From: Jerin Jacob <jerinjacobk@gmail.com>
To: Rahul Bhansali <rbhansali@marvell.com>
Cc: dpdk-dev <dev@dpdk.org>,
Pavan Nikhilesh <pbhagavatula@marvell.com>,
Shijith Thotton <sthotton@marvell.com>,
Nithin Dabilpuram <ndabilpuram@marvell.com>,
Kiran Kumar K <kirankumark@marvell.com>,
Sunil Kumar Kori <skori@marvell.com>,
Satha Rao <skoteshwar@marvell.com>,
Jerin Jacob <jerinj@marvell.com>
Subject: Re: [PATCH] net/cnxk: resolve mbuf data length update issue
Date: Wed, 16 Feb 2022 19:06:44 +0530 [thread overview]
Message-ID: <CALBAE1NEVOJk9FtVE+55zenYRaF8eDo3PAEOTz4ByJKngR8q6w@mail.gmail.com> (raw)
In-Reply-To: <20220119105203.1553714-1-rbhansali@marvell.com>
On Wed, Jan 19, 2022 at 4:22 PM Rahul Bhansali <rbhansali@marvell.com> wrote:
>
> If multi-segment is enabled and single segment/packet
> is received, then mbuf data_len is not updated in
> cn9k_nix_cqe_to_mbuf function.
> Also, in case of timestamp is enabled, mbuf data_len
> and pkt_len will be updated for all packets including
> multi segmented packets.
>
> Signed-off-by: Rahul Bhansali <rbhansali@marvell.com>
Changed the git commit as follows and Applied to
dpdk-next-net-mrvl/for-next-net. Thanks
net/cnxk: fix mbuf data length
If multi-segment is enabled and single segment/packet
is received, then mbuf data_len is not updated in
cn9k_nix_cqe_to_mbuf() function.
Also, in case of timestamp is enabled, mbuf data_len
and pkt_len will be updated for all packets including
multi segmented packets.
Fixes: 7c6bee34064f ("net/cnxk: enable PTP processing in vector Rx")
Cc: stable@dpdk.org
Signed-off-by: Rahul Bhansali <rbhansali@marvell.com>
Acked-by: Jerin Jacob <jerinj@marvell.com>
> ---
> Depends-on: Series-21246 ("[v2,1/4] net/cnxk: avoid command copy from Tx
> queue")
>
> drivers/event/cnxk/cn10k_worker.h | 2 --
> drivers/event/cnxk/cn9k_worker.h | 2 --
> drivers/net/cnxk/cn10k_rx.h | 9 +++++++--
> drivers/net/cnxk/cn9k_rx.h | 18 +++++++++++-------
> drivers/net/cnxk/cnxk_ethdev.h | 9 +++------
> 5 files changed, 21 insertions(+), 19 deletions(-)
>
> diff --git a/drivers/event/cnxk/cn10k_worker.h b/drivers/event/cnxk/cn10k_worker.h
> index 1e61a6ddf0..5c36540a43 100644
> --- a/drivers/event/cnxk/cn10k_worker.h
> +++ b/drivers/event/cnxk/cn10k_worker.h
> @@ -174,7 +174,6 @@ cn10k_process_vwqe(uintptr_t vwqe, uint16_t port_id, const uint32_t flags,
> CNXK_SSO_WQE_SG_PTR);
> cnxk_nix_mbuf_to_tstamp((struct rte_mbuf *)mbuf, tstamp,
> flags & NIX_RX_OFFLOAD_TSTAMP_F,
> - flags & NIX_RX_MULTI_SEG_F,
> (uint64_t *)tstamp_ptr);
> wqe[0] = (uint64_t *)mbuf;
> non_vec--;
> @@ -266,7 +265,6 @@ cn10k_sso_hws_get_work(struct cn10k_sso_hws *ws, struct rte_event *ev,
> cnxk_nix_mbuf_to_tstamp((struct rte_mbuf *)mbuf,
> ws->tstamp,
> flags & NIX_RX_OFFLOAD_TSTAMP_F,
> - flags & NIX_RX_MULTI_SEG_F,
> (uint64_t *)tstamp_ptr);
> gw.u64[1] = mbuf;
> } else if (CNXK_EVENT_TYPE_FROM_TAG(gw.u64[0]) ==
> diff --git a/drivers/event/cnxk/cn9k_worker.h b/drivers/event/cnxk/cn9k_worker.h
> index e44422ec25..368baae048 100644
> --- a/drivers/event/cnxk/cn9k_worker.h
> +++ b/drivers/event/cnxk/cn9k_worker.h
> @@ -208,7 +208,6 @@ cn9k_sso_hws_dual_get_work(uint64_t base, uint64_t pair_base,
> cnxk_nix_mbuf_to_tstamp((struct rte_mbuf *)mbuf,
> dws->tstamp,
> flags & NIX_RX_OFFLOAD_TSTAMP_F,
> - flags & NIX_RX_MULTI_SEG_F,
> (uint64_t *)tstamp_ptr);
> gw.u64[1] = mbuf;
> }
> @@ -285,7 +284,6 @@ cn9k_sso_hws_get_work(struct cn9k_sso_hws *ws, struct rte_event *ev,
> cnxk_nix_mbuf_to_tstamp((struct rte_mbuf *)mbuf,
> ws->tstamp,
> flags & NIX_RX_OFFLOAD_TSTAMP_F,
> - flags & NIX_RX_MULTI_SEG_F,
> (uint64_t *)tstamp_ptr);
> gw.u64[1] = mbuf;
> }
> diff --git a/drivers/net/cnxk/cn10k_rx.h b/drivers/net/cnxk/cn10k_rx.h
> index 9694a3080f..45b626b089 100644
> --- a/drivers/net/cnxk/cn10k_rx.h
> +++ b/drivers/net/cnxk/cn10k_rx.h
> @@ -364,7 +364,13 @@ cn10k_nix_cqe_to_mbuf(const struct nix_cqe_hdr_s *cq, const uint32_t tag,
> *(uint64_t *)(&mbuf->rearm_data) = val;
>
> if (flag & NIX_RX_MULTI_SEG_F)
> - nix_cqe_xtract_mseg(rx, mbuf, val, flag);
> + /*
> + * For multi segment packets, mbuf length correction according
> + * to Rx timestamp length will be handled later during
> + * timestamp data process.
