stable.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Tariq Toukan <tariqt@mellanox.com>,
	kernel-team@fb.com, Saeed Mahameed <saeedm@mellanox.com>,
	"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 4.10 03/63] net/mlx5e: Fix broken CQE compression initialization
Date: Mon, 20 Mar 2017 18:51:28 +0100	[thread overview]
Message-ID: <20170320174742.933264100@linuxfoundation.org> (raw)
In-Reply-To: <20170320174742.712298584@linuxfoundation.org>

4.10-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Tariq Toukan <tariqt@mellanox.com>


[ Upstream commit b0d4660b4cc52e6477ca3a43435351d565dfcedc ]

Some of RQ type parameters are derived from CQE compression state flag,
CQE compression flag was initialized only after RQ type parameters
setup. This leads to load RQ with stride size smaller than what we
want for when CQE compression is on.

This bug introduces no functional damage, it only makes CQE compression
occur less often, since in ConnectX4-LX CQE compression is performed
only on packets smaller than stride size.

Fix this by marking default status of CQE compression in PFLAG prior to
calling mlx5e_set_rq_priv_params(), as it inits some fields based on it.

Tested:
 load driver on systems where rx CQE compress will be on (MH)
 pktgen with  64 < pkt size < 256 and netperf TCP_STREAM (IPv4/IPv6)
 verify `ethtool -S ethxx | grep compress` are advancing more often
 (rapidly)

Fixes: 2fc4bfb7250d ("net/mlx5e: Dynamic RQ type infrastructure")
Signed-off-by: Tariq Toukan <tariqt@mellanox.com>
Cc: kernel-team@fb.com
Signed-off-by: Saeed Mahameed <saeedm@mellanox.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/net/ethernet/mellanox/mlx5/core/en_main.c |    4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

--- a/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
+++ b/drivers/net/ethernet/mellanox/mlx5/core/en_main.c
@@ -3500,6 +3500,9 @@ static void mlx5e_build_nic_netdev_priv(
 			cqe_compress_heuristic(link_speed, pci_bw);
 	}
 
+	MLX5E_SET_PFLAG(priv, MLX5E_PFLAG_RX_CQE_COMPRESS,
+			priv->params.rx_cqe_compress_def);
+
 	mlx5e_set_rq_priv_params(priv);
 	if (priv->params.rq_wq_type == MLX5_WQ_TYPE_LINKED_LIST_STRIDING_RQ)
 		priv->params.lro_en = true;
@@ -3525,7 +3528,6 @@ static void mlx5e_build_nic_netdev_priv(
 	/* Initialize pflags */
 	MLX5E_SET_PFLAG(priv, MLX5E_PFLAG_RX_CQE_BASED_MODER,
 			priv->params.rx_cq_period_mode == MLX5_CQ_PERIOD_MODE_START_FROM_CQE);
-	MLX5E_SET_PFLAG(priv, MLX5E_PFLAG_RX_CQE_COMPRESS, priv->params.rx_cqe_compress_def);
 
 	mutex_init(&priv->state_lock);
 

