Netdev Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Marc Kleine-Budde <mkl@pengutronix.de>
To: netdev@vger.kernel.org
Cc: davem@davemloft.net, kuba@kernel.org, linux-can@vger.kernel.org,
	kernel@pengutronix.de, "Marc Kleine-Budde" <mkl@pengutronix.de>,
	"Daniel Glöckner" <dg@emlix.com>,
	"Oleksij Rempel" <o.rempel@pengutronix.de>
Subject: [PATCH net-next 09/46] can: rx-offload: can_rx_offload_threaded_irq_finish(): add new function to be called from threaded interrupt
Date: Mon, 26 Jul 2021 16:11:07 +0200	[thread overview]
Message-ID: <20210726141144.862529-10-mkl@pengutronix.de> (raw)
In-Reply-To: <20210726141144.862529-1-mkl@pengutronix.de>

After reading all CAN frames from the controller in the IRQ handler
and storing them into a skb_queue, the driver calls napi_schedule().
In the napi poll function the skb from the skb_queue are then pushed
into the networking stack.

However if napi_schedule() is called from a threaded IRQ handler this
triggers the following error:

| NOHZ tick-stop error: Non-RCU local softirq work is pending, handler #08!!!

To avoid this, create a new rx-offload
function (can_rx_offload_threaded_irq_finish()) with a call to
local_bh_disable()/local_bh_enable() around the napi_schedule() call.

Convert all drivers that call can_rx_offload_irq_finish() from
threaded IRQ context to can_rx_offload_threaded_irq_finish().

Link: https://lore.kernel.org/r/20210724204745.736053-4-mkl@pengutronix.de
Suggested-by: Daniel Glöckner <dg@emlix.com>
Tested-by: Oleksij Rempel <o.rempel@pengutronix.de>
Signed-off-by: Marc Kleine-Budde <mkl@pengutronix.de>
---
 drivers/net/can/dev/rx-offload.c              | 23 +++++++++++++++++++
 drivers/net/can/m_can/m_can.c                 |  2 +-
 .../net/can/spi/mcp251xfd/mcp251xfd-core.c    |  4 ++--
 include/linux/can/rx-offload.h                |  1 +
 4 files changed, 27 insertions(+), 3 deletions(-)

diff --git a/drivers/net/can/dev/rx-offload.c b/drivers/net/can/dev/rx-offload.c
index 82ade3aa5c13..37b0cc65237b 100644
--- a/drivers/net/can/dev/rx-offload.c
+++ b/drivers/net/can/dev/rx-offload.c
@@ -299,6 +299,29 @@ void can_rx_offload_irq_finish(struct can_rx_offload *offload)
 }
 EXPORT_SYMBOL_GPL(can_rx_offload_irq_finish);
 
+void can_rx_offload_threaded_irq_finish(struct can_rx_offload *offload)
+{
+	unsigned long flags;
+	int queue_len;
+
+	if (skb_queue_empty_lockless(&offload->skb_irq_queue))
+		return;
+
+	spin_lock_irqsave(&offload->skb_queue.lock, flags);
+	skb_queue_splice_tail_init(&offload->skb_irq_queue, &offload->skb_queue);
+	spin_unlock_irqrestore(&offload->skb_queue.lock, flags);
+
+	queue_len = skb_queue_len(&offload->skb_queue);
+	if (queue_len > offload->skb_queue_len_max / 8)
+		netdev_dbg(offload->dev, "%s: queue_len=%d\n",
+			   __func__, queue_len);
+
+	local_bh_disable();
+	napi_schedule(&offload->napi);
+	local_bh_enable();
+}
+EXPORT_SYMBOL_GPL(can_rx_offload_threaded_irq_finish);
+
 static int can_rx_offload_init_queue(struct net_device *dev,
 				     struct can_rx_offload *offload,
 				     unsigned int weight)
diff --git a/drivers/net/can/m_can/m_can.c b/drivers/net/can/m_can/m_can.c
index 18461982f7a1..317cdc98c539 100644
--- a/drivers/net/can/m_can/m_can.c
+++ b/drivers/net/can/m_can/m_can.c
@@ -1059,7 +1059,7 @@ static irqreturn_t m_can_isr(int irq, void *dev_id)
 	}
 
 	if (cdev->is_peripheral)
-		can_rx_offload_irq_finish(&cdev->offload);
+		can_rx_offload_threaded_irq_finish(&cdev->offload);
 
 	return IRQ_HANDLED;
 }
diff --git a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
index f3b267ec22e0..6962ab2749df 100644
--- a/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
+++ b/drivers/net/can/spi/mcp251xfd/mcp251xfd-core.c
@@ -2196,7 +2196,7 @@ static irqreturn_t mcp251xfd_irq(int irq, void *dev_id)
 				  priv->regs_status.intf);
 
 		if (!(intf_pending)) {
-			can_rx_offload_irq_finish(&priv->offload);
+			can_rx_offload_threaded_irq_finish(&priv->offload);
 			return handled;
 		}
 
