All of lore.kernel.org
 help / color / mirror / Atom feed
From: Murali Karicheri <m-karicheri2@ti.com>
To: Bjorn Helgaas <helgaas@kernel.org>
Cc: <bhelgaas@google.com>, <devicetree@vger.kernel.org>,
	<linux-kernel@vger.kernel.org>, <linux-pci@vger.kernel.org>,
	<linux-arm-kernel@lists.infradead.org>
Subject: Re: [PATCH next v2 1/2] PCI: keystone: add pci error irq handler
Date: Wed, 20 Apr 2016 10:13:36 -0400	[thread overview]
Message-ID: <57178E90.5090205@ti.com> (raw)
In-Reply-To: <20160420010343.GI17863@localhost>

On 04/19/2016 09:03 PM, Bjorn Helgaas wrote:
> Hi Murali,
> 
> On Mon, Apr 11, 2016 at 10:50:30AM -0400, Murali Karicheri wrote:
>> Keystone PCI hardware generates error interrupts at RC using platform
>> irq instead of standard msi/legacy irq. Add a simple error handler that
>> logs the fatal interrupt status to the console.
>>
>> Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
>> Acked-by: Rob Herring <robh@kernel.org>
>> Cc: Rob Herring <robh+dt@kernel.org>
>> Cc: Pawel Moll <pawel.moll@arm.com>
>> Cc: Mark Rutland <mark.rutland@arm.com>
>> Cc: Ian Campbell <ijc+devicetree@hellion.org.uk>
>> Cc: Kumar Gala <galak@codeaurora.org>
>> Cc: Bjorn Helgaas <bhelgaas@google.com>
> 
> I applied this with minor changes to pci/host-keystone for v4.7.
> See below:
> 
>> +bool ks_dw_pcie_handle_error_irq(struct device *dev, void __iomem *reg_base)
>> +{
>> +	u32 status;
>> +	bool ret = false;
>> +
>> +	status = readl(reg_base + ERR_IRQ_STATUS_RAW) & ERR_IRQ_ALL;
>> +	if (status) {
>> +		/* The PCIESS interrupt status buts are "write 1 to clear" */
>> +		if (status & ERR_FATAL_IRQ)
>> +			dev_err(dev, "PCIE fatal error detected\n");
>> +
>> +		/* ack the irq event. */
>> +		writel(status, reg_base + ERR_IRQ_STATUS);
>> +		ret = true;
>> +	}
>> +	return ret;
>> +}
> 
> It seems pointless to me to return true/false here and then:
> 
>> +static irqreturn_t pcie_err_irq_handler(int irq, void *priv)
>> +{
>> +	struct keystone_pcie *ks_pcie = priv;
>> +
>> +	if (ks_dw_pcie_handle_error_irq(ks_pcie->pp.dev, ks_pcie->va_app_base))
>> +		return IRQ_HANDLED;
>> +
>> +	return IRQ_NONE;
>> +}
> 
> convert the true/false to IRQ_HANDLED/IRQ_NONE here.  So I changed
> ks_dw_pcie_handle_error_irq() to return IRQ_HANDLED/IRQ_NONE directly,
> resulting in the patch below.

I think that should be fine. Only reason, i did that way was that the pcie-designware
code is kept as a library of functions to configure the designware hardware common to
all platforms that has the designware core h/w and keep all of the platform specific info
such as IRQ handler to the platform glue driver such as pci-keystone.c. But that is not
true today as there is msi irq handler in this driver. So this is fine.

I will pull these from your designware pci/host-keystone and let you know if I see
any issues or send a diff patch to the list.

Once again, thanks for pulling this to v4.7.

