Merge branches 'arm/exynos', 'arm/renesas', 'arm/rockchip', 'arm/omap',...

Merge branches 'arm/exynos', 'arm/renesas', 'arm/rockchip', 'arm/omap', 'arm/mediatek', 'arm/tegra', 'arm/qcom', 'arm/smmu', 'ppc/pamu', 'x86/vt-d', 'x86/amd', 's390' and 'core' into next

Documentation/devicetree/bindings/iommu/qcom,iommu.txt

+* QCOM IOMMU v1 Implementation
+
+Qualcomm "B" family devices which are not compatible with arm-smmu have
+a similar looking IOMMU but without access to the global register space,
+and optionally requiring additional configuration to route context irqs
+to non-secure vs secure interrupt line.
+
+** Required properties:
+
+- compatible       : Should be one of:
+
+                        "qcom,msm8916-iommu"
+
+                     Followed by "qcom,msm-iommu-v1".
+
+- clock-names      : Should be a pair of "iface" (required for IOMMUs
+                     register group access) and "bus" (required for
+                     the IOMMUs underlying bus access).
+
+- clocks           : Phandles for respective clocks described by
+                     clock-names.
+
+- #address-cells   : must be 1.
+
+- #size-cells      : must be 1.
+
+- #iommu-cells     : Must be 1.  Index identifies the context-bank #.
+
+- ranges           : Base address and size of the iommu context banks.
+
+- qcom,iommu-secure-id  : secure-id.
+
+- List of sub-nodes, one per translation context bank.  Each sub-node
+  has the following required properties:
+
+  - compatible     : Should be one of:
+        - "qcom,msm-iommu-v1-ns"  : non-secure context bank
+        - "qcom,msm-iommu-v1-sec" : secure context bank
+  - reg            : Base address and size of context bank within the iommu
+  - interrupts     : The context fault irq.
+
+** Optional properties:
+
+- reg              : Base address and size of the SMMU local base, should
+                     be only specified if the iommu requires configuration
+                     for routing of context bank irq's to secure vs non-
+                     secure lines.  (Ie. if the iommu contains secure
+                     context banks)
+
+
+** Examples:
+
+	apps_iommu: iommu@1e20000 {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		#iommu-cells = <1>;
+		compatible = "qcom,msm8916-iommu", "qcom,msm-iommu-v1";
+		ranges = <0 0x1e20000 0x40000>;
+		reg = <0x1ef0000 0x3000>;
+		clocks = <&gcc GCC_SMMU_CFG_CLK>,
+			 <&gcc GCC_APSS_TCU_CLK>;
+		clock-names = "iface", "bus";
+		qcom,iommu-secure-id = <17>;
+
+		// mdp_0:
+		iommu-ctx@4000 {
+			compatible = "qcom,msm-iommu-v1-ns";
+			reg = <0x4000 0x1000>;
+			interrupts = <GIC_SPI 70 IRQ_TYPE_LEVEL_HIGH>;
+		};
+
+		// venus_ns:
+		iommu-ctx@5000 {
+			compatible = "qcom,msm-iommu-v1-sec";
+			reg = <0x5000 0x1000>;
+			interrupts = <GIC_SPI 70 IRQ_TYPE_LEVEL_HIGH>;
+		};
+	};
+
+	gpu_iommu: iommu@1f08000 {
+		#address-cells = <1>;
+		#size-cells = <1>;
+		#iommu-cells = <1>;
+		compatible = "qcom,msm8916-iommu", "qcom,msm-iommu-v1";
+		ranges = <0 0x1f08000 0x10000>;
+		clocks = <&gcc GCC_SMMU_CFG_CLK>,
+			 <&gcc GCC_GFX_TCU_CLK>;
+		clock-names = "iface", "bus";
+		qcom,iommu-secure-id = <18>;
+
+		// gfx3d_user:
+		iommu-ctx@1000 {
+			compatible = "qcom,msm-iommu-v1-ns";
+			reg = <0x1000 0x1000>;
+			interrupts = <GIC_SPI 241 IRQ_TYPE_LEVEL_HIGH>;
+		};
+
+		// gfx3d_priv:
+		iommu-ctx@2000 {
+			compatible = "qcom,msm-iommu-v1-ns";
+			reg = <0x2000 0x1000>;
+			interrupts = <GIC_SPI 242 IRQ_TYPE_LEVEL_HIGH>;
+		};
+	};
+
+	...
+
+	venus: video-codec@1d00000 {
+		...
+		iommus = <&apps_iommu 5>;
+	};
+
+	mdp: mdp@1a01000 {
+		...
+		iommus = <&apps_iommu 4>;
+	};
+
+	gpu@01c00000 {
+		...
+		iommus = <&gpu_iommu 1>, <&gpu_iommu 2>;
+	};

Documentation/devicetree/bindings/iommu/rockchip,iommu.txt

@@ -15,6 +15,11 @@ Required properties:
                     to associate with its master device.  See:
                     Documentation/devicetree/bindings/iommu/iommu.txt
 
+Optional properties:
+- rockchip,disable-mmu-reset : Don't use the mmu reset operation.
+			       Some mmu instances may produce unexpected results
+			       when the reset operation is used.
+
 Example:
 
 	vopl_mmu: iommu@ff940300 {

Documentation/devicetree/bindings/memory-controllers/mediatek,smi-larb.txt

@@ -15,6 +15,9 @@ Required properties:
 	    the register.
   - "smi" : It's the clock for transfer data and command.
 
+Required property for mt2701:
+- mediatek,larb-id :the hardware id of this larb.
+
 Example:
 	larb1: larb@16010000 {
 		compatible = "mediatek,mt8173-smi-larb";
@@ -25,3 +28,15 @@ Example:
 			 <&vdecsys CLK_VDEC_LARB_CKEN>;
 		clock-names = "apb", "smi";
 	};
+
+Example for mt2701:
+	larb0: larb@14010000 {
+		compatible = "mediatek,mt2701-smi-larb";
+		reg = <0 0x14010000 0 0x1000>;
+		mediatek,smi = <&smi_common>;
+		mediatek,larb-id = <0>;
+		clocks = <&mmsys CLK_MM_SMI_LARB0>,
+			 <&mmsys CLK_MM_SMI_LARB0>;
+		clock-names = "apb", "smi";
+		power-domains = <&scpsys MT2701_POWER_DOMAIN_DISP>;
+	};

MAINTAINERS

@@ -10940,6 +10940,13 @@ T:	git git://git.kernel.org/pub/scm/linux/kernel/git/rkuo/linux-hexagon-kernel.g
 S:	Supported
 F:	arch/hexagon/
 
+QUALCOMM IOMMU
+M:	Rob Clark <robdclark@gmail.com>
+L:	iommu@lists.linux-foundation.org
+L:	linux-arm-msm@vger.kernel.org
+S:	Maintained
+F:	drivers/iommu/qcom_iommu.c
+
 QUALCOMM VENUS VIDEO ACCELERATOR DRIVER
 M:	Stanimir Varbanov <stanimir.varbanov@linaro.org>
 L:	linux-media@vger.kernel.org

arch/s390/include/asm/pci.h

@@ -8,6 +8,7 @@
 
 #include <linux/pci.h>
 #include <linux/mutex.h>
+#include <linux/iommu.h>
 #include <asm-generic/pci.h>
 #include <asm/pci_clp.h>
 #include <asm/pci_debug.h>
@@ -122,6 +123,8 @@ struct zpci_dev {
 	unsigned long	iommu_pages;
 	unsigned int	next_bit;
 
+	struct iommu_device iommu_dev;  /* IOMMU core handle */
+
 	char res_name[16];
 	struct zpci_bar_struct bars[PCI_BAR_COUNT];
 
@@ -174,6 +177,10 @@ int clp_enable_fh(struct zpci_dev *, u8);
 int clp_disable_fh(struct zpci_dev *);
 int clp_get_state(u32 fid, enum zpci_state *state);
 
+/* IOMMU Interface */
+int zpci_init_iommu(struct zpci_dev *zdev);
+void zpci_destroy_iommu(struct zpci_dev *zdev);
+
 #ifdef CONFIG_PCI
 /* Error handling and recovery */
 void zpci_event_error(void *);

arch/s390/pci/pci.c

@@ -776,6 +776,7 @@ void pcibios_remove_bus(struct pci_bus *bus)
 
 	zpci_exit_slot(zdev);
 	zpci_cleanup_bus_resources(zdev);
+	zpci_destroy_iommu(zdev);
 	zpci_free_domain(zdev);
 
 	spin_lock(&zpci_list_lock);
@@ -848,11 +849,15 @@ int zpci_create_device(struct zpci_dev *zdev)
 	if (rc)
 		goto out;
 
+	rc = zpci_init_iommu(zdev);
+	if (rc)
+		goto out_free;
+
 	mutex_init(&zdev->lock);
 	if (zdev->state == ZPCI_FN_STATE_CONFIGURED) {
 		rc = zpci_enable_device(zdev);
 		if (rc)
-			goto out_free;
+			goto out_destroy_iommu;
 	}
 	rc = zpci_scan_bus(zdev);
 	if (rc)
@@ -869,6 +874,8 @@ int zpci_create_device(struct zpci_dev *zdev)
 out_disable:
 	if (zdev->state == ZPCI_FN_STATE_ONLINE)
 		zpci_disable_device(zdev);
+out_destroy_iommu:
+	zpci_destroy_iommu(zdev);
 out_free:
 	zpci_free_domain(zdev);
 out:

drivers/iommu/Kconfig

@@ -76,6 +76,8 @@ config IOMMU_DMA
 
 config FSL_PAMU
 	bool "Freescale IOMMU support"
+	depends on PCI
+	depends on PHYS_64BIT
 	depends on PPC_E500MC || (COMPILE_TEST && PPC)
 	select IOMMU_API
 	select GENERIC_ALLOCATOR
@@ -253,6 +255,7 @@ config TEGRA_IOMMU_SMMU
 config EXYNOS_IOMMU
 	bool "Exynos IOMMU Support"
 	depends on ARCH_EXYNOS && MMU
+	depends on !CPU_BIG_ENDIAN # revisit driver if we can enable big-endian ptes
 	select IOMMU_API
 	select ARM_DMA_USE_IOMMU
 	help
@@ -367,4 +370,14 @@ config MTK_IOMMU_V1
 
 	  if unsure, say N here.
 
+config QCOM_IOMMU
+	# Note: iommu drivers cannot (yet?) be built as modules
+	bool "Qualcomm IOMMU Support"
+	depends on ARCH_QCOM || COMPILE_TEST
+	select IOMMU_API
+	select IOMMU_IO_PGTABLE_LPAE
+	select ARM_DMA_USE_IOMMU
+	help
+	  Support for IOMMU on certain Qualcomm SoCs.
+
 endif # IOMMU_SUPPORT

drivers/iommu/Makefile

@@ -27,3 +27,4 @@ obj-$(CONFIG_TEGRA_IOMMU_SMMU) += tegra-smmu.o
 obj-$(CONFIG_EXYNOS_IOMMU) += exynos-iommu.o
 obj-$(CONFIG_FSL_PAMU) += fsl_pamu.o fsl_pamu_domain.o
 obj-$(CONFIG_S390_IOMMU) += s390-iommu.o
+obj-$(CONFIG_QCOM_IOMMU) += qcom_iommu.o

drivers/iommu/amd_iommu_proto.h

@@ -87,4 +87,6 @@ static inline bool iommu_feature(struct amd_iommu *iommu, u64 f)
 	return !!(iommu->features & f);
 }
 
+extern bool translation_pre_enabled(struct amd_iommu *iommu);
+extern struct iommu_dev_data *get_dev_data(struct device *dev);
 #endif /* _ASM_X86_AMD_IOMMU_PROTO_H  */

drivers/iommu/amd_iommu_types.h

@@ -250,6 +250,14 @@
 
 #define GA_GUEST_NR		0x1
 
+/* Bit value definition for dte irq remapping fields*/
+#define DTE_IRQ_PHYS_ADDR_MASK	(((1ULL << 45)-1) << 6)
+#define DTE_IRQ_REMAP_INTCTL_MASK	(0x3ULL << 60)
+#define DTE_IRQ_TABLE_LEN_MASK	(0xfULL << 1)
+#define DTE_IRQ_REMAP_INTCTL    (2ULL << 60)
+#define DTE_IRQ_TABLE_LEN       (8ULL << 1)
+#define DTE_IRQ_REMAP_ENABLE    1ULL
+
 #define PAGE_MODE_NONE    0x00
 #define PAGE_MODE_1_LEVEL 0x01
 #define PAGE_MODE_2_LEVEL 0x02
@@ -265,7 +273,7 @@
 #define PM_LEVEL_INDEX(x, a)	(((a) >> PM_LEVEL_SHIFT((x))) & 0x1ffULL)
 #define PM_LEVEL_ENC(x)		(((x) << 9) & 0xe00ULL)
 #define PM_LEVEL_PDE(x, a)	((a) | PM_LEVEL_ENC((x)) | \
-				 IOMMU_PTE_P | IOMMU_PTE_IR | IOMMU_PTE_IW)
+				 IOMMU_PTE_PR | IOMMU_PTE_IR | IOMMU_PTE_IW)
 #define PM_PTE_LEVEL(pte)	(((pte) >> 9) & 0x7ULL)
 
