DPDK patches and discussions
 help / color / mirror / Atom feed
From: Ferruh Yigit <ferruh.yigit@intel.com>
To: Leyi Rong <leyi.rong@intel.com>,
	qi.z.zhang@intel.com, wenzhuo.lu@intel.com,
	bruce.richardson@intel.com, beilei.xing@intel.com
Cc: dev@dpdk.org
Subject: Re: [dpdk-dev] [PATCH v3 1/3] net/i40e: remove devarg use-latest-supported-vec
Date: Fri, 15 Jan 2021 13:36:51 +0000
Message-ID: <51abc9ff-a247-cfa5-9a87-15656e424cc4@intel.com> (raw)
In-Reply-To: <20210114063951.2580-2-leyi.rong@intel.com>

On 1/14/2021 6:39 AM, Leyi Rong wrote:
> As eal parameter --force-max-simd-bitwidth is already introduced,
> to make it more clear when setting rx/tx function, remove
> devarg use-latest-supported-vec support.
> 
> Signed-off-by: Leyi Rong <leyi.rong@intel.com>
> Acked-by: Wenzhuo Lu <wenzhuo.lu@intel.com>

<...>

> -static eth_rx_burst_t
> -i40e_get_recommend_rx_vec(bool scatter)
> +static inline bool
> +get_avx_supported(bool request_avx512)
>   {
> -#if defined(RTE_ARCH_X86) && defined(CC_AVX2_SUPPORT)
> -	/*
> -	 * since AVX frequency can be different to base frequency, limit
> -	 * use of AVX2 version to later plaforms, not all those that could
> -	 * theoretically run it.
> -	 */
> -	if (rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) &&
> -			rte_vect_get_max_simd_bitwidth() >= RTE_VECT_SIMD_256)
> -		return scatter ? i40e_recv_scattered_pkts_vec_avx2 :
> -				 i40e_recv_pkts_vec_avx2;
> +#ifdef RTE_ARCH_X86
> +	if (request_avx512) {
> +		if (rte_vect_get_max_simd_bitwidth() >= RTE_VECT_SIMD_512 &&
> +		rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1 &&
> +		rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512BW) == 1)
> +#ifdef CC_AVX512_SUPPORT
> +			return true;
> +#else
> +		PMD_DRV_LOG(NOTICE,
> +			"AVX512 is not supported in build env");
> +		return false;
> +#endif
> +	} else {
> +		if (rte_vect_get_max_simd_bitwidth() >= RTE_VECT_SIMD_256 &&
> +		rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX2) == 1 &&
> +		rte_cpu_get_flag_enabled(RTE_CPUFLAG_AVX512F) == 1)
> +#ifdef CC_AVX2_SUPPORT
> +			return true;
> +#else
> +		PMD_DRV_LOG(NOTICE,
> +			"AVX2 is not supported in build env");
> +		return false;
>   #endif
> -	return scatter ? i40e_recv_scattered_pkts_vec :
> -			 i40e_recv_pkts_vec;
> +	}
> +#endif /* RTE_ARCH_X86 */
> +
> +	return false;
>   }

This was still causing build error for non x86, becuase of not used 
'request_avx512' variable, updated as below in next-net:

  diff --git a/drivers/net/i40e/i40e_rxtx.c b/drivers/net/i40e/i40e_rxtx.c
  index a4661cdd8377..ce2b0bc12ed9 100644
  --- a/drivers/net/i40e/i40e_rxtx.c
  +++ b/drivers/net/i40e/i40e_rxtx.c
  @@ -3122,6 +3122,8 @@ get_avx_supported(bool request_avx512)
                  return false;
   #endif
          }
  +#else
  +       RTE_SET_USED(request_avx512);
   #endif /* RTE_ARCH_X86 */

          return false;

