DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>
To: Shahaf Shuler <shahafs@mellanox.com>, "dev@dpdk.org" <dev@dpdk.org>
Cc: "akhil.goyal@nxp.com" <akhil.goyal@nxp.com>,
	"olivier.matz@6wind.com" <olivier.matz@6wind.com>
Subject: Re: [dpdk-dev] [PATCH v2 1/7] mbuf: new function to generate raw Tx	offload value
Date: Thu, 21 Mar 2019 13:51:39 +0000	[thread overview]
Message-ID: <2601191342CEEE43887BDE71AB977258013655E4F4@irsmsx105.ger.corp.intel.com> (raw)
Message-ID: <20190321135139.uR6lASq9gGBMsTHjgA_NdXOZSpCLEBDUgnAqXaGT7Iw@z> (raw)
In-Reply-To: <AM0PR0502MB379548ABEC4CE329A7AEAC61C3420@AM0PR0502MB3795.eurprd05.prod.outlook.com>

Hi Shahaf,

> Hi Konstantin,
> 
> Wednesday, March 20, 2019 8:47 PM, Konstantin Ananyev:
> > Subject: [dpdk-dev] [PATCH v2 1/7] mbuf: new function to generate raw Tx
> > offload value
> >
> > 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.
> >
> > Signed-off-by: Konstantin Ananyev <konstantin.ananyev@intel.com>
> > ---
> >  lib/librte_mbuf/rte_mbuf.h | 71
> > ++++++++++++++++++++++++++++++++++----
> >  1 file changed, 64 insertions(+), 7 deletions(-)
> >
> > 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 */
> >
> > +/** enum for the tx_offload bit-fields lenghts and offsets. */ enum {
> > +	RTE_MBUF_L2_LEN_BITS = 7,
> > +	RTE_MBUF_L3_LEN_BITS = 9,
> > +	RTE_MBUF_L4_LEN_BITS = 8,
> > +	RTE_MBUF_TSO_SEGSZ_BITS = 16,
> > +	RTE_MBUF_OL3_LEN_BITS = 9,
> > +	RTE_MBUF_OL2_LEN_BITS = 7,
> > +	RTE_MBUF_L2_LEN_OFS = 0,
> > +	RTE_MBUF_L3_LEN_OFS = RTE_MBUF_L2_LEN_OFS +
> > RTE_MBUF_L2_LEN_BITS,
> > +	RTE_MBUF_L4_LEN_OFS = RTE_MBUF_L3_LEN_OFS +
> > RTE_MBUF_L3_LEN_BITS,
> > +	RTE_MBUF_TSO_SEGSZ_OFS = RTE_MBUF_L4_LEN_OFS +
> > RTE_MBUF_L4_LEN_BITS,
> > +	RTE_MBUF_OL3_LEN_OFS = RTE_MBUF_TSO_SEGSZ_OFS +
> > RTE_MBUF_TSO_SEGSZ_BITS,
> > +	RTE_MBUF_OL2_LEN_OFS = RTE_MBUF_OL3_LEN_OFS +
> > RTE_MBUF_OL3_LEN_BITS,
> > +	RTE_MBUF_TXOFLD_UNUSED_OFS =
> > +		RTE_MBUF_OL2_LEN_OFS + RTE_MBUF_OL2_LEN_BITS,
> > +	RTE_MBUF_TXOFLD_UNUSED_BITS =
> > +		sizeof(uint64_t) * CHAR_BIT -
> > RTE_MBUF_TXOFLD_UNUSED_OFS, };
> > +
> >  /**
> >   * 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 */
> >
> >  			/* 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. */
> >
> > -			/* uint64_t unused:8; */
> > +			/* uint64_t
> > unused:RTE_MBUF_TXOFLD_UNUSED_BITS; */
> 
> I am not sure we want to decide now that this remaining byte is for another Tx offloads.

Not sure, I understand what is your concern here...
tx_offload field is 8B long, see 'uint64_t tx_offload' above.
All I did replaced raw number '8' with enum value ' RTE_MBUF_TXOFLD_UNUSED_BITS' which also evalueate to 8.

