| Message ID | 20211217001238.16298-1-prabhakar.mahadev-lad.rj@bp.renesas.com |
|---|---|
| State | New |
| Headers | show |
| Series | ata: pata_of_platform: Use of_irq_to_resource() to populate IRQ resource | expand |
On 12/17/21 09:12, Lad Prabhakar wrote: > platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static > allocation of IRQ resources in DT core code, this causes an issue > when using hierarchical interrupt domains using "interrupts" property > in the node as this bypassed the hierarchical setup and messed up the s/bypassed/bypasses s/messed/messes > irq chaining. > > In preparation for removal of static setup of IRQ resource from DT core > code use of_irq_to_resource(). > > Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> > --- > Hi, > > Dropping usage of platform_get_resource() was agreed based on > the discussion [0]. > > [0] https://patchwork.kernel.org/project/linux-renesas-soc/ > patch/20211209001056.29774-1-prabhakar.mahadev-lad.rj@bp.renesas.com/ > > Cheers, > Prabhakar > --- > drivers/ata/pata_of_platform.c | 11 ++++++++--- > 1 file changed, 8 insertions(+), 3 deletions(-) > > diff --git a/drivers/ata/pata_of_platform.c b/drivers/ata/pata_of_platform.c > index 35aa158fc976..557f349eb533 100644 > --- a/drivers/ata/pata_of_platform.c > +++ b/drivers/ata/pata_of_platform.c > @@ -9,6 +9,7 @@ > #include <linux/kernel.h> > #include <linux/module.h> > #include <linux/of_address.h> > +#include <linux/of_irq.h> > #include <linux/platform_device.h> > #include <linux/ata_platform.h> > #include <linux/libata.h> > @@ -25,11 +26,12 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > struct device_node *dn = ofdev->dev.of_node; > struct resource io_res; > struct resource ctl_res; > - struct resource *irq_res; > + struct resource irq_res; > unsigned int reg_shift = 0; > int pio_mode = 0; > int pio_mask; > bool use16bit; > + int irq; > > ret = of_address_to_resource(dn, 0, &io_res); > if (ret) { > @@ -45,7 +47,9 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > return -EINVAL; > } > > - irq_res = platform_get_resource(ofdev, IORESOURCE_IRQ, 0); > + irq = of_irq_to_resource(dn, 0, &irq_res); > + if (irq <= 0 && irq != -ENXIO) > + return irq ? irq : -ENXIO; Why are you making an exception for ENXIO ? I suspect this is to cover the case "there is no IRQ for this node", but then how does this differentiate from a real error case ? > > of_property_read_u32(dn, "reg-shift", ®_shift); > > @@ -63,7 +67,8 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > pio_mask = 1 << pio_mode; > pio_mask |= (1 << pio_mode) - 1; > > - return __pata_platform_probe(&ofdev->dev, &io_res, &ctl_res, irq_res, > + return __pata_platform_probe(&ofdev->dev, &io_res, &ctl_res, > + irq > 0 ? &irq_res : NULL, > reg_shift, pio_mask, &pata_platform_sht, > use16bit); > } >
On Thu, Dec 16, 2021 at 6:12 PM Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> wrote: > > platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static > allocation of IRQ resources in DT core code, this causes an issue > when using hierarchical interrupt domains using "interrupts" property > in the node as this bypassed the hierarchical setup and messed up the > irq chaining. > > In preparation for removal of static setup of IRQ resource from DT core > code use of_irq_to_resource(). > > Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> > --- > Hi, > > Dropping usage of platform_get_resource() was agreed based on > the discussion [0]. > > [0] https://patchwork.kernel.org/project/linux-renesas-soc/ > patch/20211209001056.29774-1-prabhakar.mahadev-lad.rj@bp.renesas.com/ > > Cheers, > Prabhakar > --- > drivers/ata/pata_of_platform.c | 11 ++++++++--- > 1 file changed, 8 insertions(+), 3 deletions(-) > > diff --git a/drivers/ata/pata_of_platform.c b/drivers/ata/pata_of_platform.c > index 35aa158fc976..557f349eb533 100644 > --- a/drivers/ata/pata_of_platform.c > +++ b/drivers/ata/pata_of_platform.c > @@ -9,6 +9,7 @@ > #include <linux/kernel.h> > #include <linux/module.h> > #include <linux/of_address.h> > +#include <linux/of_irq.h> > #include <linux/platform_device.h> > #include <linux/ata_platform.h> > #include <linux/libata.h> > @@ -25,11 +26,12 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > struct device_node *dn = ofdev->dev.of_node; > struct resource io_res; > struct resource ctl_res; > - struct resource *irq_res; > + struct resource irq_res; > unsigned int reg_shift = 0; > int pio_mode = 0; > int pio_mask; > bool use16bit; > + int irq; > > ret = of_address_to_resource(dn, 0, &io_res); > if (ret) { > @@ -45,7 +47,9 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > return -EINVAL; > } > > - irq_res = platform_get_resource(ofdev, IORESOURCE_IRQ, 0); > + irq = of_irq_to_resource(dn, 0, &irq_res); No, platform_get_irq() please. I really want to get rid of the 4 users of of_irq_to_resource() in the tree, not add more. Yeah, that means __pata_platform_probe() needs to change, but it's all of 2 callers. Really, I think pata_platform.c and pata_of_platform.c should be merged. I'm guessing the separation traces back to DT using a different bus type. Rob
Hi Rob, On Fri, Dec 17, 2021 at 12:39 AM Rob Herring <robh+dt@kernel.org> wrote: > > On Thu, Dec 16, 2021 at 6:12 PM Lad Prabhakar > <prabhakar.mahadev-lad.rj@bp.renesas.com> wrote: > > > > platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static > > allocation of IRQ resources in DT core code, this causes an issue > > when using hierarchical interrupt domains using "interrupts" property > > in the node as this bypassed the hierarchical setup and messed up the > > irq chaining. > > > > In preparation for removal of static setup of IRQ resource from DT core > > code use of_irq_to_resource(). > > > > Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> > > --- > > Hi, > > > > Dropping usage of platform_get_resource() was agreed based on > > the discussion [0]. > > > > [0] https://patchwork.kernel.org/project/linux-renesas-soc/ > > patch/20211209001056.29774-1-prabhakar.mahadev-lad.rj@bp.renesas.com/ > > > > Cheers, > > Prabhakar > > --- > > drivers/ata/pata_of_platform.c | 11 ++++++++--- > > 1 file changed, 8 insertions(+), 3 deletions(-) > > > > diff --git a/drivers/ata/pata_of_platform.c b/drivers/ata/pata_of_platform.c > > index 35aa158fc976..557f349eb533 100644 > > --- a/drivers/ata/pata_of_platform.c > > +++ b/drivers/ata/pata_of_platform.c > > @@ -9,6 +9,7 @@ > > #include <linux/kernel.h> > > #include <linux/module.h> > > #include <linux/of_address.h> > > +#include <linux/of_irq.h> > > #include <linux/platform_device.h> > > #include <linux/ata_platform.h> > > #include <linux/libata.h> > > @@ -25,11 +26,12 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > > struct device_node *dn = ofdev->dev.of_node; > > struct resource io_res; > > struct resource ctl_res; > > - struct resource *irq_res; > > + struct resource irq_res; > > unsigned int reg_shift = 0; > > int pio_mode = 0; > > int pio_mask; > > bool use16bit; > > + int irq; > > > > ret = of_address_to_resource(dn, 0, &io_res); > > if (ret) { > > @@ -45,7 +47,9 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > > return -EINVAL; > > } > > > > - irq_res = platform_get_resource(ofdev, IORESOURCE_IRQ, 0); > > + irq = of_irq_to_resource(dn, 0, &irq_res); > > No, platform_get_irq() please. I really want to get rid of the 4 users > of of_irq_to_resource() in the tree, not add more. > Ouch I wasn't aware of it. > Yeah, that means __pata_platform_probe() needs to change, but it's all > of 2 callers. Really, I think pata_platform.c and pata_of_platform.c > should be merged. I'm guessing the separation traces back to DT using > a different bus type. > Or we could fill in the IRQ resource by using platform_get_irq() without touching __pata_platform_probe()? Cheers, Prabhakar > Rob
