«

"Kevin Grigorenko" <> wrote in message
news:bm2k9g$mki$...
> Hello,
>
> I couldn't find an obvious answer to this in the FAQ. My basic question,
> is: Is there any difference in allocating on the heap versus the stack?
If
> heap or stack implementation is not part of the standard, then just
> disregard this question. Here's some questions I'm confused about, and if
> you can add anything else, please do so!

The c++ standard does not define "stack" in the context of the memory used
to hold automatic variables, other than stating that it holds them.

The current computer architectures, that I am familiar with, use a contguous
memory area, with a register assigned to indicate where storage can be
assigned from that area. The portion of this area used within a function is
typically called a "stack frame" and holds automatic variables plus function
return information.

This is not the only way to manage automatic type storage. The standard
convention for IBM 360... mainframes was for storage to be obtained and
released as needed via operating system calls.

The computer releated meaning of stack was derived by analogy from the stack
of clean empty trays (often in a spring loaded container so that only the
top tray could be taken) commonly used in cafeterias.

David Sachs wrote:
[...]
> This is not the only way to manage automatic type storage. The standard
> convention for IBM 360... mainframes was for storage to be obtained and
> released as needed via operating system calls.

Modern mainframes can operate in kinda "hybrid mode". Storage for stack
increments (it used for stack frames) is obtained ('getmained') and
released "as needed" but you have usual "frames stuff" and guard pages
within stack increment. And you can specify "KEEP|FREE" disposition.

http://publibz.boulder.ibm.com/epubs/pdf/ceea2130.pdf
http://publibz.boulder.ibm.com/epubs/pdf/ceea3130.pdf

regards,
alexander.