Re: crypto_memset (was: Re: Zero it if you're going to copy it out.)

[Jean-Yves Migeon <jeanyves.migeon%free.fr@localhost>]

On 22.07.2012 23:15, Matthew Mondor wrote:
> On Sun, 22 Jul 2012 19:37:26 +0200
> Alan Barrett <apb%cequrux.com@localhost> wrote:
> 
>>      errno_t
>>      memset_s(void *s, rsize_t smax, int c, rsize_t n);
> 
> Is there an issue if using something such as the following:
> 
> void *explicit_memset(void *, size_t, int);
> void explicit_bzero(void *, size_t);
> 
> To remain compatible with memset(3)/bzero(3) (or macros
> EXPLICIT_MEMSET, EXPLICIT_BZERO with the same signature)?

I agree, and they can be used in future implementations of memset_s()
when adding runtime constraints.

Using memset_s() (in a manner as expected by the standard)  requires
more heavy code changes. Having explicit_memset/bzero functions that
respect memset/bzero prototypes is still a plus IMHO.

My 2 cents, for what they're worth...

-- 
Jean-Yves Migeon
jeanyves.migeon%free.fr@localhost