From: Bruce Richardson <bruce.richardson@intel.com>
To: Jerin Jacob <jerin.jacob@caviumnetworks.com>
Cc: Jianbo Liu <jianbo.liu@linaro.org>,
dev@dpdk.org, helin.zhang@intel.com,
konstantin.ananyev@intel.com
Subject: Re: [dpdk-dev] [PATCH v3 2/4] ixgbe: implement vector PMD for arm architecture
Date: Wed, 25 May 2016 13:53:33 +0100 [thread overview]
Message-ID: <20160525125332.GA8612@bricha3-MOBL3> (raw)
In-Reply-To: <20160525122935.GA30670@localhost.localdomain>
On Wed, May 25, 2016 at 05:59:38PM +0530, Jerin Jacob wrote:
> On Fri, May 06, 2016 at 11:55:46AM +0530, Jianbo Liu wrote:
> > use ARM NEON intrinsic to implement ixgbe vPMD
> >
> > Signed-off-by: Jianbo Liu <jianbo.liu@linaro.org>
> > ---
> > drivers/net/ixgbe/Makefile | 4 +
> > drivers/net/ixgbe/ixgbe_rxtx_vec_neon.c | 561 ++++++++++++++++++++++++++++++++
> > 2 files changed, 565 insertions(+)
> > create mode 100644 drivers/net/ixgbe/ixgbe_rxtx_vec_neon.c
> >
<snip>
> > + for (pos = 0, nb_pkts_recd = 0; pos < nb_pkts;
> > + pos += RTE_IXGBE_DESCS_PER_LOOP,
> > + rxdp += RTE_IXGBE_DESCS_PER_LOOP) {
> > + uint64x2_t descs[RTE_IXGBE_DESCS_PER_LOOP];
> > + uint8x16_t pkt_mb1, pkt_mb2, pkt_mb3, pkt_mb4;
> > + uint8x16x2_t sterr_tmp1, sterr_tmp2;
> > + uint64x2_t mbp1, mbp2;
> > + uint8x16_t staterr;
> > + uint16x8_t tmp;
> > + uint32_t stat;
> > +
> > + /* B.1 load 1 mbuf point */
> > + mbp1 = vld1q_u64((uint64_t *)&sw_ring[pos]);
> > +
> > + /* Read desc statuses backwards to avoid race condition */
> > + /* A.1 load 4 pkts desc */
> > + descs[3] = vld1q_u64((uint64_t *)(rxdp + 3));
> > + rte_rmb();
>
> Any specific reason to add rte_rmb() here, If there is no performance
> drop then it makes sense to add before descs[3] uses it.i.e
> at rte_compiler_barrier() place in x86 code.
>
> > +
> > + /* B.2 copy 2 mbuf point into rx_pkts */
> > + vst1q_u64((uint64_t *)&rx_pkts[pos], mbp1);
> > +
> > + /* B.1 load 1 mbuf point */
> > + mbp2 = vld1q_u64((uint64_t *)&sw_ring[pos + 2]);
> > +
> > + descs[2] = vld1q_u64((uint64_t *)(rxdp + 2));
> > + /* B.1 load 2 mbuf point */
> > + descs[1] = vld1q_u64((uint64_t *)(rxdp + 1));
> > + descs[0] = vld1q_u64((uint64_t *)(rxdp));
> > +
> > + /* B.2 copy 2 mbuf point into rx_pkts */
> > + vst1q_u64((uint64_t *)&rx_pkts[pos + 2], mbp2);
> > +
> > + if (split_packet) {
> > + rte_prefetch_non_temporal(&rx_pkts[pos]->cacheline1);
> > + rte_prefetch_non_temporal(&rx_pkts[pos+1]->cacheline1);
> > + rte_prefetch_non_temporal(&rx_pkts[pos+2]->cacheline1);
> > + rte_prefetch_non_temporal(&rx_pkts[pos+3]->cacheline1);
>
> replace with rte_mbuf_prefetch_part2 or equivalent
>
Hi Jerin, Jianbo,
since this patch has already been applied and these are not critical issues with
it, can a new patch please be submitted to propose these additional changes on
top of what's on next-net now.
Thanks,
/Bruce
next prev parent reply other threads:[~2016-05-25 12:53 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-20 13:44 [dpdk-dev] [PATCH 1/4] ixgbe: rearrange vector PMD code for x86 Jianbo Liu
2016-04-20 13:45 ` [dpdk-dev] [PATCH 2/4] ixgbe: implement vector PMD for arm architecture Jianbo Liu
2016-04-20 13:45 ` [dpdk-dev] [PATCH 3/4] ixgbe: enable ixgbe vector PMD on ARMv8a platform Jianbo Liu
2016-04-20 13:45 ` [dpdk-dev] [PATCH 4/4] maintainers: claim responsibility for ixgbe vector PMD on ARM Jianbo Liu
2016-04-25 16:35 ` [dpdk-dev] [PATCH 1/4] ixgbe: rearrange vector PMD code for x86 Bruce Richardson
2016-04-26 8:23 ` Jianbo Liu
2016-04-26 13:50 ` [dpdk-dev] [PATCH v2 " Jianbo Liu
2016-05-03 5:51 ` Jianbo Liu
2016-05-03 16:29 ` Bruce Richardson
2016-04-26 13:55 ` [dpdk-dev] [PATCH v2 2/4] ixgbe: implement vector PMD for arm architecture Jianbo Liu
2016-04-26 13:55 ` [dpdk-dev] [PATCH v2 3/4] ixgbe: enable ixgbe vector PMD on ARMv8a platform Jianbo Liu
2016-04-26 13:55 ` [dpdk-dev] [PATCH v2 4/4] maintainers: claim responsibility for ixgbe vector PMD on ARM Jianbo Liu
2016-05-06 6:25 ` [dpdk-dev] [PATCH v3 0/4] ixgbe: enable " Jianbo Liu
2016-05-06 6:25 ` [dpdk-dev] [PATCH v3 1/4] ixgbe: rearrange vector PMD code for x86 Jianbo Liu
2016-05-06 6:25 ` [dpdk-dev] [PATCH v3 2/4] ixgbe: implement vector PMD for arm architecture Jianbo Liu
2016-05-10 14:49 ` Bruce Richardson
2016-05-11 2:40 ` Jianbo Liu
2016-05-25 12:29 ` Jerin Jacob
2016-05-25 12:53 ` Bruce Richardson [this message]
2016-05-26 1:37 ` Jianbo Liu
2016-05-06 6:25 ` [dpdk-dev] [PATCH v3 3/4] ixgbe: enable ixgbe vector PMD on ARMv8a platform Jianbo Liu
2016-05-06 6:25 ` [dpdk-dev] [PATCH v3 4/4] maintainers: claim responsibility for ixgbe vector PMD on ARM Jianbo Liu
2016-05-24 16:10 ` [dpdk-dev] [PATCH v3 0/4] ixgbe: enable " Bruce Richardson
2016-05-24 16:12 ` Bruce Richardson
2016-05-27 10:44 ` Jianbo Liu
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=20160525125332.GA8612@bricha3-MOBL3 \
--to=bruce.richardson@intel.com \
--cc=dev@dpdk.org \
--cc=helin.zhang@intel.com \
--cc=jerin.jacob@caviumnetworks.com \
--cc=jianbo.liu@linaro.org \
--cc=konstantin.ananyev@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
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).