From: <pbhagavatula@marvell.com>
To: <jerinj@marvell.com>, Pavan Nikhilesh <pbhagavatula@marvell.com>
Cc: <dev@dpdk.org>, <stable@dpdk.org>
Subject: [dpdk-stable] [dpdk-dev] [PATCH] event/octeontx2: fix unintended flush in Tx enqueue
Date: Tue, 17 Nov 2020 22:16:29 +0530 [thread overview]
Message-ID: <20201117164630.2971-1-pbhagavatula@marvell.com> (raw)
From: Pavan Nikhilesh <pbhagavatula@marvell.com>
SSO workslot shouldn't be flushed on Tx adapter enqueue as
application might use reference count to re-transmit or
forward the packet.
Fixes: cb7ee83b6365 ("event/octeontx2: improve single flow performance")
Cc: stable@dpdk.org
Signed-off-by: Pavan Nikhilesh <pbhagavatula@marvell.com>
---
drivers/event/octeontx2/otx2_evdev.h | 1 +
drivers/event/octeontx2/otx2_worker.h | 6 ++++--
drivers/event/octeontx2/otx2_worker_dual.c | 8 ++++++++
3 files changed, 13 insertions(+), 2 deletions(-)
diff --git a/drivers/event/octeontx2/otx2_evdev.h b/drivers/event/octeontx2/otx2_evdev.h
index 547e29d4a..e4e444ed7 100644
--- a/drivers/event/octeontx2/otx2_evdev.h
+++ b/drivers/event/octeontx2/otx2_evdev.h
@@ -79,6 +79,7 @@
#define SSOW_LF_GWS_OP_GWC_INVAL (0xe00ull)
#define OTX2_SSOW_GET_BASE_ADDR(_GW) ((_GW) - SSOW_LF_GWS_OP_GET_WORK)
+#define OTX2_SSOW_TT_FROM_TAG(x) (((x) >> 32) & SSO_TT_EMPTY)
#define NSEC2USEC(__ns) ((__ns) / 1E3)
#define USEC2NSEC(__us) ((__us) * 1E3)
diff --git a/drivers/event/octeontx2/otx2_worker.h b/drivers/event/octeontx2/otx2_worker.h
index 3efd3ba97..67893f669 100644
--- a/drivers/event/octeontx2/otx2_worker.h
+++ b/drivers/event/octeontx2/otx2_worker.h
@@ -198,6 +198,10 @@ otx2_ssogws_swtag_untag(struct otx2_ssogws *ws)
static __rte_always_inline void
otx2_ssogws_swtag_flush(struct otx2_ssogws *ws)
{
+ if (OTX2_SSOW_TT_FROM_TAG(otx2_read64(ws->tag_op)) == SSO_TT_EMPTY) {
+ ws->cur_tt = SSO_SYNC_EMPTY;
+ return;
+ }
otx2_write64(0, ws->swtag_flush_op);
ws->cur_tt = SSO_SYNC_EMPTY;
}
@@ -329,8 +333,6 @@ otx2_ssogws_event_tx(struct otx2_ssogws *ws, struct rte_event ev[],
}
}
- otx2_write64(0, ws->swtag_flush_op);
-
return 1;
}
diff --git a/drivers/event/octeontx2/otx2_worker_dual.c b/drivers/event/octeontx2/otx2_worker_dual.c
index 946488eab..f1823e29c 100644
--- a/drivers/event/octeontx2/otx2_worker_dual.c
+++ b/drivers/event/octeontx2/otx2_worker_dual.c
@@ -157,6 +157,8 @@ otx2_ssogws_dual_deq_ ##name(void *port, struct rte_event *ev, \
return 1; \
} \
\
+ otx2_ssogws_swtag_flush((struct otx2_ssogws *) \
+ &ws->ws_state[!ws->vws]); \
gw = otx2_ssogws_dual_get_work(&ws->ws_state[ws->vws], \
&ws->ws_state[!ws->vws], ev, \
flags, ws->lookup_mem, \
@@ -191,6 +193,8 @@ otx2_ssogws_dual_deq_timeout_ ##name(void *port, struct rte_event *ev, \
return 1; \
} \
\
+ otx2_ssogws_swtag_flush((struct otx2_ssogws *) \
+ &ws->ws_state[!ws->vws]); \
gw = otx2_ssogws_dual_get_work(&ws->ws_state[ws->vws], \
&ws->ws_state[!ws->vws], ev, \
flags, ws->lookup_mem, \
@@ -235,6 +239,8 @@ otx2_ssogws_dual_deq_seg_ ##name(void *port, struct rte_event *ev, \
return 1; \
} \
\
+ otx2_ssogws_swtag_flush((struct otx2_ssogws *) \
+ &ws->ws_state[!ws->vws]); \
gw = otx2_ssogws_dual_get_work(&ws->ws_state[ws->vws], \
&ws->ws_state[!ws->vws], ev, \
flags | NIX_RX_MULTI_SEG_F, \
@@ -273,6 +279,8 @@ otx2_ssogws_dual_deq_seg_timeout_ ##name(void *port, \
return 1; \
} \
\
+ otx2_ssogws_swtag_flush((struct otx2_ssogws *) \
+ &ws->ws_state[!ws->vws]); \
gw = otx2_ssogws_dual_get_work(&ws->ws_state[ws->vws], \
&ws->ws_state[!ws->vws], ev, \
flags | NIX_RX_MULTI_SEG_F, \
--
2.17.1
next reply other threads:[~2020-11-17 16:46 UTC|newest]
Thread overview: 3+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-17 16:46 pbhagavatula [this message]
2020-11-20 10:41 ` [dpdk-stable] [dpdk-dev] [PATCH v2] event/octeontx2: fix unconditional Tx flush pbhagavatula
2020-11-20 12:28 ` Jerin Jacob
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=20201117164630.2971-1-pbhagavatula@marvell.com \
--to=pbhagavatula@marvell.com \
--cc=dev@dpdk.org \
--cc=jerinj@marvell.com \
--cc=stable@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).