Re: [PATCH v4 2/5] crypto: AF_ALG: add AEAD support

[Stephan Mueller <smueller@chronox.de>]

Am Freitag, 5. Dezember 2014, 23:46:06 schrieb Herbert Xu:

Hi Herbert,

> > +static struct proto_ops algif_aead_ops = {
> > +	.family		=	PF_ALG,
> > +
> > +	.connect	=	sock_no_connect,
> > +	.socketpair	=	sock_no_socketpair,
> > +	.getname	=	sock_no_getname,
> > +	.ioctl		=	sock_no_ioctl,
> > +	.listen		=	sock_no_listen,
> > +	.shutdown	=	sock_no_shutdown,
> > +	.getsockopt	=	sock_no_getsockopt,
> > +	.mmap		=	sock_no_mmap,
> > +	.bind		=	sock_no_bind,
> > +	.accept		=	sock_no_accept,
> > +
> > +	.release	=	af_alg_release,
> > +	.sendmsg	=	aead_sendmsg,
> > +	.sendpage	=	aead_sendpage,
> > +	.recvmsg	=	aead_recvmsg,
> > +	.poll		=	aead_poll,
> > +	.setsockopt	=	aead_setsockopt,
> 
> No it should go into the parent setsockopt.  Perhaps add a setsockopt
> to af_alg_type in order to keep this out of the generic code.

What about adding a setauthsize to af_alg_type that is called from the parent 
setsockopt? I would like to keep all user space interfaces as close together 
as possible. With a new setsockopt callback, I fear that user space interface 
handling may be scattered around.

Currently I am simply thinking about the following:

partent setsockopt:

        case ALG_SET_AEAD_AUTHSIZE:
                if (sock->state == SS_CONNECTED)
                        goto unlock;
                if (!type->setauthsize)
                        goto unlock;
                err = type->setauthsize(ask->private, optlen);

af_alg_type:

int (*setauthsize)(void *private, unsigned int authsize);

-- 
Ciao
Stephan