Software Assurance Reference Dataset

bearophile via Digitalmars-d digitalmars-d at puremagic.com
Thu Jun 26 02:50:12 PDT 2014


Walter Bright:

> In general, stack overflow checking at compile time is the 
> halting problem. It needs a runtime check.

There are several systems, including SPARK, that perform a 
conservative and apparently acceptable stack overflow check at 
compile time. If you don't agree with what I've written in my 
post, then please give a more detailed answer to the points I've 
written above.


> Stack overflows are not safety problems when a guard page is 
> used past the end of the stack.

It's not a safety problem in Erlang/Rust, because those languages 
are designed to manage such failures in a good way. In most other 
languages it's a "safety" problem, if your program execution has 
some importance.

Bye,
bearophile


More information about the Digitalmars-d mailing list