DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ferruh Yigit <ferruh.yigit@intel.com>
To: "Min Hu (Connor)" <humin29@huawei.com>, dev@dpdk.org
Cc: Ciara Power <ciara.power@intel.com>,
	David Marchand <david.marchand@redhat.com>
Subject: Re: [dpdk-dev] [PATCH v2 1/9] net/hns3: support runtime config to select IO burst func
Date: Thu, 11 Mar 2021 17:31:36 +0000	[thread overview]
Message-ID: <ee20d440-aabe-53b5-ea62-ca57413d38c8@intel.com> (raw)
In-Reply-To: <1615356985-24722-2-git-send-email-humin29@huawei.com>

On 3/10/2021 6:16 AM, Min Hu (Connor) wrote:
> From: Chengwen Feng <fengchengwen@huawei.com>
> 
> Currently, the driver support multiple IO burst function and auto
> selection of the most appropriate function based on offload
> configuration.
> 
> Most applications such as l2fwd/l3fwd don't provide the means to
> change offload configuration, so it will use the auto selection's io
> burst function.
> 
> This patch support runtime config to select io burst function, which
> add two config: rx_func_hint and tx_func_hint, both could assign
> vec/sve/simple/common.
> 
> The driver will use the following rules to select io burst func:
> a. if hint equal vec and meet the vec Rx/Tx usage condition then use
> the neon function.
> b. if hint equal sve and meet the sve Rx/Tx usage condition then use
> the sve function.
> c. if hint equal simple and meet the simple Rx/Tx usage condition then
> use the simple function.
> d. if hint equal common then use the common function.
> e. if hint not set then:
> e.1. if meet the vec Rx/Tx usage condition then use the neon function.
> e.2. if meet the simple Rx/Tx usage condition then use the simple
> function.
> e.3. else use the common function.
> 
> Note: the sve Rx/Tx usage condition based on the vec Rx/Tx usage
> condition and runtime environment (which must support SVE).
> 
> In the previous versions, driver will preferred use the sve function
> when meet the sve Rx/Tx usage condition, but in this case driver could
> get better performance if use the neon function.
> 
> Signed-off-by: Chengwen Feng <fengchengwen@huawei.com>
> Signed-off-by: Min Hu (Connor) <humin29@huawei.com>
> ---
>   doc/guides/rel_notes/release_21_05.rst |  1 +
>   drivers/net/hns3/hns3_ethdev.c         | 74 ++++++++++++++++++++++++++++++++++
>   drivers/net/hns3/hns3_ethdev.h         | 12 ++++++
>   drivers/net/hns3/hns3_ethdev_vf.c      |  1 +
>   drivers/net/hns3/hns3_rxtx.c           | 54 ++++++++++++++++++-------
>   5 files changed, 128 insertions(+), 14 deletions(-)
> 

Hi Connor,

There are two recently added APIs to enable user to select vector datapath:
'rte_vect_set_max_simd_bitwidth()'
'rte_vect_get_max_simd_bitwidth()'

Also there is an eal parameter around those APIs: '--force-max-simd-bitwidth'

I wonder these existing APIs can be utilized to select the vector path instead 
of adding device args to the PMD, can you please check?

Thanks,
ferruh

  reply	other threads:[~2021-03-11 17:32 UTC|newest]

Thread overview: 34+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-10  6:16 [dpdk-dev] [PATCH v2 0/9] features and bugfixes for hns3 Min Hu (Connor)
2021-03-10  6:16 ` [dpdk-dev] [PATCH v2 1/9] net/hns3: support runtime config to select IO burst func Min Hu (Connor)
2021-03-11 17:31   ` Ferruh Yigit [this message]
2021-03-12  0:59     ` Min Hu (Connor)
2021-03-11 17:39   ` Ferruh Yigit
2021-03-12  1:01     ` Min Hu (Connor)
2021-03-10  6:16 ` [dpdk-dev] [PATCH v2 2/9] net/hns3: support Tx push quick doorbell to improve perf Min Hu (Connor)
2021-03-11 18:05   ` Ferruh Yigit
2021-03-12 11:02     ` Min Hu (Connor)
2021-03-10  6:16 ` [dpdk-dev] [PATCH v2 3/9] net/hns3: support for outer UDP cksum Min Hu (Connor)
2021-03-10  6:16 ` [dpdk-dev] [PATCH v2 4/9] net/hns3: adjust the format of RAS related structures Min Hu (Connor)
2021-03-11 18:25   ` Ferruh Yigit
2021-03-12  1:51     ` Min Hu (Connor)
2021-03-12 10:02       ` Ferruh Yigit
2021-03-12 10:51         ` Min Hu (Connor)
2021-03-16 11:32           ` Min Hu (Connor)
2021-03-10  6:16 ` [dpdk-dev] [PATCH v2 5/9] net/hns3: delete redundant xstats RAS statistics Min Hu (Connor)
2021-03-10  6:16 ` [dpdk-dev] [PATCH v2 6/9] net/hns3: support imissed stats for PF/VF Min Hu (Connor)
2021-03-10  6:16 ` [dpdk-dev] [PATCH v2 7/9] net/hns3: support oerrors stats in PF Min Hu (Connor)
2021-03-10  6:16 ` [dpdk-dev] [PATCH v2 8/9] net/hns3: support query Tx descriptor status Min Hu (Connor)
2021-03-10  6:16 ` [dpdk-dev] [PATCH v2 9/9] net/hns3: support query Rx " Min Hu (Connor)
2021-03-12 11:51 ` [dpdk-dev] [PATCH v3 0/8] features and bugfixes for hns3 Min Hu (Connor)
2021-03-12 11:51   ` [dpdk-dev] [PATCH v3 1/8] net/hns3: support runtime config to select IO burst func Min Hu (Connor)
2021-03-16 12:40     ` Ferruh Yigit
2021-03-17  1:14       ` Min Hu (Connor)
2021-03-17 15:28         ` Ferruh Yigit
2021-03-19  1:08           ` Min Hu (Connor)
2021-03-12 11:51   ` [dpdk-dev] [PATCH v3 2/8] net/hns3: support for outer UDP cksum Min Hu (Connor)
2021-03-12 11:51   ` [dpdk-dev] [PATCH v3 3/8] net/hns3: adjust the format of RAS related structures Min Hu (Connor)
2021-03-12 11:51   ` [dpdk-dev] [PATCH v3 4/8] net/hns3: delete redundant xstats RAS statistics Min Hu (Connor)
2021-03-12 11:51   ` [dpdk-dev] [PATCH v3 5/8] net/hns3: support imissed stats for PF/VF Min Hu (Connor)
2021-03-12 11:51   ` [dpdk-dev] [PATCH v3 6/8] net/hns3: support oerrors stats in PF Min Hu (Connor)
2021-03-12 11:51   ` [dpdk-dev] [PATCH v3 7/8] net/hns3: support query Tx descriptor status Min Hu (Connor)
2021-03-12 11:51   ` [dpdk-dev] [PATCH v3 8/8] net/hns3: support query Rx " Min Hu (Connor)

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=ee20d440-aabe-53b5-ea62-ca57413d38c8@intel.com \
    --to=ferruh.yigit@intel.com \
    --cc=ciara.power@intel.com \
    --cc=david.marchand@redhat.com \
    --cc=dev@dpdk.org \
    --cc=humin29@huawei.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).