Segmentation fault with NUMA aware ParallelScavange GC

Lokesh Gidra lokesh.gidra at yahoo.com
Fri Jul 29 05:39:03 UTC 2011




>
>Hi, Lokesh,
>
>Could you please check if 6u25 works?


You mean kernel version 2.6.25 right?

BTW, what kind of bug are you suspecting?



>
>Thanks you!
>igor
>

Thank you!
Lokesh

>On 7/28/11 3:36 PM, Lokesh Gidra wrote:
>> Hi Igor,
>>
>> Yes, its a Linux box with 2.6.35-22-server kernel version. Its running a
>> 64-bit ubuntu server.
>>
>> Yes it does reproduce with -XX:-UseLargePages option.
>>
>> Please let me know if you need any other information.
>>
>> Regards,
>> Lokesh
>>
>>     ------------------------------------------------------------------------
>>     *From:* Igor Veresov <igor.veresov at oracle.com>
>>     *To:* hotspot-gc-dev at openjdk.java.net
>>     *Sent:* Thursday, 28 July 2011 11:14 PM
>>     *Subject:* Re: Segmentation fault with NUMA aware ParallelScavange GC
>>
>>     Hi!
>>
>>     This is Linux, right?
>>     What kernel version do you have?
>>     Does it reproduce with -XX:-UseLargePages?
>>
>>     igor
>>
>>     On 7/28/11 2:23 AM, Lokesh Gidra wrote:
>>      > Hi,
>>      >
>>      > I am trying to use the NUMA-aware implementation of
>>     ParallelScavenge GC
>>      > of OpenJDK7 on a 8-node machine. Almost everytime I run it, I get a
>>      > segmentation fault trying to access a page aligned address. The
>>     test is
>>      > with DaCapo benchmarks which works fine with default setup.
>>     Although the
>>      > address lies in the range of one of the lgrp's but in the memory
>>     maps,
>>      > its not mapped anywhere. Can anyone please explain me what could
>>     be the
>>      > problem. Please let me know if some additional information is needed.
>>      >
>>      >
>>      > Best Regards,
>>      > Lokesh
>>
>>
>>
>
>
>
>



More information about the hotspot-gc-dev mailing list