From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by dpdk.org (Postfix) with ESMTP id D1CEF1B1E8 for ; Wed, 20 Mar 2019 18:53:03 +0100 (CET) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga007.fm.intel.com ([10.253.24.52]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 20 Mar 2019 10:53:02 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.60,249,1549958400"; d="scan'208";a="135942524" Received: from fmsmsx108.amr.corp.intel.com ([10.18.124.206]) by fmsmga007.fm.intel.com with ESMTP; 20 Mar 2019 10:53:02 -0700 Received: from fmsmsx117.amr.corp.intel.com ([169.254.3.142]) by FMSMSX108.amr.corp.intel.com ([169.254.9.216]) with mapi id 14.03.0415.000; Wed, 20 Mar 2019 10:53:02 -0700 From: "Wiles, Keith" To: "Ananyev, Konstantin" CC: dpdk-dev , "akhil.goyal@nxp.com" , "olivier.matz@6wind.com" Thread-Topic: [dpdk-dev] [PATCH v2 1/7] mbuf: new function to generate raw Tx offload value Thread-Index: AQHU30Huz6hfQp9e4UOCPjmhtS49dKYVQlgA Date: Wed, 20 Mar 2019 17:53:01 +0000 Message-ID: <96B7DD3D-3016-4C23-9BB0-DFB1F16B07A9@intel.com> References: <1551381661-21078-1-git-send-email-konstantin.ananyev@intel.com> <1553102679-23576-2-git-send-email-konstantin.ananyev@intel.com> In-Reply-To: <1553102679-23576-2-git-send-email-konstantin.ananyev@intel.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.144.146.157] Content-Type: text/plain; charset="us-ascii" Content-ID: <1BA5BFAD9729A649A55938C5A3317503@intel.com> Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v2 1/7] mbuf: new function to generate raw Tx offload value X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 20 Mar 2019 17:53:04 -0000 > On Mar 20, 2019, at 10:24 AM, Konstantin Ananyev wrote: >=20 > Operations to set/update bit-fields often cause compilers > to generate suboptimal code. > To help avoid such situation for tx_offload fields: > introduce new enum for tx_offload bit-fields lengths and offsets, > and new function to generate raw tx_offload value. >=20 > Signed-off-by: Konstantin Ananyev > --- > lib/librte_mbuf/rte_mbuf.h | 71 ++++++++++++++++++++++++++++++++++---- > 1 file changed, 64 insertions(+), 7 deletions(-) >=20 > diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h > index d961ccaf6..b967ad17e 100644 > --- a/lib/librte_mbuf/rte_mbuf.h > +++ b/lib/librte_mbuf/rte_mbuf.h > @@ -479,6 +479,26 @@ struct rte_mbuf_sched { > uint16_t reserved; /**< Reserved. */ > }; /**< Hierarchical scheduler */ >=20 > +/** enum for the tx_offload bit-fields lenghts and offsets. */ > +enum { > + RTE_MBUF_L2_LEN_BITS =3D 7, > + RTE_MBUF_L3_LEN_BITS =3D 9, > + RTE_MBUF_L4_LEN_BITS =3D 8, > + RTE_MBUF_TSO_SEGSZ_BITS =3D 16, > + RTE_MBUF_OL3_LEN_BITS =3D 9, > + RTE_MBUF_OL2_LEN_BITS =3D 7, > + RTE_MBUF_L2_LEN_OFS =3D 0, > + RTE_MBUF_L3_LEN_OFS =3D RTE_MBUF_L2_LEN_OFS + RTE_MBUF_L2_LEN_BITS, > + RTE_MBUF_L4_LEN_OFS =3D RTE_MBUF_L3_LEN_OFS + RTE_MBUF_L3_LEN_BITS, > + RTE_MBUF_TSO_SEGSZ_OFS =3D RTE_MBUF_L4_LEN_OFS + RTE_MBUF_L4_LEN_BITS, > + RTE_MBUF_OL3_LEN_OFS =3D RTE_MBUF_TSO_SEGSZ_OFS + RTE_MBUF_TSO_SEGSZ_BI= TS, > + RTE_MBUF_OL2_LEN_OFS =3D RTE_MBUF_OL3_LEN_OFS + RTE_MBUF_OL3_LEN_BITS, > + RTE_MBUF_TXOFLD_UNUSED_OFS =3D > + RTE_MBUF_OL2_LEN_OFS + RTE_MBUF_OL2_LEN_BITS, > + RTE_MBUF_TXOFLD_UNUSED_BITS =3D > + sizeof(uint64_t) * CHAR_BIT - RTE_MBUF_TXOFLD_UNUSED_OFS, > +}; The naming of these fields is a bit unreadable and why do we need to add RT= E_MBUF_ to everything especially adding RTE_ to everything. Then we have to shorten words like OFFSET to OFS, it would be much better f= or readability to use something like this RTE_MBUF_OL3_LEN_OFS to MBUF_OL3_LEN_OFFSET, then we have OL3 or OL2, which= does not jump out at me as well. What does OL3 or OL2 mean? OffLoad-3 or OffLoad-2 this also needs to be cle= arly defined to improve the readability of the macros. Also for TXOFLD needs to be clearer here to help the reader understand the = mean of the defines. > + > /** > * The generic rte_mbuf, containing a packet mbuf. > */ > @@ -640,19 +660,24 @@ struct rte_mbuf { > uint64_t tx_offload; /**< combined for easy fetch */ > __extension__ > struct { > - uint64_t l2_len:7; > + uint64_t l2_len:RTE_MBUF_L2_LEN_BITS; > /**< L2 (MAC) Header Length for non-tunneling pkt. > * Outer_L4_len + ... + Inner_L2_len for tunneling pkt. > */ > - uint64_t l3_len:9; /**< L3 (IP) Header Length. */ > - uint64_t l4_len:8; /**< L4 (TCP/UDP) Header Length. */ > - uint64_t tso_segsz:16; /**< TCP TSO segment size */ > + uint64_t l3_len:RTE_MBUF_L3_LEN_BITS; > + /**< L3 (IP) Header Length. */ > + uint64_t l4_len:RTE_MBUF_L4_LEN_BITS; > + /**< L4 (TCP/UDP) Header Length. */ > + uint64_t tso_segsz:RTE_MBUF_TSO_SEGSZ_BITS; > + /**< TCP TSO segment size */ >=20 > /* fields for TX offloading of tunnels */ > - uint64_t outer_l3_len:9; /**< Outer L3 (IP) Hdr Length. */ > - uint64_t outer_l2_len:7; /**< Outer L2 (MAC) Hdr Length. */ > + uint64_t outer_l3_len:RTE_MBUF_OL3_LEN_BITS; > + /**< Outer L3 (IP) Hdr Length. */ > + uint64_t outer_l2_len:RTE_MBUF_OL2_LEN_BITS; > + /**< Outer L2 (MAC) Hdr Length. */ >=20 > - /* uint64_t unused:8; */ > + /* uint64_t unused:RTE_MBUF_TXOFLD_UNUSED_BITS; */ > }; > }; >=20 > @@ -2243,6 +2268,38 @@ static inline int rte_pktmbuf_chain(struct rte_mbu= f *head, struct rte_mbuf *tail > return 0; > } >=20 > +/* > + * @warning > + * @b EXPERIMENTAL: This API may change without prior notice. > + * > + * For given input values generate raw tx_offload value. > + * @param il2 > + * l2_len value. > + * @param il3 > + * l3_len value. > + * @param il4 > + * l4_len value. > + * @param tso > + * tso_segsz value. > + * @param ol3 > + * outer_l3_len value. > + * @param ol2 > + * outer_l2_len value. > + * @return > + * raw tx_offload value. > + */ > +static inline uint64_t > +rte_mbuf_tx_offload(uint64_t il2, uint64_t il3, uint64_t il4, uint64_t t= so, > + uint64_t ol3, uint64_t ol2) > +{ > + return il2 << RTE_MBUF_L2_LEN_OFS | > + il3 << RTE_MBUF_L3_LEN_OFS | > + il4 << RTE_MBUF_L4_LEN_OFS | > + tso << RTE_MBUF_TSO_SEGSZ_OFS | > + ol3 << RTE_MBUF_OL3_LEN_OFS | > + ol2 << RTE_MBUF_OL2_LEN_OFS; > +} > + > /** > * Validate general requirements for Tx offload in mbuf. > * > --=20 > 2.17.1 >=20 Regards, Keith From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from dpdk.org (dpdk.org [92.243.14.124]) by dpdk.space (Postfix) with ESMTP id BF5ABA00E6 for ; Wed, 20 Mar 2019 18:53:05 +0100 (CET) Received: from [92.243.14.124] (localhost [127.0.0.1]) by dpdk.org (Postfix) with ESMTP id 720881B1E9; Wed, 20 Mar 2019 18:53:05 +0100 (CET) Received: from