About time for b21?

[Joe Darcy]

Dr Andrew John Hughes wrote:
> On 17:46 Fri 22 Oct     , Joe Darcy wrote:
>   
>> Dr Andrew John Hughes wrote:
>>     
>>> On 15:28 Mon 18 Oct     , Dr Andrew John Hughes wrote:
>>>   
>>>       
>>>> On 18:04 Thu 14 Oct     , Joe Darcy wrote:
>>>>     
>>>>         
>>>>> Hello.
>>>>>
>>>>> With Andrew's sync of HS19 into OpenJDK 6 and the recent batch of 
>>>>> security fixes coming real soon now, I think it is about time to have 
>>>>> b21.  I'd hold b21 for the remaining bug fixes pushed to HS19 that are 
>>>>> not quite promoted yet.
>>>>>
>>>>> Are there any other changes that should be in b21?
>>>>>
>>>>> -Joe
>>>>>       
>>>>>           
>>>> http://mail.openjdk.java.net/pipermail/jdk6-dev/2010-September/002008.html is
>>>> still awaiting review.
>>>>     
>>>>         
>> Ah, sorry for letting that fall through the cracks and for the belated 
>> reply -- reminder email just sent.
>>
>>     
>
> Thanks.
>   

There are some technical issues here I'm discussing with other members 
of the javac team.

>>>> Two of the three suggestions on http://mail.openjdk.java.net/pipermail/jdk6-dev/2010-August/001954.html
>>>> are also still pending (haven't had time to follow them up).
>>>>
>>>> 6678385 did go in, so 6853592 could do with also going in.  See:
>>>> http://mail.openjdk.java.net/pipermail/distro-pkg-dev/2010-October/010509.html
>>>> 6853592: VM test nsk.regression.b4261880 fails with "X Error of failed request: BadWindow" inconsistently.
>>>>     
>>>>         
>> 6853592 is approved to be backported in b21.
>>
>>     
>
> Pushed.
>   

Thanks.

>>>> I'll post webrevs for these three if they sound ok for OpenJDK6 b21.
>>>>
>>>> I was going to look at pushing the XRender/DirectX backport upstream, but I think that
>>>> could do with more soak time in IcedTea6.  We'll do that in b22.
>>>>     
>>>>         
>> That approach sounds sensible.
>>
>>     
>
> Ok, first webrev:
>
> http://cr.openjdk.java.net/~andrew/6668231/webrev.01/
>
> 6668231: Presence of a critical subjectAltName causes JSSE's SunX509 to fail trusted checks
> Summary: make the critical extension known to end entity checker.
> Reviewed-by: wetmore, mullan
>
> Ok for push?  The binary keystores are copied as-is from OpenJDK7.
>
>   

I approve 6668231being backported subject to Andrew (Fan)'s verification 
it is fine for this release.  Andrew (Fan), is there any issue with this 
fix going into OpenJDK 6?

Thanks,

-Joe