> Maybe it is better to change a bit the rte_mbuf_tx_offload API to receive also the mbuf and to assign the offload values w/ mask.

Could you explain a bit more what do you suggest here?
Konstantin


> 
> >  		};
> >  	};
> >
> > @@ -2243,6 +2268,38 @@ static inline int rte_pktmbuf_chain(struct
> > rte_mbuf *head, struct rte_mbuf *tail
> >  	return 0;
> >  }
> >
> > +/*
> > + * @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 tso,
> > +	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.
> >   *
> > --
> > 2.17.1


  parent reply	other threads:[~2019-03-21 13:51 UTC|newest]

Thread overview: 163+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-02-28 19:20 [dpdk-dev] [PATCH 0/6] Few small improvements for ipsec library Konstantin Ananyev
2019-02-28 19:20 ` [dpdk-dev] [PATCH 1/6] mbuf: new function to generate raw Tx offload value Konstantin Ananyev
2019-02-28 19:20 ` [dpdk-dev] [PATCH 2/6] ipsec: add Tx offload template into SA Konstantin Ananyev
2019-02-28 19:20 ` [dpdk-dev] [PATCH 3/6] ipsec: change the order in filling crypto op Konstantin Ananyev
2019-02-28 19:20 ` [dpdk-dev] [PATCH 4/6] ipsec: change the way unprocessed mbufs are accounted Konstantin Ananyev
2019-02-28 19:21 ` [dpdk-dev] [PATCH 5/6] ipsec: move inbound and outbound code into different files Konstantin Ananyev
2019-02-28 19:21 ` [dpdk-dev] [PATCH 6/6] ipsec: reorder packet check for esp inbound Konstantin Ananyev
2019-03-20 17:24 ` [dpdk-dev] [PATCH v2 0/7] Few small improvements for ipsec library Konstantin Ananyev
2019-03-20 17:24   ` Konstantin Ananyev
2019-03-20 17:24 ` [dpdk-dev] [PATCH v2 1/7] mbuf: new function to generate raw Tx offload value Konstantin Ananyev
2019-03-20 17:24   ` Konstantin Ananyev
2019-03-20 17:53   ` Wiles, Keith
2019-03-20 17:53     ` Wiles, Keith
2019-03-22 17:37     ` Ananyev, Konstantin
2019-03-22 17:37       ` Ananyev, Konstantin
2019-03-20 17:24 ` [dpdk-dev] [PATCH v2 2/7] ipsec: add Tx offload template into SA Konstantin Ananyev
2019-03-20 17:24   ` Konstantin Ananyev
2019-03-20 17:24 ` [dpdk-dev] [PATCH v2 3/7] ipsec: change the order in filling crypto op Konstantin Ananyev
2019-03-20 17:24   ` Konstantin Ananyev
2019-03-20 17:24 ` [dpdk-dev] [PATCH v2 4/7] ipsec: change the way unprocessed mbufs are accounted Konstantin Ananyev
2019-03-20 17:24   ` Konstantin Ananyev
2019-03-20 17:24 ` [dpdk-dev] [PATCH v2 5/7] ipsec: move inbound and outbound code into different files Konstantin Ananyev
2019-03-20 17:24   ` Konstantin Ananyev
2019-03-20 17:24 ` [dpdk-dev] [PATCH v2 6/7] ipsec: reorder packet check for esp inbound Konstantin Ananyev
2019-03-20 17:24   ` Konstantin Ananyev
2019-03-20 17:24 ` [dpdk-dev] [PATCH v2 7/7] ipsec: reorder packet process " Konstantin Ananyev
2019-03-20 17:24   ` Konstantin Ananyev
2019-03-20 18:46 ` [dpdk-dev] [PATCH v2 0/7] Few small improvements for ipsec library Konstantin Ananyev
2019-03-20 18:46   ` Konstantin Ananyev
2019-03-20 18:46   ` [dpdk-dev] [PATCH v2 1/7] mbuf: new function to generate raw Tx offload value Konstantin Ananyev
2019-03-20 18:46     ` Konstantin Ananyev
2019-03-21  3:33     ` Jerin Jacob Kollanukkaran
2019-03-21  3:33       ` Jerin Jacob Kollanukkaran
2019-03-21  6:04     ` Shahaf Shuler
