DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Zhang, Qi Z" <qi.z.zhang@intel.com>
To: "Guo, Jia" <jia.guo@intel.com>,
	"Yang, Qiming" <qiming.yang@intel.com>,
	"Xing, Beilei" <beilei.xing@intel.com>,
	"Wu, Jingjing" <jingjing.wu@intel.com>,
	"Wang, Haiyue" <haiyue.wang@intel.com>
Cc: "Zhao1, Wei" <wei.zhao1@intel.com>,
	"Richardson, Bruce" <bruce.richardson@intel.com>,
	"dev@dpdk.org" <dev@dpdk.org>,
	"Zhang, Helin" <helin.zhang@intel.com>,
	"mb@smartsharesystems.com" <mb@smartsharesystems.com>,
	"Yigit, Ferruh" <ferruh.yigit@intel.com>,
	"stephen@networkplumber.org" <stephen@networkplumber.org>,
	"barbette@kth.se" <barbette@kth.se>,
	"Han, YingyaX" <yingyax.han@intel.com>
Subject: Re: [dpdk-dev] [PATCH v4 4/5] net/ice: fix vector rx burst for ice
Date: Thu, 17 Sep 2020 11:03:19 +0000
Message-ID: <a4c65097e1ca42609c74b020d48a5d3f@intel.com> (raw)
In-Reply-To: <20200917075834.60034-5-jia.guo@intel.com>



> -----Original Message-----
> From: Guo, Jia <jia.guo@intel.com>
> Sent: Thursday, September 17, 2020 3:59 PM
> To: Yang, Qiming <qiming.yang@intel.com>; Xing, Beilei
> <beilei.xing@intel.com>; Zhang, Qi Z <qi.z.zhang@intel.com>; Wu, Jingjing
> <jingjing.wu@intel.com>; Wang, Haiyue <haiyue.wang@intel.com>
> Cc: Zhao1, Wei <wei.zhao1@intel.com>; Richardson, Bruce
> <bruce.richardson@intel.com>; dev@dpdk.org; Guo, Jia <jia.guo@intel.com>;
> Zhang, Helin <helin.zhang@intel.com>; mb@smartsharesystems.com; Yigit,
> Ferruh <ferruh.yigit@intel.com>; stephen@networkplumber.org;
> barbette@kth.se; Han, YingyaX <yingyax.han@intel.com>
> Subject: [PATCH v4 4/5] net/ice: fix vector rx burst for ice
> 
> The limitation of burst size in vector rx was removed, since it should retrieve as
> much received packets as possible. And also the scattered receive path should
> use a wrapper function to achieve the goal of burst maximizing. And do some
> code cleaning for vector rx path.
> 
> Bugzilla ID: 516
> Fixes: c68a52b8b38c ("net/ice: support vector SSE in Rx")
> Fixes: ae60d3c9b227 ("net/ice: support Rx AVX2 vector")
> 
> Signed-off-by: Jeff Guo <jia.guo@intel.com>
> Tested-by: Yingya Han <yingyax.han@intel.com>
> ---
>  drivers/net/ice/ice_rxtx.h          |  1 +
>  drivers/net/ice/ice_rxtx_vec_avx2.c | 23 ++++++------
> drivers/net/ice/ice_rxtx_vec_sse.c  | 56 +++++++++++++++++++----------
>  3 files changed, 49 insertions(+), 31 deletions(-)
> 
> diff --git a/drivers/net/ice/ice_rxtx.h b/drivers/net/ice/ice_rxtx.h index
> 2fdcfb7d0..3ef5f300d 100644
> --- a/drivers/net/ice/ice_rxtx.h
> +++ b/drivers/net/ice/ice_rxtx.h
> @@ -35,6 +35,7 @@
>  #define ICE_MAX_RX_BURST            ICE_RXQ_REARM_THRESH
>  #define ICE_TX_MAX_FREE_BUF_SZ      64
>  #define ICE_DESCS_PER_LOOP          4
> +#define ICE_DESCS_PER_LOOP_AVX	    8

No need to expose this if no external link, better to keep all avx stuff inside avx.c

> 
>  #define ICE_FDIR_PKT_LEN	512
> 
> diff --git a/drivers/net/ice/ice_rxtx_vec_avx2.c
> b/drivers/net/ice/ice_rxtx_vec_avx2.c
> index be50677c2..843e4f32a 100644
> --- a/drivers/net/ice/ice_rxtx_vec_avx2.c
> +++ b/drivers/net/ice/ice_rxtx_vec_avx2.c
> @@ -29,7 +29,7 @@ ice_rxq_rearm(struct ice_rx_queue *rxq)
>  			__m128i dma_addr0;
> 
>  			dma_addr0 = _mm_setzero_si128();
> -			for (i = 0; i < ICE_DESCS_PER_LOOP; i++) {
> +			for (i = 0; i < ICE_DESCS_PER_LOOP_AVX; i++) {
>  				rxep[i].mbuf = &rxq->fake_mbuf;
>  				_mm_store_si128((__m128i *)&rxdp[i].read,
>  						dma_addr0);
> @@ -132,12 +132,17 @@ ice_rxq_rearm(struct ice_rx_queue *rxq)
>  	ICE_PCI_REG_WRITE(rxq->qrx_tail, rx_id);  }
> 
> +/**
> + * vPMD raw receive routine, only accept(nb_pkts >=
> +ICE_DESCS_PER_LOOP_AVX)
> + *
> + * Notice:
> + * - nb_pkts < ICE_DESCS_PER_LOOP_AVX, just return no packet
> + * - floor align nb_pkts to a ICE_DESCS_PER_LOOP_AVX power-of-two  */

The comment is misleading, it looks like we are going to floor align nb_pkts to 2^8, better to reword .

