DPDK patches and discussions
 help / color / mirror / Atom feed
From: Olivier MATZ <olivier.matz@6wind.com>
To: Bruce Richardson <bruce.richardson@intel.com>, dev@dpdk.org
Subject: Re: [dpdk-dev] [RFC PATCH 09/14] Fix performance regression due to moved pool ptr
Date: Tue, 12 Aug 2014 13:28:24 +0200	[thread overview]
Message-ID: <53E9FA58.1060600@6wind.com> (raw)
In-Reply-To: <1407789890-17355-10-git-send-email-bruce.richardson@intel.com>

Hi Bruce,

On 08/11/2014 10:44 PM, Bruce Richardson wrote:
> Adjust the fast-path code to fix the regression caused by the pool
> pointer moving to the second cache line. This change adjusts the
> prefetching and also the way in which the mbufs are freed back to the
> mempool.
>
> Signed-off-by: Bruce Richardson <bruce.richardson@intel.com>

Just one comment here (maybe this code should be reviewed by someone
knowing the ixgbe driver better than me):


> @@ -252,14 +250,6 @@ tx_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
>   	volatile union ixgbe_adv_tx_desc *tx_r = txq->tx_ring;
>   	uint16_t n = 0;
>
> -	/*
> -	 * Begin scanning the H/W ring for done descriptors when the
> -	 * number of available descriptors drops below tx_free_thresh.  For
> -	 * each done descriptor, free the associated buffer.
> -	 */
> -	if (txq->nb_tx_free < txq->tx_free_thresh)
> -		ixgbe_tx_free_bufs(txq);
> -
>   	/* Only use descriptors that are available */
>   	nb_pkts = (uint16_t)RTE_MIN(txq->nb_tx_free, nb_pkts);
>   	if (unlikely(nb_pkts == 0))
> @@ -323,6 +313,15 @@ tx_xmit_pkts(void *tx_queue, struct rte_mbuf **tx_pkts,
>   	if (txq->tx_tail >= txq->nb_tx_desc)
>   		txq->tx_tail = 0;
>
> +	/*
> +	 * Begin scanning the H/W ring for done descriptors when the
> +	 * number of available descriptors drops below tx_free_thresh.  For
> +	 * each done descriptor, free the associated buffer.
> +	 */
> +	if (txq->nb_tx_free < txq->tx_free_thresh)
> +		ixgbe_tx_free_bufs(txq);
> +
> +
>   	/* update tail pointer */
>   	rte_wmb();
>   	IXGBE_PCI_REG_WRITE(txq->tdt_reg_addr, txq->tx_tail);

It looks like these 2 hunks are reverted in next commit. I'm not sure
this is what you expected.

  reply	other threads:[~2014-08-12 11:25 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2014-08-11 20:44 [dpdk-dev] [RFC PATCH 00/14] Extend the mbuf structure Bruce Richardson
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 01/14] mbuf: rename RTE_MBUF_SCATTER_GATHER into RTE_MBUF_REFCNT Bruce Richardson
2014-08-11 21:45   ` Stephen Hemminger
2014-08-12  7:59     ` Olivier MATZ
2014-08-12 16:25       ` Richardson, Bruce
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 02/14] mbuf: remove rte_ctrlmbuf Bruce Richardson
2014-08-12  8:27   ` Olivier MATZ
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 03/14] mbuf: remove the rte_pktmbuf structure Bruce Richardson
2014-08-12  8:38   ` Olivier MATZ
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 04/14] mbuf: replace data pointer by an offset Bruce Richardson
2014-08-12  8:55   ` Olivier MATZ
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 05/14] mbuf: rename in_port to just port Bruce Richardson
2014-08-12  9:00   ` Olivier MATZ
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 06/14] mbuf: reorder fields by time-of-use Bruce Richardson
2014-08-12 10:03   ` Olivier MATZ
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 07/14] ixgbe: rework vector pmd following mbuf changes Bruce Richardson
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 08/14] mbuf: split mbuf across two cache lines Bruce Richardson
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 09/14] Fix performance regression due to moved pool ptr Bruce Richardson
2014-08-12 11:28   ` Olivier MATZ [this message]
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 10/14] mbuf: set next pointer to NULL on mbuf free Bruce Richardson
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 11/14] ixgbe: make mbuf_initializer queue variable global Bruce Richardson
2014-08-11 21:47   ` Stephen Hemminger
2014-08-11 22:03     ` Richardson, Bruce
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 12/14] ixgbe: Make vector stores unaligned Bruce Richardson
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 13/14] mbuf: cleanup + added in additional mbuf fields Bruce Richardson
2014-08-12 14:18   ` Olivier MATZ
2014-08-11 20:44 ` [dpdk-dev] [RFC PATCH 14/14] ixgbe: Allow vector RX of scattered packets Bruce Richardson
2014-08-12 14:43 ` [dpdk-dev] [RFC PATCH 00/14] Extend the mbuf structure Olivier MATZ
2014-08-20 12:22   ` Richardson, Bruce
2014-08-20  7:08 ` Cao, Min
2014-08-20 11:11   ` Richardson, Bruce

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=53E9FA58.1060600@6wind.com \
    --to=olivier.matz@6wind.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    /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).