Received-SPF: pass (google.com: best guess record for domain of linux-crypto-owner@vger.kernel.org designates 209.132.180.67 as permitted sender) client-ip=209.132.180.67;
From:   Iuliana Prodan <iuliana.prodan@nxp.com>
To:     Horia Geanta <horia.geanta@nxp.com>,
        Herbert Xu <herbert@gondor.apana.org.au>,
        Aymen Sghaier <aymen.sghaier@nxp.com>
CC:     "David S. Miller" <davem@davemloft.net>,
        Tom Lendacky <thomas.lendacky@amd.com>,
        Gary Hook <gary.hook@amd.com>,
        "linux-crypto@vger.kernel.org" <linux-crypto@vger.kernel.org>,
        "linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
        dl-linux-imx <linux-imx@nxp.com>
Subject: Re: [PATCH 07/12] crypto: caam - refactor caam_jr_enqueue
Thread-Topic: [PATCH 07/12] crypto: caam - refactor caam_jr_enqueue
Thread-Index: AQHVnZa0BoWdCWDvSky4rKZ3/nrdZA==
Date:   Tue, 19 Nov 2019 22:49:19 +0000
Message-ID: <VI1PR04MB44452654BA9716CA43992AA68C4C0@VI1PR04MB4445.eurprd04.prod.outlook.com>
References: <1574029845-22796-1-git-send-email-iuliana.prodan@nxp.com>
 <1574029845-22796-8-git-send-email-iuliana.prodan@nxp.com>
 <VI1PR0402MB34853893505F95195F4C125B984C0@VI1PR0402MB3485.eurprd04.prod.outlook.com>
Accept-Language: en-US
Content-Language: en-US
received-spf: None (protection.outlook.com: nxp.com does not designate
 permitted sender hosts)
Content-Type: text/plain; charset="us-ascii"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: 2668063a-743e-4bd6-caa6-08d76d42b6cb
X-MS-Exchange-CrossTenant-originalarrivaltime: 19 Nov 2019 22:49:19.7828
 (UTC)
X-MS-Exchange-CrossTenant-fromentityheader: Hosted
X-MS-Exchange-CrossTenant-id: 686ea1d3-bc2b-4c6f-a92c-d99c5c301635
X-MS-Exchange-CrossTenant-mailboxtype: HOSTED
X-MS-Exchange-CrossTenant-userprincipalname: KylCcqBA0jv05a2pmS/8Obg490DsZK7lfzpPKI+oK12iJz1B3+fd8faxc4s1IciyEvQmXGnVGAff67ZW0NT/4A==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: VI1PR04MB5598
Sender: linux-crypto-owner@vger.kernel.org
Precedence: bulk
List-ID: <linux-crypto.vger.kernel.org>
X-Mailing-List: linux-crypto@vger.kernel.org

On 11/19/2019 7:55 PM, Horia Geanta wrote:=0A=
> On 11/18/2019 12:31 AM, Iuliana Prodan wrote:=0A=
>> Added a new struct - caam_jr_request_entry, to keep each request=0A=
>> information. This has a crypto_async_request, used to determine=0A=
>> the request type, and a bool to check if the request has backlog=0A=
>> flag or not.=0A=
>> This struct is passed to CAAM, via enqueue function - caam_jr_enqueue.=
=0A=
>>=0A=
>> The new added caam_jr_enqueue_no_bklog function is used to enqueue a job=
=0A=
>> descriptor head for cases like caamrng, key_gen, digest_key, where we=0A=
> Enqueuing terminology: either generic "job" or more HW-specific=0A=
> "job descriptor".=0A=
> Job descriptor *head* has no meaning.=0A=
> =0A=
>> don't have backlogged requests.=0A=
>>=0A=
> ...because the "requests" are not crypto requests - they are either comin=
g=0A=
> from hwrng (caamrng's case) or are driver-internal (key_gen, digest_key -=
=0A=
> used for key hashing / derivation during .setkey callback).=0A=
> =0A=
=0A=
Right, I'll update the patch description in v2.=0A=
