xen-devel.lists.xenproject.org archive mirror
 help / color / mirror / Atom feed
* [PATCH v2 0/8] xen/arm: Add support for SMMUv3 driver
@ 2020-11-26 17:01 Rahul Singh
  2020-11-26 17:02 ` [PATCH v2 1/8] xen/arm: Import the SMMUv3 driver from Linux Rahul Singh
                   ` (7 more replies)
  0 siblings, 8 replies; 53+ messages in thread
From: Rahul Singh @ 2020-11-26 17:01 UTC (permalink / raw)
  To: xen-devel
  Cc: bertrand.marquis, rahul.singh, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk, Andrew Cooper, George Dunlap, Ian Jackson,
	Jan Beulich, Wei Liu, Paul Durrant

This patch series is v2 of the work to add support for the SMMUv3 driver.

Approach taken is to first merge the Linux copy of the SMMUv3 driver
(tag v5.9.8) and then modify the driver to build on XEN.

Linux SMMUv3 code implements the commands-queue insertion based on atomic
operations implemented in Linux. Atomic functions used by the commands-queue
insertion is not implemented in XEN therefore we decided to revert the code
that implemented the commands-queue insertion based on atomic operations.
Once the proper atomic operations will be available in XEN the driver can be
updated to include the patch.

Remove the support for MSI and PCI ATS functionality as there is no support
available in XEN on ARM to test the functionality.

As of now only Stage-2 translation support has been tested. Stage-1
translation support is removed. Once Stage-1 translation support is tested code
can be added.

Code specific to Linux is removed from the driver to avoid dead code.

Rahul Singh (8):
  xen/arm: Import the SMMUv3 driver from Linux
  xen/arm: revert atomic operation related command-queue insertion patch
  xen/arm: revert patch related to XArray
  xen/arm: Remove support for MSI on SMMUv3
  xen/arm: Remove support for PCI ATS on SMMUv3
  xen/arm: Remove support for Stage-1 translation on SMMUv3.
  xen/arm: Remove Linux specific code that is not usable in XEN
  xen/arm: Add support for SMMUv3 driver

 MAINTAINERS                           |    6 +
 SUPPORT.md                            |    1 +
 xen/drivers/passthrough/Kconfig       |   10 +
 xen/drivers/passthrough/arm/Makefile  |    1 +
 xen/drivers/passthrough/arm/smmu-v3.c | 2954 +++++++++++++++++++++++++
 5 files changed, 2972 insertions(+)
 create mode 100644 xen/drivers/passthrough/arm/smmu-v3.c

-- 
2.17.1



^ permalink raw reply	[flat|nested] 53+ messages in thread

* [PATCH v2 1/8] xen/arm: Import the SMMUv3 driver from Linux
  2020-11-26 17:01 [PATCH v2 0/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
@ 2020-11-26 17:02 ` Rahul Singh
  2020-12-01 22:01   ` Stefano Stabellini
  2020-11-26 17:02 ` [PATCH v2 2/8] xen/arm: revert atomic operation related command-queue insertion patch Rahul Singh
                   ` (6 subsequent siblings)
  7 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-11-26 17:02 UTC (permalink / raw)
  To: xen-devel
  Cc: bertrand.marquis, rahul.singh, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

Based on tag Linux 5.9.8 commit 951cbbc386ff01b50da4f46387e994e81d9ab431

It's a copy of the Linux SMMUv3 driver. Xen specific code has not
been added yet and code has not been compiled.

Signed-off-by: Rahul Singh <rahul.singh@arm.com>
---
 xen/drivers/passthrough/arm/smmu-v3.c | 4164 +++++++++++++++++++++++++
 1 file changed, 4164 insertions(+)
 create mode 100644 xen/drivers/passthrough/arm/smmu-v3.c

diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
new file mode 100644
index 0000000000..c192544e87
--- /dev/null
+++ b/xen/drivers/passthrough/arm/smmu-v3.c
@@ -0,0 +1,4164 @@
+// SPDX-License-Identifier: GPL-2.0
+/*
+ * IOMMU API for ARM architected SMMUv3 implementations.
+ *
+ * Copyright (C) 2015 ARM Limited
+ *
+ * Author: Will Deacon <will.deacon@arm.com>
+ *
+ * This driver is powered by bad coffee and bombay mix.
+ */
+
+#include <linux/acpi.h>
+#include <linux/acpi_iort.h>
+#include <linux/bitfield.h>
+#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>
+#include <linux/iommu.h>
+#include <linux/iopoll.h>
+#include <linux/module.h>
+#include <linux/msi.h>
+#include <linux/of.h>
+#include <linux/of_address.h>
+#include <linux/of_iommu.h>
+#include <linux/of_platform.h>
+#include <linux/pci.h>
+#include <linux/pci-ats.h>
+#include <linux/platform_device.h>
+
+#include <linux/amba/bus.h>
+
+/* MMIO registers */
+#define ARM_SMMU_IDR0			0x0
+#define IDR0_ST_LVL			GENMASK(28, 27)
+#define IDR0_ST_LVL_2LVL		1
+#define IDR0_STALL_MODEL		GENMASK(25, 24)
+#define IDR0_STALL_MODEL_STALL		0
+#define IDR0_STALL_MODEL_FORCE		2
+#define IDR0_TTENDIAN			GENMASK(22, 21)
+#define IDR0_TTENDIAN_MIXED		0
+#define IDR0_TTENDIAN_LE		2
+#define IDR0_TTENDIAN_BE		3
+#define IDR0_CD2L			(1 << 19)
+#define IDR0_VMID16			(1 << 18)
+#define IDR0_PRI			(1 << 16)
+#define IDR0_SEV			(1 << 14)
+#define IDR0_MSI			(1 << 13)
+#define IDR0_ASID16			(1 << 12)
+#define IDR0_ATS			(1 << 10)
+#define IDR0_HYP			(1 << 9)
+#define IDR0_COHACC			(1 << 4)
+#define IDR0_TTF			GENMASK(3, 2)
+#define IDR0_TTF_AARCH64		2
+#define IDR0_TTF_AARCH32_64		3
+#define IDR0_S1P			(1 << 1)
+#define IDR0_S2P			(1 << 0)
+
+#define ARM_SMMU_IDR1			0x4
+#define IDR1_TABLES_PRESET		(1 << 30)
+#define IDR1_QUEUES_PRESET		(1 << 29)
+#define IDR1_REL			(1 << 28)
+#define IDR1_CMDQS			GENMASK(25, 21)
+#define IDR1_EVTQS			GENMASK(20, 16)
+#define IDR1_PRIQS			GENMASK(15, 11)
+#define IDR1_SSIDSIZE			GENMASK(10, 6)
+#define IDR1_SIDSIZE			GENMASK(5, 0)
+
+#define ARM_SMMU_IDR3			0xc
+#define IDR3_RIL			(1 << 10)
+
+#define ARM_SMMU_IDR5			0x14
+#define IDR5_STALL_MAX			GENMASK(31, 16)
+#define IDR5_GRAN64K			(1 << 6)
+#define IDR5_GRAN16K			(1 << 5)
+#define IDR5_GRAN4K			(1 << 4)
+#define IDR5_OAS			GENMASK(2, 0)
+#define IDR5_OAS_32_BIT			0
+#define IDR5_OAS_36_BIT			1
+#define IDR5_OAS_40_BIT			2
+#define IDR5_OAS_42_BIT			3
+#define IDR5_OAS_44_BIT			4
+#define IDR5_OAS_48_BIT			5
+#define IDR5_OAS_52_BIT			6
+#define IDR5_VAX			GENMASK(11, 10)
+#define IDR5_VAX_52_BIT			1
+
+#define ARM_SMMU_CR0			0x20
+#define CR0_ATSCHK			(1 << 4)
+#define CR0_CMDQEN			(1 << 3)
+#define CR0_EVTQEN			(1 << 2)
+#define CR0_PRIQEN			(1 << 1)
+#define CR0_SMMUEN			(1 << 0)
+
+#define ARM_SMMU_CR0ACK			0x24
+
+#define ARM_SMMU_CR1			0x28
+#define CR1_TABLE_SH			GENMASK(11, 10)
+#define CR1_TABLE_OC			GENMASK(9, 8)
+#define CR1_TABLE_IC			GENMASK(7, 6)
+#define CR1_QUEUE_SH			GENMASK(5, 4)
+#define CR1_QUEUE_OC			GENMASK(3, 2)
+#define CR1_QUEUE_IC			GENMASK(1, 0)
+/* CR1 cacheability fields don't quite follow the usual TCR-style encoding */
+#define CR1_CACHE_NC			0
+#define CR1_CACHE_WB			1
+#define CR1_CACHE_WT			2
+
+#define ARM_SMMU_CR2			0x2c
+#define CR2_PTM				(1 << 2)
+#define CR2_RECINVSID			(1 << 1)
+#define CR2_E2H				(1 << 0)
+
+#define ARM_SMMU_GBPA			0x44
+#define GBPA_UPDATE			(1 << 31)
+#define GBPA_ABORT			(1 << 20)
+
+#define ARM_SMMU_IRQ_CTRL		0x50
+#define IRQ_CTRL_EVTQ_IRQEN		(1 << 2)
+#define IRQ_CTRL_PRIQ_IRQEN		(1 << 1)
+#define IRQ_CTRL_GERROR_IRQEN		(1 << 0)
+
+#define ARM_SMMU_IRQ_CTRLACK		0x54
+
+#define ARM_SMMU_GERROR			0x60
+#define GERROR_SFM_ERR			(1 << 8)
+#define GERROR_MSI_GERROR_ABT_ERR	(1 << 7)
+#define GERROR_MSI_PRIQ_ABT_ERR		(1 << 6)
+#define GERROR_MSI_EVTQ_ABT_ERR		(1 << 5)
+#define GERROR_MSI_CMDQ_ABT_ERR		(1 << 4)
+#define GERROR_PRIQ_ABT_ERR		(1 << 3)
+#define GERROR_EVTQ_ABT_ERR		(1 << 2)
+#define GERROR_CMDQ_ERR			(1 << 0)
+#define GERROR_ERR_MASK			0xfd
+
+#define ARM_SMMU_GERRORN		0x64
+
+#define ARM_SMMU_GERROR_IRQ_CFG0	0x68
+#define ARM_SMMU_GERROR_IRQ_CFG1	0x70
+#define ARM_SMMU_GERROR_IRQ_CFG2	0x74
+
+#define ARM_SMMU_STRTAB_BASE		0x80
+#define STRTAB_BASE_RA			(1UL << 62)
+#define STRTAB_BASE_ADDR_MASK		GENMASK_ULL(51, 6)
+
+#define ARM_SMMU_STRTAB_BASE_CFG	0x88
+#define STRTAB_BASE_CFG_FMT		GENMASK(17, 16)
+#define STRTAB_BASE_CFG_FMT_LINEAR	0
+#define STRTAB_BASE_CFG_FMT_2LVL	1
+#define STRTAB_BASE_CFG_SPLIT		GENMASK(10, 6)
+#define STRTAB_BASE_CFG_LOG2SIZE	GENMASK(5, 0)
+
+#define ARM_SMMU_CMDQ_BASE		0x90
+#define ARM_SMMU_CMDQ_PROD		0x98
+#define ARM_SMMU_CMDQ_CONS		0x9c
+
+#define ARM_SMMU_EVTQ_BASE		0xa0
+#define ARM_SMMU_EVTQ_PROD		0x100a8
+#define ARM_SMMU_EVTQ_CONS		0x100ac
+#define ARM_SMMU_EVTQ_IRQ_CFG0		0xb0
+#define ARM_SMMU_EVTQ_IRQ_CFG1		0xb8
+#define ARM_SMMU_EVTQ_IRQ_CFG2		0xbc
+
+#define ARM_SMMU_PRIQ_BASE		0xc0
+#define ARM_SMMU_PRIQ_PROD		0x100c8
+#define ARM_SMMU_PRIQ_CONS		0x100cc
+#define ARM_SMMU_PRIQ_IRQ_CFG0		0xd0
+#define ARM_SMMU_PRIQ_IRQ_CFG1		0xd8
+#define ARM_SMMU_PRIQ_IRQ_CFG2		0xdc
+
+#define ARM_SMMU_REG_SZ			0xe00
+
+/* Common MSI config fields */
+#define MSI_CFG0_ADDR_MASK		GENMASK_ULL(51, 2)
+#define MSI_CFG2_SH			GENMASK(5, 4)
+#define MSI_CFG2_MEMATTR		GENMASK(3, 0)
+
+/* Common memory attribute values */
+#define ARM_SMMU_SH_NSH			0
+#define ARM_SMMU_SH_OSH			2
+#define ARM_SMMU_SH_ISH			3
+#define ARM_SMMU_MEMATTR_DEVICE_nGnRE	0x1
+#define ARM_SMMU_MEMATTR_OIWB		0xf
+
+#define Q_IDX(llq, p)			((p) & ((1 << (llq)->max_n_shift) - 1))
+#define Q_WRP(llq, p)			((p) & (1 << (llq)->max_n_shift))
+#define Q_OVERFLOW_FLAG			(1U << 31)
+#define Q_OVF(p)			((p) & Q_OVERFLOW_FLAG)
+#define Q_ENT(q, p)			((q)->base +			\
+					 Q_IDX(&((q)->llq), p) *	\
+					 (q)->ent_dwords)
+
+#define Q_BASE_RWA			(1UL << 62)
+#define Q_BASE_ADDR_MASK		GENMASK_ULL(51, 5)
+#define Q_BASE_LOG2SIZE			GENMASK(4, 0)
+
+/* Ensure DMA allocations are naturally aligned */
+#ifdef CONFIG_CMA_ALIGNMENT
+#define Q_MAX_SZ_SHIFT			(PAGE_SHIFT + CONFIG_CMA_ALIGNMENT)
+#else
+#define Q_MAX_SZ_SHIFT			(PAGE_SHIFT + MAX_ORDER - 1)
+#endif
+
+/*
+ * Stream table.
+ *
+ * Linear: Enough to cover 1 << IDR1.SIDSIZE entries
+ * 2lvl: 128k L1 entries,
+ *       256 lazy entries per table (each table covers a PCI bus)
+ */
+#define STRTAB_L1_SZ_SHIFT		20
+#define STRTAB_SPLIT			8
+
+#define STRTAB_L1_DESC_DWORDS		1
+#define STRTAB_L1_DESC_SPAN		GENMASK_ULL(4, 0)
+#define STRTAB_L1_DESC_L2PTR_MASK	GENMASK_ULL(51, 6)
+
+#define STRTAB_STE_DWORDS		8
+#define STRTAB_STE_0_V			(1UL << 0)
+#define STRTAB_STE_0_CFG		GENMASK_ULL(3, 1)
+#define STRTAB_STE_0_CFG_ABORT		0
+#define STRTAB_STE_0_CFG_BYPASS		4
+#define STRTAB_STE_0_CFG_S1_TRANS	5
+#define STRTAB_STE_0_CFG_S2_TRANS	6
+
+#define STRTAB_STE_0_S1FMT		GENMASK_ULL(5, 4)
+#define STRTAB_STE_0_S1FMT_LINEAR	0
+#define STRTAB_STE_0_S1FMT_64K_L2	2
+#define STRTAB_STE_0_S1CTXPTR_MASK	GENMASK_ULL(51, 6)
+#define STRTAB_STE_0_S1CDMAX		GENMASK_ULL(63, 59)
+
+#define STRTAB_STE_1_S1DSS		GENMASK_ULL(1, 0)
+#define STRTAB_STE_1_S1DSS_TERMINATE	0x0
+#define STRTAB_STE_1_S1DSS_BYPASS	0x1
+#define STRTAB_STE_1_S1DSS_SSID0	0x2
+
+#define STRTAB_STE_1_S1C_CACHE_NC	0UL
+#define STRTAB_STE_1_S1C_CACHE_WBRA	1UL
+#define STRTAB_STE_1_S1C_CACHE_WT	2UL
+#define STRTAB_STE_1_S1C_CACHE_WB	3UL
+#define STRTAB_STE_1_S1CIR		GENMASK_ULL(3, 2)
+#define STRTAB_STE_1_S1COR		GENMASK_ULL(5, 4)
+#define STRTAB_STE_1_S1CSH		GENMASK_ULL(7, 6)
+
+#define STRTAB_STE_1_S1STALLD		(1UL << 27)
+
+#define STRTAB_STE_1_EATS		GENMASK_ULL(29, 28)
+#define STRTAB_STE_1_EATS_ABT		0UL
+#define STRTAB_STE_1_EATS_TRANS		1UL
+#define STRTAB_STE_1_EATS_S1CHK		2UL
+
+#define STRTAB_STE_1_STRW		GENMASK_ULL(31, 30)
+#define STRTAB_STE_1_STRW_NSEL1		0UL
+#define STRTAB_STE_1_STRW_EL2		2UL
+
+#define STRTAB_STE_1_SHCFG		GENMASK_ULL(45, 44)
+#define STRTAB_STE_1_SHCFG_INCOMING	1UL
+
+#define STRTAB_STE_2_S2VMID		GENMASK_ULL(15, 0)
+#define STRTAB_STE_2_VTCR		GENMASK_ULL(50, 32)
+#define STRTAB_STE_2_VTCR_S2T0SZ	GENMASK_ULL(5, 0)
+#define STRTAB_STE_2_VTCR_S2SL0		GENMASK_ULL(7, 6)
+#define STRTAB_STE_2_VTCR_S2IR0		GENMASK_ULL(9, 8)
+#define STRTAB_STE_2_VTCR_S2OR0		GENMASK_ULL(11, 10)
+#define STRTAB_STE_2_VTCR_S2SH0		GENMASK_ULL(13, 12)
+#define STRTAB_STE_2_VTCR_S2TG		GENMASK_ULL(15, 14)
+#define STRTAB_STE_2_VTCR_S2PS		GENMASK_ULL(18, 16)
+#define STRTAB_STE_2_S2AA64		(1UL << 51)
+#define STRTAB_STE_2_S2ENDI		(1UL << 52)
+#define STRTAB_STE_2_S2PTW		(1UL << 54)
+#define STRTAB_STE_2_S2R		(1UL << 58)
+
+#define STRTAB_STE_3_S2TTB_MASK		GENMASK_ULL(51, 4)
+
+/*
+ * Context descriptors.
+ *
+ * Linear: when less than 1024 SSIDs are supported
+ * 2lvl: at most 1024 L1 entries,
+ *       1024 lazy entries per table.
+ */
+#define CTXDESC_SPLIT			10
+#define CTXDESC_L2_ENTRIES		(1 << CTXDESC_SPLIT)
+
+#define CTXDESC_L1_DESC_DWORDS		1
+#define CTXDESC_L1_DESC_V		(1UL << 0)
+#define CTXDESC_L1_DESC_L2PTR_MASK	GENMASK_ULL(51, 12)
+
+#define CTXDESC_CD_DWORDS		8
+#define CTXDESC_CD_0_TCR_T0SZ		GENMASK_ULL(5, 0)
+#define CTXDESC_CD_0_TCR_TG0		GENMASK_ULL(7, 6)
+#define CTXDESC_CD_0_TCR_IRGN0		GENMASK_ULL(9, 8)
+#define CTXDESC_CD_0_TCR_ORGN0		GENMASK_ULL(11, 10)
+#define CTXDESC_CD_0_TCR_SH0		GENMASK_ULL(13, 12)
+#define CTXDESC_CD_0_TCR_EPD0		(1ULL << 14)
+#define CTXDESC_CD_0_TCR_EPD1		(1ULL << 30)
+
+#define CTXDESC_CD_0_ENDI		(1UL << 15)
+#define CTXDESC_CD_0_V			(1UL << 31)
+
+#define CTXDESC_CD_0_TCR_IPS		GENMASK_ULL(34, 32)
+#define CTXDESC_CD_0_TCR_TBI0		(1ULL << 38)
+
+#define CTXDESC_CD_0_AA64		(1UL << 41)
+#define CTXDESC_CD_0_S			(1UL << 44)
+#define CTXDESC_CD_0_R			(1UL << 45)
+#define CTXDESC_CD_0_A			(1UL << 46)
+#define CTXDESC_CD_0_ASET		(1UL << 47)
+#define CTXDESC_CD_0_ASID		GENMASK_ULL(63, 48)
+
+#define CTXDESC_CD_1_TTB0_MASK		GENMASK_ULL(51, 4)
+
+/*
+ * When the SMMU only supports linear context descriptor tables, pick a
+ * reasonable size limit (64kB).
+ */
+#define CTXDESC_LINEAR_CDMAX		ilog2(SZ_64K / (CTXDESC_CD_DWORDS << 3))
+
+/* Command queue */
+#define CMDQ_ENT_SZ_SHIFT		4
+#define CMDQ_ENT_DWORDS			((1 << CMDQ_ENT_SZ_SHIFT) >> 3)
+#define CMDQ_MAX_SZ_SHIFT		(Q_MAX_SZ_SHIFT - CMDQ_ENT_SZ_SHIFT)
+
+#define CMDQ_CONS_ERR			GENMASK(30, 24)
+#define CMDQ_ERR_CERROR_NONE_IDX	0
+#define CMDQ_ERR_CERROR_ILL_IDX		1
+#define CMDQ_ERR_CERROR_ABT_IDX		2
+#define CMDQ_ERR_CERROR_ATC_INV_IDX	3
+
+#define CMDQ_PROD_OWNED_FLAG		Q_OVERFLOW_FLAG
+
+/*
+ * This is used to size the command queue and therefore must be at least
+ * BITS_PER_LONG so that the valid_map works correctly (it relies on the
+ * total number of queue entries being a multiple of BITS_PER_LONG).
+ */
+#define CMDQ_BATCH_ENTRIES		BITS_PER_LONG
+
+#define CMDQ_0_OP			GENMASK_ULL(7, 0)
+#define CMDQ_0_SSV			(1UL << 11)
+
+#define CMDQ_PREFETCH_0_SID		GENMASK_ULL(63, 32)
+#define CMDQ_PREFETCH_1_SIZE		GENMASK_ULL(4, 0)
+#define CMDQ_PREFETCH_1_ADDR_MASK	GENMASK_ULL(63, 12)
+
+#define CMDQ_CFGI_0_SSID		GENMASK_ULL(31, 12)
+#define CMDQ_CFGI_0_SID			GENMASK_ULL(63, 32)
+#define CMDQ_CFGI_1_LEAF		(1UL << 0)
+#define CMDQ_CFGI_1_RANGE		GENMASK_ULL(4, 0)
+
+#define CMDQ_TLBI_0_NUM			GENMASK_ULL(16, 12)
+#define CMDQ_TLBI_RANGE_NUM_MAX		31
+#define CMDQ_TLBI_0_SCALE		GENMASK_ULL(24, 20)
+#define CMDQ_TLBI_0_VMID		GENMASK_ULL(47, 32)
+#define CMDQ_TLBI_0_ASID		GENMASK_ULL(63, 48)
+#define CMDQ_TLBI_1_LEAF		(1UL << 0)
+#define CMDQ_TLBI_1_TTL			GENMASK_ULL(9, 8)
+#define CMDQ_TLBI_1_TG			GENMASK_ULL(11, 10)
+#define CMDQ_TLBI_1_VA_MASK		GENMASK_ULL(63, 12)
+#define CMDQ_TLBI_1_IPA_MASK		GENMASK_ULL(51, 12)
+
+#define CMDQ_ATC_0_SSID			GENMASK_ULL(31, 12)
+#define CMDQ_ATC_0_SID			GENMASK_ULL(63, 32)
+#define CMDQ_ATC_0_GLOBAL		(1UL << 9)
+#define CMDQ_ATC_1_SIZE			GENMASK_ULL(5, 0)
+#define CMDQ_ATC_1_ADDR_MASK		GENMASK_ULL(63, 12)
+
+#define CMDQ_PRI_0_SSID			GENMASK_ULL(31, 12)
+#define CMDQ_PRI_0_SID			GENMASK_ULL(63, 32)
+#define CMDQ_PRI_1_GRPID		GENMASK_ULL(8, 0)
+#define CMDQ_PRI_1_RESP			GENMASK_ULL(13, 12)
+
+#define CMDQ_SYNC_0_CS			GENMASK_ULL(13, 12)
+#define CMDQ_SYNC_0_CS_NONE		0
+#define CMDQ_SYNC_0_CS_IRQ		1
+#define CMDQ_SYNC_0_CS_SEV		2
+#define CMDQ_SYNC_0_MSH			GENMASK_ULL(23, 22)
+#define CMDQ_SYNC_0_MSIATTR		GENMASK_ULL(27, 24)
+#define CMDQ_SYNC_0_MSIDATA		GENMASK_ULL(63, 32)
+#define CMDQ_SYNC_1_MSIADDR_MASK	GENMASK_ULL(51, 2)
+
+/* Event queue */
+#define EVTQ_ENT_SZ_SHIFT		5
+#define EVTQ_ENT_DWORDS			((1 << EVTQ_ENT_SZ_SHIFT) >> 3)
+#define EVTQ_MAX_SZ_SHIFT		(Q_MAX_SZ_SHIFT - EVTQ_ENT_SZ_SHIFT)
+
+#define EVTQ_0_ID			GENMASK_ULL(7, 0)
+
+/* PRI queue */
+#define PRIQ_ENT_SZ_SHIFT		4
+#define PRIQ_ENT_DWORDS			((1 << PRIQ_ENT_SZ_SHIFT) >> 3)
+#define PRIQ_MAX_SZ_SHIFT		(Q_MAX_SZ_SHIFT - PRIQ_ENT_SZ_SHIFT)
+
+#define PRIQ_0_SID			GENMASK_ULL(31, 0)
+#define PRIQ_0_SSID			GENMASK_ULL(51, 32)
+#define PRIQ_0_PERM_PRIV		(1UL << 58)
+#define PRIQ_0_PERM_EXEC		(1UL << 59)
+#define PRIQ_0_PERM_READ		(1UL << 60)
+#define PRIQ_0_PERM_WRITE		(1UL << 61)
+#define PRIQ_0_PRG_LAST			(1UL << 62)
+#define PRIQ_0_SSID_V			(1UL << 63)
+
+#define PRIQ_1_PRG_IDX			GENMASK_ULL(8, 0)
+#define PRIQ_1_ADDR_MASK		GENMASK_ULL(63, 12)
+
+/* High-level queue structures */
+#define ARM_SMMU_POLL_TIMEOUT_US	1000000 /* 1s! */
+#define ARM_SMMU_POLL_SPIN_COUNT	10
+
+#define MSI_IOVA_BASE			0x8000000
+#define MSI_IOVA_LENGTH			0x100000
+
+static bool disable_bypass = 1;
+module_param_named(disable_bypass, disable_bypass, bool, S_IRUGO);
+MODULE_PARM_DESC(disable_bypass,
+	"Disable bypass streams such that incoming transactions from devices that are not attached to an iommu domain will report an abort back to the device and will not be allowed to pass through the SMMU.");
+
+enum pri_resp {
+	PRI_RESP_DENY = 0,
+	PRI_RESP_FAIL = 1,
+	PRI_RESP_SUCC = 2,
+};
+
+enum arm_smmu_msi_index {
+	EVTQ_MSI_INDEX,
+	GERROR_MSI_INDEX,
+	PRIQ_MSI_INDEX,
+	ARM_SMMU_MAX_MSIS,
+};
+
+static phys_addr_t arm_smmu_msi_cfg[ARM_SMMU_MAX_MSIS][3] = {
+	[EVTQ_MSI_INDEX] = {
+		ARM_SMMU_EVTQ_IRQ_CFG0,
+		ARM_SMMU_EVTQ_IRQ_CFG1,
+		ARM_SMMU_EVTQ_IRQ_CFG2,
+	},
+	[GERROR_MSI_INDEX] = {
+		ARM_SMMU_GERROR_IRQ_CFG0,
+		ARM_SMMU_GERROR_IRQ_CFG1,
+		ARM_SMMU_GERROR_IRQ_CFG2,
+	},
+	[PRIQ_MSI_INDEX] = {
+		ARM_SMMU_PRIQ_IRQ_CFG0,
+		ARM_SMMU_PRIQ_IRQ_CFG1,
+		ARM_SMMU_PRIQ_IRQ_CFG2,
+	},
+};
+
+struct arm_smmu_cmdq_ent {
+	/* Common fields */
+	u8				opcode;
+	bool				substream_valid;
+
+	/* Command-specific fields */
+	union {
+		#define CMDQ_OP_PREFETCH_CFG	0x1
+		struct {
+			u32			sid;
+			u8			size;
+			u64			addr;
+		} prefetch;
+
+		#define CMDQ_OP_CFGI_STE	0x3
+		#define CMDQ_OP_CFGI_ALL	0x4
+		#define CMDQ_OP_CFGI_CD		0x5
+		#define CMDQ_OP_CFGI_CD_ALL	0x6
+		struct {
+			u32			sid;
+			u32			ssid;
+			union {
+				bool		leaf;
+				u8		span;
+			};
+		} cfgi;
+
+		#define CMDQ_OP_TLBI_NH_ASID	0x11
+		#define CMDQ_OP_TLBI_NH_VA	0x12
+		#define CMDQ_OP_TLBI_EL2_ALL	0x20
+		#define CMDQ_OP_TLBI_S12_VMALL	0x28
+		#define CMDQ_OP_TLBI_S2_IPA	0x2a
+		#define CMDQ_OP_TLBI_NSNH_ALL	0x30
+		struct {
+			u8			num;
+			u8			scale;
+			u16			asid;
+			u16			vmid;
+			bool			leaf;
+			u8			ttl;
+			u8			tg;
+			u64			addr;
+		} tlbi;
+
+		#define CMDQ_OP_ATC_INV		0x40
+		#define ATC_INV_SIZE_ALL	52
+		struct {
+			u32			sid;
+			u32			ssid;
+			u64			addr;
+			u8			size;
+			bool			global;
+		} atc;
+
+		#define CMDQ_OP_PRI_RESP	0x41
+		struct {
+			u32			sid;
+			u32			ssid;
+			u16			grpid;
+			enum pri_resp		resp;
+		} pri;
+
+		#define CMDQ_OP_CMD_SYNC	0x46
+		struct {
+			u64			msiaddr;
+		} sync;
+	};
+};
+
+struct arm_smmu_ll_queue {
+	union {
+		u64			val;
+		struct {
+			u32		prod;
+			u32		cons;
+		};
+		struct {
+			atomic_t	prod;
+			atomic_t	cons;
+		} atomic;
+		u8			__pad[SMP_CACHE_BYTES];
+	} ____cacheline_aligned_in_smp;
+	u32				max_n_shift;
+};
+
+struct arm_smmu_queue {
+	struct arm_smmu_ll_queue	llq;
+	int				irq; /* Wired interrupt */
+
+	__le64				*base;
+	dma_addr_t			base_dma;
+	u64				q_base;
+
+	size_t				ent_dwords;
+
+	u32 __iomem			*prod_reg;
+	u32 __iomem			*cons_reg;
+};
+
+struct arm_smmu_queue_poll {
+	ktime_t				timeout;
+	unsigned int			delay;
+	unsigned int			spin_cnt;
+	bool				wfe;
+};
+
+struct arm_smmu_cmdq {
+	struct arm_smmu_queue		q;
+	atomic_long_t			*valid_map;
+	atomic_t			owner_prod;
+	atomic_t			lock;
+};
+
+struct arm_smmu_cmdq_batch {
+	u64				cmds[CMDQ_BATCH_ENTRIES * CMDQ_ENT_DWORDS];
+	int				num;
+};
+
+struct arm_smmu_evtq {
+	struct arm_smmu_queue		q;
+	u32				max_stalls;
+};
+
+struct arm_smmu_priq {
+	struct arm_smmu_queue		q;
+};
+
+/* High-level stream table and context descriptor structures */
+struct arm_smmu_strtab_l1_desc {
+	u8				span;
+
+	__le64				*l2ptr;
+	dma_addr_t			l2ptr_dma;
+};
+
+struct arm_smmu_ctx_desc {
+	u16				asid;
+	u64				ttbr;
+	u64				tcr;
+	u64				mair;
+};
+
+struct arm_smmu_l1_ctx_desc {
+	__le64				*l2ptr;
+	dma_addr_t			l2ptr_dma;
+};
+
+struct arm_smmu_ctx_desc_cfg {
+	__le64				*cdtab;
+	dma_addr_t			cdtab_dma;
+	struct arm_smmu_l1_ctx_desc	*l1_desc;
+	unsigned int			num_l1_ents;
+};
+
+struct arm_smmu_s1_cfg {
+	struct arm_smmu_ctx_desc_cfg	cdcfg;
+	struct arm_smmu_ctx_desc	cd;
+	u8				s1fmt;
+	u8				s1cdmax;
+};
+
+struct arm_smmu_s2_cfg {
+	u16				vmid;
+	u64				vttbr;
+	u64				vtcr;
+};
+
+struct arm_smmu_strtab_cfg {
+	__le64				*strtab;
+	dma_addr_t			strtab_dma;
+	struct arm_smmu_strtab_l1_desc	*l1_desc;
+	unsigned int			num_l1_ents;
+
+	u64				strtab_base;
+	u32				strtab_base_cfg;
+};
+
+/* An SMMUv3 instance */
+struct arm_smmu_device {
+	struct device			*dev;
+	void __iomem			*base;
+	void __iomem			*page1;
+
+#define ARM_SMMU_FEAT_2_LVL_STRTAB	(1 << 0)
+#define ARM_SMMU_FEAT_2_LVL_CDTAB	(1 << 1)
+#define ARM_SMMU_FEAT_TT_LE		(1 << 2)
+#define ARM_SMMU_FEAT_TT_BE		(1 << 3)
+#define ARM_SMMU_FEAT_PRI		(1 << 4)
+#define ARM_SMMU_FEAT_ATS		(1 << 5)
+#define ARM_SMMU_FEAT_SEV		(1 << 6)
+#define ARM_SMMU_FEAT_MSI		(1 << 7)
+#define ARM_SMMU_FEAT_COHERENCY		(1 << 8)
+#define ARM_SMMU_FEAT_TRANS_S1		(1 << 9)
+#define ARM_SMMU_FEAT_TRANS_S2		(1 << 10)
+#define ARM_SMMU_FEAT_STALLS		(1 << 11)
+#define ARM_SMMU_FEAT_HYP		(1 << 12)
+#define ARM_SMMU_FEAT_STALL_FORCE	(1 << 13)
+#define ARM_SMMU_FEAT_VAX		(1 << 14)
+#define ARM_SMMU_FEAT_RANGE_INV		(1 << 15)
+	u32				features;
+
+#define ARM_SMMU_OPT_SKIP_PREFETCH	(1 << 0)
+#define ARM_SMMU_OPT_PAGE0_REGS_ONLY	(1 << 1)
+	u32				options;
+
+	struct arm_smmu_cmdq		cmdq;
+	struct arm_smmu_evtq		evtq;
+	struct arm_smmu_priq		priq;
+
+	int				gerr_irq;
+	int				combined_irq;
+
+	unsigned long			ias; /* IPA */
+	unsigned long			oas; /* PA */
+	unsigned long			pgsize_bitmap;
+
+#define ARM_SMMU_MAX_ASIDS		(1 << 16)
+	unsigned int			asid_bits;
+
+#define ARM_SMMU_MAX_VMIDS		(1 << 16)
+	unsigned int			vmid_bits;
+	DECLARE_BITMAP(vmid_map, ARM_SMMU_MAX_VMIDS);
+
+	unsigned int			ssid_bits;
+	unsigned int			sid_bits;
+
+	struct arm_smmu_strtab_cfg	strtab_cfg;
+
+	/* IOMMU core code handle */
+	struct iommu_device		iommu;
+};
+
+/* SMMU private data for each master */
+struct arm_smmu_master {
+	struct arm_smmu_device		*smmu;
+	struct device			*dev;
+	struct arm_smmu_domain		*domain;
+	struct list_head		domain_head;
+	u32				*sids;
+	unsigned int			num_sids;
+	bool				ats_enabled;
+	unsigned int			ssid_bits;
+};
+
+/* SMMU private data for an IOMMU domain */
+enum arm_smmu_domain_stage {
+	ARM_SMMU_DOMAIN_S1 = 0,
+	ARM_SMMU_DOMAIN_S2,
+	ARM_SMMU_DOMAIN_NESTED,
+	ARM_SMMU_DOMAIN_BYPASS,
+};
+
+struct arm_smmu_domain {
+	struct arm_smmu_device		*smmu;
+	struct mutex			init_mutex; /* Protects smmu pointer */
+
+	struct io_pgtable_ops		*pgtbl_ops;
+	bool				non_strict;
+	atomic_t			nr_ats_masters;
+
+	enum arm_smmu_domain_stage	stage;
+	union {
+		struct arm_smmu_s1_cfg	s1_cfg;
+		struct arm_smmu_s2_cfg	s2_cfg;
+	};
+
+	struct iommu_domain		domain;
+
+	struct list_head		devices;
+	spinlock_t			devices_lock;
+};
+
+struct arm_smmu_option_prop {
+	u32 opt;
+	const char *prop;
+};
+
+static DEFINE_XARRAY_ALLOC1(asid_xa);
+
+static struct arm_smmu_option_prop arm_smmu_options[] = {
+	{ ARM_SMMU_OPT_SKIP_PREFETCH, "hisilicon,broken-prefetch-cmd" },
+	{ ARM_SMMU_OPT_PAGE0_REGS_ONLY, "cavium,cn9900-broken-page1-regspace"},
+	{ 0, NULL},
+};
+
+static inline void __iomem *arm_smmu_page1_fixup(unsigned long offset,
+						 struct arm_smmu_device *smmu)
+{
+	if (offset > SZ_64K)
+		return smmu->page1 + offset - SZ_64K;
+
+	return smmu->base + offset;
+}
+
+static struct arm_smmu_domain *to_smmu_domain(struct iommu_domain *dom)
+{
+	return container_of(dom, struct arm_smmu_domain, domain);
+}
+
+static void parse_driver_options(struct arm_smmu_device *smmu)
+{
+	int i = 0;
+
+	do {
+		if (of_property_read_bool(smmu->dev->of_node,
+						arm_smmu_options[i].prop)) {
+			smmu->options |= arm_smmu_options[i].opt;
+			dev_notice(smmu->dev, "option %s\n",
+				arm_smmu_options[i].prop);
+		}
+	} while (arm_smmu_options[++i].opt);
+}
+
+/* Low-level queue manipulation functions */
+static bool queue_has_space(struct arm_smmu_ll_queue *q, u32 n)
+{
+	u32 space, prod, cons;
+
+	prod = Q_IDX(q, q->prod);
+	cons = Q_IDX(q, q->cons);
+
+	if (Q_WRP(q, q->prod) == Q_WRP(q, q->cons))
+		space = (1 << q->max_n_shift) - (prod - cons);
+	else
+		space = cons - prod;
+
+	return space >= n;
+}
+
+static bool queue_full(struct arm_smmu_ll_queue *q)
+{
+	return Q_IDX(q, q->prod) == Q_IDX(q, q->cons) &&
+	       Q_WRP(q, q->prod) != Q_WRP(q, q->cons);
+}
+
+static bool queue_empty(struct arm_smmu_ll_queue *q)
+{
+	return Q_IDX(q, q->prod) == Q_IDX(q, q->cons) &&
+	       Q_WRP(q, q->prod) == Q_WRP(q, q->cons);
+}
+
+static bool queue_consumed(struct arm_smmu_ll_queue *q, u32 prod)
+{
+	return ((Q_WRP(q, q->cons) == Q_WRP(q, prod)) &&
+		(Q_IDX(q, q->cons) > Q_IDX(q, prod))) ||
+	       ((Q_WRP(q, q->cons) != Q_WRP(q, prod)) &&
+		(Q_IDX(q, q->cons) <= Q_IDX(q, prod)));
+}
+
+static void queue_sync_cons_out(struct arm_smmu_queue *q)
+{
+	/*
+	 * Ensure that all CPU accesses (reads and writes) to the queue
+	 * are complete before we update the cons pointer.
+	 */
+	mb();
+	writel_relaxed(q->llq.cons, q->cons_reg);
+}
+
+static void queue_inc_cons(struct arm_smmu_ll_queue *q)
+{
+	u32 cons = (Q_WRP(q, q->cons) | Q_IDX(q, q->cons)) + 1;
+	q->cons = Q_OVF(q->cons) | Q_WRP(q, cons) | Q_IDX(q, cons);
+}
+
+static int queue_sync_prod_in(struct arm_smmu_queue *q)
+{
+	int ret = 0;
+	u32 prod = readl_relaxed(q->prod_reg);
+
+	if (Q_OVF(prod) != Q_OVF(q->llq.prod))
+		ret = -EOVERFLOW;
+
+	q->llq.prod = prod;
+	return ret;
+}
+
+static u32 queue_inc_prod_n(struct arm_smmu_ll_queue *q, int n)
+{
+	u32 prod = (Q_WRP(q, q->prod) | Q_IDX(q, q->prod)) + n;
+	return Q_OVF(q->prod) | Q_WRP(q, prod) | Q_IDX(q, prod);
+}
+
+static void queue_poll_init(struct arm_smmu_device *smmu,
+			    struct arm_smmu_queue_poll *qp)
+{
+	qp->delay = 1;
+	qp->spin_cnt = 0;
+	qp->wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
+	qp->timeout = ktime_add_us(ktime_get(), ARM_SMMU_POLL_TIMEOUT_US);
+}
+
+static int queue_poll(struct arm_smmu_queue_poll *qp)
+{
+	if (ktime_compare(ktime_get(), qp->timeout) > 0)
+		return -ETIMEDOUT;
+
+	if (qp->wfe) {
+		wfe();
+	} else if (++qp->spin_cnt < ARM_SMMU_POLL_SPIN_COUNT) {
+		cpu_relax();
+	} else {
+		udelay(qp->delay);
+		qp->delay *= 2;
+		qp->spin_cnt = 0;
+	}
+
+	return 0;
+}
+
+static void queue_write(__le64 *dst, u64 *src, size_t n_dwords)
+{
+	int i;
+
+	for (i = 0; i < n_dwords; ++i)
+		*dst++ = cpu_to_le64(*src++);
+}
+
+static void queue_read(__le64 *dst, u64 *src, size_t n_dwords)
+{
+	int i;
+
+	for (i = 0; i < n_dwords; ++i)
+		*dst++ = le64_to_cpu(*src++);
+}
+
+static int queue_remove_raw(struct arm_smmu_queue *q, u64 *ent)
+{
+	if (queue_empty(&q->llq))
+		return -EAGAIN;
+
+	queue_read(ent, Q_ENT(q, q->llq.cons), q->ent_dwords);
+	queue_inc_cons(&q->llq);
+	queue_sync_cons_out(q);
+	return 0;
+}
+
+/* High-level queue accessors */
+static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
+{
+	memset(cmd, 0, 1 << CMDQ_ENT_SZ_SHIFT);
+	cmd[0] |= FIELD_PREP(CMDQ_0_OP, ent->opcode);
+
+	switch (ent->opcode) {
+	case CMDQ_OP_TLBI_EL2_ALL:
+	case CMDQ_OP_TLBI_NSNH_ALL:
+		break;
+	case CMDQ_OP_PREFETCH_CFG:
+		cmd[0] |= FIELD_PREP(CMDQ_PREFETCH_0_SID, ent->prefetch.sid);
+		cmd[1] |= FIELD_PREP(CMDQ_PREFETCH_1_SIZE, ent->prefetch.size);
+		cmd[1] |= ent->prefetch.addr & CMDQ_PREFETCH_1_ADDR_MASK;
+		break;
+	case CMDQ_OP_CFGI_CD:
+		cmd[0] |= FIELD_PREP(CMDQ_CFGI_0_SSID, ent->cfgi.ssid);
+		fallthrough;
+	case CMDQ_OP_CFGI_STE:
+		cmd[0] |= FIELD_PREP(CMDQ_CFGI_0_SID, ent->cfgi.sid);
+		cmd[1] |= FIELD_PREP(CMDQ_CFGI_1_LEAF, ent->cfgi.leaf);
+		break;
+	case CMDQ_OP_CFGI_CD_ALL:
+		cmd[0] |= FIELD_PREP(CMDQ_CFGI_0_SID, ent->cfgi.sid);
+		break;
+	case CMDQ_OP_CFGI_ALL:
+		/* Cover the entire SID range */
+		cmd[1] |= FIELD_PREP(CMDQ_CFGI_1_RANGE, 31);
+		break;
+	case CMDQ_OP_TLBI_NH_VA:
+		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_NUM, ent->tlbi.num);
+		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_SCALE, ent->tlbi.scale);
+		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
+		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_ASID, ent->tlbi.asid);
+		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_LEAF, ent->tlbi.leaf);
+		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TTL, ent->tlbi.ttl);
+		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TG, ent->tlbi.tg);
+		cmd[1] |= ent->tlbi.addr & CMDQ_TLBI_1_VA_MASK;
+		break;
+	case CMDQ_OP_TLBI_S2_IPA:
+		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_NUM, ent->tlbi.num);
+		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_SCALE, ent->tlbi.scale);
+		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
+		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_LEAF, ent->tlbi.leaf);
+		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TTL, ent->tlbi.ttl);
+		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TG, ent->tlbi.tg);
+		cmd[1] |= ent->tlbi.addr & CMDQ_TLBI_1_IPA_MASK;
+		break;
+	case CMDQ_OP_TLBI_NH_ASID:
+		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_ASID, ent->tlbi.asid);
+		fallthrough;
+	case CMDQ_OP_TLBI_S12_VMALL:
+		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
+		break;
+	case CMDQ_OP_ATC_INV:
+		cmd[0] |= FIELD_PREP(CMDQ_0_SSV, ent->substream_valid);
+		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_GLOBAL, ent->atc.global);
+		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_SSID, ent->atc.ssid);
+		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_SID, ent->atc.sid);
+		cmd[1] |= FIELD_PREP(CMDQ_ATC_1_SIZE, ent->atc.size);
+		cmd[1] |= ent->atc.addr & CMDQ_ATC_1_ADDR_MASK;
+		break;
+	case CMDQ_OP_PRI_RESP:
+		cmd[0] |= FIELD_PREP(CMDQ_0_SSV, ent->substream_valid);
+		cmd[0] |= FIELD_PREP(CMDQ_PRI_0_SSID, ent->pri.ssid);
+		cmd[0] |= FIELD_PREP(CMDQ_PRI_0_SID, ent->pri.sid);
+		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_GRPID, ent->pri.grpid);
+		switch (ent->pri.resp) {
+		case PRI_RESP_DENY:
+		case PRI_RESP_FAIL:
+		case PRI_RESP_SUCC:
+			break;
+		default:
+			return -EINVAL;
+		}
+		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_RESP, ent->pri.resp);
+		break;
+	case CMDQ_OP_CMD_SYNC:
+		if (ent->sync.msiaddr) {
+			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_IRQ);
+			cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
+		} else {
+			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
+		}
+		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSH, ARM_SMMU_SH_ISH);
+		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIATTR, ARM_SMMU_MEMATTR_OIWB);
+		break;
+	default:
+		return -ENOENT;
+	}
+
+	return 0;
+}
+
+static void arm_smmu_cmdq_build_sync_cmd(u64 *cmd, struct arm_smmu_device *smmu,
+					 u32 prod)
+{
+	struct arm_smmu_queue *q = &smmu->cmdq.q;
+	struct arm_smmu_cmdq_ent ent = {
+		.opcode = CMDQ_OP_CMD_SYNC,
+	};
+
+	/*
+	 * Beware that Hi16xx adds an extra 32 bits of goodness to its MSI
+	 * payload, so the write will zero the entire command on that platform.
+	 */
+	if (smmu->features & ARM_SMMU_FEAT_MSI &&
+	    smmu->features & ARM_SMMU_FEAT_COHERENCY) {
+		ent.sync.msiaddr = q->base_dma + Q_IDX(&q->llq, prod) *
+				   q->ent_dwords * 8;
+	}
+
+	arm_smmu_cmdq_build_cmd(cmd, &ent);
+}
+
+static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
+{
+	static const char *cerror_str[] = {
+		[CMDQ_ERR_CERROR_NONE_IDX]	= "No error",
+		[CMDQ_ERR_CERROR_ILL_IDX]	= "Illegal command",
+		[CMDQ_ERR_CERROR_ABT_IDX]	= "Abort on command fetch",
+		[CMDQ_ERR_CERROR_ATC_INV_IDX]	= "ATC invalidate timeout",
+	};
+
+	int i;
+	u64 cmd[CMDQ_ENT_DWORDS];
+	struct arm_smmu_queue *q = &smmu->cmdq.q;
+	u32 cons = readl_relaxed(q->cons_reg);
+	u32 idx = FIELD_GET(CMDQ_CONS_ERR, cons);
+	struct arm_smmu_cmdq_ent cmd_sync = {
+		.opcode = CMDQ_OP_CMD_SYNC,
+	};
+
