PF> My method is more simple -- I 'stage' every 
 PF> read/write in a static local
 PF> buffer (ie: avoid dynamic memory allocs,  which are

 CS> Do you have a seperate static buffer for every named pipe,
 CS> or is a single buffer shared?  How big do you make the

I use one buffer in my PipeWrite() function,  which is allocated at init
time with DosAllocMem(... OBJ_TILED ...).

This implies that my function either serialize itself,  or never be called
from more than one thread at a time.  If I wanted to get around this,  I'd
have to make sure each thread had its own stage buffer (no big deal,  but I
didn't need it).


 CS> Distributing subsystems on networks is still a few months away,but by 
 CS> that time we might even consider some other IPC 
 CS> mechanism such as TCP/IP or DSOM.  I don't have any 
 CS> experience with DSOM yet,however -- do you know how it

Me either.


--- Maximus/2 3.00