=0A=
>> diff --git a/drivers/crypto/caam/caamalg.c b/drivers/crypto/caam/caamalg=
.c=0A=
>> index 21b6172..abebcfc 100644=0A=
>> --- a/drivers/crypto/caam/caamalg.c=0A=
>> +++ b/drivers/crypto/caam/caamalg.c=0A=
> [...]=0A=
>> @@ -1416,7 +1424,7 @@ static inline int chachapoly_crypt(struct aead_req=
uest *req, bool encrypt)=0A=
>>   			     DUMP_PREFIX_ADDRESS, 16, 4, desc, desc_bytes(desc),=0A=
>>   			     1);=0A=
>>   =0A=
>> -	ret =3D caam_jr_enqueue(jrdev, desc, aead_crypt_done, req);=0A=
>> +	ret =3D caam_jr_enqueue(jrdev, desc, aead_crypt_done, &edesc->jrentry)=
;=0A=
>>   	if (ret !=3D -EINPROGRESS) {=0A=
>>   		aead_unmap(jrdev, edesc, req);=0A=
>>   		kfree(edesc);=0A=
>> @@ -1440,6 +1448,7 @@ static inline int aead_crypt(struct aead_request *=
req, bool encrypt)=0A=
>>   	struct aead_edesc *edesc;=0A=
>>   	struct crypto_aead *aead =3D crypto_aead_reqtfm(req);=0A=
>>   	struct caam_ctx *ctx =3D crypto_aead_ctx(aead);=0A=
>> +	struct caam_jr_request_entry *jrentry;=0A=
>>   	struct device *jrdev =3D ctx->jrdev;=0A=
>>   	bool all_contig;=0A=
>>   	u32 *desc;=0A=
>> @@ -1459,7 +1468,9 @@ static inline int aead_crypt(struct aead_request *=
req, bool encrypt)=0A=
>>   			     desc_bytes(edesc->hw_desc), 1);=0A=
>>   =0A=
>>   	desc =3D edesc->hw_desc;=0A=
>> -	ret =3D caam_jr_enqueue(jrdev, desc, aead_crypt_done, req);=0A=
>> +	jrentry =3D &edesc->jrentry;=0A=
>> +=0A=
>> +	ret =3D caam_jr_enqueue(jrdev, desc, aead_crypt_done, jrentry);=0A=
> Let's avoid adding a new local variable by using &edesc->jrentry directly=
,=0A=
> like in chachapoly_crypt().=0A=
> Similar for the other places.=0A=
> =0A=
I've removed jrentry and req (as mentioned below) in patch #11 of this =0A=
series, but I'll remove them, from this patch, in v2.=0A=
=0A=
>> diff --git a/drivers/crypto/caam/caamhash.c b/drivers/crypto/caam/caamha=
sh.c=0A=
>> index baf4ab1..d9de3dc 100644=0A=
>> --- a/drivers/crypto/caam/caamhash.c=0A=
>> +++ b/drivers/crypto/caam/caamhash.c=0A=
> [...]=0A=
>> @@ -933,11 +943,13 @@ static int ahash_final_ctx(struct ahash_request *r=
eq)=0A=
>>   			     DUMP_PREFIX_ADDRESS, 16, 4, desc, desc_bytes(desc),=0A=
>>   			     1);=0A=
>>   =0A=
>> -	ret =3D caam_jr_enqueue(jrdev, desc, ahash_done_ctx_src, req);=0A=
>> +	jrentry =3D &edesc->jrentry;=0A=
>> +=0A=
>> +	ret =3D caam_jr_enqueue(jrdev, desc, ahash_done_ctx_src, jrentry);=0A=
>>   	if (ret =3D=3D -EINPROGRESS)=0A=
>>   		return ret;=0A=
>>   =0A=
>> - unmap_ctx:=0A=
>> +unmap_ctx:=0A=
> That's correct, however whitespace fixing should be done separately.=0A=
> =0A=
Should I make a separate patch for these two whitespaces?=0A=
=0A=
>> @@ -1009,11 +1022,13 @@ static int ahash_finup_ctx(struct ahash_request =
*req)=0A=
>>   			     DUMP_PREFIX_ADDRESS, 16, 4, desc, desc_bytes(desc),=0A=
>>   			     1);=0A=
>>   =0A=
>> -	ret =3D caam_jr_enqueue(jrdev, desc, ahash_done_ctx_src, req);=0A=
>> +	jrentry =3D &edesc->jrentry;=0A=
>> +=0A=
>> +	ret =3D caam_jr_enqueue(jrdev, desc, ahash_done_ctx_src, jrentry);=0A=
>>   	if (ret =3D=3D -EINPROGRESS)=0A=
>>   		return ret;=0A=
>>   =0A=
>> - unmap_ctx:=0A=