 #define PM_MAP_4k		0
@@ -314,19 +322,29 @@
 #define PTE_LEVEL_PAGE_SIZE(level)			\
 	(1ULL << (12 + (9 * (level))))
 
-#define IOMMU_PTE_P  (1ULL << 0)
-#define IOMMU_PTE_TV (1ULL << 1)
+/*
+ * Bit value definition for I/O PTE fields
+ */
+#define IOMMU_PTE_PR (1ULL << 0)
 #define IOMMU_PTE_U  (1ULL << 59)
 #define IOMMU_PTE_FC (1ULL << 60)
 #define IOMMU_PTE_IR (1ULL << 61)
 #define IOMMU_PTE_IW (1ULL << 62)
 
+/*
+ * Bit value definition for DTE fields
+ */
+#define DTE_FLAG_V  (1ULL << 0)
+#define DTE_FLAG_TV (1ULL << 1)
+#define DTE_FLAG_IR (1ULL << 61)
+#define DTE_FLAG_IW (1ULL << 62)
+
 #define DTE_FLAG_IOTLB	(1ULL << 32)
-#define DTE_FLAG_SA	(1ULL << 34)
 #define DTE_FLAG_GV	(1ULL << 55)
 #define DTE_FLAG_MASK	(0x3ffULL << 32)
 #define DTE_GLX_SHIFT	(56)
 #define DTE_GLX_MASK	(3)
+#define DEV_DOMID_MASK	0xffffULL
 
 #define DTE_GCR3_VAL_A(x)	(((x) >> 12) & 0x00007ULL)
 #define DTE_GCR3_VAL_B(x)	(((x) >> 15) & 0x0ffffULL)
@@ -343,7 +361,7 @@
 #define GCR3_VALID		0x01ULL
 
 #define IOMMU_PAGE_MASK (((1ULL << 52) - 1) & ~0xfffULL)
-#define IOMMU_PTE_PRESENT(pte) ((pte) & IOMMU_PTE_P)
+#define IOMMU_PTE_PRESENT(pte) ((pte) & IOMMU_PTE_PR)
 #define IOMMU_PTE_PAGE(pte) (phys_to_virt((pte) & IOMMU_PAGE_MASK))
 #define IOMMU_PTE_MODE(pte) (((pte) >> 9) & 0x07)
 
@@ -435,6 +453,8 @@ struct iommu_domain;
 struct irq_domain;
 struct amd_irte_ops;
 
+#define AMD_IOMMU_FLAG_TRANS_PRE_ENABLED      (1 << 0)
+
 /*
  * This structure contains generic data for  IOMMU protection domains
  * independent of their use.
@@ -569,6 +589,7 @@ struct amd_iommu {
 	struct amd_irte_ops *irte_ops;
 #endif
 
+	u32 flags;
 	volatile u64 __aligned(8) cmd_sem;
 };
 
@@ -599,6 +620,30 @@ struct devid_map {
 	bool cmd_line;
 };
 
+/*
+ * This struct contains device specific data for the IOMMU
+ */
+struct iommu_dev_data {
+	struct list_head list;		  /* For domain->dev_list */
+	struct list_head dev_data_list;	  /* For global dev_data_list */
+	struct protection_domain *domain; /* Domain the device is bound to */
+	u16 devid;			  /* PCI Device ID */
+	u16 alias;			  /* Alias Device ID */
+	bool iommu_v2;			  /* Device can make use of IOMMUv2 */
+	bool passthrough;		  /* Device is identity mapped */
+	struct {
+		bool enabled;
+		int qdep;
+	} ats;				  /* ATS state */
+	bool pri_tlp;			  /* PASID TLB required for
+					     PPR completions */
+	u32 errata;			  /* Bitmap for errata to apply */
+	bool use_vapic;			  /* Enable device to use vapic mode */
+	bool defer_attach;
+
+	struct ratelimit_state rs;        /* Ratelimit IOPF messages */
+};
+
 /* Map HPET and IOAPIC ids to the devid used by the IOMMU */
 extern struct list_head ioapic_map;
 extern struct list_head hpet_map;

drivers/iommu/amd_iommu_v2.c

@@ -562,14 +562,30 @@ static int ppr_notifier(struct notifier_block *nb, unsigned long e, void *data)
 	unsigned long flags;
 	struct fault *fault;
 	bool finish;
-	u16 tag;
+	u16 tag, devid;
 	int ret;
+	struct iommu_dev_data *dev_data;
+	struct pci_dev *pdev = NULL;
 
 	iommu_fault = data;
 	tag         = iommu_fault->tag & 0x1ff;
 	finish      = (iommu_fault->tag >> 9) & 1;
 
+	devid = iommu_fault->device_id;
+	pdev = pci_get_bus_and_slot(PCI_BUS_NUM(devid), devid & 0xff);
+	if (!pdev)
+		return -ENODEV;
+	dev_data = get_dev_data(&pdev->dev);
+
+	/* In kdump kernel pci dev is not initialized yet -> send INVALID */
 	ret = NOTIFY_DONE;
+	if (translation_pre_enabled(amd_iommu_rlookup_table[devid])
+		&& dev_data->defer_attach) {
+		amd_iommu_complete_ppr(pdev, iommu_fault->pasid,
+				       PPR_INVALID, tag);
+		goto out;
+	}
+
 	dev_state = get_device_state(iommu_fault->device_id);
 	if (dev_state == NULL)
 		goto out;

drivers/iommu/arm-smmu-regs.h

+/*
+ * IOMMU API for ARM architected SMMU implementations.
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ *
+ * This program is distributed in the hope that it will be useful,
+ * but WITHOUT ANY WARRANTY; without even the implied warranty of
+ * MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ *
+ * You should have received a copy of the GNU General Public License
+ * along with this program; if not, write to the Free Software
+ * Foundation, Inc., 59 Temple Place - Suite 330, Boston, MA 02111-1307, USA.
+ *
+ * Copyright (C) 2013 ARM Limited
+ *
+ * Author: Will Deacon <will.deacon@arm.com>
+ */
+
+#ifndef _ARM_SMMU_REGS_H
+#define _ARM_SMMU_REGS_H
+
+/* Configuration registers */
+#define ARM_SMMU_GR0_sCR0		0x0
+#define sCR0_CLIENTPD			(1 << 0)
+#define sCR0_GFRE			(1 << 1)
+#define sCR0_GFIE			(1 << 2)
+#define sCR0_EXIDENABLE			(1 << 3)
+#define sCR0_GCFGFRE			(1 << 4)
+#define sCR0_GCFGFIE			(1 << 5)
+#define sCR0_USFCFG			(1 << 10)
+#define sCR0_VMIDPNE			(1 << 11)
+#define sCR0_PTM			(1 << 12)
+#define sCR0_FB				(1 << 13)
+#define sCR0_VMID16EN			(1 << 31)
+#define sCR0_BSU_SHIFT			14
+#define sCR0_BSU_MASK			0x3
+
+/* Auxiliary Configuration register */
+#define ARM_SMMU_GR0_sACR		0x10
+
+/* Identification registers */
+#define ARM_SMMU_GR0_ID0		0x20
+#define ARM_SMMU_GR0_ID1		0x24
+#define ARM_SMMU_GR0_ID2		0x28
+#define ARM_SMMU_GR0_ID3		0x2c
+#define ARM_SMMU_GR0_ID4		0x30
+#define ARM_SMMU_GR0_ID5		0x34
+#define ARM_SMMU_GR0_ID6		0x38
+#define ARM_SMMU_GR0_ID7		0x3c
+#define ARM_SMMU_GR0_sGFSR		0x48
+#define ARM_SMMU_GR0_sGFSYNR0		0x50
+#define ARM_SMMU_GR0_sGFSYNR1		0x54
+#define ARM_SMMU_GR0_sGFSYNR2		0x58
+
+#define ID0_S1TS			(1 << 30)
+#define ID0_S2TS			(1 << 29)
+#define ID0_NTS				(1 << 28)
+#define ID0_SMS				(1 << 27)
+#define ID0_ATOSNS			(1 << 26)
+#define ID0_PTFS_NO_AARCH32		(1 << 25)
+#define ID0_PTFS_NO_AARCH32S		(1 << 24)
+#define ID0_CTTW			(1 << 14)
+#define ID0_NUMIRPT_SHIFT		16
+#define ID0_NUMIRPT_MASK		0xff
+#define ID0_NUMSIDB_SHIFT		9
+#define ID0_NUMSIDB_MASK		0xf
+#define ID0_EXIDS			(1 << 8)
+#define ID0_NUMSMRG_SHIFT		0
+#define ID0_NUMSMRG_MASK		0xff
+
+#define ID1_PAGESIZE			(1 << 31)
+#define ID1_NUMPAGENDXB_SHIFT		28
+#define ID1_NUMPAGENDXB_MASK		7
+#define ID1_NUMS2CB_SHIFT		16
+#define ID1_NUMS2CB_MASK		0xff
+#define ID1_NUMCB_SHIFT			0
+#define ID1_NUMCB_MASK			0xff
+
+#define ID2_OAS_SHIFT			4
+#define ID2_OAS_MASK			0xf
+#define ID2_IAS_SHIFT			0
+#define ID2_IAS_MASK			0xf
+#define ID2_UBS_SHIFT			8
+#define ID2_UBS_MASK			0xf
+#define ID2_PTFS_4K			(1 << 12)
+#define ID2_PTFS_16K			(1 << 13)
+#define ID2_PTFS_64K			(1 << 14)
+#define ID2_VMID16			(1 << 15)
+
+#define ID7_MAJOR_SHIFT			4
+#define ID7_MAJOR_MASK			0xf
+
+/* Global TLB invalidation */
+#define ARM_SMMU_GR0_TLBIVMID		0x64
+#define ARM_SMMU_GR0_TLBIALLNSNH	0x68
+#define ARM_SMMU_GR0_TLBIALLH		0x6c
+#define ARM_SMMU_GR0_sTLBGSYNC		0x70
+#define ARM_SMMU_GR0_sTLBGSTATUS	0x74
+#define sTLBGSTATUS_GSACTIVE		(1 << 0)
+
+/* Stream mapping registers */
+#define ARM_SMMU_GR0_SMR(n)		(0x800 + ((n) << 2))
+#define SMR_VALID			(1 << 31)
+#define SMR_MASK_SHIFT			16
+#define SMR_ID_SHIFT			0
+
+#define ARM_SMMU_GR0_S2CR(n)		(0xc00 + ((n) << 2))
+#define S2CR_CBNDX_SHIFT		0
+#define S2CR_CBNDX_MASK			0xff
+#define S2CR_EXIDVALID			(1 << 10)
+#define S2CR_TYPE_SHIFT			16
+#define S2CR_TYPE_MASK			0x3
+enum arm_smmu_s2cr_type {
+	S2CR_TYPE_TRANS,
+	S2CR_TYPE_BYPASS,
+	S2CR_TYPE_FAULT,
+};
+
+#define S2CR_PRIVCFG_SHIFT		24
+#define S2CR_PRIVCFG_MASK		0x3
+enum arm_smmu_s2cr_privcfg {
+	S2CR_PRIVCFG_DEFAULT,
+	S2CR_PRIVCFG_DIPAN,
+	S2CR_PRIVCFG_UNPRIV,
+	S2CR_PRIVCFG_PRIV,
+};
+
+/* Context bank attribute registers */
+#define ARM_SMMU_GR1_CBAR(n)		(0x0 + ((n) << 2))
+#define CBAR_VMID_SHIFT			0
+#define CBAR_VMID_MASK			0xff
+#define CBAR_S1_BPSHCFG_SHIFT		8
+#define CBAR_S1_BPSHCFG_MASK		3
+#define CBAR_S1_BPSHCFG_NSH		3
+#define CBAR_S1_MEMATTR_SHIFT		12
+#define CBAR_S1_MEMATTR_MASK		0xf
+#define CBAR_S1_MEMATTR_WB		0xf
+#define CBAR_TYPE_SHIFT			16
+#define CBAR_TYPE_MASK			0x3
+#define CBAR_TYPE_S2_TRANS		(0 << CBAR_TYPE_SHIFT)
+#define CBAR_TYPE_S1_TRANS_S2_BYPASS	(1 << CBAR_TYPE_SHIFT)
+#define CBAR_TYPE_S1_TRANS_S2_FAULT	(2 << CBAR_TYPE_SHIFT)
+#define CBAR_TYPE_S1_TRANS_S2_TRANS	(3 << CBAR_TYPE_SHIFT)
+#define CBAR_IRPTNDX_SHIFT		24
+#define CBAR_IRPTNDX_MASK		0xff
+
+#define ARM_SMMU_GR1_CBA2R(n)		(0x800 + ((n) << 2))
+#define CBA2R_RW64_32BIT		(0 << 0)
+#define CBA2R_RW64_64BIT		(1 << 0)
+#define CBA2R_VMID_SHIFT		16
+#define CBA2R_VMID_MASK			0xffff
+
+#define ARM_SMMU_CB_SCTLR		0x0
+#define ARM_SMMU_CB_ACTLR		0x4
+#define ARM_SMMU_CB_RESUME		0x8
+#define ARM_SMMU_CB_TTBCR2		0x10
+#define ARM_SMMU_CB_TTBR0		0x20
+#define ARM_SMMU_CB_TTBR1		0x28
+#define ARM_SMMU_CB_TTBCR		0x30
+#define ARM_SMMU_CB_CONTEXTIDR		0x34
+#define ARM_SMMU_CB_S1_MAIR0		0x38
+#define ARM_SMMU_CB_S1_MAIR1		0x3c
+#define ARM_SMMU_CB_PAR			0x50
+#define ARM_SMMU_CB_FSR			0x58
+#define ARM_SMMU_CB_FAR			0x60
+#define ARM_SMMU_CB_FSYNR0		0x68
+#define ARM_SMMU_CB_S1_TLBIVA		0x600
+#define ARM_SMMU_CB_S1_TLBIASID		0x610
+#define ARM_SMMU_CB_S1_TLBIVAL		0x620
+#define ARM_SMMU_CB_S2_TLBIIPAS2	0x630
+#define ARM_SMMU_CB_S2_TLBIIPAS2L	0x638
+#define ARM_SMMU_CB_TLBSYNC		0x7f0
+#define ARM_SMMU_CB_TLBSTATUS		0x7f4
+#define ARM_SMMU_CB_ATS1PR		0x800
+#define ARM_SMMU_CB_ATSR		0x8f0
+
+#define SCTLR_S1_ASIDPNE		(1 << 12)
+#define SCTLR_CFCFG			(1 << 7)
+#define SCTLR_CFIE			(1 << 6)
+#define SCTLR_CFRE			(1 << 5)
+#define SCTLR_E				(1 << 4)
+#define SCTLR_AFE			(1 << 2)
+#define SCTLR_TRE			(1 << 1)
+#define SCTLR_M				(1 << 0)
+
+#define CB_PAR_F			(1 << 0)
+
+#define ATSR_ACTIVE			(1 << 0)
+
+#define RESUME_RETRY			(0 << 0)
+#define RESUME_TERMINATE		(1 << 0)
+
+#define TTBCR2_SEP_SHIFT		15
+#define TTBCR2_SEP_UPSTREAM		(0x7 << TTBCR2_SEP_SHIFT)
+#define TTBCR2_AS			(1 << 4)
+
+#define TTBRn_ASID_SHIFT		48
+
+#define FSR_MULTI			(1 << 31)
+#define FSR_SS				(1 << 30)
+#define FSR_UUT				(1 << 8)
+#define FSR_ASF				(1 << 7)
+#define FSR_TLBLKF			(1 << 6)
+#define FSR_TLBMCF			(1 << 5)
+#define FSR_EF				(1 << 4)
+#define FSR_PF				(1 << 3)
+#define FSR_AFF				(1 << 2)
+#define FSR_TF				(1 << 1)
+
+#define FSR_IGN				(FSR_AFF | FSR_ASF | \
+					 FSR_TLBMCF | FSR_TLBLKF)
+#define FSR_FAULT			(FSR_MULTI | FSR_SS | FSR_UUT | \
+					 FSR_EF | FSR_PF | FSR_TF | FSR_IGN)
+
+#define FSYNR0_WNR			(1 << 4)
+
+#endif /* _ARM_SMMU_REGS_H */

