Bug 1557511

Summary: annobin: failure building wxGTK3 in F28
Product: [Fedora] Fedora Reporter: Scott Talbert <swt>
Component: annobinAssignee: Nick Clifton <nickc>
Status: CLOSED ERRATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 28CC: belegdol, fweimer, fzatlouk, mhroncok, ngompa13, nickc, sergio
Target Milestone: ---   
Target Release: ---   
Hardware: Unspecified   
OS: Unspecified   
Whiteboard: AcceptedFreezeException
Fixed In Version: annobin-5.0-1.fc28 annobin-5.1-1.fc28 Doc Type: If docs needed, set a value
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2018-03-26 22:29:34 UTC Type: Bug
Regression: --- Mount Type: ---
Documentation: --- CRM:
Verified Versions: Category: ---
oVirt Team: --- RHEL 7.3 requirements from Atomic Host:
Cloudforms Team: --- Target Upstream Version:
Embargoed:
Bug Depends On:    
Bug Blocks: 1469205    

Description Scott Talbert 2018-03-16 18:23:02 UTC
I'm currently unable to build wxGTK3 in F28 (exact same package builds fine in Rawhide).  In the logs, I see this error repeatedly:

annobin: sound_sdl.cpp: CET values have changed from 0:1954187174:20:1000000000000 to 0:100824:20:1000000000000

And then later, it fails to link:

`_ZN8wxThread8OnDeleteEv_end' referenced in section `.gnu.build.attributes' of advdll_sound_sdl.o: defined in discarded section `.text._ZN8wxThread8OnDeleteEv[_ZN8wxThread8OnDeleteEv]' of advdll_sound_sdl.o

See the full build log here:
https://kojipkgs.fedoraproject.org//work/tasks/4989/25734989/build.log

Comment 1 Julian Sikorski 2018-03-17 11:05:10 UTC
mame is affected by what looks to be the same issue. I get the following message:

annobin: model1.cpp: CET values have changed from 32764:0:20:1000000000000 to 16:0:20:1000000000001

linking failure occurs later:

`_ZSt20__replacement_assertPKciS0_S0__end' referenced in section `.gnu.build.attributes' of ../../../../linux_gcc/bin/x64/Release/mame_mame/libsega.a(model1.o): defined in discarded section `.text._ZSt20__replacement_assertPKciS0_S0_[_ZSt20__replacement_assertPKciS0_S0_]' of ../../../../linux_gcc/bin/x64/Release/mame_mame/libsega.a(model1.o)

Same as mentioned by Scott, rawhide build works as expected. Full log is here:

https://koji.fedoraproject.org/koji/taskinfo?taskID=25582153

As you can see, the issue only affects x86_64.

Comment 2 Neal Gompa 2018-03-17 15:11:59 UTC
This is afflicting Mir pre-release builds too: https://copr-be.cloud.fedoraproject.org/results/ngompa/mir-prerel-fedora/fedora-28-x86_64/00728338-mir/build.log.gz

Comment 3 Nick Clifton 2018-03-19 10:23:44 UTC
Hi Guys,

  Hmm, well I am not sure why rawhide should be working, when F28 is not,
  but the solution seems obvious - sync F28 annobin with rawhide.  So this
  is what I have done.  Please could you try:  annobin-5.0-1.fc28  and let
  me know if the problem still persists.

Cheers
  Nick

Comment 4 Nick Clifton 2018-03-19 10:40:03 UTC
Oops - I should have also mentioned that the message about "CET values have changed" is just a debugging output that I accidentally left in the plugin.
It should have no effect on plugin's behaviour and it should not be causing
the problem.

Comment 5 Scott Talbert 2018-03-20 01:20:54 UTC
I was able to build wxGTK3 successfully with annobin-5.0-1.fc28, so that seems to have resolved the problem.  As as aside, there seems to be some different debug output now:

BUILDSTDERR: url.cpp: ICE: unknown dwarf version level -1
BUILDSTDERR: annobin: url.cpp: ICE: unknown dwarf version level -1
BUILDSTDERR: annobin: zlibstream.cpp: ICE: unknown dwarf version level -1

Comment 6 Nick Clifton 2018-03-20 11:43:58 UTC
Hi Scott,

  Thanks for letting me know about this error message.  I am not sure
  how gcc can have a value of -1 for the DWARF version, but I have now
  added code to the annobin plugin to handle it.  Please try: 
  annobin-5.1-1.fc29 or annobin-5.1-1.fc28 and see it this makes things
  better.

Cheers
  Nick

Comment 7 Miro Hrončok 2018-03-20 13:09:12 UTC
Nick, could you please create an update for this?

Comment 8 Fedora Update System 2018-03-20 13:15:20 UTC
annobin-5.1-1.fc28 has been submitted as an update to Fedora 28. https://bodhi.fedoraproject.org/updates/FEDORA-2018-2c415304da

Comment 9 Neal Gompa 2018-03-20 14:52:17 UTC
It wasn't working in a scratch build earlier today: https://koji.fedoraproject.org/koji/taskinfo?taskID=25834766

I assume it's because the annobin is horrifically out of date there...

Comment 10 Fedora Update System 2018-03-20 14:55:47 UTC
annobin-5.1-1.fc28 has been pushed to the Fedora 28 testing repository. If problems still persist, please make note of it in this bug report.
See https://fedoraproject.org/wiki/QA:Updates_Testing for
instructions on how to install test updates.
You can provide feedback for this update here: https://bodhi.fedoraproject.org/updates/FEDORA-2018-2c415304da

Comment 11 Scott Talbert 2018-03-20 23:34:03 UTC
Yes, annobin-5.1-1.fc28 gets rid of the unknown dwarf version level error when building wxGTK3.

Comment 12 Sergio Basto 2018-03-21 00:14:13 UTC
to get it into the buildroot permanently during the current beta freeze, you have to submit it as an update and nominate for freeze exception: https://qa.fedoraproject.org/blockerbugs/propose_bug (https://fedoraproject.org/wiki/QA:SOP_freeze_exception_bug_process).

Comment 13 Fedora Blocker Bugs Application 2018-03-21 00:24:12 UTC
Proposed as a Freeze Exception for 28-beta by Fedora user sergiomb using the blocker tracking app because:

 annobin, fix failure buildings

Comment 14 František Zatloukal 2018-03-22 18:44:20 UTC
Discussed during blocker review [1]:

AcceptedFreezeException (Beta) - this is accepted as a freeze exception on the basis it's needed to build some packages, and pulling it in as an FE rather than using a buildroot override keeps the contents of the composes more consistent

[1] https://meetbot-raw.fedoraproject.org/fedora-meeting-1/2018-03-22/

Comment 15 Fedora Update System 2018-03-26 22:29:34 UTC
annobin-5.1-1.fc28 has been pushed to the Fedora 28 stable repository. If problems still persist, please make note of it in this bug report.