+	dev_err(smmu->dev, "CMDQ error (cons 0x%08x): %s\n", cons,
+		idx < ARRAY_SIZE(cerror_str) ?  cerror_str[idx] : "Unknown");
+
+	switch (idx) {
+	case CMDQ_ERR_CERROR_ABT_IDX:
+		dev_err(smmu->dev, "retrying command fetch\n");
+	case CMDQ_ERR_CERROR_NONE_IDX:
+		return;
+	case CMDQ_ERR_CERROR_ATC_INV_IDX:
+		/*
+		 * ATC Invalidation Completion timeout. CONS is still pointing
+		 * at the CMD_SYNC. Attempt to complete other pending commands
+		 * by repeating the CMD_SYNC, though we might well end up back
+		 * here since the ATC invalidation may still be pending.
+		 */
+		return;
+	case CMDQ_ERR_CERROR_ILL_IDX:
+	default:
+		break;
+	}
+
+	/*
+	 * We may have concurrent producers, so we need to be careful
+	 * not to touch any of the shadow cmdq state.
+	 */
+	queue_read(cmd, Q_ENT(q, cons), q->ent_dwords);
+	dev_err(smmu->dev, "skipping command in error state:\n");
+	for (i = 0; i < ARRAY_SIZE(cmd); ++i)
+		dev_err(smmu->dev, "\t0x%016llx\n", (unsigned long long)cmd[i]);
+
+	/* Convert the erroneous command into a CMD_SYNC */
+	if (arm_smmu_cmdq_build_cmd(cmd, &cmd_sync)) {
+		dev_err(smmu->dev, "failed to convert to CMD_SYNC\n");
+		return;
+	}
+
+	queue_write(Q_ENT(q, cons), cmd, q->ent_dwords);
+}
+
+/*
+ * Command queue locking.
+ * This is a form of bastardised rwlock with the following major changes:
+ *
+ * - The only LOCK routines are exclusive_trylock() and shared_lock().
+ *   Neither have barrier semantics, and instead provide only a control
+ *   dependency.
+ *
+ * - The UNLOCK routines are supplemented with shared_tryunlock(), which
+ *   fails if the caller appears to be the last lock holder (yes, this is
+ *   racy). All successful UNLOCK routines have RELEASE semantics.
+ */
+static void arm_smmu_cmdq_shared_lock(struct arm_smmu_cmdq *cmdq)
+{
+	int val;
+
+	/*
+	 * We can try to avoid the cmpxchg() loop by simply incrementing the
+	 * lock counter. When held in exclusive state, the lock counter is set
+	 * to INT_MIN so these increments won't hurt as the value will remain
+	 * negative.
+	 */
+	if (atomic_fetch_inc_relaxed(&cmdq->lock) >= 0)
+		return;
+
+	do {
+		val = atomic_cond_read_relaxed(&cmdq->lock, VAL >= 0);
+	} while (atomic_cmpxchg_relaxed(&cmdq->lock, val, val + 1) != val);
+}
+
+static void arm_smmu_cmdq_shared_unlock(struct arm_smmu_cmdq *cmdq)
+{
+	(void)atomic_dec_return_release(&cmdq->lock);
+}
+
+static bool arm_smmu_cmdq_shared_tryunlock(struct arm_smmu_cmdq *cmdq)
+{
+	if (atomic_read(&cmdq->lock) == 1)
+		return false;
+
+	arm_smmu_cmdq_shared_unlock(cmdq);
+	return true;
+}
+
+#define arm_smmu_cmdq_exclusive_trylock_irqsave(cmdq, flags)		\
+({									\
+	bool __ret;							\
+	local_irq_save(flags);						\
+	__ret = !atomic_cmpxchg_relaxed(&cmdq->lock, 0, INT_MIN);	\
+	if (!__ret)							\
+		local_irq_restore(flags);				\
+	__ret;								\
+})
+
+#define arm_smmu_cmdq_exclusive_unlock_irqrestore(cmdq, flags)		\
+({									\
+	atomic_set_release(&cmdq->lock, 0);				\
+	local_irq_restore(flags);					\
+})
+
+
+/*
+ * Command queue insertion.
+ * This is made fiddly by our attempts to achieve some sort of scalability
+ * since there is one queue shared amongst all of the CPUs in the system.  If
+ * you like mixed-size concurrency, dependency ordering and relaxed atomics,
+ * then you'll *love* this monstrosity.
+ *
+ * The basic idea is to split the queue up into ranges of commands that are
+ * owned by a given CPU; the owner may not have written all of the commands
+ * itself, but is responsible for advancing the hardware prod pointer when
+ * the time comes. The algorithm is roughly:
+ *
+ * 	1. Allocate some space in the queue. At this point we also discover
+ *	   whether the head of the queue is currently owned by another CPU,
+ *	   or whether we are the owner.
+ *
+ *	2. Write our commands into our allocated slots in the queue.
+ *
+ *	3. Mark our slots as valid in arm_smmu_cmdq.valid_map.
+ *
+ *	4. If we are an owner:
+ *		a. Wait for the previous owner to finish.
+ *		b. Mark the queue head as unowned, which tells us the range
+ *		   that we are responsible for publishing.
+ *		c. Wait for all commands in our owned range to become valid.
+ *		d. Advance the hardware prod pointer.
+ *		e. Tell the next owner we've finished.
+ *
+ *	5. If we are inserting a CMD_SYNC (we may or may not have been an
+ *	   owner), then we need to stick around until it has completed:
+ *		a. If we have MSIs, the SMMU can write back into the CMD_SYNC
+ *		   to clear the first 4 bytes.
+ *		b. Otherwise, we spin waiting for the hardware cons pointer to
+ *		   advance past our command.
+ *
+ * The devil is in the details, particularly the use of locking for handling
+ * SYNC completion and freeing up space in the queue before we think that it is
+ * full.
+ */
+static void __arm_smmu_cmdq_poll_set_valid_map(struct arm_smmu_cmdq *cmdq,
+					       u32 sprod, u32 eprod, bool set)
+{
+	u32 swidx, sbidx, ewidx, ebidx;
+	struct arm_smmu_ll_queue llq = {
+		.max_n_shift	= cmdq->q.llq.max_n_shift,
+		.prod		= sprod,
+	};
+
+	ewidx = BIT_WORD(Q_IDX(&llq, eprod));
+	ebidx = Q_IDX(&llq, eprod) % BITS_PER_LONG;
+
+	while (llq.prod != eprod) {
+		unsigned long mask;
+		atomic_long_t *ptr;
+		u32 limit = BITS_PER_LONG;
+
+		swidx = BIT_WORD(Q_IDX(&llq, llq.prod));
+		sbidx = Q_IDX(&llq, llq.prod) % BITS_PER_LONG;
+
+		ptr = &cmdq->valid_map[swidx];
+
+		if ((swidx == ewidx) && (sbidx < ebidx))
+			limit = ebidx;
+
+		mask = GENMASK(limit - 1, sbidx);
+
+		/*
+		 * The valid bit is the inverse of the wrap bit. This means
+		 * that a zero-initialised queue is invalid and, after marking
+		 * all entries as valid, they become invalid again when we
+		 * wrap.
+		 */
+		if (set) {
+			atomic_long_xor(mask, ptr);
+		} else { /* Poll */
+			unsigned long valid;
+
+			valid = (ULONG_MAX + !!Q_WRP(&llq, llq.prod)) & mask;
+			atomic_long_cond_read_relaxed(ptr, (VAL & mask) == valid);
+		}
+
+		llq.prod = queue_inc_prod_n(&llq, limit - sbidx);
+	}
+}
+
+/* Mark all entries in the range [sprod, eprod) as valid */
+static void arm_smmu_cmdq_set_valid_map(struct arm_smmu_cmdq *cmdq,
+					u32 sprod, u32 eprod)
+{
+	__arm_smmu_cmdq_poll_set_valid_map(cmdq, sprod, eprod, true);
+}
+
+/* Wait for all entries in the range [sprod, eprod) to become valid */
+static void arm_smmu_cmdq_poll_valid_map(struct arm_smmu_cmdq *cmdq,
+					 u32 sprod, u32 eprod)
+{
+	__arm_smmu_cmdq_poll_set_valid_map(cmdq, sprod, eprod, false);
+}
+
+/* Wait for the command queue to become non-full */
+static int arm_smmu_cmdq_poll_until_not_full(struct arm_smmu_device *smmu,
+					     struct arm_smmu_ll_queue *llq)
+{
+	unsigned long flags;
+	struct arm_smmu_queue_poll qp;
+	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
+	int ret = 0;
+
+	/*
+	 * Try to update our copy of cons by grabbing exclusive cmdq access. If
+	 * that fails, spin until somebody else updates it for us.
+	 */
+	if (arm_smmu_cmdq_exclusive_trylock_irqsave(cmdq, flags)) {
+		WRITE_ONCE(cmdq->q.llq.cons, readl_relaxed(cmdq->q.cons_reg));
+		arm_smmu_cmdq_exclusive_unlock_irqrestore(cmdq, flags);
+		llq->val = READ_ONCE(cmdq->q.llq.val);
+		return 0;
+	}
+
+	queue_poll_init(smmu, &qp);
+	do {
+		llq->val = READ_ONCE(smmu->cmdq.q.llq.val);
+		if (!queue_full(llq))
+			break;
+
+		ret = queue_poll(&qp);
+	} while (!ret);
+
+	return ret;
+}
+
+/*
+ * Wait until the SMMU signals a CMD_SYNC completion MSI.
+ * Must be called with the cmdq lock held in some capacity.
+ */
+static int __arm_smmu_cmdq_poll_until_msi(struct arm_smmu_device *smmu,
+					  struct arm_smmu_ll_queue *llq)
+{
+	int ret = 0;
+	struct arm_smmu_queue_poll qp;
+	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
+	u32 *cmd = (u32 *)(Q_ENT(&cmdq->q, llq->prod));
+
+	queue_poll_init(smmu, &qp);
+
+	/*
+	 * The MSI won't generate an event, since it's being written back
+	 * into the command queue.
+	 */
+	qp.wfe = false;
+	smp_cond_load_relaxed(cmd, !VAL || (ret = queue_poll(&qp)));
+	llq->cons = ret ? llq->prod : queue_inc_prod_n(llq, 1);
+	return ret;
+}
+
+/*
+ * Wait until the SMMU cons index passes llq->prod.
+ * Must be called with the cmdq lock held in some capacity.
+ */
+static int __arm_smmu_cmdq_poll_until_consumed(struct arm_smmu_device *smmu,
+					       struct arm_smmu_ll_queue *llq)
+{
+	struct arm_smmu_queue_poll qp;
+	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
+	u32 prod = llq->prod;
+	int ret = 0;
+
+	queue_poll_init(smmu, &qp);
+	llq->val = READ_ONCE(smmu->cmdq.q.llq.val);
+	do {
+		if (queue_consumed(llq, prod))
+			break;
+
+		ret = queue_poll(&qp);
+
+		/*
+		 * This needs to be a readl() so that our subsequent call
+		 * to arm_smmu_cmdq_shared_tryunlock() can fail accurately.
+		 *
+		 * Specifically, we need to ensure that we observe all
+		 * shared_lock()s by other CMD_SYNCs that share our owner,
+		 * so that a failing call to tryunlock() means that we're
+		 * the last one out and therefore we can safely advance
+		 * cmdq->q.llq.cons. Roughly speaking:
+		 *
+		 * CPU 0		CPU1			CPU2 (us)
+		 *
+		 * if (sync)
+		 * 	shared_lock();
+		 *
+		 * dma_wmb();
+		 * set_valid_map();
+		 *
+		 * 			if (owner) {
+		 *				poll_valid_map();
+		 *				<control dependency>
+		 *				writel(prod_reg);
+		 *
+		 *						readl(cons_reg);
+		 *						tryunlock();
+		 *
+		 * Requires us to see CPU 0's shared_lock() acquisition.
+		 */
+		llq->cons = readl(cmdq->q.cons_reg);
+	} while (!ret);
+
+	return ret;
+}
+
+static int arm_smmu_cmdq_poll_until_sync(struct arm_smmu_device *smmu,
+					 struct arm_smmu_ll_queue *llq)
+{
+	if (smmu->features & ARM_SMMU_FEAT_MSI &&
+	    smmu->features & ARM_SMMU_FEAT_COHERENCY)
+		return __arm_smmu_cmdq_poll_until_msi(smmu, llq);
+
+	return __arm_smmu_cmdq_poll_until_consumed(smmu, llq);
+}
+
+static void arm_smmu_cmdq_write_entries(struct arm_smmu_cmdq *cmdq, u64 *cmds,
+					u32 prod, int n)
+{
+	int i;
+	struct arm_smmu_ll_queue llq = {
+		.max_n_shift	= cmdq->q.llq.max_n_shift,
+		.prod		= prod,
+	};
+
+	for (i = 0; i < n; ++i) {
+		u64 *cmd = &cmds[i * CMDQ_ENT_DWORDS];
+
+		prod = queue_inc_prod_n(&llq, i);
+		queue_write(Q_ENT(&cmdq->q, prod), cmd, CMDQ_ENT_DWORDS);
+	}
+}
+
+/*
+ * This is the actual insertion function, and provides the following
+ * ordering guarantees to callers:
+ *
+ * - There is a dma_wmb() before publishing any commands to the queue.
+ *   This can be relied upon to order prior writes to data structures
+ *   in memory (such as a CD or an STE) before the command.
+ *
+ * - On completion of a CMD_SYNC, there is a control dependency.
+ *   This can be relied upon to order subsequent writes to memory (e.g.
+ *   freeing an IOVA) after completion of the CMD_SYNC.
+ *
+ * - Command insertion is totally ordered, so if two CPUs each race to
+ *   insert their own list of commands then all of the commands from one
+ *   CPU will appear before any of the commands from the other CPU.
+ */
+static int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
+				       u64 *cmds, int n, bool sync)
+{
+	u64 cmd_sync[CMDQ_ENT_DWORDS];
+	u32 prod;
+	unsigned long flags;
+	bool owner;
+	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
+	struct arm_smmu_ll_queue llq = {
+		.max_n_shift = cmdq->q.llq.max_n_shift,
+	}, head = llq;
+	int ret = 0;
+
+	/* 1. Allocate some space in the queue */
+	local_irq_save(flags);
+	llq.val = READ_ONCE(cmdq->q.llq.val);
+	do {
+		u64 old;
+
+		while (!queue_has_space(&llq, n + sync)) {
+			local_irq_restore(flags);
+			if (arm_smmu_cmdq_poll_until_not_full(smmu, &llq))
+				dev_err_ratelimited(smmu->dev, "CMDQ timeout\n");
+			local_irq_save(flags);
+		}
+
+		head.cons = llq.cons;
+		head.prod = queue_inc_prod_n(&llq, n + sync) |
+					     CMDQ_PROD_OWNED_FLAG;
+
+		old = cmpxchg_relaxed(&cmdq->q.llq.val, llq.val, head.val);
+		if (old == llq.val)
+			break;
+
+		llq.val = old;
+	} while (1);
+	owner = !(llq.prod & CMDQ_PROD_OWNED_FLAG);
+	head.prod &= ~CMDQ_PROD_OWNED_FLAG;
+	llq.prod &= ~CMDQ_PROD_OWNED_FLAG;
+
+	/*
+	 * 2. Write our commands into the queue
+	 * Dependency ordering from the cmpxchg() loop above.
+	 */
+	arm_smmu_cmdq_write_entries(cmdq, cmds, llq.prod, n);
+	if (sync) {
+		prod = queue_inc_prod_n(&llq, n);
+		arm_smmu_cmdq_build_sync_cmd(cmd_sync, smmu, prod);
+		queue_write(Q_ENT(&cmdq->q, prod), cmd_sync, CMDQ_ENT_DWORDS);
+
+		/*
+		 * In order to determine completion of our CMD_SYNC, we must
+		 * ensure that the queue can't wrap twice without us noticing.
+		 * We achieve that by taking the cmdq lock as shared before
+		 * marking our slot as valid.
+		 */
+		arm_smmu_cmdq_shared_lock(cmdq);
+	}
+
+	/* 3. Mark our slots as valid, ensuring commands are visible first */
+	dma_wmb();
+	arm_smmu_cmdq_set_valid_map(cmdq, llq.prod, head.prod);
+
+	/* 4. If we are the owner, take control of the SMMU hardware */
+	if (owner) {
+		/* a. Wait for previous owner to finish */
+		atomic_cond_read_relaxed(&cmdq->owner_prod, VAL == llq.prod);
+
+		/* b. Stop gathering work by clearing the owned flag */
+		prod = atomic_fetch_andnot_relaxed(CMDQ_PROD_OWNED_FLAG,
+						   &cmdq->q.llq.atomic.prod);
+		prod &= ~CMDQ_PROD_OWNED_FLAG;
+
+		/*
+		 * c. Wait for any gathered work to be written to the queue.
+		 * Note that we read our own entries so that we have the control
+		 * dependency required by (d).
+		 */
+		arm_smmu_cmdq_poll_valid_map(cmdq, llq.prod, prod);
+
+		/*
+		 * d. Advance the hardware prod pointer
+		 * Control dependency ordering from the entries becoming valid.
+		 */
+		writel_relaxed(prod, cmdq->q.prod_reg);
+
+		/*
+		 * e. Tell the next owner we're done
+		 * Make sure we've updated the hardware first, so that we don't
+		 * race to update prod and potentially move it backwards.
+		 */
+		atomic_set_release(&cmdq->owner_prod, prod);
+	}
+
+	/* 5. If we are inserting a CMD_SYNC, we must wait for it to complete */
+	if (sync) {
+		llq.prod = queue_inc_prod_n(&llq, n);
+		ret = arm_smmu_cmdq_poll_until_sync(smmu, &llq);
+		if (ret) {
+			dev_err_ratelimited(smmu->dev,
+					    "CMD_SYNC timeout at 0x%08x [hwprod 0x%08x, hwcons 0x%08x]\n",
+					    llq.prod,
+					    readl_relaxed(cmdq->q.prod_reg),
+					    readl_relaxed(cmdq->q.cons_reg));
+		}
+
+		/*
+		 * Try to unlock the cmdq lock. This will fail if we're the last
+		 * reader, in which case we can safely update cmdq->q.llq.cons
+		 */
+		if (!arm_smmu_cmdq_shared_tryunlock(cmdq)) {
+			WRITE_ONCE(cmdq->q.llq.cons, llq.cons);
+			arm_smmu_cmdq_shared_unlock(cmdq);
+		}
+	}
+
+	local_irq_restore(flags);
+	return ret;
+}
+
+static int arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
+				   struct arm_smmu_cmdq_ent *ent)
+{
+	u64 cmd[CMDQ_ENT_DWORDS];
+
+	if (arm_smmu_cmdq_build_cmd(cmd, ent)) {
+		dev_warn(smmu->dev, "ignoring unknown CMDQ opcode 0x%x\n",
+			 ent->opcode);
+		return -EINVAL;
+	}
+
+	return arm_smmu_cmdq_issue_cmdlist(smmu, cmd, 1, false);
+}
+
+static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
+{
+	return arm_smmu_cmdq_issue_cmdlist(smmu, NULL, 0, true);
+}
+
+static void arm_smmu_cmdq_batch_add(struct arm_smmu_device *smmu,
+				    struct arm_smmu_cmdq_batch *cmds,
+				    struct arm_smmu_cmdq_ent *cmd)
+{
+	if (cmds->num == CMDQ_BATCH_ENTRIES) {
+		arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, false);
+		cmds->num = 0;
+	}
+	arm_smmu_cmdq_build_cmd(&cmds->cmds[cmds->num * CMDQ_ENT_DWORDS], cmd);
+	cmds->num++;
+}
+
+static int arm_smmu_cmdq_batch_submit(struct arm_smmu_device *smmu,
+				      struct arm_smmu_cmdq_batch *cmds)
+{
+	return arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, true);
+}
+
+/* Context descriptor manipulation functions */
+static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
+			     int ssid, bool leaf)
+{
+	size_t i;
+	unsigned long flags;
+	struct arm_smmu_master *master;
+	struct arm_smmu_cmdq_batch cmds = {};
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_cmdq_ent cmd = {
+		.opcode	= CMDQ_OP_CFGI_CD,
+		.cfgi	= {
+			.ssid	= ssid,
+			.leaf	= leaf,
+		},
+	};
+
+	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
+	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
+		for (i = 0; i < master->num_sids; i++) {
+			cmd.cfgi.sid = master->sids[i];
+			arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
+		}
+	}
+	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+
+	arm_smmu_cmdq_batch_submit(smmu, &cmds);
+}
+
+static int arm_smmu_alloc_cd_leaf_table(struct arm_smmu_device *smmu,
+					struct arm_smmu_l1_ctx_desc *l1_desc)
+{
+	size_t size = CTXDESC_L2_ENTRIES * (CTXDESC_CD_DWORDS << 3);
+
+	l1_desc->l2ptr = dmam_alloc_coherent(smmu->dev, size,
+					     &l1_desc->l2ptr_dma, GFP_KERNEL);
+	if (!l1_desc->l2ptr) {
+		dev_warn(smmu->dev,
+			 "failed to allocate context descriptor table\n");
+		return -ENOMEM;
+	}
+	return 0;
+}
+
+static void arm_smmu_write_cd_l1_desc(__le64 *dst,
+				      struct arm_smmu_l1_ctx_desc *l1_desc)
+{
+	u64 val = (l1_desc->l2ptr_dma & CTXDESC_L1_DESC_L2PTR_MASK) |
+		  CTXDESC_L1_DESC_V;
+
+	/* See comment in arm_smmu_write_ctx_desc() */
+	WRITE_ONCE(*dst, cpu_to_le64(val));
+}
+
+static __le64 *arm_smmu_get_cd_ptr(struct arm_smmu_domain *smmu_domain,
+				   u32 ssid)
+{
+	__le64 *l1ptr;
+	unsigned int idx;
+	struct arm_smmu_l1_ctx_desc *l1_desc;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_ctx_desc_cfg *cdcfg = &smmu_domain->s1_cfg.cdcfg;
+
+	if (smmu_domain->s1_cfg.s1fmt == STRTAB_STE_0_S1FMT_LINEAR)
+		return cdcfg->cdtab + ssid * CTXDESC_CD_DWORDS;
+
+	idx = ssid >> CTXDESC_SPLIT;
+	l1_desc = &cdcfg->l1_desc[idx];
+	if (!l1_desc->l2ptr) {
+		if (arm_smmu_alloc_cd_leaf_table(smmu, l1_desc))
+			return NULL;
+
+		l1ptr = cdcfg->cdtab + idx * CTXDESC_L1_DESC_DWORDS;
+		arm_smmu_write_cd_l1_desc(l1ptr, l1_desc);
+		/* An invalid L1CD can be cached */
+		arm_smmu_sync_cd(smmu_domain, ssid, false);
+	}
+	idx = ssid & (CTXDESC_L2_ENTRIES - 1);
+	return l1_desc->l2ptr + idx * CTXDESC_CD_DWORDS;
+}
+
+static int arm_smmu_write_ctx_desc(struct arm_smmu_domain *smmu_domain,
+				   int ssid, struct arm_smmu_ctx_desc *cd)
+{
+	/*
+	 * This function handles the following cases:
+	 *
+	 * (1) Install primary CD, for normal DMA traffic (SSID = 0).
+	 * (2) Install a secondary CD, for SID+SSID traffic.
+	 * (3) Update ASID of a CD. Atomically write the first 64 bits of the
+	 *     CD, then invalidate the old entry and mappings.
+	 * (4) Remove a secondary CD.
+	 */
+	u64 val;
+	bool cd_live;
+	__le64 *cdptr;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+
+	if (WARN_ON(ssid >= (1 << smmu_domain->s1_cfg.s1cdmax)))
+		return -E2BIG;
+
+	cdptr = arm_smmu_get_cd_ptr(smmu_domain, ssid);
+	if (!cdptr)
+		return -ENOMEM;
+
+	val = le64_to_cpu(cdptr[0]);
+	cd_live = !!(val & CTXDESC_CD_0_V);
+
+	if (!cd) { /* (4) */
+		val = 0;
+	} else if (cd_live) { /* (3) */
+		val &= ~CTXDESC_CD_0_ASID;
+		val |= FIELD_PREP(CTXDESC_CD_0_ASID, cd->asid);
+		/*
+		 * Until CD+TLB invalidation, both ASIDs may be used for tagging
+		 * this substream's traffic
+		 */
+	} else { /* (1) and (2) */
+		cdptr[1] = cpu_to_le64(cd->ttbr & CTXDESC_CD_1_TTB0_MASK);
+		cdptr[2] = 0;
+		cdptr[3] = cpu_to_le64(cd->mair);
+
+		/*
+		 * STE is live, and the SMMU might read dwords of this CD in any
+		 * order. Ensure that it observes valid values before reading
+		 * V=1.
+		 */
+		arm_smmu_sync_cd(smmu_domain, ssid, true);
+
+		val = cd->tcr |
+#ifdef __BIG_ENDIAN
+			CTXDESC_CD_0_ENDI |
+#endif
+			CTXDESC_CD_0_R | CTXDESC_CD_0_A | CTXDESC_CD_0_ASET |
+			CTXDESC_CD_0_AA64 |
+			FIELD_PREP(CTXDESC_CD_0_ASID, cd->asid) |
+			CTXDESC_CD_0_V;
+
+		/* STALL_MODEL==0b10 && CD.S==0 is ILLEGAL */
+		if (smmu->features & ARM_SMMU_FEAT_STALL_FORCE)
+			val |= CTXDESC_CD_0_S;
+	}
+
+	/*
+	 * The SMMU accesses 64-bit values atomically. See IHI0070Ca 3.21.3
+	 * "Configuration structures and configuration invalidation completion"
+	 *
+	 *   The size of single-copy atomic reads made by the SMMU is
+	 *   IMPLEMENTATION DEFINED but must be at least 64 bits. Any single
+	 *   field within an aligned 64-bit span of a structure can be altered
+	 *   without first making the structure invalid.
+	 */
+	WRITE_ONCE(cdptr[0], cpu_to_le64(val));
+	arm_smmu_sync_cd(smmu_domain, ssid, true);
+	return 0;
+}
+
+static int arm_smmu_alloc_cd_tables(struct arm_smmu_domain *smmu_domain)
+{
+	int ret;
+	size_t l1size;
+	size_t max_contexts;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
+	struct arm_smmu_ctx_desc_cfg *cdcfg = &cfg->cdcfg;
+
+	max_contexts = 1 << cfg->s1cdmax;
+
+	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB) ||
+	    max_contexts <= CTXDESC_L2_ENTRIES) {
+		cfg->s1fmt = STRTAB_STE_0_S1FMT_LINEAR;
+		cdcfg->num_l1_ents = max_contexts;
+
+		l1size = max_contexts * (CTXDESC_CD_DWORDS << 3);
+	} else {
+		cfg->s1fmt = STRTAB_STE_0_S1FMT_64K_L2;
+		cdcfg->num_l1_ents = DIV_ROUND_UP(max_contexts,
+						  CTXDESC_L2_ENTRIES);
+
+		cdcfg->l1_desc = devm_kcalloc(smmu->dev, cdcfg->num_l1_ents,
+					      sizeof(*cdcfg->l1_desc),
+					      GFP_KERNEL);
+		if (!cdcfg->l1_desc)
+			return -ENOMEM;
+
+		l1size = cdcfg->num_l1_ents * (CTXDESC_L1_DESC_DWORDS << 3);
+	}
+
+	cdcfg->cdtab = dmam_alloc_coherent(smmu->dev, l1size, &cdcfg->cdtab_dma,
+					   GFP_KERNEL);
+	if (!cdcfg->cdtab) {
+		dev_warn(smmu->dev, "failed to allocate context descriptor\n");
+		ret = -ENOMEM;
+		goto err_free_l1;
+	}
+
+	return 0;
+
+err_free_l1:
+	if (cdcfg->l1_desc) {
+		devm_kfree(smmu->dev, cdcfg->l1_desc);
+		cdcfg->l1_desc = NULL;
+	}
+	return ret;
+}
+
+static void arm_smmu_free_cd_tables(struct arm_smmu_domain *smmu_domain)
+{
+	int i;
+	size_t size, l1size;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_ctx_desc_cfg *cdcfg = &smmu_domain->s1_cfg.cdcfg;
+
+	if (cdcfg->l1_desc) {
+		size = CTXDESC_L2_ENTRIES * (CTXDESC_CD_DWORDS << 3);
+
+		for (i = 0; i < cdcfg->num_l1_ents; i++) {
+			if (!cdcfg->l1_desc[i].l2ptr)
+				continue;
+
+			dmam_free_coherent(smmu->dev, size,
+					   cdcfg->l1_desc[i].l2ptr,
+					   cdcfg->l1_desc[i].l2ptr_dma);
+		}
+		devm_kfree(smmu->dev, cdcfg->l1_desc);
+		cdcfg->l1_desc = NULL;
+
+		l1size = cdcfg->num_l1_ents * (CTXDESC_L1_DESC_DWORDS << 3);
+	} else {
+		l1size = cdcfg->num_l1_ents * (CTXDESC_CD_DWORDS << 3);
+	}
+
+	dmam_free_coherent(smmu->dev, l1size, cdcfg->cdtab, cdcfg->cdtab_dma);
+	cdcfg->cdtab_dma = 0;
+	cdcfg->cdtab = NULL;
+}
+
+static void arm_smmu_free_asid(struct arm_smmu_ctx_desc *cd)
+{
+	if (!cd->asid)
+		return;
+
+	xa_erase(&asid_xa, cd->asid);
+}
+
+/* Stream table manipulation functions */
+static void
+arm_smmu_write_strtab_l1_desc(__le64 *dst, struct arm_smmu_strtab_l1_desc *desc)
+{
+	u64 val = 0;
+
+	val |= FIELD_PREP(STRTAB_L1_DESC_SPAN, desc->span);
+	val |= desc->l2ptr_dma & STRTAB_L1_DESC_L2PTR_MASK;
+
+	/* See comment in arm_smmu_write_ctx_desc() */
+	WRITE_ONCE(*dst, cpu_to_le64(val));
+}
+
+static void arm_smmu_sync_ste_for_sid(struct arm_smmu_device *smmu, u32 sid)
+{
+	struct arm_smmu_cmdq_ent cmd = {
+		.opcode	= CMDQ_OP_CFGI_STE,
+		.cfgi	= {
+			.sid	= sid,
+			.leaf	= true,
+		},
+	};
+
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	arm_smmu_cmdq_issue_sync(smmu);
+}
+
+static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
+				      __le64 *dst)
+{
+	/*
+	 * This is hideously complicated, but we only really care about
+	 * three cases at the moment:
+	 *
+	 * 1. Invalid (all zero) -> bypass/fault (init)
+	 * 2. Bypass/fault -> translation/bypass (attach)
+	 * 3. Translation/bypass -> bypass/fault (detach)
+	 *
+	 * Given that we can't update the STE atomically and the SMMU
+	 * doesn't read the thing in a defined order, that leaves us
+	 * with the following maintenance requirements:
+	 *
+	 * 1. Update Config, return (init time STEs aren't live)
+	 * 2. Write everything apart from dword 0, sync, write dword 0, sync
+	 * 3. Update Config, sync
+	 */
+	u64 val = le64_to_cpu(dst[0]);
+	bool ste_live = false;
+	struct arm_smmu_device *smmu = NULL;
+	struct arm_smmu_s1_cfg *s1_cfg = NULL;
+	struct arm_smmu_s2_cfg *s2_cfg = NULL;
+	struct arm_smmu_domain *smmu_domain = NULL;
+	struct arm_smmu_cmdq_ent prefetch_cmd = {
+		.opcode		= CMDQ_OP_PREFETCH_CFG,
+		.prefetch	= {
+			.sid	= sid,
+		},
+	};
+
+	if (master) {
+		smmu_domain = master->domain;
+		smmu = master->smmu;
+	}
+
+	if (smmu_domain) {
+		switch (smmu_domain->stage) {
+		case ARM_SMMU_DOMAIN_S1:
+			s1_cfg = &smmu_domain->s1_cfg;
+			break;
+		case ARM_SMMU_DOMAIN_S2:
+		case ARM_SMMU_DOMAIN_NESTED:
+			s2_cfg = &smmu_domain->s2_cfg;
+			break;
+		default:
+			break;
+		}
+	}
+
+	if (val & STRTAB_STE_0_V) {
+		switch (FIELD_GET(STRTAB_STE_0_CFG, val)) {
+		case STRTAB_STE_0_CFG_BYPASS:
+			break;
+		case STRTAB_STE_0_CFG_S1_TRANS:
+		case STRTAB_STE_0_CFG_S2_TRANS:
+			ste_live = true;
+			break;
+		case STRTAB_STE_0_CFG_ABORT:
+			BUG_ON(!disable_bypass);
+			break;
+		default:
+			BUG(); /* STE corruption */
+		}
+	}
+
+	/* Nuke the existing STE_0 value, as we're going to rewrite it */
+	val = STRTAB_STE_0_V;
+
+	/* Bypass/fault */
+	if (!smmu_domain || !(s1_cfg || s2_cfg)) {
+		if (!smmu_domain && disable_bypass)
+			val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_ABORT);
+		else
+			val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_BYPASS);
+
+		dst[0] = cpu_to_le64(val);
+		dst[1] = cpu_to_le64(FIELD_PREP(STRTAB_STE_1_SHCFG,
+						STRTAB_STE_1_SHCFG_INCOMING));
+		dst[2] = 0; /* Nuke the VMID */
+		/*
+		 * The SMMU can perform negative caching, so we must sync
+		 * the STE regardless of whether the old value was live.
+		 */
+		if (smmu)
+			arm_smmu_sync_ste_for_sid(smmu, sid);
+		return;
+	}
+
+	if (s1_cfg) {
+		BUG_ON(ste_live);
+		dst[1] = cpu_to_le64(
+			 FIELD_PREP(STRTAB_STE_1_S1DSS, STRTAB_STE_1_S1DSS_SSID0) |
+			 FIELD_PREP(STRTAB_STE_1_S1CIR, STRTAB_STE_1_S1C_CACHE_WBRA) |
+			 FIELD_PREP(STRTAB_STE_1_S1COR, STRTAB_STE_1_S1C_CACHE_WBRA) |
+			 FIELD_PREP(STRTAB_STE_1_S1CSH, ARM_SMMU_SH_ISH) |
+			 FIELD_PREP(STRTAB_STE_1_STRW, STRTAB_STE_1_STRW_NSEL1));
+
+		if (smmu->features & ARM_SMMU_FEAT_STALLS &&
+		   !(smmu->features & ARM_SMMU_FEAT_STALL_FORCE))
+			dst[1] |= cpu_to_le64(STRTAB_STE_1_S1STALLD);
+
+		val |= (s1_cfg->cdcfg.cdtab_dma & STRTAB_STE_0_S1CTXPTR_MASK) |
+			FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_S1_TRANS) |
+			FIELD_PREP(STRTAB_STE_0_S1CDMAX, s1_cfg->s1cdmax) |
+			FIELD_PREP(STRTAB_STE_0_S1FMT, s1_cfg->s1fmt);
+	}
+
+	if (s2_cfg) {
+		BUG_ON(ste_live);
+		dst[2] = cpu_to_le64(
+			 FIELD_PREP(STRTAB_STE_2_S2VMID, s2_cfg->vmid) |
+			 FIELD_PREP(STRTAB_STE_2_VTCR, s2_cfg->vtcr) |
+#ifdef __BIG_ENDIAN
+			 STRTAB_STE_2_S2ENDI |
+#endif
+			 STRTAB_STE_2_S2PTW | STRTAB_STE_2_S2AA64 |
+			 STRTAB_STE_2_S2R);
+
+		dst[3] = cpu_to_le64(s2_cfg->vttbr & STRTAB_STE_3_S2TTB_MASK);
+
+		val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_S2_TRANS);
+	}
+
+	if (master->ats_enabled)
+		dst[1] |= cpu_to_le64(FIELD_PREP(STRTAB_STE_1_EATS,
+						 STRTAB_STE_1_EATS_TRANS));
+
+	arm_smmu_sync_ste_for_sid(smmu, sid);
+	/* See comment in arm_smmu_write_ctx_desc() */
+	WRITE_ONCE(dst[0], cpu_to_le64(val));
+	arm_smmu_sync_ste_for_sid(smmu, sid);
+
+	/* It's likely that we'll want to use the new STE soon */
+	if (!(smmu->options & ARM_SMMU_OPT_SKIP_PREFETCH))
+		arm_smmu_cmdq_issue_cmd(smmu, &prefetch_cmd);
+}
+
+static void arm_smmu_init_bypass_stes(u64 *strtab, unsigned int nent)
+{
+	unsigned int i;
+
+	for (i = 0; i < nent; ++i) {
+		arm_smmu_write_strtab_ent(NULL, -1, strtab);
+		strtab += STRTAB_STE_DWORDS;
+	}
+}
+
+static int arm_smmu_init_l2_strtab(struct arm_smmu_device *smmu, u32 sid)
+{
+	size_t size;
+	void *strtab;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+	struct arm_smmu_strtab_l1_desc *desc = &cfg->l1_desc[sid >> STRTAB_SPLIT];
+
+	if (desc->l2ptr)
+		return 0;
+
+	size = 1 << (STRTAB_SPLIT + ilog2(STRTAB_STE_DWORDS) + 3);
+	strtab = &cfg->strtab[(sid >> STRTAB_SPLIT) * STRTAB_L1_DESC_DWORDS];
+
+	desc->span = STRTAB_SPLIT + 1;
+	desc->l2ptr = dmam_alloc_coherent(smmu->dev, size, &desc->l2ptr_dma,
+					  GFP_KERNEL);
+	if (!desc->l2ptr) {
+		dev_err(smmu->dev,
+			"failed to allocate l2 stream table for SID %u\n",
+			sid);
+		return -ENOMEM;
+	}
+
+	arm_smmu_init_bypass_stes(desc->l2ptr, 1 << STRTAB_SPLIT);
+	arm_smmu_write_strtab_l1_desc(strtab, desc);
+	return 0;
+}
+
+/* IRQ and event handlers */
+static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
+{
+	int i;
+	struct arm_smmu_device *smmu = dev;
+	struct arm_smmu_queue *q = &smmu->evtq.q;
+	struct arm_smmu_ll_queue *llq = &q->llq;
+	u64 evt[EVTQ_ENT_DWORDS];
+
+	do {
+		while (!queue_remove_raw(q, evt)) {
+			u8 id = FIELD_GET(EVTQ_0_ID, evt[0]);
+
+			dev_info(smmu->dev, "event 0x%02x received:\n", id);
+			for (i = 0; i < ARRAY_SIZE(evt); ++i)
+				dev_info(smmu->dev, "\t0x%016llx\n",