drivers/iommu/arm-smmu-v3.c

@@ -2852,9 +2852,15 @@ static int arm_smmu_device_remove(struct platform_device *pdev)
 	struct arm_smmu_device *smmu = platform_get_drvdata(pdev);
 
 	arm_smmu_device_disable(smmu);
+
 	return 0;
 }
 
+static void arm_smmu_device_shutdown(struct platform_device *pdev)
+{
+	arm_smmu_device_remove(pdev);
+}
+
 static const struct of_device_id arm_smmu_of_match[] = {
 	{ .compatible = "arm,smmu-v3", },
 	{ },
@@ -2868,6 +2874,7 @@ static struct platform_driver arm_smmu_driver = {
 	},
 	.probe	= arm_smmu_device_probe,
 	.remove	= arm_smmu_device_remove,
+	.shutdown = arm_smmu_device_shutdown,
 };
 module_platform_driver(arm_smmu_driver);
 

drivers/iommu/dmar.c

@@ -1343,7 +1343,7 @@ void qi_flush_dev_iotlb(struct intel_iommu *iommu, u16 sid, u16 qdep,
 
 	if (mask) {
 		BUG_ON(addr & ((1 << (VTD_PAGE_SHIFT + mask)) - 1));
-		addr |= (1 << (VTD_PAGE_SHIFT + mask - 1)) - 1;
+		addr |= (1ULL << (VTD_PAGE_SHIFT + mask - 1)) - 1;
 		desc.high = QI_DEV_IOTLB_ADDR(addr) | QI_DEV_IOTLB_SIZE;
 	} else
 		desc.high = QI_DEV_IOTLB_ADDR(addr);

drivers/iommu/exynos-iommu.c

@@ -54,10 +54,6 @@ typedef u32 sysmmu_pte_t;
 #define lv2ent_small(pent) ((*(pent) & 2) == 2)
 #define lv2ent_large(pent) ((*(pent) & 3) == 1)
 
-#ifdef CONFIG_BIG_ENDIAN
-#warning "revisit driver if we can enable big-endian ptes"
-#endif
-
 /*
  * v1.x - v3.x SYSMMU supports 32bit physical and 32bit virtual address spaces
  * v5.0 introduced support for 36bit physical address space by shifting
@@ -569,7 +565,7 @@ static void sysmmu_tlb_invalidate_entry(struct sysmmu_drvdata *data,
 	spin_unlock_irqrestore(&data->lock, flags);
 }
 
-static struct iommu_ops exynos_iommu_ops;
+static const struct iommu_ops exynos_iommu_ops;
 
 static int __init exynos_sysmmu_probe(struct platform_device *pdev)
 {
@@ -659,6 +655,13 @@ static int __init exynos_sysmmu_probe(struct platform_device *pdev)
 		}
 	}
 
+	/*
+	 * use the first registered sysmmu device for performing
+	 * dma mapping operations on iommu page tables (cpu cache flush)
+	 */
+	if (!dma_dev)
+		dma_dev = &pdev->dev;
+
 	pm_runtime_enable(dev);
 
 	return 0;
@@ -1323,7 +1326,7 @@ static int exynos_iommu_of_xlate(struct device *dev,
 	return 0;
 }
 
-static struct iommu_ops exynos_iommu_ops = {
+static const struct iommu_ops exynos_iommu_ops = {
 	.domain_alloc = exynos_iommu_domain_alloc,
 	.domain_free = exynos_iommu_domain_free,
 	.attach_dev = exynos_iommu_attach_device,
@@ -1339,8 +1342,6 @@ static struct iommu_ops exynos_iommu_ops = {
 	.of_xlate = exynos_iommu_of_xlate,
 };
 
-static bool init_done;
-
 static int __init exynos_iommu_init(void)
 {
 	int ret;
@@ -1373,8 +1374,6 @@ static int __init exynos_iommu_init(void)
 		goto err_set_iommu;
 	}
 
-	init_done = true;
-
 	return 0;
 err_set_iommu:
 	kmem_cache_free(lv2table_kmem_cache, zero_lv2_table);
@@ -1384,27 +1383,6 @@ static int __init exynos_iommu_init(void)
 	kmem_cache_destroy(lv2table_kmem_cache);
 	return ret;
 }
+core_initcall(exynos_iommu_init);
 
-static int __init exynos_iommu_of_setup(struct device_node *np)
-{
-	struct platform_device *pdev;
-
-	if (!init_done)
-		exynos_iommu_init();
-
-	pdev = of_platform_device_create(np, NULL, platform_bus_type.dev_root);
-	if (!pdev)
-		return -ENODEV;
-
-	/*
-	 * use the first registered sysmmu device for performing
-	 * dma mapping operations on iommu page tables (cpu cache flush)
-	 */
-	if (!dma_dev)
-		dma_dev = &pdev->dev;
-
-	return 0;
-}
-
-IOMMU_OF_DECLARE(exynos_iommu_of, "samsung,exynos-sysmmu",
-		 exynos_iommu_of_setup);
+IOMMU_OF_DECLARE(exynos_iommu_of, "samsung,exynos-sysmmu", NULL);

drivers/iommu/fsl_pamu.c

@@ -42,6 +42,8 @@ struct pamu_isr_data {
 static struct paace *ppaact;
 static struct paace *spaact;
 
+static bool probed;			/* Has PAMU been probed? */
+
 /*
  * Table for matching compatible strings, for device tree
  * guts node, for QorIQ SOCs.
@@ -530,8 +532,8 @@ u32 get_stash_id(u32 stash_dest_hint, u32 vcpu)
 		if (node) {
 			prop = of_get_property(node, "cache-stash-id", NULL);
 			if (!prop) {
-				pr_debug("missing cache-stash-id at %s\n",
-					 node->full_name);
+				pr_debug("missing cache-stash-id at %pOF\n",
+					 node);
 				of_node_put(node);
 				return ~(u32)0;
 			}
@@ -557,8 +559,8 @@ u32 get_stash_id(u32 stash_dest_hint, u32 vcpu)
 		if (stash_dest_hint == cache_level) {
 			prop = of_get_property(node, "cache-stash-id", NULL);
 			if (!prop) {
-				pr_debug("missing cache-stash-id at %s\n",
-					 node->full_name);
+				pr_debug("missing cache-stash-id at %pOF\n",
+					 node);
 				of_node_put(node);
 				return ~(u32)0;
 			}
@@ -568,8 +570,7 @@ u32 get_stash_id(u32 stash_dest_hint, u32 vcpu)
 
 		prop = of_get_property(node, "next-level-cache", NULL);
 		if (!prop) {
-			pr_debug("can't find next-level-cache at %s\n",
-				 node->full_name);
+			pr_debug("can't find next-level-cache at %pOF\n", node);
 			of_node_put(node);
 			return ~(u32)0;  /* can't traverse any further */
 		}
@@ -1033,6 +1034,9 @@ static int fsl_pamu_probe(struct platform_device *pdev)
 	 * NOTE : All PAMUs share the same LIODN tables.
 	 */
 
+	if (WARN_ON(probed))
+		return -EBUSY;
+
 	pamu_regs = of_iomap(dev->of_node, 0);
 	if (!pamu_regs) {
 		dev_err(dev, "ioremap of PAMU node failed\n");
@@ -1063,8 +1067,7 @@ static int fsl_pamu_probe(struct platform_device *pdev)
 
 	guts_node = of_find_matching_node(NULL, guts_device_ids);
 	if (!guts_node) {
-		dev_err(dev, "could not find GUTS node %s\n",
-			dev->of_node->full_name);
+		dev_err(dev, "could not find GUTS node %pOF\n", dev->of_node);
 		ret = -ENODEV;
 		goto error;
 	}
@@ -1172,6 +1175,8 @@ static int fsl_pamu_probe(struct platform_device *pdev)
 
 	setup_liodns();
 
+	probed = true;
+
 	return 0;
 
 error_genpool:
@@ -1246,8 +1251,7 @@ static __init int fsl_pamu_init(void)
 
 	pdev = platform_device_alloc("fsl-of-pamu", 0);
 	if (!pdev) {
-		pr_err("could not allocate device %s\n",
-		       np->full_name);
+		pr_err("could not allocate device %pOF\n", np);
 		ret = -ENOMEM;
 		goto error_device_alloc;
 	}
@@ -1259,8 +1263,7 @@ static __init int fsl_pamu_init(void)
 
 	ret = platform_device_add(pdev);
 	if (ret) {
-		pr_err("could not add device %s (err=%i)\n",
-		       np->full_name, ret);
+		pr_err("could not add device %pOF (err=%i)\n", np, ret);
 		goto error_device_add;
 	}
 

drivers/iommu/fsl_pamu_domain.c

@@ -33,6 +33,8 @@ static struct kmem_cache *fsl_pamu_domain_cache;
 static struct kmem_cache *iommu_devinfo_cache;
 static DEFINE_SPINLOCK(device_domain_lock);
 
+struct iommu_device pamu_iommu;	/* IOMMU core code handle */
+
 static struct fsl_dma_domain *to_fsl_dma_domain(struct iommu_domain *dom)
 {
 	return container_of(dom, struct fsl_dma_domain, iommu_domain);
@@ -619,8 +621,8 @@ static int handle_attach_device(struct fsl_dma_domain *dma_domain,
 	for (i = 0; i < num; i++) {
 		/* Ensure that LIODN value is valid */
 		if (liodn[i] >= PAACE_NUMBER_ENTRIES) {
-			pr_debug("Invalid liodn %d, attach device failed for %s\n",
-				 liodn[i], dev->of_node->full_name);
+			pr_debug("Invalid liodn %d, attach device failed for %pOF\n",
+				 liodn[i], dev->of_node);
 			ret = -EINVAL;
 			break;
 		}
@@ -684,8 +686,7 @@ static int fsl_pamu_attach_device(struct iommu_domain *domain,
 		liodn_cnt = len / sizeof(u32);
 		ret = handle_attach_device(dma_domain, dev, liodn, liodn_cnt);
 	} else {
-		pr_debug("missing fsl,liodn property at %s\n",
-			 dev->of_node->full_name);
+		pr_debug("missing fsl,liodn property at %pOF\n", dev->of_node);
 		ret = -EINVAL;
 	}
 
