From 6e9d554fa6e481a848358c215f129432262123c0 Mon Sep 17 00:00:00 2001 From: Wolfgang Grandegger Date: Mon, 12 Dec 2011 16:09:28 +0100 Subject: can: flexcan: fix irq flooding by clearing all interrupt sources MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit As pointed out by Reuben Dowle and Lothar Waßmann, the TWRN_INT, RWRN_INT, BOFF_INT interrupt sources need to be cleared as well to avoid interrupt flooding, at least for the Flexcan on i.MX28 SOCs. Furthermore, the interrupts are only cleared, if really one of those interrupt sources are pending (which is not the case for rx and tx done). Cc: Reuben Dowle Cc: Lothar Waßmann Signed-off-by: Wolfgang Grandegger Signed-off-by: Marc Kleine-Budde --- drivers/net/can/flexcan.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-) (limited to 'drivers/net/can') diff --git a/drivers/net/can/flexcan.c b/drivers/net/can/flexcan.c index 7fd8089946fb..96d235799ec1 100644 --- a/drivers/net/can/flexcan.c +++ b/drivers/net/can/flexcan.c @@ -118,6 +118,9 @@ (FLEXCAN_ESR_TWRN_INT | FLEXCAN_ESR_RWRN_INT | FLEXCAN_ESR_BOFF_INT) #define FLEXCAN_ESR_ERR_ALL \ (FLEXCAN_ESR_ERR_BUS | FLEXCAN_ESR_ERR_STATE) +#define FLEXCAN_ESR_ALL_INT \ + (FLEXCAN_ESR_TWRN_INT | FLEXCAN_ESR_RWRN_INT | \ + FLEXCAN_ESR_BOFF_INT | FLEXCAN_ESR_ERR_INT) /* FLEXCAN interrupt flag register (IFLAG) bits */ #define FLEXCAN_TX_BUF_ID 8 @@ -577,7 +580,9 @@ static irqreturn_t flexcan_irq(int irq, void *dev_id) reg_iflag1 = flexcan_read(®s->iflag1); reg_esr = flexcan_read(®s->esr); - flexcan_write(FLEXCAN_ESR_ERR_INT, ®s->esr); /* ACK err IRQ */ + /* ACK all bus error and state change IRQ sources */ + if (reg_esr & FLEXCAN_ESR_ALL_INT) + flexcan_write(reg_esr & FLEXCAN_ESR_ALL_INT, ®s->esr); /* * schedule NAPI in case of: -- cgit v1.2.1 From 7bb4db93ae59e0faf810a83a8578f56bc968ab01 Mon Sep 17 00:00:00 2001 From: Wolfgang Grandegger Date: Mon, 12 Dec 2011 16:07:16 +0100 Subject: can: cc770: store echo skb before starting the transfer On slow systems and high CAN bitrates, the error message "can_put_echo_skb: BUG! echo_skb is occupied!" did show up because can_put_echo_skb() was called after starting the transfer. Signed-off-by: Wolfgang Grandegger Signed-off-by: Marc Kleine-Budde --- drivers/net/can/cc770/cc770.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'drivers/net/can') diff --git a/drivers/net/can/cc770/cc770.c b/drivers/net/can/cc770/cc770.c index 766896747643..c30f0e6f1048 100644 --- a/drivers/net/can/cc770/cc770.c +++ b/drivers/net/can/cc770/cc770.c @@ -440,12 +440,14 @@ static netdev_tx_t cc770_start_xmit(struct sk_buff *skb, struct net_device *dev) for (i = 0; i < dlc; i++) cc770_write_reg(priv, msgobj[mo].data[i], cf->data[i]); + /* Store echo skb before starting the transfer */ + can_put_echo_skb(skb, dev, 0); + cc770_write_reg(priv, msgobj[mo].ctrl1, RMTPND_RES | TXRQST_SET | CPUUPD_RES | NEWDAT_UNC); stats->tx_bytes += dlc; - can_put_echo_skb(skb, dev, 