Merge tag 'iommu-updates-v6.1' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu

Pull iommu updates from Joerg Roedel:

 - remove the bus_set_iommu() interface which became unnecesary because
   of IOMMU per-device probing

 - make the dma-iommu.h header private

 - Intel VT-d changes from Lu Baolu:
	  - Decouple PASID and PRI from SVA
	  - Add ESRTPS & ESIRTPS capability check
	  - Cleanups

 - Apple DART support for the M1 Pro/MAX SOCs

 - support for AMD IOMMUv2 page-tables for the DMA-API layer.

   The v2 page-tables are compatible with the x86 CPU page-tables. Using
   them for DMA-API prepares support for hardware-assisted IOMMU
   virtualization

 - support for MT6795 Helio X10 M4Us in the Mediatek IOMMU driver

 - some smaller fixes and cleanups

* tag 'iommu-updates-v6.1' of git://git.kernel.org/pub/scm/linux/kernel/git/joro/iommu: (59 commits)
  iommu/vt-d: Avoid unnecessary global DMA cache invalidation
  iommu/vt-d: Avoid unnecessary global IRTE cache invalidation
  iommu/vt-d: Rename cap_5lp_support to cap_fl5lp_support
  iommu/vt-d: Remove pasid_set_eafe()
  iommu/vt-d: Decouple PASID & PRI enabling from SVA
  iommu/vt-d: Remove unnecessary SVA data accesses in page fault path
  dt-bindings: iommu: arm,smmu-v3: Relax order of interrupt names
  iommu: dart: Support t6000 variant
  iommu/io-pgtable-dart: Add DART PTE support for t6000
  iommu/io-pgtable: Add DART subpage protection support
  iommu/io-pgtable: Move Apple DART support to its own file
  iommu/mediatek: Add support for MT6795 Helio X10 M4Us
  iommu/mediatek: Introduce new flag TF_PORT_TO_ADDR_MT8173
  dt-bindings: mediatek: Add bindings for MT6795 M4U
  iommu/iova: Fix module config properly
  iommu/amd: Fix sparse warning
  iommu/amd: Remove outdated comment
  iommu/amd: Free domain ID after domain_flush_pages
  iommu/amd: Free domain id in error path
  iommu/virtio: Fix compile error with viommu_capable()
  ...

3 files changed, 17 insertions, 148 deletions

diff --git a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
index 71f7edded9cf..ba47c73f5b8c 100644
--- a/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+++ b/drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
@@ -14,7 +14,6 @@
 #include <linux/bitops.h>
 #include <linux/crash_dump.h>
 #include <linux/delay.h>
-#include <linux/dma-iommu.h>
 #include <linux/err.h>
 #include <linux/interrupt.h>
 #include <linux/io-pgtable.h>
@@ -28,9 +27,8 @@
 #include <linux/pci-ats.h>
 #include <linux/platform_device.h>
 
-#include <linux/amba/bus.h>
-
 #include "arm-smmu-v3.h"
+#include "../../dma-iommu.h"
 #include "../../iommu-sva-lib.h"
 
 static bool disable_bypass = true;
@@ -1992,11 +1990,14 @@ static const struct iommu_flush_ops arm_smmu_flush_ops = {
 };
 
 /* IOMMU API */
-static bool arm_smmu_capable(enum iommu_cap cap)
+static bool arm_smmu_capable(struct device *dev, enum iommu_cap cap)
 {
+	struct arm_smmu_master *master = dev_iommu_priv_get(dev);
+
 	switch (cap) {
 	case IOMMU_CAP_CACHE_COHERENCY:
-		return true;
+		/* Assume that a coherent TCU implies coherent TBUs */
+		return master->smmu->features & ARM_SMMU_FEAT_COHERENCY;
 	case IOMMU_CAP_NOEXEC:
 		return true;
 	default:
@@ -3694,43 +3695,6 @@ static unsigned long arm_smmu_resource_size(struct arm_smmu_device *smmu)
 		return SZ_128K;
 }
 
-static int arm_smmu_set_bus_ops(struct iommu_ops *ops)
-{
-	int err;
-
-#ifdef CONFIG_PCI
-	if (pci_bus_type.iommu_ops != ops) {
-		err = bus_set_iommu(&pci_bus_type, ops);
-		if (err)
-			return err;
-	}
-#endif
-#ifdef CONFIG_ARM_AMBA
-	if (amba_bustype.iommu_ops != ops) {
-		err = bus_set_iommu(&amba_bustype, ops);
-		if (err)
-			goto err_reset_pci_ops;
-	}
-#endif
-	if (platform_bus_type.iommu_ops != ops) {
-		err = bus_set_iommu(&platform_bus_type, ops);
-		if (err)
-			goto err_reset_amba_ops;
-	}
-
-	return 0;
-
-err_reset_amba_ops:
-#ifdef CONFIG_ARM_AMBA
-	bus_set_iommu(&amba_bustype, NULL);
-#endif
-err_reset_pci_ops: __maybe_unused;
-#ifdef CONFIG_PCI
-	bus_set_iommu(&pci_bus_type, NULL);
-#endif
-	return err;
-}
-
 static void __iomem *arm_smmu_ioremap(struct device *dev, resource_size_t start,
 				      resource_size_t size)
 {
@@ -3869,27 +3833,17 @@ static int arm_smmu_device_probe(struct platform_device *pdev)
 	ret = iommu_device_register(&smmu->iommu, &arm_smmu_ops, dev);
 	if (ret) {
 		dev_err(dev, "Failed to register iommu\n");
-		goto err_sysfs_remove;
+		iommu_device_sysfs_remove(&smmu->iommu);
+		return ret;
 	}
 
