>>A key reason for the last is so that someone can use LSQA for an SRB 
that 
>>is not associated with a TCB, for which you might not want private 
storage 
>>obtained by the SRB to be freed just because some task terminates.

>Seems like a weak reason. Are there other ways SRBs are protected from 
poor
>subpool choices?

It is far from a weak reason. It is an extremely strong reason.
If this is a "poor subpool choice" then what would be a good subpool 
choice? 

Any task-related subpool will have its storage freed when that task 
terminates. An SRB *can* (but need not) be tied to a particular task. When 
it is, it can rely on a task-owned subpool associated with that task (or a 
parent of that task). When it is not, it ought not to rely on such.

The SRB, in that case, needs to use a subpool that has no task 
association. That is either LSQA or common. If anything, I'd say that 
common would be a "poor subpool choice" if addressability from multiple 
address spaces is not needed.

Peter Relson
z/OS Core Technology Design


----------------------------------------------------------------------
For IBM-MAIN subscribe / signoff / archive access instructions,
send email to [email protected] with the message: INFO IBM-MAIN

Reply via email to