From: Ruifeng Wang <Ruifeng.Wang@arm.com>
To: Slava Ovsiienko <viacheslavo@nvidia.com>,
Raslan Darawsheh <rasland@nvidia.com>,
Matan Azrad <matan@nvidia.com>,
Shahaf Shuler <shahafs@nvidia.com>
Cc: "dev@dpdk.org" <dev@dpdk.org>,
"jerinj@marvell.com" <jerinj@marvell.com>, nd <nd@arm.com>,
Honnappa Nagarahalli <Honnappa.Nagarahalli@arm.com>,
nd <nd@arm.com>
Subject: Re: [dpdk-dev] [PATCH 2/2] net/mlx5: reduce unnecessary memory access
Date: Fri, 2 Jul 2021 07:28:43 +0000 [thread overview]
Message-ID: <AM5PR0802MB2465A7909DDF505A5B95F1329E1F9@AM5PR0802MB2465.eurprd08.prod.outlook.com> (raw)
In-Reply-To: <DM6PR12MB37536960736695A7A3911DD2DF1F9@DM6PR12MB3753.namprd12.prod.outlook.com>
> -----Original Message-----
> From: Slava Ovsiienko <viacheslavo@nvidia.com>
> Sent: Friday, July 2, 2021 3:06 PM
> To: Ruifeng Wang <Ruifeng.Wang@arm.com>; Raslan Darawsheh
> <rasland@nvidia.com>; Matan Azrad <matan@nvidia.com>; Shahaf Shuler
> <shahafs@nvidia.com>
> Cc: dev@dpdk.org; jerinj@marvell.com; nd <nd@arm.com>; Honnappa
> Nagarahalli <Honnappa.Nagarahalli@arm.com>
> Subject: RE: [PATCH 2/2] net/mlx5: reduce unnecessary memory access
>
> Hi, Ruifeng
>
> Could we go further and implement loop inside the conditional?
> Like this:
> if (mlx5_mr_btree_len(&rxq->mr_ctrl.cache_bh) > 1) {
> for (i = 0; i < n; ++i) {
> void *buf_addr = elts[i]->buf_addr;
>
> wq[i].addr = rte_cpu_to_be_64((uintptr_t)buf_addr +
> RTE_PKTMBUF_HEADROOM);
> wq[i].lkey = mlx5_rx_mb2mr(rxq, elts[i]);
> }
> } else {
> for (i = 0; i < n; ++i) {
> void *buf_addr = elts[i]->buf_addr;
>
> wq[i].addr = rte_cpu_to_be_64((uintptr_t)buf_addr +
> RTE_PKTMBUF_HEADROOM);
> }
> }
> What do you think?
Agree. Loop inside the conditional should be more efficient.
> Also, we should check the performance on other archs is not affected.
I will also test on x86 platform that I have.
>
> With best regards,
> Slava
>
> > -----Original Message-----
> > From: Ruifeng Wang <ruifeng.wang@arm.com>
> > Sent: Tuesday, June 1, 2021 11:31
> > To: Raslan Darawsheh <rasland@nvidia.com>; Matan Azrad
> > <matan@nvidia.com>; Shahaf Shuler <shahafs@nvidia.com>; Slava
> > Ovsiienko <viacheslavo@nvidia.com>
> > Cc: dev@dpdk.org; jerinj@marvell.com; nd@arm.com;
> > honnappa.nagarahalli@arm.com; Ruifeng Wang <ruifeng.wang@arm.com>
> > Subject: [PATCH 2/2] net/mlx5: reduce unnecessary memory access
> >
> > MR btree len is a constant during Rx replenish.
> > Moved retrieve of the value out of loop to reduce data loads.
> > Slight performance uplift was measured on N1SDP.
