public inbox for libabigail@sourceware.org
 help / color / mirror / Atom feed
* [Bug default/29650] New: "Assertion `__abg_cond__' failed." with  with libabigail 2.1
@ 2022-10-03 19:14 romain.geissler at amadeus dot com
  2022-10-03 19:22 ` [Bug default/29650] " romain.geissler at amadeus dot com
                   ` (11 more replies)
  0 siblings, 12 replies; 14+ messages in thread
From: romain.geissler at amadeus dot com @ 2022-10-03 19:14 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

            Bug ID: 29650
           Summary: "Assertion `__abg_cond__' failed." with  with
                    libabigail 2.1
           Product: libabigail
           Version: unspecified
            Status: UNCONFIRMED
          Severity: normal
          Priority: P2
         Component: default
          Assignee: dodji at redhat dot com
          Reporter: romain.geissler at amadeus dot com
                CC: libabigail at sourceware dot org
  Target Milestone: ---

Hi,

I am having issues with libabigail 2.1 we didn't have with the previous
version. We get in quite some places errors like:

abidiff: /workdir/src/libabigail-2.1/src/abg-default-reporter.cc:375: virtual
void abigail::comparison::default_reporter::report(const
abigail::comparison::qualified_type_diff&, std::ostream&, const string&) const:
Assertion `__abg_cond__' failed.


I will try to attach (if it's not too big) an reproducer with the xalan open
source library 1.11 vs 1.12, but we also have similar failures with proprietary
software I can't really upload here. Just untar the tarball into a directory
and run:

abidiff --no-unreferenced-symbols libxalan-c.so.111.0 libxalan-c.so.112.0

Cheers,
Romain

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
@ 2022-10-03 19:22 ` romain.geissler at amadeus dot com
  2022-10-03 19:23 ` romain.geissler at amadeus dot com
                   ` (10 subsequent siblings)
  11 siblings, 0 replies; 14+ messages in thread
From: romain.geissler at amadeus dot com @ 2022-10-03 19:22 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

--- Comment #1 from Romain Geissler <romain.geissler at amadeus dot com> ---
Created attachment 14375
  --> https://sourceware.org/bugzilla/attachment.cgi?id=14375&action=edit
abidiff.tar.gz part 1

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
  2022-10-03 19:22 ` [Bug default/29650] " romain.geissler at amadeus dot com
@ 2022-10-03 19:23 ` romain.geissler at amadeus dot com
  2022-10-03 19:24 ` romain.geissler at amadeus dot com
                   ` (9 subsequent siblings)
  11 siblings, 0 replies; 14+ messages in thread
From: romain.geissler at amadeus dot com @ 2022-10-03 19:23 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

--- Comment #2 from Romain Geissler <romain.geissler at amadeus dot com> ---
Created attachment 14376
  --> https://sourceware.org/bugzilla/attachment.cgi?id=14376&action=edit
abidiff.tar.gz part 2

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
  2022-10-03 19:22 ` [Bug default/29650] " romain.geissler at amadeus dot com
  2022-10-03 19:23 ` romain.geissler at amadeus dot com
@ 2022-10-03 19:24 ` romain.geissler at amadeus dot com
  2022-10-03 19:25 ` romain.geissler at amadeus dot com
                   ` (8 subsequent siblings)
  11 siblings, 0 replies; 14+ messages in thread
From: romain.geissler at amadeus dot com @ 2022-10-03 19:24 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

--- Comment #3 from Romain Geissler <romain.geissler at amadeus dot com> ---
Created attachment 14377
  --> https://sourceware.org/bugzilla/attachment.cgi?id=14377&action=edit
abidiff.tar.gz part 3

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
                   ` (2 preceding siblings ...)
  2022-10-03 19:24 ` romain.geissler at amadeus dot com
@ 2022-10-03 19:25 ` romain.geissler at amadeus dot com
  2022-10-07 20:30 ` dodji at redhat dot com
                   ` (7 subsequent siblings)
  11 siblings, 0 replies; 14+ messages in thread
From: romain.geissler at amadeus dot com @ 2022-10-03 19:25 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

--- Comment #4 from Romain Geissler <romain.geissler at amadeus dot com> ---
The above tar.gz was splitted in 3 using this command: split -b 9800k
abidiff.tar.gz abidiff.tar.gz.part.