+					 (unsigned long long)evt[i]);
+
+		}
+
+		/*
+		 * Not much we can do on overflow, so scream and pretend we're
+		 * trying harder.
+		 */
+		if (queue_sync_prod_in(q) == -EOVERFLOW)
+			dev_err(smmu->dev, "EVTQ overflow detected -- events lost\n");
+	} while (!queue_empty(llq));
+
+	/* Sync our overflow flag, as we believe we're up to speed */
+	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
+		    Q_IDX(llq, llq->cons);
+	return IRQ_HANDLED;
+}
+
+static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
+{
+	u32 sid, ssid;
+	u16 grpid;
+	bool ssv, last;
+
+	sid = FIELD_GET(PRIQ_0_SID, evt[0]);
+	ssv = FIELD_GET(PRIQ_0_SSID_V, evt[0]);
+	ssid = ssv ? FIELD_GET(PRIQ_0_SSID, evt[0]) : 0;
+	last = FIELD_GET(PRIQ_0_PRG_LAST, evt[0]);
+	grpid = FIELD_GET(PRIQ_1_PRG_IDX, evt[1]);
+
+	dev_info(smmu->dev, "unexpected PRI request received:\n");
+	dev_info(smmu->dev,
+		 "\tsid 0x%08x.0x%05x: [%u%s] %sprivileged %s%s%s access at iova 0x%016llx\n",
+		 sid, ssid, grpid, last ? "L" : "",
+		 evt[0] & PRIQ_0_PERM_PRIV ? "" : "un",
+		 evt[0] & PRIQ_0_PERM_READ ? "R" : "",
+		 evt[0] & PRIQ_0_PERM_WRITE ? "W" : "",
+		 evt[0] & PRIQ_0_PERM_EXEC ? "X" : "",
+		 evt[1] & PRIQ_1_ADDR_MASK);
+
+	if (last) {
+		struct arm_smmu_cmdq_ent cmd = {
+			.opcode			= CMDQ_OP_PRI_RESP,
+			.substream_valid	= ssv,
+			.pri			= {
+				.sid	= sid,
+				.ssid	= ssid,
+				.grpid	= grpid,
+				.resp	= PRI_RESP_DENY,
+			},
+		};
+
+		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	}
+}
+
+static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
+{
+	struct arm_smmu_device *smmu = dev;
+	struct arm_smmu_queue *q = &smmu->priq.q;
+	struct arm_smmu_ll_queue *llq = &q->llq;
+	u64 evt[PRIQ_ENT_DWORDS];
+
+	do {
+		while (!queue_remove_raw(q, evt))
+			arm_smmu_handle_ppr(smmu, evt);
+
+		if (queue_sync_prod_in(q) == -EOVERFLOW)
+			dev_err(smmu->dev, "PRIQ overflow detected -- requests lost\n");
+	} while (!queue_empty(llq));
+
+	/* Sync our overflow flag, as we believe we're up to speed */
+	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
+		      Q_IDX(llq, llq->cons);
+	queue_sync_cons_out(q);
+	return IRQ_HANDLED;
+}
+
+static int arm_smmu_device_disable(struct arm_smmu_device *smmu);
+
+static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
+{
+	u32 gerror, gerrorn, active;
+	struct arm_smmu_device *smmu = dev;
+
+	gerror = readl_relaxed(smmu->base + ARM_SMMU_GERROR);
+	gerrorn = readl_relaxed(smmu->base + ARM_SMMU_GERRORN);
+
+	active = gerror ^ gerrorn;
+	if (!(active & GERROR_ERR_MASK))
+		return IRQ_NONE; /* No errors pending */
+
+	dev_warn(smmu->dev,
+		 "unexpected global error reported (0x%08x), this could be serious\n",
+		 active);
+
+	if (active & GERROR_SFM_ERR) {
+		dev_err(smmu->dev, "device has entered Service Failure Mode!\n");
+		arm_smmu_device_disable(smmu);
+	}
+
+	if (active & GERROR_MSI_GERROR_ABT_ERR)
+		dev_warn(smmu->dev, "GERROR MSI write aborted\n");
+
+	if (active & GERROR_MSI_PRIQ_ABT_ERR)
+		dev_warn(smmu->dev, "PRIQ MSI write aborted\n");
+
+	if (active & GERROR_MSI_EVTQ_ABT_ERR)
+		dev_warn(smmu->dev, "EVTQ MSI write aborted\n");
+
+	if (active & GERROR_MSI_CMDQ_ABT_ERR)
+		dev_warn(smmu->dev, "CMDQ MSI write aborted\n");
+
+	if (active & GERROR_PRIQ_ABT_ERR)
+		dev_err(smmu->dev, "PRIQ write aborted -- events may have been lost\n");
+
+	if (active & GERROR_EVTQ_ABT_ERR)
+		dev_err(smmu->dev, "EVTQ write aborted -- events may have been lost\n");
+
+	if (active & GERROR_CMDQ_ERR)
+		arm_smmu_cmdq_skip_err(smmu);
+
+	writel(gerror, smmu->base + ARM_SMMU_GERRORN);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t arm_smmu_combined_irq_thread(int irq, void *dev)
+{
+	struct arm_smmu_device *smmu = dev;
+
+	arm_smmu_evtq_thread(irq, dev);
+	if (smmu->features & ARM_SMMU_FEAT_PRI)
+		arm_smmu_priq_thread(irq, dev);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
+{
+	arm_smmu_gerror_handler(irq, dev);
+	return IRQ_WAKE_THREAD;
+}
+
+static void
+arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size,
+			struct arm_smmu_cmdq_ent *cmd)
+{
+	size_t log2_span;
+	size_t span_mask;
+	/* ATC invalidates are always on 4096-bytes pages */
+	size_t inval_grain_shift = 12;
+	unsigned long page_start, page_end;
+
+	*cmd = (struct arm_smmu_cmdq_ent) {
+		.opcode			= CMDQ_OP_ATC_INV,
+		.substream_valid	= !!ssid,
+		.atc.ssid		= ssid,
+	};
+
+	if (!size) {
+		cmd->atc.size = ATC_INV_SIZE_ALL;
+		return;
+	}
+
+	page_start	= iova >> inval_grain_shift;
+	page_end	= (iova + size - 1) >> inval_grain_shift;
+
+	/*
+	 * In an ATS Invalidate Request, the address must be aligned on the
+	 * range size, which must be a power of two number of page sizes. We
+	 * thus have to choose between grossly over-invalidating the region, or
+	 * splitting the invalidation into multiple commands. For simplicity
+	 * we'll go with the first solution, but should refine it in the future
+	 * if multiple commands are shown to be more efficient.
+	 *
+	 * Find the smallest power of two that covers the range. The most
+	 * significant differing bit between the start and end addresses,
+	 * fls(start ^ end), indicates the required span. For example:
+	 *
+	 * We want to invalidate pages [8; 11]. This is already the ideal range:
+	 *		x = 0b1000 ^ 0b1011 = 0b11
+	 *		span = 1 << fls(x) = 4
+	 *
+	 * To invalidate pages [7; 10], we need to invalidate [0; 15]:
+	 *		x = 0b0111 ^ 0b1010 = 0b1101
+	 *		span = 1 << fls(x) = 16
+	 */
+	log2_span	= fls_long(page_start ^ page_end);
+	span_mask	= (1ULL << log2_span) - 1;
+
+	page_start	&= ~span_mask;
+
+	cmd->atc.addr	= page_start << inval_grain_shift;
+	cmd->atc.size	= log2_span;
+}
+
+static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
+{
+	int i;
+	struct arm_smmu_cmdq_ent cmd;
+
+	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
+
+	for (i = 0; i < master->num_sids; i++) {
+		cmd.atc.sid = master->sids[i];
+		arm_smmu_cmdq_issue_cmd(master->smmu, &cmd);
+	}
+
+	return arm_smmu_cmdq_issue_sync(master->smmu);
+}
+
+static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
+				   int ssid, unsigned long iova, size_t size)
+{
+	int i;
+	unsigned long flags;
+	struct arm_smmu_cmdq_ent cmd;
+	struct arm_smmu_master *master;
+	struct arm_smmu_cmdq_batch cmds = {};
+
+	if (!(smmu_domain->smmu->features & ARM_SMMU_FEAT_ATS))
+		return 0;
+
+	/*
+	 * Ensure that we've completed prior invalidation of the main TLBs
+	 * before we read 'nr_ats_masters' in case of a concurrent call to
+	 * arm_smmu_enable_ats():
+	 *
+	 *	// unmap()			// arm_smmu_enable_ats()
+	 *	TLBI+SYNC			atomic_inc(&nr_ats_masters);
+	 *	smp_mb();			[...]
+	 *	atomic_read(&nr_ats_masters);	pci_enable_ats() // writel()
+	 *
+	 * Ensures that we always see the incremented 'nr_ats_masters' count if
+	 * ATS was enabled at the PCI device before completion of the TLBI.
+	 */
+	smp_mb();
+	if (!atomic_read(&smmu_domain->nr_ats_masters))
+		return 0;
+
+	arm_smmu_atc_inv_to_cmd(ssid, iova, size, &cmd);
+
+	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
+	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
+		if (!master->ats_enabled)
+			continue;
+
+		for (i = 0; i < master->num_sids; i++) {
+			cmd.atc.sid = master->sids[i];
+			arm_smmu_cmdq_batch_add(smmu_domain->smmu, &cmds, &cmd);
+		}
+	}
+	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+
+	return arm_smmu_cmdq_batch_submit(smmu_domain->smmu, &cmds);
+}
+
+/* IO_PGTABLE API */
+static void arm_smmu_tlb_inv_context(void *cookie)
+{
+	struct arm_smmu_domain *smmu_domain = cookie;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_cmdq_ent cmd;
+
+	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
+		cmd.opcode	= CMDQ_OP_TLBI_NH_ASID;
+		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
+		cmd.tlbi.vmid	= 0;
+	} else {
+		cmd.opcode	= CMDQ_OP_TLBI_S12_VMALL;
+		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
+	}
+
+	/*
+	 * NOTE: when io-pgtable is in non-strict mode, we may get here with
+	 * PTEs previously cleared by unmaps on the current CPU not yet visible
+	 * to the SMMU. We are relying on the dma_wmb() implicit during cmd
+	 * insertion to guarantee those are observed before the TLBI. Do be
+	 * careful, 007.
+	 */
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	arm_smmu_cmdq_issue_sync(smmu);
+	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
+}
+
+static void arm_smmu_tlb_inv_range(unsigned long iova, size_t size,
+				   size_t granule, bool leaf,
+				   struct arm_smmu_domain *smmu_domain)
+{
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	unsigned long start = iova, end = iova + size, num_pages = 0, tg = 0;
+	size_t inv_range = granule;
+	struct arm_smmu_cmdq_batch cmds = {};
+	struct arm_smmu_cmdq_ent cmd = {
+		.tlbi = {
+			.leaf	= leaf,
+		},
+	};
+
+	if (!size)
+		return;
+
+	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
+		cmd.opcode	= CMDQ_OP_TLBI_NH_VA;
+		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
+	} else {
+		cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
+		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
+	}
+
+	if (smmu->features & ARM_SMMU_FEAT_RANGE_INV) {
+		/* Get the leaf page size */
+		tg = __ffs(smmu_domain->domain.pgsize_bitmap);
+
+		/* Convert page size of 12,14,16 (log2) to 1,2,3 */
+		cmd.tlbi.tg = (tg - 10) / 2;
+
+		/* Determine what level the granule is at */
+		cmd.tlbi.ttl = 4 - ((ilog2(granule) - 3) / (tg - 3));
+
+		num_pages = size >> tg;
+	}
+
+	while (iova < end) {
+		if (smmu->features & ARM_SMMU_FEAT_RANGE_INV) {
+			/*
+			 * On each iteration of the loop, the range is 5 bits
+			 * worth of the aligned size remaining.
+			 * The range in pages is:
+			 *
+			 * range = (num_pages & (0x1f << __ffs(num_pages)))
+			 */
+			unsigned long scale, num;
+
+			/* Determine the power of 2 multiple number of pages */
+			scale = __ffs(num_pages);
+			cmd.tlbi.scale = scale;
+
+			/* Determine how many chunks of 2^scale size we have */
+			num = (num_pages >> scale) & CMDQ_TLBI_RANGE_NUM_MAX;
+			cmd.tlbi.num = num - 1;
+
+			/* range is num * 2^scale * pgsize */
+			inv_range = num << (scale + tg);
+
+			/* Clear out the lower order bits for the next iteration */
+			num_pages -= num << scale;
+		}
+
+		cmd.tlbi.addr = iova;
+		arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
+		iova += inv_range;
+	}
+	arm_smmu_cmdq_batch_submit(smmu, &cmds);
+
+	/*
+	 * Unfortunately, this can't be leaf-only since we may have
+	 * zapped an entire table.
+	 */
+	arm_smmu_atc_inv_domain(smmu_domain, 0, start, size);
+}
+
+static void arm_smmu_tlb_inv_page_nosync(struct iommu_iotlb_gather *gather,
+					 unsigned long iova, size_t granule,
+					 void *cookie)
+{
+	struct arm_smmu_domain *smmu_domain = cookie;
+	struct iommu_domain *domain = &smmu_domain->domain;
+
+	iommu_iotlb_gather_add_page(domain, gather, iova, granule);
+}
+
+static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,
+				  size_t granule, void *cookie)
+{
+	arm_smmu_tlb_inv_range(iova, size, granule, false, cookie);
+}
+
+static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,
+				  size_t granule, void *cookie)
+{
+	arm_smmu_tlb_inv_range(iova, size, granule, true, cookie);
+}
+
+static const struct iommu_flush_ops arm_smmu_flush_ops = {
+	.tlb_flush_all	= arm_smmu_tlb_inv_context,
+	.tlb_flush_walk = arm_smmu_tlb_inv_walk,
+	.tlb_flush_leaf = arm_smmu_tlb_inv_leaf,
+	.tlb_add_page	= arm_smmu_tlb_inv_page_nosync,
+};
+
+/* IOMMU API */
+static bool arm_smmu_capable(enum iommu_cap cap)
+{
+	switch (cap) {
+	case IOMMU_CAP_CACHE_COHERENCY:
+		return true;
+	case IOMMU_CAP_NOEXEC:
+		return true;
+	default:
+		return false;
+	}
+}
+
+static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
+{
+	struct arm_smmu_domain *smmu_domain;
+
+	if (type != IOMMU_DOMAIN_UNMANAGED &&
+	    type != IOMMU_DOMAIN_DMA &&
+	    type != IOMMU_DOMAIN_IDENTITY)
+		return NULL;
+
+	/*
+	 * Allocate the domain and initialise some of its data structures.
+	 * We can't really do anything meaningful until we've added a
+	 * master.
+	 */
+	smmu_domain = kzalloc(sizeof(*smmu_domain), GFP_KERNEL);
+	if (!smmu_domain)
+		return NULL;
+
+	if (type == IOMMU_DOMAIN_DMA &&
+	    iommu_get_dma_cookie(&smmu_domain->domain)) {
+		kfree(smmu_domain);
+		return NULL;
+	}
+
+	mutex_init(&smmu_domain->init_mutex);
+	INIT_LIST_HEAD(&smmu_domain->devices);
+	spin_lock_init(&smmu_domain->devices_lock);
+
+	return &smmu_domain->domain;
+}
+
+static int arm_smmu_bitmap_alloc(unsigned long *map, int span)
+{
+	int idx, size = 1 << span;
+
+	do {
+		idx = find_first_zero_bit(map, size);
+		if (idx == size)
+			return -ENOSPC;
+	} while (test_and_set_bit(idx, map));
+
+	return idx;
+}
+
+static void arm_smmu_bitmap_free(unsigned long *map, int idx)
+{
+	clear_bit(idx, map);
+}
+
+static void arm_smmu_domain_free(struct iommu_domain *domain)
+{
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+
+	iommu_put_dma_cookie(domain);
+	free_io_pgtable_ops(smmu_domain->pgtbl_ops);
+
+	/* Free the CD and ASID, if we allocated them */
+	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
+		struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
+
+		if (cfg->cdcfg.cdtab)
+			arm_smmu_free_cd_tables(smmu_domain);
+		arm_smmu_free_asid(&cfg->cd);
+	} else {
+		struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
+		if (cfg->vmid)
+			arm_smmu_bitmap_free(smmu->vmid_map, cfg->vmid);
+	}
+
+	kfree(smmu_domain);
+}
+
+static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
+				       struct arm_smmu_master *master,
+				       struct io_pgtable_cfg *pgtbl_cfg)
+{
+	int ret;
+	u32 asid;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
+	typeof(&pgtbl_cfg->arm_lpae_s1_cfg.tcr) tcr = &pgtbl_cfg->arm_lpae_s1_cfg.tcr;
+
+	ret = xa_alloc(&asid_xa, &asid, &cfg->cd,
+		       XA_LIMIT(1, (1 << smmu->asid_bits) - 1), GFP_KERNEL);
+	if (ret)
+		return ret;
+
+	cfg->s1cdmax = master->ssid_bits;
+
+	ret = arm_smmu_alloc_cd_tables(smmu_domain);
+	if (ret)
+		goto out_free_asid;
+
+	cfg->cd.asid	= (u16)asid;
+	cfg->cd.ttbr	= pgtbl_cfg->arm_lpae_s1_cfg.ttbr;
+	cfg->cd.tcr	= FIELD_PREP(CTXDESC_CD_0_TCR_T0SZ, tcr->tsz) |
+			  FIELD_PREP(CTXDESC_CD_0_TCR_TG0, tcr->tg) |
+			  FIELD_PREP(CTXDESC_CD_0_TCR_IRGN0, tcr->irgn) |
+			  FIELD_PREP(CTXDESC_CD_0_TCR_ORGN0, tcr->orgn) |
+			  FIELD_PREP(CTXDESC_CD_0_TCR_SH0, tcr->sh) |
+			  FIELD_PREP(CTXDESC_CD_0_TCR_IPS, tcr->ips) |
+			  CTXDESC_CD_0_TCR_EPD1 | CTXDESC_CD_0_AA64;
+	cfg->cd.mair	= pgtbl_cfg->arm_lpae_s1_cfg.mair;
+
+	/*
+	 * Note that this will end up calling arm_smmu_sync_cd() before
+	 * the master has been added to the devices list for this domain.
+	 * This isn't an issue because the STE hasn't been installed yet.
+	 */
+	ret = arm_smmu_write_ctx_desc(smmu_domain, 0, &cfg->cd);
+	if (ret)
+		goto out_free_cd_tables;
+
+	return 0;
+
+out_free_cd_tables:
+	arm_smmu_free_cd_tables(smmu_domain);
+out_free_asid:
+	arm_smmu_free_asid(&cfg->cd);
+	return ret;
+}
+
+static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
+				       struct arm_smmu_master *master,
+				       struct io_pgtable_cfg *pgtbl_cfg)
+{
+	int vmid;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
+	typeof(&pgtbl_cfg->arm_lpae_s2_cfg.vtcr) vtcr;
+
+	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
+	if (vmid < 0)
+		return vmid;
+
+	vtcr = &pgtbl_cfg->arm_lpae_s2_cfg.vtcr;
+	cfg->vmid	= (u16)vmid;
+	cfg->vttbr	= pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
+	cfg->vtcr	= FIELD_PREP(STRTAB_STE_2_VTCR_S2T0SZ, vtcr->tsz) |
+			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SL0, vtcr->sl) |
+			  FIELD_PREP(STRTAB_STE_2_VTCR_S2IR0, vtcr->irgn) |
+			  FIELD_PREP(STRTAB_STE_2_VTCR_S2OR0, vtcr->orgn) |
+			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SH0, vtcr->sh) |
+			  FIELD_PREP(STRTAB_STE_2_VTCR_S2TG, vtcr->tg) |
+			  FIELD_PREP(STRTAB_STE_2_VTCR_S2PS, vtcr->ps);
+	return 0;
+}
+
+static int arm_smmu_domain_finalise(struct iommu_domain *domain,
+				    struct arm_smmu_master *master)
+{
+	int ret;
+	unsigned long ias, oas;
+	enum io_pgtable_fmt fmt;
+	struct io_pgtable_cfg pgtbl_cfg;
+	struct io_pgtable_ops *pgtbl_ops;
+	int (*finalise_stage_fn)(struct arm_smmu_domain *,
+				 struct arm_smmu_master *,
+				 struct io_pgtable_cfg *);
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+
+	if (domain->type == IOMMU_DOMAIN_IDENTITY) {
+		smmu_domain->stage = ARM_SMMU_DOMAIN_BYPASS;
+		return 0;
+	}
+
+	/* Restrict the stage to what we can actually support */
+	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S1))
+		smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
+	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S2))
+		smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
+
+	switch (smmu_domain->stage) {
+	case ARM_SMMU_DOMAIN_S1:
+		ias = (smmu->features & ARM_SMMU_FEAT_VAX) ? 52 : 48;
+		ias = min_t(unsigned long, ias, VA_BITS);
+		oas = smmu->ias;
+		fmt = ARM_64_LPAE_S1;
+		finalise_stage_fn = arm_smmu_domain_finalise_s1;
+		break;
+	case ARM_SMMU_DOMAIN_NESTED:
+	case ARM_SMMU_DOMAIN_S2:
+		ias = smmu->ias;
+		oas = smmu->oas;
+		fmt = ARM_64_LPAE_S2;
+		finalise_stage_fn = arm_smmu_domain_finalise_s2;
+		break;
+	default:
+		return -EINVAL;
+	}
+
+	pgtbl_cfg = (struct io_pgtable_cfg) {
+		.pgsize_bitmap	= smmu->pgsize_bitmap,
+		.ias		= ias,
+		.oas		= oas,
+		.coherent_walk	= smmu->features & ARM_SMMU_FEAT_COHERENCY,
+		.tlb		= &arm_smmu_flush_ops,
+		.iommu_dev	= smmu->dev,
+	};
+
+	if (smmu_domain->non_strict)
+		pgtbl_cfg.quirks |= IO_PGTABLE_QUIRK_NON_STRICT;
+
+	pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
+	if (!pgtbl_ops)
+		return -ENOMEM;
+
+	domain->pgsize_bitmap = pgtbl_cfg.pgsize_bitmap;
+	domain->geometry.aperture_end = (1UL << pgtbl_cfg.ias) - 1;
+	domain->geometry.force_aperture = true;
+
+	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
+	if (ret < 0) {
+		free_io_pgtable_ops(pgtbl_ops);
+		return ret;
+	}
+
+	smmu_domain->pgtbl_ops = pgtbl_ops;
+	return 0;
+}
+
+static __le64 *arm_smmu_get_step_for_sid(struct arm_smmu_device *smmu, u32 sid)
+{
+	__le64 *step;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+
+	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
+		struct arm_smmu_strtab_l1_desc *l1_desc;
+		int idx;
+
+		/* Two-level walk */
+		idx = (sid >> STRTAB_SPLIT) * STRTAB_L1_DESC_DWORDS;
+		l1_desc = &cfg->l1_desc[idx];
+		idx = (sid & ((1 << STRTAB_SPLIT) - 1)) * STRTAB_STE_DWORDS;
+		step = &l1_desc->l2ptr[idx];
+	} else {
+		/* Simple linear lookup */
+		step = &cfg->strtab[sid * STRTAB_STE_DWORDS];
+	}
+
+	return step;
+}
+
+static void arm_smmu_install_ste_for_dev(struct arm_smmu_master *master)
+{
+	int i, j;
+	struct arm_smmu_device *smmu = master->smmu;
+
+	for (i = 0; i < master->num_sids; ++i) {
+		u32 sid = master->sids[i];
+		__le64 *step = arm_smmu_get_step_for_sid(smmu, sid);
+
+		/* Bridged PCI devices may end up with duplicated IDs */
+		for (j = 0; j < i; j++)
+			if (master->sids[j] == sid)
+				break;
+		if (j < i)
+			continue;
+
+		arm_smmu_write_strtab_ent(master, sid, step);
+	}
+}
+
+static bool arm_smmu_ats_supported(struct arm_smmu_master *master)
+{
+	struct device *dev = master->dev;
+	struct arm_smmu_device *smmu = master->smmu;
+	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
+
+	if (!(smmu->features & ARM_SMMU_FEAT_ATS))
+		return false;
+
+	if (!(fwspec->flags & IOMMU_FWSPEC_PCI_RC_ATS))
+		return false;
+
+	return dev_is_pci(dev) && pci_ats_supported(to_pci_dev(dev));
+}
+
+static void arm_smmu_enable_ats(struct arm_smmu_master *master)
+{
+	size_t stu;
+	struct pci_dev *pdev;
+	struct arm_smmu_device *smmu = master->smmu;
+	struct arm_smmu_domain *smmu_domain = master->domain;
+
+	/* Don't enable ATS at the endpoint if it's not enabled in the STE */
+	if (!master->ats_enabled)
+		return;
+
+	/* Smallest Translation Unit: log2 of the smallest supported granule */
+	stu = __ffs(smmu->pgsize_bitmap);
+	pdev = to_pci_dev(master->dev);
+
+	atomic_inc(&smmu_domain->nr_ats_masters);
+	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
+	if (pci_enable_ats(pdev, stu))
+		dev_err(master->dev, "Failed to enable ATS (STU %zu)\n", stu);
+}
+
+static void arm_smmu_disable_ats(struct arm_smmu_master *master)
+{
+	struct arm_smmu_domain *smmu_domain = master->domain;
+
+	if (!master->ats_enabled)
+		return;
+
+	pci_disable_ats(to_pci_dev(master->dev));
+	/*
+	 * Ensure ATS is disabled at the endpoint before we issue the
+	 * ATC invalidation via the SMMU.
+	 */
+	wmb();
+	arm_smmu_atc_inv_master(master);
+	atomic_dec(&smmu_domain->nr_ats_masters);
+}
+
+static int arm_smmu_enable_pasid(struct arm_smmu_master *master)
+{
+	int ret;
+	int features;
+	int num_pasids;
+	struct pci_dev *pdev;
+
+	if (!dev_is_pci(master->dev))
+		return -ENODEV;
+
+	pdev = to_pci_dev(master->dev);
+
+	features = pci_pasid_features(pdev);
+	if (features < 0)
+		return features;
+
+	num_pasids = pci_max_pasids(pdev);
+	if (num_pasids <= 0)
+		return num_pasids;
+
+	ret = pci_enable_pasid(pdev, features);
+	if (ret) {
+		dev_err(&pdev->dev, "Failed to enable PASID\n");
+		return ret;
+	}
+
+	master->ssid_bits = min_t(u8, ilog2(num_pasids),
+				  master->smmu->ssid_bits);
+	return 0;
+}
+
+static void arm_smmu_disable_pasid(struct arm_smmu_master *master)
+{
+	struct pci_dev *pdev;
+
+	if (!dev_is_pci(master->dev))
+		return;
+
+	pdev = to_pci_dev(master->dev);
+
+	if (!pdev->pasid_enabled)
+		return;
+
+	master->ssid_bits = 0;
+	pci_disable_pasid(pdev);
+}
+
+static void arm_smmu_detach_dev(struct arm_smmu_master *master)
+{
+	unsigned long flags;
+	struct arm_smmu_domain *smmu_domain = master->domain;
+
+	if (!smmu_domain)
+		return;
+
+	arm_smmu_disable_ats(master);
+
+	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
+	list_del(&master->domain_head);
+	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+
+	master->domain = NULL;
+	master->ats_enabled = false;
+	arm_smmu_install_ste_for_dev(master);
+}
+
+static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
+{
+	int ret = 0;
+	unsigned long flags;
+	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
+	struct arm_smmu_device *smmu;
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+	struct arm_smmu_master *master;
+
+	if (!fwspec)
+		return -ENOENT;
+
+	master = dev_iommu_priv_get(dev);
+	smmu = master->smmu;
+
+	arm_smmu_detach_dev(master);
+
+	mutex_lock(&smmu_domain->init_mutex);
+
+	if (!smmu_domain->smmu) {
+		smmu_domain->smmu = smmu;
+		ret = arm_smmu_domain_finalise(domain, master);
+		if (ret) {
+			smmu_domain->smmu = NULL;
+			goto out_unlock;
+		}
+	} else if (smmu_domain->smmu != smmu) {
+		dev_err(dev,
+			"cannot attach to SMMU %s (upstream of %s)\n",
+			dev_name(smmu_domain->smmu->dev),
+			dev_name(smmu->dev));
+		ret = -ENXIO;
+		goto out_unlock;
+	} else if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1 &&
+		   master->ssid_bits != smmu_domain->s1_cfg.s1cdmax) {
+		dev_err(dev,
+			"cannot attach to incompatible domain (%u SSID bits != %u)\n",
+			smmu_domain->s1_cfg.s1cdmax, master->ssid_bits);
+		ret = -EINVAL;
+		goto out_unlock;
+	}
+
+	master->domain = smmu_domain;
+
+	if (smmu_domain->stage != ARM_SMMU_DOMAIN_BYPASS)
+		master->ats_enabled = arm_smmu_ats_supported(master);
+
+	arm_smmu_install_ste_for_dev(master);
+
+	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
+	list_add(&master->domain_head, &smmu_domain->devices);
+	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
+
+	arm_smmu_enable_ats(master);
+
+out_unlock:
+	mutex_unlock(&smmu_domain->init_mutex);
+	return ret;
+}
+
+static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
+			phys_addr_t paddr, size_t size, int prot, gfp_t gfp)
+{
+	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
+
+	if (!ops)
+		return -ENODEV;
+
+	return ops->map(ops, iova, paddr, size, prot, gfp);
+}
+
+static size_t arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova,
+			     size_t size, struct iommu_iotlb_gather *gather)
+{
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
+
+	if (!ops)
+		return 0;
+
+	return ops->unmap(ops, iova, size, gather);
+}
+
+static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain)
+{
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+
+	if (smmu_domain->smmu)
+		arm_smmu_tlb_inv_context(smmu_domain);
+}
+
+static void arm_smmu_iotlb_sync(struct iommu_domain *domain,
+				struct iommu_iotlb_gather *gather)
+{
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+
+	arm_smmu_tlb_inv_range(gather->start, gather->end - gather->start,
+			       gather->pgsize, true, smmu_domain);
+}
+
+static phys_addr_t
+arm_smmu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
+{
+	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
+
+	if (domain->type == IOMMU_DOMAIN_IDENTITY)
+		return iova;
+
+	if (!ops)
+		return 0;
+
+	return ops->iova_to_phys(ops, iova);
+}
+
+static struct platform_driver arm_smmu_driver;
+
+static
+struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle *fwnode)
+{
+	struct device *dev = driver_find_device_by_fwnode(&arm_smmu_driver.driver,
+							  fwnode);
+	put_device(dev);
+	return dev ? dev_get_drvdata(dev) : NULL;
+}
+
+static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
+{
+	unsigned long limit = smmu->strtab_cfg.num_l1_ents;
+
+	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB)
+		limit *= 1UL << STRTAB_SPLIT;
+
+	return sid < limit;
+}
+
+static struct iommu_ops arm_smmu_ops;
+
+static struct iommu_device *arm_smmu_probe_device(struct device *dev)
+{
+	int i, ret;
+	struct arm_smmu_device *smmu;
+	struct arm_smmu_master *master;
+	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
+
+	if (!fwspec || fwspec->ops != &arm_smmu_ops)
+		return ERR_PTR(-ENODEV);
+
+	if (WARN_ON_ONCE(dev_iommu_priv_get(dev)))
+		return ERR_PTR(-EBUSY);
+
+	smmu = arm_smmu_get_by_fwnode(fwspec->iommu_fwnode);
+	if (!smmu)
+		return ERR_PTR(-ENODEV);
+
+	master = kzalloc(sizeof(*master), GFP_KERNEL);
+	if (!master)
+		return ERR_PTR(-ENOMEM);
+
+	master->dev = dev;
+	master->smmu = smmu;
+	master->sids = fwspec->ids;
+	master->num_sids = fwspec->num_ids;
+	dev_iommu_priv_set(dev, master);
+
+	/* Check the SIDs are in range of the SMMU and our stream table */
+	for (i = 0; i < master->num_sids; i++) {
+		u32 sid = master->sids[i];
+
+		if (!arm_smmu_sid_in_range(smmu, sid)) {
+			ret = -ERANGE;
+			goto err_free_master;
+		}
+
+		/* Ensure l2 strtab is initialised */
+		if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
+			ret = arm_smmu_init_l2_strtab(smmu, sid);
+			if (ret)
+				goto err_free_master;
+		}
+	}
+
+	master->ssid_bits = min(smmu->ssid_bits, fwspec->num_pasid_bits);
+
+	/*
+	 * Note that PASID must be enabled before, and disabled after ATS:
+	 * PCI Express Base 4.0r1.0 - 10.5.1.3 ATS Control Register
+	 *
+	 *   Behavior is undefined if this bit is Set and the value of the PASID
+	 *   Enable, Execute Requested Enable, or Privileged Mode Requested bits
+	 *   are changed.
+	 */
+	arm_smmu_enable_pasid(master);
+
+	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB))
+		master->ssid_bits = min_t(u8, master->ssid_bits,
+					  CTXDESC_LINEAR_CDMAX);
+
+	return &smmu->iommu;
+
+err_free_master:
+	kfree(master);
+	dev_iommu_priv_set(dev, NULL);
+	return ERR_PTR(ret);
+}
+
+static void arm_smmu_release_device(struct device *dev)
+{
+	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
+	struct arm_smmu_master *master;
+
+	if (!fwspec || fwspec->ops != &arm_smmu_ops)
+		return;
+
+	master = dev_iommu_priv_get(dev);
+	arm_smmu_detach_dev(master);
+	arm_smmu_disable_pasid(master);
+	kfree(master);
+	iommu_fwspec_free(dev);
+}
+
+static struct iommu_group *arm_smmu_device_group(struct device *dev)
+{
+	struct iommu_group *group;
+
+	/*
+	 * We don't support devices sharing stream IDs other than PCI RID
+	 * aliases, since the necessary ID-to-device lookup becomes rather
+	 * impractical given a potential sparse 32-bit stream ID space.
+	 */
+	if (dev_is_pci(dev))
+		group = pci_device_group(dev);
+	else
+		group = generic_device_group(dev);
+
+	return group;
+}
+
+static int arm_smmu_domain_get_attr(struct iommu_domain *domain,
+				    enum iommu_attr attr, void *data)
+{
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+
+	switch (domain->type) {
+	case IOMMU_DOMAIN_UNMANAGED:
+		switch (attr) {
+		case DOMAIN_ATTR_NESTING:
+			*(int *)data = (smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED);
+			return 0;
+		default:
+			return -ENODEV;
+		}
+		break;
+	case IOMMU_DOMAIN_DMA:
+		switch (attr) {
+		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
+			*(int *)data = smmu_domain->non_strict;
+			return 0;
+		default:
+			return -ENODEV;
+		}
+		break;
+	default:
+		return -EINVAL;
+	}
+}
+
+static int arm_smmu_domain_set_attr(struct iommu_domain *domain,
