patches for DPDK stable branches
 help / color / mirror / Atom feed
From: Kevin Traynor <ktraynor@redhat.com>
To: "Mattias Rönnblom" <mattias.ronnblom@ericsson.com>
Cc: dpdk stable <stable@dpdk.org>
Subject: [dpdk-stable] patch 'event/dsw: avoid credit leak on oversized enqueue bursts' has been queued to LTS release 18.11.7
Date: Fri,  7 Feb 2020 15:12:22 +0000	[thread overview]
Message-ID: <20200207151248.29804-10-ktraynor@redhat.com> (raw)
In-Reply-To: <20200207151248.29804-1-ktraynor@redhat.com>

Hi,

FYI, your patch has been queued to LTS release 18.11.7

Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
It will be pushed if I get no objections before 02/13/20. So please
shout if anyone has objections.

Also note that after the patch there's a diff of the upstream commit vs the
patch applied to the branch. This will indicate if there was any rebasing
needed to apply to the stable branch. If there were code changes for rebasing
(ie: not only metadata diffs), please double check that the rebase was
correctly done.

Queued patches are on a temporary branch at:
https://github.com/kevintraynor/dpdk-stable-queue

This queued commit can be viewed at:
https://github.com/kevintraynor/dpdk-stable-queue/commit/1caa6035fa2a96f7de79d24d15b07b6f598d1697

Thanks.

Kevin.

---
From 1caa6035fa2a96f7de79d24d15b07b6f598d1697 Mon Sep 17 00:00:00 2001
From: =?UTF-8?q?Mattias=20R=C3=B6nnblom?= <mattias.ronnblom@ericsson.com>
Date: Tue, 14 Jan 2020 19:03:38 +0100
Subject: [PATCH] event/dsw: avoid credit leak on oversized enqueue bursts
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit

[ upstream commit 0c4155c7b5b51a60f676c1d7279bfcd1b14acd38 ]

If an application issues rte_event_enqueue_new_burst() or
rte_event_enqueue_forward_burst() call with a burst of events longer
than the configured max enqueue burst size, DSW allocates credits not
only for events actually enqueued, but for the complete burst. If this
process is repeated, enough credits will have leaked to cause the
event device to backpressure (i.e. disallow) any new enqueue
operations.

In addition, the port-level enqueue xstats will log the wrong number
of events enqueued for oversized enqueues.

This patch makes DSW gracefully handle oversized enqueue bursts.

Fixes: 1c8e3caa3bfb ("event/dsw: add event scheduling and device start/stop")

Signed-off-by: Mattias Rönnblom <mattias.ronnblom@ericsson.com>
---
 drivers/event/dsw/dsw_event.c | 38 ++++++++++++++++++++++++-----------
 1 file changed, 26 insertions(+), 12 deletions(-)

diff --git a/drivers/event/dsw/dsw_event.c b/drivers/event/dsw/dsw_event.c
index 1641c2d06e..eae53b2404 100644
--- a/drivers/event/dsw/dsw_event.c
+++ b/drivers/event/dsw/dsw_event.c
@@ -1019,10 +1019,10 @@ dsw_event_enqueue(void *port, const struct rte_event *ev)
 
 static __rte_always_inline uint16_t
-dsw_event_enqueue_burst_generic(void *port, const struct rte_event events[],
+dsw_event_enqueue_burst_generic(struct dsw_port *source_port,
+				const struct rte_event events[],
 				uint16_t events_len, bool op_types_known,
 				uint16_t num_new, uint16_t num_release,
 				uint16_t num_non_release)
 {
-	struct dsw_port *source_port = port;
 	struct dsw_evdev *dsw = source_port->dsw;
 	bool enough_credits;
@@ -1048,11 +1048,8 @@ dsw_event_enqueue_burst_generic(void *port, const struct rte_event events[],
 	if (unlikely(events_len == 0)) {
 		dsw_port_note_op(source_port, DSW_MAX_PORT_OPS_PER_BG_TASK);
-		dsw_port_flush_out_buffers(dsw, port);
+		dsw_port_flush_out_buffers(dsw, source_port);
 		return 0;
 	}
 
-	if (unlikely(events_len > source_port->enqueue_depth))
-		events_len = source_port->enqueue_depth;
-
 	dsw_port_note_op(source_port, events_len);
 