> >
> > Signed-off-by: Ruifeng Wang <ruifeng.wang@arm.com>
> > ---
> > drivers/net/mlx5/mlx5_rxtx_vec.c | 6 ++++--
> > 1 file changed, 4 insertions(+), 2 deletions(-)
> >
> > diff --git a/drivers/net/mlx5/mlx5_rxtx_vec.c
> > b/drivers/net/mlx5/mlx5_rxtx_vec.c
> > index d5af2d91ff..fc7e2a7f41 100644
> > --- a/drivers/net/mlx5/mlx5_rxtx_vec.c
> > +++ b/drivers/net/mlx5/mlx5_rxtx_vec.c
> > @@ -95,6 +95,7 @@ mlx5_rx_replenish_bulk_mbuf(struct mlx5_rxq_data
> > *rxq)
> > volatile struct mlx5_wqe_data_seg *wq =
> > &((volatile struct mlx5_wqe_data_seg *)rxq-
> >wqes)[elts_idx];
> > unsigned int i;
> > + uint16_t btree_len;
> >
> > if (n >= rxq->rq_repl_thresh) {
> > MLX5_ASSERT(n >=
> > MLX5_VPMD_RXQ_RPLNSH_THRESH(q_n));
> > @@ -106,6 +107,8 @@ mlx5_rx_replenish_bulk_mbuf(struct
> mlx5_rxq_data
> > *rxq)
> > rxq->stats.rx_nombuf += n;
> > return;
> > }
> > +
> > + btree_len = mlx5_mr_btree_len(&rxq->mr_ctrl.cache_bh);
> > for (i = 0; i < n; ++i) {
> > void *buf_addr;
> >
> > @@ -119,8 +122,7 @@ mlx5_rx_replenish_bulk_mbuf(struct
> mlx5_rxq_data
> > *rxq)
> > wq[i].addr = rte_cpu_to_be_64((uintptr_t)buf_addr
> +
> >
> > RTE_PKTMBUF_HEADROOM);
> > /* If there's a single MR, no need to replace LKey. */
> > - if (unlikely(mlx5_mr_btree_len(&rxq-
> > >mr_ctrl.cache_bh)
> > - > 1))
> > + if (unlikely(btree_len > 1))
> > wq[i].lkey = mlx5_rx_mb2mr(rxq, elts[i]);
> > }
> > rxq->rq_ci += n;
> > --
> > 2.25.1
next prev parent reply other threads:[~2021-07-02 7:28 UTC|newest]
Thread overview: 16+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-06-01 8:30 [dpdk-dev] [PATCH 0/2] MLX5 PMD tuning Ruifeng Wang
2021-06-01 8:30 ` [dpdk-dev] [PATCH 1/2] net/mlx5: remove redundant operations Ruifeng Wang
2021-07-02 8:12 ` Slava Ovsiienko
2021-07-02 10:30 ` Ruifeng Wang
2021-07-05 10:01 ` Slava Ovsiienko
2021-07-07 8:00 ` Ruifeng Wang
2021-06-01 8:30 ` [dpdk-dev] [PATCH 2/2] net/mlx5: reduce unnecessary memory access Ruifeng Wang
2021-07-02 7:05 ` Slava Ovsiienko
2021-07-02 7:28 ` Ruifeng Wang [this message]
2021-06-30 7:22 ` [dpdk-dev] [PATCH 0/2] MLX5 PMD tuning Ruifeng Wang
2021-07-07 9:03 ` [dpdk-dev] [PATCH v2 " Ruifeng Wang
2021-07-07 9:03 ` [dpdk-dev] [PATCH v2 1/2] net/mlx5: remove redundant operations Ruifeng Wang
2021-07-12 15:31 ` Slava Ovsiienko
2021-07-07 9:03 ` [dpdk-dev] [PATCH v2 2/2] net/mlx5: reduce unnecessary memory access Ruifeng Wang
2021-07-12 15:33 ` Slava Ovsiienko
2021-07-13 9:32 ` [dpdk-dev] [PATCH v2 0/2] MLX5 PMD tuning Raslan Darawsheh
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=AM5PR0802MB2465A7909DDF505A5B95F1329E1F9@AM5PR0802MB2465.eurprd08.prod.outlook.com \
--to=ruifeng.wang@arm.com \
--cc=Honnappa.Nagarahalli@arm.com \
--cc=dev@dpdk.org \
--cc=jerinj@marvell.com \
--cc=matan@nvidia.com \
--cc=nd@arm.com \
--cc=rasland@nvidia.com \
--cc=shahafs@nvidia.com \
--cc=viacheslavo@nvidia.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).