1. Patchew
  2. linux
  3. [PATCH v3 0/5] of/irq: Misc msi-parent handling fixes/clean-ups
  4. View patch

[PATCH v3 1/5] of/irq: Add msi-parent check to of_msi_xlate()

Lorenzo Pieralisi posted 5 patches 3 months, 3 weeks ago
There is a newer version of this series
[PATCH v3 1/5] of/irq: Add msi-parent check to of_msi_xlate()
Posted by Lorenzo Pieralisi 3 months, 3 weeks ago 
In some legacy platforms the MSI controller for a PCI host bridge is
identified by an msi-parent property whose phandle points at an MSI
controller node with no #msi-cells property, that implicitly
means #msi-cells == 0.

For such platforms, mapping a device ID and retrieving the MSI controller
node becomes simply a matter of checking whether in the device hierarchy
there is an msi-parent property pointing at an MSI controller node with
such characteristics.

Add a helper function to of_msi_xlate() to check the msi-parent property in
addition to msi-map and retrieve the MSI controller node (with a 1:1 ID
deviceID-IN<->deviceID-OUT  mapping) to provide support for deviceID
mapping and MSI controller node retrieval for such platforms.

Fixes: 57d72196dfc8 ("irqchip/gic-v5: Add GICv5 ITS support")
Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
Cc: Sascha Bischoff <sascha.bischoff@arm.com>
Cc: Rob Herring <robh@kernel.org>
Cc: Marc Zyngier <maz@kernel.org>
---
 drivers/of/irq.c | 38 +++++++++++++++++++++++++++++++++++---
 1 file changed, 35 insertions(+), 3 deletions(-)

diff --git a/drivers/of/irq.c b/drivers/of/irq.c
index 65c3c23255b7..e67b2041e73b 100644
--- a/drivers/of/irq.c
+++ b/drivers/of/irq.c
@@ -671,6 +671,35 @@ void __init of_irq_init(const struct of_device_id *matches)
 	}
 }
 
+static int of_check_msi_parent(struct device_node *dev_node, struct device_node **msi_node)
+{
+	struct of_phandle_args msi_spec;
+	int ret;
+
+	/*
+	 * An msi-parent phandle with a missing or == 0 #msi-cells
+	 * property identifies a 1:1 ID translation mapping.
+	 *
+	 * Set the msi controller node if the firmware matches this
+	 * condition.
+	 */
+	ret = of_parse_phandle_with_optional_args(dev_node, "msi-parent", "#msi-cells",
+						  0, &msi_spec);
+	if (!ret) {
+		if ((*msi_node && *msi_node != msi_spec.np) || msi_spec.args_count != 0)
+			ret = -EINVAL;
+
+		if (!ret) {
+			/* Return with a node reference held */
+			*msi_node = msi_spec.np;
+			return 0;
+		}
+		of_node_put(msi_spec.np);
+	}
+
+	return ret;
+}
+
 /**
  * of_msi_xlate - map a MSI ID and find relevant MSI controller node
  * @dev: device for which the mapping is to be done.
@@ -678,7 +707,7 @@ void __init of_irq_init(const struct of_device_id *matches)
  * @id_in: Device ID.
  *
  * Walk up the device hierarchy looking for devices with a "msi-map"
- * property. If found, apply the mapping to @id_in.
+ * or "msi-parent" property. If found, apply the mapping to @id_in.
  * If @msi_np points to a non-NULL device node pointer, only entries targeting
  * that node will be matched; if it points to a NULL value, it will receive the
  * device node of the first matching target phandle, with a reference held.
@@ -692,12 +721,15 @@ u32 of_msi_xlate(struct device *dev, struct device_node **msi_np, u32 id_in)
 
 	/*
 	 * Walk up the device parent links looking for one with a
-	 * "msi-map" property.
+	 * "msi-map" or an "msi-parent" property.
 	 */
-	for (parent_dev = dev; parent_dev; parent_dev = parent_dev->parent)
+	for (parent_dev = dev; parent_dev; parent_dev = parent_dev->parent) {
 		if (!of_map_id(parent_dev->of_node, id_in, "msi-map",
 				"msi-map-mask", msi_np, &id_out))
 			break;
+		if (!of_check_msi_parent(parent_dev->of_node, msi_np))
+			break;
+	}
 	return id_out;
 }
 