+				    enum iommu_attr attr, void *data)
+{
+	int ret = 0;
+	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+
+	mutex_lock(&smmu_domain->init_mutex);
+
+	switch (domain->type) {
+	case IOMMU_DOMAIN_UNMANAGED:
+		switch (attr) {
+		case DOMAIN_ATTR_NESTING:
+			if (smmu_domain->smmu) {
+				ret = -EPERM;
+				goto out_unlock;
+			}
+
+			if (*(int *)data)
+				smmu_domain->stage = ARM_SMMU_DOMAIN_NESTED;
+			else
+				smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
+			break;
+		default:
+			ret = -ENODEV;
+		}
+		break;
+	case IOMMU_DOMAIN_DMA:
+		switch(attr) {
+		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
+			smmu_domain->non_strict = *(int *)data;
+			break;
+		default:
+			ret = -ENODEV;
+		}
+		break;
+	default:
+		ret = -EINVAL;
+	}
+
+out_unlock:
+	mutex_unlock(&smmu_domain->init_mutex);
+	return ret;
+}
+
+static int arm_smmu_of_xlate(struct device *dev, struct of_phandle_args *args)
+{
+	return iommu_fwspec_add_ids(dev, args->args, 1);
+}
+
+static void arm_smmu_get_resv_regions(struct device *dev,
+				      struct list_head *head)
+{
+	struct iommu_resv_region *region;
+	int prot = IOMMU_WRITE | IOMMU_NOEXEC | IOMMU_MMIO;
+
+	region = iommu_alloc_resv_region(MSI_IOVA_BASE, MSI_IOVA_LENGTH,
+					 prot, IOMMU_RESV_SW_MSI);
+	if (!region)
+		return;
+
+	list_add_tail(&region->list, head);
+
+	iommu_dma_get_resv_regions(dev, head);
+}
+
+static struct iommu_ops arm_smmu_ops = {
+	.capable		= arm_smmu_capable,
+	.domain_alloc		= arm_smmu_domain_alloc,
+	.domain_free		= arm_smmu_domain_free,
+	.attach_dev		= arm_smmu_attach_dev,
+	.map			= arm_smmu_map,
+	.unmap			= arm_smmu_unmap,
+	.flush_iotlb_all	= arm_smmu_flush_iotlb_all,
+	.iotlb_sync		= arm_smmu_iotlb_sync,
+	.iova_to_phys		= arm_smmu_iova_to_phys,
+	.probe_device		= arm_smmu_probe_device,
+	.release_device		= arm_smmu_release_device,
+	.device_group		= arm_smmu_device_group,
+	.domain_get_attr	= arm_smmu_domain_get_attr,
+	.domain_set_attr	= arm_smmu_domain_set_attr,
+	.of_xlate		= arm_smmu_of_xlate,
+	.get_resv_regions	= arm_smmu_get_resv_regions,
+	.put_resv_regions	= generic_iommu_put_resv_regions,
+	.pgsize_bitmap		= -1UL, /* Restricted during device attach */
+};
+
+/* Probing and initialisation functions */
+static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
+				   struct arm_smmu_queue *q,
+				   unsigned long prod_off,
+				   unsigned long cons_off,
+				   size_t dwords, const char *name)
+{
+	size_t qsz;
+
+	do {
+		qsz = ((1 << q->llq.max_n_shift) * dwords) << 3;
+		q->base = dmam_alloc_coherent(smmu->dev, qsz, &q->base_dma,
+					      GFP_KERNEL);
+		if (q->base || qsz < PAGE_SIZE)
+			break;
+
+		q->llq.max_n_shift--;
+	} while (1);
+
+	if (!q->base) {
+		dev_err(smmu->dev,
+			"failed to allocate queue (0x%zx bytes) for %s\n",
+			qsz, name);
+		return -ENOMEM;
+	}
+
+	if (!WARN_ON(q->base_dma & (qsz - 1))) {
+		dev_info(smmu->dev, "allocated %u entries for %s\n",
+			 1 << q->llq.max_n_shift, name);
+	}
+
+	q->prod_reg	= arm_smmu_page1_fixup(prod_off, smmu);
+	q->cons_reg	= arm_smmu_page1_fixup(cons_off, smmu);
+	q->ent_dwords	= dwords;
+
+	q->q_base  = Q_BASE_RWA;
+	q->q_base |= q->base_dma & Q_BASE_ADDR_MASK;
+	q->q_base |= FIELD_PREP(Q_BASE_LOG2SIZE, q->llq.max_n_shift);
+
+	q->llq.prod = q->llq.cons = 0;
+	return 0;
+}
+
+static void arm_smmu_cmdq_free_bitmap(void *data)
+{
+	unsigned long *bitmap = data;
+	bitmap_free(bitmap);
+}
+
+static int arm_smmu_cmdq_init(struct arm_smmu_device *smmu)
+{
+	int ret = 0;
+	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
+	unsigned int nents = 1 << cmdq->q.llq.max_n_shift;
+	atomic_long_t *bitmap;
+
+	atomic_set(&cmdq->owner_prod, 0);
+	atomic_set(&cmdq->lock, 0);
+
+	bitmap = (atomic_long_t *)bitmap_zalloc(nents, GFP_KERNEL);
+	if (!bitmap) {
+		dev_err(smmu->dev, "failed to allocate cmdq bitmap\n");
+		ret = -ENOMEM;
+	} else {
+		cmdq->valid_map = bitmap;
+		devm_add_action(smmu->dev, arm_smmu_cmdq_free_bitmap, bitmap);
+	}
+
+	return ret;
+}
+
+static int arm_smmu_init_queues(struct arm_smmu_device *smmu)
+{
+	int ret;
+
+	/* cmdq */
+	ret = arm_smmu_init_one_queue(smmu, &smmu->cmdq.q, ARM_SMMU_CMDQ_PROD,
+				      ARM_SMMU_CMDQ_CONS, CMDQ_ENT_DWORDS,
+				      "cmdq");
+	if (ret)
+		return ret;
+
+	ret = arm_smmu_cmdq_init(smmu);
+	if (ret)
+		return ret;
+
+	/* evtq */
+	ret = arm_smmu_init_one_queue(smmu, &smmu->evtq.q, ARM_SMMU_EVTQ_PROD,
+				      ARM_SMMU_EVTQ_CONS, EVTQ_ENT_DWORDS,
+				      "evtq");
+	if (ret)
+		return ret;
+
+	/* priq */
+	if (!(smmu->features & ARM_SMMU_FEAT_PRI))
+		return 0;
+
+	return arm_smmu_init_one_queue(smmu, &smmu->priq.q, ARM_SMMU_PRIQ_PROD,
+				       ARM_SMMU_PRIQ_CONS, PRIQ_ENT_DWORDS,
+				       "priq");
+}
+
+static int arm_smmu_init_l1_strtab(struct arm_smmu_device *smmu)
+{
+	unsigned int i;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+	size_t size = sizeof(*cfg->l1_desc) * cfg->num_l1_ents;
+	void *strtab = smmu->strtab_cfg.strtab;
+
+	cfg->l1_desc = devm_kzalloc(smmu->dev, size, GFP_KERNEL);
+	if (!cfg->l1_desc) {
+		dev_err(smmu->dev, "failed to allocate l1 stream table desc\n");
+		return -ENOMEM;
+	}
+
+	for (i = 0; i < cfg->num_l1_ents; ++i) {
+		arm_smmu_write_strtab_l1_desc(strtab, &cfg->l1_desc[i]);
+		strtab += STRTAB_L1_DESC_DWORDS << 3;
+	}
+
+	return 0;
+}
+
+static int arm_smmu_init_strtab_2lvl(struct arm_smmu_device *smmu)
+{
+	void *strtab;
+	u64 reg;
+	u32 size, l1size;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+
+	/* Calculate the L1 size, capped to the SIDSIZE. */
+	size = STRTAB_L1_SZ_SHIFT - (ilog2(STRTAB_L1_DESC_DWORDS) + 3);
+	size = min(size, smmu->sid_bits - STRTAB_SPLIT);
+	cfg->num_l1_ents = 1 << size;
+
+	size += STRTAB_SPLIT;
+	if (size < smmu->sid_bits)
+		dev_warn(smmu->dev,
+			 "2-level strtab only covers %u/%u bits of SID\n",
+			 size, smmu->sid_bits);
+
+	l1size = cfg->num_l1_ents * (STRTAB_L1_DESC_DWORDS << 3);
+	strtab = dmam_alloc_coherent(smmu->dev, l1size, &cfg->strtab_dma,
+				     GFP_KERNEL);
+	if (!strtab) {
+		dev_err(smmu->dev,
+			"failed to allocate l1 stream table (%u bytes)\n",
+			size);
+		return -ENOMEM;
+	}
+	cfg->strtab = strtab;
+
+	/* Configure strtab_base_cfg for 2 levels */
+	reg  = FIELD_PREP(STRTAB_BASE_CFG_FMT, STRTAB_BASE_CFG_FMT_2LVL);
+	reg |= FIELD_PREP(STRTAB_BASE_CFG_LOG2SIZE, size);
+	reg |= FIELD_PREP(STRTAB_BASE_CFG_SPLIT, STRTAB_SPLIT);
+	cfg->strtab_base_cfg = reg;
+
+	return arm_smmu_init_l1_strtab(smmu);
+}
+
+static int arm_smmu_init_strtab_linear(struct arm_smmu_device *smmu)
+{
+	void *strtab;
+	u64 reg;
+	u32 size;
+	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
+
+	size = (1 << smmu->sid_bits) * (STRTAB_STE_DWORDS << 3);
+	strtab = dmam_alloc_coherent(smmu->dev, size, &cfg->strtab_dma,
+				     GFP_KERNEL);
+	if (!strtab) {
+		dev_err(smmu->dev,
+			"failed to allocate linear stream table (%u bytes)\n",
+			size);
+		return -ENOMEM;
+	}
+	cfg->strtab = strtab;
+	cfg->num_l1_ents = 1 << smmu->sid_bits;
+
+	/* Configure strtab_base_cfg for a linear table covering all SIDs */
+	reg  = FIELD_PREP(STRTAB_BASE_CFG_FMT, STRTAB_BASE_CFG_FMT_LINEAR);
+	reg |= FIELD_PREP(STRTAB_BASE_CFG_LOG2SIZE, smmu->sid_bits);
+	cfg->strtab_base_cfg = reg;
+
+	arm_smmu_init_bypass_stes(strtab, cfg->num_l1_ents);
+	return 0;
+}
+
+static int arm_smmu_init_strtab(struct arm_smmu_device *smmu)
+{
+	u64 reg;
+	int ret;
+
+	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB)
+		ret = arm_smmu_init_strtab_2lvl(smmu);
+	else
+		ret = arm_smmu_init_strtab_linear(smmu);
+
+	if (ret)
+		return ret;
+
+	/* Set the strtab base address */
+	reg  = smmu->strtab_cfg.strtab_dma & STRTAB_BASE_ADDR_MASK;
+	reg |= STRTAB_BASE_RA;
+	smmu->strtab_cfg.strtab_base = reg;
+
+	/* Allocate the first VMID for stage-2 bypass STEs */
+	set_bit(0, smmu->vmid_map);
+	return 0;
+}
+
+static int arm_smmu_init_structures(struct arm_smmu_device *smmu)
+{
+	int ret;
+
+	ret = arm_smmu_init_queues(smmu);
+	if (ret)
+		return ret;
+
+	return arm_smmu_init_strtab(smmu);
+}
+
+static int arm_smmu_write_reg_sync(struct arm_smmu_device *smmu, u32 val,
+				   unsigned int reg_off, unsigned int ack_off)
+{
+	u32 reg;
+
+	writel_relaxed(val, smmu->base + reg_off);
+	return readl_relaxed_poll_timeout(smmu->base + ack_off, reg, reg == val,
+					  1, ARM_SMMU_POLL_TIMEOUT_US);
+}
+
+/* GBPA is "special" */
+static int arm_smmu_update_gbpa(struct arm_smmu_device *smmu, u32 set, u32 clr)
+{
+	int ret;
+	u32 reg, __iomem *gbpa = smmu->base + ARM_SMMU_GBPA;
+
+	ret = readl_relaxed_poll_timeout(gbpa, reg, !(reg & GBPA_UPDATE),
+					 1, ARM_SMMU_POLL_TIMEOUT_US);
+	if (ret)
+		return ret;
+
+	reg &= ~clr;
+	reg |= set;
+	writel_relaxed(reg | GBPA_UPDATE, gbpa);
+	ret = readl_relaxed_poll_timeout(gbpa, reg, !(reg & GBPA_UPDATE),
+					 1, ARM_SMMU_POLL_TIMEOUT_US);
+
+	if (ret)
+		dev_err(smmu->dev, "GBPA not responding to update\n");
+	return ret;
+}
+
+static void arm_smmu_free_msis(void *data)
+{
+	struct device *dev = data;
+	platform_msi_domain_free_irqs(dev);
+}
+
+static void arm_smmu_write_msi_msg(struct msi_desc *desc, struct msi_msg *msg)
+{
+	phys_addr_t doorbell;
+	struct device *dev = msi_desc_to_dev(desc);
+	struct arm_smmu_device *smmu = dev_get_drvdata(dev);
+	phys_addr_t *cfg = arm_smmu_msi_cfg[desc->platform.msi_index];
+
+	doorbell = (((u64)msg->address_hi) << 32) | msg->address_lo;
+	doorbell &= MSI_CFG0_ADDR_MASK;
+
+	writeq_relaxed(doorbell, smmu->base + cfg[0]);
+	writel_relaxed(msg->data, smmu->base + cfg[1]);
+	writel_relaxed(ARM_SMMU_MEMATTR_DEVICE_nGnRE, smmu->base + cfg[2]);
+}
+
+static void arm_smmu_setup_msis(struct arm_smmu_device *smmu)
+{
+	struct msi_desc *desc;
+	int ret, nvec = ARM_SMMU_MAX_MSIS;
+	struct device *dev = smmu->dev;
+
+	/* Clear the MSI address regs */
+	writeq_relaxed(0, smmu->base + ARM_SMMU_GERROR_IRQ_CFG0);
+	writeq_relaxed(0, smmu->base + ARM_SMMU_EVTQ_IRQ_CFG0);
+
+	if (smmu->features & ARM_SMMU_FEAT_PRI)
+		writeq_relaxed(0, smmu->base + ARM_SMMU_PRIQ_IRQ_CFG0);
+	else
+		nvec--;
+
+	if (!(smmu->features & ARM_SMMU_FEAT_MSI))
+		return;
+
+	if (!dev->msi_domain) {
+		dev_info(smmu->dev, "msi_domain absent - falling back to wired irqs\n");
+		return;
+	}
+
+	/* Allocate MSIs for evtq, gerror and priq. Ignore cmdq */
+	ret = platform_msi_domain_alloc_irqs(dev, nvec, arm_smmu_write_msi_msg);
+	if (ret) {
+		dev_warn(dev, "failed to allocate MSIs - falling back to wired irqs\n");
+		return;
+	}
+
+	for_each_msi_entry(desc, dev) {
+		switch (desc->platform.msi_index) {
+		case EVTQ_MSI_INDEX:
+			smmu->evtq.q.irq = desc->irq;
+			break;
+		case GERROR_MSI_INDEX:
+			smmu->gerr_irq = desc->irq;
+			break;
+		case PRIQ_MSI_INDEX:
+			smmu->priq.q.irq = desc->irq;
+			break;
+		default:	/* Unknown */
+			continue;
+		}
+	}
+
+	/* Add callback to free MSIs on teardown */
+	devm_add_action(dev, arm_smmu_free_msis, dev);
+}
+
+static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
+{
+	int irq, ret;
+
+	arm_smmu_setup_msis(smmu);
+
+	/* Request interrupt lines */
+	irq = smmu->evtq.q.irq;
+	if (irq) {
+		ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
+						arm_smmu_evtq_thread,
+						IRQF_ONESHOT,
+						"arm-smmu-v3-evtq", smmu);
+		if (ret < 0)
+			dev_warn(smmu->dev, "failed to enable evtq irq\n");
+	} else {
+		dev_warn(smmu->dev, "no evtq irq - events will not be reported!\n");
+	}
+
+	irq = smmu->gerr_irq;
+	if (irq) {
+		ret = devm_request_irq(smmu->dev, irq, arm_smmu_gerror_handler,
+				       0, "arm-smmu-v3-gerror", smmu);
+		if (ret < 0)
+			dev_warn(smmu->dev, "failed to enable gerror irq\n");
+	} else {
+		dev_warn(smmu->dev, "no gerr irq - errors will not be reported!\n");
+	}
+
+	if (smmu->features & ARM_SMMU_FEAT_PRI) {
+		irq = smmu->priq.q.irq;
+		if (irq) {
+			ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
+							arm_smmu_priq_thread,
+							IRQF_ONESHOT,
+							"arm-smmu-v3-priq",
+							smmu);
+			if (ret < 0)
+				dev_warn(smmu->dev,
+					 "failed to enable priq irq\n");
+		} else {
+			dev_warn(smmu->dev, "no priq irq - PRI will be broken\n");
+		}
+	}
+}
+
+static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
+{
+	int ret, irq;
+	u32 irqen_flags = IRQ_CTRL_EVTQ_IRQEN | IRQ_CTRL_GERROR_IRQEN;
+
+	/* Disable IRQs first */
+	ret = arm_smmu_write_reg_sync(smmu, 0, ARM_SMMU_IRQ_CTRL,
+				      ARM_SMMU_IRQ_CTRLACK);
+	if (ret) {
+		dev_err(smmu->dev, "failed to disable irqs\n");
+		return ret;
+	}
+
+	irq = smmu->combined_irq;
+	if (irq) {
+		/*
+		 * Cavium ThunderX2 implementation doesn't support unique irq
+		 * lines. Use a single irq line for all the SMMUv3 interrupts.
+		 */
+		ret = devm_request_threaded_irq(smmu->dev, irq,
+					arm_smmu_combined_irq_handler,
+					arm_smmu_combined_irq_thread,
+					IRQF_ONESHOT,
+					"arm-smmu-v3-combined-irq", smmu);
+		if (ret < 0)
+			dev_warn(smmu->dev, "failed to enable combined irq\n");
+	} else
+		arm_smmu_setup_unique_irqs(smmu);
+
+	if (smmu->features & ARM_SMMU_FEAT_PRI)
+		irqen_flags |= IRQ_CTRL_PRIQ_IRQEN;
+
+	/* Enable interrupt generation on the SMMU */
+	ret = arm_smmu_write_reg_sync(smmu, irqen_flags,
+				      ARM_SMMU_IRQ_CTRL, ARM_SMMU_IRQ_CTRLACK);
+	if (ret)
+		dev_warn(smmu->dev, "failed to enable irqs\n");
+
+	return 0;
+}
+
+static int arm_smmu_device_disable(struct arm_smmu_device *smmu)
+{
+	int ret;
+
+	ret = arm_smmu_write_reg_sync(smmu, 0, ARM_SMMU_CR0, ARM_SMMU_CR0ACK);
+	if (ret)
+		dev_err(smmu->dev, "failed to clear cr0\n");
+
+	return ret;
+}
+
+static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
+{
+	int ret;
+	u32 reg, enables;
+	struct arm_smmu_cmdq_ent cmd;
+
+	/* Clear CR0 and sync (disables SMMU and queue processing) */
+	reg = readl_relaxed(smmu->base + ARM_SMMU_CR0);
+	if (reg & CR0_SMMUEN) {
+		dev_warn(smmu->dev, "SMMU currently enabled! Resetting...\n");
+		WARN_ON(is_kdump_kernel() && !disable_bypass);
+		arm_smmu_update_gbpa(smmu, GBPA_ABORT, 0);
+	}
+
+	ret = arm_smmu_device_disable(smmu);
+	if (ret)
+		return ret;
+
+	/* CR1 (table and queue memory attributes) */
+	reg = FIELD_PREP(CR1_TABLE_SH, ARM_SMMU_SH_ISH) |
+	      FIELD_PREP(CR1_TABLE_OC, CR1_CACHE_WB) |
+	      FIELD_PREP(CR1_TABLE_IC, CR1_CACHE_WB) |
+	      FIELD_PREP(CR1_QUEUE_SH, ARM_SMMU_SH_ISH) |
+	      FIELD_PREP(CR1_QUEUE_OC, CR1_CACHE_WB) |
+	      FIELD_PREP(CR1_QUEUE_IC, CR1_CACHE_WB);
+	writel_relaxed(reg, smmu->base + ARM_SMMU_CR1);
+
+	/* CR2 (random crap) */
+	reg = CR2_PTM | CR2_RECINVSID | CR2_E2H;
+	writel_relaxed(reg, smmu->base + ARM_SMMU_CR2);
+
+	/* Stream table */
+	writeq_relaxed(smmu->strtab_cfg.strtab_base,
+		       smmu->base + ARM_SMMU_STRTAB_BASE);
+	writel_relaxed(smmu->strtab_cfg.strtab_base_cfg,
+		       smmu->base + ARM_SMMU_STRTAB_BASE_CFG);
+
+	/* Command queue */
+	writeq_relaxed(smmu->cmdq.q.q_base, smmu->base + ARM_SMMU_CMDQ_BASE);
+	writel_relaxed(smmu->cmdq.q.llq.prod, smmu->base + ARM_SMMU_CMDQ_PROD);
+	writel_relaxed(smmu->cmdq.q.llq.cons, smmu->base + ARM_SMMU_CMDQ_CONS);
+
+	enables = CR0_CMDQEN;
+	ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
+				      ARM_SMMU_CR0ACK);
+	if (ret) {
+		dev_err(smmu->dev, "failed to enable command queue\n");
+		return ret;
+	}
+
+	/* Invalidate any cached configuration */
+	cmd.opcode = CMDQ_OP_CFGI_ALL;
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	arm_smmu_cmdq_issue_sync(smmu);
+
+	/* Invalidate any stale TLB entries */
+	if (smmu->features & ARM_SMMU_FEAT_HYP) {
+		cmd.opcode = CMDQ_OP_TLBI_EL2_ALL;
+		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	}
+
+	cmd.opcode = CMDQ_OP_TLBI_NSNH_ALL;
+	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+	arm_smmu_cmdq_issue_sync(smmu);
+
+	/* Event queue */
+	writeq_relaxed(smmu->evtq.q.q_base, smmu->base + ARM_SMMU_EVTQ_BASE);
+	writel_relaxed(smmu->evtq.q.llq.prod,
+		       arm_smmu_page1_fixup(ARM_SMMU_EVTQ_PROD, smmu));
+	writel_relaxed(smmu->evtq.q.llq.cons,
+		       arm_smmu_page1_fixup(ARM_SMMU_EVTQ_CONS, smmu));
+
+	enables |= CR0_EVTQEN;
+	ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
+				      ARM_SMMU_CR0ACK);
+	if (ret) {
+		dev_err(smmu->dev, "failed to enable event queue\n");
+		return ret;
+	}
+
+	/* PRI queue */
+	if (smmu->features & ARM_SMMU_FEAT_PRI) {
+		writeq_relaxed(smmu->priq.q.q_base,
+			       smmu->base + ARM_SMMU_PRIQ_BASE);
+		writel_relaxed(smmu->priq.q.llq.prod,
+			       arm_smmu_page1_fixup(ARM_SMMU_PRIQ_PROD, smmu));
+		writel_relaxed(smmu->priq.q.llq.cons,
+			       arm_smmu_page1_fixup(ARM_SMMU_PRIQ_CONS, smmu));
+
+		enables |= CR0_PRIQEN;
+		ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
+					      ARM_SMMU_CR0ACK);
+		if (ret) {
+			dev_err(smmu->dev, "failed to enable PRI queue\n");
+			return ret;
+		}
+	}
+
+	if (smmu->features & ARM_SMMU_FEAT_ATS) {
+		enables |= CR0_ATSCHK;
+		ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
+					      ARM_SMMU_CR0ACK);
+		if (ret) {
+			dev_err(smmu->dev, "failed to enable ATS check\n");
+			return ret;
+		}
+	}
+
+	ret = arm_smmu_setup_irqs(smmu);
+	if (ret) {
+		dev_err(smmu->dev, "failed to setup irqs\n");
+		return ret;
+	}
+
+	if (is_kdump_kernel())
+		enables &= ~(CR0_EVTQEN | CR0_PRIQEN);
+
+	/* Enable the SMMU interface, or ensure bypass */
+	if (!bypass || disable_bypass) {
+		enables |= CR0_SMMUEN;
+	} else {
+		ret = arm_smmu_update_gbpa(smmu, 0, GBPA_ABORT);
+		if (ret)
+			return ret;
+	}
+	ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
+				      ARM_SMMU_CR0ACK);
+	if (ret) {
+		dev_err(smmu->dev, "failed to enable SMMU interface\n");
+		return ret;
+	}
+
+	return 0;
+}
+
+static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
+{
+	u32 reg;
+	bool coherent = smmu->features & ARM_SMMU_FEAT_COHERENCY;
+
+	/* IDR0 */
+	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR0);
+
+	/* 2-level structures */
+	if (FIELD_GET(IDR0_ST_LVL, reg) == IDR0_ST_LVL_2LVL)
+		smmu->features |= ARM_SMMU_FEAT_2_LVL_STRTAB;
+
+	if (reg & IDR0_CD2L)
+		smmu->features |= ARM_SMMU_FEAT_2_LVL_CDTAB;
+
+	/*
+	 * Translation table endianness.
+	 * We currently require the same endianness as the CPU, but this
+	 * could be changed later by adding a new IO_PGTABLE_QUIRK.
+	 */
+	switch (FIELD_GET(IDR0_TTENDIAN, reg)) {
+	case IDR0_TTENDIAN_MIXED:
+		smmu->features |= ARM_SMMU_FEAT_TT_LE | ARM_SMMU_FEAT_TT_BE;
+		break;
+#ifdef __BIG_ENDIAN
+	case IDR0_TTENDIAN_BE:
+		smmu->features |= ARM_SMMU_FEAT_TT_BE;
+		break;
+#else
+	case IDR0_TTENDIAN_LE:
+		smmu->features |= ARM_SMMU_FEAT_TT_LE;
+		break;
+#endif
+	default:
+		dev_err(smmu->dev, "unknown/unsupported TT endianness!\n");
+		return -ENXIO;
+	}
+
+	/* Boolean feature flags */
+	if (IS_ENABLED(CONFIG_PCI_PRI) && reg & IDR0_PRI)
+		smmu->features |= ARM_SMMU_FEAT_PRI;
+
+	if (IS_ENABLED(CONFIG_PCI_ATS) && reg & IDR0_ATS)
+		smmu->features |= ARM_SMMU_FEAT_ATS;
+
+	if (reg & IDR0_SEV)
+		smmu->features |= ARM_SMMU_FEAT_SEV;
+
+	if (reg & IDR0_MSI)
+		smmu->features |= ARM_SMMU_FEAT_MSI;
+
+	if (reg & IDR0_HYP)
+		smmu->features |= ARM_SMMU_FEAT_HYP;
+
+	/*
+	 * The coherency feature as set by FW is used in preference to the ID
+	 * register, but warn on mismatch.
+	 */
+	if (!!(reg & IDR0_COHACC) != coherent)
+		dev_warn(smmu->dev, "IDR0.COHACC overridden by FW configuration (%s)\n",
+			 coherent ? "true" : "false");
+
+	switch (FIELD_GET(IDR0_STALL_MODEL, reg)) {
+	case IDR0_STALL_MODEL_FORCE:
+		smmu->features |= ARM_SMMU_FEAT_STALL_FORCE;
+		fallthrough;
+	case IDR0_STALL_MODEL_STALL:
+		smmu->features |= ARM_SMMU_FEAT_STALLS;
+	}
+
+	if (reg & IDR0_S1P)
+		smmu->features |= ARM_SMMU_FEAT_TRANS_S1;
+
+	if (reg & IDR0_S2P)
+		smmu->features |= ARM_SMMU_FEAT_TRANS_S2;
+
+	if (!(reg & (IDR0_S1P | IDR0_S2P))) {
+		dev_err(smmu->dev, "no translation support!\n");
+		return -ENXIO;
+	}
+
+	/* We only support the AArch64 table format at present */
+	switch (FIELD_GET(IDR0_TTF, reg)) {
+	case IDR0_TTF_AARCH32_64:
+		smmu->ias = 40;
+		fallthrough;
+	case IDR0_TTF_AARCH64:
+		break;
+	default:
+		dev_err(smmu->dev, "AArch64 table format not supported!\n");
+		return -ENXIO;
+	}
+
+	/* ASID/VMID sizes */
+	smmu->asid_bits = reg & IDR0_ASID16 ? 16 : 8;
+	smmu->vmid_bits = reg & IDR0_VMID16 ? 16 : 8;
+
+	/* IDR1 */
+	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR1);
+	if (reg & (IDR1_TABLES_PRESET | IDR1_QUEUES_PRESET | IDR1_REL)) {
+		dev_err(smmu->dev, "embedded implementation not supported\n");
+		return -ENXIO;
+	}
+
+	/* Queue sizes, capped to ensure natural alignment */
+	smmu->cmdq.q.llq.max_n_shift = min_t(u32, CMDQ_MAX_SZ_SHIFT,
+					     FIELD_GET(IDR1_CMDQS, reg));
+	if (smmu->cmdq.q.llq.max_n_shift <= ilog2(CMDQ_BATCH_ENTRIES)) {
+		/*
+		 * We don't support splitting up batches, so one batch of
+		 * commands plus an extra sync needs to fit inside the command
+		 * queue. There's also no way we can handle the weird alignment
+		 * restrictions on the base pointer for a unit-length queue.
+		 */
+		dev_err(smmu->dev, "command queue size <= %d entries not supported\n",
+			CMDQ_BATCH_ENTRIES);
+		return -ENXIO;
+	}
+
+	smmu->evtq.q.llq.max_n_shift = min_t(u32, EVTQ_MAX_SZ_SHIFT,
+					     FIELD_GET(IDR1_EVTQS, reg));
+	smmu->priq.q.llq.max_n_shift = min_t(u32, PRIQ_MAX_SZ_SHIFT,
+					     FIELD_GET(IDR1_PRIQS, reg));
+
+	/* SID/SSID sizes */
+	smmu->ssid_bits = FIELD_GET(IDR1_SSIDSIZE, reg);
+	smmu->sid_bits = FIELD_GET(IDR1_SIDSIZE, reg);
+
+	/*
+	 * If the SMMU supports fewer bits than would fill a single L2 stream
+	 * table, use a linear table instead.
+	 */
+	if (smmu->sid_bits <= STRTAB_SPLIT)
+		smmu->features &= ~ARM_SMMU_FEAT_2_LVL_STRTAB;
+
+	/* IDR3 */
+	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR3);
+	if (FIELD_GET(IDR3_RIL, reg))
+		smmu->features |= ARM_SMMU_FEAT_RANGE_INV;
+
+	/* IDR5 */
+	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR5);
+
+	/* Maximum number of outstanding stalls */
+	smmu->evtq.max_stalls = FIELD_GET(IDR5_STALL_MAX, reg);
+
+	/* Page sizes */
+	if (reg & IDR5_GRAN64K)
+		smmu->pgsize_bitmap |= SZ_64K | SZ_512M;
+	if (reg & IDR5_GRAN16K)
+		smmu->pgsize_bitmap |= SZ_16K | SZ_32M;
+	if (reg & IDR5_GRAN4K)
+		smmu->pgsize_bitmap |= SZ_4K | SZ_2M | SZ_1G;
+
+	/* Input address size */
+	if (FIELD_GET(IDR5_VAX, reg) == IDR5_VAX_52_BIT)
+		smmu->features |= ARM_SMMU_FEAT_VAX;
+
+	/* Output address size */
+	switch (FIELD_GET(IDR5_OAS, reg)) {
+	case IDR5_OAS_32_BIT:
+		smmu->oas = 32;
+		break;
+	case IDR5_OAS_36_BIT:
+		smmu->oas = 36;
+		break;
+	case IDR5_OAS_40_BIT:
+		smmu->oas = 40;
+		break;
+	case IDR5_OAS_42_BIT:
+		smmu->oas = 42;
+		break;
+	case IDR5_OAS_44_BIT:
+		smmu->oas = 44;
+		break;
+	case IDR5_OAS_52_BIT:
+		smmu->oas = 52;
+		smmu->pgsize_bitmap |= 1ULL << 42; /* 4TB */
+		break;
+	default:
+		dev_info(smmu->dev,
+			"unknown output address size. Truncating to 48-bit\n");
+		fallthrough;
+	case IDR5_OAS_48_BIT:
+		smmu->oas = 48;
+	}
+
+	if (arm_smmu_ops.pgsize_bitmap == -1UL)
+		arm_smmu_ops.pgsize_bitmap = smmu->pgsize_bitmap;
+	else
+		arm_smmu_ops.pgsize_bitmap |= smmu->pgsize_bitmap;
+
+	/* Set the DMA mask for our table walker */
+	if (dma_set_mask_and_coherent(smmu->dev, DMA_BIT_MASK(smmu->oas)))
+		dev_warn(smmu->dev,
+			 "failed to set DMA mask for table walker\n");
+
+	smmu->ias = max(smmu->ias, smmu->oas);
+
+	dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n",
+		 smmu->ias, smmu->oas, smmu->features);
+	return 0;
+}
+
+#ifdef CONFIG_ACPI
+static void acpi_smmu_get_options(u32 model, struct arm_smmu_device *smmu)
+{
+	switch (model) {
+	case ACPI_IORT_SMMU_V3_CAVIUM_CN99XX:
+		smmu->options |= ARM_SMMU_OPT_PAGE0_REGS_ONLY;
+		break;
+	case ACPI_IORT_SMMU_V3_HISILICON_HI161X:
+		smmu->options |= ARM_SMMU_OPT_SKIP_PREFETCH;
+		break;
+	}
+
+	dev_notice(smmu->dev, "option mask 0x%x\n", smmu->options);
+}
+
+static int arm_smmu_device_acpi_probe(struct platform_device *pdev,
+				      struct arm_smmu_device *smmu)
+{
+	struct acpi_iort_smmu_v3 *iort_smmu;
+	struct device *dev = smmu->dev;
+	struct acpi_iort_node *node;
+
+	node = *(struct acpi_iort_node **)dev_get_platdata(dev);
+
+	/* Retrieve SMMUv3 specific data */
+	iort_smmu = (struct acpi_iort_smmu_v3 *)node->node_data;
+
+	acpi_smmu_get_options(iort_smmu->model, smmu);
+
+	if (iort_smmu->flags & ACPI_IORT_SMMU_V3_COHACC_OVERRIDE)
+		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
+
+	return 0;
+}
+#else
+static inline int arm_smmu_device_acpi_probe(struct platform_device *pdev,
+					     struct arm_smmu_device *smmu)
+{
+	return -ENODEV;
+}
+#endif
+
+static int arm_smmu_device_dt_probe(struct platform_device *pdev,
+				    struct arm_smmu_device *smmu)
+{
+	struct device *dev = &pdev->dev;
+	u32 cells;
+	int ret = -EINVAL;
+
+	if (of_property_read_u32(dev->of_node, "#iommu-cells", &cells))
+		dev_err(dev, "missing #iommu-cells property\n");
+	else if (cells != 1)
+		dev_err(dev, "invalid #iommu-cells value (%d)\n", cells);
+	else
+		ret = 0;
+
+	parse_driver_options(smmu);
+
+	if (of_dma_is_coherent(dev->of_node))
+		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
+
+	return ret;
+}
+
+static unsigned long arm_smmu_resource_size(struct arm_smmu_device *smmu)
+{
+	if (smmu->options & ARM_SMMU_OPT_PAGE0_REGS_ONLY)
+		return SZ_64K;
+	else
+		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)
+{
+	struct resource res = {
+		.flags = IORESOURCE_MEM,
+		.start = start,
+		.end = start + size - 1,
+	};
+
+	return devm_ioremap_resource(dev, &res);
+}
+
+static int arm_smmu_device_probe(struct platform_device *pdev)
+{
+	int irq, ret;
+	struct resource *res;
+	resource_size_t ioaddr;
+	struct arm_smmu_device *smmu;
+	struct device *dev = &pdev->dev;
+	bool bypass;
+
+	smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
+	if (!smmu) {
+		dev_err(dev, "failed to allocate arm_smmu_device\n");
+		return -ENOMEM;
+	}
+	smmu->dev = dev;
+
+	if (dev->of_node) {
+		ret = arm_smmu_device_dt_probe(pdev, smmu);
+	} else {
+		ret = arm_smmu_device_acpi_probe(pdev, smmu);
+		if (ret == -ENODEV)
+			return ret;
+	}
+
+	/* Set bypass mode according to firmware probing result */
+	bypass = !!ret;
+
+	/* Base address */
+	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
+	if (resource_size(res) < arm_smmu_resource_size(smmu)) {
+		dev_err(dev, "MMIO region too small (%pr)\n", res);
+		return -EINVAL;
+	}
+	ioaddr = res->start;
+
+	/*
+	 * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
+	 * the PMCG registers which are reserved by the PMU driver.
+	 */
+	smmu->base = arm_smmu_ioremap(dev, ioaddr, ARM_SMMU_REG_SZ);
+	if (IS_ERR(smmu->base))
+		return PTR_ERR(smmu->base);
+
+	if (arm_smmu_resource_size(smmu) > SZ_64K) {
+		smmu->page1 = arm_smmu_ioremap(dev, ioaddr + SZ_64K,
+					       ARM_SMMU_REG_SZ);
+		if (IS_ERR(smmu->page1))
+			return PTR_ERR(smmu->page1);
+	} else {
+		smmu->page1 = smmu->base;
+	}
+
+	/* Interrupt lines */
+
+	irq = platform_get_irq_byname_optional(pdev, "combined");
+	if (irq > 0)
+		smmu->combined_irq = irq;
+	else {
+		irq = platform_get_irq_byname_optional(pdev, "eventq");
+		if (irq > 0)
+			smmu->evtq.q.irq = irq;
+
+		irq = platform_get_irq_byname_optional(pdev, "priq");
+		if (irq > 0)
+			smmu->priq.q.irq = irq;
+
+		irq = platform_get_irq_byname_optional(pdev, "gerror");
+		if (irq > 0)
+			smmu->gerr_irq = irq;
+	}
+	/* Probe the h/w */
+	ret = arm_smmu_device_hw_probe(smmu);
+	if (ret)
+		return ret;
+
+	/* Initialise in-memory data structures */
+	ret = arm_smmu_init_structures(smmu);
+	if (ret)
+		return ret;
+
+	/* Record our private device structure */
+	platform_set_drvdata(pdev, smmu);
+
+	/* Reset the device */
+	ret = arm_smmu_device_reset(smmu, bypass);
+	if (ret)
+		return ret;
+
+	/* And we're up. Go go go! */
+	ret = iommu_device_sysfs_add(&smmu->iommu, dev, NULL,
+				     "smmu3.%pa", &ioaddr);
+	if (ret)
+		return ret;
+
+	iommu_device_set_ops(&smmu->iommu, &arm_smmu_ops);
+	iommu_device_set_fwnode(&smmu->iommu, dev->fwnode);
+
+	ret = iommu_device_register(&smmu->iommu);
+	if (ret) {
+		dev_err(dev, "Failed to register iommu\n");
+		return ret;
+	}
+
+	return arm_smmu_set_bus_ops(&arm_smmu_ops);
+}
+
+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);
+
+	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", },
+	{ },
+};
+MODULE_DEVICE_TABLE(of, arm_smmu_of_match);
+
+static struct platform_driver arm_smmu_driver = {
+	.driver	= {
+		.name			= "arm-smmu-v3",
+		.of_match_table		= arm_smmu_of_match,
+		.suppress_bind_attrs	= true,
+	},
+	.probe	= arm_smmu_device_probe,
+	.remove	= arm_smmu_device_remove,
+	.shutdown = arm_smmu_device_shutdown,
+};
+module_platform_driver(arm_smmu_driver);
+
+MODULE_DESCRIPTION("IOMMU API for ARM architected SMMUv3 implementations");
+MODULE_AUTHOR("Will Deacon <will@kernel.org>");
+MODULE_ALIAS("platform:arm-smmu-v3");
+MODULE_LICENSE("GPL v2");
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 53+ messages in thread

