public inbox for libabigail@sourceware.org
 help / color / mirror / Atom feed
From: "szhorvat at gmail dot com" <sourceware-bugzilla@sourceware.org>
To: libabigail@sourceware.org
Subject: [Bug default/27700] New: Assertion failure when using --drop-private-types
Date: Tue, 06 Apr 2021 11:00:02 +0000	[thread overview]
Message-ID: <bug-27700-9487@http.sourceware.org/bugzilla/> (raw)

https://sourceware.org/bugzilla/show_bug.cgi?id=27700

            Bug ID: 27700
           Summary: Assertion failure when using --drop-private-types
           Product: libabigail
           Version: unspecified
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: default
          Assignee: dodji at redhat dot com
          Reporter: szhorvat at gmail dot com
                CC: libabigail at sourceware dot org
  Target Milestone: ---

Created attachment 13351
  --> https://sourceware.org/bugzilla/attachment.cgi?id=13351&action=edit
Offending shared library

I see an assertion failure when using abidw with the --drop-private-types flag:

abidw: ../../src/abg-ir.cc:23638: size_t
abigail::ir::hash_as_canonical_type_or_constant(const abigail::ir::type_base*):
Assertion `__abg_cond__' failed.

This is with the current master branch from the git repo (as of today). With
version 1.8, I see a segmentation fault instead.

I am working on Ubuntu 20.04 x86_64, and using abidw on the igraph library
(version 0.9.1). https://github.com/igraph/igraph/  While I realize that the
following is far from a *minimal* example, here is how to reproduce the
problem:

Compile igraph:

git clone https://github.com/igraph/igraph.git
cd igraph && mkdir build && cd build
cmake .. -DBUILD_SHARED_LIBS=ON -DCMAKE_BUILD_TYPE=Debug
make

Run abidw:

touch foo.h
abidw --hf foo.h --drop-private-types src/libigraph.so > /dev/null

abidw: ../../src/abg-ir.cc:23638: size_t
abigail::ir::hash_as_canonical_type_or_constant(const abigail::ir::type_base*):
Assertion `__abg_cond__' failed.
Aborted (core dumped)

foo.h is a dummy empty header. The --hf or --hd options are necessary to
trigger the problem. Originally, I used --hd to specify the location of the
igraph headers, however, it seems to be sufficient to provide an empty header.
This is what I did in the example above.

I am attaching the offending shared library.

P.S. The igraph build system auto-detects some external dependencies, so the
above method to build igraph produces system-dependent results. However, it
does not seem to matter which external libraries I link to—it crashes all the
same.

-- 
You are receiving this mail because:
You are on the CC list for the bug.

             reply	other threads:[~2021-04-06 11:00 UTC|newest]

Thread overview: 3+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-04-06 11:00 szhorvat at gmail dot com [this message]
2021-04-07 17:46 ` [Bug default/27700] " dodji at redhat dot com
2021-04-12 15:56 ` dodji at redhat dot com

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=bug-27700-9487@http.sourceware.org/bugzilla/ \
    --to=sourceware-bugzilla@sourceware.org \
    --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).