@@ -720,8 +721,7 @@ static void fsl_pamu_detach_device(struct iommu_domain *domain,
 	if (prop)
 		detach_device(dev, dma_domain);
 	else
-		pr_debug("missing fsl,liodn property at %s\n",
-			 dev->of_node->full_name);
+		pr_debug("missing fsl,liodn property at %pOF\n", dev->of_node);
 }
 
 static  int configure_domain_geometry(struct iommu_domain *domain, void *data)
@@ -983,11 +983,14 @@ static int fsl_pamu_add_device(struct device *dev)
 
 	iommu_group_put(group);
 
+	iommu_device_link(&pamu_iommu, dev);
+
 	return 0;
 }
 
 static void fsl_pamu_remove_device(struct device *dev)
 {
+	iommu_device_unlink(&pamu_iommu, dev);
 	iommu_group_remove_device(dev);
 }
 
@@ -1073,6 +1076,19 @@ int __init pamu_domain_init(void)
 	if (ret)
 		return ret;
 
+	ret = iommu_device_sysfs_add(&pamu_iommu, NULL, NULL, "iommu0");
+	if (ret)
+		return ret;
+
+	iommu_device_set_ops(&pamu_iommu, &fsl_pamu_ops);
+
+	ret = iommu_device_register(&pamu_iommu);
+	if (ret) {
+		iommu_device_sysfs_remove(&pamu_iommu);
+		pr_err("Can't register iommu device\n");
+		return ret;
+	}
+
 	bus_set_iommu(&platform_bus_type, &fsl_pamu_ops);
 	bus_set_iommu(&pci_bus_type, &fsl_pamu_ops);
 

drivers/iommu/intel-svm.c

@@ -24,6 +24,7 @@
 #include <linux/pci-ats.h>
 #include <linux/dmar.h>
 #include <linux/interrupt.h>
+#include <asm/page.h>
 
 static irqreturn_t prq_event_thread(int irq, void *d);
 
@@ -555,6 +556,14 @@ static bool access_error(struct vm_area_struct *vma, struct page_req_dsc *req)
 	return (requested & ~vma->vm_flags) != 0;
 }
 
+static bool is_canonical_address(u64 addr)
+{
+	int shift = 64 - (__VIRTUAL_MASK_SHIFT + 1);
+	long saddr = (long) addr;
+
+	return (((saddr << shift) >> shift) == saddr);
+}
+
 static irqreturn_t prq_event_thread(int irq, void *d)
 {
 	struct intel_iommu *iommu = d;
@@ -612,6 +621,11 @@ static irqreturn_t prq_event_thread(int irq, void *d)
 		/* If the mm is already defunct, don't handle faults. */
 		if (!mmget_not_zero(svm->mm))
 			goto bad_req;
+
+		/* If address is not canonical, return invalid response */
+		if (!is_canonical_address(address))
+			goto bad_req;
+
 		down_read(&svm->mm->mmap_sem);
 		vma = find_extend_vma(svm->mm, address);
 		if (!vma || address < vma->vm_start)

drivers/iommu/iommu.c

@@ -527,6 +527,8 @@ static int iommu_group_create_direct_mappings(struct iommu_group *group,
 
 	}
 
+	iommu_flush_tlb_all(domain);
+
 out:
 	iommu_put_resv_regions(dev, &mappings);
 
@@ -1005,11 +1007,10 @@ struct iommu_group *iommu_group_get_for_dev(struct device *dev)
 	if (group)
 		return group;
 
-	group = ERR_PTR(-EINVAL);
-
-	if (ops && ops->device_group)
-		group = ops->device_group(dev);
+	if (!ops)
+		return ERR_PTR(-EINVAL);
 
+	group = ops->device_group(dev);
 	if (WARN_ON_ONCE(group == NULL))
 		return ERR_PTR(-EINVAL);
 
@@ -1283,6 +1284,10 @@ static int __iommu_attach_device(struct iommu_domain *domain,
 				 struct device *dev)
 {
 	int ret;
+	if ((domain->ops->is_attach_deferred != NULL) &&
+	    domain->ops->is_attach_deferred(domain, dev))
+		return 0;
+
 	if (unlikely(domain->ops->attach_dev == NULL))
 		return -ENODEV;
 
@@ -1298,12 +1303,8 @@ int iommu_attach_device(struct iommu_domain *domain, struct device *dev)
 	int ret;
 
 	group = iommu_group_get(dev);
-	/* FIXME: Remove this when groups a mandatory for iommu drivers */
-	if (group == NULL)
-		return __iommu_attach_device(domain, dev);
-
 	/*
-	 * We have a group - lock it to make sure the device-count doesn't
+	 * Lock the group to make sure the device-count doesn't
 	 * change while we are attaching
 	 */
 	mutex_lock(&group->mutex);
@@ -1324,6 +1325,10 @@ EXPORT_SYMBOL_GPL(iommu_attach_device);
 static void __iommu_detach_device(struct iommu_domain *domain,
 				  struct device *dev)
 {
+	if ((domain->ops->is_attach_deferred != NULL) &&
+	    domain->ops->is_attach_deferred(domain, dev))
+		return;
+
 	if (unlikely(domain->ops->detach_dev == NULL))
 		return;
 
@@ -1336,9 +1341,6 @@ void iommu_detach_device(struct iommu_domain *domain, struct device *dev)
 	struct iommu_group *group;
 
 	group = iommu_group_get(dev);
-	/* FIXME: Remove this when groups a mandatory for iommu drivers */
-	if (group == NULL)
-		return __iommu_detach_device(domain, dev);
 
 	mutex_lock(&group->mutex);
 	if (iommu_group_device_count(group) != 1) {
@@ -1360,8 +1362,7 @@ struct iommu_domain *iommu_get_domain_for_dev(struct device *dev)
 	struct iommu_group *group;
 
 	group = iommu_group_get(dev);
-	/* FIXME: Remove this when groups a mandatory for iommu drivers */
-	if (group == NULL)
+	if (!group)
 		return NULL;
 
 	domain = group->domain;
@@ -1556,13 +1557,16 @@ int iommu_map(struct iommu_domain *domain, unsigned long iova,
 }
 EXPORT_SYMBOL_GPL(iommu_map);
 
-size_t iommu_unmap(struct iommu_domain *domain, unsigned long iova, size_t size)
+static size_t __iommu_unmap(struct iommu_domain *domain,
+			    unsigned long iova, size_t size,
+			    bool sync)
 {
+	const struct iommu_ops *ops = domain->ops;
 	size_t unmapped_page, unmapped = 0;
-	unsigned int min_pagesz;
 	unsigned long orig_iova = iova;
+	unsigned int min_pagesz;
 
-	if (unlikely(domain->ops->unmap == NULL ||
+	if (unlikely(ops->unmap == NULL ||
 		     domain->pgsize_bitmap == 0UL))
 		return -ENODEV;
 
@@ -1592,10 +1596,13 @@ size_t iommu_unmap(struct iommu_domain *domain, unsigned long iova, size_t size)
 	while (unmapped < size) {
 		size_t pgsize = iommu_pgsize(domain, iova, size - unmapped);
 
-		unmapped_page = domain->ops->unmap(domain, iova, pgsize);
+		unmapped_page = ops->unmap(domain, iova, pgsize);
 		if (!unmapped_page)
 			break;
 
+		if (sync && ops->iotlb_range_add)
+			ops->iotlb_range_add(domain, iova, pgsize);
+
 		pr_debug("unmapped: iova 0x%lx size 0x%zx\n",
 			 iova, unmapped_page);
 
@@ -1603,11 +1610,27 @@ size_t iommu_unmap(struct iommu_domain *domain, unsigned long iova, size_t size)
 		unmapped += unmapped_page;
 	}
 
+	if (sync && ops->iotlb_sync)
+		ops->iotlb_sync(domain);
+
 	trace_unmap(orig_iova, size, unmapped);
 	return unmapped;
 }
+
+size_t iommu_unmap(struct iommu_domain *domain,
+		   unsigned long iova, size_t size)
+{
+	return __iommu_unmap(domain, iova, size, true);
+}
 EXPORT_SYMBOL_GPL(iommu_unmap);
 
+size_t iommu_unmap_fast(struct iommu_domain *domain,
+			unsigned long iova, size_t size)
+{
+	return __iommu_unmap(domain, iova, size, false);
+}
+EXPORT_SYMBOL_GPL(iommu_unmap_fast);
+
 size_t default_iommu_map_sg(struct iommu_domain *domain, unsigned long iova,
 			 struct scatterlist *sg, unsigned int nents, int prot)
 {

drivers/iommu/iova.c

@@ -32,6 +32,8 @@ static unsigned long iova_rcache_get(struct iova_domain *iovad,
 				     unsigned long limit_pfn);
 static void init_iova_rcaches(struct iova_domain *iovad);
 static void free_iova_rcaches(struct iova_domain *iovad);
+static void fq_destroy_all_entries(struct iova_domain *iovad);
+static void fq_flush_timeout(unsigned long data);
 
 void
 init_iova_domain(struct iova_domain *iovad, unsigned long granule,
@@ -50,10 +52,61 @@ init_iova_domain(struct iova_domain *iovad, unsigned long granule,
 	iovad->granule = granule;
 	iovad->start_pfn = start_pfn;
 	iovad->dma_32bit_pfn = pfn_32bit + 1;
+	iovad->flush_cb = NULL;
+	iovad->fq = NULL;
 	init_iova_rcaches(iovad);
 }
 EXPORT_SYMBOL_GPL(init_iova_domain);
 
+static void free_iova_flush_queue(struct iova_domain *iovad)
+{
+	if (!iovad->fq)
+		return;
+
+	if (timer_pending(&iovad->fq_timer))
+		del_timer(&iovad->fq_timer);
+
+	fq_destroy_all_entries(iovad);
+
+	free_percpu(iovad->fq);
+
+	iovad->fq         = NULL;
+	iovad->flush_cb   = NULL;
+	iovad->entry_dtor = NULL;
+}
+
+int init_iova_flush_queue(struct iova_domain *iovad,
+			  iova_flush_cb flush_cb, iova_entry_dtor entry_dtor)
+{
+	int cpu;
+
+	atomic64_set(&iovad->fq_flush_start_cnt,  0);
+	atomic64_set(&iovad->fq_flush_finish_cnt, 0);
+
+	iovad->fq = alloc_percpu(struct iova_fq);
+	if (!iovad->fq)
+		return -ENOMEM;
+
+	iovad->flush_cb   = flush_cb;
+	iovad->entry_dtor = entry_dtor;
+
+	for_each_possible_cpu(cpu) {
+		struct iova_fq *fq;
+
+		fq = per_cpu_ptr(iovad->fq, cpu);
+		fq->head = 0;
+		fq->tail = 0;
+
+		spin_lock_init(&fq->lock);
+	}
+
+	setup_timer(&iovad->fq_timer, fq_flush_timeout, (unsigned long)iovad);
+	atomic_set(&iovad->fq_timer_on, 0);
+
+	return 0;
+}
+EXPORT_SYMBOL_GPL(init_iova_flush_queue);
+
 static struct rb_node *
 __get_cached_rbnode(struct iova_domain *iovad, unsigned long *limit_pfn)
 {
@@ -423,6 +476,135 @@ free_iova_fast(struct iova_domain *iovad, unsigned long pfn, unsigned long size)
 }
 EXPORT_SYMBOL_GPL(free_iova_fast);
 