Hi Damien, Thank you for the review. On Fri, Dec 17, 2021 at 12:23 AM Damien Le Moal <damien.lemoal@opensource.wdc.com> wrote: > > On 12/17/21 09:12, Lad Prabhakar wrote: > > platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static > > allocation of IRQ resources in DT core code, this causes an issue > > when using hierarchical interrupt domains using "interrupts" property > > in the node as this bypassed the hierarchical setup and messed up the > > s/bypassed/bypasses > s/messed/messes > Ok will update this. > > irq chaining. > > > > In preparation for removal of static setup of IRQ resource from DT core > > code use of_irq_to_resource(). > > > > Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> > > --- > > Hi, > > > > Dropping usage of platform_get_resource() was agreed based on > > the discussion [0]. > > > > [0] https://patchwork.kernel.org/project/linux-renesas-soc/ > > patch/20211209001056.29774-1-prabhakar.mahadev-lad.rj@bp.renesas.com/ > > > > Cheers, > > Prabhakar > > --- > > drivers/ata/pata_of_platform.c | 11 ++++++++--- > > 1 file changed, 8 insertions(+), 3 deletions(-) > > > > diff --git a/drivers/ata/pata_of_platform.c b/drivers/ata/pata_of_platform.c > > index 35aa158fc976..557f349eb533 100644 > > --- a/drivers/ata/pata_of_platform.c > > +++ b/drivers/ata/pata_of_platform.c > > @@ -9,6 +9,7 @@ > > #include <linux/kernel.h> > > #include <linux/module.h> > > #include <linux/of_address.h> > > +#include <linux/of_irq.h> > > #include <linux/platform_device.h> > > #include <linux/ata_platform.h> > > #include <linux/libata.h> > > @@ -25,11 +26,12 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > > struct device_node *dn = ofdev->dev.of_node; > > struct resource io_res; > > struct resource ctl_res; > > - struct resource *irq_res; > > + struct resource irq_res; > > unsigned int reg_shift = 0; > > int pio_mode = 0; > > int pio_mask; > > bool use16bit; > > + int irq; > > > > ret = of_address_to_resource(dn, 0, &io_res); > > if (ret) { > > @@ -45,7 +47,9 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > > return -EINVAL; > > } > > > > - irq_res = platform_get_resource(ofdev, IORESOURCE_IRQ, 0); > > + irq = of_irq_to_resource(dn, 0, &irq_res); > > + if (irq <= 0 && irq != -ENXIO) > > + return irq ? irq : -ENXIO; > > Why are you making an exception for ENXIO ? I suspect this is to cover > the case "there is no IRQ for this node", but then how does this > differentiate from a real error case ? > -ENXIO to handle no interrupt case. In case of actual error case we fallback to polling mode.. > > > > of_property_read_u32(dn, "reg-shift", ®_shift); > > > > @@ -63,7 +67,8 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > > pio_mask = 1 << pio_mode; > > pio_mask |= (1 << pio_mode) - 1; > > > > - return __pata_platform_probe(&ofdev->dev, &io_res, &ctl_res, irq_res, > > + return __pata_platform_probe(&ofdev->dev, &io_res, &ctl_res, > > + irq > 0 ? &irq_res : NULL, .. here is check for irq number again. Cheers, Prabhakar > > reg_shift, pio_mask, &pata_platform_sht, > > use16bit); > > } > > > > > -- > Damien Le Moal > Western Digital Research
Hello! On 17.12.2021 3:12, Lad Prabhakar wrote: > platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static > allocation of IRQ resources in DT core code, this causes an issue > when using hierarchical interrupt domains using "interrupts" property > in the node as this bypassed the hierarchical setup and messed up the > irq chaining. > > In preparation for removal of static setup of IRQ resource from DT core > code use of_irq_to_resource(). > > Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> > --- > Hi, > > Dropping usage of platform_get_resource() was agreed based on > the discussion [0]. > > [0] https://patchwork.kernel.org/project/linux-renesas-soc/ > patch/20211209001056.29774-1-prabhakar.mahadev-lad.rj@bp.renesas.com/ > > Cheers, > Prabhakar > --- > drivers/ata/pata_of_platform.c | 11 ++++++++--- > 1 file changed, 8 insertions(+), 3 deletions(-) > > diff --git a/drivers/ata/pata_of_platform.c b/drivers/ata/pata_of_platform.c > index 35aa158fc976..557f349eb533 100644 > --- a/drivers/ata/pata_of_platform.c > +++ b/drivers/ata/pata_of_platform.c [...] > @@ -45,7 +47,9 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > return -EINVAL; > } > > - irq_res = platform_get_resource(ofdev, IORESOURCE_IRQ, 0); > + irq = of_irq_to_resource(dn, 0, &irq_res); Why not use platform_get_irq_optional()? > + if (irq <= 0 && irq != -ENXIO) > + return irq ? irq : -ENXIO; > > of_property_read_u32(dn, "reg-shift", ®_shift); > [...] MBR, Sergey
