Re: Tests in a software release

["Dmitry A. Kazakov" <mailbox@dmitry-kazakov.de>]

On 25/10/2017 21:30, Victor Porton wrote:
> Do you agree that a release (that is software for which debugging was
> finished) should have integer overflow tests but not array out of bounds
> tests (because array out of bounds is always a programming error, but
> integer overflow may happen in an innocent program)?

Neither check can be disabled.

1. Both are contracted behavior. Disabling checks breaks the contract 
unless proven otherwise and *statically*. In the latter case it is an 
optimization issue to remove the check when that does not change the 
program behavior.

2. Any error must be detected as early as possible, so better bounds 
check now than a memory corruption check later. Therefore no difference 
if check failure were expected or not.

-- 
Regards,
Dmitry A. Kazakov
http://www.dmitry-kazakov.de