+#define fq_ring_for_each(i, fq) \
+	for ((i) = (fq)->head; (i) != (fq)->tail; (i) = ((i) + 1) % IOVA_FQ_SIZE)
+
+static inline bool fq_full(struct iova_fq *fq)
+{
+	assert_spin_locked(&fq->lock);
+	return (((fq->tail + 1) % IOVA_FQ_SIZE) == fq->head);
+}
+
+static inline unsigned fq_ring_add(struct iova_fq *fq)
+{
+	unsigned idx = fq->tail;
+
+	assert_spin_locked(&fq->lock);
+
+	fq->tail = (idx + 1) % IOVA_FQ_SIZE;
+
+	return idx;
+}
+
+static void fq_ring_free(struct iova_domain *iovad, struct iova_fq *fq)
+{
+	u64 counter = atomic64_read(&iovad->fq_flush_finish_cnt);
+	unsigned idx;
+
+	assert_spin_locked(&fq->lock);
+
+	fq_ring_for_each(idx, fq) {
+
+		if (fq->entries[idx].counter >= counter)
+			break;
+
+		if (iovad->entry_dtor)
+			iovad->entry_dtor(fq->entries[idx].data);
+
+		free_iova_fast(iovad,
+			       fq->entries[idx].iova_pfn,
+			       fq->entries[idx].pages);
+
+		fq->head = (fq->head + 1) % IOVA_FQ_SIZE;
+	}
+}
+
+static void iova_domain_flush(struct iova_domain *iovad)
+{
+	atomic64_inc(&iovad->fq_flush_start_cnt);
+	iovad->flush_cb(iovad);
+	atomic64_inc(&iovad->fq_flush_finish_cnt);
+}
+
+static void fq_destroy_all_entries(struct iova_domain *iovad)
+{
+	int cpu;
+
+	/*
+	 * This code runs when the iova_domain is being detroyed, so don't
+	 * bother to free iovas, just call the entry_dtor on all remaining
+	 * entries.
+	 */
+	if (!iovad->entry_dtor)
+		return;
+
+	for_each_possible_cpu(cpu) {
+		struct iova_fq *fq = per_cpu_ptr(iovad->fq, cpu);
+		int idx;
+
+		fq_ring_for_each(idx, fq)
+			iovad->entry_dtor(fq->entries[idx].data);
+	}
+}
+
+static void fq_flush_timeout(unsigned long data)
+{
+	struct iova_domain *iovad = (struct iova_domain *)data;
+	int cpu;
+
+	atomic_set(&iovad->fq_timer_on, 0);
+	iova_domain_flush(iovad);
+
+	for_each_possible_cpu(cpu) {
+		unsigned long flags;
+		struct iova_fq *fq;
+
+		fq = per_cpu_ptr(iovad->fq, cpu);
+		spin_lock_irqsave(&fq->lock, flags);
+		fq_ring_free(iovad, fq);
+		spin_unlock_irqrestore(&fq->lock, flags);
+	}
+}
+
+void queue_iova(struct iova_domain *iovad,
+		unsigned long pfn, unsigned long pages,
+		unsigned long data)
+{
+	struct iova_fq *fq = get_cpu_ptr(iovad->fq);
+	unsigned long flags;
+	unsigned idx;
+
+	spin_lock_irqsave(&fq->lock, flags);
+
+	/*
+	 * First remove all entries from the flush queue that have already been
+	 * flushed out on another CPU. This makes the fq_full() check below less
+	 * likely to be true.
+	 */
+	fq_ring_free(iovad, fq);
+
+	if (fq_full(fq)) {
+		iova_domain_flush(iovad);
+		fq_ring_free(iovad, fq);
+	}
+
+	idx = fq_ring_add(fq);
+
+	fq->entries[idx].iova_pfn = pfn;
+	fq->entries[idx].pages    = pages;
+	fq->entries[idx].data     = data;
+	fq->entries[idx].counter  = atomic64_read(&iovad->fq_flush_start_cnt);
+
+	spin_unlock_irqrestore(&fq->lock, flags);
+
+	if (atomic_cmpxchg(&iovad->fq_timer_on, 0, 1) == 0)
+		mod_timer(&iovad->fq_timer,
+			  jiffies + msecs_to_jiffies(IOVA_FQ_TIMEOUT));
+
+	put_cpu_ptr(iovad->fq);
+}
+EXPORT_SYMBOL_GPL(queue_iova);
+
 /**
  * put_iova_domain - destroys the iova doamin
  * @iovad: - iova domain in question.
@@ -433,6 +615,7 @@ void put_iova_domain(struct iova_domain *iovad)
 	struct rb_node *node;
 	unsigned long flags;
 
+	free_iova_flush_queue(iovad);
 	free_iova_rcaches(iovad);
 	spin_lock_irqsave(&iovad->iova_rbtree_lock, flags);
 	node = rb_first(&iovad->rbroot);

drivers/iommu/msm_iommu.c

@@ -393,6 +393,7 @@ static struct msm_iommu_dev *find_iommu_for_dev(struct device *dev)
 static int msm_iommu_add_device(struct device *dev)
 {
 	struct msm_iommu_dev *iommu;
+	struct iommu_group *group;
 	unsigned long flags;
 	int ret = 0;
 
@@ -406,7 +407,16 @@ static int msm_iommu_add_device(struct device *dev)
 
 	spin_unlock_irqrestore(&msm_iommu_lock, flags);
 
-	return ret;
+	if (ret)
+		return ret;
+
+	group = iommu_group_get_for_dev(dev);
+	if (IS_ERR(group))
+		return PTR_ERR(group);
+
+	iommu_group_put(group);
+
+	return 0;
 }
 
 static void msm_iommu_remove_device(struct device *dev)
@@ -421,6 +431,8 @@ static void msm_iommu_remove_device(struct device *dev)
 		iommu_device_unlink(&iommu->iommu, dev);
 
 	spin_unlock_irqrestore(&msm_iommu_lock, flags);
+
+	iommu_group_remove_device(dev);
 }
 
 static int msm_iommu_attach_dev(struct iommu_domain *domain, struct device *dev)
@@ -700,6 +712,7 @@ static struct iommu_ops msm_iommu_ops = {
 	.iova_to_phys = msm_iommu_iova_to_phys,
 	.add_device = msm_iommu_add_device,
 	.remove_device = msm_iommu_remove_device,
+	.device_group = generic_device_group,
 	.pgsize_bitmap = MSM_IOMMU_PGSIZES,
 	.of_xlate = qcom_iommu_of_xlate,
 };

drivers/iommu/mtk_iommu.h

@@ -34,6 +34,12 @@ struct mtk_iommu_suspend_reg {
 	u32				int_main_control;
 };
 
+enum mtk_iommu_plat {
+	M4U_MT2701,
+	M4U_MT2712,
+	M4U_MT8173,
+};
+
 struct mtk_iommu_domain;
 
 struct mtk_iommu_data {
@@ -50,6 +56,9 @@ struct mtk_iommu_data {
 	bool				tlb_flush_active;
 
 	struct iommu_device		iommu;
+	enum mtk_iommu_plat		m4u_plat;
+
+	struct list_head		list;
 };
 
 static inline int compare_of(struct device *dev, void *data)

drivers/iommu/of_iommu.c

@@ -25,6 +25,8 @@
 #include <linux/of_pci.h>
 #include <linux/slab.h>
 
+#define NO_IOMMU	1
+
 static const struct of_device_id __iommu_of_table_sentinel
 	__used __section(__iommu_of_table_end);
 
@@ -109,8 +111,8 @@ static bool of_iommu_driver_present(struct device_node *np)
 	return of_match_node(&__iommu_of_table, np);
 }
 
-static const struct iommu_ops
-*of_iommu_xlate(struct device *dev, struct of_phandle_args *iommu_spec)
+static int of_iommu_xlate(struct device *dev,
+			  struct of_phandle_args *iommu_spec)
 {
 	const struct iommu_ops *ops;
 	struct fwnode_handle *fwnode = &iommu_spec->np->fwnode;
@@ -120,95 +122,53 @@ static const struct iommu_ops
 	if ((ops && !ops->of_xlate) ||
 	    !of_device_is_available(iommu_spec->np) ||
 	    (!ops && !of_iommu_driver_present(iommu_spec->np)))
-		return NULL;
+		return NO_IOMMU;
 
 	err = iommu_fwspec_init(dev, &iommu_spec->np->fwnode, ops);
 	if (err)
-		return ERR_PTR(err);
+		return err;
 	/*
 	 * The otherwise-empty fwspec handily serves to indicate the specific
 	 * IOMMU device we're waiting for, which will be useful if we ever get
 	 * a proper probe-ordering dependency mechanism in future.
 	 */
 	if (!ops)
-		return ERR_PTR(-EPROBE_DEFER);
-
-	err = ops->of_xlate(dev, iommu_spec);
-	if (err)
-		return ERR_PTR(err);
+		return -EPROBE_DEFER;
 
-	return ops;
+	return ops->of_xlate(dev, iommu_spec);
 }
 
-static int __get_pci_rid(struct pci_dev *pdev, u16 alias, void *data)
-{
-	struct of_phandle_args *iommu_spec = data;
-
-	iommu_spec->args[0] = alias;
-	return iommu_spec->np == pdev->bus->dev.of_node;
-}
+struct of_pci_iommu_alias_info {
+	struct device *dev;
+	struct device_node *np;
+};
 
-static const struct iommu_ops
-*of_pci_iommu_init(struct pci_dev *pdev, struct device_node *bridge_np)
+static int of_pci_iommu_init(struct pci_dev *pdev, u16 alias, void *data)
 {
-	const struct iommu_ops *ops;
-	struct of_phandle_args iommu_spec;
+	struct of_pci_iommu_alias_info *info = data;
+	struct of_phandle_args iommu_spec = { .args_count = 1 };
 	int err;
 
-	/*
-	 * Start by tracing the RID alias down the PCI topology as
-	 * far as the host bridge whose OF node we have...
-	 * (we're not even attempting to handle multi-alias devices yet)
-	 */
-	iommu_spec.args_count = 1;
-	iommu_spec.np = bridge_np;
-	pci_for_each_dma_alias(pdev, __get_pci_rid, &iommu_spec);
-	/*
-	 * ...then find out what that becomes once it escapes the PCI
-	 * bus into the system beyond, and which IOMMU it ends up at.
-	 */
-	iommu_spec.np = NULL;
-	err = of_pci_map_rid(bridge_np, iommu_spec.args[0], "iommu-map",
+	err = of_pci_map_rid(info->np, alias, "iommu-map",
 			     "iommu-map-mask", &iommu_spec.np,
 			     iommu_spec.args);
 	if (err)
-		return err == -ENODEV ? NULL : ERR_PTR(err);
-
-	ops = of_iommu_xlate(&pdev->dev, &iommu_spec);
+		return err == -ENODEV ? NO_IOMMU : err;
 
+	err = of_iommu_xlate(info->dev, &iommu_spec);
 	of_node_put(iommu_spec.np);
-	return ops;
-}
-
-static const struct iommu_ops
-*of_platform_iommu_init(struct device *dev, struct device_node *np)
-{
-	struct of_phandle_args iommu_spec;
-	const struct iommu_ops *ops = NULL;
-	int idx = 0;
-
-	/*
-	 * We don't currently walk up the tree looking for a parent IOMMU.
-	 * See the `Notes:' section of
-	 * Documentation/devicetree/bindings/iommu/iommu.txt
-	 */
-	while (!of_parse_phandle_with_args(np, "iommus", "#iommu-cells",
-					   idx, &iommu_spec)) {
-		ops = of_iommu_xlate(dev, &iommu_spec);
-		of_node_put(iommu_spec.np);
-		idx++;
-		if (IS_ERR_OR_NULL(ops))
-			break;
-	}
+	if (err)
+		return err;
 
-	return ops;
+	return info->np == pdev->bus->dev.of_node;
 }
 
 const struct iommu_ops *of_iommu_configure(struct device *dev,
 					   struct device_node *master_np)
 {
-	const struct iommu_ops *ops;
+	const struct iommu_ops *ops = NULL;
 	struct iommu_fwspec *fwspec = dev->iommu_fwspec;
+	int err = NO_IOMMU;
 
 	if (!master_np)
 		return NULL;
@@ -221,25 +181,54 @@ const struct iommu_ops *of_iommu_configure(struct device *dev,
 		iommu_fwspec_free(dev);
 	}
 
-	if (dev_is_pci(dev))
-		ops = of_pci_iommu_init(to_pci_dev(dev), master_np);
-	else
-		ops = of_platform_iommu_init(dev, master_np);
+	/*
+	 * We don't currently walk up the tree looking for a parent IOMMU.
+	 * See the `Notes:' section of
+	 * Documentation/devicetree/bindings/iommu/iommu.txt
+	 */
+	if (dev_is_pci(dev)) {
+		struct of_pci_iommu_alias_info info = {
+			.dev = dev,
+			.np = master_np,
+		};
+
+		err = pci_for_each_dma_alias(to_pci_dev(dev),
+					     of_pci_iommu_init, &info);
+	} else {
+		struct of_phandle_args iommu_spec;
+		int idx = 0;
+
+		while (!of_parse_phandle_with_args(master_np, "iommus",
+						   "#iommu-cells",
+						   idx, &iommu_spec)) {
+			err = of_iommu_xlate(dev, &iommu_spec);
+			of_node_put(iommu_spec.np);
+			idx++;
+			if (err)
+				break;
+		}
+	}
+
+	/*
+	 * Two success conditions can be represented by non-negative err here:
+	 * >0 : there is no IOMMU, or one was unavailable for non-fatal reasons
+	 *  0 : we found an IOMMU, and dev->fwspec is initialised appropriately
+	 * <0 : any actual error
+	 */
+	if (!err)
+		ops = dev->iommu_fwspec->ops;
 	/*
 	 * If we have reason to believe the IOMMU driver missed the initial
 	 * add_device callback for dev, replay it to get things in order.
 	 */
-	if (!IS_ERR_OR_NULL(ops) && ops->add_device &&
-	    dev->bus && !dev->iommu_group) {
-		int err = ops->add_device(dev);
-
-		if (err)
-			ops = ERR_PTR(err);
-	}
+	if (ops && ops->add_device && dev->bus && !dev->iommu_group)
+		err = ops->add_device(dev);
 
