[PATCH v1 1/1] PCI: keystone: Use device_get_match_data()

[PATCH v1 1/1] PCI: keystone: Use device_get_match_data()

[Andy Shevchenko]

Instead of manipulations with OF APIs, use device_get_match_data().

While at it, drop of_match_ptr() completely and make compiler happy,
otherwise it complains:

  pci-keystone.c:1069:34: warning: ‘ks_pcie_of_match’ defined but not used [-Wunused-const-variable=]

Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
---
 drivers/pci/controller/dwc/pci-keystone.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)

diff --git a/drivers/pci/controller/dwc/pci-keystone.c b/drivers/pci/controller/dwc/pci-keystone.c
index bde3b2824e89..f36ea618a248 100644
--- a/drivers/pci/controller/dwc/pci-keystone.c
+++ b/drivers/pci/controller/dwc/pci-keystone.c
@@ -24,6 +24,7 @@
 #include <linux/of_pci.h>
 #include <linux/phy/phy.h>
 #include <linux/platform_device.h>
+#include <linux/property.h>
 #include <linux/regmap.h>
 #include <linux/resource.h>
 #include <linux/signal.h>
@@ -1091,7 +1092,6 @@ static int __init ks_pcie_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct device_node *np = dev->of_node;
 	const struct ks_pcie_of_data *data;
-	const struct of_device_id *match;
 	enum dw_pcie_device_mode mode;
 	struct dw_pcie *pci;
 	struct keystone_pcie *ks_pcie;
@@ -1108,8 +1108,7 @@ static int __init ks_pcie_probe(struct platform_device *pdev)
 	int irq;
 	int i;
 
-	match = of_match_device(of_match_ptr(ks_pcie_of_match), dev);
-	data = (struct ks_pcie_of_data *)match->data;
+	data = device_get_match_data(dev);
 	if (!data)
 		return -EINVAL;
 
@@ -1309,7 +1308,7 @@ static struct platform_driver ks_pcie_driver __refdata = {
 	.remove = __exit_p(ks_pcie_remove),
 	.driver = {
 		.name	= "keystone-pcie",
-		.of_match_table = of_match_ptr(ks_pcie_of_match),
+		.of_match_table = ks_pcie_of_match,
 	},
 };
 builtin_platform_driver(ks_pcie_driver);
-- 
2.30.2

Re: [PATCH v1 1/1] PCI: keystone: Use device_get_match_data()

[Bjorn Helgaas]

On Wed, Jul 28, 2021 at 01:55:58PM +0300, Andy Shevchenko wrote:
> Instead of manipulations with OF APIs, use device_get_match_data().
> 
> While at it, drop of_match_ptr() completely and make compiler happy,
> otherwise it complains:
> 
>   pci-keystone.c:1069:34: warning: ‘ks_pcie_of_match’ defined but not used [-Wunused-const-variable=]

These are two separate things and I'd prefer two separate patches.

I have a to-do item on my list to replace of_match_device(), as you
did here.  I originally suggested replacing with
device_get_match_data(), but I think Rob prefers
of_device_get_match_data() because there's really no benefit to the
extra indirection of device_get_match_data().  These are not drivers
that may potentially be used with either ACPI or OF; they're just OF.

Either way, I'd like to see a patch that does this for all drivers in
drivers/pci/controller/ at the same time so they get slightly more
consistent.

Same for the .of_match_table update; a good change that I'd like to
apply universally.  It looks like pcie-spear13xx.c, pcie-armada8k.c,
pci-ftpci100.c, pci-v3-semi.c, pci-xgene.c, pcie-iproc-platform.c also
have the same issue.

> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
> ---
>  drivers/pci/controller/dwc/pci-keystone.c | 7 +++----
>  1 file changed, 3 insertions(+), 4 deletions(-)
> 
> diff --git a/drivers/pci/controller/dwc/pci-keystone.c b/drivers/pci/controller/dwc/pci-keystone.c
> index bde3b2824e89..f36ea618a248 100644
> --- a/drivers/pci/controller/dwc/pci-keystone.c
> +++ b/drivers/pci/controller/dwc/pci-keystone.c
> @@ -24,6 +24,7 @@
>  #include <linux/of_pci.h>
>  #include <linux/phy/phy.h>
>  #include <linux/platform_device.h>
> +#include <linux/property.h>
>  #include <linux/regmap.h>
>  #include <linux/resource.h>
>  #include <linux/signal.h>
> @@ -1091,7 +1092,6 @@ static int __init ks_pcie_probe(struct platform_device *pdev)
>  	struct device *dev = &pdev->dev;
>  	struct device_node *np = dev->of_node;
>  	const struct ks_pcie_of_data *data;
> -	const struct of_device_id *match;
>  	enum dw_pcie_device_mode mode;
>  	struct dw_pcie *pci;
>  	struct keystone_pcie *ks_pcie;
> @@ -1108,8 +1108,7 @@ static int __init ks_pcie_probe(struct platform_device *pdev)
>  	int irq;
>  	int i;
>  
> -	match = of_match_device(of_match_ptr(ks_pcie_of_match), dev);
> -	data = (struct ks_pcie_of_data *)match->data;
> +	data = device_get_match_data(dev);
>  	if (!data)
>  		return -EINVAL;
>  
> @@ -1309,7 +1308,7 @@ static struct platform_driver ks_pcie_driver __refdata = {
>  	.remove = __exit_p(ks_pcie_remove),
>  	.driver = {
>  		.name	= "keystone-pcie",
> -		.of_match_table = of_match_ptr(ks_pcie_of_match),
> +		.of_match_table = ks_pcie_of_match,
>  	},
>  };
>  builtin_platform_driver(ks_pcie_driver);
> -- 
> 2.30.2
> 

Re: [PATCH v1 1/1] PCI: keystone: Use device_get_match_data()

[Andy Shevchenko]

On Wed, Jul 28, 2021 at 6:51 PM Bjorn Helgaas <helgaas@kernel.org> wrote:
> On Wed, Jul 28, 2021 at 01:55:58PM +0300, Andy Shevchenko wrote:
> > Instead of manipulations with OF APIs, use device_get_match_data().
> >
> > While at it, drop of_match_ptr() completely and make compiler happy,
> > otherwise it complains:
> >
> >   pci-keystone.c:1069:34: warning: ‘ks_pcie_of_match’ defined but not used [-Wunused-const-variable=]
>
> These are two separate things and I'd prefer two separate patches.
>
> I have a to-do item on my list to replace of_match_device(), as you
> did here.  I originally suggested replacing with
> device_get_match_data(), but I think Rob prefers
> of_device_get_match_data() because there's really no benefit to the
> extra indirection of device_get_match_data().  These are not drivers
> that may potentially be used with either ACPI or OF; they're just OF.
>
> Either way, I'd like to see a patch that does this for all drivers in
> drivers/pci/controller/ at the same time so they get slightly more
> consistent.
>
> Same for the .of_match_table update; a good change that I'd like to
> apply universally.  It looks like pcie-spear13xx.c, pcie-armada8k.c,
> pci-ftpci100.c, pci-v3-semi.c, pci-xgene.c, pcie-iproc-platform.c also
> have the same issue.

Thanks for the review, I will drop this.

-- 
With Best Regards,
Andy Shevchenko