* [PATCH v2 2/8] xen/arm: revert atomic operation related command-queue insertion patch
  2020-11-26 17:01 [PATCH v2 0/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
  2020-11-26 17:02 ` [PATCH v2 1/8] xen/arm: Import the SMMUv3 driver from Linux Rahul Singh
@ 2020-11-26 17:02 ` Rahul Singh
  2020-12-01 22:23   ` Stefano Stabellini
  2020-12-02 13:44   ` Julien Grall
  2020-11-26 17:02 ` [PATCH v2 3/8] xen/arm: revert patch related to XArray Rahul Singh
                   ` (5 subsequent siblings)
  7 siblings, 2 replies; 53+ messages in thread
From: Rahul Singh @ 2020-11-26 17:02 UTC (permalink / raw)
  To: xen-devel
  Cc: bertrand.marquis, rahul.singh, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

Linux SMMUv3 code implements the commands-queue insertion based on
atomic operations implemented in Linux. Atomic functions used by the
commands-queue insertion is not implemented in XEN therefore revert the
patch that implemented the commands-queue insertion based on atomic
operations.

Once the proper atomic operations will be available in XEN the driver
can be updated.

Reverted the commit 587e6c10a7ce89a5924fdbeff2ec524fbd6a124b
iommu/arm-smmu-v3: Reduce contention during command-queue insertion

