Re: The cost of pthread-mutexes or corresponding...

[Sean Walton]

Richard Levitte - VMS Whacker wrote:

> Using dynamic locks would create fragmentation of the heap, but there
> are solutions to that, to some level of intricacy.  However, there's
> something else that disturbs me a bit, and it's the thought that locks
> may be a limited resource and can't therefore be littered all over the
> heap without some performance hit or something else.

You could a centralized locking system instead of heap.  Imagine 
creating a custom mutex table that is fixed (or even dynamic).  As each 
mutex is allocated and freed, it would come from this space.  Since it 
is centralized, it does not fragment the heap, and sections of it could 
be swapped out of RAM without intervention from the library (please note 
that this is demand paging -- something Win2K has only begun to play 
with).  Of course this would mean writing a new set of mutex functions...

-Sean Walton

______________________________________________________________________
OpenSSL Project                                 http://www.openssl.org
Development Mailing List                       [EMAIL PROTECTED]
Automated List Manager                           [EMAIL PROTECTED]