patches for DPDK stable branches
 help / color / mirror / Atom feed
From: christian.ehrhardt@canonical.com
To: Kathleen Capella <kathleen.capella@arm.com>
Cc: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>,
	Qi Zhang <qi.z.zhang@intel.com>, dpdk stable <stable@dpdk.org>
Subject: patch 'net/iavf: count continuous DD bits for Arm' has been queued to stable release 19.11.12
Date: Fri, 25 Feb 2022 18:15:47 +0100	[thread overview]
Message-ID: <20220225171550.3499040-54-christian.ehrhardt@canonical.com> (raw)
In-Reply-To: <20220225171550.3499040-1-christian.ehrhardt@canonical.com>

Hi,

FYI, your patch has been queued to stable release 19.11.12

Note it hasn't been pushed to http://dpdk.org/browse/dpdk-stable yet.
It will be pushed if I get no objections before 02/27/22. 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/cpaelzer/dpdk-stable-queue

This queued commit can be viewed at:
https://github.com/cpaelzer/dpdk-stable-queue/commit/55f2ea35c7cafdb178e59862d66eaf73b3282706

Thanks.

Christian Ehrhardt <christian.ehrhardt@canonical.com>

---
From 55f2ea35c7cafdb178e59862d66eaf73b3282706 Mon Sep 17 00:00:00 2001
From: Kathleen Capella <kathleen.capella@arm.com>
Date: Sat, 5 Feb 2022 00:26:29 +0000
Subject: [PATCH] net/iavf: count continuous DD bits for Arm

[ upstream commit 3d88d5e401f6d04882415866012a98bced0c3cb8 ]

On Arm platforms, reading of descriptors may be re-ordered causing the
status of DD bits to be discontinuous. Add logic to only process
continuous descriptors by checking DD bits.

Fixes: 1060591eada5 ("net/avf: enable bulk allocate Rx")

Signed-off-by: Kathleen Capella <kathleen.capella@arm.com>
Reviewed-by: Honnappa Nagarahalli <honnappa.nagarahalli@arm.com>
Reviewed-by: Qi Zhang <qi.z.zhang@intel.com>
---
 drivers/net/iavf/iavf_rxtx.c | 26 ++++++++++++++++++++++----
 1 file changed, 22 insertions(+), 4 deletions(-)

diff --git a/drivers/net/iavf/iavf_rxtx.c b/drivers/net/iavf/iavf_rxtx.c
index bc67712d9c..d406f0d992 100644
--- a/drivers/net/iavf/iavf_rxtx.c
+++ b/drivers/net/iavf/iavf_rxtx.c
@@ -1101,7 +1101,7 @@ iavf_rx_scan_hw_ring(struct iavf_rx_queue *rxq)
 	uint16_t pkt_len;
 	uint64_t qword1;
 	uint32_t rx_status;
-	int32_t s[IAVF_LOOK_AHEAD], nb_dd;
+	int32_t s[IAVF_LOOK_AHEAD], var, nb_dd;
 	int32_t i, j, nb_rx = 0;
 	uint64_t pkt_flags;
 	static const uint32_t ptype_tbl[UINT8_MAX + 1] __rte_cache_aligned = {
@@ -1150,9 +1150,27 @@ iavf_rx_scan_hw_ring(struct iavf_rx_queue *rxq)
 
 		rte_smp_rmb();
 
-		/* Compute how many status bits were set */
-		for (j = 0, nb_dd = 0; j < IAVF_LOOK_AHEAD; j++)
-			nb_dd += s[j] & (1 << IAVF_RX_DESC_STATUS_DD_SHIFT);
+		/* Compute how many contiguous DD bits were set */
+		for (j = 0, nb_dd = 0; j < IAVF_LOOK_AHEAD; j++) {
+			var = s[j] & (1 << IAVF_RX_DESC_STATUS_DD_SHIFT);
+#ifdef RTE_ARCH_ARM
+			/* For Arm platforms, count only contiguous descriptors
+			 * whose DD bit is set to 1. On Arm platforms, reads of
+			 * descriptors can be reordered. Since the CPU may
+			 * be reading the descriptors as the NIC updates them
+			 * in memory, it is possbile that the DD bit for a
+			 * descriptor earlier in the queue is read as not set
+			 * while the DD bit for a descriptor later in the queue
+			 * is read as set.
+			 */
+			if (var)
+				nb_dd += 1;
+			else
+				break;
+#else
+			nb_dd += var;
+#endif
+		}
 
 		nb_rx += nb_dd;
 
-- 
2.35.0

---
  Diff of the applied patch vs upstream commit (please double-check if non-empty:
---
--- -	2022-02-25 16:58:46.315135552 +0100
+++ 0054-net-iavf-count-continuous-DD-bits-for-Arm.patch	2022-02-25 16:58:44.300990467 +0100
@@ -1 +1 @@
-From 3d88d5e401f6d04882415866012a98bced0c3cb8 Mon Sep 17 00:00:00 2001
+From 55f2ea35c7cafdb178e59862d66eaf73b3282706 Mon Sep 17 00:00:00 2001
@@ -5,0 +6,2 @@
+[ upstream commit 3d88d5e401f6d04882415866012a98bced0c3cb8 ]
+
@@ -11 +12,0 @@
-Cc: stable@dpdk.org
@@ -21 +22 @@
-index 7158938643..f07d886821 100644
+index bc67712d9c..d406f0d992 100644
@@ -24 +25 @@
-@@ -1897,7 +1897,7 @@ iavf_rx_scan_hw_ring(struct iavf_rx_queue *rxq)
+@@ -1101,7 +1101,7 @@ iavf_rx_scan_hw_ring(struct iavf_rx_queue *rxq)
@@ -32,2 +33,2 @@
- 	const uint32_t *ptype_tbl = rxq->vsi->adapter->ptype_tbl;
-@@ -1928,9 +1928,27 @@ iavf_rx_scan_hw_ring(struct iavf_rx_queue *rxq)
+ 	static const uint32_t ptype_tbl[UINT8_MAX + 1] __rte_cache_aligned = {
+@@ -1150,9 +1150,27 @@ iavf_rx_scan_hw_ring(struct iavf_rx_queue *rxq)

  parent reply	other threads:[~2022-02-25 17:18 UTC|newest]