>  static inline uint16_t
>  _ice_recv_raw_pkts_vec_avx2(struct ice_rx_queue *rxq, struct rte_mbuf
> **rx_pkts,
>  			    uint16_t nb_pkts, uint8_t *split_packet)  { -#define
> ICE_DESCS_PER_LOOP_AVX 8
> -
>  	const uint32_t *ptype_tbl = rxq->vsi->adapter->ptype_tbl;
>  	const __m256i mbuf_init = _mm256_set_epi64x(0, 0,
>  			0, rxq->mbuf_initializer);
> @@ -603,10 +608,6 @@ _ice_recv_raw_pkts_vec_avx2(struct ice_rx_queue
> *rxq, struct rte_mbuf **rx_pkts,
>  	return received;
>  }
> 
> -/*
> - * Notice:
> - * - nb_pkts < ICE_DESCS_PER_LOOP, just return no packet
> - */
>  uint16_t
>  ice_recv_pkts_vec_avx2(void *rx_queue, struct rte_mbuf **rx_pkts,
>  		       uint16_t nb_pkts)
> @@ -616,8 +617,6 @@ ice_recv_pkts_vec_avx2(void *rx_queue, struct
> rte_mbuf **rx_pkts,
> 
>  /**
>   * vPMD receive routine that reassembles single burst of 32 scattered
> packets
> - * Notice:
> - * - nb_pkts < ICE_DESCS_PER_LOOP, just return no packet
>   */

Why we need to remove this? is it still true for this function?

>  static uint16_t
>  ice_recv_scattered_burst_vec_avx2(void *rx_queue, struct rte_mbuf
> **rx_pkts, @@ -626,6 +625,9 @@ ice_recv_scattered_burst_vec_avx2(void
> *rx_queue, struct rte_mbuf **rx_pkts,
>  	struct ice_rx_queue *rxq = rx_queue;
>  	uint8_t split_flags[ICE_VPMD_RX_BURST] = {0};
> 
> +	/* split_flags only can support max of ICE_VPMD_RX_BURST */
> +	nb_pkts = RTE_MIN(nb_pkts, ICE_VPMD_RX_BURST);

Is this necessary?  the only consumer of this function is ice_recv_scattered_pkts_vec_avx2, 
I think nb_pkts <= ICE_VPMD_RX_BURST it already be guaranteed.
> +
>  	/* get some new buffers */
>  	uint16_t nb_bufs = _ice_recv_raw_pkts_vec_avx2(rxq, rx_pkts, nb_pkts,
>  						       split_flags);
> @@ -657,9 +659,6 @@ ice_recv_scattered_burst_vec_avx2(void *rx_queue,
> struct rte_mbuf **rx_pkts,
> 
>  /**
>   * vPMD receive routine that reassembles scattered packets.
> - * Main receive routine that can handle arbitrary burst sizes
> - * Notice:
> - * - nb_pkts < ICE_DESCS_PER_LOOP, just return no packet
>   */

Why we need to remove this? isn't it the main routine that be able to handle arbitrary burst size?

Btw, I will suggest all AVX2 changes can be in a separate patch, because this looks like some code clean and fix.
its not related with the main purpose of the patch set.