So you just need to cat the 3 parts into a single file to have the full file.

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
                   ` (3 preceding siblings ...)
  2022-10-03 19:25 ` romain.geissler at amadeus dot com
@ 2022-10-07 20:30 ` dodji at redhat dot com
  2022-10-10 15:19 ` dodji at redhat dot com
                   ` (6 subsequent siblings)
  11 siblings, 0 replies; 14+ messages in thread
From: dodji at redhat dot com @ 2022-10-07 20:30 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

dodji at redhat dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
     Ever confirmed|0                           |1
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2022-10-07

--- Comment #5 from dodji at redhat dot com ---
I have been able to reproduce the issue, thank you for providing the
reproducer!

I am looking into this.

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
                   ` (4 preceding siblings ...)
  2022-10-07 20:30 ` dodji at redhat dot com
@ 2022-10-10 15:19 ` dodji at redhat dot com
  2022-10-10 15:20 ` dodji at redhat dot com
                   ` (5 subsequent siblings)
  11 siblings, 0 replies; 14+ messages in thread
From: dodji at redhat dot com @ 2022-10-10 15:19 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

--- Comment #6 from dodji at redhat dot com ---
This should now be fixed in the master branch of the libabigail project by the
commit
https://sourceware.org/git/?p=libabigail.git;a=commit;h=0d49c09e9cb4a5f45a697013a6e54fa708714cdc.

The fix should be available in the coming libabigail 2.2 version.

Thanks for filling this issue, and sorry for the inconvenience.

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
                   ` (5 preceding siblings ...)
  2022-10-10 15:19 ` dodji at redhat dot com
@ 2022-10-10 15:20 ` dodji at redhat dot com
  2022-10-10 15:24 ` romain.geissler at amadeus dot com
                   ` (4 subsequent siblings)
  11 siblings, 0 replies; 14+ messages in thread
From: dodji at redhat dot com @ 2022-10-10 15:20 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

dodji at redhat dot com changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|ASSIGNED                    |RESOLVED

--- Comment #7 from dodji at redhat dot com ---
closing the bug as fixed.

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
                   ` (6 preceding siblings ...)
  2022-10-10 15:20 ` dodji at redhat dot com
@ 2022-10-10 15:24 ` romain.geissler at amadeus dot com
  2022-10-10 16:22 ` dodji at redhat dot com
                   ` (3 subsequent siblings)
  11 siblings, 0 replies; 14+ messages in thread
From: romain.geissler at amadeus dot com @ 2022-10-10 15:24 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

--- Comment #8 from Romain Geissler <romain.geissler at amadeus dot com> ---
Thanks !

When I upgraded to abigail 2.1 I had actually not just one project hitting
this, but quite some among all the ones I maintain. Maybe it's not the exact
same issue, but in all case it was an assertion failure (the root cause of the
failure may be different for each).

Do you plan on cutting a "bugfix" release in the coming days or this is not
critical and it can way for a bigger release 2.2 in some months ? If you plan
for a release soon I will wait for it, otherwise I will migrate again to 2.1
with a cherry-pick of your patch for this bug.

Romain

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
                   ` (7 preceding siblings ...)
  2022-10-10 15:24 ` romain.geissler at amadeus dot com
@ 2022-10-10 16:22 ` dodji at redhat dot com
  2022-10-10 18:36 ` romain.geissler at amadeus dot com
                   ` (2 subsequent siblings)
  11 siblings, 0 replies; 14+ messages in thread
From: dodji at redhat dot com @ 2022-10-10 16:22 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

--- Comment #9 from dodji at redhat dot com ---
"romain.geissler at amadeus dot com"
<sourceware-bugzilla@sourceware.org> writes:

> https://sourceware.org/bugzilla/show_bug.cgi?id=29650
>
> --- Comment #8 from Romain Geissler <romain.geissler at amadeus dot com> ---
> Thanks !
>
> When I upgraded to abigail 2.1 I had actually not just one project hitting
> this, but quite some among all the ones I maintain. Maybe it's not the exact
> same issue, but in all case it was an assertion failure (the root cause of the
> failure may be different for each).

