Re: [Lit.] Buffer overruns

From: Andrew Swallow (am.swallow_at_btopenworld.com)
Date: 02/02/05 

Date: Wed, 2 Feb 2005 19:14:42 +0000 (UTC)

David Wagner wrote:

> Trevor L. Jackson, III wrote:
>
>>This is an artifact of the quest for provability. In most cases C
>>esoterica is unnecessary to the prevention of buffer management errors.
>> But when your start stressing completeness, perfection, and/or
>>mathematical proofs you drag in all the minuscule details that are not
>>relevant to the errors under consideration, but might confound an
>>attempt to achieve perfection or prove the absence of flaws.
>
>
> Yes, I agree with that. Any theorem prover for C had better know
> every little last detail of the specification.

Or ban use of that part of the specification. Making things provable
may mean placing restrictions on what we do.

Andrew Swallow

Loading