> + * Hence, flag argument is not required.
> + */
> + nix_cqe_xtract_mseg(rx, mbuf, val, 0);
> else
> mbuf->next = NULL;
> }
> @@ -452,7 +458,6 @@ cn10k_nix_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t pkts,
> flags);
> cnxk_nix_mbuf_to_tstamp(mbuf, rxq->tstamp,
> (flags & NIX_RX_OFFLOAD_TSTAMP_F),
> - (flags & NIX_RX_MULTI_SEG_F),
> (uint64_t *)((uint8_t *)mbuf
> + data_off));
> rx_pkts[packets++] = mbuf;
> diff --git a/drivers/net/cnxk/cn9k_rx.h b/drivers/net/cnxk/cn9k_rx.h
> index fa4efbf80a..c883eaeb07 100644
> --- a/drivers/net/cnxk/cn9k_rx.h
> +++ b/drivers/net/cnxk/cn9k_rx.h
> @@ -345,13 +345,18 @@ cn9k_nix_cqe_to_mbuf(const struct nix_cqe_hdr_s *cq, const uint32_t tag,
> mbuf->ol_flags = ol_flags;
> *(uint64_t *)(&mbuf->rearm_data) = val;
> mbuf->pkt_len = len;
> -
> - if (flag & NIX_RX_MULTI_SEG_F) {
> - nix_cqe_xtract_mseg(rx, mbuf, val, flag);
> - } else {
> - mbuf->data_len = len;
> + mbuf->data_len = len;
> +
> + if (flag & NIX_RX_MULTI_SEG_F)
> + /*
> + * For multi segment packets, mbuf length correction according
> + * to Rx timestamp length will be handled later during
> + * timestamp data process.
> + * Hence, flag argument is not required.
> + */
> + nix_cqe_xtract_mseg(rx, mbuf, val, 0);
> + else
> mbuf->next = NULL;
> - }
> }
>
> static inline uint16_t
> @@ -414,7 +419,6 @@ cn9k_nix_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts, uint16_t pkts,
> flags);
> cnxk_nix_mbuf_to_tstamp(mbuf, rxq->tstamp,
> (flags & NIX_RX_OFFLOAD_TSTAMP_F),
> - (flags & NIX_RX_MULTI_SEG_F),
> (uint64_t *)((uint8_t *)mbuf
> + data_off));
> rx_pkts[packets++] = mbuf;
> diff --git a/drivers/net/cnxk/cnxk_ethdev.h b/drivers/net/cnxk/cnxk_ethdev.h
> index 5bfda3d815..480cc6dfa4 100644
> --- a/drivers/net/cnxk/cnxk_ethdev.h
> +++ b/drivers/net/cnxk/cnxk_ethdev.h
> @@ -685,14 +685,11 @@ cnxk_nix_timestamp_dynfield(struct rte_mbuf *mbuf,
> static __rte_always_inline void
> cnxk_nix_mbuf_to_tstamp(struct rte_mbuf *mbuf,
> struct cnxk_timesync_info *tstamp,
> - const uint8_t ts_enable, const uint8_t mseg_enable,
> - uint64_t *tstamp_ptr)
> + const uint8_t ts_enable, uint64_t *tstamp_ptr)
> {
> if (ts_enable) {
> - if (!mseg_enable) {
> - mbuf->pkt_len -= CNXK_NIX_TIMESYNC_RX_OFFSET;
> - mbuf->data_len -= CNXK_NIX_TIMESYNC_RX_OFFSET;
> - }
> + mbuf->pkt_len -= CNXK_NIX_TIMESYNC_RX_OFFSET;
> + mbuf->data_len -= CNXK_NIX_TIMESYNC_RX_OFFSET;
>
> /* Reading the rx timestamp inserted by CGX, viz at
> * starting of the packet data.
> --
> 2.25.1
>
prev parent reply other threads:[~2022-02-16 13:37 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-01-19 10:52 Rahul Bhansali
2022-02-16 13:36 ` Jerin Jacob [this message]
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=CALBAE1NEVOJk9FtVE+55zenYRaF8eDo3PAEOTz4ByJKngR8q6w@mail.gmail.com \
--to=jerinjacobk@gmail.com \
--cc=dev@dpdk.org \
--cc=jerinj@marvell.com \
--cc=kirankumark@marvell.com \
--cc=ndabilpuram@marvell.com \
--cc=pbhagavatula@marvell.com \
--cc=rbhansali@marvell.com \
--cc=skori@marvell.com \
--cc=skoteshwar@marvell.com \
--cc=sthotton@marvell.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).