If they are all C++ libraries, then maybe it's the same bug (if the
assert is at the same location).  Sorry for that.

> Do you plan on cutting a "bugfix" release in the coming days or this is not
> critical and it can way for a bigger release 2.2 in some months ?

I don't have a planned release really soon, but I might cherry-pick the
fix into the Fedora and EPEL packages of libabigail if that can be
useful, leading to a libabigail-2.1-2 package.  Would that be helpful?

> If you plan for a release soon I will wait for it, otherwise I will
> migrate again to 2.1 with a cherry-pick of your patch for this bug.

Ah, if you can cherry-pick the fix into your 2.1, then it might be best
to do that until we release 2.2, which I hope won't be too far down the
road.  I definitely wouldn't want to wait as long what we did for 2.1.

Please let me know what you'd prefer.

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
                   ` (8 preceding siblings ...)
  2022-10-10 16:22 ` dodji at redhat dot com
@ 2022-10-10 18:36 ` romain.geissler at amadeus dot com
  2022-10-11  9:49   ` Dodji Seketeli
  2022-10-10 21:23 ` woodard at redhat dot com
  2022-10-11  9:49 ` dodji at seketeli dot org
  11 siblings, 1 reply; 14+ messages in thread
From: romain.geissler at amadeus dot com @ 2022-10-10 18:36 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

--- Comment #10 from Romain Geissler <romain.geissler at amadeus dot com> ---
Yes we use almost only C++.

For abigail we don't use any official package, we rebuilt it from source, so I
will chose the cherry-pick path. I don't know if anyone else than us is
impacted by this problem (others might want the fedora package to have this fix
if they are impacted).

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
                   ` (9 preceding siblings ...)
  2022-10-10 18:36 ` romain.geissler at amadeus dot com
@ 2022-10-10 21:23 ` woodard at redhat dot com
  2022-10-11  9:49 ` dodji at seketeli dot org
  11 siblings, 0 replies; 14+ messages in thread
From: woodard at redhat dot com @ 2022-10-10 21:23 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

Ben Woodard <woodard at redhat dot com> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |woodard at redhat dot com

--- Comment #11 from Ben Woodard <woodard at redhat dot com> ---
Dodji can we consider a 2.1.1 fork beginning with this patch. This impacted us
here too since the majority of our apps and tools are C++. I'll run another
batch of large scale testing and we can use 2.1.1 to pick up any remaining
problems like this assert that we are seeing with the current 2.1.

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

* Re: [Bug default/29650] "Assertion `__abg_cond__' failed." with with libabigail 2.1
  2022-10-10 18:36 ` romain.geissler at amadeus dot com
@ 2022-10-11  9:49   ` Dodji Seketeli
  0 siblings, 0 replies; 14+ messages in thread
From: Dodji Seketeli @ 2022-10-11  9:49 UTC (permalink / raw)
  To: romain.geissler at amadeus dot com via Libabigail
  Cc: romain.geissler at amadeus dot com

Hello Ben,

woodard at redhat dot com via Libabigail <libabigail@sourceware.org> a
écrit:

> Dodji can we consider a 2.1.1 fork beginning with this patch.

There is always a tag for the latest released version.  The last one is
libabigail-2.1.

From that tag, if there the need for bugfix branch, I create a branch
which name is <tag>-branch, and I cherry-pick the bug fixes into that
branch.

So in this case, the branch name would be libabigail-2.1-branch.

I have just created it, with the relevant patches cherry-picked in it.
I'll be maintaining that branch for this cycle, until 2.2.  I try to
avoid doing this because it's a bit more work, but when it's necessary,
well, it's necessary.

So it's available at
https://sourceware.org/git/?p=libabigail.git;a=shortlog;h=refs/heads/libabigail-2.1-branch.

