[Cryptography] bounded pointers in C

[Tom Mitchell]

On Mon, Apr 21, 2014 at 9:46 AM, Nemo <nemo at self-evident.org> wrote:
> Patrick Chkoreff <patrick at rayservers.net> writes:
>
>> According to the C Language Reference Manual,
.....
>> (C Language Reference Manual, document 007-0701-150, Appendix A, Section
>> F.3.7 "Arrays and Pointers")
>
> "A" for effort, but that is not the C standard.
>
> http://csweb.cs.wfu.edu/~torgerse/Kokua/More_SGI/007-0701-150/sgi_html/pr02.html
>
> "This manual contains a summary of the syntax and semantics of the C
> programming language as implemented on SGI workstations."















>
> The C standard looks more like this:
>
> http://www.open-std.org/jtc1/sc22/wg14/www/docs/

There are three needed things here....
a) the standard
b) the product/ compiler documentation
c) test case to verify the functionality you depend on.

Without all three it is too easy to miss the mark.
The last is interesting in that it is the framework
for a bug report to get the product corrected.

One additional lump in the road is how to code
when a & b are at odds with each other.

And BTW the SGI compiler was open sourced way back and lives on.
PathScale and others use that source it as a product base and publish
source changes.  It is a highly optimizing
compiler (more so for Fortran) and at one point was
called a EKO for every known optimization ;)  The internals
of the compiler are all about efficient optimization and over
the years more than one chunk of code fell on its face
because the authors and maintainers depended on something
"clever" in common compilers but not in a standard or ......



-- 
  T o m    M i t c h e l l