<?xml version="1.0" encoding="UTF-8" standalone="yes" ?>
<!DOCTYPE bugzilla SYSTEM "https://bugzilla.yoctoproject.org/page.cgi?id=bugzilla.dtd">

<bugzilla version="5.0.6"
          urlbase="https://bugzilla.yoctoproject.org/"
          
          maintainer="it-coreprojects-helpdesk@linuxfoundation.org"
>

    <bug>
          <bug_id>1713</bug_id>
          
          <creation_ts>2011-11-02 17:37:27 +0000</creation_ts>
          <short_desc>[NAS] Oprofile will cause kernel hang when profiling.</short_desc>
          <delta_ts>2012-08-16 21:00:21 +0000</delta_ts>
          <reporter_accessible>1</reporter_accessible>
          <cclist_accessible>1</cclist_accessible>
          <classification_id>7</classification_id>
          <classification>Build System, Metadata &amp; Runtime</classification>
          <product>BSPs</product>
          <component>bsps-runtime</component>
          <version>unspecified</version>
          <rep_platform>Blacksand</rep_platform>
          <op_sys>x86</op_sys>
          <bug_status>RESOLVED</bug_status>
          <resolution>WONTFIX</resolution>
          
          
          <bug_file_loc></bug_file_loc>
          <status_whiteboard>(P3)</status_whiteboard>
          <keywords></keywords>
          <priority>Medium</priority>
          <bug_severity>normal</bug_severity>
          <target_milestone>1.3</target_milestone>
          
          
          <everconfirmed>1</everconfirmed>
          <reporter name="Shane Wang">shane.wang</reporter>
          <assigned_to name="Lianhao Lu">lianhao.lu</assigned_to>
          <cc>alexandru.c.georgescu</cc>
    
    <cc>dvhart</cc>
    
    <cc>jessica.zhang</cc>
    
    <cc>lianhao.lu</cc>
    
    <cc>meta.mr.watcher</cc>
    
    <cc>meta.watcher</cc>
    
    <cc>poky.adt.watcher</cc>
    
    <cc>poky.watcher</cc>
    
    <cc>yp.kernel.watcher</cc>
    
    <cc>yp.watcher</cc>
          
          
          <cf_os>---</cf_os>
          <cf_regression_type>---</cf_regression_type>
          
          <cf_docchange>---</cf_docchange>

      

      

      

          <comment_sort_order>oldest_to_newest</comment_sort_order>  
          <long_desc isprivate="0" >
    <commentid>17047</commentid>
    <comment_count>0</comment_count>
    <who name="Shane Wang">shane.wang</who>
    <bug_when>2011-11-02 17:37:27 +0000</bug_when>
    <thetext>During the profiling of NAS we implemented, when copying a big file (say 4GB) from the NAS samba server with ADT Oprofile plug-in, the NAS kernel will hang mostly. But without Oprofile, the issue disappears.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>17051</commentid>
    <comment_count>1</comment_count>
    <who name="Jessica">jessica.zhang</who>
    <bug_when>2011-11-02 18:27:53 +0000</bug_when>
    <thetext>So what&apos;s the symptom of kernel hang, and core dump file? By the way, have you tried just with oprofile-viewer and oprofile-server instead of through eclipse plug-in, does it hang as well?  &quot;But without Oprofile, the issue disappears&quot;, what exactly does this mean, you stop oprofile-server or not doing oprofile at all?

I have a feeling it may not related to ADT but Oprofile itself.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>17052</commentid>
    <comment_count>2</comment_count>
    <who name="Lianhao Lu">lianhao.lu</who>
    <bug_when>2011-11-02 18:43:05 +0000</bug_when>
    <thetext>It&apos;s should be a bug in oprofile itself, not ADT. By saying &quot;But without Oprofile, the issue disappears&quot;, it means doing the copy operation without oprofile at all.

In most cases, the NAS just hang silently without any core dump or any out output on the screen. How ever we noticed a kernel panic once and by reading the panic output on the screen, we found an oprofile related function in kernel named something like &quot;xxx_oprofile_add_sample_xxx&quot; was in the call stack back trace.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>17060</commentid>
    <comment_count>3</comment_count>
    <who name="Shane Wang">shane.wang</who>
    <bug_when>2011-11-02 20:40:56 +0000</bug_when>
    <thetext>The procedures to reproduce are: 

$opcontrol --start-daemon --separate=lib -c 6
$opcontrol --start

Then on a client, copying a 4GB file from NAS, NAS system will hang. Without Oprofile, the error has never happened.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>17082</commentid>
    <comment_count>4</comment_count>
    <who name="Darren Hart">dvhart</who>
    <bug_when>2011-11-03 15:58:20 +0000</bug_when>
    <thetext>Please provide the text of the kernel panic.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>17086</commentid>
    <comment_count>5</comment_count>
    <who name="Lianhao Lu">lianhao.lu</who>
    <bug_when>2011-11-03 20:19:01 +0000</bug_when>
    <thetext>It seems we can no longer reproduce the kernel panic, but the system just hang silently without any output on screen. Here is a script which could trigger the hang on NAS/blacksand:

#!/bin/sh
opcontrol --init
opcontrol --start-daemon --no-vmlinux --separate=lib -c 6
opcontrol --start
rm -f testfile
dd if=/dev/zero of=./testfile bs=1G count=4</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>17101</commentid>
    <comment_count>6</comment_count>
    <who name="Darren Hart">dvhart</who>
    <bug_when>2011-11-04 05:45:33 +0000</bug_when>
    <thetext>How long do you allow the command to run?</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>17117</commentid>
    <comment_count>7</comment_count>
    <who name="Lianhao Lu">lianhao.lu</who>
    <bug_when>2011-11-06 16:47:38 +0000</bug_when>
    <thetext>Normally, it will hang the system in less then 3 minutes.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>17757</commentid>
    <comment_count>8</comment_count>
    <who name="Jessica">jessica.zhang</who>
    <bug_when>2011-12-20 11:29:50 +0000</bug_when>
    <thetext>Per today&apos;s bug triage, decide to re-assign the bug back to PRC since they have the HW setup to further investigate the issue and come up a clear context on the issue and then seek help from Darren...</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>24087</commentid>
    <comment_count>9</comment_count>
    <who name="Alexandru Georgescu">alexandru.c.georgescu</who>
    <bug_when>2012-08-16 10:42:14 +0000</bug_when>
    <thetext>Hi Jessica,
Is there any update regarding the bug? Do you know if it&apos;s still reproducing?

Thanks,
Alex.</thetext>
  </long_desc><long_desc isprivate="0" >
    <commentid>24127</commentid>
    <comment_count>10</comment_count>
    <who name="Jessica">jessica.zhang</who>
    <bug_when>2012-08-16 21:00:21 +0000</bug_when>
    <thetext>Since we no longer have the setup and lack of info on how to reproduce it</thetext>
  </long_desc>
      
      

    </bug>

</bugzilla>