Bug 911 - [LTP] /usr/include/libio.h segfault with nm case on qemuppc
Summary: [LTP] /usr/include/libio.h segfault with nm case on qemuppc
Status: RESOLVED WONTFIX
Alias: None
Product: Kernel
Classification: Yocto Project Subprojects
Component: kernel-tooling (show other bugs)
Version: unspecified
Hardware: x86 Multiple
: Medium normal
Target Milestone: 1.1
Assignee: Bruce Ashfield
QA Contact:
URL:
Whiteboard: not looked at yet (may 16, 2011)
Depends on:
Blocks:
 
Reported: 2011-03-18 07:01 UTC by Jiajun Xu
Modified: 2011-07-31 20:01 UTC (History)
8 users (show)

See Also:
OS type for building Yocto: ---
Type of Regression: ---
Verified:
Documentation change: ---


Attachments
source file of nm01 (4.61 KB, text/plain)
2011-03-18 07:01 UTC, Jiajun Xu
no flags Details

Note You need to log in before you can comment on or make changes to this bug.
Description Jiajun Xu 2011-03-18 07:01:11 UTC
Created attachment 122 [details]
source file of nm01

Image location:
http://autobuilder.pokylinux.org/nightly/20110305-4/machines/qemu/qemuppc/
Tree/Branch: Distro/Bernard
Commit: commit c192fefd868be53c20dfd8bc44e3e815cf53a10a
Author: Saul Wold <sgw@linux.intel.com>
Date:   Fri Mar 4 18:13:50 2011 -0800

    lsbsetup: Fix LIC_FILE_CHKSUM

    Signed-off-by: Saul Wold <sgw@linux.intel.com>


The test case could pass in v1.0 M2 on qemuppc. Meanwhile, on other platform, it could pass also. The error log is as below:
####
tag=nm stime=1299398989
cmdline="export TCdat=$LTPROOT/testcases/bin; nm01 "
contacts=""
analysis=exit
<<<test_output>>>
In file included from /usr/include/stdio.h:75:0,
                 from /opt/ltp/testcases/bin/nmfile1.c:1:
/usr/include/libio.h:338:1: internal compiler error: Segmentation fault
Please submit a full bug report,
with preprocessed source if appropriate.
See <http://gcc.gnu.org/bugs.html> for instructions.
-------System test for nm command------
-)1
nm: '/tmp/nm12685/nmfile1': No such file
-)2
nm: '/tmp/nm12685/nmfile1': No such file
-)3
nm: '/tmp/nm12685/nmfile1': No such file
nm -f sysv: FAIL
<<<execution_status>>>
initiation_status="ok"
duration=40 termination_type=exited termination_id=1 corefile=no
cutime=3197 cstime=566
####
Comment 1 Bruce Ashfield 2011-03-21 13:18:13 UTC
Have we run this on the h/w ppc target ? There aren't
any kernel changes that would trigger this, so it is either
configuration, memory or something in the flags that might
be triggering this.

If you use the M2 kernel with the M4 userspace, what is
the result ?
Comment 2 Jiajun Xu 2011-03-23 07:35:27 UTC
(In reply to comment #1)
> Have we run this on the h/w ppc target ? There aren't
> any kernel changes that would trigger this, so it is either
> configuration, memory or something in the flags that might
> be triggering this.
> If you use the M2 kernel with the M4 userspace, what is
> the result ?

The same case could pass on real ppc target -- mpc8315e-rdb with RC2 build.
cc Yi Zhao to check if he could help to try M2 kernel with M4 userspace.
Comment 3 Bruce Ashfield 2011-03-23 12:12:18 UTC
Thanks for the update. This smells like another qemuppc limitation jumping up.
Comment 4 Yi Zhao 2011-04-11 00:28:28 UTC
I re-tested it on 3 different lsb-sdk build images:
M2 kernel + M4 userspace (20101218 kernel + 20110305 RC2 rootfs)
20110305 RC2 build image (I tested it again)
1.0 final release image (20110325 build)

And I found the case can *pass* on all of these images.

<<<test_start>>>
tag=nm stime=1301226684
cmdline="export TCdat=$LTPROOT/testcases/bin; nm01 "
contacts=""
analysis=exit
<<<test_output>>>
-------System test for nm command------
-)1
-)2
-)3
-)4
-)5
-)6
-)7
-)8
nm01: PASS
<<<execution_status>>>
initiation_status="ok"
duration=35 termination_type=exited termination_id=0 corefile=no
cutime=3023 cstime=392
<<<test_end>>>

So I think this issue maybe relate to bug 780
Comment 6 Bruce Ashfield 2011-05-24 08:54:38 UTC
Looks like this can be closed. Do you agree ? I'm assuming
that this continues to pass on 1.1 ..
Comment 7 Jingdong Lu 2011-07-31 20:01:04 UTC
Close this bug now and re-open it if needed.