Note: This is a public test instance of Red Hat Bugzilla. The data contained within is a snapshot of the live data so any changes you make will not be reflected in the production Bugzilla. Email is disabled so feel free to test any aspect of the site that you want. File any problems you find or give feedback at bugzilla.redhat.com.
Bug 1197577 - FTBFS: cutter-1.2.4-4 on F-23
Summary: FTBFS: cutter-1.2.4-4 on F-23
Keywords:
Status: CLOSED RAWHIDE
Alias: None
Product: Fedora
Classification: Fedora
Component: cutter
Version: 23
Hardware: Unspecified
OS: Unspecified
unspecified
unspecified
Target Milestone: ---
Assignee: Kentaro Hayashi
QA Contact: Fedora Extras Quality Assurance
URL: https://github.com/clear-code/cutter/...
Whiteboard:
Depends On: 1182957
Blocks: harden-failure
TreeView+ depends on / blocked
 
Reported: 2015-03-02 05:36 UTC by Mamoru TASAKA
Modified: 2016-02-16 07:33 UTC (History)
4 users (show)

Fixed In Version: cutter-1.2.5-1.fc24
Doc Type: Bug Fix
Doc Text:
Clone Of: 1182957
Environment:
Last Closed: 2016-02-16 07:33:46 UTC
Type: Bug
Embargoed:

Attachments (Terms of Use)

Description Mamoru TASAKA 2015-03-02 05:36:03 UTC 
+++ This bug was initially created as a clone of Bug #1182957 +++

Description of problem:
Currently cutter [cutter-1.2.4-2.fc21] FTBFS (Fail To Build From Source) on F-21 and above. Please fix this.

http://koji.fedoraproject.org/koji/buildinfo?buildID=555444


Additional info:
This fix is needed:
https://github.com/clear-code/cutter/issues/19

--- Additional comment from HAYASHI Kentaro on 2015-02-14 14:12:16 JST ---

Thank you for reporting!

It needs one more patch to fix "make check" failure.

Here is the commit to support newer GLib error message.
https://github.com/clear-code/cutter/commit/3947d292996458f3800e93021b86e1476f85922b

By this commit, it seems well for F-21, but F-22 still have problem.
It is a bit strange because mock build is ok, but koji --scrach build fails.

Crash: test_limit_block (default context)
https://kojipkgs.fedoraproject.org//work/tasks/4273/8934273/build.log

--- Additional comment from HAYASHI Kentaro on 2015-02-14 14:59:03 JST ---

(In reply to HAYASHI Kentaro from comment #1)
> It is a bit strange because mock build is ok, but koji --scrach build fails.
> 
> Crash: test_limit_block (default context)
> https://kojipkgs.fedoraproject.org//work/tasks/4273/8934273/build.log

It needs more patches which are already fixed in upstream.

Then, make check process is a bit more succeeded now.
https://kojipkgs.fedoraproject.org//work/tasks/4737/8934737/build.log

--- Additional comment from HAYASHI Kentaro on 2015-02-17 23:45:49 JST ---



--- Additional comment from HAYASHI Kentaro on 2015-02-18 00:20:54 JST ---

(In reply to HAYASHI Kentaro from comment #3)
> Created attachment 992728 [details]
> try to fix "make check" failure patch for rawhide

F-21: it's ok without this patch
rawhide: it's ok with this patch

F-21: gcc 4.9.2-6
rawhide: gcc 5.0.0-0.13

rawhide default: faild
rawhide with --enable-debug: failed

So, there is nothing to do with optimization.

And, there is no difference compile option for test-cppcut-assertions.cpp.

--- Additional comment from HAYASHI Kentaro on 2015-02-18 01:30:49 JST ---

(In reply to HAYASHI Kentaro from comment #4)
> (In reply to HAYASHI Kentaro from comment #3)
> > Created attachment 992728 [details]
> > try to fix "make check" failure patch for rawhide
> 
> F-21: it's ok without this patch
> rawhide: it's ok with this patch
> 

typo, It's ok for f-21 and f22, not rawhide.

--- Additional comment from Mamoru TASAKA on 2015-02-20 20:03:39 JST ---

By the way, I tried git head (3947d292996458f3) on F-22 (both i386,
x86_64) and make check fails like:

FAIL: run-test.sh
============================================================================
Testsuite summary for cutter 1.2.4
============================================================================
# TOTAL: 1
# PASS:  0
# SKIP:  0
# XFAIL: 0
# FAIL:  1
# XPASS: 0
# ERROR: 0
============================================================================
See test/test-suite.log
Please report to kou
============================================================================

test/test-suite.log attached. (While I have not investigated in detail) it looks like (again just my guess) this segv is derived from cutter/test-cut-test.c:323. and the signal wasn't caught properly...???

