RTEMS 4.11.2 milestone date

[Chris Johns]

On 17/03/2017 10:58, Gedare Bloom wrote:
> On Thu, Mar 16, 2017 at 7:52 PM, Joel Sherrill <joel at rtems.org> wrote:
>>
>>
>> On Thu, Mar 16, 2017 at 6:22 PM, Chris Johns <chrisj at rtems.org> wrote:
>>>
>>> On 17/03/2017 07:48, Joel Sherrill wrote:
>>>>
>>>> On Thu, Mar 16, 2017 at 3:22 PM, Chris Johns <chrisj at rtems.org
>>>> <mailto:chrisj at rtems.org>> wrote:
>>>>
>>>>     On 17/03/2017 01:06, Joel Sherrill wrote:
>>>>
>>>>         On Thu, Mar 16, 2017 at 4:43 AM, Chris Johns <chrisj at rtems.org
>>>>         <mailto:chrisj at rtems.org>
>>>>         <mailto:chrisj at rtems.org <mailto:chrisj at rtems.org>>> wrote:
>>>>             If you have anything you would like in the release please
>>>>         make sure you
>>>>             have created a ticket and assigned it to the milestone.
>>>>
>>>>         Are the 4.11 tools buildable on the newer hosts that you and I
>>>>         have had troubles with the 4.12 tools? I think specifically
>>>>         qemu and gdb are troublesome.
>>>>
>>>>
>>>>     Good point, I am not sure. I will check FreeBSD 11.0 and raise a
>>>>     ticket if there is a problem.
>>>>
>>>>
>>>> FWIW assuming qemu on the master is the same as on 4.11, it built find
>>>> on a new CentOS 7 install.
>>>
>>>
>>> I do not know if they are the same. I will check what there is in 4.11 on
>>> FreeBSD 10.3 and 11.0.
>>>
>>>> But that's still at gcc 4.8. Something with a new
>>>> enough gcc will have trouble.
>>>
>>>
>>> If tickets are raised and patches supplied in time they may make RC1
>>> otherwise they will have to wait for the next release.
>>
>>
>> Agreed.
>>
>>>
>>>
>>>
>>>>
>>>> And you have the gdb powerpc issue on FreeBSD which might occur.
>>>>
>>>
>>> Is there a ticket for this? I am struggling to remember what the problem
>>> is. I do not use any PPC devices and it is not a primary focus for me.
>>>
>>
>> You couldn't build powerpc gdb because of the error symbol. There was
>> something

Ah yes something in the simulator is referencing a C symbol when it will 
now be a C++ symbol, ie missing include somewhere. I wonder if this is a 
clang vs gcc issue. This was using the latest gdb so I will see if it is 
a problem with the 4.11 gdb.

>> with max_align_t but I think that got resolved.
>>
>>
> https://sourceware.org/bugzilla/show_bug.cgi?id=21206

Yes, and this is fixed on 4.12 (master). I may need to backport the fix. 
It is isolated to FreeBSD hosts so a low impact fix.

>
>>>>
>>>> I don't know of any code issues. We all are good at filing those. :)
>>>
>>>
>>> I would like to package and release what we have at the set dates rather
>>> than hold the release while everything possible is fixed.
>>>
>>
>> +1
>>
> 4.12 should be on the table, too

This need some more discussion before a date is set so maybe another 
thread. It is an important question and raises some roadmap related 
questions.

Chris