Bug 1212314

Summary: [abrt] ltrace: type_pointer_destroy(): ltrace killed by SIGABRT
Product: [Fedora] Fedora Reporter: Daniel Vacek <dvacek>
Component: ltraceAssignee: Jeff Law <law>
Status: CLOSED ERRATA QA Contact: Fedora Extras Quality Assurance <extras-qa>
Severity: unspecified Docs Contact:
Priority: unspecified    
Version: 20CC: law, pmachata
Target Milestone: ---   
Target Release: ---   
Hardware: x86_64   
OS: Unspecified   
URL: https://retrace.fedoraproject.org/faf/reports/bthash/9eae5ee27f60f25bce8b1945f7e272b88e510a8f
Whiteboard: abrt_hash:83837430d0fb84382c6f0af24e84872818a9ba7c
Fixed In Version: ltrace-0.7.2-10.fc20 Doc Type: Bug Fix
Doc Text:
Story Points: ---
Clone Of: Environment:
Last Closed: 2015-04-26 12:59:16 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:
Attachments:
Description Flags
File: backtrace
none
File: cgroup
none
File: core_backtrace
none
File: dso_list
none
File: limits
none
File: maps
none
File: open_fds
none
File: proc_pid_status
none
complete console output none

Description Daniel Vacek 2015-04-16 07:12:04 UTC 
Description of problem:
just running it as usual

Version-Release number of selected component:
ltrace-0.7.2-9.fc20

Additional info:
reporter:       libreport-2.2.3
backtrace_rating: 4
cmdline:        ltrace -SfttTs 8192 ./ocl
crash_function: type_pointer_destroy
executable:     /usr/bin/ltrace
kernel:         3.19.3-100.fc20.x86_64+debug
type:           CCpp

Truncated backtrace:
Thread no. 1 (9 frames)
 #5 type_pointer_destroy at type.c:214
 #6 type_destroy at type.c:234
 #7 value_set_type at value.c:64
 #8 vect_destroy at vect.c:147
 #9 val_dict_destroy at value_dict.c:146
 #10 callstack_pop at handle_event.c:749
 #11 handle_breakpoint at handle_event.c:616
 #12 handle_event at handle_event.c:179
 #13 ltrace_main at libltrace.c:194
Comment 1 Daniel Vacek 2015-04-16 07:12:07 UTC 
Created attachment 1015051 [details]
File: backtrace
Comment 2 Daniel Vacek 2015-04-16 07:12:08 UTC 
Created attachment 1015052 [details]
File: cgroup
Comment 3 Daniel Vacek 2015-04-16 07:12:09 UTC 
Created attachment 1015053 [details]
File: core_backtrace
Comment 4 Daniel Vacek 2015-04-16 07:12:10 UTC 
Created attachment 1015054 [details]
File: dso_list
Comment 5 Daniel Vacek 2015-04-16 07:12:12 UTC 
Created attachment 1015055 [details]
File: limits
Comment 6 Daniel Vacek 2015-04-16 07:12:13 UTC 
Created attachment 1015056 [details]
File: maps
Comment 7 Daniel Vacek 2015-04-16 07:12:14 UTC 
Created attachment 1015057 [details]
File: open_fds
Comment 8 Daniel Vacek 2015-04-16 07:12:16 UTC 
Created attachment 1015058 [details]
File: proc_pid_status
Comment 9 Petr Machata 2015-04-16 09:18:31 UTC 
That looks like it might be a double free.  Do you happen to remember what the error message roughly said, before glibc killed the process?

Can you reproduce this?  And what is ocl?  Do you know if it's a multi-threaded program?
Comment 10 Daniel Vacek 2015-04-16 09:53:32 UTC 
Created attachment 1015114 [details]
complete console output

Hi Petr,

  luckily I still have the terminal from yesterday opened, output attached.

  It may very well be double free indeed. The actual message printed was:

*** Error in `ltrace': free(): invalid pointer: 0x000000000062a100 ***

  ocl is my project using OpenCL. AFAIK it is not multithreaded (I do not
believe pocl library runs it's own threads looking at the output).
  I was actually trying to trace out why is pocl failing, I ended up with
updating to new version and it is working now. ltrace is still crashing
though. Moreover everytime on the same pointer, I tried just now again.

Have a nice one,

--nX
Comment 11 Petr Machata 2015-04-16 10:33:49 UTC 
I think this might be
  http://permalink.gmane.org/gmane.comp.debugging.ltrace.devel/691

Let me try to spin a quick update to see if it helps.
Comment 12 Petr Machata 2015-04-16 10:45:04 UTC 
Could you please try the following build?

  http://koji.fedoraproject.org/koji/taskinfo?taskID=9492661

It's something of a blind shot, the address seems to come from an unmapped region, which is unexpected, and possibly indicates a different problem, but worth a try.
Comment 13 Daniel Vacek 2015-04-16 11:31:07 UTC 
Cool, good spot, Petře.

  That indeed matches and I can confirm koji build works like a charm.

Thanks,

--nX
Comment 14 Petr Machata 2015-04-16 11:41:33 UTC 
Good stuff, thanks for testing.  I'll spin an update.
Comment 15 Fedora Update System 2015-04-16 11:51:54 UTC 
ltrace-0.7.2-10.fc20 has been submitted as an update for Fedora 20.
https://admin.fedoraproject.org/updates/ltrace-0.7.2-10.fc20
Comment 16 Fedora Update System 2015-04-18 09:47:12 UTC 
Package ltrace-0.7.2-10.fc20:
* should fix your issue,
* was pushed to the Fedora 20 testing repository,
* should be available at your local mirror within two days.
Update it with:
# su -c 'yum update --enablerepo=updates-testing ltrace-0.7.2-10.fc20'
as soon as you are able to.
Please go to the following url:
https://admin.fedoraproject.org/updates/FEDORA-2015-6351/ltrace-0.7.2-10.fc20
then log in and leave karma (feedback).
Comment 17 Fedora Update System 2015-04-26 12:59:16 UTC 
ltrace-0.7.2-10.fc20 has been pushed to the Fedora 20 stable repository.  If problems still persist, please make note of it in this bug report.