Allow HotSpot to build on Linux 3.0+

[Erik Trimble]

On 7/28/2011 5:31 PM, Keith McGuigan wrote:
>
> On Jul 28, 2011, at 8:28 PM, Dr Andrew John Hughes wrote:
>
>> On 10:45 Thu 28 Jul     , Keith McGuigan wrote:
>>>
>>> On Jul 28, 2011, at 9:54 AM, Dr Andrew John Hughes wrote:
>>>
>>>> On 09:17 Thu 28 Jul     , Keith McGuigan wrote:
>>>>>
>>>>> Ok, thanks.  Code looks good to me then.
>>>>>
>>>>
>>>> Ok, can I push this or do you still need to do this via JPRT?
>>>>
>>>> Either way, I need a bug ID please.
>>>
>>> Yes, this will need to be pushed via JPRT (as all hotspot pushes
>>> must).  I can create an ID for this too.  One more review would be
>>> good too.
>>>
>>
>> Ok.  Going forward, non-Oracle OpenJDK contributors either need to be
>> able to push via JPRT or some other method needs to be used which is
>> accessible by all.
>
> We have top men working on it now.  Top men.
>
> (At least, I'm pretty sure we do...)
>
> -- 
> - Keith

I swear, I though I heard my name mentioned. Someone talking about me 
here? You talkin' to me?

:-)

Obviously, making a JPRT-like system available to *all* authorized 
committers is something that's definitely on the agenda - it would 
benefit everyone in terms of turnaround and increased pace of 
development, not to mention a better sense of community.

Now that IBM is onboard, I'd also hope to start seeing direct 
contributions from them.

There's several big obstacles I can see:

     (a) A better Bug system needs to be in place, one that can meet the 
needs to both the OpenJDK community, and the various sponsors
     (b) A well-thought-out process for accepting both new contributors 
*and* for properly evaluating and integrating fixes from committers has 
to be worked out
      (c) Proper hardware to equip the build/test system has to be 
purchased or allocated. Part of this step is to make it clear what the 
requirements of such a system are, and then get folks to contribute 
appropriate resources in line with what the system needs.  I think it's 
fair to expect that much of this system needs to come from folks other 
than Oracle, especially since Oracle most likely would be tasked with 
maintaining and servicing it (and paying for the ongoing support and 
running costs).


Without speaking out-of-turn, or making any Official Oracle Statement or 
Commitment, (a) is well underway, (b) is being discussed (and should 
likely be brought out into the OpenJDK discussion lists sometime here), 
and (c) hasn't even been looked at.



-- 
Erik Trimble
Java Platform Group Infrastructure
Mailstop:  usca22-317
Phone:  x67195
Santa Clara, CA
Timezone: US/Pacific (UTC-0800)