Murali
> 
> This is applied to pci/host-keystone for v4.7.
> 
> 
> commit 7be92716c1aa40f5690e6a5f01fc0d385dd72303
> Author: Murali Karicheri <m-karicheri2@ti.com>
> Date:   Mon Apr 11 10:50:30 2016 -0400
> 
>     PCI: keystone: Add error IRQ handler
>     
>     Keystone PCI hardware generates error interrupts at RC using a platform IRQ
>     instead of a standard MSI or legacy IRQ.  Add a simple error handler that
>     logs the fatal interrupt status to the console.
>     
>     [bhelgaas: tidy comments, return irqreturn_t directly]
>     Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
>     Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
>     Acked-by: Rob Herring <robh@kernel.org>
>     CC: Pawel Moll <pawel.moll@arm.com>
>     CC: Mark Rutland <mark.rutland@arm.com>
>     CC: Ian Campbell <ijc+devicetree@hellion.org.uk>
>     CC: Kumar Gala <galak@codeaurora.org>
> 
> diff --git a/Documentation/devicetree/bindings/pci/pci-keystone.txt b/Documentation/devicetree/bindings/pci/pci-keystone.txt
> index 54eae29..d08a4d5 100644
> --- a/Documentation/devicetree/bindings/pci/pci-keystone.txt
> +++ b/Documentation/devicetree/bindings/pci/pci-keystone.txt
> @@ -56,6 +56,7 @@ Optional properties:-
>  	phy-names: name of the Generic Keystine SerDes phy for PCI
>  	  - If boot loader already does PCI link establishment, then phys and
>  	    phy-names shouldn't be present.
> +	interrupts: platform interrupt for error interrupts.
>  
>  Designware DT Properties not applicable for Keystone PCI
>  
> diff --git a/drivers/pci/host/pci-keystone-dw.c b/drivers/pci/host/pci-keystone-dw.c
> index 6153853..4151509 100644
> --- a/drivers/pci/host/pci-keystone-dw.c
> +++ b/drivers/pci/host/pci-keystone-dw.c
> @@ -14,6 +14,7 @@
>  
>  #include <linux/irq.h>
>  #include <linux/irqdomain.h>
> +#include <linux/irqreturn.h>
>  #include <linux/module.h>
>  #include <linux/of.h>
>  #include <linux/of_pci.h>
> @@ -53,6 +54,21 @@
>  #define IRQ_STATUS			0x184
>  #define MSI_IRQ_OFFSET			4
>  
> +/* Error IRQ bits */
> +#define ERR_AER		BIT(5)	/* ECRC error */
> +#define ERR_AXI		BIT(4)	/* AXI tag lookup fatal error */
> +#define ERR_CORR	BIT(3)	/* Correctable error */
> +#define ERR_NONFATAL	BIT(2)	/* Non-fatal error */
> +#define ERR_FATAL	BIT(1)	/* Fatal error */
> +#define ERR_SYS		BIT(0)	/* System (fatal, non-fatal, or correctable) */
> +#define ERR_IRQ_ALL	(ERR_AER | ERR_AXI | ERR_CORR | \
> +			 ERR_NONFATAL | ERR_FATAL | ERR_SYS)
> +#define ERR_FATAL_IRQ	(ERR_FATAL | ERR_AXI)
> +#define ERR_IRQ_STATUS_RAW		0x1c0
> +#define ERR_IRQ_STATUS			0x1c4
> +#define ERR_IRQ_ENABLE_SET		0x1c8
> +#define ERR_IRQ_ENABLE_CLR		0x1cc
> +
>  /* Config space registers */
>  #define DEBUG0				0x728
>  
> @@ -243,6 +259,28 @@ void ks_dw_pcie_handle_legacy_irq(struct keystone_pcie *ks_pcie, int offset)
>  	writel(offset, ks_pcie->va_app_base + IRQ_EOI);
>  }
>  
> +void ks_dw_pcie_enable_error_irq(void __iomem *reg_base)
> +{
> +	writel(ERR_IRQ_ALL, reg_base + ERR_IRQ_ENABLE_SET);
> +}
> +
> +irqreturn_t ks_dw_pcie_handle_error_irq(struct device *dev,
> +					void __iomem *reg_base)
> +{
> +	u32 status;
> +
> +	status = readl(reg_base + ERR_IRQ_STATUS_RAW) & ERR_IRQ_ALL;
> +	if (!status)
> +		return IRQ_NONE;
> +
> +	if (status & ERR_FATAL_IRQ)
> +		dev_err(dev, "fatal error (status %#010x)\n", status);
> +
> +	/* Ack the IRQ; status bits are RW1C */
> +	writel(status, reg_base + ERR_IRQ_STATUS);
> +	return IRQ_HANDLED;
> +}
> +
>  static void ks_dw_pcie_ack_legacy_irq(struct irq_data *d)
>  {
>  }
> diff --git a/drivers/pci/host/pci-keystone.c b/drivers/pci/host/pci-keystone.c
> index b71f55b..6868918 100644
> --- a/drivers/pci/host/pci-keystone.c
> +++ b/drivers/pci/host/pci-keystone.c
> @@ -15,6 +15,7 @@
>  #include <linux/irqchip/chained_irq.h>
>  #include <linux/clk.h>
>  #include <linux/delay.h>
> +#include <linux/interrupt.h>
>  #include <linux/irqdomain.h>
>  #include <linux/module.h>
>  #include <linux/msi.h>
> @@ -226,6 +227,9 @@ static void ks_pcie_setup_interrupts(struct keystone_pcie *ks_pcie)
>  							 ks_pcie);
>  		}
>  	}
> +
> +	if (ks_pcie->error_irq > 0)
> +		ks_dw_pcie_enable_error_irq(ks_pcie->va_app_base);
>  }
>  
>  /*
> @@ -289,6 +293,14 @@ static struct pcie_host_ops keystone_pcie_host_ops = {
>  	.scan_bus = ks_dw_pcie_v3_65_scan_bus,
>  };
>  
> +static irqreturn_t pcie_err_irq_handler(int irq, void *priv)
> +{
> +	struct keystone_pcie *ks_pcie = priv;
> +
> +	return ks_dw_pcie_handle_error_irq(ks_pcie->pp.dev,
> +					   ks_pcie->va_app_base);
> +}
> +
>  static int __init ks_add_pcie_port(struct keystone_pcie *ks_pcie,
>  			 struct platform_device *pdev)
>  {
> @@ -309,6 +321,22 @@ static int __init ks_add_pcie_port(struct keystone_pcie *ks_pcie,
>  			return ret;
>  	}
>  
> +	/*
> +	 * Index 0 is the platform interrupt for error interrupt
> +	 * from RC.  This is optional.
> +	 */
> +	ks_pcie->error_irq = irq_of_parse_and_map(ks_pcie->np, 0);
> +	if (ks_pcie->error_irq <= 0)
> +		dev_info(&pdev->dev, "no error IRQ defined\n");
> +	else {
> +		if (request_irq(ks_pcie->error_irq, pcie_err_irq_handler,
> +				IRQF_SHARED, "pcie-error-irq", ks_pcie) < 0) {
> +			dev_err(&pdev->dev, "failed to request error IRQ %d\n",
> +				ks_pcie->error_irq);
> +			return ret;
> +		}
> +	}
> +
>  	pp->root_bus_nr = -1;
>  	pp->ops = &keystone_pcie_host_ops;
>  	ret = ks_dw_pcie_host_init(ks_pcie, ks_pcie->msi_intc_np);
> @@ -376,6 +404,7 @@ static int __init ks_pcie_probe(struct platform_device *pdev)
>  	devm_release_mem_region(dev, res->start, resource_size(res));
>  
>  	pp->dev = dev;
> +	ks_pcie->np = node;
>  	platform_set_drvdata(pdev, ks_pcie);
>  	ks_pcie->clk = devm_clk_get(dev, "pcie");
>  	if (IS_ERR(ks_pcie->clk)) {
> diff --git a/drivers/pci/host/pci-keystone.h b/drivers/pci/host/pci-keystone.h
> index f0944e8..a5b0cb2 100644
> --- a/drivers/pci/host/pci-keystone.h
> +++ b/drivers/pci/host/pci-keystone.h
> @@ -29,6 +29,9 @@ struct keystone_pcie {
>  	int			msi_host_irqs[MAX_MSI_HOST_IRQS];
>  	struct			device_node *msi_intc_np;
>  	struct irq_domain	*legacy_irq_domain;
> +	struct device_node	*np;
> +
> +	int error_irq;
>  
>  	/* Application register space */
>  	void __iomem		*va_app_base;
> @@ -42,6 +45,9 @@ phys_addr_t ks_dw_pcie_get_msi_addr(struct pcie_port *pp);
>  /* Keystone specific PCI controller APIs */
>  void ks_dw_pcie_enable_legacy_irqs(struct keystone_pcie *ks_pcie);
>  void ks_dw_pcie_handle_legacy_irq(struct keystone_pcie *ks_pcie, int offset);
> +void ks_dw_pcie_enable_error_irq(void __iomem *reg_base);
> +irqreturn_t ks_dw_pcie_handle_error_irq(struct device *dev,
> +					void __iomem *reg_base);
>  int  ks_dw_pcie_host_init(struct keystone_pcie *ks_pcie,
>  			struct device_node *msi_intc_np);
>  int ks_dw_pcie_wr_other_conf(struct pcie_port *pp, struct pci_bus *bus,
> 