-	ret = arm_smmu_set_bus_ops(&arm_smmu_ops);
-	if (ret)
-		goto err_unregister_device;
-
 	return 0;
-
-err_unregister_device:
-	iommu_device_unregister(&smmu->iommu);
-err_sysfs_remove:
-	iommu_device_sysfs_remove(&smmu->iommu);
-	return ret;
 }
 
 static int arm_smmu_device_remove(struct platform_device *pdev)
 {
 	struct arm_smmu_device *smmu = platform_get_drvdata(pdev);
 
-	arm_smmu_set_bus_ops(NULL);
 	iommu_device_unregister(&smmu->iommu);
 	iommu_device_sysfs_remove(&smmu->iommu);
 	arm_smmu_device_disable(smmu);
diff --git a/drivers/iommu/arm/arm-smmu/arm-smmu.c b/drivers/iommu/arm/arm-smmu/arm-smmu.c
index dfa82df00342..6c1114a4d6cc 100644
--- a/drivers/iommu/arm/arm-smmu/arm-smmu.c
+++ b/drivers/iommu/arm/arm-smmu/arm-smmu.c
@@ -21,7 +21,6 @@
 #include <linux/acpi_iort.h>
 #include <linux/bitfield.h>
 #include <linux/delay.h>
-#include <linux/dma-iommu.h>
 #include <linux/dma-mapping.h>
 #include <linux/err.h>
 #include <linux/interrupt.h>
@@ -37,10 +36,10 @@
 #include <linux/ratelimit.h>
 #include <linux/slab.h>
 
-#include <linux/amba/bus.h>
 #include <linux/fsl/mc.h>
 
 #include "arm-smmu.h"
+#include "../../dma-iommu.h"
 
 /*
  * Apparently, some Qualcomm arm64 platforms which appear to expose their SMMU
@@ -93,8 +92,6 @@ static struct platform_driver arm_smmu_driver;
 static struct iommu_ops arm_smmu_ops;
 
 #ifdef CONFIG_ARM_SMMU_LEGACY_DT_BINDINGS
-static int arm_smmu_bus_init(struct iommu_ops *ops);
-
 static struct device_node *dev_get_dev_node(struct device *dev)
 {
 	if (dev_is_pci(dev)) {
@@ -180,20 +177,6 @@ static int arm_smmu_register_legacy_master(struct device *dev,
 	kfree(sids);
 	return err;
 }
-
-/*
- * With the legacy DT binding in play, we have no guarantees about
- * probe order, but then we're also not doing default domains, so we can
- * delay setting bus ops until we're sure every possible SMMU is ready,
- * and that way ensure that no probe_device() calls get missed.
- */
-static int arm_smmu_legacy_bus_init(void)
-{
-	if (using_legacy_binding)
-		return arm_smmu_bus_init(&arm_smmu_ops);
-	return 0;
-}
-device_initcall_sync(arm_smmu_legacy_bus_init);
 #else
 static int arm_smmu_register_legacy_master(struct device *dev,
 					   struct arm_smmu_device **smmu)
@@ -1330,15 +1313,14 @@ static phys_addr_t arm_smmu_iova_to_phys(struct iommu_domain *domain,
 	return ops->iova_to_phys(ops, iova);
 }
 
-static bool arm_smmu_capable(enum iommu_cap cap)
+static bool arm_smmu_capable(struct device *dev, enum iommu_cap cap)
 {
+	struct arm_smmu_master_cfg *cfg = dev_iommu_priv_get(dev);
+
 	switch (cap) {
 	case IOMMU_CAP_CACHE_COHERENCY:
-		/*
-		 * Return true here as the SMMU can always send out coherent
-		 * requests.
-		 */
-		return true;
+		/* Assume that a coherent TCU implies coherent TBUs */
+		return cfg->smmu->features & ARM_SMMU_FEAT_COHERENT_WALK;
 	case IOMMU_CAP_NOEXEC:
 		return true;
 	default:
@@ -2016,52 +1998,6 @@ static int arm_smmu_device_dt_probe(struct arm_smmu_device *smmu,
 	return 0;
 }
 