Signed-off-by: Rahul Singh <rahul.singh@arm.com>
---
 xen/drivers/passthrough/arm/smmu-v3.c | 847 ++++++--------------------
 1 file changed, 180 insertions(+), 667 deletions(-)

diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
index c192544e87..97eac61ea4 100644
--- a/xen/drivers/passthrough/arm/smmu-v3.c
+++ b/xen/drivers/passthrough/arm/smmu-v3.c
@@ -330,15 +330,6 @@
 #define CMDQ_ERR_CERROR_ABT_IDX		2
 #define CMDQ_ERR_CERROR_ATC_INV_IDX	3
 
-#define CMDQ_PROD_OWNED_FLAG		Q_OVERFLOW_FLAG
-
-/*
- * This is used to size the command queue and therefore must be at least
- * BITS_PER_LONG so that the valid_map works correctly (it relies on the
- * total number of queue entries being a multiple of BITS_PER_LONG).
- */
-#define CMDQ_BATCH_ENTRIES		BITS_PER_LONG
-
 #define CMDQ_0_OP			GENMASK_ULL(7, 0)
 #define CMDQ_0_SSV			(1UL << 11)
 
@@ -407,8 +398,9 @@
 #define PRIQ_1_ADDR_MASK		GENMASK_ULL(63, 12)
 
 /* High-level queue structures */
-#define ARM_SMMU_POLL_TIMEOUT_US	1000000 /* 1s! */
-#define ARM_SMMU_POLL_SPIN_COUNT	10
+#define ARM_SMMU_POLL_TIMEOUT_US	100
+#define ARM_SMMU_CMDQ_SYNC_TIMEOUT_US	1000000 /* 1s! */
+#define ARM_SMMU_CMDQ_SYNC_SPIN_COUNT	10
 
 #define MSI_IOVA_BASE			0x8000000
 #define MSI_IOVA_LENGTH			0x100000
@@ -513,24 +505,15 @@ struct arm_smmu_cmdq_ent {
 
 		#define CMDQ_OP_CMD_SYNC	0x46
 		struct {
+			u32			msidata;
 			u64			msiaddr;
 		} sync;
 	};
 };
 
 struct arm_smmu_ll_queue {
-	union {
-		u64			val;
-		struct {
-			u32		prod;
-			u32		cons;
-		};
-		struct {
-			atomic_t	prod;
-			atomic_t	cons;
-		} atomic;
-		u8			__pad[SMP_CACHE_BYTES];
-	} ____cacheline_aligned_in_smp;
+	u32				prod;
+	u32				cons;
 	u32				max_n_shift;
 };
 
@@ -548,23 +531,9 @@ struct arm_smmu_queue {
 	u32 __iomem			*cons_reg;
 };
 
-struct arm_smmu_queue_poll {
-	ktime_t				timeout;
-	unsigned int			delay;
-	unsigned int			spin_cnt;
-	bool				wfe;
-};
-
 struct arm_smmu_cmdq {
 	struct arm_smmu_queue		q;
-	atomic_long_t			*valid_map;
-	atomic_t			owner_prod;
-	atomic_t			lock;
-};
-
-struct arm_smmu_cmdq_batch {
-	u64				cmds[CMDQ_BATCH_ENTRIES * CMDQ_ENT_DWORDS];
-	int				num;
+	spinlock_t			lock;
 };
 
 struct arm_smmu_evtq {
@@ -660,6 +629,8 @@ struct arm_smmu_device {
 
 	int				gerr_irq;
 	int				combined_irq;
+	u32				sync_nr;
+	u8				prev_cmd_opcode;
 
 	unsigned long			ias; /* IPA */
 	unsigned long			oas; /* PA */
@@ -677,6 +648,12 @@ struct arm_smmu_device {
 
 	struct arm_smmu_strtab_cfg	strtab_cfg;
 
+	/* Hi16xx adds an extra 32 bits of goodness to its MSI payload */
+	union {
+		u32			sync_count;
+		u64			padding;
+	};
+
 	/* IOMMU core code handle */
 	struct iommu_device		iommu;
 };
@@ -763,21 +740,6 @@ static void parse_driver_options(struct arm_smmu_device *smmu)
 }
 
 /* Low-level queue manipulation functions */
-static bool queue_has_space(struct arm_smmu_ll_queue *q, u32 n)
-{
-	u32 space, prod, cons;
-
-	prod = Q_IDX(q, q->prod);
-	cons = Q_IDX(q, q->cons);
-
-	if (Q_WRP(q, q->prod) == Q_WRP(q, q->cons))
-		space = (1 << q->max_n_shift) - (prod - cons);
-	else
-		space = cons - prod;
-
-	return space >= n;
-}
-
 static bool queue_full(struct arm_smmu_ll_queue *q)
 {
 	return Q_IDX(q, q->prod) == Q_IDX(q, q->cons) &&
@@ -790,12 +752,9 @@ static bool queue_empty(struct arm_smmu_ll_queue *q)
 	       Q_WRP(q, q->prod) == Q_WRP(q, q->cons);
 }
 
-static bool queue_consumed(struct arm_smmu_ll_queue *q, u32 prod)
+static void queue_sync_cons_in(struct arm_smmu_queue *q)
 {
-	return ((Q_WRP(q, q->cons) == Q_WRP(q, prod)) &&
-		(Q_IDX(q, q->cons) > Q_IDX(q, prod))) ||
-	       ((Q_WRP(q, q->cons) != Q_WRP(q, prod)) &&
-		(Q_IDX(q, q->cons) <= Q_IDX(q, prod)));
+	q->llq.cons = readl_relaxed(q->cons_reg);
 }
 
 static void queue_sync_cons_out(struct arm_smmu_queue *q)
@@ -826,34 +785,46 @@ static int queue_sync_prod_in(struct arm_smmu_queue *q)
 	return ret;
 }
 
-static u32 queue_inc_prod_n(struct arm_smmu_ll_queue *q, int n)
+static void queue_sync_prod_out(struct arm_smmu_queue *q)
 {
-	u32 prod = (Q_WRP(q, q->prod) | Q_IDX(q, q->prod)) + n;
-	return Q_OVF(q->prod) | Q_WRP(q, prod) | Q_IDX(q, prod);
+	writel(q->llq.prod, q->prod_reg);
 }
 
-static void queue_poll_init(struct arm_smmu_device *smmu,
-			    struct arm_smmu_queue_poll *qp)
+static void queue_inc_prod(struct arm_smmu_ll_queue *q)
 {
-	qp->delay = 1;
-	qp->spin_cnt = 0;
-	qp->wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
-	qp->timeout = ktime_add_us(ktime_get(), ARM_SMMU_POLL_TIMEOUT_US);
+	u32 prod = (Q_WRP(q, q->prod) | Q_IDX(q, q->prod)) + 1;
+	q->prod = Q_OVF(q->prod) | Q_WRP(q, prod) | Q_IDX(q, prod);
 }
 
-static int queue_poll(struct arm_smmu_queue_poll *qp)
+/*
+ * Wait for the SMMU to consume items. If sync is true, wait until the queue
+ * is empty. Otherwise, wait until there is at least one free slot.
+ */
+static int queue_poll_cons(struct arm_smmu_queue *q, bool sync, bool wfe)
 {
-	if (ktime_compare(ktime_get(), qp->timeout) > 0)
-		return -ETIMEDOUT;
+	ktime_t timeout;
+	unsigned int delay = 1, spin_cnt = 0;
 
-	if (qp->wfe) {
-		wfe();
-	} else if (++qp->spin_cnt < ARM_SMMU_POLL_SPIN_COUNT) {
-		cpu_relax();
-	} else {
-		udelay(qp->delay);
-		qp->delay *= 2;
-		qp->spin_cnt = 0;
+	/* Wait longer if it's a CMD_SYNC */
+	timeout = ktime_add_us(ktime_get(), sync ?
+					    ARM_SMMU_CMDQ_SYNC_TIMEOUT_US :
+					    ARM_SMMU_POLL_TIMEOUT_US);
+
+	while (queue_sync_cons_in(q),
+	      (sync ? !queue_empty(&q->llq) : queue_full(&q->llq))) {
+		if (ktime_compare(ktime_get(), timeout) > 0)
+			return -ETIMEDOUT;
+
+		if (wfe) {
+			wfe();
+		} else if (++spin_cnt < ARM_SMMU_CMDQ_SYNC_SPIN_COUNT) {
+			cpu_relax();
+			continue;
+		} else {
+			udelay(delay);
+			delay *= 2;
+			spin_cnt = 0;
+		}
 	}
 
 	return 0;
@@ -867,6 +838,17 @@ static void queue_write(__le64 *dst, u64 *src, size_t n_dwords)
 		*dst++ = cpu_to_le64(*src++);
 }
 
