From: Corentin LABBE <clabbe.montjoie@gmail.com>
Subject: Re: [PATCH v4 3/3] crypto: Add Allwinner Security System crypto accelerator
Date: Thu, 24 Jul 2014 13:04:55 +0200
Message-ID: <53D0E857.8000405@gmail.com>
References: <1405169953-13695-1-git-send-email-clabbe.montjoie@gmail.com> <1405169953-13695-4-git-send-email-clabbe.montjoie@gmail.com> <20140724060054.GA6545@gondor.apana.org.au>
Mime-Version: 1.0
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: QUOTED-PRINTABLE
Cc: robh+dt@kernel.org, pawel.moll@arm.com, mark.rutland@arm.com,
	ijc+devicetree@hellion.org.uk, galak@codeaurora.org,
	rdunlap@infradead.org, maxime.ripard@free-electrons.com,
	linux@arm.linux.org.uk, davem@davemloft.net,
	grant.likely@linaro.org, devicetree@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org, linux-kernel@vger.kernel.org,
	linux-crypto@vger.kernel.org
To: Herbert Xu <herbert@gondor.apana.org.au>
In-Reply-To: <20140724060054.GA6545@gondor.apana.org.au>
Sender: linux-crypto-owner@vger.kernel.org

Le 24/07/2014 08:00, Herbert Xu a =E9crit :
> On Sat, Jul 12, 2014 at 02:59:13PM +0200, LABBE Corentin wrote:
>>
>> +/* sunxi_hash_init: initialize request context
>> + * Activate the SS, and configure it for MD5 or SHA1
>> + */
>> +int sunxi_hash_init(struct ahash_request *areq)
>> +{
>> +	const char *hash_type;
>> +	struct crypto_ahash *tfm =3D crypto_ahash_reqtfm(areq);
>> +	struct sunxi_req_ctx *op =3D crypto_ahash_ctx(tfm);
>> +
>> +	mutex_lock(&ss->lock);
>> +
>> +	hash_type =3D crypto_tfm_alg_name(areq->base.tfm);
>> +
>> +	op->byte_count =3D 0;
>> +	op->nbwait =3D 0;
>> +	op->waitbuf =3D 0;
>> +
>> +	/* Enable and configure SS for MD5 or SHA1 */
>> +	if (strcmp(hash_type, "sha1") =3D=3D 0)
>> +		op->mode =3D SS_OP_SHA1;
>> +	else
>> +		op->mode =3D SS_OP_MD5;
>> +
>> +	writel(op->mode | SS_ENABLED, ss->base + SS_CTL);
>> +	return 0;
>=20
> The hash driver is completely broken.  You are modifying tfm
> ctx data which is shared by all users of a single tfm.  So
> if two users conduct hashes in parallel they will step all
> over each other.

So where can I store data for each request ?

>=20
> Worse, the unpaired mutex_lock will quickly lead to dead locks.
>=20
> You cannot assume that final will be called.

An user reported an equivalent problem when using openssl speed test wi=
th cryptodev.
Does cryptoqueue is a good answer to that problem since the device coul=
d handle only one transformation at a time ?
And perhaps with cryptoqueue,  my first question is useless.