Bug 48251 - guality_check hangs indefinitely on Tru64 UNIX
Summary: guality_check hangs indefinitely on Tru64 UNIX
Status: RESOLVED WONTFIX
Alias: None
Product: gcc
Classification: Unclassified
Component: testsuite (show other bugs)
Version: 4.7.0
: P3 normal
Target Milestone: ---
Assignee: Not yet assigned to anyone
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2011-03-23 11:42 UTC by Rainer Orth
Modified: 2014-06-13 11:52 UTC (History)
3 users (show)

See Also:
Host: alpha-dec-osf5.1b
Target: alpha-dec-osf5.1b
Build: alpha-dec-osf5.1b
Known to work:
Known to fail: 4.6.0, 4.7.0
Last reconfirmed: 2011-03-23 12:00:31


Attachments

Note You need to log in before you can comment on or make changes to this bug.
Description Rainer Orth 2011-03-23 11:42:31 UTC
The guality_check in g++.dg/guality/guality.exp hangs indefinitely on Tru64 UNIX
V5.1B, hanging the whole testsuite run.  It seems to be a gdb issue, but still
occurs with CVS gdb.

Here's what happens when I use a simple looping test program as target of the
procedure used in guality.h:

> gdb -nx -nw --quiet 
(gdb) set height
Argument required (integer to set it to.).
(gdb) set height 0
(gdb) attach 216346
Attaching to process 216346
[New process 216346]
Retry #1:
Retry #2:
Retry #3:
Retry #4:
0x00000001200010e0 in ?? ()
(gdb) set guality_attached = 1
No symbol table is loaded.  Use the "file" command.
(gdb) b 1
No symbol table is loaded.  Use the "file" command.

In g++.log, I find that the test is run with the regular timeout of 300 seconds,
but this has no effect for some reason.  Might be a bug in expect (5.38.2).

The guality_check test in gcc.dg/guality/guality.exp isn't affected for some
reason.

I'll disable the test on Tru64 UNIX until this is resolved.

This is somewhat related to PR testsuite/46529, but there the timeout does
work.
Comment 1 Rainer Orth 2011-03-23 11:56:09 UTC
Author: ro
Date: Wed Mar 23 11:55:51 2011
New Revision: 171346

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=171346
Log:
	PR testsuite/48251
	* g++.dg/guality/guality.exp: Disable on alpha*-dec-osf*.

Modified:
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/testsuite/g++.dg/guality/guality.exp
Comment 2 Rainer Orth 2011-03-23 12:00:31 UTC
Worked around for now.
Comment 3 Rainer Orth 2011-03-28 16:46:33 UTC
Author: ro
Date: Mon Mar 28 16:46:27 2011
New Revision: 171616

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=171616
Log:
	Backport from mainline:
	2011-03-23  Rainer Orth  <ro@CeBiTec.Uni-Bielefeld.DE>

	PR testsuite/48251
	* g++.dg/guality/guality.exp: Disable on alpha*-dec-osf*.

Modified:
    branches/gcc-4_6-branch/gcc/testsuite/ChangeLog
    branches/gcc-4_6-branch/gcc/testsuite/g++.dg/guality/guality.exp
Comment 4 Rainer Orth 2011-04-18 15:54:59 UTC
Author: ro
Date: Mon Apr 18 15:54:50 2011
New Revision: 172657

URL: http://gcc.gnu.org/viewcvs?root=gcc&view=rev&rev=172657
Log:
	PR testsuite/48251
	* gcc.dg/guality/guality.exp: Disable on alpha*-dec-osf*.
	* gfortran.dg/guality/guality.exp: Likewise.

Modified:
    trunk/gcc/testsuite/ChangeLog
    trunk/gcc/testsuite/gcc.dg/guality/guality.exp
    trunk/gcc/testsuite/gfortran.dg/guality/guality.exp
Comment 5 Jakub Jelinek 2011-05-19 10:38:20 UTC
Can this be closed now?
Comment 6 ro@CeBiTec.Uni-Bielefeld.DE 2011-05-19 11:07:14 UTC
> --- Comment #5 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-05-19 10:38:20 UTC ---
> Can this be closed now?

No, skipping the tests is just a workaround.  I mean to look into the
root cause of this and would like the PR to be kept open until then.

	Rainer
Comment 7 Jakub Jelinek 2011-05-19 11:28:25 UTC
If you can't attach to a running process, that sounds like either a gdb bug, or kernel bug, certainly not a gcc fault.
Comment 8 ro@CeBiTec.Uni-Bielefeld.DE 2011-05-19 12:16:25 UTC
> --- Comment #7 from Jakub Jelinek <jakub at gcc dot gnu.org> 2011-05-19 11:28:25 UTC ---
> If you can't attach to a running process, that sounds like either a gdb bug, or
> kernel bug, certainly not a gcc fault.

I'd bet it's a gdb bug, but I'd still keep the gcc testsuite bug open as
a reminder.

	Rainer
Comment 9 Hin-Tak Leung 2012-08-30 17:54:25 UTC
Thread issue? 
Many of the newer mpfr make check also takes a long time/time out if it is left to the default tls enabled:
 
https://gforge.inria.fr/tracker/?func=detail&atid=619&aid=14806&group_id=136
Comment 10 Rainer Orth 2014-06-13 11:52:23 UTC
4.7 branch closed.