-- 
2.50.1
Re: [PATCH v3 1/5] of/irq: Add msi-parent check to of_msi_xlate()
Posted by Rob Herring 3 months, 2 weeks ago 
On Fri, Oct 17, 2025 at 10:47:48AM +0200, Lorenzo Pieralisi wrote:
> In some legacy platforms the MSI controller for a PCI host bridge is
> identified by an msi-parent property whose phandle points at an MSI
> controller node with no #msi-cells property, that implicitly
> means #msi-cells == 0.
> 
> For such platforms, mapping a device ID and retrieving the MSI controller
> node becomes simply a matter of checking whether in the device hierarchy
> there is an msi-parent property pointing at an MSI controller node with
> such characteristics.
> 
> Add a helper function to of_msi_xlate() to check the msi-parent property in
> addition to msi-map and retrieve the MSI controller node (with a 1:1 ID
> deviceID-IN<->deviceID-OUT  mapping) to provide support for deviceID
> mapping and MSI controller node retrieval for such platforms.
> 
> Fixes: 57d72196dfc8 ("irqchip/gic-v5: Add GICv5 ITS support")
> Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
> Cc: Sascha Bischoff <sascha.bischoff@arm.com>
> Cc: Rob Herring <robh@kernel.org>
> Cc: Marc Zyngier <maz@kernel.org>
> ---
>  drivers/of/irq.c | 38 +++++++++++++++++++++++++++++++++++---
>  1 file changed, 35 insertions(+), 3 deletions(-)

It all looks good to me other than 1 nit below. How do you propose 
merging the series? I can take the first 3 for 6.18 and then patches 4 
and 5 can go via their respective trees for 6.19?

> 
> diff --git a/drivers/of/irq.c b/drivers/of/irq.c
> index 65c3c23255b7..e67b2041e73b 100644
> --- a/drivers/of/irq.c
> +++ b/drivers/of/irq.c
> @@ -671,6 +671,35 @@ void __init of_irq_init(const struct of_device_id *matches)
>  	}
>  }
>  
> +static int of_check_msi_parent(struct device_node *dev_node, struct device_node **msi_node)
> +{
> +	struct of_phandle_args msi_spec;
> +	int ret;
> +
> +	/*
> +	 * An msi-parent phandle with a missing or == 0 #msi-cells
> +	 * property identifies a 1:1 ID translation mapping.
> +	 *
> +	 * Set the msi controller node if the firmware matches this
> +	 * condition.
> +	 */
> +	ret = of_parse_phandle_with_optional_args(dev_node, "msi-parent", "#msi-cells",
> +						  0, &msi_spec);
> +	if (!ret) {

if (ret)
	return ret;

And then save a level of indentation.

> +		if ((*msi_node && *msi_node != msi_spec.np) || msi_spec.args_count != 0)
> +			ret = -EINVAL;
> +
> +		if (!ret) {
> +			/* Return with a node reference held */
> +			*msi_node = msi_spec.np;
> +			return 0;
> +		}
> +		of_node_put(msi_spec.np);
> +	}
> +
> +	return ret;
> +}
> +
>  /**
>   * of_msi_xlate - map a MSI ID and find relevant MSI controller node
>   * @dev: device for which the mapping is to be done.
> @@ -678,7 +707,7 @@ void __init of_irq_init(const struct of_device_id *matches)
>   * @id_in: Device ID.
>   *
>   * Walk up the device hierarchy looking for devices with a "msi-map"
> - * property. If found, apply the mapping to @id_in.
> + * or "msi-parent" property. If found, apply the mapping to @id_in.
>   * If @msi_np points to a non-NULL device node pointer, only entries targeting
>   * that node will be matched; if it points to a NULL value, it will receive the
>   * device node of the first matching target phandle, with a reference held.
> @@ -692,12 +721,15 @@ u32 of_msi_xlate(struct device *dev, struct device_node **msi_np, u32 id_in)
>  
>  	/*
>  	 * Walk up the device parent links looking for one with a
> -	 * "msi-map" property.
> +	 * "msi-map" or an "msi-parent" property.
>  	 */
> -	for (parent_dev = dev; parent_dev; parent_dev = parent_dev->parent)
> +	for (parent_dev = dev; parent_dev; parent_dev = parent_dev->parent) {
>  		if (!of_map_id(parent_dev->of_node, id_in, "msi-map",
>  				"msi-map-mask", msi_np, &id_out))
>  			break;
> +		if (!of_check_msi_parent(parent_dev->of_node, msi_np))
> +			break;
> +	}
>  	return id_out;
>  }
>  
> -- 
> 2.50.1
>
Re: [PATCH v3 1/5] of/irq: Add msi-parent check to of_msi_xlate()
Posted by Lorenzo Pieralisi 3 months, 2 weeks ago 
On Tue, Oct 21, 2025 at 06:55:17AM -0500, Rob Herring wrote:
> On Fri, Oct 17, 2025 at 10:47:48AM +0200, Lorenzo Pieralisi wrote:
> > In some legacy platforms the MSI controller for a PCI host bridge is
> > identified by an msi-parent property whose phandle points at an MSI
> > controller node with no #msi-cells property, that implicitly
> > means #msi-cells == 0.
> > 
> > For such platforms, mapping a device ID and retrieving the MSI controller
> > node becomes simply a matter of checking whether in the device hierarchy
> > there is an msi-parent property pointing at an MSI controller node with
> > such characteristics.
> > 
> > Add a helper function to of_msi_xlate() to check the msi-parent property in
> > addition to msi-map and retrieve the MSI controller node (with a 1:1 ID
> > deviceID-IN<->deviceID-OUT  mapping) to provide support for deviceID
> > mapping and MSI controller node retrieval for such platforms.
> > 
> > Fixes: 57d72196dfc8 ("irqchip/gic-v5: Add GICv5 ITS support")
> > Signed-off-by: Lorenzo Pieralisi <lpieralisi@kernel.org>
> > Cc: Sascha Bischoff <sascha.bischoff@arm.com>
> > Cc: Rob Herring <robh@kernel.org>
> > Cc: Marc Zyngier <maz@kernel.org>
> > ---
> >  drivers/of/irq.c | 38 +++++++++++++++++++++++++++++++++++---
> >  1 file changed, 35 insertions(+), 3 deletions(-)
> 
> It all looks good to me other than 1 nit below. How do you propose 
> merging the series? I can take the first 3 for 6.18 and then patches 4 
> and 5 can go via their respective trees for 6.19?

