Project proposal: Remove the Permanent Generation

[Jon Masamitsu]

Tomas.

I put up a second version that I think is slightly
better (in terms of a cleaner webrev).

http://cr.openjdk.java.net/~jmasa/6964458/webrev.01

The .00 version had code in it that wouldn't compile
(from patches that failed) and will be delete in a day or two.

Jon

On 06/04/12 12:06, Jon Masamitsu wrote:
> Tomas,
>
> I redid the webrev (overwrote the previous one).  The changes
> for attachListerner_solaris.cpp that you saw previously are not
> in the new webrev (i.e., not difference between hotspot-metadata
> and hotspot-gc).  When you get a chance, could you take
> a look at a few more files as a second sanity check?
>
> Thanks.
>
> Jon
>
>
> On 06/02/12 02:47, Tomas Hurka wrote:
>> Hi Jon,
>>
>> On 2 Jun 2012, at 00:48, Jon Masamitsu wrote:
>>
>>> This is the webrev for the latest sync with hotspot-gc.   Most 
>>> changes are cleanups
>>> or bug fixes.  One large change is the removal of the (unused for a 
>>> while now)
>>> perm gen from the generational hierarchy.
>>>
>>> http://cr.openjdk.java.net/~jmasa/6964458/webrev.00/
>>>
>>> As we've said the pervasive renaming of data structures will be done
>>> just before we integrate in order to ease the burden of the syncs.
>>>
>>> Now's the time to look at the code in your area of interest.
>> I just took a brief look at the webrev above and I was surprised by 
>> the changes in
>> attachListener_solaris.cpp<http://cr.openjdk.java.net/~jmasa/6964458/webrev.00/src/os/solaris/vm/attachListener_solaris.cpp.udiff.html> 
>>
>> and 
>> os_windows.cpp<http://cr.openjdk.java.net/~jmasa/6964458/webrev.00/src/os/windows/vm/os_windows.cpp.udiff.html>.
>> It looks like they have nothing to do with the removal of the 
>> Permanent Generation.
>>
>> Bye,
>> -- 
>> Tomas Hurka<mailto:tomas.hurka at oracle.com>
>> NetBeans Profiler http://profiler.netbeans.org
>> VisualVM http://visualvm.java.net
>> Software Developer
>> Oracle, Praha Czech Republic
>>
>>
>>