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, Luo bin <luobin9@huawei.com>,
	Jakub Kicinski <kuba@kernel.org>,
	"David S. Miller" <davem@davemloft.net>
Subject: [PATCH 5.8 48/56] hinic: fix rewaking txq after netif_tx_disable
Date: Fri, 25 Sep 2020 14:48:38 +0200	[thread overview]
Message-ID: <20200925124735.028887932@linuxfoundation.org> (raw)
In-Reply-To: <20200925124727.878494124@linuxfoundation.org>

From: Luo bin <luobin9@huawei.com>

[ Upstream commit a1b80e0143a1b878f8e21d82fd55f3f46f0014be ]

When calling hinic_close in hinic_set_channels, all queues are
stopped after netif_tx_disable, but some queue may be rewaken in
free_tx_poll by mistake while drv is handling tx irq. If one queue
is rewaken core may call hinic_xmit_frame to send pkt after
netif_tx_disable within a short time which may results in accessing
memory that has been already freed in hinic_close. So we call
napi_disable before netif_tx_disable in hinic_close to fix this bug.

Fixes: 2eed5a8b614b ("hinic: add set_channels ethtool_ops support")
Signed-off-by: Luo bin <luobin9@huawei.com>
Reviewed-by: Jakub Kicinski <kuba@kernel.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
---
 drivers/net/ethernet/huawei/hinic/hinic_main.c |   24 ++++++++++++++++++++++++
 drivers/net/ethernet/huawei/hinic/hinic_tx.c   |   18 +++---------------
 2 files changed, 27 insertions(+), 15 deletions(-)

--- a/drivers/net/ethernet/huawei/hinic/hinic_main.c
+++ b/drivers/net/ethernet/huawei/hinic/hinic_main.c
@@ -168,6 +168,24 @@ err_init_txq:
 	return err;
 }
 
+static void enable_txqs_napi(struct hinic_dev *nic_dev)
+{
+	int num_txqs = hinic_hwdev_num_qps(nic_dev->hwdev);
+	int i;
+
+	for (i = 0; i < num_txqs; i++)
+		napi_enable(&nic_dev->txqs[i].napi);
+}
+
+static void disable_txqs_napi(struct hinic_dev *nic_dev)
+{
+	int num_txqs = hinic_hwdev_num_qps(nic_dev->hwdev);
+	int i;
+
+	for (i = 0; i < num_txqs; i++)
+		napi_disable(&nic_dev->txqs[i].napi);
+}
+
 /**
  * free_txqs - Free the Logical Tx Queues of specific NIC device
  * @nic_dev: the specific NIC device
@@ -394,6 +412,8 @@ int hinic_open(struct net_device *netdev
 		goto err_create_txqs;
 	}
 
+	enable_txqs_napi(nic_dev);
+
 	err = create_rxqs(nic_dev);
 	if (err) {
 		netif_err(nic_dev, drv, netdev,
@@ -475,6 +495,7 @@ err_port_state:
 	}
 
 err_create_rxqs:
+	disable_txqs_napi(nic_dev);
 	free_txqs(nic_dev);
 
 err_create_txqs:
@@ -488,6 +509,9 @@ int hinic_close(struct net_device *netde
 	struct hinic_dev *nic_dev = netdev_priv(netdev);
 	unsigned int flags;
 
+	/* Disable txq napi firstly to aviod rewaking txq in free_tx_poll */
+	disable_txqs_napi(nic_dev);
+
 	down(&nic_dev->mgmt_lock);
 
 	flags = nic_dev->flags;
--- a/drivers/net/ethernet/huawei/hinic/hinic_tx.c
+++ b/drivers/net/ethernet/huawei/hinic/hinic_tx.c
@@ -684,18 +684,6 @@ static int free_tx_poll(struct napi_stru
 	return budget;
 }
 