Hi Sergey, On Fri, Dec 17, 2021 at 9:33 AM Sergey Shtylyov <s.shtylyov@omp.ru> wrote: > > Hello! > > On 17.12.2021 3:12, Lad Prabhakar wrote: > > > platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static > > allocation of IRQ resources in DT core code, this causes an issue > > when using hierarchical interrupt domains using "interrupts" property > > in the node as this bypassed the hierarchical setup and messed up the > > irq chaining. > > > > In preparation for removal of static setup of IRQ resource from DT core > > code use of_irq_to_resource(). > > > > Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> > > --- > > Hi, > > > > Dropping usage of platform_get_resource() was agreed based on > > the discussion [0]. > > > > [0] https://patchwork.kernel.org/project/linux-renesas-soc/ > > patch/20211209001056.29774-1-prabhakar.mahadev-lad.rj@bp.renesas.com/ > > > > Cheers, > > Prabhakar > > --- > > drivers/ata/pata_of_platform.c | 11 ++++++++--- > > 1 file changed, 8 insertions(+), 3 deletions(-) > > > > diff --git a/drivers/ata/pata_of_platform.c b/drivers/ata/pata_of_platform.c > > index 35aa158fc976..557f349eb533 100644 > > --- a/drivers/ata/pata_of_platform.c > > +++ b/drivers/ata/pata_of_platform.c > [...] > > @@ -45,7 +47,9 @@ static int pata_of_platform_probe(struct platform_device *ofdev) > > return -EINVAL; > > } > > > > - irq_res = platform_get_resource(ofdev, IORESOURCE_IRQ, 0); > > + irq = of_irq_to_resource(dn, 0, &irq_res); > > Why not use platform_get_irq_optional()? > Agreed. Cheers, Prabhakar > > + if (irq <= 0 && irq != -ENXIO) > > + return irq ? irq : -ENXIO; > > > > of_property_read_u32(dn, "reg-shift", ®_shift); > > > [...] > > MBR, Sergey
diff --git a/drivers/ata/pata_of_platform.c b/drivers/ata/pata_of_platform.c index 35aa158fc976..557f349eb533 100644 --- a/drivers/ata/pata_of_platform.c +++ b/drivers/ata/pata_of_platform.c @@ -9,6 +9,7 @@ #include <linux/kernel.h> #include <linux/module.h> #include <linux/of_address.h> +#include <linux/of_irq.h> #include <linux/platform_device.h> #include <linux/ata_platform.h> #include <linux/libata.h> @@ -25,11 +26,12 @@ static int pata_of_platform_probe(struct platform_device *ofdev) struct device_node *dn = ofdev->dev.of_node; struct resource io_res; struct resource ctl_res; - struct resource *irq_res; + struct resource irq_res; unsigned int reg_shift = 0; int pio_mode = 0; int pio_mask; bool use16bit; + int irq; ret = of_address_to_resource(dn, 0, &io_res); if (ret) { @@ -45,7 +47,9 @@ static int pata_of_platform_probe(struct platform_device *ofdev) return -EINVAL; } - irq_res = platform_get_resource(ofdev, IORESOURCE_IRQ, 0); + irq = of_irq_to_resource(dn, 0, &irq_res); + if (irq <= 0 && irq != -ENXIO) + return irq ? irq : -ENXIO; of_property_read_u32(dn, "reg-shift", ®_shift); @@ -63,7 +67,8 @@ static int pata_of_platform_probe(struct platform_device *ofdev) pio_mask = 1 << pio_mode; pio_mask |= (1 << pio_mode) - 1; - return __pata_platform_probe(&ofdev->dev, &io_res, &ctl_res, irq_res, + return __pata_platform_probe(&ofdev->dev, &io_res, &ctl_res, + irq > 0 ? &irq_res : NULL, reg_shift, pio_mask, &pata_platform_sht, use16bit); }
platform_get_resource(pdev, IORESOURCE_IRQ, ..) relies on static allocation of IRQ resources in DT core code, this causes an issue when using hierarchical interrupt domains using "interrupts" property in the node as this bypassed the hierarchical setup and messed up the irq chaining. In preparation for removal of static setup of IRQ resource from DT core code use of_irq_to_resource(). Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@bp.renesas.com> --- Hi, Dropping usage of platform_get_resource() was agreed based on the discussion [0]. [0] https://patchwork.kernel.org/project/linux-renesas-soc/ patch/20211209001056.29774-1-prabhakar.mahadev-lad.rj@bp.renesas.com/ Cheers, Prabhakar --- drivers/ata/pata_of_platform.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-)