[SciPy-user] shared memory machines

Robert Kern robert.kern@gmail....
Mon Feb 9 12:28:59 CST 2009


On Mon, Feb 9, 2009 at 12:24, Sturla Molden <sturla@molden.no> wrote:
> On 2/9/2009 12:20 PM, Sturla Molden wrote:
>
>> I have been working on that. Basically using what Robert Kern posted a
>> while ago and ripping out some code from multiprocessing's heap object.
>
> Here is a first draft. The ftok issue is not fixed.
>
> I am not sure if Robert Kern's use of copy_reg affects ndarrays in general,
> or just the ones we create here.

It affects ndarrays in general. The reduce function should ideally be
written to detect whether the ndarray is shared, or is a view
eventually leading back to a shared ndarray, or is just a regular
ndarray.

-- 
Robert Kern

"I have come to believe that the whole world is an enigma, a harmless
enigma that is made terrible by our own mad attempt to interpret it as
though it had an underlying truth."
  -- Umberto Eco


More information about the SciPy-user mailing list