-- 
Murali Karicheri
Linux Kernel, Keystone

WARNING: multiple messages have this Message-ID (diff)
From: Murali Karicheri <m-karicheri2@ti.com>
To: Bjorn Helgaas <helgaas@kernel.org>
Cc: bhelgaas@google.com, devicetree@vger.kernel.org,
	linux-kernel@vger.kernel.org, linux-pci@vger.kernel.org,
	linux-arm-kernel@lists.infradead.org
Subject: Re: [PATCH next v2 1/2] PCI: keystone: add pci error irq handler
Date: Wed, 20 Apr 2016 10:13:36 -0400	[thread overview]
Message-ID: <57178E90.5090205@ti.com> (raw)
In-Reply-To: <20160420010343.GI17863@localhost>

On 04/19/2016 09:03 PM, Bjorn Helgaas wrote:
> Hi Murali,
> 
> On Mon, Apr 11, 2016 at 10:50:30AM -0400, Murali Karicheri wrote:
>> Keystone PCI hardware generates error interrupts at RC using platform
>> irq instead of standard msi/legacy irq. Add a simple error handler that
>> logs the fatal interrupt status to the console.
>>
>> Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
>> Acked-by: Rob Herring <robh@kernel.org>
>> Cc: Rob Herring <robh+dt@kernel.org>
>> Cc: Pawel Moll <pawel.moll@arm.com>
>> Cc: Mark Rutland <mark.rutland@arm.com>
>> Cc: Ian Campbell <ijc+devicetree@hellion.org.uk>
>> Cc: Kumar Gala <galak@codeaurora.org>
>> Cc: Bjorn Helgaas <bhelgaas@google.com>
> 
> I applied this with minor changes to pci/host-keystone for v4.7.
> See below:
> 
>> +bool ks_dw_pcie_handle_error_irq(struct device *dev, void __iomem *reg_base)
>> +{
>> +	u32 status;
>> +	bool ret = false;
>> +
>> +	status = readl(reg_base + ERR_IRQ_STATUS_RAW) & ERR_IRQ_ALL;
>> +	if (status) {
>> +		/* The PCIESS interrupt status buts are "write 1 to clear" */
>> +		if (status & ERR_FATAL_IRQ)
>> +			dev_err(dev, "PCIE fatal error detected\n");
>> +
>> +		/* ack the irq event. */
>> +		writel(status, reg_base + ERR_IRQ_STATUS);
>> +		ret = true;
>> +	}
>> +	return ret;
>> +}
> 
> It seems pointless to me to return true/false here and then:
> 
>> +static irqreturn_t pcie_err_irq_handler(int irq, void *priv)
>> +{
>> +	struct keystone_pcie *ks_pcie = priv;
>> +
>> +	if (ks_dw_pcie_handle_error_irq(ks_pcie->pp.dev, ks_pcie->va_app_base))
>> +		return IRQ_HANDLED;
>> +
>> +	return IRQ_NONE;
>> +}
> 
> convert the true/false to IRQ_HANDLED/IRQ_NONE here.  So I changed
> ks_dw_pcie_handle_error_irq() to return IRQ_HANDLED/IRQ_NONE directly,
> resulting in the patch below.