  reply	other threads:[~2020-09-17 11:03 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-08-27  7:54 [dpdk-dev] [PATCH v1 0/4] maximize vector rx burst for PMDs Jeff Guo
2020-08-27  7:54 ` [dpdk-dev] [PATCH v1 1/4] net/ixgbe: maximize vector rx burst for ixgbe Jeff Guo
2020-08-27  7:54 ` [dpdk-dev] [PATCH v1 2/4] net/i40e: maximize vector rx burst for i40e Jeff Guo
2020-08-27  7:54 ` [dpdk-dev] [PATCH v1 3/4] net/ice: maximize vector rx burst for ice Jeff Guo
2020-08-27  7:54 ` [dpdk-dev] [PATCH v1 4/4] net/iavf: maximize vector rx burst for iavf Jeff Guo
2020-08-27  8:40 ` [dpdk-dev] [RFC] ethdev: rte_eth_rx_burst() requirements for nb_pkts Morten Brørup
2020-08-27  9:09   ` Bruce Richardson
2020-08-27  9:31     ` Morten Brørup
2020-08-27  9:43       ` Bruce Richardson
2020-08-27 10:13         ` [dpdk-dev] [RFC] ethdev: rte_eth_rx_burst() requirements fornb_pkts Morten Brørup
2020-08-27 11:41           ` Bruce Richardson
2020-08-28  9:03             ` Morten Brørup
2020-08-28 10:07               ` Bruce Richardson
2020-08-28 10:50                 ` Morten Brørup
2020-08-29 10:15                 ` Morten Brørup
2020-09-09  6:36 ` [dpdk-dev] [PATCH v3 0/5] fix vector rx burst for PMDs Jeff Guo
2020-09-09  6:36   ` [dpdk-dev] [PATCH v3 1/5] net/iavf: fix vector rx burst for iavf Jeff Guo
2020-09-09  6:36   ` [dpdk-dev] [PATCH v3 2/5] net/ixgbe: fix vector rx burst for ixgbe Jeff Guo
     [not found]     ` <VI1PR0802MB23518C6B517B6EAD8E018CD49E260@VI1PR0802MB2351.eurprd08.prod.outlook.com>
2020-09-09  9:54       ` [dpdk-dev] 回复: " Feifei Wang
2020-09-09  6:36   ` [dpdk-dev] [PATCH v3 3/5] net/i40e: fix vector rx burst for i40e Jeff Guo
2020-09-09  6:36   ` [dpdk-dev] [PATCH v3 4/5] net/ice: fix vector rx burst for ice Jeff Guo
2020-09-15  7:10     ` Han, YingyaX
2020-09-09  6:36   ` [dpdk-dev] [PATCH v3 5/5] net/fm10k: fix vector rx burst for fm10k Jeff Guo
2020-09-09  6:45   ` [dpdk-dev] [PATCH v3 0/5] fix vector rx burst for PMDs Wang, Haiyue
2020-09-09  7:03     ` Guo, Jia
2020-09-09  7:05       ` Wang, Haiyue
2020-09-09  7:43         ` Morten Brørup
2020-09-09  7:55           ` Wang, Haiyue
2020-09-09  8:01             ` Guo, Jia
2020-09-17  7:58 ` [dpdk-dev] [PATCH v4 " Jeff Guo
2020-09-17  7:58   ` [dpdk-dev] [PATCH v4 1/5] net/iavf: fix vector rx burst for iavf Jeff Guo
2020-09-17  7:58   ` [dpdk-dev] [PATCH v4 2/5] net/ixgbe: fix vector rx burst for ixgbe Jeff Guo
2020-09-17  7:58   ` [dpdk-dev] [PATCH v4 3/5] net/i40e: fix vector rx burst for i40e Jeff Guo
2020-09-17  7:58   ` [dpdk-dev] [PATCH v4 4/5] net/ice: fix vector rx burst for ice Jeff Guo
2020-09-17 11:03     ` Zhang, Qi Z [this message]
2020-09-18  3:20       ` Guo, Jia
2020-09-18  3:41         ` Zhang, Qi Z
2020-09-18  4:41           ` Guo, Jia
2020-09-18  5:39             ` Zhang, Qi Z
2020-09-17  7:58   ` [dpdk-dev] [PATCH v4 5/5] net/fm10k: fix vector rx burst for fm10k Jeff Guo
2020-10-16  9:44 ` [dpdk-dev] [PATCH v5 0/5] fix vector rx burst for PMDs Jeff Guo
2020-10-16  9:44   ` [dpdk-dev] [PATCH v5 1/5] net/ixgbe: fix vector rx burst for ixgbe Jeff Guo
2020-10-16  9:44   ` [dpdk-dev] [PATCH v5 2/5] net/i40e: fix vector rx burst for i40e Jeff Guo
2020-10-16  9:44   ` [dpdk-dev] [PATCH v5 3/5] net/ice: fix vector rx burst for ice Jeff Guo
2020-10-16  9:44   ` [dpdk-dev] [PATCH v5 4/5] net/fm10k: fix vector rx burst for fm10k Jeff Guo
2020-10-16  9:44   ` [dpdk-dev] [PATCH v5 5/5] net/iavf: fix vector rx burst for iavf Jeff Guo
2020-10-23  5:09     ` Ling, WeiX
2020-10-23 10:11   ` [dpdk-dev] [PATCH v5 0/5] fix vector rx burst for PMDs Zhang, Qi Z

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=a4c65097e1ca42609c74b020d48a5d3f@intel.com \
    --to=qi.z.zhang@intel.com \
    --cc=barbette@kth.se \
    --cc=beilei.xing@intel.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    --cc=ferruh.yigit@intel.com \
    --cc=haiyue.wang@intel.com \
    --cc=helin.zhang@intel.com \
    --cc=jia.guo@intel.com \
    --cc=jingjing.wu@intel.com \
    --cc=mb@smartsharesystems.com \
    --cc=qiming.yang@intel.com \
    --cc=stephen@networkplumber.org \
    --cc=wei.zhao1@intel.com \
    --cc=yingyax.han@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

DPDK patches and discussions

This inbox may be cloned and mirrored by anyone:

	git clone --mirror http://inbox.dpdk.org/dev/0 dev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 dev dev/ http://inbox.dpdk.org/dev \
		dev@dpdk.org
	public-inbox-index dev

Example config snippet for mirrors.
Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.dev


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git