2019-03-21  6:04       ` Shahaf Shuler
2019-03-21 13:51       ` Ananyev, Konstantin [this message]
2019-03-21 13:51         ` Ananyev, Konstantin
2019-03-24  8:00         ` Shahaf Shuler
2019-03-24  8:00           ` Shahaf Shuler
2019-03-26 15:43     ` [dpdk-dev] [PATCH v3 0/8] Few small improvements for ipsec library Konstantin Ananyev
2019-03-26 15:43       ` Konstantin Ananyev
2019-03-26 15:43       ` [dpdk-dev] [PATCH v3 1/8] mbuf: new function to generate raw Tx offload value Konstantin Ananyev
2019-03-26 15:43         ` Konstantin Ananyev
2019-03-28  8:16         ` Akhil Goyal
2019-03-28  8:16           ` Akhil Goyal
2019-03-26 15:43       ` [dpdk-dev] [PATCH v3 2/8] ipsec: add Tx offload template into SA Konstantin Ananyev
2019-03-26 15:43         ` Konstantin Ananyev
2019-03-28  8:52         ` Akhil Goyal
2019-03-28  8:52           ` Akhil Goyal
2019-03-26 15:43       ` [dpdk-dev] [PATCH v3 3/8] ipsec: change the order in filling crypto op Konstantin Ananyev
2019-03-26 15:43         ` Konstantin Ananyev
2019-03-28  9:02         ` Akhil Goyal
2019-03-28  9:02           ` Akhil Goyal
2019-03-26 15:43       ` [dpdk-dev] [PATCH v3 4/8] ipsec: change the way unprocessed mbufs are accounted Konstantin Ananyev
2019-03-26 15:43         ` Konstantin Ananyev
2019-03-28 10:52         ` Akhil Goyal
2019-03-28 10:52           ` Akhil Goyal
2019-03-26 15:43       ` [dpdk-dev] [PATCH v3 5/8] ipsec: move inbound and outbound code into different files Konstantin Ananyev
2019-03-26 15:43         ` Konstantin Ananyev
2019-03-28 11:20         ` Akhil Goyal
2019-03-28 11:20           ` Akhil Goyal
2019-03-26 15:43       ` [dpdk-dev] [PATCH v3 6/8] ipsec: reorder packet check for esp inbound Konstantin Ananyev
2019-03-26 15:43         ` Konstantin Ananyev
2019-03-28 11:27         ` Akhil Goyal
2019-03-28 11:27           ` Akhil Goyal
2019-03-26 15:43       ` [dpdk-dev] [PATCH v3 7/8] ipsec: reorder packet process " Konstantin Ananyev
2019-03-26 15:43         ` Konstantin Ananyev
2019-03-26 15:43       ` [dpdk-dev] [PATCH v3 8/8] ipsec: de-duplicate crypto op prepare code-path Konstantin Ananyev
2019-03-26 15:43         ` Konstantin Ananyev
2019-03-28 11:35         ` Akhil Goyal
2019-03-28 11:35           ` Akhil Goyal
2019-03-28 11:21       ` [dpdk-dev] [PATCH v3 0/8] Few small improvements for ipsec library Akhil Goyal
2019-03-28 11:21         ` Akhil Goyal
2019-03-28 11:49         ` Ananyev, Konstantin
2019-03-28 11:49           ` Ananyev, Konstantin
2019-03-29 10:27       ` [dpdk-dev] [PATCH v4 0/9] " Konstantin Ananyev
2019-03-29 10:27         ` Konstantin Ananyev
2019-03-29 10:27         ` [dpdk-dev] [PATCH v4 1/9] mbuf: new function to generate raw Tx offload value Konstantin Ananyev
2019-03-29 10:27           ` Konstantin Ananyev
2019-03-29 12:54           ` Olivier Matz
2019-03-29 12:54             ` Olivier Matz
2019-03-30 14:20             ` Ananyev, Konstantin
2019-03-30 14:20               ` Ananyev, Konstantin
2019-03-29 10:27         ` [dpdk-dev] [PATCH v4 2/9] ipsec: add Tx offload template into SA Konstantin Ananyev
2019-03-29 10:27           ` Konstantin Ananyev
2019-03-29 10:27         ` [dpdk-dev] [PATCH v4 3/9] ipsec: change the order in filling crypto op Konstantin Ananyev