I think that should be fine. Only reason, i did that way was that the pcie-designware
code is kept as a library of functions to configure the designware hardware common to
all platforms that has the designware core h/w and keep all of the platform specific info
such as IRQ handler to the platform glue driver such as pci-keystone.c. But that is not
true today as there is msi irq handler in this driver. So this is fine.

I will pull these from your designware pci/host-keystone and let you know if I see
any issues or send a diff patch to the list.

Once again, thanks for pulling this to v4.7.

Murali
> 
> This is applied to pci/host-keystone for v4.7.
> 
> 
> commit 7be92716c1aa40f5690e6a5f01fc0d385dd72303
> Author: Murali Karicheri <m-karicheri2@ti.com>
> Date:   Mon Apr 11 10:50:30 2016 -0400
> 
>     PCI: keystone: Add error IRQ handler
>     
>     Keystone PCI hardware generates error interrupts at RC using a platform IRQ
>     instead of a standard MSI or legacy IRQ.  Add a simple error handler that
>     logs the fatal interrupt status to the console.
>     
>     [bhelgaas: tidy comments, return irqreturn_t directly]
>     Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
>     Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
>     Acked-by: Rob Herring <robh@kernel.org>
>     CC: Pawel Moll <pawel.moll@arm.com>
>     CC: Mark Rutland <mark.rutland@arm.com>
>     CC: Ian Campbell <ijc+devicetree@hellion.org.uk>
>     CC: Kumar Gala <galak@codeaurora.org>
> 
> diff --git a/Documentation/devicetree/bindings/pci/pci-keystone.txt b/Documentation/devicetree/bindings/pci/pci-keystone.txt
> index 54eae29..d08a4d5 100644
> --- a/Documentation/devicetree/bindings/pci/pci-keystone.txt
> +++ b/Documentation/devicetree/bindings/pci/pci-keystone.txt
> @@ -56,6 +56,7 @@ Optional properties:-
>  	phy-names: name of the Generic Keystine SerDes phy for PCI
>  	  - If boot loader already does PCI link establishment, then phys and
>  	    phy-names shouldn't be present.
> +	interrupts: platform interrupt for error interrupts.
>  
>  Designware DT Properties not applicable for Keystone PCI
>  
> diff --git a/drivers/pci/host/pci-keystone-dw.c b/drivers/pci/host/pci-keystone-dw.c
> index 6153853..4151509 100644
> --- a/drivers/pci/host/pci-keystone-dw.c
> +++ b/drivers/pci/host/pci-keystone-dw.c
> @@ -14,6 +14,7 @@
>  
>  #include <linux/irq.h>
>  #include <linux/irqdomain.h>
> +#include <linux/irqreturn.h>
>  #include <linux/module.h>
>  #include <linux/of.h>
>  #include <linux/of_pci.h>
> @@ -53,6 +54,21 @@
>  #define IRQ_STATUS			0x184
>  #define MSI_IRQ_OFFSET			4
>  
> +/* Error IRQ bits */
> +#define ERR_AER		BIT(5)	/* ECRC error */
> +#define ERR_AXI		BIT(4)	/* AXI tag lookup fatal error */
> +#define ERR_CORR	BIT(3)	/* Correctable error */
> +#define ERR_NONFATAL	BIT(2)	/* Non-fatal error */
> +#define ERR_FATAL	BIT(1)	/* Fatal error */
> +#define ERR_SYS		BIT(0)	/* System (fatal, non-fatal, or correctable) */
> +#define ERR_IRQ_ALL	(ERR_AER | ERR_AXI | ERR_CORR | \
> +			 ERR_NONFATAL | ERR_FATAL | ERR_SYS)
> +#define ERR_FATAL_IRQ	(ERR_FATAL | ERR_AXI)
> +#define ERR_IRQ_STATUS_RAW		0x1c0
> +#define ERR_IRQ_STATUS			0x1c4
> +#define ERR_IRQ_ENABLE_SET		0x1c8
> +#define ERR_IRQ_ENABLE_CLR		0x1cc
> +
>  /* Config space registers */
>  #define DEBUG0				0x728
>  
> @@ -243,6 +259,28 @@ void ks_dw_pcie_handle_legacy_irq(struct keystone_pcie *ks_pcie, int offset)
>  	writel(offset, ks_pcie->va_app_base + IRQ_EOI);
>  }
>  
> +void ks_dw_pcie_enable_error_irq(void __iomem *reg_base)
> +{
> +	writel(ERR_IRQ_ALL, reg_base + ERR_IRQ_ENABLE_SET);
> +}
> +
> +irqreturn_t ks_dw_pcie_handle_error_irq(struct device *dev,
> +					void __iomem *reg_base)
