From: Akhil Goyal <akhil.goyal@nxp.com>
To: Pablo de Lara <pablo.de.lara.guarch@intel.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>
Subject: Re: [dpdk-dev] [PATCH v2 1/2] crypto/zuc: support IPSec Multi-buffer lib v0.54
Date: Mon, 11 May 2020 07:43:47 +0000 [thread overview]
Message-ID: <VI1PR04MB31689FE43F1B090799918E4AE6A10@VI1PR04MB3168.eurprd04.prod.outlook.com> (raw)
In-Reply-To: <1589139650-195685-1-git-send-email-pablo.de.lara.guarch@intel.com>
Hi Pablo,
It seems there is compilation issue in Clang with this patch.
Regards,
Akhil
> Signed-off-by: Pablo de Lara <pablo.de.lara.guarch@intel.com>
> ---
>
> v2:
> - Simplified logic in process_hash_op per Akhil's comments
>
> doc/guides/cryptodevs/zuc.rst | 6 ++--
> doc/guides/rel_notes/release_20_05.rst | 7 ++++
> drivers/crypto/zuc/rte_zuc_pmd.c | 58 ++++++++++++++++++++--------------
> 3 files changed, 44 insertions(+), 27 deletions(-)
>
> diff --git a/doc/guides/cryptodevs/zuc.rst b/doc/guides/cryptodevs/zuc.rst
> index 38ea999..c384f3d 100644
> --- a/doc/guides/cryptodevs/zuc.rst
> +++ b/doc/guides/cryptodevs/zuc.rst
> @@ -35,8 +35,8 @@ Installation
> To build DPDK with the ZUC_PMD the user is required to download the multi-
> buffer
> library from `here
> <https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.c
> om%2F01org%2Fintel-ipsec-
> mb&data=02%7C01%7Cakhil.goyal%40nxp.com%7C5e2b3704135b4de6d5
> a708d7f51a140b%7C686ea1d3bc2b4c6fa92cd99c5c301635%7C0%7C0%7C6372
> 47364662642602&sdata=gncK8ob%2FbtOTpjZH2luy4%2Fa4TcfQYGsvI6M%
> 2BRQdkfBk%3D&reserved=0>`_
> and compile it on their user system before building DPDK.
> -The latest version of the library supported by this PMD is v0.53, which
> -can be downloaded from
> `<https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.
> com%2F01org%2Fintel-ipsec-
> mb%2Farchive%2Fv0.53.zip&data=02%7C01%7Cakhil.goyal%40nxp.com%7
> C5e2b3704135b4de6d5a708d7f51a140b%7C686ea1d3bc2b4c6fa92cd99c5c301
> 635%7C0%7C0%7C637247364662642602&sdata=JuMgmXftLm62TGvaaSo
> OUpSOWGhT9eVQ%2Bb8r3NEFUow%3D&reserved=0>`_.
> +The latest version of the library supported by this PMD is v0.54, which
> +can be downloaded from
> `<https://eur01.safelinks.protection.outlook.com/?url=https%3A%2F%2Fgithub.
> com%2F01org%2Fintel-ipsec-
> mb%2Farchive%2Fv0.54.zip&data=02%7C01%7Cakhil.goyal%40nxp.com%7
> C5e2b3704135b4de6d5a708d7f51a140b%7C686ea1d3bc2b4c6fa92cd99c5c301
> 635%7C0%7C0%7C637247364662642602&sdata=9oDdUezguCEPTCe5FfQS
> w7zJKt5%2BiSUe5v4Ypm5YBIw%3D&reserved=0>`_.
>
> After downloading the library, the user needs to unpack and compile it
> on their system before building DPDK:
> @@ -63,7 +63,7 @@ and the external crypto libraries supported by them:
> DPDK version Crypto library version
> ============= ================================
> 16.11 - 19.11 LibSSO ZUC
> - 20.02+ Multi-buffer library 0.53
> + 20.02+ Multi-buffer library 0.53 - 0.54
> ============= ================================
>
>
> diff --git a/doc/guides/rel_notes/release_20_05.rst
> b/doc/guides/rel_notes/release_20_05.rst
> index fe6c75e..7d3a4bf 100644
> --- a/doc/guides/rel_notes/release_20_05.rst
> +++ b/doc/guides/rel_notes/release_20_05.rst
> @@ -155,6 +155,13 @@ New Features
>
> * Added support for intel-ipsec-mb version 0.54.
>
> +* **Updated the ZUC crypto PMD.**
> +
> + * Added support for intel-ipsec-mb version 0.54.
