RE: [Xen-ia64-devel] Weekly benchmark results [12/4rd week]

["yo.fujita" <yo.fujita@xxxxxxxxxxxxxxxx>]

Dan,

I retested gcc loop on the changeset 8474.
To our regret the same phenomenon occurred, this seems to be the same bug as
you cited in the previous mail.

I ran the test for about 2 hours and saw "segmentation fault" 3 times.

In addition, I got the following kernel oops.
" Unable to handle kernel paging request at virtual address 32000f0002000000
collect2[22056]: Oops 11012296146944 [1] "
I attach the result log to this mail.

I'll try to see if I can narrow down the condition that causes this bug.

Thanks
Fujita

> -----Original Message-----
> From: Magenheimer, Dan (HP Labs Fort Collins)
[mailto:dan.magenheimer@xxxxxx]
> Sent: Saturday, January 07, 2006 6:13 AM
> To: yo.fujita; xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> Subject: RE: [Xen-ia64-devel] Weekly benchmark results [12/4rd week]
> 
> I suspect that this is the same problem as reported here:
> 
> http://lists.xensource.com/archives/html/xen-ia64-devel/2005-12/msg00185.h
> tml
> 
> This is the first report on domU. I've only observed it on dom0.
> Because it seems random, it may be very difficult to track
> down.
> 
> Dan
> 
> > -----Original Message-----
> > From: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx
> > [mailto:xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx] On Behalf
> > Of yo.fujita
> > Sent: Friday, January 06, 2006 3:32 AM
> > To: 'yo.fujita'; xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> > Subject: RE: [Xen-ia64-devel] Weekly benchmark results [12/4rd week]
> >
> > Hi, all
> >
> > I had put out the following mail before.
> > And, it was pointed out that there was an unexpected value in the
> > bench result of DomU.
> > Thanks, Eddie.
> >
> > Please look at the following value.
> > dom0:
> >             C Compiler Throughput                       702.0 lpm
> > (60 secs, 3 samples)
> > domU:
> >             C Compiler Throughput                     64655.7 lpm
> > (62 secs, 2 samples)
> >
> > dom0:
> >             Shell Scripts (1 concurrent)               2276.3 lpm
> > (60 secs, 3 samples)
> > domU:
> >             Shell Scripts (1 concurrent)             473647.0 lpm
> > (61 secs, 2 samples)
> >
> > The value of dom0 and domU is greatly different.
> > I investigated this phenomenon.
> > As for the understood thing, domU uncommonly abort in gcc.
> > Please look at cc_loop.txt of the attached file.
> > This is a result of loop only C compile of unixbench.
> > When the result is Bad wait status:0xb or 0x100, the value will become
> > abnormal.
> > As for "Shell Scripts", it might be similar.
> > the next benchmark will be executed with changeset8474.
> >
> > Thanks,
> > Fujita
> >
> > > -----Original Message-----
> > > From: xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx
> > > [mailto:xen-ia64-devel-bounces@xxxxxxxxxxxxxxxxxxx] On
> > Behalf Of yo.fujita
> > > Sent: Thursday, December 29, 2005 3:59 PM
> > > To: xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
> > > Subject: [Xen-ia64-devel] Weekly benchmark results [12/4rd week]
> > >
> > > Hi,
> > >
> > > I am investigating stability of Dom0 and DomU by using
> > various benchmark
> > > tools.
> > > We Fujitsu contribute with the results of benchmarks to
> > this community
> > > periodically from now on.
> > >
> > > Although it is only the detail result of unixbench this
> > time, we will post
> > > the detail result of all benchmarks someday.
> > >
> > > The tool used now is as follows.
> > >  - unixbench4.1.0
> > >  - bonnie++-1.03
> > >  - ltp-full-20051103
> > >  - iozone3_191
> > >  - lmbench-3.0-a5
> > >
> > > TEST ENVIRONMENT
> > >     Machine          : Tiger4
> > >     CPU              : Itanium2 1.5GHz *2 (Madison)
> > >     MEM              : 8GB
> > >     HDD              : Ultra 320 SCSI 36.7GB(MAS3367NC)
> > >     KERN             : 2.6.12-xen0
> > >     changeset        : 8372:7ef565a7cc86 (12/15 latest)
> > >     Dom0 OS          : RHEL4 U2 (no SMP)
> > >     DomU OS          : RHEL4 U2 (no SMP)
> > >     No. of DomU's    : 1
> > >
> > > SUMMARY
> > >  - ltp-full-20051103 :hang in bench testing.
> > >  - Iozone3_191       :hang in bench testing.
> > >  - bonnie++-1.03     :bench is pass. but hang immediately after.
> > >
> > > issues:
> > >      - If changeset is after 8732, DomU will not start.
> > >
> > > TEST RESULT
> > >     unixbench4.1.0    : PASS
> > >     bonnie++-1.03　   : FAIL
> > >     ltp-full-20051103 : FAIL
> > >     iozone3_191       : PASS
> > >     lmbench-3.0-a5    : PASS
> > > Thanks and best regards,
> > > Fujita and Fujisu members
> >



_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel