On Thursday, 97/03/27, Denis Tonn wrote to David Noon about "Trap
0005" as follows:

DT>  It *is* possible for compiler runtime to automaticly include an 
DT> exception handler and do "sparse stack allocation" transparently,
DT> but I don't see how the linker would have any control over the
DT> "stack size" settings for this case. 

Hi Denis,

The above is absolutely correct, and I think it applies here.

As I recall, Murray's program was a multi-threaded PL/I program, where
the stack for the secondary thread is allocated by the ATTACH statement
and, therefore, controlled by the PL/I RTL, not the linker. Since the
compiler defaults to generating stack probes it works a treat.

Regards

Dave


 * KWQ/2 1.2i * "... and Morgo, as the friendly drelb." -- G. Owen

--- Maximus/2 3.01