> + * Updated the PMD to support Multi-buffer ZUC-EIA3,
> + improving performance significantly, when using
> + intel-ipsec-mb version 0.54
> +
> * **Added a new driver for Intel Foxville I225 devices.**
>
> Added the new ``igc`` net driver for Intel Foxville I225 devices. See the
> diff --git a/drivers/crypto/zuc/rte_zuc_pmd.c
> b/drivers/crypto/zuc/rte_zuc_pmd.c
> index 17926b4..45ae04b 100644
> --- a/drivers/crypto/zuc/rte_zuc_pmd.c
> +++ b/drivers/crypto/zuc/rte_zuc_pmd.c
> @@ -237,12 +237,13 @@ process_zuc_hash_op(struct zuc_qp *qp, struct
> rte_crypto_op **ops,
> struct zuc_session **sessions,
> uint8_t num_ops)
> {
> - unsigned i;
> + unsigned int i;
> uint8_t processed_ops = 0;
> - uint8_t *src;
> - uint32_t *dst;
> - uint32_t length_in_bits;
> - uint8_t *iv;
> + uint8_t *src[ZUC_MAX_BURST];
> + uint32_t *dst[ZUC_MAX_BURST];
> + uint32_t length_in_bits[ZUC_MAX_BURST];
> + uint8_t *iv[ZUC_MAX_BURST];
> + const void *hash_keys[ZUC_MAX_BURST];
> struct zuc_session *sess;
>
> for (i = 0; i < num_ops; i++) {
> @@ -255,33 +256,42 @@ process_zuc_hash_op(struct zuc_qp *qp, struct
> rte_crypto_op **ops,
>
> sess = sessions[i];
>
> - length_in_bits = ops[i]->sym->auth.data.length;
> + length_in_bits[i] = ops[i]->sym->auth.data.length;
>
> - src = rte_pktmbuf_mtod(ops[i]->sym->m_src, uint8_t *) +
> + src[i] = rte_pktmbuf_mtod(ops[i]->sym->m_src, uint8_t *) +
> (ops[i]->sym->auth.data.offset >> 3);
> - iv = rte_crypto_op_ctod_offset(ops[i], uint8_t *,
> + iv[i] = rte_crypto_op_ctod_offset(ops[i], uint8_t *,
> sess->auth_iv_offset);
>
> - if (sess->auth_op == RTE_CRYPTO_AUTH_OP_VERIFY) {
> - dst = (uint32_t *)qp->temp_digest;
> -
> - IMB_ZUC_EIA3_1_BUFFER(qp->mb_mgr, sess-
> >pKey_hash,
> - iv, src,
> - length_in_bits, dst);
> - /* Verify digest. */
> - if (memcmp(dst, ops[i]->sym->auth.digest.data,
> - ZUC_DIGEST_LENGTH) != 0)
> - ops[i]->status =
> RTE_CRYPTO_OP_STATUS_AUTH_FAILED;
> - } else {
> - dst = (uint32_t *)ops[i]->sym->auth.digest.data;
> + hash_keys[i] = sess->pKey_hash;
> + if (sess->auth_op == RTE_CRYPTO_AUTH_OP_VERIFY)
> + dst[i] = (uint32_t *)qp->temp_digest;
> + else
> + dst[i] = (uint32_t *)ops[i]->sym->auth.digest.data;
>
> - IMB_ZUC_EIA3_1_BUFFER(qp->mb_mgr, sess-
> >pKey_hash,
> - iv, src,
> - length_in_bits, dst);
> - }
> +#if IMB_VERSION_NUM < IMB_VERSION(0, 53, 3)
> + IMB_ZUC_EIA3_1_BUFFER(qp->mb_mgr, hash_keys[i],
> + iv[i], src[i], length_in_bits[i], dst[i]);
> +#endif
> processed_ops++;
> }
>
> +#if IMB_VERSION_NUM >= IMB_VERSION(0, 53, 3)
> + IMB_ZUC_EIA3_N_BUFFER(qp->mb_mgr, (const void **)hash_keys,
> + (const void **)iv, (const void **)src, length_in_bits,
> + dst, processed_ops);
> +#endif
> +
> + /*
> + * If tag needs to be verified, compare generated tag
> + * with attached tag
> + */
> + for (i = 0; i < processed_ops; i++)
> + if (sessions[i]->auth_op == RTE_CRYPTO_AUTH_OP_VERIFY)
> + if (memcmp(dst[i], ops[i]->sym->auth.digest.data,
> + ZUC_DIGEST_LENGTH) != 0)
> + ops[i]->status =
> RTE_CRYPTO_OP_STATUS_AUTH_FAILED;
> +
> return processed_ops;
> }
>
> --
> 2.7.5
next prev parent reply other threads:[~2020-05-11 7:43 UTC|newest]
Thread overview: 11+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-04-21 16:56 [dpdk-dev] [PATCH " Pablo de Lara
2020-04-21 16:56 ` [dpdk-dev] [PATCH 2/2] doc: support IPsec " Pablo de Lara
2020-05-09 19:05 ` [dpdk-dev] [PATCH 1/2] crypto/zuc: support IPSec " Akhil Goyal
2020-05-10 19:33 ` De Lara Guarch, Pablo
2020-05-10 19:40 ` [dpdk-dev] [PATCH v2 " Pablo de Lara
2020-05-10 19:40 ` [dpdk-dev] [PATCH v2 2/2] doc: support IPsec " Pablo de Lara
2020-05-11 7:43 ` Akhil Goyal [this message]
2020-05-11 9:14 ` [dpdk-dev] [PATCH v3 1/2] crypto/zuc: support IPSec " Pablo de Lara
2020-05-11 9:14 ` [dpdk-dev] [PATCH v3 2/2] doc: support IPsec " Pablo de Lara
2020-05-11 10:28 ` [dpdk-dev] [PATCH v3 1/2] crypto/zuc: support IPSec " Akhil Goyal
2020-05-11 12:06 ` De Lara Guarch, Pablo
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=VI1PR04MB31689FE43F1B090799918E4AE6A10@VI1PR04MB3168.eurprd04.prod.outlook.com \
--to=akhil.goyal@nxp.com \
--cc=dev@dpdk.org \
--cc=pablo.de.lara.guarch@intel.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).