[Mpi3-rma] Alternative Proposal for Shared Memory Support

[Rajeev Thakur]

You could, but it starts looking more and more like RMA. The purpose of SHM_allocate was to provide an easy way to allocate shared memory and access it directly with memory references, instead of the user having to do that himself using POSIX or System V shared memory allocation functions. Now if he has to read the whole RMA chapter for that... :-)

Rajeev


On Mar 16, 2011, at 8:22 AM, James Dinan wrote:

> Hi Rajeev,
> 
> Why not use lock/unlock?  Lock_all seems like a good fit here.
> 
> ~Jim.
> 
> On 3/16/11 8:09 AM, Rajeev Thakur wrote:
>> Right. Some things that need to be touched up are that flush can only be called within lock-unlock currently. Also flush takes a target rank, which is the rank passed to the put/get. Flush_all has the notion of all ranks used in preceding puts/gets. Such things would need to be addressed.
>> 
>> Rajeev
>> 
>> 
>> On Mar 15, 2011, at 11:22 PM, Torsten Hoefler wrote:
>> 
>>> On Tue, Mar 15, 2011 at 09:39:24PM -0500, Rajeev Thakur wrote:
>>>> That's what I meant in my first mail. We need to say whether this example will work or not.
>>>> 
>>>> A=10
>>>> barrier         barrier
>>>>                    x = A
>>>> 
>>>> If not, which synchronization functions need to be used and how? The
>>>> current text does not automatically cover this example.
>>> Right, I never claimed that the proposal is complete :-). It's basically
>>> a quick poll to see if there are major flaws and I should retract or if
>>> it is worth discussing at the Forum (to flesh out such things like you
>>> mentioned).
>>> 
>>> To answer your question: I think this should either be:
>>> 
>>> A=10
>>> MPI_Win_flush
>>> barrier           barrier
>>>                  x=A
>>> 
>>> This would basically insert a membar in MPI_Win_flush which makes
>>> sure that A is committed before the barrier completes (a barrier flag is
>>> set and read).
>>> 
>>> This fits our current semantics where we suggest that win_flush can be
>>> used for ordering. Keep in mind that it's still the unified model, i.e.,
>>> the value would commit eventually, the flush is just needed to enforce
>>> ordering with regards to the barrier.
>>> 
>>> Thanks&  Best,
>>>  Torsten
>>> 
>>> --
>>> bash$ :(){ :|:&};: --------------------- http://www.unixer.de/ -----
>>> "A couple of months in the laboratory can frequently save a couple of
>>> hours in the library." - Westheimer (contemporary)
>>> _______________________________________________
>>> mpi3-rma mailing list
>>> mpi3-rma at lists.mpi-forum.org
>>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-rma
>> 
>> 
>> _______________________________________________
>> mpi3-rma mailing list
>> mpi3-rma at lists.mpi-forum.org
>> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-rma
> 
> _______________________________________________
> mpi3-rma mailing list
> mpi3-rma at lists.mpi-forum.org
> http://lists.mpi-forum.org/mailman/listinfo.cgi/mpi3-rma