@@ -2298,7 +2298,7 @@ static irqreturn_t mcp251xfd_irq(int irq, void *dev_id)
 	} while (1);
 
  out_fail:
-	can_rx_offload_irq_finish(&priv->offload);
+	can_rx_offload_threaded_irq_finish(&priv->offload);
 
 	netdev_err(priv->ndev, "IRQ handler returned %d (intf=0x%08x).\n",
 		   err, priv->regs_status.intf);
diff --git a/include/linux/can/rx-offload.h b/include/linux/can/rx-offload.h
index 516f64df0ebc..c11477620403 100644
--- a/include/linux/can/rx-offload.h
+++ b/include/linux/can/rx-offload.h
@@ -50,6 +50,7 @@ unsigned int can_rx_offload_get_echo_skb(struct can_rx_offload *offload,
 int can_rx_offload_queue_tail(struct can_rx_offload *offload,
 			      struct sk_buff *skb);
 void can_rx_offload_irq_finish(struct can_rx_offload *offload);
+void can_rx_offload_threaded_irq_finish(struct can_rx_offload *offload);
 void can_rx_offload_del(struct can_rx_offload *offload);
 void can_rx_offload_enable(struct can_rx_offload *offload);
 
-- 
2.30.2



  parent reply	other threads:[~2021-07-26 14:13 UTC|newest]

Thread overview: 48+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-26 14:10 pull-request: can-next 2021-07-25 Marc Kleine-Budde
2021-07-26 14:10 ` [PATCH net-next 01/46] can: j1939: j1939_sk_sock_destruct(): correct a grammatical error Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 02/46] can: j1939: fix checkpatch warnings Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 03/46] can: j1939: replace fall through comment by fallthrough pseudo-keyword Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 04/46] can: j1939: j1939_session_completed(): use consistent name se_skb for the session skb Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 05/46] can: j1939: j1939_session_tx_dat(): use consistent name se_skcb for session skb control buffer Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 06/46] can: j1939: j1939_xtp_rx_dat_one(): use separate pointer " Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 07/46] can: rx-offload: add skb queue for use during ISR Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 08/46] can: rx-offload: can_rx_offload_irq_finish(): directly call napi_schedule() Marc Kleine-Budde
2021-07-26 14:11 ` Marc Kleine-Budde [this message]
2021-07-26 14:11 ` [PATCH net-next 10/46] can: bittiming: fix documentation for struct can_tdc Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 11/46] can: netlink: clear data_bittiming if FD is turned off Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 12/46] can: netlink: remove redundant check in can_validate() Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 13/46] dt-bindings: net: can: Document transceiver implementation as phy Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 14/46] can: m_can: Add support for transceiver " Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 15/46] can: m_can: use devm_platform_ioremap_resource_byname Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 16/46] can: m_can: remove support for custom bit timing Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 17/46] can: mcp251xfd: mcp251xfd_probe(): try to get crystal clock rate from property Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 18/46] can: mcp251xfd: Fix header block to clarify independence from OF Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 19/46] can: mcp251xfd: mcp251xfd_open(): request IRQ as shared Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 20/46] can: esd_usb2: use DEVICE_ATTR_RO() helper macro Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 21/46] can: janz-ican3: use DEVICE_ATTR_RO/RW() " Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 22/46] can: at91_can: use DEVICE_ATTR_RW() " Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 23/46] net: at91_can: remove redundant blank lines Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 24/46] net: at91_can: add blank line after declarations Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 25/46] net: at91_can: fix the code style issue about macro Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 26/46] net: at91_can: use BIT macro Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 27/46] net: at91_can: fix the alignment issue Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 28/46] net: at91_can: add braces {} to all arms of the statement Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 29/46] net: at91_can: remove redundant space Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 30/46] net: at91_can: fix the comments style issue Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 31/46] can: peak_pci: convert comments to network style comments Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 32/46] can: peak_pci: fix checkpatch warnings Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 33/46] can: peak_pci: Add name and FW version of the card in kernel buffer Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 34/46] can: peak_usb: pcan_usb_get_device_id(): read value only in case of success Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 35/46] can: peak_usb: PCAN-USB: add support of loopback and one-shot mode Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 36/46] can: peak_usb: pcan_usb_encode_msg(): add information Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 37/46] can: peak_usb: pcan_usb_decode_error(): upgrade handling of bus state changes Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 38/46] can: etas_es58x: fix three typos in author name and documentation Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 39/46] can: etas_es58x: use error pointer during device probing Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 40/46] can: etas_es58x: use devm_kzalloc() to allocate device resources Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 41/46] can: etas_es58x: add es58x_free_netdevs() to factorize code Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 42/46] can: etas_es58x: use sizeof and sizeof_field macros instead of constant values Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 43/46] can: etas_es58x: rewrite the message cast in es58{1,_fd}_tx_can_msg to increase readability Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 44/46] can: flexcan: add platform data header Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 45/46] can: flexcan: add mcf5441x support Marc Kleine-Budde
2021-07-26 14:11 ` [PATCH net-next 46/46] can: flexcan: update Kconfig to enable coldfire Marc Kleine-Budde
2021-07-26 21:20 ` pull-request: can-next 2021-07-25 patchwork-bot+netdevbpf

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=20210726141144.862529-10-mkl@pengutronix.de \
    --to=mkl@pengutronix.de \
    --cc=davem@davemloft.net \
    --cc=dg@emlix.com \
    --cc=kernel@pengutronix.de \
    --cc=kuba@kernel.org \
    --cc=linux-can@vger.kernel.org \
    --cc=netdev@vger.kernel.org \
    --cc=o.rempel@pengutronix.de \
    --subject='Re: [PATCH net-next 09/46] can: rx-offload: can_rx_offload_threaded_irq_finish(): add new function to be called from threaded interrupt' \
    /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

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).