Hi y'all,
Quick question about RSA_new() I wanted to run by interested parties;
namely, RSA_new() does a bunch of initialisations of a new RSA structure
prior to calling the RSA_METHOD's "init()" handler to do any tweaking to
the new RSA structure that it may choose to do. However, the final few
lines go something like this;
if ((meth->init != NULL) && !meth->init(ret))
{
OPENSSL_free(ret);
ret=NULL;
}
else
CRYPTO_new_ex_data(rsa_meth,ret,&ret->ex_data);
return(ret);
Why doesn't CRYPTO_new_ex_data(...) get called *prior* to the RSA_METHOD's
"init()" handler - isn't the ex_data member of an RSA structure a perfect
place for a custom method place information? By this scheme, the ex_data
member of the structure is uninitialised when the init() handler is
called, and even if the handler initialised it, it would be erased by the
following CRYPTO_new_ex_data() call.
Would fixing this break anything that anyone is aware of? (Ie. putting the
CRYPTO_new_ex_data() prior to the meth->init() call).
Cheers,
Geoff
______________________________________________________________________
OpenSSL Project http://www.openssl.org
Development Mailing List [EMAIL PROTECTED]
Automated List Manager [EMAIL PROTECTED]
