After an evaluation, GNOME has moved from Bugzilla to GitLab. Learn more about GitLab.
No new issues can be reported in GNOME Bugzilla anymore.
To report an issue in a GNOME project, go to GNOME GitLab.
Do not go to GNOME Gitlab for: Bluefish, Doxygen, GnuCash, GStreamer, java-gnome, LDTP, NetworkManager, Tomboy.
Bug 160384 - "make check" of the libgail-gnome-1.1.0 package fails with assertion failures
"make check" of the libgail-gnome-1.1.0 package fails with assertion failures
Status: RESOLVED OBSOLETE
Product: atk
Classification: Platform
Component: gail
git master
Other other
: Low minor
: ---
Assigned To: padraig.obriain
padraig.obriain
AP3
Depends on:
Blocks:
 
 
Reported: 2004-12-03 21:25 UTC by Randy McMurchy
Modified: 2011-01-21 06:46 UTC
See Also:
GNOME target: ---
GNOME version: 2.11/2.12


Attachments
turns off the test-gail-gnome test since it segfaults anyway (555 bytes, patch)
2005-07-14 18:27 UTC, Brent Smith (smitten)
none Details | Review
updated, specified gnome bugzilla in the comment (574 bytes, patch)
2005-07-14 18:34 UTC, Brent Smith (smitten)
committed Details | Review

Description Randy McMurchy 2004-12-03 21:25:17 UTC
To: submit@bugs.gnome.org
X-Mailer: bug-buddy 2.8.0
Subject: "make check" of the libgail-gnome-1.1.0 package fails with assertion failures

Distribution: LinuxFromScratch
Package: atk
Severity: normal
Version: GNOME2.8.1 unspecified
Gnome-Distributor: GNOME.Org
Synopsis: "make check" of the libgail-gnome-1.1.0 package fails with assertion failures
Bugzilla-Product: atk
Bugzilla-Component: gail
Bugzilla-Version: unspecified
BugBuddy-GnomeVersion: 2.0 (2.8.0)
Description:
Description of the crash:

During the "make check" of the libgail-gnome package (1.1.0), it crashes
due to the following:

** ERROR **: file test-gail-gnome.c: line 87 (test_spi_stuff): assertion
failed: (SPI_IS_REMOTE_OBJECT (socket_obj))
aborting...


Steps to reproduce the crash:
1. Install GNOME platform, desktop applications, including AT-SPI-1.6.0
2. build libgail-gnome, then issue "make check"
3. It produces this error

Expected Results:
Run the assertion tests correctly

How often does this happen?
Always

Additional Information:

Commenting out the assertion test on line 87 produces error at the
next test, and so on and so on.





Debugging Information:

Backtrace was generated from
'/home/rml/build/libgail-gnome-1.1.0/test/.libs/lt-test-gail-gnome'

Using host libthread_db library "/lib/libthread_db.so.1".
[Thread debugging using libthread_db enabled]
[New Thread 1089089120 (LWP 30683)]
0xffffe410 in ?? ()

Thread 1 (Thread 1089089120 (LWP 30683))

  • #0 ??
  • #1 ??
  • #2 ??
  • #3 ??
  • #4 __waitpid_nocancel
    from /lib/libpthread.so.0
  • #5 libgnomeui_segv_handle
    from /opt/gnome-2.8/lib/libgnomeui-2.so.0
  • #6 <signal handler called>
  • #7 ??
  • #8 ??
  • #9 ??
  • #10 ??
  • #11 raise
    from /lib/libc.so.6
  • #12 abort
    from /lib/libc.so.6
  • #13 g_logv
    from /usr/lib/libglib-2.0.so.0
  • #14 g_log
    from /usr/lib/libglib-2.0.so.0
  • #15 test_spi_stuff
    at test-gail-gnome.c line 87
  • #16 test_gail_gnome
    at test-gail-gnome.c line 125
  • #17 main
    at test-gail-gnome.c line 149
  • #0 ??




------- Bug moved to this database by unknown@bugzilla.gnome.org 2004-12-03 16:25 -------


Unknown platform unknown. Setting to default platform "Other".
Unknown milestone "unknown" in product "atk".
   Setting to default milestone for this product, '---'
Setting to default status "UNCONFIRMED".
Setting qa contact to the default for this product.
   This bug either had no qa contact or an invalid one.

Comment 1 padraig.obriain 2004-12-04 10:47:32 UTC
If the accessibility Gconf key is set then this problem does not occur.

gconftool-2 -s -t bool /desktop/gnome/interface/accessibility true.

However I get a crash on line 90. I will investigate further.
Comment 2 padraig.obriain 2004-12-06 09:26:37 UTC
The procewss bonobo-sample-controls-2 does not load libgnomeui so libgail-gnome
is not loaded.
Comment 3 padraig.obriain 2004-12-06 11:55:02 UTC
It looks like this has not worked since March 2002. See CVS Log entry for
version 1.5.
Comment 4 bill.haneman 2004-12-06 12:34:14 UTC
lowering priority and severity (because this package is very stable, i.e. almost
no code changes are taking place).  Therefore leaving this as-is is low risk.  A
patch to fix this would be welcome.

In the meantime perhaps we should modify Makefile.am so that this test is not
attempted if someone runs 'make check'.
Comment 5 Brent Smith (smitten) 2005-07-14 18:25:09 UTC
still get this is CVS HEAD, changing version to 2.11/2.12 and attaching an
*untested* patch
Comment 6 Brent Smith (smitten) 2005-07-14 18:27:09 UTC
Created attachment 49185 [details] [review]
turns off the test-gail-gnome test since it segfaults anyway
Comment 7 Brent Smith (smitten) 2005-07-14 18:34:27 UTC
Created attachment 49186 [details] [review]
updated, specified gnome bugzilla in the comment
Comment 8 bill.haneman 2005-07-15 13:16:28 UTC
Comment on attachment 49186 [details] [review]
updated, specified gnome bugzilla in the comment

thanks Brent
Comment 9 Brent Smith (smitten) 2005-12-29 17:40:54 UTC
Bill, can we close this?  Or are we keeping it around b/c of the segfault?  It seems like if it's never going to be fixed, we should just close it.

2005-06-15  Luis Villa <luis.villa@gmail.com>
        * tests/Makefile.am disable 'make check' as it currently
        segfaults- bug #160384. Patch from Brent Smith.
Comment 10 Kjartan Maraas 2006-01-08 15:08:08 UTC
The patch was commited. Marking as such.
Comment 11 Calum Benson 2006-04-26 17:05:38 UTC
Apologies for spam... ensuring Sun a11y folks are cc'ed on all current accessibility bugs.