  reply	other threads:[~2021-01-15 13:37 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-12-15  2:19 [dpdk-dev] [PATCH 0/3] AVX512 vPMD on i40e Leyi Rong
2020-12-15  2:19 ` [dpdk-dev] [PATCH 1/3] net/i40e: remove devarg use-latest-supported-vec Leyi Rong
2020-12-15  2:19 ` [dpdk-dev] [PATCH 2/3] net/i40e: add AVX512 vector path Leyi Rong
2020-12-15  2:19 ` [dpdk-dev] [PATCH 3/3] net/i40e: optimize Tx by using AVX512 Leyi Rong
2021-01-07  7:44 ` [dpdk-dev] [PATCH v2 0/3] AVX512 vPMD on i40e Leyi Rong
2021-01-07  7:44   ` [dpdk-dev] [PATCH v2 1/3] net/i40e: remove devarg use-latest-supported-vec Leyi Rong
2021-01-13  6:12     ` Lu, Wenzhuo
2021-01-13 13:40     ` Ferruh Yigit
2021-01-07  7:44   ` [dpdk-dev] [PATCH v2 2/3] net/i40e: add AVX512 vector path Leyi Rong
2021-01-13  6:13     ` Lu, Wenzhuo
2021-01-07  7:44   ` [dpdk-dev] [PATCH v2 3/3] net/i40e: optimize Tx by using AVX512 Leyi Rong
2021-01-13  6:12     ` Lu, Wenzhuo
2021-01-13  9:53   ` [dpdk-dev] [PATCH v2 0/3] AVX512 vPMD on i40e Zhang, Qi Z
2021-01-14  6:39 ` [dpdk-dev] [PATCH v3 " Leyi Rong
2021-01-14  6:39   ` [dpdk-dev] [PATCH v3 1/3] net/i40e: remove devarg use-latest-supported-vec Leyi Rong
2021-01-15 13:36     ` Ferruh Yigit [this message]
2021-01-14  6:39   ` [dpdk-dev] [PATCH v3 2/3] net/i40e: add AVX512 vector path Leyi Rong
2021-01-14  6:39   ` [dpdk-dev] [PATCH v3 3/3] net/i40e: optimize Tx by using AVX512 Leyi Rong
2021-01-14  7:37   ` [dpdk-dev] [PATCH v3 0/3] AVX512 vPMD on i40e Zhang, Qi Z
2021-01-17 11:26     ` Odi Assli
2021-01-18 13:58       ` Rong, Leyi
2021-01-18 14:24         ` Ferruh Yigit
2021-01-18 14:53           ` Odi Assli
2021-01-18 16:36             ` Ferruh Yigit
2021-01-19 13:46               ` Ali Alnubani
2021-01-20  6:25                 ` Tal Shnaiderman
2021-01-20  8:36                   ` David Marchand
2021-01-20  9:18                     ` Ferruh Yigit
2021-01-20  9:23                     ` Thomas Monjalon
2021-01-20  9:53                       ` David Marchand
2021-01-20 10:05                         ` Ali Alnubani
2021-01-20 17:51                           ` Ferruh Yigit
2021-01-20 18:04                             ` Ferruh Yigit
2021-01-21  5:01                             ` Kadam, Pallavi
2021-01-25 14:35                               ` David Marchand
2021-01-26 16:17                                 ` Rong, Leyi
2021-01-26 16:22                                   ` Thomas Monjalon
2021-01-26 16:39                                     ` Ferruh Yigit
2021-01-26 16:48                                       ` Thomas Monjalon
2021-01-26 16:51                                         ` Ferruh Yigit
2021-01-28 20:35                                           ` Dmitry Kozlyuk
2021-01-28 21:24                                             ` Thomas Monjalon

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=51abc9ff-a247-cfa5-9a87-15656e424cc4@intel.com \
    --to=ferruh.yigit@intel.com \
    --cc=beilei.xing@intel.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    --cc=leyi.rong@intel.com \
    --cc=qi.z.zhang@intel.com \
    --cc=wenzhuo.lu@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 https://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/ https://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