@@ -1110,6 +1107,11 @@ dsw_event_enqueue_burst(void *port, const struct rte_event events[],
 			uint16_t events_len)
 {
-	return dsw_event_enqueue_burst_generic(port, events, events_len, false,
-					       0, 0, 0);
+	struct dsw_port *source_port = port;
+
+	if (unlikely(events_len > source_port->enqueue_depth))
+		events_len = source_port->enqueue_depth;
+
+	return dsw_event_enqueue_burst_generic(source_port, events,
+					       events_len, false, 0, 0, 0);
 }
 
@@ -1118,6 +1120,12 @@ dsw_event_enqueue_new_burst(void *port, const struct rte_event events[],
 			    uint16_t events_len)
 {
-	return dsw_event_enqueue_burst_generic(port, events, events_len, true,
-					       events_len, 0, events_len);
+	struct dsw_port *source_port = port;
+
+	if (unlikely(events_len > source_port->enqueue_depth))
+		events_len = source_port->enqueue_depth;
+
+	return dsw_event_enqueue_burst_generic(source_port, events,
+					       events_len, true, events_len,
+					       0, events_len);
 }
 
@@ -1126,6 +1134,12 @@ dsw_event_enqueue_forward_burst(void *port, const struct rte_event events[],
 				uint16_t events_len)
 {
-	return dsw_event_enqueue_burst_generic(port, events, events_len, true,
-					       0, 0, events_len);
+	struct dsw_port *source_port = port;
+
+	if (unlikely(events_len > source_port->enqueue_depth))
+		events_len = source_port->enqueue_depth;
+
+	return dsw_event_enqueue_burst_generic(source_port, events,
+					       events_len, true, 0, 0,
+					       events_len);
 }
 
-- 
2.21.1

---
  Diff of the applied patch vs upstream commit (please double-check if non-empty:
---
--- -	2020-02-07 15:08:18.054801011 +0000
+++ 0010-event-dsw-avoid-credit-leak-on-oversized-enqueue-bur.patch	2020-02-07 15:08:17.495063202 +0000
@@ -1 +1 @@
-From 0c4155c7b5b51a60f676c1d7279bfcd1b14acd38 Mon Sep 17 00:00:00 2001
+From 1caa6035fa2a96f7de79d24d15b07b6f598d1697 Mon Sep 17 00:00:00 2001
@@ -8,0 +9,2 @@
+[ upstream commit 0c4155c7b5b51a60f676c1d7279bfcd1b14acd38 ]
+
@@ -23 +24,0 @@
-Cc: stable@dpdk.org


  parent reply	other threads:[~2020-02-07 15:13 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-02-07 15:12 [dpdk-stable] patch 'eal/linux: fix uninitialized data valgrind warning' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'eal/linux: fix build error on RHEL 7.6' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'doc: fix build with python 3.8' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'build: explicitly enable sse4 for meson' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'kni: fix meson warning about console keyword' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'doc: fix warning with meson' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'doc: reduce whitespace in meson build file' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'doc: reduce indentation " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'event/dsw: flush buffers immediately on zero-sized enqueue' " Kevin Traynor
2020-02-07 15:12 ` Kevin Traynor [this message]
2020-02-07 15:12 ` [dpdk-stable] patch 'bus/fslmc: remove conflicting memory barrier macro' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'examples/ipsec-secgw: fix crash on unsupported algo' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'app/testpmd: fix device mcast list error handling' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/fm10k: fix descriptor VLAN field filling in Tx' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/iavf: fix Rx total stats' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/iavf: add TSO offload use basic path' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/ixgbe: fix link status' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/ixgbe: fix link up in FreeBSD' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/iavf/base: fix command buffer memory leak' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/iavf/base: fix adminq return' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/iavf: fix virtual channel " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/i40e: fix Tx when TSO is enabled' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/bnxt: fix link during port toggle' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'app/testpmd: call cleanup on exit' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/vhost: check creation failure' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/virtio-user: check file descriptor before closing' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'vhost: fix socket initial value' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/i40e/base: fix buffer address' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/i40e/base: fix error message' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/i40e/base: fix Tx descriptors number' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/i40e/base: fix retrying logic' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'ethdev: fix callback unregister with wildcard argument list' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/netvsc: fix crash in secondary process' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'net/cxgbe: announce Tx multi-segments offload' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'service: avoid false sharing on core state' " Kevin Traynor
2020-02-07 15:12 ` [dpdk-stable] patch 'service: don't walk out of bounds when checking services' " Kevin Traynor

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=20200207151248.29804-10-ktraynor@redhat.com \
    --to=ktraynor@redhat.com \
    --cc=mattias.ronnblom@ericsson.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).