You wrote (in your message from Tue 10)
 > 
 > Why does the internal malloc of XFree make the list grow to 157 entries in
 > one day when the other allocation method behaves a lot better?  

Because the XFree86 internal allocator uses mmap(/dev/zero) instead of sbrk()
to get memory from the system, in order to be able to release
arbitrary chunks when they are freezed...

 > And why do
 > things become so slow when the list length grows to 150+ entries?

Good question. Nobody knows if other OS don't see the problem because
no one runs applix on them or if they are more efficient in handling
lots of mmap() calls.


					Matthieu