Bug 434147

Summary: libunwind failed massrebuild attempt for GCC 4.3
Product: [Fedora] Fedora Reporter: Jesse Keating <jkeating>
Component: libunwindAssignee: Jan Kratochvil <jan.kratochvil>
Status: CLOSED RAWHIDE QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: medium Docs Contact:
Priority: medium    
Version: rawhideCC: dcantrell
Target Milestone: ---   
Target Release: ---   
Hardware: All   
OS: Linux   
Whiteboard:
Fixed In Version: libunwind-0.99-0.4.frysk20070405cvs.fc9 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2008-02-24 05:17:14 UTC Type: ---
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: 427850    
Bug Blocks: 432425    
Attachments:
Description Flags
Output of "ps -e -o pid,ppid,pgid,cmd fww" during a hung lt-test-ptrace none

Description Jesse Keating 2008-02-22 13:10:11 UTC
This is an automatically filed bug for a failed rebuild attempt for GCC 4.3.

http://fedoraproject.org/wiki/JesseKeating/gcc43MassRebuildProposal

Please verify why this build failed and fix it.
http://koji.fedoraproject.org/koji/taskinfo?taskID=440103
Unknown exit code, check the failed task.

Comment 1 Mike Bonnet 2008-02-22 21:27:50 UTC
The i386 and x86_64 builds were hung during the lt-test-ptrace test.  By hung I
mean they were consuming 100% cpu time.  I'm attaching the output of a "ps -e -o
pid,ppid,pgid,cmd fww" to the bug.  The second to last entry (lt-test-ptrace)
was taking up most of the cpu time, with lt-test-ptrace-stepper occasionally
taking a few %.

Could this be a kernel/xen issue?  The builders are running 2.6.18-53.1.6.el5xen.


Comment 2 Mike Bonnet 2008-02-22 21:29:40 UTC
Created attachment 295678 [details]
Output of "ps -e -o pid,ppid,pgid,cmd fww" during a hung lt-test-ptrace

Comment 3 Jan Kratochvil 2008-02-22 21:42:03 UTC
It is in fact a DUPLICATE of Bug 427850.
Going to fix it ASAP.
(No matter why the test(s) got locked up no test should be allowed to busy-loop
forever.)


Comment 4 Jan Kratochvil 2008-02-24 05:17:14 UTC
Committed to Rawhide:
* Sun Feb 24 2008 Jan Kratochvil <jan.kratochvil> -
0.99-0.4.frysk20070405cvs
- Abort the possibly hung up testcases after 120 seconds (BZ 427850, BZ 434147).

It works for me on F8, though.  The right way is to remove the testing from the
build and move it to a separate testing system (RHTS) as has been done with GDB.

./orphanripper: 21633 Timed out after 120 seconds: ./run-ptrace-stepper
./orphanripper: Killed -9 orphan PID 21651:
/builddir/build/BUILD/libunwind-0.99-alpha/tests/.libs/lt-test-ptrace-stepper
./orphanripper: Killed -9 orphan PID 21635:
/builddir/build/BUILD/libunwind-0.99-alpha/tests/.libs/lt-test-ptrace -t -b jump
foo+ __kernel_sigreturn lockup prefoo+ main+ __libc_start_main+ _start+
./test-ptrace-stepper
FAIL: run-ptrace-stepper
Backtrace does not match on level==2, want "bowler+", got "main"
Backtrace does not match on level==3, want "main+", got "__libc_start_main"
Backtrace does not match on level==4, want "__libc_start_main+", got "_start"
Too short backtrace on level==5, want "_start+"
FAILURE: detected 4 errors