https://bugzilla.novell.com/show_bug.cgi?id=215634
Summary: openposix.pthread_spin_lock_1_1 does not finish Product: SUSE Linux 10.1 Version: Final Platform: i686 OS/Version: Other Status: NEW Severity: Normal Priority: P5 - None Component: Other AssignedTo: erik.hamera@novell.com ReportedBy: pkirsch@novell.com QAContact: qa@suse.de CC: tarnold@suse.de
running the LTP-testsuite on zert169 especially the testcase /usr/share/qa/ltp/tmp/ltp_temp/openposix.pthread_spin_lock_1_1 from openposix.tcf (located in /usr/share/qa/ltp/tcf) does cause a 100% cpu loop. The loop is caused by the file /usr/lib/ltp/testcases/bin/openposix/conformance/interfaces/pthread_spin_lock/1-1.test (see strace file attached)
zert169:~ # rpm -q ltp ltp-20060412-2 zert169:~ # uname -a Linux zert169 2.6.16.21-0.25-default #1 Tue Sep 19 07:26:15 UTC 2006 i686 i686 i386 GNU/Linux
https://bugzilla.novell.com/show_bug.cgi?id=215634
------- Comment #1 from pkirsch@novell.com 2006-10-27 05:03 MST ------- Created an attachment (id=102804) --> (https://bugzilla.novell.com/attachment.cgi?id=102804&action=view) strace 1-1.test
https://bugzilla.novell.com/show_bug.cgi?id=215634
------- Comment #2 from pkirsch@novell.com 2006-10-27 05:04 MST ------- Sorry, i forgot to mention it is the remastered version of SL 10.1
https://bugzilla.novell.com/show_bug.cgi?id=215634
erik.hamera@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |ASSIGNED
------- Comment #3 from erik.hamera@novell.com 2006-10-27 12:36 MST ------- It hangs on my SL10.1 too ... strace -tt 20:23:10.845854 execve("./1-1.test", ["./1-1.test"], [/* 58 vars */]) = 0 .. 20:23:12.857262 munmap(0xb7579000, 27974) = 0 20:23:12.857419 futex(0xb7578644, FUTEX_WAKE, 2147483647) = 0 20:23:12.857552 _exit(0) = ? .. and now it call nothing ...
I'll see on it
https://bugzilla.novell.com/show_bug.cgi?id=215634
------- Comment #4 from yxu@novell.com 2006-11-14 05:06 MST ------- I would suggest we skip this test by removing this paragraph:
timer 300 fg 1 pthread_spin_lock_1_1 /usr/lib/ltp/runalltests.sh -f /usr/share/qa/ltp/tmp/ltp_temp/openposix.pthread_spin_lock_1_1 wait
in the file /usr/share/qa/ltp/tcf/openposix.tcf
on that certain machine for a temorary solution.
Erik or I will look into the source code and maybe make a patch and apply the patch to ltp and build a new RPM package, but it may take some time, and right now, ltp is in a freezing status. So I come up with the idea of skipping the test that makes the cpu loop.
( Another way could be update your ltp, since the new ltp package is available for sles10. /mounts/work/built/dists/all/sles10-i386/packs-i386/ltp/ltp-20061017/ltp-20061017-1.4.i586.rpm But i am not sure if there is anything new about pthread_spin_lock_1_1 in the new ltp package. )
https://bugzilla.novell.com/show_bug.cgi?id=215634
------- Comment #5 from pkirsch@novell.com 2006-11-14 06:03 MST ------- There is a similar harder situation in Bug #145160 with testcase sched_yield_1_1. This testcase makes (under special circumstances) machines freeze. So I'm not sure if it is a good idea to skip this testcase. Is there a possibility to ask the author of the testcase of some 'besides effects', if she/he knows one.
https://bugzilla.novell.com/show_bug.cgi?id=215634
Martin Mrazik mmrazik@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Priority|P5 - None |P4 - Low
https://bugzilla.novell.com/show_bug.cgi?id=215634
Martin Mrazik mmrazik@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- AssignedTo|erik.hamera@novell.com |dsterba@novell.com Status|ASSIGNED |NEW
https://bugzilla.novell.com/show_bug.cgi?id=215634
User dsterba@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=215634#c8
David Sterba dsterba@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- CC| |erik.hamera@novell.com
--- Comment #8 from David Sterba dsterba@novell.com 2008-10-29 08:22:37 MDT --- I've reproduced the bug with
SLES10 SP2 2.6.16.60-0.21-default LTP 20060412
and found out that LTP 20080930 has updated version of the test (with some more signal handling) and the test does not hang anymore. I'll submit the fix to our LTP.
https://bugzilla.novell.com/show_bug.cgi?id=215634
User dsterba@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=215634#c9
David Sterba dsterba@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEW |NEEDINFO Info Provider| |erik.hamera@novell.com
--- Comment #9 from David Sterba dsterba@novell.com 2008-10-29 08:42:48 MDT --- Hm, the fix is already there (LTP 20071231).
Erik, or Yi Xu, can you please retest again if you still encounter this problem? Else I consider this fixed.
https://bugzilla.novell.com/show_bug.cgi?id=215634
User mmrazik@novell.com added comment https://bugzilla.novell.com/show_bug.cgi?id=215634#c10
Martin Mrazik mmrazik@novell.com changed:
What |Removed |Added ---------------------------------------------------------------------------- Status|NEEDINFO |RESOLVED Info Provider|erik.hamera@novell.com | Resolution| |FIXED
--- Comment #10 from Martin Mrazik mmrazik@novell.com 2009-01-06 03:21:41 MST --- closing as of comment #9
-
bugzilla_noreply＠novell.com