> +{
> +	u32 status;
> +
> +	status = readl(reg_base + ERR_IRQ_STATUS_RAW) & ERR_IRQ_ALL;
> +	if (!status)
> +		return IRQ_NONE;
> +
> +	if (status & ERR_FATAL_IRQ)
> +		dev_err(dev, "fatal error (status %#010x)\n", status);
> +
> +	/* Ack the IRQ; status bits are RW1C */
> +	writel(status, reg_base + ERR_IRQ_STATUS);
> +	return IRQ_HANDLED;
> +}
> +
>  static void ks_dw_pcie_ack_legacy_irq(struct irq_data *d)
>  {
>  }
> diff --git a/drivers/pci/host/pci-keystone.c b/drivers/pci/host/pci-keystone.c
> index b71f55b..6868918 100644
> --- a/drivers/pci/host/pci-keystone.c
> +++ b/drivers/pci/host/pci-keystone.c
> @@ -15,6 +15,7 @@
>  #include <linux/irqchip/chained_irq.h>
>  #include <linux/clk.h>
>  #include <linux/delay.h>
> +#include <linux/interrupt.h>
>  #include <linux/irqdomain.h>
>  #include <linux/module.h>
>  #include <linux/msi.h>
> @@ -226,6 +227,9 @@ static void ks_pcie_setup_interrupts(struct keystone_pcie *ks_pcie)
>  							 ks_pcie);
>  		}
>  	}
> +
> +	if (ks_pcie->error_irq > 0)
> +		ks_dw_pcie_enable_error_irq(ks_pcie->va_app_base);
>  }
>  
>  /*
> @@ -289,6 +293,14 @@ static struct pcie_host_ops keystone_pcie_host_ops = {
>  	.scan_bus = ks_dw_pcie_v3_65_scan_bus,
>  };
>  
> +static irqreturn_t pcie_err_irq_handler(int irq, void *priv)
> +{
> +	struct keystone_pcie *ks_pcie = priv;
> +
> +	return ks_dw_pcie_handle_error_irq(ks_pcie->pp.dev,
> +					   ks_pcie->va_app_base);
> +}
> +
>  static int __init ks_add_pcie_port(struct keystone_pcie *ks_pcie,
>  			 struct platform_device *pdev)
>  {
> @@ -309,6 +321,22 @@ static int __init ks_add_pcie_port(struct keystone_pcie *ks_pcie,
>  			return ret;
>  	}
>  
> +	/*
> +	 * Index 0 is the platform interrupt for error interrupt
> +	 * from RC.  This is optional.
> +	 */
> +	ks_pcie->error_irq = irq_of_parse_and_map(ks_pcie->np, 0);
> +	if (ks_pcie->error_irq <= 0)
> +		dev_info(&pdev->dev, "no error IRQ defined\n");
> +	else {
> +		if (request_irq(ks_pcie->error_irq, pcie_err_irq_handler,
> +				IRQF_SHARED, "pcie-error-irq", ks_pcie) < 0) {
> +			dev_err(&pdev->dev, "failed to request error IRQ %d\n",
> +				ks_pcie->error_irq);
> +			return ret;
> +		}
> +	}
> +
>  	pp->root_bus_nr = -1;
>  	pp->ops = &keystone_pcie_host_ops;
>  	ret = ks_dw_pcie_host_init(ks_pcie, ks_pcie->msi_intc_np);
> @@ -376,6 +404,7 @@ static int __init ks_pcie_probe(struct platform_device *pdev)
>  	devm_release_mem_region(dev, res->start, resource_size(res));
>  
>  	pp->dev = dev;
> +	ks_pcie->np = node;
>  	platform_set_drvdata(pdev, ks_pcie);
>  	ks_pcie->clk = devm_clk_get(dev, "pcie");
>  	if (IS_ERR(ks_pcie->clk)) {
> diff --git a/drivers/pci/host/pci-keystone.h b/drivers/pci/host/pci-keystone.h
> index f0944e8..a5b0cb2 100644
> --- a/drivers/pci/host/pci-keystone.h
> +++ b/drivers/pci/host/pci-keystone.h
> @@ -29,6 +29,9 @@ struct keystone_pcie {
>  	int			msi_host_irqs[MAX_MSI_HOST_IRQS];
>  	struct			device_node *msi_intc_np;
>  	struct irq_domain	*legacy_irq_domain;
> +	struct device_node	*np;
> +
> +	int error_irq;
>  
>  	/* Application register space */
>  	void __iomem		*va_app_base;
> @@ -42,6 +45,9 @@ phys_addr_t ks_dw_pcie_get_msi_addr(struct pcie_port *pp);
>  /* Keystone specific PCI controller APIs */
>  void ks_dw_pcie_enable_legacy_irqs(struct keystone_pcie *ks_pcie);
>  void ks_dw_pcie_handle_legacy_irq(struct keystone_pcie *ks_pcie, int offset);
> +void ks_dw_pcie_enable_error_irq(void __iomem *reg_base);
> +irqreturn_t ks_dw_pcie_handle_error_irq(struct device *dev,
> +					void __iomem *reg_base);
>  int  ks_dw_pcie_host_init(struct keystone_pcie *ks_pcie,
>  			struct device_node *msi_intc_np);
>  int ks_dw_pcie_wr_other_conf(struct pcie_port *pp, struct pci_bus *bus,
> 


