Minimum stack size in java6 and java7

[David Holmes]

On 12/06/2012 6:57 AM, Dean Long wrote:
> Could be related to this change:
>
> http://mail.openjdk.java.net/pipermail/net-dev/2009-May/000868.html

Which lead to:

http://mail.openjdk.java.net/pipermail/hotspot-runtime-dev/2012-February/003092.html

though this only affects 6u32 onwards, 7u4 onwards and 8.

David

> dl
>
> On 6/6/2012 8:58 AM, Brandon Williams wrote:
>> Hello,
>>
>> I'm a member of the Apache Cassandra project, and we're having an
>> issue upgrading to java7:
>>
>> https://issues.apache.org/jira/browse/CASSANDRA-4275
>>
>> Could someone explain why the Xss minimum is so high? It seems like
>> if we're willing to
>> live without recursion we should be allowed to set it much smaller than
>> the 128K default (java6) or 160 (java7). What motivated increasing the
>> minimum to 160K for java7?
>>
>> We could accept the 32K increase to the java7 minimum, but having to
>> double the stack size as we currently do seems a little unreasonable.
>>
>> Thanks in advance,
>> -Brandon