Greg Harrison (harrisog++at++geocities.com)
Wed, 16 Dec 1998 19:11:09 -0500
>
> There are no plans to move Performer to a pthreads base. To do so would
> break just about every application out there. Pthreads assumes implicit
> sharing and contains no support for the explicit shared memory model
> adopted by Performer. We'd still have to use arenas, but as I already
> stated, this doesn't work with pthreads.
>
> "Fixing" this would require a fundamental change in the way at least the
> C library works for multi-threaded applications and the way
> multi-threaded applications are scheduled by the kernel. This
> incompatibility won't be fixed.
I accept the inherent problems with changing the inner workings of Performer
or trying to get them to peacefully co-exist with pthreads, but what about
making Farenheight Scene Graph and its 'add-on modules' (that will
eventually replace Performer) be pthreaded? If FSG is to be a joint MS-SGI
standard this would appear to be a good/necessary thing, since MS is
certainly not going to implement sproc. Anyone thought this one out already
at SGI eng?
Greg Harrison
Air Force Institute of Technology
This archive was generated by hypermail 2.0b2 on Wed Dec 16 1998 - 16:14:53 PST