0); /* * HM: We had some cases of repeated IRQs so make sure the -- cgit v1.2.1 From 2d5091e08c684912ed6b9ca03d1f7b01501b7bf6 Mon Sep 17 00:00:00 2001 From: Wolfgang Zarre Date: Sun, 15 Jan 2012 13:21:34 +0100 Subject: can: cc770: Fix indirect access deadlock on ISA cards This fix avoids a deadlock if an interrupt occurs during consecutive port operations on ISA cards utilising indirect access via address and data port. Tested on a B&R ISA card. Cc: linux-can@vger.kernel.org Cc: netdev@vger.kernel.org Signed-off-by: Wolfgang Zarre Acked-by: Wolfgang Grandegger Signed-off-by: Marc Kleine-Budde --- drivers/net/can/cc770/cc770_isa.c | 16 +++++++++++++++- 1 file changed, 15 insertions(+), 1 deletion(-) (limited to 'drivers/net/can') diff --git a/drivers/net/can/cc770/cc770_isa.c b/drivers/net/can/cc770/cc770_isa.c index 4be5fe2c40a5..9f3a25ccd665 100644 --- a/drivers/net/can/cc770/cc770_isa.c +++ b/drivers/net/can/cc770/cc770_isa.c @@ -110,6 +110,11 @@ MODULE_PARM_DESC(bcr, "Bus configuration register (default=0x40 [CBY])"); #define CC770_IOSIZE 0x20 #define CC770_IOSIZE_INDIRECT 0x02 +/* Spinlock for cc770_isa_port_write_reg_indirect + * and cc770_isa_port_read_reg_indirect + */ +static DEFINE_SPINLOCK(cc770_isa_port_lock); + static struct platform_device *cc770_isa_devs[MAXDEV]; static u8 cc770_isa_mem_read_reg(const struct cc770_priv *priv, int reg) @@ -138,18 +143,27 @@ static u8 cc770_isa_port_read_reg_indirect(const struct cc770_priv *priv, int reg) { unsigned long base = (unsigned long)priv->reg_base; + unsigned long flags; + u8 val; + spin_lock_irqsave(&cc770_isa_port_lock, flags); outb(reg, base); - return inb(base + 1); + val = inb(base + 1); + spin_unlock_irqrestore(&cc770_isa_port_lock, flags); + + return val; } static void cc770_isa_port_write_reg_indirect(const struct cc770_priv *priv, int reg, u8 val) { unsigned long base = (unsigned long)priv->reg_base; + unsigned long flags; + spin_lock_irqsave(&cc770_isa_port_lock, flags); outb(reg, base); outb(val, base + 1); + spin_unlock_irqrestore(&cc770_isa_port_lock, flags); } static int __devinit cc770_isa_probe(struct platform_device *pdev) -- cgit v1.2.1 From e3f240f460a36b158217944b52a85f304914c1a6 Mon Sep 17 00:00:00 2001 From: Wolfgang Grandegger Date: Wed, 1 Feb 2012 10:50:23 +0100 Subject: can: ti_hecc: use netif_rx in the interrupt This patch fixes locking problems by using netif_rx() instead of netif_receive_skb() in ti_hecc_error() called from the interrupt context. Signed-off-by: Wolfgang Grandegger Signed-off-by: Marc Kleine-Budde --- drivers/net/can/ti_hecc.