lists.openwall.net   lists  /  announce  owl-users  owl-dev  john-users  john-dev  passwdqc-users  yescrypt  popa3d-users  /  oss-security  kernel-hardening  musl  sabotage  tlsify  passwords  /  crypt-dev  xvendor  /  Bugtraq  Full-Disclosure  linux-kernel  linux-netdev  linux-ext4  linux-hardening  linux-cve-announce  PHC 
Open Source and information security mailing list archives
 
Hash Suite: Windows password security audit tool. GUI, reports in PDF.
[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20250922142610.80200-1-lpieralisi@kernel.org>
Date: Mon, 22 Sep 2025 16:26:10 +0200
From: Lorenzo Pieralisi <lpieralisi@...nel.org>
To: linux-kernel@...r.kernel.org
Cc: linux-arm-kernel@...ts.infradead.org,
	devicetree@...r.kernel.org,
	Sascha Bischoff <sascha.bischoff@....com>,
	Thomas Gleixner <tglx@...utronix.de>,
	Rob Herring <robh@...nel.org>,
	Marc Zyngier <maz@...nel.org>
Subject: [PATCH] irqchip/gic-v5: Add PCI bus msi-parent property handling

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 its_v5_pci_msi_prepare() 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@...nel.org>
Cc: Sascha Bischoff <sascha.bischoff@....com>
Cc: Thomas Gleixner <tglx@...utronix.de>
Cc: Rob Herring <robh@...nel.org>
Cc: Marc Zyngier <maz@...nel.org>
---
Follow-up to [1] in that it is a fix and too risky to fix generic OF code at
this stage of development since it might affect other platforms.

Apply a fix to GIC ITS v5 MSI parent code - follow-up will clean up
the msi-parent parsing in the kernel tree.

[1] https://lore.kernel.org/lkml/20250916091858.257868-1-lpieralisi@kernel.org/

 drivers/irqchip/irq-gic-its-msi-parent.c | 34 ++++++++++++++++++++++--
 1 file changed, 32 insertions(+), 2 deletions(-)

diff --git a/drivers/irqchip/irq-gic-its-msi-parent.c b/drivers/irqchip/irq-gic-its-msi-parent.c
index eb1473f1448a..198fb4e9a22d 100644
--- a/drivers/irqchip/irq-gic-its-msi-parent.c
+++ b/drivers/irqchip/irq-gic-its-msi-parent.c
@@ -101,6 +101,33 @@ static int its_pci_msi_prepare(struct irq_domain *domain, struct device *dev,
 	return msi_info->ops->msi_prepare(domain->parent, dev, nvec, info);
 }
 
+static int its_v5_get_msi_parent(struct device *dev, struct device_node **msi_np)
+{
+	struct of_phandle_args out_msi;
+	struct device *parent_dev;
+	int ret;
+
+	/*
+	 * Walk up the device parent links looking for one with a
+	 *  "msi-parent" property.
+	 */
+	for (parent_dev = dev; parent_dev; parent_dev = parent_dev->parent) {
+		ret = of_parse_phandle_with_optional_args(parent_dev->of_node, "msi-parent",
+							  "#msi-cells",
+							  0, &out_msi);
+		if (!ret) {
+			if (!out_msi.args_count) {
+				/* Return with a node reference held */
+				*msi_np = out_msi.np;
+				return 0;
+			}
+			of_node_put(out_msi.np);
+		}
+	}
+
+	return -ENODEV;
+}
+
 static int its_v5_pci_msi_prepare(struct irq_domain *domain, struct device *dev,
 				  int nvec, msi_alloc_info_t *info)
 {
@@ -117,8 +144,11 @@ static int its_v5_pci_msi_prepare(struct irq_domain *domain, struct device *dev,
 	pdev = to_pci_dev(dev);
 
 	rid = pci_msi_map_rid_ctlr_node(pdev, &msi_node);
-	if (!msi_node)
-		return -ENODEV;
+	if (!msi_node) {
+		ret = its_v5_get_msi_parent(&pdev->dev, &msi_node);
+		if (ret)
+			return ret;
+	}
 
 	ret = its_translate_frame_address(msi_node, &pa);
 	if (ret)
-- 
2.48.0


Powered by blists - more mailing lists

Powered by Openwall GNU/*/Linux Powered by OpenVZ