 	/* Ignore all other errors apart from EPROBE_DEFER */
-	if (IS_ERR(ops) && (PTR_ERR(ops) != -EPROBE_DEFER)) {
-		dev_dbg(dev, "Adding to IOMMU failed: %ld\n", PTR_ERR(ops));
+	if (err == -EPROBE_DEFER) {
+		ops = ERR_PTR(err);
+	} else if (err < 0) {
+		dev_dbg(dev, "Adding to IOMMU failed: %d\n", err);
 		ops = NULL;
 	}
 
@@ -255,8 +244,7 @@ static int __init of_iommu_init(void)
 		const of_iommu_init_fn init_fn = match->data;
 
 		if (init_fn && init_fn(np))
-			pr_err("Failed to initialise IOMMU %s\n",
-				of_node_full_name(np));
+			pr_err("Failed to initialise IOMMU %pOF\n", np);
 	}
 
 	return 0;

drivers/iommu/omap-iommu.c

@@ -11,6 +11,7 @@
  * published by the Free Software Foundation.
  */
 
+#include <linux/dma-mapping.h>
 #include <linux/err.h>
 #include <linux/slab.h>
 #include <linux/interrupt.h>
@@ -29,8 +30,6 @@
 #include <linux/regmap.h>
 #include <linux/mfd/syscon.h>
 
-#include <asm/cacheflush.h>
-
 #include <linux/platform_data/iommu-omap.h>
 
 #include "omap-iopgtable.h"
@@ -454,36 +453,35 @@ static void flush_iotlb_all(struct omap_iommu *obj)
 /*
  *	H/W pagetable operations
  */
-static void flush_iopgd_range(u32 *first, u32 *last)
+static void flush_iopte_range(struct device *dev, dma_addr_t dma,
+			      unsigned long offset, int num_entries)
 {
-	/* FIXME: L2 cache should be taken care of if it exists */
-	do {
-		asm("mcr	p15, 0, %0, c7, c10, 1 @ flush_pgd"
-		    : : "r" (first));
-		first += L1_CACHE_BYTES / sizeof(*first);
-	} while (first <= last);
-}
+	size_t size = num_entries * sizeof(u32);
 
-static void flush_iopte_range(u32 *first, u32 *last)
-{
-	/* FIXME: L2 cache should be taken care of if it exists */
-	do {
-		asm("mcr	p15, 0, %0, c7, c10, 1 @ flush_pte"
-		    : : "r" (first));
-		first += L1_CACHE_BYTES / sizeof(*first);
-	} while (first <= last);
+	dma_sync_single_range_for_device(dev, dma, offset, size, DMA_TO_DEVICE);
 }
 
-static void iopte_free(u32 *iopte)
+static void iopte_free(struct omap_iommu *obj, u32 *iopte, bool dma_valid)
 {
+	dma_addr_t pt_dma;
+
 	/* Note: freed iopte's must be clean ready for re-use */
-	if (iopte)
+	if (iopte) {
+		if (dma_valid) {
+			pt_dma = virt_to_phys(iopte);
+			dma_unmap_single(obj->dev, pt_dma, IOPTE_TABLE_SIZE,
+					 DMA_TO_DEVICE);
+		}
+
 		kmem_cache_free(iopte_cachep, iopte);
+	}
 }
 
-static u32 *iopte_alloc(struct omap_iommu *obj, u32 *iopgd, u32 da)
+static u32 *iopte_alloc(struct omap_iommu *obj, u32 *iopgd,
+			dma_addr_t *pt_dma, u32 da)
 {
 	u32 *iopte;
+	unsigned long offset = iopgd_index(da) * sizeof(da);
 
 	/* a table has already existed */
 	if (*iopgd)
@@ -500,18 +498,38 @@ static u32 *iopte_alloc(struct omap_iommu *obj, u32 *iopgd, u32 da)
 		if (!iopte)
 			return ERR_PTR(-ENOMEM);
 
+		*pt_dma = dma_map_single(obj->dev, iopte, IOPTE_TABLE_SIZE,
+					 DMA_TO_DEVICE);
+		if (dma_mapping_error(obj->dev, *pt_dma)) {
+			dev_err(obj->dev, "DMA map error for L2 table\n");
+			iopte_free(obj, iopte, false);
+			return ERR_PTR(-ENOMEM);
+		}
+
+		/*
+		 * we rely on dma address and the physical address to be
+		 * the same for mapping the L2 table
+		 */
+		if (WARN_ON(*pt_dma != virt_to_phys(iopte))) {
+			dev_err(obj->dev, "DMA translation error for L2 table\n");
+			dma_unmap_single(obj->dev, *pt_dma, IOPTE_TABLE_SIZE,
+					 DMA_TO_DEVICE);
+			iopte_free(obj, iopte, false);
+			return ERR_PTR(-ENOMEM);
+		}
+
 		*iopgd = virt_to_phys(iopte) | IOPGD_TABLE;
-		flush_iopgd_range(iopgd, iopgd);
 
+		flush_iopte_range(obj->dev, obj->pd_dma, offset, 1);
 		dev_vdbg(obj->dev, "%s: a new pte:%p\n", __func__, iopte);
 	} else {
 		/* We raced, free the reduniovant table */
-		iopte_free(iopte);
+		iopte_free(obj, iopte, false);
 	}
 
 pte_ready:
 	iopte = iopte_offset(iopgd, da);
-
+	*pt_dma = virt_to_phys(iopte);
 	dev_vdbg(obj->dev,
 		 "%s: da:%08x pgd:%p *pgd:%08x pte:%p *pte:%08x\n",
 		 __func__, da, iopgd, *iopgd, iopte, *iopte);
@@ -522,6 +540,7 @@ static u32 *iopte_alloc(struct omap_iommu *obj, u32 *iopgd, u32 da)
 static int iopgd_alloc_section(struct omap_iommu *obj, u32 da, u32 pa, u32 prot)
 {
 	u32 *iopgd = iopgd_offset(obj, da);
+	unsigned long offset = iopgd_index(da) * sizeof(da);
 
 	if ((da | pa) & ~IOSECTION_MASK) {
 		dev_err(obj->dev, "%s: %08x:%08x should aligned on %08lx\n",
@@ -530,13 +549,14 @@ static int iopgd_alloc_section(struct omap_iommu *obj, u32 da, u32 pa, u32 prot)
 	}
 
 	*iopgd = (pa & IOSECTION_MASK) | prot | IOPGD_SECTION;
-	flush_iopgd_range(iopgd, iopgd);
+	flush_iopte_range(obj->dev, obj->pd_dma, offset, 1);
 	return 0;
 }
 
 static int iopgd_alloc_super(struct omap_iommu *obj, u32 da, u32 pa, u32 prot)
 {
 	u32 *iopgd = iopgd_offset(obj, da);
+	unsigned long offset = iopgd_index(da) * sizeof(da);
 	int i;
 
 	if ((da | pa) & ~IOSUPER_MASK) {
@@ -547,20 +567,22 @@ static int iopgd_alloc_super(struct omap_iommu *obj, u32 da, u32 pa, u32 prot)
 
 	for (i = 0; i < 16; i++)
 		*(iopgd + i) = (pa & IOSUPER_MASK) | prot | IOPGD_SUPER;
-	flush_iopgd_range(iopgd, iopgd + 15);
+	flush_iopte_range(obj->dev, obj->pd_dma, offset, 16);
 	return 0;
 }
 
 static int iopte_alloc_page(struct omap_iommu *obj, u32 da, u32 pa, u32 prot)
 {
 	u32 *iopgd = iopgd_offset(obj, da);
-	u32 *iopte = iopte_alloc(obj, iopgd, da);
+	dma_addr_t pt_dma;
+	u32 *iopte = iopte_alloc(obj, iopgd, &pt_dma, da);
+	unsigned long offset = iopte_index(da) * sizeof(da);
 
 	if (IS_ERR(iopte))
 		return PTR_ERR(iopte);
 
 	*iopte = (pa & IOPAGE_MASK) | prot | IOPTE_SMALL;
-	flush_iopte_range(iopte, iopte);
+	flush_iopte_range(obj->dev, pt_dma, offset, 1);
 
 	dev_vdbg(obj->dev, "%s: da:%08x pa:%08x pte:%p *pte:%08x\n",
 		 __func__, da, pa, iopte, *iopte);
@@ -571,7 +593,9 @@ static int iopte_alloc_page(struct omap_iommu *obj, u32 da, u32 pa, u32 prot)
 static int iopte_alloc_large(struct omap_iommu *obj, u32 da, u32 pa, u32 prot)
 {
 	u32 *iopgd = iopgd_offset(obj, da);
-	u32 *iopte = iopte_alloc(obj, iopgd, da);
+	dma_addr_t pt_dma;
+	u32 *iopte = iopte_alloc(obj, iopgd, &pt_dma, da);
+	unsigned long offset = iopte_index(da) * sizeof(da);
 	int i;
 
 	if ((da | pa) & ~IOLARGE_MASK) {
@@ -585,7 +609,7 @@ static int iopte_alloc_large(struct omap_iommu *obj, u32 da, u32 pa, u32 prot)
 
 	for (i = 0; i < 16; i++)
 		*(iopte + i) = (pa & IOLARGE_MASK) | prot | IOPTE_LARGE;
-	flush_iopte_range(iopte, iopte + 15);
+	flush_iopte_range(obj->dev, pt_dma, offset, 16);
 	return 0;
 }
 
@@ -674,6 +698,9 @@ static size_t iopgtable_clear_entry_core(struct omap_iommu *obj, u32 da)
 	size_t bytes;
 	u32 *iopgd = iopgd_offset(obj, da);
 	int nent = 1;
+	dma_addr_t pt_dma;
+	unsigned long pd_offset = iopgd_index(da) * sizeof(da);
+	unsigned long pt_offset = iopte_index(da) * sizeof(da);
 
 	if (!*iopgd)
 		return 0;
@@ -690,7 +717,8 @@ static size_t iopgtable_clear_entry_core(struct omap_iommu *obj, u32 da)
 		}
 		bytes *= nent;
 		memset(iopte, 0, nent * sizeof(*iopte));
-		flush_iopte_range(iopte, iopte + (nent - 1) * sizeof(*iopte));
+		pt_dma = virt_to_phys(iopte);
+		flush_iopte_range(obj->dev, pt_dma, pt_offset, nent);
 
 		/*
 		 * do table walk to check if this table is necessary or not
@@ -700,7 +728,7 @@ static size_t iopgtable_clear_entry_core(struct omap_iommu *obj, u32 da)
 			if (iopte[i])
 				goto out;
 
-		iopte_free(iopte);
+		iopte_free(obj, iopte, true);
 		nent = 1; /* for the next L1 entry */
 	} else {
 		bytes = IOPGD_SIZE;
@@ -712,7 +740,7 @@ static size_t iopgtable_clear_entry_core(struct omap_iommu *obj, u32 da)
 		bytes *= nent;
 	}
 	memset(iopgd, 0, nent * sizeof(*iopgd));
-	flush_iopgd_range(iopgd, iopgd + (nent - 1) * sizeof(*iopgd));
+	flush_iopte_range(obj->dev, obj->pd_dma, pd_offset, nent);
 out:
 	return bytes;
 }
@@ -738,6 +766,7 @@ static size_t iopgtable_clear_entry(struct omap_iommu *obj, u32 da)
 
 static void iopgtable_clear_entry_all(struct omap_iommu *obj)
 {
+	unsigned long offset;
 	int i;
 
 	spin_lock(&obj->page_table_lock);
@@ -748,15 +777,16 @@ static void iopgtable_clear_entry_all(struct omap_iommu *obj)
 
 		da = i << IOPGD_SHIFT;
 		iopgd = iopgd_offset(obj, da);
+		offset = iopgd_index(da) * sizeof(da);
 
 		if (!*iopgd)
 			continue;
 
 		if (iopgd_is_table(*iopgd))
-			iopte_free(iopte_offset(iopgd, 0));
+			iopte_free(obj, iopte_offset(iopgd, 0), true);
 
 		*iopgd = 0;
-		flush_iopgd_range(iopgd, iopgd);
+		flush_iopte_range(obj->dev, obj->pd_dma, offset, 1);
 	}
 
 	flush_iotlb_all(obj);
@@ -786,7 +816,7 @@ static irqreturn_t iommu_fault_handler(int irq, void *data)
 	if (!report_iommu_fault(domain, obj->dev, da, 0))
 		return IRQ_HANDLED;
 
-	iommu_disable(obj);
+	iommu_write_reg(obj, 0, MMU_IRQENABLE);
 
 	iopgd = iopgd_offset(obj, da);
 
@@ -815,10 +845,18 @@ static int omap_iommu_attach(struct omap_iommu *obj, u32 *iopgd)
 
 	spin_lock(&obj->iommu_lock);
 