Thread overview: 57+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-02-25 17:14 patch 'net/hns3: fix residual MAC after setting default MAC' " christian.ehrhardt
2022-02-25 17:14 ` patch 'net/hns3: fix secondary process reference count' " christian.ehrhardt
2022-02-25 17:14 ` patch 'net/hns3: fix multi-process action register and unregister' " christian.ehrhardt
2022-02-25 17:14 ` patch 'net/hns3: unregister MP action on close for secondary' " christian.ehrhardt
2022-02-25 17:14 ` patch 'net/ice: build failure with make and GCC > 11' " christian.ehrhardt
2022-02-25 17:14 ` patch 'config/ppc: fix build with GCC >= 10' " christian.ehrhardt
2022-02-25 17:15 ` patch 'maintainers: update for stable branches' " christian.ehrhardt
2022-02-25 17:15 ` patch 'bus/ifpga: remove useless check while browsing devices' " christian.ehrhardt
2022-02-25 17:15 ` patch 'eal/linux: log hugepage create errors with filename' " christian.ehrhardt
2022-02-25 17:15 ` patch 'devtools: fix comment detection in forbidden token check' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/ixgbe: add vector Rx parameter " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/bnxt: fix xstats query' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/bonding: fix mode type mismatch' " christian.ehrhardt
2022-02-25 17:15 ` patch 'app/testpmd: fix dereference before null check' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/cxgbe: fix dangling pointer by mailbox access rework' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/mlx5: fix maximum packet headers size for TSO' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/nfp: remove useless range checks' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/sfc: validate queue span when parsing flow action RSS' " christian.ehrhardt
2022-02-25 17:15 ` patch 'raw/ifpga/base: fix SPI transaction' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/ice: fix link up when starting device' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/bnxt: handle ring cleanup in case of error' " christian.ehrhardt
2022-02-25 17:15 ` patch 'raw/ifpga/base: fix port feature ID' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/memif: remove unnecessary Rx interrupt stub' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/bonding: fix RSS with early configure' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/hns3: fix using enum as boolean' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/virtio: fix Tx queue 0 overriden by queue 128' " christian.ehrhardt
2022-02-25 17:15 ` patch 'vdpa/ifc: fix log info mismatch' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/virtio-user: check FD flags getting failure' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/mlx5: reject jump to root table' " christian.ehrhardt
2022-02-25 17:15 ` patch 'build: fix warning about using -Wextra flag' " christian.ehrhardt
2022-02-25 17:15 ` patch 'kni: fix ioctl signature' " christian.ehrhardt
2022-02-25 17:15 ` patch 'doc: fix KNI PMD name typo' " christian.ehrhardt
2022-02-25 17:15 ` patch 'ring: fix error code when creating ring' " christian.ehrhardt
2022-02-25 17:15 ` patch 'test/mem: fix error check' " christian.ehrhardt
2022-02-25 17:15 ` patch 'bus/dpaa: fix C++ include guard' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/cxgbe: remove useless " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/dpaa2: " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/hns3: fix max packet size rollback in PF' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/hns3: fix RSS key with null' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/ixgbe: check filter init failure' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/bonding: fix promiscuous and allmulticast state' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/bonding: fix reference count on mbufs' " christian.ehrhardt
2022-02-25 17:15 ` patch 'app/testpmd: fix bonding mode set' " christian.ehrhardt
2022-02-25 17:15 ` patch 'stack: fix stubs header export' " christian.ehrhardt
2022-02-25 17:15 ` patch 'ipsec: fix C++ include' " christian.ehrhardt
2022-02-25 17:15 ` patch 'table: " christian.ehrhardt
2022-02-25 17:15 ` patch 'vhost: " christian.ehrhardt
2022-02-25 17:15 ` patch 'test/mbuf: fix mbuf data content check' " christian.ehrhardt
2022-02-25 17:15 ` patch 'ipc: end multiprocess thread during cleanup' " christian.ehrhardt
2022-02-25 17:15 ` patch 'vfio: cleanup the multiprocess sync handle' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/memif: remove pointer deference before null check' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/sfc: do not push fast free offload to default TxQ config' " christian.ehrhardt
2022-02-25 17:15 ` patch 'net/sfc: demand Tx fast free offload on EF10 simple datapath' " christian.ehrhardt
2022-02-25 17:15 ` christian.ehrhardt [this message]
2022-02-25 17:15 ` patch 'net/mlx5: fix committed bucket size' " christian.ehrhardt
2022-02-25 17:15 ` patch 'compress/octeontx: fix null pointer dereference' " christian.ehrhardt
2022-02-25 17:15 ` patch 'raw/ntb: clear all valid doorbell bits on init' " christian.ehrhardt

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=20220225171550.3499040-54-christian.ehrhardt@canonical.com \
    --to=christian.ehrhardt@canonical.com \
    --cc=honnappa.nagarahalli@arm.com \
    --cc=kathleen.capella@arm.com \
    --cc=qi.z.zhang@intel.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).