mga09.intel.com (mga09.intel.com [134.134.136.24]) by dpdk.org (Postfix) with ESMTP id D1CEF1B1E8 for ; Wed, 20 Mar 2019 18:53:03 +0100 (CET) X-Amp-Result: SKIPPED(no attachment in message) X-Amp-File-Uploaded: False Received: from fmsmga007.fm.intel.com ([10.253.24.52]) by orsmga102.jf.intel.com with ESMTP/TLS/DHE-RSA-AES256-GCM-SHA384; 20 Mar 2019 10:53:02 -0700 X-ExtLoop1: 1 X-IronPort-AV: E=Sophos;i="5.60,249,1549958400"; d="scan'208";a="135942524" Received: from fmsmsx108.amr.corp.intel.com ([10.18.124.206]) by fmsmga007.fm.intel.com with ESMTP; 20 Mar 2019 10:53:02 -0700 Received: from fmsmsx117.amr.corp.intel.com ([169.254.3.142]) by FMSMSX108.amr.corp.intel.com ([169.254.9.216]) with mapi id 14.03.0415.000; Wed, 20 Mar 2019 10:53:02 -0700 From: "Wiles, Keith" To: "Ananyev, Konstantin" CC: dpdk-dev , "akhil.goyal@nxp.com" , "olivier.matz@6wind.com" Thread-Topic: [dpdk-dev] [PATCH v2 1/7] mbuf: new function to generate raw Tx offload value Thread-Index: AQHU30Huz6hfQp9e4UOCPjmhtS49dKYVQlgA Date: Wed, 20 Mar 2019 17:53:01 +0000 Message-ID: <96B7DD3D-3016-4C23-9BB0-DFB1F16B07A9@intel.com> References: <1551381661-21078-1-git-send-email-konstantin.ananyev@intel.com> <1553102679-23576-2-git-send-email-konstantin.ananyev@intel.com> In-Reply-To: <1553102679-23576-2-git-send-email-konstantin.ananyev@intel.com> Accept-Language: en-US Content-Language: en-US X-MS-Has-Attach: X-MS-TNEF-Correlator: x-originating-ip: [10.144.146.157] Content-Type: text/plain; charset="UTF-8" Content-ID: <1BA5BFAD9729A649A55938C5A3317503@intel.com> Content-Transfer-Encoding: quoted-printable MIME-Version: 1.0 Subject: Re: [dpdk-dev] [PATCH v2 1/7] mbuf: new function to generate raw Tx offload value X-BeenThere: dev@dpdk.org X-Mailman-Version: 2.1.15 Precedence: list List-Id: DPDK patches and discussions List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: dev-bounces@dpdk.org Sender: "dev" Message-ID: <20190320175301.wrr_OsN42tbGKv2LgsJb4DWLrrg6AN3ANkKa6yJLj58@z> > On Mar 20, 2019, at 10:24 AM, Konstantin Ananyev wrote: >=20 > Operations to set/update bit-fields often cause compilers > to generate suboptimal code. > To help avoid such situation for tx_offload fields: > introduce new enum for tx_offload bit-fields lengths and offsets, > and new function to generate raw tx_offload value. >=20 > Signed-off-by: Konstantin Ananyev > --- > lib/librte_mbuf/rte_mbuf.h | 71 ++++++++++++++++++++++++++++++++++---- > 1 file changed, 64 insertions(+), 7 deletions(-) >=20 > diff --git a/lib/librte_mbuf/rte_mbuf.h b/lib/librte_mbuf/rte_mbuf.h > index d961ccaf6..b967ad17e 100644 > --- a/lib/librte_mbuf/rte_mbuf.h > +++ b/lib/librte_mbuf/rte_mbuf.h > @@ -479,6 +479,26 @@ struct rte_mbuf_sched { > uint16_t reserved; /**< Reserved. */ > }; /**< Hierarchical scheduler */ >=20 > +/** enum for the tx_offload bit-fields lenghts and offsets. */ > +enum { > + RTE_MBUF_L2_LEN_BITS =3D 7, > + RTE_MBUF_L3_LEN_BITS =3D 9, > + RTE_MBUF_L4_LEN_BITS =3D 8, > + RTE_MBUF_TSO_SEGSZ_BITS =3D 16, > + RTE_MBUF_OL3_LEN_BITS =3D 9, > + RTE_MBUF_OL2_LEN_BITS =3D 7, > + RTE_MBUF_L2_LEN_OFS =3D 0, > + RTE_MBUF_L3_LEN_OFS =3D RTE_MBUF_L2_LEN_OFS + RTE_MBUF_L2_LEN_BITS, > + RTE_MBUF_L4_LEN_OFS =3D RTE_MBUF_L3_LEN_OFS + RTE_MBUF_L3_LEN_BITS, > + RTE_MBUF_TSO_SEGSZ_OFS =3D RTE_MBUF_L4_LEN_OFS + RTE_MBUF_L4_LEN_BITS, > + RTE_MBUF_OL3_LEN_OFS =3D RTE_MBUF_TSO_SEGSZ_OFS + RTE_MBUF_TSO_SEGSZ_BI= TS, > + RTE_MBUF_OL2_LEN_OFS =3D RTE_MBUF_OL3_LEN_OFS + RTE_MBUF_OL3_LEN_BITS, > + RTE_MBUF_TXOFLD_UNUSED_OFS =3D > + RTE_MBUF_OL2_LEN_OFS + RTE_MBUF_OL2_LEN_BITS, > + RTE_MBUF_TXOFLD_UNUSED_BITS =3D > + sizeof(uint64_t) * CHAR_BIT - RTE_MBUF_TXOFLD_UNUSED_OFS, > +}; The naming of these fields is a bit unreadable and why do we need to add RT= E_MBUF_ to everything especially adding RTE_ to everything. Then we have to shorten words like OFFSET to OFS, it would be much better f= or readability to use something like this RTE_MBUF_OL3_LEN_OFS to MBUF_OL3_LEN_OFFSET, then we have OL3 or OL2, which= does not jump out at me as well. What does OL3 or OL2 mean? OffLoad-3 or OffLoad-2 this also needs to be cle= arly defined to improve the readability of the macros. Also for TXOFLD needs to be clearer here to help the reader understand the = mean of the defines. > + > /** > * The generic rte_mbuf, containing a packet mbuf. > */ > @@ -640,19 +660,24 @@ struct rte_mbuf { > uint64_t tx_offload; /**< combined for easy fetch */ > __extension__ > struct { > - uint64_t l2_len:7; > + uint64_t l2_len:RTE_MBUF_L2_LEN_BITS; > /**< L2 (MAC) Header Length for non-tunneling pkt. > * Outer_L4_len + ... + Inner_L2_len for tunneling pkt. > */ > - uint64_t l3_len:9; /**< L3 (IP) Header Length. */ > - uint64_t l4_len:8; /**< L4 (TCP/UDP) Header Length. */ > - uint64_t tso_segsz:16; /**< TCP TSO segment size */ > + uint64_t l3_len:RTE_MBUF_L3_LEN_BITS; > + /**< L3 (IP) Header Length. */ > + uint64_t l4_len:RTE_MBUF_L4_LEN_BITS; > + /**< L4 (TCP/UDP) Header Length. */ > + uint64_t tso_segsz:RTE_MBUF_TSO_SEGSZ_BITS; > + /**< TCP TSO segment size */ >=20 > /* fields for TX offloading of tunnels */ > - uint64_t outer_l3_len:9; /**< Outer L3 (IP) Hdr Length. */ > - uint64_t outer_l2_len:7; /**< Outer L2 (MAC) Hdr Length. */ > + uint64_t outer_l3_len:RTE_MBUF_OL3_LEN_BITS; > + /**< Outer L3 (IP) Hdr Length. */ > + uint64_t outer_l2_len:RTE_MBUF_OL2_LEN_BITS; > + /**< Outer L2 (MAC) Hdr Length. */ >=20 > - /* uint64_t unused:8; */ > + /* uint64_t unused:RTE_MBUF_TXOFLD_UNUSED_BITS; */ > }; > }; >=20 > @@ -2243,6 +2268,38 @@ static inline int rte_pktmbuf_chain(struct rte_mbu= f *head, struct rte_mbuf *tail > return 0; > } >=20 > +/* > + * @warning > + * @b EXPERIMENTAL: This API may change without prior notice. > + * > + * For given input values generate raw tx_offload value. > + * @param il2 > + * l2_len value. > + * @param il3 > + * l3_len value. > + * @param il4 > + * l4_len value. > + * @param tso > + * tso_segsz value. > + * @param ol3 > + * outer_l3_len value. > + * @param ol2 > + * outer_l2_len value. > + * @return > + * raw tx_offload value. > + */ > +static inline uint64_t > +rte_mbuf_tx_offload(uint64_t il2, uint64_t il3, uint64_t il4, uint64_t t= so, > + uint64_t ol3, uint64_t ol2) > +{ > + return il2 << RTE_MBUF_L2_LEN_OFS | > + il3 << RTE_MBUF_L3_LEN_OFS | > + il4 << RTE_MBUF_L4_LEN_OFS | > + tso << RTE_MBUF_TSO_SEGSZ_OFS | > + ol3 << RTE_MBUF_OL3_LEN_OFS | > + ol2 << RTE_MBUF_OL2_LEN_OFS; > +} > + > /** > * Validate general requirements for Tx offload in mbuf. > * > --=20 > 2.17.1 >=20 Regards, Keith