From: Jerin Jacob <jerinjacobk@gmail.com>
To: Pavan Nikhilesh <pbhagavatula@marvell.com>
Cc: Jerin Jacob <jerinj@marvell.com>,
Shijith Thotton <sthotton@marvell.com>, dpdk-dev <dev@dpdk.org>
Subject: Re: [PATCH 1/2] event/cnxk: remove deschedule usage in CN9K
Date: Tue, 22 Feb 2022 15:21:20 +0530 [thread overview]
Message-ID: <CALBAE1MtFRnw7CR4PXM9U9bJ16r0WAf7uPivGC8+W0Ac1BR1jA@mail.gmail.com> (raw)
In-Reply-To: <20220219121338.2438-1-pbhagavatula@marvell.com>
On Sat, Feb 19, 2022 at 6:05 PM <pbhagavatula@marvell.com> wrote:
>
> From: Pavan Nikhilesh <pbhagavatula@marvell.com>
>
> Using deschedule cmd might incorrectly ignore updates to WQE, GGRP
> on CN9K.
> Use addwork to pipeline work instead.
>
> Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
Series applied to dpdk-next-net-eventdev/for-main. Thanks
> ---
> drivers/event/cnxk/cn9k_worker.h | 41 +++++++++++++++++++++++++-------
> 1 file changed, 32 insertions(+), 9 deletions(-)
>
> diff --git a/drivers/event/cnxk/cn9k_worker.h b/drivers/event/cnxk/cn9k_worker.h
> index 79374b8d95..0905d744cc 100644
> --- a/drivers/event/cnxk/cn9k_worker.h
> +++ b/drivers/event/cnxk/cn9k_worker.h
> @@ -63,15 +63,18 @@ cn9k_sso_hws_fwd_swtag(uint64_t base, const struct rte_event *ev)
> }
>
> static __rte_always_inline void
> -cn9k_sso_hws_fwd_group(uint64_t base, const struct rte_event *ev,
> - const uint16_t grp)
> +cn9k_sso_hws_new_event_wait(struct cn9k_sso_hws *ws, const struct rte_event *ev)
> {
> const uint32_t tag = (uint32_t)ev->event;
> const uint8_t new_tt = ev->sched_type;
> + const uint64_t event_ptr = ev->u64;
> + const uint16_t grp = ev->queue_id;
>
> - plt_write64(ev->u64, base + SSOW_LF_GWS_OP_UPD_WQP_GRP1);
> - cnxk_sso_hws_swtag_desched(tag, new_tt, grp,
> - base + SSOW_LF_GWS_OP_SWTAG_DESCHED);
> + while (ws->xaq_lmt <= __atomic_load_n(ws->fc_mem, __ATOMIC_RELAXED))
> + ;
> +
> + cnxk_sso_hws_add_work(event_ptr, tag, new_tt,
> + ws->grp_base + (grp << 12));
> }
>
> static __rte_always_inline void
> @@ -86,10 +89,12 @@ cn9k_sso_hws_forward_event(struct cn9k_sso_hws *ws, const struct rte_event *ev)
> } else {
> /*
> * Group has been changed for group based work pipelining,
> - * Use deschedule/add_work operation to transfer the event to
> + * Use add_work operation to transfer the event to
> * new group/core
> */
> - cn9k_sso_hws_fwd_group(ws->base, ev, grp);
> + rte_atomic_thread_fence(__ATOMIC_RELEASE);
> + roc_sso_hws_head_wait(ws->base);
> + cn9k_sso_hws_new_event_wait(ws, ev);
> }
> }
>
> @@ -113,6 +118,22 @@ cn9k_sso_hws_dual_new_event(struct cn9k_sso_hws_dual *dws,
> return 1;
> }
>
> +static __rte_always_inline void
> +cn9k_sso_hws_dual_new_event_wait(struct cn9k_sso_hws_dual *dws,
> + const struct rte_event *ev)
> +{
> + const uint32_t tag = (uint32_t)ev->event;
> + const uint8_t new_tt = ev->sched_type;
> + const uint64_t event_ptr = ev->u64;
> + const uint16_t grp = ev->queue_id;
> +
> + while (dws->xaq_lmt <= __atomic_load_n(dws->fc_mem, __ATOMIC_RELAXED))
> + ;
> +
> + cnxk_sso_hws_add_work(event_ptr, tag, new_tt,
> + dws->grp_base + (grp << 12));
> +}
> +
> static __rte_always_inline void
> cn9k_sso_hws_dual_forward_event(struct cn9k_sso_hws_dual *dws, uint64_t base,
> const struct rte_event *ev)
> @@ -126,10 +147,12 @@ cn9k_sso_hws_dual_forward_event(struct cn9k_sso_hws_dual *dws, uint64_t base,
> } else {
> /*
> * Group has been changed for group based work pipelining,
> - * Use deschedule/add_work operation to transfer the event to
> + * Use add_work operation to transfer the event to
> * new group/core
> */
> - cn9k_sso_hws_fwd_group(base, ev, grp);
> + rte_atomic_thread_fence(__ATOMIC_RELEASE);
> + roc_sso_hws_head_wait(base);
> + cn9k_sso_hws_dual_new_event_wait(dws, ev);
> }
> }
>
> --
> 2.17.1
>
prev parent reply other threads:[~2022-02-22 9:51 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-02-19 12:13 pbhagavatula
2022-02-19 12:13 ` [PATCH 2/2] event/cnxk: update SQB fc check for Tx adapter pbhagavatula
2022-02-22 9:51 ` 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=CALBAE1MtFRnw7CR4PXM9U9bJ16r0WAf7uPivGC8+W0Ac1BR1jA@mail.gmail.com \
--to=jerinjacobk@gmail.com \
--cc=dev@dpdk.org \
--cc=jerinj@marvell.com \
--cc=pbhagavatula@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).