Merge patch series "Add OP-TEE based RPMB driver for UFS devices"

Bean Huo <beanhuo@iokpp.de> says:

This patch series introduces OP-TEE based RPMB (Replay Protected
Memory Block) support for UFS devices, extending the kernel-level
secure storage capabilities that are currently available for eMMC
devices.

Previously, OP-TEE required a userspace supplicant to access RPMB
partitions, which created complex dependencies and reliability issues,
especially during early boot scenarios. Recent work by Linaro has
moved core supplicant functionality directly into the Linux kernel for
eMMC devices, eliminating userspace dependencies and enabling
immediate secure storage access. This series extends the same approach
to UFS devices, which are used in enterprise and mobile applications
that require secure storage capabilities.

Benefits:

 - Eliminates dependency on userspace supplicant for UFS RPMB access

 - Enables early boot secure storage access (e.g., fTPM, secure UEFI
   variables)

 - Provides kernel-level RPMB access as soon as UFS driver is
   initialized

 - Removes complex initramfs dependencies and boot ordering
   requirements

 - Ensures reliable and deterministic secure storage operations

 - Supports both built-in and modular fTPM configurations.

Prerequisites:
--------------

This patch series depends on commit 7e8242405b94 ("rpmb: move struct
rpmb_frame to common header") which has been merged into mainline
v6.18-rc2.

Link: https://patch.msgid.link/20251107230518.4060231-1-beanhuo@iokpp.de
Signed-off-by: Martin K. Petersen <martin.petersen@oracle.com>

2 files changed, 14 insertions, 1 deletions

diff --git a/drivers/pci/controller/cadence/pcie-cadence-ep.c b/drivers/pci/controller/cadence/pcie-cadence-ep.c
index 1eac012a8226..c0e1194a936b 100644
--- a/drivers/pci/controller/cadence/pcie-cadence-ep.c
+++ b/drivers/pci/controller/cadence/pcie-cadence-ep.c
@@ -255,7 +255,7 @@ static int cdns_pcie_ep_get_msi(struct pci_epc *epc, u8 fn, u8 vfn)
 	u16 flags, mme;
 	u8 cap;
 
-	cap = cdns_pcie_find_capability(pcie, PCI_CAP_ID_MSIX);
+	cap = cdns_pcie_find_capability(pcie, PCI_CAP_ID_MSI);
 	fn = cdns_pcie_get_fn_from_vfn(pcie, fn, vfn);
 
 	/* Validate that the MSI feature is actually enabled. */
diff --git a/drivers/pci/controller/vmd.c b/drivers/pci/controller/vmd.c
index 1bd5bf4a6097..b4b62b9ccc45 100644
--- a/drivers/pci/controller/vmd.c
+++ b/drivers/pci/controller/vmd.c
@@ -192,6 +192,12 @@ static void vmd_pci_msi_enable(struct irq_data *data)
 	data->chip->irq_unmask(data);
 }
 
+static unsigned int vmd_pci_msi_startup(struct irq_data *data)
+{
+	vmd_pci_msi_enable(data);
+	return 0;
+}
+
 static void vmd_irq_disable(struct irq_data *data)
 {
 	struct vmd_irq *vmdirq = data->chip_data;
@@ -210,6 +216,11 @@ static void vmd_pci_msi_disable(struct irq_data *data)
 	vmd_irq_disable(data->parent_data);
 }
 
+static void vmd_pci_msi_shutdown(struct irq_data *data)
+{
+	vmd_pci_msi_disable(data);
+}
+
 static struct irq_chip vmd_msi_controller = {
 	.name			= "VMD-MSI",
 	.irq_compose_msi_msg	= vmd_compose_msi_msg,
@@ -309,6 +320,8 @@ static bool vmd_init_dev_msi_info(struct device *dev, struct irq_domain *domain,
 	if (!msi_lib_init_dev_msi_info(dev, domain, real_parent, info))
 		return false;
 
+	info->chip->irq_startup		= vmd_pci_msi_startup;
+	info->chip->irq_shutdown	= vmd_pci_msi_shutdown;
 	info->chip->irq_enable		= vmd_pci_msi_enable;
 	info->chip->irq_disable		= vmd_pci_msi_disable;
 	return true;