Pradeep,
        Any word on this?
Thanks,
Jeff

Jeff Burke wrote:
> Pradeep,
>       Yes the machines it fails to build on are numa systems. Here are  the
> numa stats from one of the systems that failed.
>> ***************** NUMA STAT*****************
>>                            node0           node1           node2           
>> node3
>> numa_hit                  192430          120743          120737          
>> 123810
>> numa_miss                      0               0               0             
>>   0
>> numa_foreign                   0               0               0             
>>   0
>> interleave_hit              1780            1840            1772            
>> 1829
>> local_node                191476          118522          117800          
>> 120935
>> other_node                   954            2221            2937            
>> 2875
>> ***************** NUMA SHOW *****************
>> policy: default
>> preferred node: current
>> physcpubind: 0 1 2 3 4 5 6 7 
>> cpubind: 0 1 2 3 
>> nodebind: 0 1 2 3 
>> membind: 0 1 2 3 
>> ***************** NUMA HARDWARE *****************
>> available: 4 nodes (0-3)
>> node 0 size: 16215 MB
>> node 0 free: 14905 MB
>> node 1 size: 16384 MB
>> node 1 free: 15499 MB
>> node 2 size: 16384 MB
>> node 2 free: 15490 MB
>> node 3 size: 12287 MB
>> node 3 free: 11602 MB
>> node distances:
>> node   0   1   2   3 
>>   0:  10  20  20  20 
>>   1:  20  10  20  20 
>>   2:  20  20  10  20 
>>   3:  20  20  20  10 
>       
>       When you tested your changes was that from a clean tree? Or was it
> previously used to build? Looking at the error it can't link in libnuma.so.
> 
> Thanks,
> Jeff
> 
> Pradeep K Surisetty wrote:
>> Hi Jeff check weather the machine is a NUMA machine or not. Its work fine 
>> with me
>> on different machines. i tried on
>> SLES10SP1 x86-64, RHEL5 x86-64, RHEL5 i386, SLES10 SP1 PPC.
>>
>> Generally /sys/devices/system/node contains some folders with the name of
>> nodes present in that machine. For exapmle node0,node1...etc
>> If nodes are more than one , its a NUMA machine otherwise NON NUMA machine.
>>
>> Could you provide me , what are all the folders(nodes) are present in
>> '/sys/devices/system/node' directory.
>>
>>
>> Regards
>> Pradeep
>>
>>
>> Subrata Modak wrote:
>>> On Tue, 2008-01-08 at 17:34 -0500, Jeff Burke wrote:
>>>> Subrata,
>>>>           Was the attached patch.gz file tested successfully?
>>>> After applying the patch I still get the following error. I am running
>>>> on a RHEL5.1 x86_64 system.
>>> I thought Rishiś problem was solved as he acked this Patch. Pradeep, can
>>> you please address Jeffś issue ?
>> Pradeep,
>>              Any word one this? Should I revert and apply the patch from
>> Subrata
>> that skips building numa?
>>
>> -Jeff
>>> -Subrata
>>>
>>>>> patch -p0 <numa-build.patch
>>>>> patching file ltp-full-20071231/testcases/kernel/numa/numa01.sh
>>>>> patching file ltp-full-20071231/testcases/kernel/numa/numa_node_size.c
>>>>> patching file ltp-full-20071231/testcases/kernel/numa/test.sh
>>>> < --- snip --->
>>>>> make[3]: Entering directory
>> `/mnt/tests/kernel/distribution/ltp/20071231/ltp-full-20071231/testcases/kernel'
>>
>>>>> make[4]: Entering directory
>> `/mnt/tests/kernel/distribution/ltp/20071231/ltp-full-20071231/testcases/kernel/numa'
>>
>>>>> cc -Wall  -I../../../include -Wall     numa_node_size.c  -L../../../lib
>> -lltp  -L../../../lib  -lnuma -o numa_node_size
>>>>> numa_node_size.c: In function ‘hardware’:
>>>>> numa_node_size.c:49: warning: implicit declaration of function
>> ‘numa_max_node’
>>>>> numa_node_size.c:54: warning: implicit declaration of function
>> ‘numa_node_size’
>>>>> numa_node_size.c: In function ‘main’:
>>>>> numa_node_size.c:63: warning: implicit declaration of function
>> ‘numa_available’
>>>>> /usr/bin/ld: cannot find -lnuma
>>>>> collect2: ld returned 1 exit status
>>>>> make[4]: *** [numa_node_size] Error 1
>>>>> make[4]: Leaving directory
>> `/mnt/tests/kernel/distribution/ltp/20071231/ltp-full-20071231/testcases/kernel/numa'
>>
>>>>> make[3]: *** [all] Error 2
>>>>> make[3]: Leaving directory
>> `/mnt/tests/kernel/distribution/ltp/20071231/ltp-full-20071231/testcases/kernel'
>>
>>>>> make[2]: *** [all] Error 2
>>>>> make[2]: Leaving directory
>> `/mnt/tests/kernel/distribution/ltp/20071231/ltp-full-20071231/testcases'
>>>>> make[1]: *** [all] Error 2
>>>>> make[1]: Leaving directory
>> `/mnt/tests/kernel/distribution/ltp/20071231/ltp-full-20071231'
>>>>> make: *** [build] Error 2
>>>> Thanks,
>>>> Jeff
>>>>
>>>> Subrata Modak wrote:
>>>>> Hi All,
>>>>>
>>>>> Please find the fix for the NUMA problem reported by Rishi.
>>>>>
>>>>> --Subrata
>>>>>
>>>>> -------- Forwarded Message --------
>>>>> From: [EMAIL PROTECTED]
>>>>> To: [EMAIL PROTECTED]
>>>>> Subject: Re: [Fwd: Re: [LTP] ltp-full-20071231: Build Failure on Numa
>>>>> node enable machine]
>>>>> Date: Wed, 2 Jan 2008 08:42:45 -0500
>>>>>
>>>>> Hi subratha/rishi
>>>>>
>>>>> Thank you rishi
>>>>> I have done the changes and patch is attached
>>>>>
>>>>> Regards
>>>>> pradeep
>>>>>
>>
>>
> 
> 
> -------------------------------------------------------------------------
> Check out the new SourceForge.net Marketplace.
> It's the best place to buy or sell services for
> just about anything Open Source.
> http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
> _______________________________________________
> Ltp-list mailing list
> [email protected]
> https://lists.sourceforge.net/lists/listinfo/ltp-list


-------------------------------------------------------------------------
This SF.net email is sponsored by: Microsoft
Defy all challenges. Microsoft(R) Visual Studio 2008.
http://clk.atdmt.com/MRT/go/vse0120000070mrt/direct/01/
_______________________________________________
Ltp-list mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/ltp-list

Reply via email to