-static void tx_napi_add(struct hinic_txq *txq, int weight)
-{
-	netif_napi_add(txq->netdev, &txq->napi, free_tx_poll, weight);
-	napi_enable(&txq->napi);
-}
-
-static void tx_napi_del(struct hinic_txq *txq)
-{
-	napi_disable(&txq->napi);
-	netif_napi_del(&txq->napi);
-}
-
 static irqreturn_t tx_irq(int irq, void *data)
 {
 	struct hinic_txq *txq = data;
@@ -724,7 +712,7 @@ static int tx_request_irq(struct hinic_t
 	struct hinic_sq *sq = txq->sq;
 	int err;
 
-	tx_napi_add(txq, nic_dev->tx_weight);
+	netif_napi_add(txq->netdev, &txq->napi, free_tx_poll, nic_dev->tx_weight);
 
 	hinic_hwdev_msix_set(nic_dev->hwdev, sq->msix_entry,
 			     TX_IRQ_NO_PENDING, TX_IRQ_NO_COALESC,
@@ -734,7 +722,7 @@ static int tx_request_irq(struct hinic_t
 	err = request_irq(sq->irq, tx_irq, 0, txq->irq_name, txq);
 	if (err) {
 		dev_err(&pdev->dev, "Failed to request Tx irq\n");
-		tx_napi_del(txq);
+		netif_napi_del(&txq->napi);
 		return err;
 	}
 
@@ -746,7 +734,7 @@ static void tx_free_irq(struct hinic_txq
 	struct hinic_sq *sq = txq->sq;
 
 	free_irq(sq->irq, txq);
-	tx_napi_del(txq);
+	netif_napi_del(&txq->napi);
 }
 
 /**



  parent reply	other threads:[~2020-09-25 12:50 UTC|newest]