  parent reply	other threads:[~2017-03-20 17:51 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-03-20 17:51 [PATCH 4.10 00/63] 4.10.5-stable review Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 01/63] net/mlx5e: Register/unregister vport representors on interface attach/detach Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 02/63] net/mlx5e: Do not reduce LRO WQE size when not using build_skb Greg Kroah-Hartman
2017-03-20 17:51 ` Greg Kroah-Hartman [this message]
2017-03-20 17:51 ` [PATCH 4.10 04/63] net/mlx5e: Update MPWQE stride size when modifying CQE compress state Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 05/63] net/mlx5e: Fix wrong CQE decompression Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 06/63] sctp: deny peeloff operation on asocs with threads sleeping on it Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 07/63] vxlan: correctly validate VXLAN ID against VXLAN_N_VID Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 08/63] vti6: return GRE_KEY for vti6 Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 09/63] vxlan: dont allow overwrite of config src addr Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 10/63] ipv4: add missing initialization for flowi4_uid Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 11/63] ipv4: mask tos for input route Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 12/63] sctp: set sin_port for addr param when checking duplicate address Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 13/63] net sched actions: decrement module reference count after table flush Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 15/63] vxlan: lock RCU on TX path Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 16/63] geneve: " Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 17/63] mlxsw: spectrum_router: Avoid potential packets loss Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 18/63] tcp/dccp: block BH for SYN processing Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 19/63] net: bridge: allow IPv6 when multicast flood is disabled Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 20/63] net: dont call strlen() on the user buffer in packet_bind_spkt() Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 21/63] net: net_enable_timestamp() can be called from irq contexts Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 22/63] ipv6: orphan skbs in reassembly unit Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 23/63] dccp: Unlock sock before calling sk_free() Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 24/63] amd-xgbe: Stop the PHY before releasing interrupts Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 25/63] amd-xgbe: Be sure to set MDIO modes on device (re)start Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 26/63] amd-xgbe: Dont overwrite SFP PHY mod_absent settings Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 27/63] bonding: use ETH_MAX_MTU as max mtu Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 28/63] strparser: destroy workqueue on module exit Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 29/63] tcp: fix various issues for sockets morphing to listen state Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 30/63] net: fix socket refcounting in skb_complete_wifi_ack() Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 31/63] net: fix socket refcounting in skb_complete_tx_timestamp() Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 32/63] net/sched: act_skbmod: remove unneeded rcu_read_unlock in tcf_skbmod_dump Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 33/63] dccp: fix use-after-free in dccp_feat_activate_values Greg Kroah-Hartman
2017-03-20 17:51 ` [PATCH 4.10 34/63] team: use ETH_MAX_MTU as max mtu Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 35/63] vrf: Fix use-after-free in vrf_xmit Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 36/63] net/tunnel: set inner protocol in network gro hooks Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 37/63] uapi: fix linux/packet_diag.h userspace compilation error Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 38/63] amd-xgbe: Enable IRQs only if napi_complete_done() is true Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 40/63] mpls: Send route delete notifications when router module is unloaded Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 41/63] mpls: Do not decrement alive counter for unregister events Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 42/63] ipv6: make ECMP route replacement less greedy Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 43/63] ipv6: avoid write to a possibly cloned skb Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 44/63] bridge: drop netfilter fake rtable unconditionally Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 48/63] dccp: fix memory leak during tear-down of unsuccessful connection request Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 49/63] arm64: KVM: VHE: Clear HCR_TGE when invalidating guest TLBs Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 50/63] irqchip/gicv3-its: Add workaround for QDF2400 ITS erratum 0065 Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 54/63] x86/unwind: Fix last frame check for aligned function stacks Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 55/63] x86/tsc: Fix ART for TSC_KNOWN_FREQ Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 56/63] x86/kasan: Fix boot with KASAN=y and PROFILE_ANNOTATED_BRANCHES=y Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 57/63] x86/intel_rdt: Put group node in rdtgroup_kn_unlock Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 58/63] x86/perf: Fix CR4.PCE propagation to use active_mm instead of mm Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 59/63] futex: Fix potential use-after-free in FUTEX_REQUEUE_PI Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 60/63] futex: Add missing error handling to FUTEX_REQUEUE_PI Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 61/63] locking/rwsem: Fix down_write_killable() for CONFIG_RWSEM_GENERIC_SPINLOCK=y Greg Kroah-Hartman
2017-03-20 17:52 ` [PATCH 4.10 62/63] crypto: powerpc - Fix initialisation of crc32c context Greg Kroah-Hartman
2017-03-21  0:13 ` [PATCH 4.10 00/63] 4.10.5-stable review Shuah Khan
2017-03-21  4:29   ` Greg Kroah-Hartman
2017-03-21  2:14 ` Guenter Roeck
2017-03-21  4:36   ` Greg Kroah-Hartman

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=20170320174742.933264100@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=davem@davemloft.net \
    --cc=kernel-team@fb.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=saeedm@mellanox.com \
    --cc=stable@vger.kernel.org \
    --cc=tariqt@mellanox.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).