From: Herbert Xu Subject: Re: [PATCH 1/2] crypto: cbc - Propagate NEED_FALLBACK bit Date: Mon, 27 Feb 2017 17:51:36 +0800 Message-ID: <20170227095136.GA31553@gondor.apana.org.au> References: <1488157399-3587-1-git-send-email-marcelo.cerri@canonical.com> <1488157399-3587-2-git-send-email-marcelo.cerri@canonical.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: "David S. Miller" , linux-crypto@vger.kernel.org, linux-kernel@vger.kernel.org To: Marcelo Henrique Cerri Return-path: Content-Disposition: inline In-Reply-To: <1488157399-3587-2-git-send-email-marcelo.cerri@canonical.com> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-crypto.vger.kernel.org On Sun, Feb 26, 2017 at 10:03:18PM -0300, Marcelo Henrique Cerri wrote: > When requesting a fallback algorithm, we should propagate the > NEED_FALLBACK bit when search for the underlying algorithm. > > This will prevents drivers from allocating unnecessary fallbacks that > are never called. For instance, currently the vmx-crypto driver will use > the following chain of calls when calling the fallback implementation: > > p8_aes_cbc -> cbc(p8_aes) -> aes-generic > > However p8_aes will always delegate its calls to aes-generic. With this > patch, p8_aes_cbc will be able to use cbc(aes-generic) directly as its > fallback. The same applies to aes_s390. > > Signed-off-by: Marcelo Henrique Cerri > --- > crypto/cbc.c | 20 ++++++++++++++------ > 1 file changed, 14 insertions(+), 6 deletions(-) > > diff --git a/crypto/cbc.c b/crypto/cbc.c > index bc160a3..7147842 100644 > --- a/crypto/cbc.c > +++ b/crypto/cbc.c > @@ -108,24 +108,32 @@ static void crypto_cbc_free(struct skcipher_instance *inst) > static int crypto_cbc_create(struct crypto_template *tmpl, struct rtattr **tb) > { > struct skcipher_instance *inst; > + struct crypto_attr_type *algt; > struct crypto_spawn *spawn; > struct crypto_alg *alg; > + u32 mask; > int err; > > err = crypto_check_attr_type(tb, CRYPTO_ALG_TYPE_SKCIPHER); > if (err) > return err; > > + algt = crypto_get_attr_type(tb); > + if (IS_ERR(algt)) > + return PTR_ERR(algt); > + > + mask = CRYPTO_ALG_TYPE_MASK | > + crypto_requires_off(algt->type, algt->mask, > + CRYPTO_ALG_NEED_FALLBACK); > + > + alg = crypto_get_attr_alg(tb, CRYPTO_ALG_TYPE_CIPHER, mask); > + if (IS_ERR(alg)) > + return PTR_ERR(alg); > + > inst = kzalloc(sizeof(*inst) + sizeof(*spawn), GFP_KERNEL); > if (!inst) > return -ENOMEM; You're leaking alg if the kzalloc of inst fails. Easiest fix would be to do crypto_get_attr_alg after the kzalloc as is the status quo. Thanks, -- Email: Herbert Xu Home Page: http://gondor.apana.org.au/~herbert/ PGP Key: http://gondor.apana.org.au/~herbert/pubkey.txt