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers/net/can') diff --git a/drivers/net/can/ti_hecc.c b/drivers/net/can/ti_hecc.c index df809e3f130e..5a2e1e3588a1 100644 --- a/drivers/net/can/ti_hecc.c +++ b/drivers/net/can/ti_hecc.c @@ -745,9 +745,10 @@ static int ti_hecc_error(struct net_device *ndev, int int_status, } } - netif_receive_skb(skb); + netif_rx(skb); stats->rx_packets++; stats->rx_bytes += cf->can_dlc; + return 0; } -- cgit v1.2.1 From 44b0052c5cb4e75389ed3eb9e98c29295a7dadfb Mon Sep 17 00:00:00 2001 From: Xi Wang Date: Mon, 12 Dec 2011 02:16:20 -0500 Subject: can: pch_can: fix error passive level test The test (((errc & PCH_REC) >> 8) > 127) would always be false because the receive error counter ((errc & PCH_REC) >> 8) is at most 127, where PCH_REC is defined as 0x7f00. To test whether the receive error counter has reached the error passive level, the RP bit (15) should be used. Signed-off-by: Xi Wang Acked-by: Wolfgang Grandegger Signed-off-by: Marc Kleine-Budde --- drivers/net/can/pch_can.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'drivers/net/can') diff --git a/drivers/net/can/pch_can.c b/drivers/net/can/pch_can.c index d11fbb2b95ff..6edc25e0dd15 100644 --- a/drivers/net/can/pch_can.c +++ b/drivers/net/can/pch_can.c @@ -66,6 +66,7 @@ #define PCH_IF_CREQ_BUSY BIT(15) #define PCH_STATUS_INT 0x8000 +#define PCH_RP 0x00008000 #define PCH_REC 0x00007f00 #define PCH_TEC 0x000000ff @@ -527,7 +528,7 @@ static void pch_can_error(struct net_device *ndev, u32 status) priv->can.can_stats.error_passive++; state = CAN_STATE_ERROR_PASSIVE; cf->can_id |= CAN_ERR_CRTL; - if (((errc & PCH_REC) >> 8) > 127) + if (errc & PCH_RP) cf->data[1] |= CAN_ERR_CRTL_RX_PASSIVE; if ((errc & PCH_TEC) > 127) cf->data[1] |= CAN_ERR_CRTL_TX_PASSIVE; -- cgit v1.2.1 From d0a71a7e6de0e0ce9f86c8ba6e13414a9df63e0b Mon Sep 17 00:00:00 2001 From: Sebastian Haas Date: Thu, 22 Dec 2011 23:57:49 +0100 Subject: can: ems_usb: Removed double netif_device_detach netif_device_attched is called twice when ems_usb_start fails with -ENODEV Signed-off-by: Sebastian Haas Signed-off-by: Marc Kleine-Budde --- drivers/net/can/usb/ems_usb.c | 9 --------- 1 file changed, 9 deletions(-) (limited to 'drivers/net/can') diff --git a/drivers/net/can/usb/ems_usb.c b/drivers/net/can/usb/ems_usb.c index 9697c14b8dc6..7dae64d44e83 100644 --- a/drivers/net/can/usb/ems_usb.c +++ b/drivers/net/can/usb/ems_usb.c @@ -627,9 +627,6 @@ static int ems_usb_start(struct ems_usb *dev) err = usb_submit_urb(urb, GFP_KERNEL); if (err) { - if (err == -ENODEV) - netif_device_detach(dev->netdev); - usb_unanchor_urb(urb); usb_free_coherent(dev->udev, RX_BUFFER_SIZE, buf, urb->transfer_dma); @@ -659,9 +656,6 @@ static int ems_usb_start(struct ems_usb *dev) err = usb_submit_urb(dev->intr_urb, GFP_KERNEL); if (err) { - if (err == -ENODEV) - netif_device_detach(dev->netdev); - dev_warn(netdev->dev.parent, "intr URB submit failed: %d\n", err); @@ -692,9 +686,6 @@ static int ems_usb_start(struct ems_usb *dev) return 0; failed: - if (err == -ENODEV) - netif_device_detach(dev->netdev); - dev_warn(netdev->dev.parent, "couldn't submit control: %d\n", err); return err; -- cgit v1.2.1 From 29830406415c227a54af429d7b300aabd4754237 Mon Sep 17 00:00:00 2001 From: Stephane Grosjean Date: Wed, 1 Feb 2012 11:05:48 +0100 Subject: can: peak_pci: Fix the way channels are linked together Change the way channels objects are linked together by peak_pci_probe() avoiding any kernel oops when driver is removed. Side effect is that the list is now browsed from last to first channel. Signed-off-by: Stephane Grosjean Signed-off-by: Marc Kleine-Budde --- drivers/net/can/sja1000/peak_pci.c | 23 +++++++++-------------- 1 file changed, 9 insertions(+), 14 deletions(-) (limited to 'drivers/net/can') diff --git a/drivers/net/can/sja1000/peak_pci.c b/drivers/net/can/sja1000/peak_pci.c index 2c7f5036f570..214795945bc4 100644 --- a/drivers/net/can/sja1000/peak_pci.c +++ b/drivers/net/can/sja1000/peak_pci.c @@ -39,9 +39,9 @@ MODULE_LICENSE("GPL v2"); #define DRV_NAME "peak_pci" struct peak_pci_chan { - void __iomem *cfg_base; /* Common for all channels */ - struct net_device *next_dev; /* Chain of network devices */ - u16 icr_mask; /* Interrupt mask for fast ack */ + void __iomem *cfg_base; /* Common for all channels */ + struct net_device *prev_dev; /* Chain of network devices */ + u16 icr_mask; /* Interrupt mask for fast ack */ }; #define PEAK_PCI_CAN_CLOCK (16000000 / 2) @@ -98,7 +98,7 @@ static int __devinit peak_pci_probe(struct pci_dev *pdev, { struct sja1000_priv *priv; struct peak_pci_chan *chan; - struct net_device *dev, *dev0 = NULL; + struct net_device *dev; void __iomem *cfg_base, *reg_base; u16 sub_sys_id, icr; int i, err, channels; @@ -196,18 +196,14 @@ static int __devinit peak_pci_probe(struct pci_dev *pdev, } /* Create chain of SJA1000 devices */ - if (i == 0) - dev0 = dev; - else - chan->next_dev = dev; + chan->prev_dev = pci_get_drvdata(pdev); + pci_set_drvdata(pdev, dev); dev_info(&pdev->dev, "%s at reg_base=0x%p cfg_base=0x%p irq=%d\n", dev->name, priv->reg_base, chan->cfg_base, dev->irq); } - pci_set_drvdata(pdev, dev0); - /* Enable interrupts */ writew(icr, cfg_base + PITA_ICR + 2); @@ -217,12 +213,11 @@ failure_remove_channels: /* Disable interrupts */ writew(0x0, cfg_base + PITA_ICR + 2); - for (dev = dev0; dev; dev = chan->next_dev) { + for (dev = pci_get_drvdata(pdev); dev; dev = chan->prev_dev) { unregister_sja1000dev(dev); free_sja1000dev(dev); priv = netdev_priv(dev); chan = priv->priv; - dev = chan->next_dev; } pci_iounmap(pdev, reg_base); @@ -241,7 +236,7 @@ failure_disable_pci: static void __devexit peak_pci_remove(struct pci_dev *pdev) { - struct net_device *dev = pci_get_drvdata(pdev); /* First device */ + struct net_device *dev = pci_get_drvdata(pdev); /* Last device */ struct sja1000_priv *priv = netdev_priv(dev); struct peak_pci_chan *chan = priv->priv; void __iomem *cfg_base = chan->cfg_base; @@ -255,7 +250,7 @@ static void __devexit peak_pci_remove(struct pci_dev *pdev) dev_info(&pdev->dev, "removing device %s\n", dev->name); unregister_sja1000dev(dev); free_sja1000dev(dev); - dev = chan->next_dev; + dev = chan->prev_dev; if (!dev) break; priv = netdev_priv(dev); -- cgit v1.2.1