-- 
Murali Karicheri
Linux Kernel, Keystone

WARNING: multiple messages have this Message-ID (diff)
From: m-karicheri2@ti.com (Murali Karicheri)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH next v2 1/2] PCI: keystone: add pci error irq handler
Date: Wed, 20 Apr 2016 10:13:36 -0400	[thread overview]
Message-ID: <57178E90.5090205@ti.com> (raw)
In-Reply-To: <20160420010343.GI17863@localhost>

On 04/19/2016 09:03 PM, Bjorn Helgaas wrote:
> Hi Murali,
> 
> On Mon, Apr 11, 2016 at 10:50:30AM -0400, Murali Karicheri wrote:
>> Keystone PCI hardware generates error interrupts at RC using platform
>> irq instead of standard msi/legacy irq. Add a simple error handler that
>> logs the fatal interrupt status to the console.
>>
>> Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
>> Acked-by: Rob Herring <robh@kernel.org>
>> Cc: Rob Herring <robh+dt@kernel.org>
>> Cc: Pawel Moll <pawel.moll@arm.com>
>> Cc: Mark Rutland <mark.rutland@arm.com>
>> Cc: Ian Campbell <ijc+devicetree@hellion.org.uk>
>> Cc: Kumar Gala <galak@codeaurora.org>
>> Cc: Bjorn Helgaas <bhelgaas@google.com>
> 
> I applied this with minor changes to pci/host-keystone for v4.7.
> See below:
> 
>> +bool ks_dw_pcie_handle_error_irq(struct device *dev, void __iomem *reg_base)
>> +{
>> +	u32 status;
>> +	bool ret = false;
>> +
>> +	status = readl(reg_base + ERR_IRQ_STATUS_RAW) & ERR_IRQ_ALL;
>> +	if (status) {
>> +		/* The PCIESS interrupt status buts are "write 1 to clear" */
>> +		if (status & ERR_FATAL_IRQ)
>> +			dev_err(dev, "PCIE fatal error detected\n");
>> +
>> +		/* ack the irq event. */
>> +		writel(status, reg_base + ERR_IRQ_STATUS);
>> +		ret = true;
>> +	}
>> +	return ret;
>> +}
> 
> It seems pointless to me to return true/false here and then:
> 
>> +static irqreturn_t pcie_err_irq_handler(int irq, void *priv)
>> +{
>> +	struct keystone_pcie *ks_pcie = priv;
>> +
>> +	if (ks_dw_pcie_handle_error_irq(ks_pcie->pp.dev, ks_pcie->va_app_base))
>> +		return IRQ_HANDLED;
>> +
>> +	return IRQ_NONE;
>> +}
> 
> convert the true/false to IRQ_HANDLED/IRQ_NONE here.  So I changed
> ks_dw_pcie_handle_error_irq() to return IRQ_HANDLED/IRQ_NONE directly,
> resulting in the patch below.

I think that should be fine. Only reason, i did that way was that the pcie-designware
code is kept as a library of functions to configure the designware hardware common to
all platforms that has the designware core h/w and keep all of the platform specific info
such as IRQ handler to the platform glue driver such as pci-keystone.c. But that is not
true today as there is msi irq handler in this driver. So this is fine.

I will pull these from your designware pci/host-keystone and let you know if I see
any issues or send a diff patch to the list.

Once again, thanks for pulling this to v4.7.