Yep, though patch (3) isn't really a fix so not sure it is v6.18 material
but that's up to you.

I'd leave patch (1) brewing in -next if possible a little bit - it should
not cause any issues but it might.

> > 
> > diff --git a/drivers/of/irq.c b/drivers/of/irq.c
> > index 65c3c23255b7..e67b2041e73b 100644
> > --- a/drivers/of/irq.c
> > +++ b/drivers/of/irq.c
> > @@ -671,6 +671,35 @@ void __init of_irq_init(const struct of_device_id *matches)
> >  	}
> >  }
> >  
> > +static int of_check_msi_parent(struct device_node *dev_node, struct device_node **msi_node)
> > +{
> > +	struct of_phandle_args msi_spec;
> > +	int ret;
> > +
> > +	/*
> > +	 * An msi-parent phandle with a missing or == 0 #msi-cells
> > +	 * property identifies a 1:1 ID translation mapping.
> > +	 *
> > +	 * Set the msi controller node if the firmware matches this
> > +	 * condition.
> > +	 */
> > +	ret = of_parse_phandle_with_optional_args(dev_node, "msi-parent", "#msi-cells",
> > +						  0, &msi_spec);
> > +	if (!ret) {
> 
> if (ret)
> 	return ret;
> 
> And then save a level of indentation.

Fixed - sent v4.

Thanks !
Lorenzo

> > +		if ((*msi_node && *msi_node != msi_spec.np) || msi_spec.args_count != 0)
> > +			ret = -EINVAL;
> > +
> > +		if (!ret) {
> > +			/* Return with a node reference held */
> > +			*msi_node = msi_spec.np;
> > +			return 0;
> > +		}
> > +		of_node_put(msi_spec.np);
> > +	}
> > +
> > +	return ret;
> > +}
> > +
> >  /**
> >   * of_msi_xlate - map a MSI ID and find relevant MSI controller node
> >   * @dev: device for which the mapping is to be done.
> > @@ -678,7 +707,7 @@ void __init of_irq_init(const struct of_device_id *matches)
> >   * @id_in: Device ID.
> >   *
> >   * Walk up the device hierarchy looking for devices with a "msi-map"
> > - * property. If found, apply the mapping to @id_in.
> > + * or "msi-parent" property. If found, apply the mapping to @id_in.
> >   * If @msi_np points to a non-NULL device node pointer, only entries targeting
> >   * that node will be matched; if it points to a NULL value, it will receive the
> >   * device node of the first matching target phandle, with a reference held.
> > @@ -692,12 +721,15 @@ u32 of_msi_xlate(struct device *dev, struct device_node **msi_np, u32 id_in)
> >  
> >  	/*
> >  	 * Walk up the device parent links looking for one with a
> > -	 * "msi-map" property.
> > +	 * "msi-map" or an "msi-parent" property.
> >  	 */
> > -	for (parent_dev = dev; parent_dev; parent_dev = parent_dev->parent)
> > +	for (parent_dev = dev; parent_dev; parent_dev = parent_dev->parent) {
> >  		if (!of_map_id(parent_dev->of_node, id_in, "msi-map",
> >  				"msi-map-mask", msi_np, &id_out))
> >  			break;
> > +		if (!of_check_msi_parent(parent_dev->of_node, msi_np))
> > +			break;
> > +	}
> >  	return id_out;
> >  }
> >  
> > -- 
> > 2.50.1
> >

Patchew 2.1-devel-b67e4c2

© 2016 - 2026 Red Hat, Inc.