+	obj->pd_dma = dma_map_single(obj->dev, iopgd, IOPGD_TABLE_SIZE,
+				     DMA_TO_DEVICE);
+	if (dma_mapping_error(obj->dev, obj->pd_dma)) {
+		dev_err(obj->dev, "DMA map error for L1 table\n");
+		err = -ENOMEM;
+		goto out_err;
+	}
+
 	obj->iopgd = iopgd;
 	err = iommu_enable(obj);
 	if (err)
-		goto err_enable;
+		goto out_err;
 	flush_iotlb_all(obj);
 
 	spin_unlock(&obj->iommu_lock);
@@ -827,7 +865,7 @@ static int omap_iommu_attach(struct omap_iommu *obj, u32 *iopgd)
 
 	return 0;
 
-err_enable:
+out_err:
 	spin_unlock(&obj->iommu_lock);
 
 	return err;
@@ -844,7 +882,10 @@ static void omap_iommu_detach(struct omap_iommu *obj)
 
 	spin_lock(&obj->iommu_lock);
 
+	dma_unmap_single(obj->dev, obj->pd_dma, IOPGD_TABLE_SIZE,
+			 DMA_TO_DEVICE);
 	iommu_disable(obj);
+	obj->pd_dma = 0;
 	obj->iopgd = NULL;
 
 	spin_unlock(&obj->iommu_lock);
@@ -1008,11 +1049,6 @@ static struct platform_driver omap_iommu_driver = {
 	},
 };
 
-static void iopte_cachep_ctor(void *iopte)
-{
-	clean_dcache_area(iopte, IOPTE_TABLE_SIZE);
-}
-
 static u32 iotlb_init_entry(struct iotlb_entry *e, u32 da, u32 pa, int pgsz)
 {
 	memset(e, 0, sizeof(*e));
@@ -1159,7 +1195,6 @@ static struct iommu_domain *omap_iommu_domain_alloc(unsigned type)
 	if (WARN_ON(!IS_ALIGNED((long)omap_domain->pgtable, IOPGD_TABLE_SIZE)))
 		goto fail_align;
 
-	clean_dcache_area(omap_domain->pgtable, IOPGD_TABLE_SIZE);
 	spin_lock_init(&omap_domain->lock);
 
 	omap_domain->domain.geometry.aperture_start = 0;
@@ -1347,7 +1382,7 @@ static int __init omap_iommu_init(void)
 	of_node_put(np);
 
 	p = kmem_cache_create("iopte_cache", IOPTE_TABLE_SIZE, align, flags,
-			      iopte_cachep_ctor);
+			      NULL);
 	if (!p)
 		return -ENOMEM;
 	iopte_cachep = p;

drivers/iommu/omap-iommu.h

@@ -61,6 +61,7 @@ struct omap_iommu {
 	 */
 	u32		*iopgd;
 	spinlock_t	page_table_lock; /* protect iopgd */
+	dma_addr_t	pd_dma;
 
 	int		nr_tlb_entries;
 

drivers/iommu/rockchip-iommu.c

@@ -90,7 +90,9 @@ struct rk_iommu {
 	struct device *dev;
 	void __iomem **bases;
 	int num_mmu;
-	int irq;
+	int *irq;
+	int num_irq;
+	bool reset_disabled;
 	struct iommu_device iommu;
 	struct list_head node; /* entry in rk_iommu_domain.iommus */
 	struct iommu_domain *domain; /* domain to which iommu is attached */
@@ -414,6 +416,9 @@ static int rk_iommu_force_reset(struct rk_iommu *iommu)
 	int ret, i;
 	u32 dte_addr;
 
+	if (iommu->reset_disabled)
+		return 0;
+
 	/*
 	 * Check if register DTE_ADDR is working by writing DTE_ADDR_DUMMY
 	 * and verifying that upper 5 nybbles are read back.
@@ -825,10 +830,12 @@ static int rk_iommu_attach_device(struct iommu_domain *domain,
 
 	iommu->domain = domain;
 
-	ret = devm_request_irq(iommu->dev, iommu->irq, rk_iommu_irq,
-			       IRQF_SHARED, dev_name(dev), iommu);
-	if (ret)
-		return ret;
+	for (i = 0; i < iommu->num_irq; i++) {
+		ret = devm_request_irq(iommu->dev, iommu->irq[i], rk_iommu_irq,
+				       IRQF_SHARED, dev_name(dev), iommu);
+		if (ret)
+			return ret;
+	}
 
 	for (i = 0; i < iommu->num_mmu; i++) {
 		rk_iommu_write(iommu->bases[i], RK_MMU_DTE_ADDR,
@@ -878,7 +885,8 @@ static void rk_iommu_detach_device(struct iommu_domain *domain,
 	}
 	rk_iommu_disable_stall(iommu);
 
-	devm_free_irq(iommu->dev, iommu->irq, iommu);
+	for (i = 0; i < iommu->num_irq; i++)
+		devm_free_irq(iommu->dev, iommu->irq[i], iommu);
 
 	iommu->domain = NULL;
 
@@ -1008,20 +1016,20 @@ static int rk_iommu_group_set_iommudata(struct iommu_group *group,
 	ret = of_parse_phandle_with_args(np, "iommus", "#iommu-cells", 0,
 					 &args);
 	if (ret) {
-		dev_err(dev, "of_parse_phandle_with_args(%s) => %d\n",
-			np->full_name, ret);
+		dev_err(dev, "of_parse_phandle_with_args(%pOF) => %d\n",
+			np, ret);
 		return ret;
 	}
 	if (args.args_count != 0) {
-		dev_err(dev, "incorrect number of iommu params found for %s (found %d, expected 0)\n",
-			args.np->full_name, args.args_count);
+		dev_err(dev, "incorrect number of iommu params found for %pOF (found %d, expected 0)\n",
+			args.np, args.args_count);
 		return -EINVAL;
 	}
 
 	pd = of_find_device_by_node(args.np);
 	of_node_put(args.np);
 	if (!pd) {
-		dev_err(dev, "iommu %s not found\n", args.np->full_name);
+		dev_err(dev, "iommu %pOF not found\n", args.np);
 		return -EPROBE_DEFER;
 	}
 
@@ -1157,12 +1165,28 @@ static int rk_iommu_probe(struct platform_device *pdev)
 	if (iommu->num_mmu == 0)
 		return PTR_ERR(iommu->bases[0]);
 
-	iommu->irq = platform_get_irq(pdev, 0);
-	if (iommu->irq < 0) {
-		dev_err(dev, "Failed to get IRQ, %d\n", iommu->irq);
+	iommu->num_irq = platform_irq_count(pdev);
+	if (iommu->num_irq < 0)
+		return iommu->num_irq;
+	if (iommu->num_irq == 0)
 		return -ENXIO;
+
+	iommu->irq = devm_kcalloc(dev, iommu->num_irq, sizeof(*iommu->irq),
+				  GFP_KERNEL);
+	if (!iommu->irq)
+		return -ENOMEM;
+
+	for (i = 0; i < iommu->num_irq; i++) {
+		iommu->irq[i] = platform_get_irq(pdev, i);
+		if (iommu->irq[i] < 0) {
+			dev_err(dev, "Failed to get IRQ, %d\n", iommu->irq[i]);
+			return -ENXIO;
+		}
 	}
 
+	iommu->reset_disabled = device_property_read_bool(dev,
+					"rockchip,disable-mmu-reset");
+
 	err = iommu_device_sysfs_add(&iommu->iommu, dev, NULL, dev_name(dev));
 	if (err)
 		return err;

drivers/iommu/s390-iommu.c

@@ -18,6 +18,8 @@
  */
 #define S390_IOMMU_PGSIZES	(~0xFFFUL)
 
+static const struct iommu_ops s390_iommu_ops;
+
 struct s390_domain {
 	struct iommu_domain	domain;
 	struct list_head	devices;
@@ -166,11 +168,13 @@ static void s390_iommu_detach_device(struct iommu_domain *domain,
 static int s390_iommu_add_device(struct device *dev)
 {
 	struct iommu_group *group = iommu_group_get_for_dev(dev);
+	struct zpci_dev *zdev = to_pci_dev(dev)->sysdata;
 
 	if (IS_ERR(group))
 		return PTR_ERR(group);
 
 	iommu_group_put(group);
+	iommu_device_link(&zdev->iommu_dev, dev);
 
 	return 0;
 }
@@ -197,6 +201,7 @@ static void s390_iommu_remove_device(struct device *dev)
 			s390_iommu_detach_device(domain, dev);
 	}
 
+	iommu_device_unlink(&zdev->iommu_dev, dev);
 	iommu_group_remove_device(dev);
 }
 
@@ -327,7 +332,37 @@ static size_t s390_iommu_unmap(struct iommu_domain *domain,
 	return size;
 }
 
-static struct iommu_ops s390_iommu_ops = {
+int zpci_init_iommu(struct zpci_dev *zdev)
+{
+	int rc = 0;
+
+	rc = iommu_device_sysfs_add(&zdev->iommu_dev, NULL, NULL,
+				    "s390-iommu.%08x", zdev->fid);
+	if (rc)
+		goto out_err;
+
+	iommu_device_set_ops(&zdev->iommu_dev, &s390_iommu_ops);
+
+	rc = iommu_device_register(&zdev->iommu_dev);
+	if (rc)
+		goto out_sysfs;
+
+	return 0;
+
+out_sysfs:
+	iommu_device_sysfs_remove(&zdev->iommu_dev);
+
+out_err:
+	return rc;
+}
+
+void zpci_destroy_iommu(struct zpci_dev *zdev)
+{
+	iommu_device_unregister(&zdev->iommu_dev);
+	iommu_device_sysfs_remove(&zdev->iommu_dev);
+}
+
+static const struct iommu_ops s390_iommu_ops = {
 	.capable = s390_iommu_capable,
 	.domain_alloc = s390_domain_alloc,
 	.domain_free = s390_domain_free,

drivers/iommu/tegra-gart.c

@@ -61,6 +61,8 @@ struct gart_device {
 	struct list_head	client;
 	spinlock_t		client_lock;	/* for client list */
 	struct device		*dev;
+
+	struct iommu_device	iommu;		/* IOMMU Core handle */
 };
 
 struct gart_domain {
@@ -334,12 +336,35 @@ static bool gart_iommu_capable(enum iommu_cap cap)
 	return false;
 }
 
+static int gart_iommu_add_device(struct device *dev)
+{
+	struct iommu_group *group = iommu_group_get_for_dev(dev);
+
+	if (IS_ERR(group))
+		return PTR_ERR(group);
+
+	iommu_group_put(group);
+
+	iommu_device_link(&gart_handle->iommu, dev);
+
+	return 0;
+}
+
+static void gart_iommu_remove_device(struct device *dev)
+{
+	iommu_group_remove_device(dev);
+	iommu_device_unlink(&gart_handle->iommu, dev);
+}
+
 static const struct iommu_ops gart_iommu_ops = {
 	.capable	= gart_iommu_capable,
 	.domain_alloc	= gart_iommu_domain_alloc,
 	.domain_free	= gart_iommu_domain_free,
 	.attach_dev	= gart_iommu_attach_dev,
 	.detach_dev	= gart_iommu_detach_dev,
+	.add_device	= gart_iommu_add_device,
+	.remove_device	= gart_iommu_remove_device,
+	.device_group	= generic_device_group,
 	.map		= gart_iommu_map,
 	.map_sg		= default_iommu_map_sg,
 	.unmap		= gart_iommu_unmap,
@@ -378,6 +403,7 @@ static int tegra_gart_probe(struct platform_device *pdev)
 	struct resource *res, *res_remap;
 	void __iomem *gart_regs;
 	struct device *dev = &pdev->dev;
+	int ret;
 
 	if (gart_handle)
 		return -EIO;
@@ -404,6 +430,22 @@ static int tegra_gart_probe(struct platform_device *pdev)
 		return -ENXIO;
 	}
 
+	ret = iommu_device_sysfs_add(&gart->iommu, &pdev->dev, NULL,
+				     dev_name(&pdev->dev));
+	if (ret) {
+		dev_err(dev, "Failed to register IOMMU in sysfs\n");
+		return ret;
+	}
+
+	iommu_device_set_ops(&gart->iommu, &gart_iommu_ops);
+
+	ret = iommu_device_register(&gart->iommu);
+	if (ret) {
+		dev_err(dev, "Failed to register IOMMU\n");
+		iommu_device_sysfs_remove(&gart->iommu);
+		return ret;
+	}
+
 	gart->dev = &pdev->dev;
 	spin_lock_init(&gart->pte_lock);
 	spin_lock_init(&gart->client_lock);
@@ -430,6 +472,9 @@ static int tegra_gart_remove(struct platform_device *pdev)
 {
 	struct gart_device *gart = platform_get_drvdata(pdev);
 
+	iommu_device_unregister(&gart->iommu);
+	iommu_device_sysfs_remove(&gart->iommu);
+
 	writel(0, gart->regs + GART_CONFIG);
 	if (gart->savedata)
 		vfree(gart->savedata);

drivers/iommu/tegra-smmu.c

@@ -36,6 +36,8 @@ struct tegra_smmu {
 	struct list_head list;
 
 	struct dentry *debugfs;
+
+	struct iommu_device iommu;	/* IOMMU Core code handle */
 };
 
 struct tegra_smmu_as {
@@ -704,6 +706,7 @@ static struct tegra_smmu *tegra_smmu_find(struct device_node *np)
 static int tegra_smmu_add_device(struct device *dev)
 {
 	struct device_node *np = dev->of_node;
+	struct iommu_group *group;
 	struct of_phandle_args args;
 	unsigned int index = 0;
 
