representation
issues and thus make 'abidw --abidiff' and 'abipkgdiff --self-check'
emit an error for all changes detected when comparing the ABI of a
binary against its own ABIXML representation.
The first four patches fix four classes of these issues and the last two
patches make 'abidw --abidiff' and 'abipkgdiff --self-check' emit
error upon any ABI change.
Dodji Seketeli (6):
writer: Emit definitions of declarations when they are present
ir: Introduce internal pretty representation for anonymous classes
reader: Don't lose anonymous-ness of decl-only classes
dwarf-reader: Avoid having several functions with the same symbol
abidw: make --abidiff report any change against own ABIXML
abipkgdiff: make --self-check to fail on any change against own ABIXML
include/abg-fwd.h | 3 +
src/abg-default-reporter.cc | 2 +
src/abg-dwarf-reader.cc | 53 +-
src/abg-ir.cc | 40 +-
src/abg-reader.cc | 2 +
src/abg-writer.cc | 16 +-
tests/data/test-annotate/libtest23.so.abi | 140 +-
.../data/test-annotate/test13-pr18894.so.abi | 1662 ++--
.../data/test-annotate/test15-pr18892.so.abi | 2277 +++---
.../data/test-annotate/test21-pr19092.so.abi | 2829 +++----
.../test-read-dwarf/PR22122-libftdc.so.abi | 5747 ++++++-------
tests/data/test-read-dwarf/libtest23.so.abi | 138 +-
.../test-read-dwarf/test-libandroid.so.abi | 196 +-
.../test-read-dwarf/test10-pr18818-gcc.so.abi | 236 +-
.../test-read-dwarf/test11-pr18828.so.abi | 4550 +++++------
.../test-read-dwarf/test12-pr18844.so.abi | 7086 ++++++++---------
.../test-read-dwarf/test13-pr18894.so.abi | 1511 ++--
.../test-read-dwarf/test15-pr18892.so.abi | 2269 +++---
.../test-read-dwarf/test16-pr18904.so.abi | 48 +-
.../test-read-dwarf/test21-pr19092.so.abi | 2782 +++----
.../test22-pr19097-libstdc++.so.6.0.17.so.abi | 140 +-
.../test9-pr18818-clang.so.abi | 340 +-
tools/abidw.cc | 2 +-
tools/abipkgdiff.cc | 2 +-
24 files changed, 16697 insertions(+), 15374 deletions(-)
--
2.29.2
--
Dodji
next reply other threads:[~2020-11-27 16:56 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-27 16:56 Dodji Seketeli [this message]
2020-11-27 17:03 ` [PATCH 1/6] writer: Emit definitions of declarations when they are present Dodji Seketeli
2020-11-30 13:49 ` Giuliano Procida
2020-11-30 15:56 ` Dodji Seketeli
2020-11-30 18:15 ` Giuliano Procida
2020-12-01 9:51 ` Dodji Seketeli
2020-11-27 17:05 ` [PATCH 2/6] ir: Introduce internal pretty representation for anonymous classes Dodji Seketeli
2020-11-27 17:06 ` [PATCH 3/6] reader: Don't lose anonymous-ness of decl-only classes Dodji Seketeli
2020-11-27 17:07 ` [PATCH 4/6] dwarf-reader: Avoid having several functions with the same symbol Dodji Seketeli
2020-11-27 17:08 ` [PATCH 5/6] abidw: make --abidiff report any change against own ABIXML Dodji Seketeli
2020-11-27 17:08 ` [PATCH 6/6] abipkgdiff: make --self-check to fail on " Dodji Seketeli
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=87h7pa7n8c.fsf@redhat.com \
--to=dodji@redhat.com \
--cc=libabigail@sourceware.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).