>> +unmap_ctx:=0A=
> Again, unrelated whitespace fix.=0A=
> =0A=
>> diff --git a/drivers/crypto/caam/caampkc.c b/drivers/crypto/caam/caampkc=
.c=0A=
>> index 7f7ea32..bb0e4b9 100644=0A=
>> --- a/drivers/crypto/caam/caampkc.c=0A=
>> +++ b/drivers/crypto/caam/caampkc.c=0A=
> [...]=0A=
>> @@ -315,6 +317,8 @@ static struct rsa_edesc *rsa_edesc_alloc(struct akci=
pher_request *req,=0A=
>>   	edesc->mapped_src_nents =3D mapped_src_nents;=0A=
>>   	edesc->mapped_dst_nents =3D mapped_dst_nents;=0A=
>>   =0A=
>> +	edesc->jrentry.base =3D &req->base;=0A=
>> +=0A=
>>   	edesc->sec4_sg_dma =3D dma_map_single(dev, edesc->sec4_sg,=0A=
>>   					    sec4_sg_bytes, DMA_TO_DEVICE);=0A=
>>   	if (dma_mapping_error(dev, edesc->sec4_sg_dma)) {=0A=
> [...]=0A=
>> @@ -633,7 +638,10 @@ static int caam_rsa_enc(struct akcipher_request *re=
q)=0A=
>>   	/* Initialize Job Descriptor */=0A=
>>   	init_rsa_pub_desc(edesc->hw_desc, &edesc->pdb.pub);=0A=
>>   =0A=
>> -	ret =3D caam_jr_enqueue(jrdev, edesc->hw_desc, rsa_pub_done, req);=0A=
>> +	jrentry =3D &edesc->jrentry;=0A=
>> +	jrentry->base =3D &req->base;=0A=
> This field is already set in rsa_edesc_alloc().=0A=
> =0A=
>> @@ -666,7 +675,10 @@ static int caam_rsa_dec_priv_f1(struct akcipher_req=
uest *req)=0A=
>>   	/* Initialize Job Descriptor */=0A=
>>   	init_rsa_priv_f1_desc(edesc->hw_desc, &edesc->pdb.priv_f1);=0A=
>>   =0A=
>> -	ret =3D caam_jr_enqueue(jrdev, edesc->hw_desc, rsa_priv_f_done, req);=
=0A=
>> +	jrentry =3D &edesc->jrentry;=0A=
>> +	jrentry->base =3D &req->base;=0A=
> The same here.=0A=
> =0A=
>> @@ -699,7 +712,10 @@ static int caam_rsa_dec_priv_f2(struct akcipher_req=
uest *req)=0A=
>>   	/* Initialize Job Descriptor */=0A=
>>   	init_rsa_priv_f2_desc(edesc->hw_desc, &edesc->pdb.priv_f2);=0A=
>>   =0A=
>> -	ret =3D caam_jr_enqueue(jrdev, edesc->hw_desc, rsa_priv_f_done, req);=
=0A=
>> +	jrentry =3D &edesc->jrentry;=0A=
>> +	jrentry->base =3D &req->base;=0A=
> And here.=0A=
> =0A=
>> @@ -732,7 +749,10 @@ static int caam_rsa_dec_priv_f3(struct akcipher_req=
uest *req)=0A=
>>   	/* Initialize Job Descriptor */=0A=
>>   	init_rsa_priv_f3_desc(edesc->hw_desc, &edesc->pdb.priv_f3);=0A=
>>   =0A=
>> -	ret =3D caam_jr_enqueue(jrdev, edesc->hw_desc, rsa_priv_f_done, req);=
=0A=
>> +	jrentry =3D &edesc->jrentry;=0A=
>> +	jrentry->base =3D &req->base;=0A=
> Also here.=0A=
> =0A=
>> diff --git a/drivers/crypto/caam/intern.h b/drivers/crypto/caam/intern.h=
=0A=
>> index c7c10c9..58be66c 100644=0A=
>> --- a/drivers/crypto/caam/intern.h=0A=
>> +++ b/drivers/crypto/caam/intern.h=0A=
> [...]=0A=
>> @@ -104,6 +105,15 @@ struct caam_drv_private {=0A=
>>   #endif=0A=
>>   };=0A=
>>   =0A=
>> +/*=0A=
>> + * Storage for tracking each request that is processed by a ring=0A=
>> + */=0A=
>> +struct caam_jr_request_entry {=0A=
>> +	/* Common attributes for async crypto requests */=0A=
>> +	struct crypto_async_request *base;=0A=
>> +	bool bklog;	/* Stored to determine if the request needs backlog */=0A=
>> +};=0A=
>> +=0A=
> Could we use kernel-doc here?=0A=
=0A=
Sure, will do in v2.=0A=
=0A=
Thanks,=0A=
Iulia=0A=
=0A=
> =0A=
> Horia=0A=
> =0A=
=0A=