@@ -719,18 +722,33 @@ static int tegra_smmu_add_device(struct device *dev)
 			 * first match.
 			 */
 			dev->archdata.iommu = smmu;
+
+			iommu_device_link(&smmu->iommu, dev);
+
 			break;
 		}
 
 		index++;
 	}
 
+	group = iommu_group_get_for_dev(dev);
+	if (IS_ERR(group))
+		return PTR_ERR(group);
+
+	iommu_group_put(group);
+
 	return 0;
 }
 
 static void tegra_smmu_remove_device(struct device *dev)
 {
+	struct tegra_smmu *smmu = dev->archdata.iommu;
+
+	if (smmu)
+		iommu_device_unlink(&smmu->iommu, dev);
+
 	dev->archdata.iommu = NULL;
+	iommu_group_remove_device(dev);
 }
 
 static const struct iommu_ops tegra_smmu_ops = {
@@ -741,6 +759,7 @@ static const struct iommu_ops tegra_smmu_ops = {
 	.detach_dev = tegra_smmu_detach_dev,
 	.add_device = tegra_smmu_add_device,
 	.remove_device = tegra_smmu_remove_device,
+	.device_group = generic_device_group,
 	.map = tegra_smmu_map,
 	.unmap = tegra_smmu_unmap,
 	.map_sg = default_iommu_map_sg,
@@ -930,9 +949,24 @@ struct tegra_smmu *tegra_smmu_probe(struct device *dev,
 
 	tegra_smmu_ahb_enable();
 
+	err = iommu_device_sysfs_add(&smmu->iommu, dev, NULL, dev_name(dev));
+	if (err)
+		return ERR_PTR(err);
+
+	iommu_device_set_ops(&smmu->iommu, &tegra_smmu_ops);
+
+	err = iommu_device_register(&smmu->iommu);
+	if (err) {
+		iommu_device_sysfs_remove(&smmu->iommu);
+		return ERR_PTR(err);
+	}
+
 	err = bus_set_iommu(&platform_bus_type, &tegra_smmu_ops);
-	if (err < 0)
+	if (err < 0) {
+		iommu_device_unregister(&smmu->iommu);
+		iommu_device_sysfs_remove(&smmu->iommu);
 		return ERR_PTR(err);
+	}
 
 	if (IS_ENABLED(CONFIG_DEBUG_FS))
 		tegra_smmu_debugfs_init(smmu);
@@ -942,6 +976,9 @@ struct tegra_smmu *tegra_smmu_probe(struct device *dev,
 
 void tegra_smmu_remove(struct tegra_smmu *smmu)
 {
+	iommu_device_unregister(&smmu->iommu);
+	iommu_device_sysfs_remove(&smmu->iommu);
+
 	if (IS_ENABLED(CONFIG_DEBUG_FS))
 		tegra_smmu_debugfs_exit(smmu);
 }

drivers/memory/mtk-smi.c

@@ -16,6 +16,7 @@
 #include <linux/device.h>
 #include <linux/err.h>
 #include <linux/io.h>
+#include <linux/module.h>
 #include <linux/of.h>
 #include <linux/of_platform.h>
 #include <linux/platform_device.h>
@@ -23,7 +24,10 @@
 #include <soc/mediatek/smi.h>
 #include <dt-bindings/memory/mt2701-larb-port.h>
 
+/* mt8173 */
 #define SMI_LARB_MMU_EN		0xf00
+
+/* mt2701 */
 #define REG_SMI_SECUR_CON_BASE		0x5c0
 
 /* every register control 8 port, register offset 0x4 */
@@ -41,7 +45,12 @@
 /* mt2701 domain should be set to 3 */
 #define SMI_SECUR_CON_VAL_DOMAIN(id)	(0x3 << ((((id) & 0x7) << 2) + 1))
 
+/* mt2712 */
+#define SMI_LARB_NONSEC_CON(id)	(0x380 + ((id) * 4))
+#define F_MMU_EN		BIT(0)
+
 struct mtk_smi_larb_gen {
+	bool need_larbid;
 	int port_in_larb[MTK_LARB_NR_MAX + 1];
 	void (*config_port)(struct device *);
 };
@@ -148,6 +157,15 @@ mtk_smi_larb_bind(struct device *dev, struct device *master, void *data)
 	struct mtk_smi_iommu *smi_iommu = data;
 	unsigned int         i;
 
+	if (larb->larb_gen->need_larbid) {
+		larb->mmu = &smi_iommu->larb_imu[larb->larbid].mmu;
+		return 0;
+	}
+
+	/*
+	 * If there is no larbid property, Loop to find the corresponding
+	 * iommu information.
+	 */
 	for (i = 0; i < smi_iommu->larb_nr; i++) {
 		if (dev == smi_iommu->larb_imu[i].dev) {
 			/* The 'mmu' may be updated in iommu-attach/detach. */
@@ -158,13 +176,32 @@ mtk_smi_larb_bind(struct device *dev, struct device *master, void *data)
 	return -ENODEV;
 }
 
-static void mtk_smi_larb_config_port(struct device *dev)
+static void mtk_smi_larb_config_port_mt2712(struct device *dev)
 {
 	struct mtk_smi_larb *larb = dev_get_drvdata(dev);
+	u32 reg;
+	int i;
 
-	writel(*larb->mmu, larb->base + SMI_LARB_MMU_EN);
+	/*
+	 * larb 8/9 is the bdpsys larb, the iommu_en is enabled defaultly.
+	 * Don't need to set it again.
+	 */
+	if (larb->larbid == 8 || larb->larbid == 9)
+		return;
+
+	for_each_set_bit(i, (unsigned long *)larb->mmu, 32) {
+		reg = readl_relaxed(larb->base + SMI_LARB_NONSEC_CON(i));
+		reg |= F_MMU_EN;
+		writel(reg, larb->base + SMI_LARB_NONSEC_CON(i));
+	}
 }
 
+static void mtk_smi_larb_config_port_mt8173(struct device *dev)
+{
+	struct mtk_smi_larb *larb = dev_get_drvdata(dev);
+
+	writel(*larb->mmu, larb->base + SMI_LARB_MMU_EN);
+}
 
 static void mtk_smi_larb_config_port_gen1(struct device *dev)
 {
@@ -210,10 +247,11 @@ static const struct component_ops mtk_smi_larb_component_ops = {
 
 static const struct mtk_smi_larb_gen mtk_smi_larb_mt8173 = {
 	/* mt8173 do not need the port in larb */
-	.config_port = mtk_smi_larb_config_port,
+	.config_port = mtk_smi_larb_config_port_mt8173,
 };
 
 static const struct mtk_smi_larb_gen mtk_smi_larb_mt2701 = {
+	.need_larbid = true,
 	.port_in_larb = {
 		LARB0_PORT_OFFSET, LARB1_PORT_OFFSET,
 		LARB2_PORT_OFFSET, LARB3_PORT_OFFSET
@@ -221,6 +259,11 @@ static const struct mtk_smi_larb_gen mtk_smi_larb_mt2701 = {
 	.config_port = mtk_smi_larb_config_port_gen1,
 };
 
+static const struct mtk_smi_larb_gen mtk_smi_larb_mt2712 = {
+	.need_larbid = true,
+	.config_port = mtk_smi_larb_config_port_mt2712,
+};
+
 static const struct of_device_id mtk_smi_larb_of_ids[] = {
 	{
 		.compatible = "mediatek,mt8173-smi-larb",
@@ -230,6 +273,10 @@ static const struct of_device_id mtk_smi_larb_of_ids[] = {
 		.compatible = "mediatek,mt2701-smi-larb",
 		.data = &mtk_smi_larb_mt2701
 	},
+	{
+		.compatible = "mediatek,mt2712-smi-larb",
+		.data = &mtk_smi_larb_mt2712
+	},
 	{}
 };
 
@@ -240,20 +287,13 @@ static int mtk_smi_larb_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct device_node *smi_node;
 	struct platform_device *smi_pdev;
-	const struct of_device_id *of_id;
-
-	if (!dev->pm_domain)
-		return -EPROBE_DEFER;
-
-	of_id = of_match_node(mtk_smi_larb_of_ids, pdev->dev.of_node);
-	if (!of_id)
-		return -EINVAL;
+	int err;
 
 	larb = devm_kzalloc(dev, sizeof(*larb), GFP_KERNEL);
 	if (!larb)
 		return -ENOMEM;
 
-	larb->larb_gen = of_id->data;
+	larb->larb_gen = of_device_get_match_data(dev);
 	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 	larb->base = devm_ioremap_resource(dev, res);
 	if (IS_ERR(larb->base))
@@ -268,6 +308,15 @@ static int mtk_smi_larb_probe(struct platform_device *pdev)
 		return PTR_ERR(larb->smi.clk_smi);
 	larb->smi.dev = dev;
 
+	if (larb->larb_gen->need_larbid) {
+		err = of_property_read_u32(dev->of_node, "mediatek,larb-id",
+					   &larb->larbid);
+		if (err) {
+			dev_err(dev, "missing larbid property\n");
+			return err;
+		}
+	}
+
 	smi_node = of_parse_phandle(dev->of_node, "mediatek,smi", 0);
 	if (!smi_node)
 		return -EINVAL;
@@ -275,6 +324,8 @@ static int mtk_smi_larb_probe(struct platform_device *pdev)
 	smi_pdev = of_find_device_by_node(smi_node);
 	of_node_put(smi_node);
 	if (smi_pdev) {
+		if (!platform_get_drvdata(smi_pdev))
+			return -EPROBE_DEFER;
 		larb->smi_common_dev = &smi_pdev->dev;
 	} else {
 		dev_err(dev, "Failed to get the smi_common device\n");
@@ -311,6 +362,10 @@ static const struct of_device_id mtk_smi_common_of_ids[] = {
 		.compatible = "mediatek,mt2701-smi-common",
 		.data = (void *)MTK_SMI_GEN1
 	},
+	{
+		.compatible = "mediatek,mt2712-smi-common",
+		.data = (void *)MTK_SMI_GEN2
+	},
 	{}
 };
 
@@ -319,11 +374,8 @@ static int mtk_smi_common_probe(struct platform_device *pdev)
 	struct device *dev = &pdev->dev;
 	struct mtk_smi *common;
 	struct resource *res;
-	const struct of_device_id *of_id;
 	enum mtk_smi_gen smi_gen;
-
-	if (!dev->pm_domain)
-		return -EPROBE_DEFER;
+	int ret;
 
 	common = devm_kzalloc(dev, sizeof(*common), GFP_KERNEL);
 	if (!common)
@@ -338,17 +390,13 @@ static int mtk_smi_common_probe(struct platform_device *pdev)
 	if (IS_ERR(common->clk_smi))
 		return PTR_ERR(common->clk_smi);
 
-	of_id = of_match_node(mtk_smi_common_of_ids, pdev->dev.of_node);
-	if (!of_id)
-		return -EINVAL;
-
 	/*
 	 * for mtk smi gen 1, we need to get the ao(always on) base to config
 	 * m4u port, and we need to enable the aync clock for transform the smi
 	 * clock into emi clock domain, but for mtk smi gen2, there's no smi ao
 	 * base.
 	 */
-	smi_gen = (enum mtk_smi_gen)of_id->data;
+	smi_gen = (enum mtk_smi_gen)of_device_get_match_data(dev);
 	if (smi_gen == MTK_SMI_GEN1) {
 		res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
 		common->smi_ao_base = devm_ioremap_resource(dev, res);
@@ -359,7 +407,9 @@ static int mtk_smi_common_probe(struct platform_device *pdev)
 		if (IS_ERR(common->clk_async))
 			return PTR_ERR(common->clk_async);
 
-		clk_prepare_enable(common->clk_async);
+		ret = clk_prepare_enable(common->clk_async);
+		if (ret)
+			return ret;
 	}
 	pm_runtime_enable(dev);
 	platform_set_drvdata(pdev, common);
@@ -403,4 +453,4 @@ static int __init mtk_smi_init(void)
 	return ret;
 }
 
-subsys_initcall(mtk_smi_init);
+module_init(mtk_smi_init);

include/dt-bindings/memory/mt8173-larb-port.h

@@ -15,10 +15,6 @@
 #define __DTS_IOMMU_PORT_MT8173_H
 
 #define MTK_M4U_ID(larb, port)		(((larb) << 5) | (port))
-/* Local arbiter ID */
-#define MTK_M4U_TO_LARB(id)		(((id) >> 5) & 0x7)
-/* PortID within the local arbiter */
-#define MTK_M4U_TO_PORT(id)		((id) & 0x1f)
 
 #define M4U_LARB0_ID			0
 #define M4U_LARB1_ID			1

include/soc/mediatek/smi.h

@@ -19,7 +19,7 @@
 
 #ifdef CONFIG_MTK_SMI
 
-#define MTK_LARB_NR_MAX		8
+#define MTK_LARB_NR_MAX		16
 
 #define MTK_SMI_MMU_EN(port)	BIT(port)