dual processor

Robin Becker robin at reportlab.com
Mon Sep 12 15:54:33 CEST 2005


Robin Becker wrote:
> Robin Becker wrote:
> 
>>Paul Rubin wrote:
>>
> 
> 
>>>This module might be of interest:  http://poshmodule.sf.net
>>>
>>
>>It seems it might be a bit out of date. I've emailed the author via sf, but no 
>>reply. Does anyone know if poshmodule works with latest stuff?
> 
from the horse's mouth comes confirmation that POSH is in a Norwegian Blue condition

from steffenv at cs.uit.no

> Hi,
> 
> Sorry for the late reply; You're right in that the project is close to dead.
> That's simply because I haven't had the time and motivation to maintain it,
> or rather, to improve it to the point where it's more usable and bug-free.
> Currently, I view POSH as a proof of concept that transparent access to
> Python objects allocated in shared memory is feasible.  The main limitation
> of POSH is that it relies on the semantics of fork() to preserve the exact
> same memory layout in parent and child processes; porting POSH to Windows or
> other platforms without a fork() system call would be a challenge.  Besides
> that, the implementation could use a more efficient memory allocator, and
> some features that sacrifice performance for flexibility probably need to be
> rethought.   It also badly needs a shared-memory blocking queue
> implementation to support common programming models involving queues, and a
> portable lock implementation for more architectures.  These latter
> improvements should be rather straightforward, but I don't know how to avoid
> the fundamental limitation of relying on fork().
> 
> Feel free to repost these comments on comp.lang.python if you want.
> 
> Cheers,
> Steffen


-- 
Robin Becker




More information about the Python-list mailing list