--- Additional comment from Mamoru TASAKA on 2015-02-20 20:09:05 JST ---

I reported the issue on my comment 6 to the upstream:
https://github.com/clear-code/cutter/issues/20

--- Additional comment from Fedora Update System on 2015-02-25 23:02:51 JST ---

cutter-1.2.4-4.fc22 has been submitted as an update for Fedora 22.
https://admin.fedoraproject.org/updates/cutter-1.2.4-4.fc22

--- Additional comment from Fedora Update System on 2015-02-25 23:33:55 JST ---

cutter-1.2.4-4.fc21 has been submitted as an update for Fedora 21.
https://admin.fedoraproject.org/updates/cutter-1.2.4-4.fc21

--- Additional comment from HAYASHI Kentaro on 2015-02-25 23:48:35 JST ---

I've checked again build status for http://pkgs.fedoraproject.org/cgit/cutter.git/.

koji reports:
  rawhide stil fail
  f22 succeeded
  f21 succeeded

Here is the build logs:

  rawhide makecheck failed
  http://koji.fedoraproject.org/koji/taskinfo?taskID=9066948

  f22 make check succeeded
  http://koji.fedoraproject.org/koji/taskinfo?taskID=9066950

  f21 make check succeeded
  http://koji.fedoraproject.org/koji/taskinfo?taskID=9066960

The situation is different because it doesn't segv on f-22.

--- Additional comment from Fedora Update System on 2015-02-26 00:15:30 JST ---

Package cutter-1.2.4-4.fc22:
* should fix your issue,
* was pushed to the Fedora 22 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing cutter-1.2.4-4.fc22'
as soon as you are able to.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2015-2555/cutter-1.2.4-4.fc22
then log in and leave karma (feedback).

--- Additional comment from Mamoru TASAKA on 2015-02-26 14:35:57 JST ---

Currently I have no idea the cause of test failure on F-23.
Comment 1 Mamoru TASAKA 2015-03-02 05:41:02 UTC 
So even on F-22 (currently build is successful), enabling harneded build make test suite fail,
http://koji.fedoraproject.org/koji/taskinfo?taskID=9115889

(note: disabling hardened build on F-22 makes test successful
 http://koji.fedoraproject.org/koji/taskinfo?taskID=9115897
)

So this is related to -pie issue.
However currently even if I disable hardened build on glib2 and cutter, still test fails, I guess I have disable hardededn build on more packages.

Guessing harned build affects dlopen'ed libraries in an abnormal ways...

https://lists.fedoraproject.org/pipermail/devel/2015-March/208636.html
Comment 2 Moez Roy 2015-03-14 14:49:26 UTC 
Can you please add '%global _hardened_build 1' to the spec file (from the master branch).

Does it build with target to F21 with the above change to the spec file?
Comment 3 Mamoru TASAKA 2015-03-14 15:03:20 UTC 
Read my comment 1.

(In reply to Mamoru TASAKA from comment #1)
> So even on F-22 (currently build is successful), enabling harneded build
> make test suite fail,
> http://koji.fedoraproject.org/koji/taskinfo?taskID=9115889
> 
> (note: disabling hardened build on F-22 makes test successful
>  http://koji.fedoraproject.org/koji/taskinfo?taskID=9115897
> )
> 
> So this is related to -pie issue.
> However currently even if I disable hardened build on glib2 and cutter,
> still test fails, I guess I have disable hardededn build on more packages.
> 
> Guessing harned build affects dlopen'ed libraries in an abnormal ways...
> 
> https://lists.fedoraproject.org/pipermail/devel/2015-March/208636.html
Comment 4 Mamoru TASAKA 2015-03-14 15:07:02 UTC 
This is very difficult because we have to figure what fpie'ed packages affects this.
Comment 5 Jan Kurik 2015-07-15 14:28:07 UTC 
This bug appears to have been reported against 'rawhide' during the Fedora 23 development cycle.
Changing version to '23'.

(As we did not run this process for some time, it could affect also pre-Fedora 23 development
cycle bugs. We are very sorry. It will help us with cleanup during Fedora 23 End Of Life. Thank you.)

More information and reason for this action is here:
https://fedoraproject.org/wiki/BugZappers/HouseKeeping/Fedora23
Comment 6 Kentaro Hayashi 2016-01-21 17:11:55 UTC 
This bug will be solved by cutter 1.2.5-1
Comment 7 Mamoru TASAKA 2016-02-16 07:33:46 UTC 
Thank you for fixing.
Note You need to log in before you can comment on or make changes to this bug.