Murali
> 
> This is applied to pci/host-keystone for v4.7.
> 
> 
> commit 7be92716c1aa40f5690e6a5f01fc0d385dd72303
> Author: Murali Karicheri <m-karicheri2@ti.com>
> Date:   Mon Apr 11 10:50:30 2016 -0400
> 
>     PCI: keystone: Add error IRQ handler
>     
>     Keystone PCI hardware generates error interrupts at RC using a platform IRQ
>     instead of a standard MSI or legacy IRQ.  Add a simple error handler that
>     logs the fatal interrupt status to the console.
>     
>     [bhelgaas: tidy comments, return irqreturn_t directly]
>     Signed-off-by: Murali Karicheri <m-karicheri2@ti.com>
>     Signed-off-by: Bjorn Helgaas <bhelgaas@google.com>
>     Acked-by: Rob Herring <robh@kernel.org>
>     CC: Pawel Moll <pawel.moll@arm.com>
>     CC: Mark Rutland <mark.rutland@arm.com>
>     CC: Ian Campbell <ijc+devicetree@hellion.org.uk>
>     CC: Kumar Gala <galak@codeaurora.org>
> 
> diff --git a/Documentation/devicetree/bindings/pci/pci-keystone.txt b/Documentation/devicetree/bindings/pci/pci-keystone.txt
> index 54eae29..d08a4d5 100644
> --- a/Documentation/devicetree/bindings/pci/pci-keystone.txt
> +++ b/Documentation/devicetree/bindings/pci/pci-keystone.txt
> @@ -56,6 +56,7 @@ Optional properties:-
>  	phy-names: name of the Generic Keystine SerDes phy for PCI
>  	  - If boot loader already does PCI link establishment, then phys and
>  	    phy-names shouldn't be present.
> +	interrupts: platform interrupt for error interrupts.
>  
>  Designware DT Properties not applicable for Keystone PCI
>  
> diff --git a/drivers/pci/host/pci-keystone-dw.c b/drivers/pci/host/pci-keystone-dw.c
> index 6153853..4151509 100644
> --- a/drivers/pci/host/pci-keystone-dw.c
> +++ b/drivers/pci/host/pci-keystone-dw.c
> @@ -14,6 +14,7 @@
>  
>  #include <linux/irq.h>
>  #include <linux/irqdomain.h>
> +#include <linux/irqreturn.h>
>  #include <linux/module.h>
>  #include <linux/of.h>
>  #include <linux/of_pci.h>
> @@ -53,6 +54,21 @@
>  #define IRQ_STATUS			0x184
>  #define MSI_IRQ_OFFSET			4
>  
> +/* Error IRQ bits */
> +#define ERR_AER		BIT(5)	/* ECRC error */
> +#define ERR_AXI		BIT(4)	/* AXI tag lookup fatal error */
> +#define ERR_CORR	BIT(3)	/* Correctable error */
> +#define ERR_NONFATAL	BIT(2)	/* Non-fatal error */
> +#define ERR_FATAL	BIT(1)	/* Fatal error */
> +#define ERR_SYS		BIT(0)	/* System (fatal, non-fatal, or correctable) */
> +#define ERR_IRQ_ALL	(ERR_AER | ERR_AXI | ERR_CORR | \
> +			 ERR_NONFATAL | ERR_FATAL | ERR_SYS)
> +#define ERR_FATAL_IRQ	(ERR_FATAL | ERR_AXI)
> +#define ERR_IRQ_STATUS_RAW		0x1c0
> +#define ERR_IRQ_STATUS			0x1c4
> +#define ERR_IRQ_ENABLE_SET		0x1c8
> +#define ERR_IRQ_ENABLE_CLR		0x1cc
> +
>  /* Config space registers */
>  #define DEBUG0				0x728
>  
> @@ -243,6 +259,28 @@ void ks_dw_pcie_handle_legacy_irq(struct keystone_pcie *ks_pcie, int offset)
>  	writel(offset, ks_pcie->va_app_base + IRQ_EOI);
>  }
>  
> +void ks_dw_pcie_enable_error_irq(void __iomem *reg_base)
> +{
> +	writel(ERR_IRQ_ALL, reg_base + ERR_IRQ_ENABLE_SET);
> +}
> +
> +irqreturn_t ks_dw_pcie_handle_error_irq(struct device *dev,
> +					void __iomem *reg_base)
> +{
> +	u32 status;
> +
> +	status = readl(reg_base + ERR_IRQ_STATUS_RAW) & ERR_IRQ_ALL;
> +	if (!status)
> +		return IRQ_NONE;
> +
> +	if (status & ERR_FATAL_IRQ)
> +		dev_err(dev, "fatal error (status %#010x)\n", status);
> +
> +	/* Ack the IRQ; status bits are RW1C */
> +	writel(status, reg_base + ERR_IRQ_STATUS);
> +	return IRQ_HANDLED;
> +}
> +
>  static void ks_dw_pcie_ack_legacy_irq(struct irq_data *d)
>  {
>  }
> diff --git a/drivers/pci/host/pci-keystone.c b/drivers/pci/host/pci-keystone.c
> index b71f55b..6868918 100644
> --- a/drivers/pci/host/pci-keystone.c
> +++ b/drivers/pci/host/pci-keystone.c
> @@ -15,6 +15,7 @@
>  #include <linux/irqchip/chained_irq.h>
>  #include <linux/clk.h>
>  #include <linux/delay.h>
> +#include <linux/interrupt.h>
>  #include <linux/irqdomain.h>
>  #include <linux/module.h>
>  #include <linux/msi.h>
> @@ -226,6 +227,9 @@ static void ks_pcie_setup_interrupts(struct keystone_pcie *ks_pcie)
>  							 ks_pcie);
>  		}
>  	}
> +
> +	if (ks_pcie->error_irq > 0)
> +		ks_dw_pcie_enable_error_irq(ks_pcie->va_app_base);
>  }
>  
>  /*
> @@ -289,6 +293,14 @@ static struct pcie_host_ops keystone_pcie_host_ops = {
>  	.scan_bus = ks_dw_pcie_v3_65_scan_bus,
>  };
>  
> +static irqreturn_t pcie_err_irq_handler(int irq, void *priv)
> +{
> +	struct keystone_pcie *ks_pcie = priv;
> +
> +	return ks_dw_pcie_handle_error_irq(ks_pcie->pp.dev,
> +					   ks_pcie->va_app_base);
> +}
> +
>  static int __init ks_add_pcie_port(struct keystone_pcie *ks_pcie,
>  			 struct platform_device *pdev)
>  {
> @@ -309,6 +321,22 @@ static int __init ks_add_pcie_port(struct keystone_pcie *ks_pcie,
>  			return ret;
>  	}
>  
> +	/*
> +	 * Index 0 is the platform interrupt for error interrupt
> +	 * from RC.  This is optional.
> +	 */
> +	ks_pcie->error_irq = irq_of_parse_and_map(ks_pcie->np, 0);
> +	if (ks_pcie->error_irq <= 0)
> +		dev_info(&pdev->dev, "no error IRQ defined\n");
> +	else {
> +		if (request_irq(ks_pcie->error_irq, pcie_err_irq_handler,
> +				IRQF_SHARED, "pcie-error-irq", ks_pcie) < 0) {
> +			dev_err(&pdev->dev, "failed to request error IRQ %d\n",
> +				ks_pcie->error_irq);
> +			return ret;
> +		}
> +	}
> +
>  	pp->root_bus_nr = -1;
>  	pp->ops = &keystone_pcie_host_ops;
>  	ret = ks_dw_pcie_host_init(ks_pcie, ks_pcie->msi_intc_np);
> @@ -376,6 +404,7 @@ static int __init ks_pcie_probe(struct platform_device *pdev)
>  	devm_release_mem_region(dev, res->start, resource_size(res));
>  
>  	pp->dev = dev;
> +	ks_pcie->np = node;
>  	platform_set_drvdata(pdev, ks_pcie);
>  	ks_pcie->clk = devm_clk_get(dev, "pcie");
>  	if (IS_ERR(ks_pcie->clk)) {
> diff --git a/drivers/pci/host/pci-keystone.h b/drivers/pci/host/pci-keystone.h
> index f0944e8..a5b0cb2 100644
> --- a/drivers/pci/host/pci-keystone.h
> +++ b/drivers/pci/host/pci-keystone.h
> @@ -29,6 +29,9 @@ struct keystone_pcie {
>  	int			msi_host_irqs[MAX_MSI_HOST_IRQS];
>  	struct			device_node *msi_intc_np;
>  	struct irq_domain	*legacy_irq_domain;
> +	struct device_node	*np;
> +
> +	int error_irq;
>  
>  	/* Application register space */
>  	void __iomem		*va_app_base;
> @@ -42,6 +45,9 @@ phys_addr_t ks_dw_pcie_get_msi_addr(struct pcie_port *pp);
>  /* Keystone specific PCI controller APIs */
>  void ks_dw_pcie_enable_legacy_irqs(struct keystone_pcie *ks_pcie);
>  void ks_dw_pcie_handle_legacy_irq(struct keystone_pcie *ks_pcie, int offset);
> +void ks_dw_pcie_enable_error_irq(void __iomem *reg_base);
> +irqreturn_t ks_dw_pcie_handle_error_irq(struct device *dev,
> +					void __iomem *reg_base);
>  int  ks_dw_pcie_host_init(struct keystone_pcie *ks_pcie,
>  			struct device_node *msi_intc_np);
>  int ks_dw_pcie_wr_other_conf(struct pcie_port *pp, struct pci_bus *bus,
> 