When I roll out libabigail-2.1-2 Fedora packages, they'll be cut from
that branch.

[...]

Hello Romain,

"romain.geissler at amadeus dot com via Libabigail"
<libabigail@sourceware.org> a écrit:

[...]

> For abigail we don't use any official package, we rebuilt it from source, so I
> will chose the cherry-pick path.

[...]

Ok, so if you like, you might want to get the content of the branch
'libabigail-2.1-branch', browsable at
https://sourceware.org/git/?p=libabigail.git;a=shortlog;h=refs/heads/libabigail-2.1-branch,
which contains all bugfixes that will be accumulated until 2.2 is released.

[...]

Cheers.

-- 
		Dodji

^ permalink raw reply	[flat|nested] 14+ messages in thread

* [Bug default/29650] "Assertion `__abg_cond__' failed." with  with libabigail 2.1
  2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
                   ` (10 preceding siblings ...)
  2022-10-10 21:23 ` woodard at redhat dot com
@ 2022-10-11  9:49 ` dodji at seketeli dot org
  11 siblings, 0 replies; 14+ messages in thread
From: dodji at seketeli dot org @ 2022-10-11  9:49 UTC (permalink / raw)
  To: libabigail

https://sourceware.org/bugzilla/show_bug.cgi?id=29650

--- Comment #12 from dodji at seketeli dot org ---
Hello Ben,

woodard at redhat dot com via Libabigail <libabigail@sourceware.org> a
écrit:

> Dodji can we consider a 2.1.1 fork beginning with this patch.

There is always a tag for the latest released version.  The last one is
libabigail-2.1.

From that tag, if there the need for bugfix branch, I create a branch
which name is <tag>-branch, and I cherry-pick the bug fixes into that
branch.

So in this case, the branch name would be libabigail-2.1-branch.

I have just created it, with the relevant patches cherry-picked in it.
I'll be maintaining that branch for this cycle, until 2.2.  I try to
avoid doing this because it's a bit more work, but when it's necessary,
well, it's necessary.

So it's available at
https://sourceware.org/git/?p=libabigail.git;a=shortlog;h=refs/heads/libabigail-2.1-branch.

When I roll out libabigail-2.1-2 Fedora packages, they'll be cut from
that branch.

[...]

Hello Romain,

"romain.geissler at amadeus dot com via Libabigail"
<libabigail@sourceware.org> a écrit:

[...]

> For abigail we don't use any official package, we rebuilt it from source, so I
> will chose the cherry-pick path.

[...]

Ok, so if you like, you might want to get the content of the branch
'libabigail-2.1-branch', browsable at
https://sourceware.org/git/?p=libabigail.git;a=shortlog;h=refs/heads/libabigail-2.1-branch,
which contains all bugfixes that will be accumulated until 2.2 is released.

[...]

Cheers.

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

^ permalink raw reply	[flat|nested] 14+ messages in thread

end of thread, other threads:[~2022-10-11  9:49 UTC | newest]

Thread overview: 14+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-10-03 19:14 [Bug default/29650] New: "Assertion `__abg_cond__' failed." with with libabigail 2.1 romain.geissler at amadeus dot com
2022-10-03 19:22 ` [Bug default/29650] " romain.geissler at amadeus dot com
2022-10-03 19:23 ` romain.geissler at amadeus dot com
2022-10-03 19:24 ` romain.geissler at amadeus dot com
2022-10-03 19:25 ` romain.geissler at amadeus dot com
2022-10-07 20:30 ` dodji at redhat dot com
2022-10-10 15:19 ` dodji at redhat dot com
2022-10-10 15:20 ` dodji at redhat dot com
2022-10-10 15:24 ` romain.geissler at amadeus dot com
2022-10-10 16:22 ` dodji at redhat dot com
2022-10-10 18:36 ` romain.geissler at amadeus dot com
2022-10-11  9:49   ` Dodji Seketeli
2022-10-10 21:23 ` woodard at redhat dot com
2022-10-11  9:49 ` dodji at seketeli dot org

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).