[Guy Harris <guy@xxxxxxxxxxxx>]

ronnie sahlberg wrote:

> since guids are "never" reused and also since we might preload the 
> mapping table from a file  
> that is why i recently changed the se_tree  code  to make it trivial to 
> add a
> pe_tree_...   type.   PErmanent tree   a tree with storage scope that is 
> never released.

Should we have versions of the routines that allocate ep_ and se_ things 
that take, as an argument, an "emem_chunk_t *", and allocate from that 
storage pool, and have routines to create and destroy storage pools?