-- 
Murali Karicheri
Linux Kernel, Keystone

  reply	other threads:[~2016-04-20 14:14 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-04-11 14:50 [PATCH next v2 1/2] PCI: keystone: add pci error irq handler Murali Karicheri
2016-04-11 14:50 ` Murali Karicheri
2016-04-11 14:50 ` Murali Karicheri
2016-04-11 14:50 ` [PATCH next v2 2/2] PCI: keystone: remove unnecessary goto statement Murali Karicheri
2016-04-11 14:50   ` Murali Karicheri
2016-04-11 14:50   ` Murali Karicheri
2016-04-20  1:06   ` Bjorn Helgaas
2016-04-20  1:06     ` Bjorn Helgaas
2016-04-20 14:29     ` Murali Karicheri
2016-04-20 14:29       ` Murali Karicheri
2016-04-20 14:29       ` Murali Karicheri
2016-04-20  1:03 ` [PATCH next v2 1/2] PCI: keystone: add pci error irq handler Bjorn Helgaas
2016-04-20  1:03   ` Bjorn Helgaas
2016-04-20  1:03   ` Bjorn Helgaas
2016-04-20 14:13   ` Murali Karicheri [this message]
2016-04-20 14:13     ` Murali Karicheri
2016-04-20 14:13     ` Murali Karicheri

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=57178E90.5090205@ti.com \
    --to=m-karicheri2@ti.com \
    --cc=bhelgaas@google.com \
    --cc=devicetree@vger.kernel.org \
    --cc=helgaas@kernel.org \
    --cc=linux-arm-kernel@lists.infradead.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-pci@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.