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:   Horia Geanta <horia.geanta@nxp.com>
To:     Herbert Xu <herbert@gondor.apana.org.au>
CC:     "David S. Miller" <davem@davemloft.net>,
        Aymen Sghaier <aymen.sghaier@nxp.com>,
        Valentin Ciocoi Radulescu <valentin.ciocoi@nxp.com>,
        "linux-crypto@vger.kernel.org" <linux-crypto@vger.kernel.org>,
        dl-linux-imx <linux-imx@nxp.com>
Subject: Re: [PATCH] crypto: caam/qi2 - fix chacha20 data size error
Thread-Topic: [PATCH] crypto: caam/qi2 - fix chacha20 data size error
Thread-Index: AQHV7gHsrEsorQs4ikqBr5jkreFUxw==
Date:   Fri, 28 Feb 2020 06:40:12 +0000
Message-ID: <VI1PR0402MB34858272BF3305BFCBC21BFB98E80@VI1PR0402MB3485.eurprd04.prod.outlook.com>
References: <20200221075201.5725-1-horia.geanta@nxp.com>
 <20200228004235.GB9163@gondor.apana.org.au>
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="iso-8859-2"
Content-Transfer-Encoding: quoted-printable
MIME-Version: 1.0
X-MS-Exchange-CrossTenant-Network-Message-Id: dff58d88-c25e-4416-e1d6-08d7bc190fd7
X-MS-Exchange-CrossTenant-originalarrivaltime: 28 Feb 2020 06:40:12.0216
 (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: Z9wDSdD9mULDBV/6GTfm1nSp4Wks1WHAtU9STEoqSw6kQ0G4yw5G0Kln00HNve8g6DBVn8Qmh/wxXywKpp9RYA==
X-MS-Exchange-Transport-CrossTenantHeadersStamped: VI1PR0402MB3695
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 2/28/2020 2:42 AM, Herbert Xu wrote:=0A=
> On Fri, Feb 21, 2020 at 09:52:01AM +0200, Horia Geant=E3 wrote:=0A=
>> HW generates a Data Size error for chacha20 requests that are not=0A=
>> a multiple of 64B, since algorithm state (AS) does not have=0A=
>> the FINAL bit set.=0A=
>>=0A=
>> Since updating req->iv (for chaining) is not required,=0A=
>> modify skcipher descriptors to set the FINAL bit for chacha20.=0A=
>>=0A=
>> [Note that for skcipher decryption we know that ctx1_iv_off is 0,=0A=
>> which allows for an optimization by not checking algorithm type,=0A=
>> since append_dec_op1() sets FINAL bit for all algorithms except AES.]=0A=
>>=0A=
>> Also drop the descriptor operations that save the IV.=0A=
>> However, in order to keep code logic simple, things like=0A=
>> S/G tables generation etc. are not touched.=0A=
>>=0A=
>> Cc: <stable@vger.kernel.org> # v5.3+=0A=
>> Fixes: 334d37c9e263 ("crypto: caam - update IV using HW support")=0A=
>> Signed-off-by: Horia Geant=E3 <horia.geanta@nxp.com>=0A=
>> ---=0A=
>>  drivers/crypto/caam/caamalg_desc.c | 14 ++++++++++----=0A=
>>  1 file changed, 10 insertions(+), 4 deletions(-)=0A=
>>=0A=
>> diff --git a/drivers/crypto/caam/caamalg_desc.c b/drivers/crypto/caam/ca=
amalg_desc.c=0A=
>> index aa9ccca67045..372d3d4ed6c5 100644=0A=
>> --- a/drivers/crypto/caam/caamalg_desc.c=0A=
>> +++ b/drivers/crypto/caam/caamalg_desc.c=0A=
>> @@ -1379,6 +1379,8 @@ void cnstr_shdsc_skcipher_encap(u32 * const desc, =
struct alginfo *cdata,=0A=
>>  				const u32 ctx1_iv_off)=0A=
>>  {=0A=
>>  	u32 *key_jump_cmd;=0A=
>> +	bool is_chacha20 =3D ((cdata->algtype & OP_ALG_ALGSEL_MASK) =3D=3D=0A=
>> +			    OP_ALG_ALGSEL_CHACHA20);=0A=
>>  =0A=
>>  	init_sh_desc(desc, HDR_SHARE_SERIAL | HDR_SAVECTX);=0A=
>>  	/* Skip if already shared */=0A=
>> @@ -1417,14 +1419,15 @@ void cnstr_shdsc_skcipher_encap(u32 * const desc=
, struct alginfo *cdata,=0A=
>>  				      LDST_OFFSET_SHIFT));=0A=
>>  =0A=
>>  	/* Load operation */=0A=
>> -	append_operation(desc, cdata->algtype | OP_ALG_AS_INIT |=0A=
>> -			 OP_ALG_ENCRYPT);=0A=
>> +	if (is_chacha20)=0A=
>> +		options |=3D OP_ALG_AS_FINALIZE;=0A=
>> +	append_operation(desc, options);=0A=
> =0A=
> This patch doesn't compile.=0A=
> =0A=
Thanks for catching this.=0A=
Not sure how this slipped through, probably it happened due to=0A=
piling up too many *.patch files...=0A=
Will fix in v2.=0A=
=0A=
Horia=0A=