-static int arm_smmu_bus_init(struct iommu_ops *ops)
-{
-	int err;
-
-	/* Oh, for a proper bus abstraction */
-	if (!iommu_present(&platform_bus_type)) {
-		err = bus_set_iommu(&platform_bus_type, ops);
-		if (err)
-			return err;
-	}
-#ifdef CONFIG_ARM_AMBA
-	if (!iommu_present(&amba_bustype)) {
-		err = bus_set_iommu(&amba_bustype, ops);
-		if (err)
-			goto err_reset_platform_ops;
-	}
-#endif
-#ifdef CONFIG_PCI
-	if (!iommu_present(&pci_bus_type)) {
-		err = bus_set_iommu(&pci_bus_type, ops);
-		if (err)
-			goto err_reset_amba_ops;
-	}
-#endif
-#ifdef CONFIG_FSL_MC_BUS
-	if (!iommu_present(&fsl_mc_bus_type)) {
-		err = bus_set_iommu(&fsl_mc_bus_type, ops);
-		if (err)
-			goto err_reset_pci_ops;
-	}
-#endif
-	return 0;
-
-err_reset_pci_ops: __maybe_unused;
-#ifdef CONFIG_PCI
-	bus_set_iommu(&pci_bus_type, NULL);
-#endif
-err_reset_amba_ops: __maybe_unused;
-#ifdef CONFIG_ARM_AMBA
-	bus_set_iommu(&amba_bustype, NULL);
-#endif
-err_reset_platform_ops: __maybe_unused;
-	bus_set_iommu(&platform_bus_type, NULL);
-	return err;
-}
-
 static void arm_smmu_rmr_install_bypass_smr(struct arm_smmu_device *smmu)
 {
 	struct list_head rmr_list;
@@ -2226,7 +2162,8 @@ static int arm_smmu_device_probe(struct platform_device *pdev)
 	err = iommu_device_register(&smmu->iommu, &arm_smmu_ops, dev);
 	if (err) {
 		dev_err(dev, "Failed to register iommu\n");
-		goto err_sysfs_remove;
+		iommu_device_sysfs_remove(&smmu->iommu);
+		return err;
 	}
 
 	platform_set_drvdata(pdev, smmu);
@@ -2248,24 +2185,7 @@ static int arm_smmu_device_probe(struct platform_device *pdev)
 		pm_runtime_enable(dev);
 	}
 
-	/*
-	 * For ACPI and generic DT bindings, an SMMU will be probed before
-	 * any device which might need it, so we want the bus ops in place
-	 * ready to handle default domain setup as soon as any SMMU exists.
-	 */
-	if (!using_legacy_binding) {
-		err = arm_smmu_bus_init(&arm_smmu_ops);
-		if (err)
-			goto err_unregister_device;
-	}
-
 	return 0;
-
-err_unregister_device:
-	iommu_device_unregister(&smmu->iommu);
-err_sysfs_remove:
-	iommu_device_sysfs_remove(&smmu->iommu);
-	return err;
 }
 
 static int arm_smmu_device_remove(struct platform_device *pdev)
@@ -2278,7 +2198,6 @@ static int arm_smmu_device_remove(struct platform_device *pdev)
 	if (!bitmap_empty(smmu->context_map, ARM_SMMU_MAX_CBS))
 		dev_notice(&pdev->dev, "disabling translation\n");
 
-	arm_smmu_bus_init(NULL);
 	iommu_device_unregister(&smmu->iommu);
 	iommu_device_sysfs_remove(&smmu->iommu);
 
diff --git a/drivers/iommu/arm/arm-smmu/qcom_iommu.c b/drivers/iommu/arm/arm-smmu/qcom_iommu.c
index 17235116d3bb..3869c3ecda8c 100644
--- a/drivers/iommu/arm/arm-smmu/qcom_iommu.c
+++ b/drivers/iommu/arm/arm-smmu/qcom_iommu.c
@@ -493,7 +493,7 @@ static phys_addr_t qcom_iommu_iova_to_phys(struct iommu_domain *domain,
 	return ret;
 }
 
-static bool qcom_iommu_capable(enum iommu_cap cap)
+static bool qcom_iommu_capable(struct device *dev, enum iommu_cap cap)
 {
 	switch (cap) {
 	case IOMMU_CAP_CACHE_COHERENCY:
@@ -837,8 +837,6 @@ static int qcom_iommu_device_probe(struct platform_device *pdev)
 		goto err_pm_disable;
 	}
 
-	bus_set_iommu(&platform_bus_type, &qcom_iommu_ops);
-
 	if (qcom_iommu->local_base) {
 		pm_runtime_get_sync(dev);
 		writel_relaxed(0xffffffff, qcom_iommu->local_base + SMMU_INTR_SEL_NS);
@@ -856,8 +854,6 @@ static int qcom_iommu_device_remove(struct platform_device *pdev)
 {
 	struct qcom_iommu_dev *qcom_iommu = platform_get_drvdata(pdev);
 
-	bus_set_iommu(&platform_bus_type, NULL);
-
 	pm_runtime_force_suspend(&pdev->dev);
 	platform_set_drvdata(pdev, NULL);
 	iommu_device_sysfs_remove(&qcom_iommu->iommu);