Thread overview: 67+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2020-09-25 12:47 [PATCH 5.8 00/56] 5.8.12-rc1 review Greg Kroah-Hartman
2020-09-25 12:47 ` [PATCH 5.8 01/56] ibmvnic fix NULL tx_pools and rx_tools issue at do_reset Greg Kroah-Hartman
2020-09-25 12:47 ` [PATCH 5.8 02/56] ibmvnic: add missing parenthesis in do_reset() Greg Kroah-Hartman
2020-09-25 12:47 ` [PATCH 5.8 03/56] act_ife: load meta modules before tcf_idr_check_alloc() Greg Kroah-Hartman
2020-09-25 12:47 ` [PATCH 5.8 04/56] bnxt_en: Avoid sending firmware messages when AER error is detected Greg Kroah-Hartman
2020-09-25 12:47 ` [PATCH 5.8 05/56] bnxt_en: Fix NULL ptr dereference crash in bnxt_fw_reset_task() Greg Kroah-Hartman
2020-09-25 12:47 ` [PATCH 5.8 06/56] cxgb4: fix memory leak during module unload Greg Kroah-Hartman
2020-09-25 12:47 ` [PATCH 5.8 07/56] cxgb4: Fix offset when clearing filter byte counters Greg Kroah-Hartman
2020-09-25 12:47 ` [PATCH 5.8 08/56] geneve: add transport ports in route lookup for geneve Greg Kroah-Hartman
2020-09-25 12:47 ` [PATCH 5.8 09/56] hdlc_ppp: add range checks in ppp_cp_parse_cr() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 10/56] hinic: bump up the timeout of SET_FUNC_STATE cmd Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 11/56] ip: fix tos reflection in ack and reset packets Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 12/56] ipv4: Initialize flowi4_multipath_hash in data path Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 13/56] ipv4: Update exception handling for multipath routes via same device Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 14/56] ipv6: avoid lockdep issue in fib6_del() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 15/56] net: bridge: br_vlan_get_pvid_rcu() should dereference the VLAN group under RCU Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 16/56] net: DCB: Validate DCB_ATTR_DCB_BUFFER argument Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 17/56] net: dsa: rtl8366: Properly clear member config Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 18/56] net: Fix bridge enslavement failure Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 19/56] net: ipv6: fix kconfig dependency warning for IPV6_SEG6_HMAC Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 20/56] net/mlx5: Fix FTE cleanup Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 21/56] net: phy: call phy_disable_interrupts() in phy_attach_direct() instead Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 22/56] net: sched: initialize with 0 before setting erspan md->u Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 23/56] net: sch_generic: aviod concurrent reset and enqueue op for lockless qdisc Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 24/56] net: sctp: Fix IPv6 ancestor_size calc in sctp_copy_descendant Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 25/56] nfp: use correct define to return NONE fec Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 26/56] taprio: Fix allowing too small intervals Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 27/56] tipc: Fix memory leak in tipc_group_create_member() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 28/56] tipc: fix shutdown() of connection oriented socket Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 29/56] tipc: use skb_unshare() instead in tipc_buf_append() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 30/56] net/mlx5e: Enable adding peer miss rules only if merged eswitch is supported Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 31/56] net/mlx5e: TLS, Do not expose FPGA TLS counter if not supported Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 32/56] bnxt_en: Use memcpy to copy VPD field info Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 33/56] bnxt_en: return proper error codes in bnxt_show_temp Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 34/56] bnxt_en: Protect bnxt_set_eee() and bnxt_set_pauseparam() with mutex Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 35/56] net: lantiq: Wake TX queue again Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 36/56] net: lantiq: use netif_tx_napi_add() for TX NAPI Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 37/56] net: lantiq: Use napi_complete_done() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 38/56] net: lantiq: Disable IRQs only if NAPI gets scheduled Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 39/56] net: phy: Avoid NPD upon phy_detach() when driver is unbound Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 40/56] net: phy: Do not warn in phy_stop() on PHY_DOWN Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 41/56] net: qrtr: check skb_put_padto() return value Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 42/56] net: add __must_check to skb_put_padto() Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 43/56] net: ethernet: ti: cpsw_new: fix suspend/resume Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 44/56] wireguard: noise: take lock when removing handshake entry from table Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 45/56] wireguard: peerlookup: take lock before checking hash in replace operation Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 46/56] net: ipa: fix u32_replace_bits by u32p_xxx version Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 47/56] net/mlx5e: Fix memory leak of tunnel info when rule under multipath not ready Greg Kroah-Hartman
2020-09-25 12:48 ` Greg Kroah-Hartman [this message]
2020-09-25 12:48 ` [PATCH 5.8 49/56] hv_netvsc: Fix hibernation for mlx5 VF driver Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 50/56] net: dsa: link interfaces with the DSA master to get rid of lockdep warnings Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 51/56] net: dsa: microchip: ksz8795: really set the correct number of ports Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 52/56] net: macb: fix for pause frame receive enable bit Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 53/56] Revert "netns: dont disable BHs when locking "nsid_lock"" Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 54/56] net/mlx5e: Use RCU to protect rq->xdp_prog Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 55/56] net/mlx5e: Use synchronize_rcu to sync with NAPI Greg Kroah-Hartman
2020-09-25 12:48 ` [PATCH 5.8 56/56] net/mlx5e: Fix endianness when calculating pedit mask first bit Greg Kroah-Hartman
2020-09-25 18:01 ` [PATCH 5.8 00/56] 5.8.12-rc1 review Jon Hunter
2020-09-26 16:09   ` Greg Kroah-Hartman
2020-09-25 18:32 ` Naresh Kamboju
2020-09-26 16:10   ` Greg Kroah-Hartman
2020-09-25 20:01 ` Shuah Khan
2020-09-26 16:10   ` Greg Kroah-Hartman
2020-09-25 21:22 ` Jeffrin Jose T
2020-09-26 16:10   ` Greg Kroah-Hartman
2020-09-26 15:44 ` Guenter Roeck
2020-09-26 16:10   ` 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=20200925124735.028887932@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=davem@davemloft.net \
    --cc=kuba@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luobin9@huawei.com \
    --cc=stable@vger.kernel.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).