2019-03-29 10:27           ` Konstantin Ananyev
2019-03-29 10:27         ` [dpdk-dev] [PATCH v4 4/9] ipsec: change the way unprocessed mbufs are accounted Konstantin Ananyev
2019-03-29 10:27           ` Konstantin Ananyev
2019-03-29 10:27         ` [dpdk-dev] [PATCH v4 5/9] ipsec: move inbound and outbound code into different files Konstantin Ananyev
2019-03-29 10:27           ` Konstantin Ananyev
2019-03-29 10:27         ` [dpdk-dev] [PATCH v4 6/9] ipsec: reorder packet check for esp inbound Konstantin Ananyev
2019-03-29 10:27           ` Konstantin Ananyev
2019-03-29 10:27         ` [dpdk-dev] [PATCH v4 7/9] ipsec: reorder packet process " Konstantin Ananyev
2019-03-29 10:27           ` Konstantin Ananyev
2019-03-29 10:27         ` [dpdk-dev] [PATCH v4 8/9] ipsec: de-duplicate crypto op prepare code-path Konstantin Ananyev
2019-03-29 10:27           ` Konstantin Ananyev
2019-03-29 10:27         ` [dpdk-dev] [PATCH v4 9/9] doc: add ipsec lib into shared libraries list Konstantin Ananyev
2019-03-29 10:27           ` Konstantin Ananyev
2019-03-29 16:03           ` Akhil Goyal
2019-03-29 16:03             ` Akhil Goyal
2019-04-01 12:56         ` [dpdk-dev] [PATCH v5 0/9] Few small improvements for ipsec library Konstantin Ananyev
2019-04-01 12:56           ` Konstantin Ananyev
2019-04-01 12:56           ` [dpdk-dev] [PATCH v5 1/9] mbuf: new function to generate raw Tx offload value Konstantin Ananyev
2019-04-01 12:56             ` Konstantin Ananyev
2019-04-01 13:18             ` Akhil Goyal
2019-04-01 13:18               ` Akhil Goyal
2019-04-01 13:22             ` Olivier Matz
2019-04-01 13:22               ` Olivier Matz
2019-04-01 13:55               ` Ananyev, Konstantin
2019-04-01 13:55                 ` Ananyev, Konstantin
2019-04-01 12:56           ` [dpdk-dev] [PATCH v5 2/9] ipsec: add Tx offload template into SA Konstantin Ananyev
2019-04-01 12:56             ` Konstantin Ananyev
2019-04-01 12:56           ` [dpdk-dev] [PATCH v5 3/9] ipsec: change the order in filling crypto op Konstantin Ananyev
2019-04-01 12:56             ` Konstantin Ananyev
2019-04-01 12:56           ` [dpdk-dev] [PATCH v5 4/9] ipsec: change the way unprocessed mbufs are accounted Konstantin Ananyev
2019-04-01 12:56             ` Konstantin Ananyev
2019-04-01 12:56           ` [dpdk-dev] [PATCH v5 5/9] ipsec: move inbound and outbound code into different files Konstantin Ananyev
2019-04-01 12:56             ` Konstantin Ananyev
2019-04-01 12:56           ` [dpdk-dev] [PATCH v5 6/9] ipsec: reorder packet check for esp inbound Konstantin Ananyev
2019-04-01 12:56             ` Konstantin Ananyev
2019-04-01 12:56           ` [dpdk-dev] [PATCH v5 7/9] ipsec: reorder packet process " Konstantin Ananyev
2019-04-01 12:56             ` Konstantin Ananyev
2019-04-01 12:56           ` [dpdk-dev] [PATCH v5 8/9] ipsec: de-duplicate crypto op prepare code-path Konstantin Ananyev
2019-04-01 12:56             ` Konstantin Ananyev
2019-04-01 12:56           ` [dpdk-dev] [PATCH v5 9/9] doc: add ipsec lib into shared libraries list Konstantin Ananyev
2019-04-01 12:56             ` Konstantin Ananyev
2019-04-02  8:34           ` [dpdk-dev] [PATCH v6 0/9] Few small improvements for ipsec library Konstantin Ananyev
2019-04-02  8:34             ` Konstantin Ananyev
2019-04-02  8:34             ` [dpdk-dev] [PATCH v6 1/9] mbuf: new function to generate raw Tx offload value Konstantin Ananyev
2019-04-02  8:34               ` Konstantin Ananyev
2019-04-02  8:49               ` Olivier Matz
2019-04-02  8:49                 ` Olivier Matz
2019-04-02  8:34             ` [dpdk-dev] [PATCH v6 2/9] ipsec: add Tx offload template into SA Konstantin Ananyev
2019-04-02  8:34               ` Konstantin Ananyev
2019-04-02  8:34             ` [dpdk-dev] [PATCH v6 3/9] ipsec: change the order in filling crypto op Konstantin Ananyev
2019-04-02  8:34               ` Konstantin Ananyev
2019-04-02  8:34             ` [dpdk-dev] [PATCH v6 4/9] ipsec: change the way unprocessed mbufs are accounted Konstantin Ananyev
2019-04-02  8:34               ` Konstantin Ananyev
2019-04-02  8:34             ` [dpdk-dev] [PATCH v6 5/9] ipsec: move inbound and outbound code into different files Konstantin Ananyev
2019-04-02  8:34               ` Konstantin Ananyev
2019-04-02  8:34             ` [dpdk-dev] [PATCH v6 6/9] ipsec: reorder packet check for esp inbound Konstantin Ananyev
2019-04-02  8:34               ` Konstantin Ananyev
2019-04-02  8:34             ` [dpdk-dev] [PATCH v6 7/9] ipsec: reorder packet process " Konstantin Ananyev
2019-04-02  8:34               ` Konstantin Ananyev
2019-04-02  8:34             ` [dpdk-dev] [PATCH v6 8/9] ipsec: de-duplicate crypto op prepare code-path Konstantin Ananyev
2019-04-02  8:34               ` Konstantin Ananyev
2019-04-02  8:34             ` [dpdk-dev] [PATCH v6 9/9] doc: add ipsec lib into shared libraries list Konstantin Ananyev
2019-04-02  8:34               ` Konstantin Ananyev
2019-04-02 15:36             ` [dpdk-dev] [PATCH v6 0/9] Few small improvements for ipsec library Akhil Goyal
2019-04-02 15:36               ` Akhil Goyal
2019-03-20 18:46   ` [dpdk-dev] [PATCH v2 2/7] ipsec: add Tx offload template into SA Konstantin Ananyev
2019-03-20 18:46     ` Konstantin Ananyev
2019-03-20 18:46   ` [dpdk-dev] [PATCH v2 3/7] ipsec: change the order in filling crypto op Konstantin Ananyev
2019-03-20 18:46     ` Konstantin Ananyev
2019-03-20 18:46   ` [dpdk-dev] [PATCH v2 4/7] ipsec: change the way unprocessed mbufs are accounted Konstantin Ananyev
2019-03-20 18:46     ` Konstantin Ananyev
2019-03-20 18:46   ` [dpdk-dev] [PATCH v2 5/7] ipsec: move inbound and outbound code into different files Konstantin Ananyev
2019-03-20 18:46     ` Konstantin Ananyev
2019-03-20 18:46   ` [dpdk-dev] [PATCH v2 6/7] ipsec: reorder packet check for esp inbound Konstantin Ananyev
2019-03-20 18:46     ` Konstantin Ananyev
2019-03-20 18:46   ` [dpdk-dev] [PATCH v2 7/7] ipsec: reorder packet process " Konstantin Ananyev
2019-03-20 18:46     ` Konstantin Ananyev

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=2601191342CEEE43887BDE71AB977258013655E4F4@irsmsx105.ger.corp.intel.com \
    --to=konstantin.ananyev@intel.com \
    --cc=akhil.goyal@nxp.com \
    --cc=dev@dpdk.org \
    --cc=olivier.matz@6wind.com \
    --cc=shahafs@mellanox.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).