> "mac.diff" patch work on some system, but probably incomplete. (Also,
> for platforms with 32-bit filesystem with POSIX options:
> _POSIX_V6_ILP32_OFF32 or _POSIX_V7_ILP32_OFF32, probably incorrect).
Related to this 32-bit file system problem. As far as I can see count
filed is used only to track every 128-th invocation of the functions in
question in c->key_meshing case. And the concern is that it will
incorrectly handle overflows. Is it correctly understood? In such case
wouldn't it be more appropriate to do exactly what it's supposed to do,
i.e. trace every 128-th invocation, e.g.
if (c->count&&c->key_meshing && c->count%1024==0)
{
cryptopro_key_meshing(&(c->cctx),iv);
c->count=1024;
}
...
c->count+=8;
In this case it would be possible to declare count as
platform-independent [and therefore preferred] unsigned int. Well, one
can even do unsigned short...
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List [email protected]
Automated List Manager [email protected]