+static int queue_insert_raw(struct arm_smmu_queue *q, u64 *ent)
+{
+	if (queue_full(&q->llq))
+		return -ENOSPC;
+
+	queue_write(Q_ENT(q, q->llq.prod), ent, q->ent_dwords);
+	queue_inc_prod(&q->llq);
+	queue_sync_prod_out(q);
+	return 0;
+}
+
 static void queue_read(__le64 *dst, u64 *src, size_t n_dwords)
 {
 	int i;
@@ -964,14 +946,20 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
 		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_RESP, ent->pri.resp);
 		break;
 	case CMDQ_OP_CMD_SYNC:
-		if (ent->sync.msiaddr) {
+		if (ent->sync.msiaddr)
 			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_IRQ);
-			cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
-		} else {
+		else
 			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
-		}
 		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSH, ARM_SMMU_SH_ISH);
 		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIATTR, ARM_SMMU_MEMATTR_OIWB);
+		/*
+		 * Commands are written little-endian, but we want the SMMU to
+		 * receive MSIData, and thus write it back to memory, in CPU
+		 * byte order, so big-endian needs an extra byteswap here.
+		 */
+		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIDATA,
+				     cpu_to_le32(ent->sync.msidata));
+		cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
 		break;
 	default:
 		return -ENOENT;
@@ -980,27 +968,6 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
 	return 0;
 }
 
-static void arm_smmu_cmdq_build_sync_cmd(u64 *cmd, struct arm_smmu_device *smmu,
-					 u32 prod)
-{
-	struct arm_smmu_queue *q = &smmu->cmdq.q;
-	struct arm_smmu_cmdq_ent ent = {
-		.opcode = CMDQ_OP_CMD_SYNC,
-	};
-
-	/*
-	 * Beware that Hi16xx adds an extra 32 bits of goodness to its MSI
-	 * payload, so the write will zero the entire command on that platform.
-	 */
-	if (smmu->features & ARM_SMMU_FEAT_MSI &&
-	    smmu->features & ARM_SMMU_FEAT_COHERENCY) {
-		ent.sync.msiaddr = q->base_dma + Q_IDX(&q->llq, prod) *
-				   q->ent_dwords * 8;
-	}
-
-	arm_smmu_cmdq_build_cmd(cmd, &ent);
-}
-
 static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
 {
 	static const char *cerror_str[] = {
@@ -1058,474 +1025,109 @@ static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
 	queue_write(Q_ENT(q, cons), cmd, q->ent_dwords);
 }
 
-/*
- * Command queue locking.
- * This is a form of bastardised rwlock with the following major changes:
- *
- * - The only LOCK routines are exclusive_trylock() and shared_lock().
- *   Neither have barrier semantics, and instead provide only a control
- *   dependency.
- *
- * - The UNLOCK routines are supplemented with shared_tryunlock(), which
- *   fails if the caller appears to be the last lock holder (yes, this is
- *   racy). All successful UNLOCK routines have RELEASE semantics.
- */
-static void arm_smmu_cmdq_shared_lock(struct arm_smmu_cmdq *cmdq)
+static void arm_smmu_cmdq_insert_cmd(struct arm_smmu_device *smmu, u64 *cmd)
 {
-	int val;
-
-	/*
-	 * We can try to avoid the cmpxchg() loop by simply incrementing the
-	 * lock counter. When held in exclusive state, the lock counter is set
-	 * to INT_MIN so these increments won't hurt as the value will remain
-	 * negative.
-	 */
-	if (atomic_fetch_inc_relaxed(&cmdq->lock) >= 0)
-		return;
-
-	do {
-		val = atomic_cond_read_relaxed(&cmdq->lock, VAL >= 0);
-	} while (atomic_cmpxchg_relaxed(&cmdq->lock, val, val + 1) != val);
-}
-
-static void arm_smmu_cmdq_shared_unlock(struct arm_smmu_cmdq *cmdq)
-{
-	(void)atomic_dec_return_release(&cmdq->lock);
-}
-
-static bool arm_smmu_cmdq_shared_tryunlock(struct arm_smmu_cmdq *cmdq)
-{
-	if (atomic_read(&cmdq->lock) == 1)
-		return false;
-
-	arm_smmu_cmdq_shared_unlock(cmdq);
-	return true;
-}
-
-#define arm_smmu_cmdq_exclusive_trylock_irqsave(cmdq, flags)		\
-({									\
-	bool __ret;							\
-	local_irq_save(flags);						\
-	__ret = !atomic_cmpxchg_relaxed(&cmdq->lock, 0, INT_MIN);	\
-	if (!__ret)							\
-		local_irq_restore(flags);				\
-	__ret;								\
-})
-
-#define arm_smmu_cmdq_exclusive_unlock_irqrestore(cmdq, flags)		\
-({									\
-	atomic_set_release(&cmdq->lock, 0);				\
-	local_irq_restore(flags);					\
-})
-
-
-/*
- * Command queue insertion.
- * This is made fiddly by our attempts to achieve some sort of scalability
- * since there is one queue shared amongst all of the CPUs in the system.  If
- * you like mixed-size concurrency, dependency ordering and relaxed atomics,
- * then you'll *love* this monstrosity.
- *
- * The basic idea is to split the queue up into ranges of commands that are
- * owned by a given CPU; the owner may not have written all of the commands
- * itself, but is responsible for advancing the hardware prod pointer when
- * the time comes. The algorithm is roughly:
- *
- * 	1. Allocate some space in the queue. At this point we also discover
- *	   whether the head of the queue is currently owned by another CPU,
- *	   or whether we are the owner.
- *
- *	2. Write our commands into our allocated slots in the queue.
- *
- *	3. Mark our slots as valid in arm_smmu_cmdq.valid_map.
- *
- *	4. If we are an owner:
- *		a. Wait for the previous owner to finish.
- *		b. Mark the queue head as unowned, which tells us the range
- *		   that we are responsible for publishing.
- *		c. Wait for all commands in our owned range to become valid.
- *		d. Advance the hardware prod pointer.
- *		e. Tell the next owner we've finished.
- *
- *	5. If we are inserting a CMD_SYNC (we may or may not have been an
- *	   owner), then we need to stick around until it has completed:
- *		a. If we have MSIs, the SMMU can write back into the CMD_SYNC
- *		   to clear the first 4 bytes.
- *		b. Otherwise, we spin waiting for the hardware cons pointer to
- *		   advance past our command.
- *
- * The devil is in the details, particularly the use of locking for handling
- * SYNC completion and freeing up space in the queue before we think that it is
- * full.
- */
-static void __arm_smmu_cmdq_poll_set_valid_map(struct arm_smmu_cmdq *cmdq,
-					       u32 sprod, u32 eprod, bool set)
-{
-	u32 swidx, sbidx, ewidx, ebidx;
-	struct arm_smmu_ll_queue llq = {
-		.max_n_shift	= cmdq->q.llq.max_n_shift,
-		.prod		= sprod,
-	};
-
-	ewidx = BIT_WORD(Q_IDX(&llq, eprod));
-	ebidx = Q_IDX(&llq, eprod) % BITS_PER_LONG;
-
-	while (llq.prod != eprod) {
-		unsigned long mask;
-		atomic_long_t *ptr;
-		u32 limit = BITS_PER_LONG;
-
-		swidx = BIT_WORD(Q_IDX(&llq, llq.prod));
-		sbidx = Q_IDX(&llq, llq.prod) % BITS_PER_LONG;
-
-		ptr = &cmdq->valid_map[swidx];
-
-		if ((swidx == ewidx) && (sbidx < ebidx))
-			limit = ebidx;
-
-		mask = GENMASK(limit - 1, sbidx);
-
-		/*
-		 * The valid bit is the inverse of the wrap bit. This means
-		 * that a zero-initialised queue is invalid and, after marking
-		 * all entries as valid, they become invalid again when we
-		 * wrap.
-		 */
-		if (set) {
-			atomic_long_xor(mask, ptr);
-		} else { /* Poll */
-			unsigned long valid;
+	struct arm_smmu_queue *q = &smmu->cmdq.q;
+	bool wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
 
-			valid = (ULONG_MAX + !!Q_WRP(&llq, llq.prod)) & mask;
-			atomic_long_cond_read_relaxed(ptr, (VAL & mask) == valid);
-		}
+	smmu->prev_cmd_opcode = FIELD_GET(CMDQ_0_OP, cmd[0]);
 
-		llq.prod = queue_inc_prod_n(&llq, limit - sbidx);
+	while (queue_insert_raw(q, cmd) == -ENOSPC) {
+		if (queue_poll_cons(q, false, wfe))
+			dev_err_ratelimited(smmu->dev, "CMDQ timeout\n");
 	}
 }
 
-/* Mark all entries in the range [sprod, eprod) as valid */
-static void arm_smmu_cmdq_set_valid_map(struct arm_smmu_cmdq *cmdq,
-					u32 sprod, u32 eprod)
-{
-	__arm_smmu_cmdq_poll_set_valid_map(cmdq, sprod, eprod, true);
-}
-
-/* Wait for all entries in the range [sprod, eprod) to become valid */
-static void arm_smmu_cmdq_poll_valid_map(struct arm_smmu_cmdq *cmdq,
-					 u32 sprod, u32 eprod)
-{
-	__arm_smmu_cmdq_poll_set_valid_map(cmdq, sprod, eprod, false);
-}
-
-/* Wait for the command queue to become non-full */
-static int arm_smmu_cmdq_poll_until_not_full(struct arm_smmu_device *smmu,
-					     struct arm_smmu_ll_queue *llq)
+static void arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
+				    struct arm_smmu_cmdq_ent *ent)
 {
+	u64 cmd[CMDQ_ENT_DWORDS];
 	unsigned long flags;
-	struct arm_smmu_queue_poll qp;
-	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
-	int ret = 0;
 
-	/*
-	 * Try to update our copy of cons by grabbing exclusive cmdq access. If
-	 * that fails, spin until somebody else updates it for us.
-	 */
-	if (arm_smmu_cmdq_exclusive_trylock_irqsave(cmdq, flags)) {
-		WRITE_ONCE(cmdq->q.llq.cons, readl_relaxed(cmdq->q.cons_reg));
-		arm_smmu_cmdq_exclusive_unlock_irqrestore(cmdq, flags);
-		llq->val = READ_ONCE(cmdq->q.llq.val);
-		return 0;
+	if (arm_smmu_cmdq_build_cmd(cmd, ent)) {
+		dev_warn(smmu->dev, "ignoring unknown CMDQ opcode 0x%x\n",
+			 ent->opcode);
+		return;
 	}
 
-	queue_poll_init(smmu, &qp);
-	do {
-		llq->val = READ_ONCE(smmu->cmdq.q.llq.val);
-		if (!queue_full(llq))
-			break;
-
-		ret = queue_poll(&qp);
-	} while (!ret);
-
-	return ret;
-}
-
-/*
- * Wait until the SMMU signals a CMD_SYNC completion MSI.
- * Must be called with the cmdq lock held in some capacity.
- */
-static int __arm_smmu_cmdq_poll_until_msi(struct arm_smmu_device *smmu,
-					  struct arm_smmu_ll_queue *llq)
-{
-	int ret = 0;
-	struct arm_smmu_queue_poll qp;
-	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
-	u32 *cmd = (u32 *)(Q_ENT(&cmdq->q, llq->prod));
-
-	queue_poll_init(smmu, &qp);
-
-	/*
-	 * The MSI won't generate an event, since it's being written back
-	 * into the command queue.
-	 */
-	qp.wfe = false;
-	smp_cond_load_relaxed(cmd, !VAL || (ret = queue_poll(&qp)));
-	llq->cons = ret ? llq->prod : queue_inc_prod_n(llq, 1);
-	return ret;
+	spin_lock_irqsave(&smmu->cmdq.lock, flags);
+	arm_smmu_cmdq_insert_cmd(smmu, cmd);
+	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
 }
 
 /*
- * Wait until the SMMU cons index passes llq->prod.
- * Must be called with the cmdq lock held in some capacity.
+ * The difference between val and sync_idx is bounded by the maximum size of
+ * a queue at 2^20 entries, so 32 bits is plenty for wrap-safe arithmetic.
  */
-static int __arm_smmu_cmdq_poll_until_consumed(struct arm_smmu_device *smmu,
-					       struct arm_smmu_ll_queue *llq)
+static int __arm_smmu_sync_poll_msi(struct arm_smmu_device *smmu, u32 sync_idx)
 {
-	struct arm_smmu_queue_poll qp;
-	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
-	u32 prod = llq->prod;
-	int ret = 0;
+	ktime_t timeout;
+	u32 val;
 
-	queue_poll_init(smmu, &qp);
-	llq->val = READ_ONCE(smmu->cmdq.q.llq.val);
-	do {
-		if (queue_consumed(llq, prod))
-			break;
-
-		ret = queue_poll(&qp);
-
-		/*
-		 * This needs to be a readl() so that our subsequent call
-		 * to arm_smmu_cmdq_shared_tryunlock() can fail accurately.
-		 *
-		 * Specifically, we need to ensure that we observe all
-		 * shared_lock()s by other CMD_SYNCs that share our owner,
-		 * so that a failing call to tryunlock() means that we're
-		 * the last one out and therefore we can safely advance
-		 * cmdq->q.llq.cons. Roughly speaking:
-		 *
-		 * CPU 0		CPU1			CPU2 (us)
-		 *
-		 * if (sync)
-		 * 	shared_lock();
-		 *
-		 * dma_wmb();
-		 * set_valid_map();
-		 *
-		 * 			if (owner) {
-		 *				poll_valid_map();
-		 *				<control dependency>
-		 *				writel(prod_reg);
-		 *
-		 *						readl(cons_reg);
-		 *						tryunlock();
-		 *
-		 * Requires us to see CPU 0's shared_lock() acquisition.
-		 */
-		llq->cons = readl(cmdq->q.cons_reg);
-	} while (!ret);
+	timeout = ktime_add_us(ktime_get(), ARM_SMMU_CMDQ_SYNC_TIMEOUT_US);
+	val = smp_cond_load_acquire(&smmu->sync_count,
+				    (int)(VAL - sync_idx) >= 0 ||
+				    !ktime_before(ktime_get(), timeout));
 
-	return ret;
+	return (int)(val - sync_idx) < 0 ? -ETIMEDOUT : 0;
 }
 
-static int arm_smmu_cmdq_poll_until_sync(struct arm_smmu_device *smmu,
-					 struct arm_smmu_ll_queue *llq)
+static int __arm_smmu_cmdq_issue_sync_msi(struct arm_smmu_device *smmu)
 {
-	if (smmu->features & ARM_SMMU_FEAT_MSI &&
-	    smmu->features & ARM_SMMU_FEAT_COHERENCY)
-		return __arm_smmu_cmdq_poll_until_msi(smmu, llq);
-
-	return __arm_smmu_cmdq_poll_until_consumed(smmu, llq);
-}
-
-static void arm_smmu_cmdq_write_entries(struct arm_smmu_cmdq *cmdq, u64 *cmds,
-					u32 prod, int n)
-{
-	int i;
-	struct arm_smmu_ll_queue llq = {
-		.max_n_shift	= cmdq->q.llq.max_n_shift,
-		.prod		= prod,
-	};
-
-	for (i = 0; i < n; ++i) {
-		u64 *cmd = &cmds[i * CMDQ_ENT_DWORDS];
-
-		prod = queue_inc_prod_n(&llq, i);
-		queue_write(Q_ENT(&cmdq->q, prod), cmd, CMDQ_ENT_DWORDS);
-	}
-}
-
-/*
- * This is the actual insertion function, and provides the following
- * ordering guarantees to callers:
- *
- * - There is a dma_wmb() before publishing any commands to the queue.
- *   This can be relied upon to order prior writes to data structures
- *   in memory (such as a CD or an STE) before the command.
- *
- * - On completion of a CMD_SYNC, there is a control dependency.
- *   This can be relied upon to order subsequent writes to memory (e.g.
- *   freeing an IOVA) after completion of the CMD_SYNC.
- *
- * - Command insertion is totally ordered, so if two CPUs each race to
- *   insert their own list of commands then all of the commands from one
- *   CPU will appear before any of the commands from the other CPU.
- */
-static int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
-				       u64 *cmds, int n, bool sync)
-{
-	u64 cmd_sync[CMDQ_ENT_DWORDS];
-	u32 prod;
+	u64 cmd[CMDQ_ENT_DWORDS];
 	unsigned long flags;
-	bool owner;
-	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
-	struct arm_smmu_ll_queue llq = {
-		.max_n_shift = cmdq->q.llq.max_n_shift,
-	}, head = llq;
-	int ret = 0;
-
-	/* 1. Allocate some space in the queue */
-	local_irq_save(flags);
-	llq.val = READ_ONCE(cmdq->q.llq.val);
-	do {
-		u64 old;
-
-		while (!queue_has_space(&llq, n + sync)) {
-			local_irq_restore(flags);
-			if (arm_smmu_cmdq_poll_until_not_full(smmu, &llq))
-				dev_err_ratelimited(smmu->dev, "CMDQ timeout\n");
-			local_irq_save(flags);
-		}
-
-		head.cons = llq.cons;
-		head.prod = queue_inc_prod_n(&llq, n + sync) |
-					     CMDQ_PROD_OWNED_FLAG;
-
-		old = cmpxchg_relaxed(&cmdq->q.llq.val, llq.val, head.val);
-		if (old == llq.val)
-			break;
-
-		llq.val = old;
-	} while (1);
-	owner = !(llq.prod & CMDQ_PROD_OWNED_FLAG);
-	head.prod &= ~CMDQ_PROD_OWNED_FLAG;
-	llq.prod &= ~CMDQ_PROD_OWNED_FLAG;
-
-	/*
-	 * 2. Write our commands into the queue
-	 * Dependency ordering from the cmpxchg() loop above.
-	 */
-	arm_smmu_cmdq_write_entries(cmdq, cmds, llq.prod, n);
-	if (sync) {
-		prod = queue_inc_prod_n(&llq, n);
-		arm_smmu_cmdq_build_sync_cmd(cmd_sync, smmu, prod);
-		queue_write(Q_ENT(&cmdq->q, prod), cmd_sync, CMDQ_ENT_DWORDS);
-
-		/*
-		 * In order to determine completion of our CMD_SYNC, we must
-		 * ensure that the queue can't wrap twice without us noticing.
-		 * We achieve that by taking the cmdq lock as shared before
-		 * marking our slot as valid.
-		 */
-		arm_smmu_cmdq_shared_lock(cmdq);
-	}
-
-	/* 3. Mark our slots as valid, ensuring commands are visible first */
-	dma_wmb();
-	arm_smmu_cmdq_set_valid_map(cmdq, llq.prod, head.prod);
-
-	/* 4. If we are the owner, take control of the SMMU hardware */
-	if (owner) {
-		/* a. Wait for previous owner to finish */
-		atomic_cond_read_relaxed(&cmdq->owner_prod, VAL == llq.prod);
-
-		/* b. Stop gathering work by clearing the owned flag */
-		prod = atomic_fetch_andnot_relaxed(CMDQ_PROD_OWNED_FLAG,
-						   &cmdq->q.llq.atomic.prod);
-		prod &= ~CMDQ_PROD_OWNED_FLAG;
+	struct arm_smmu_cmdq_ent  ent = {
+		.opcode = CMDQ_OP_CMD_SYNC,
+		.sync	= {
+			.msiaddr = virt_to_phys(&smmu->sync_count),
+		},
+	};
 
-		/*
-		 * c. Wait for any gathered work to be written to the queue.
-		 * Note that we read our own entries so that we have the control
-		 * dependency required by (d).
-		 */
-		arm_smmu_cmdq_poll_valid_map(cmdq, llq.prod, prod);
+	spin_lock_irqsave(&smmu->cmdq.lock, flags);
 
-		/*
-		 * d. Advance the hardware prod pointer
-		 * Control dependency ordering from the entries becoming valid.
-		 */
-		writel_relaxed(prod, cmdq->q.prod_reg);
-
-		/*
-		 * e. Tell the next owner we're done
-		 * Make sure we've updated the hardware first, so that we don't
-		 * race to update prod and potentially move it backwards.
-		 */
-		atomic_set_release(&cmdq->owner_prod, prod);
+	/* Piggy-back on the previous command if it's a SYNC */
+	if (smmu->prev_cmd_opcode == CMDQ_OP_CMD_SYNC) {
+		ent.sync.msidata = smmu->sync_nr;
+	} else {
+		ent.sync.msidata = ++smmu->sync_nr;
+		arm_smmu_cmdq_build_cmd(cmd, &ent);
+		arm_smmu_cmdq_insert_cmd(smmu, cmd);
 	}
 
-	/* 5. If we are inserting a CMD_SYNC, we must wait for it to complete */
-	if (sync) {
-		llq.prod = queue_inc_prod_n(&llq, n);
-		ret = arm_smmu_cmdq_poll_until_sync(smmu, &llq);
-		if (ret) {
-			dev_err_ratelimited(smmu->dev,
-					    "CMD_SYNC timeout at 0x%08x [hwprod 0x%08x, hwcons 0x%08x]\n",
-					    llq.prod,
-					    readl_relaxed(cmdq->q.prod_reg),
-					    readl_relaxed(cmdq->q.cons_reg));
-		}
-
-		/*
-		 * Try to unlock the cmdq lock. This will fail if we're the last
-		 * reader, in which case we can safely update cmdq->q.llq.cons
-		 */
-		if (!arm_smmu_cmdq_shared_tryunlock(cmdq)) {
-			WRITE_ONCE(cmdq->q.llq.cons, llq.cons);
-			arm_smmu_cmdq_shared_unlock(cmdq);
-		}
-	}
+	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
 
-	local_irq_restore(flags);
-	return ret;
+	return __arm_smmu_sync_poll_msi(smmu, ent.sync.msidata);
 }
 
-static int arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
-				   struct arm_smmu_cmdq_ent *ent)
+static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
 {
 	u64 cmd[CMDQ_ENT_DWORDS];
+	unsigned long flags;
+	bool wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
+	struct arm_smmu_cmdq_ent ent = { .opcode = CMDQ_OP_CMD_SYNC };
+	int ret;
 
-	if (arm_smmu_cmdq_build_cmd(cmd, ent)) {
-		dev_warn(smmu->dev, "ignoring unknown CMDQ opcode 0x%x\n",
-			 ent->opcode);
-		return -EINVAL;
-	}
+	arm_smmu_cmdq_build_cmd(cmd, &ent);
 
-	return arm_smmu_cmdq_issue_cmdlist(smmu, cmd, 1, false);
-}
+	spin_lock_irqsave(&smmu->cmdq.lock, flags);
+	arm_smmu_cmdq_insert_cmd(smmu, cmd);
+	ret = queue_poll_cons(&smmu->cmdq.q, true, wfe);
+	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
 
-static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
-{
-	return arm_smmu_cmdq_issue_cmdlist(smmu, NULL, 0, true);
+	return ret;
 }
 
-static void arm_smmu_cmdq_batch_add(struct arm_smmu_device *smmu,
-				    struct arm_smmu_cmdq_batch *cmds,
-				    struct arm_smmu_cmdq_ent *cmd)
+static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
 {
-	if (cmds->num == CMDQ_BATCH_ENTRIES) {
-		arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, false);
-		cmds->num = 0;
-	}
-	arm_smmu_cmdq_build_cmd(&cmds->cmds[cmds->num * CMDQ_ENT_DWORDS], cmd);
-	cmds->num++;
-}
+	int ret;
+	bool msi = (smmu->features & ARM_SMMU_FEAT_MSI) &&
+		   (smmu->features & ARM_SMMU_FEAT_COHERENCY);
 
-static int arm_smmu_cmdq_batch_submit(struct arm_smmu_device *smmu,
-				      struct arm_smmu_cmdq_batch *cmds)
-{
-	return arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, true);
+	ret = msi ? __arm_smmu_cmdq_issue_sync_msi(smmu)
+		  : __arm_smmu_cmdq_issue_sync(smmu);
+	if (ret)
+		dev_err_ratelimited(smmu->dev, "CMD_SYNC timeout\n");
+	return ret;
 }
 
 /* Context descriptor manipulation functions */
@@ -1535,7 +1137,6 @@ static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
 	size_t i;
 	unsigned long flags;
 	struct arm_smmu_master *master;
-	struct arm_smmu_cmdq_batch cmds = {};
 	struct arm_smmu_device *smmu = smmu_domain->smmu;
 	struct arm_smmu_cmdq_ent cmd = {
 		.opcode	= CMDQ_OP_CFGI_CD,
@@ -1549,12 +1150,12 @@ static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
 	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
 		for (i = 0; i < master->num_sids; i++) {
 			cmd.cfgi.sid = master->sids[i];
-			arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
+			arm_smmu_cmdq_issue_cmd(smmu, &cmd);
 		}
 	}
 	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
 
-	arm_smmu_cmdq_batch_submit(smmu, &cmds);
+	arm_smmu_cmdq_issue_sync(smmu);
 }
 
 static int arm_smmu_alloc_cd_leaf_table(struct arm_smmu_device *smmu,
@@ -2189,16 +1790,17 @@ arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size,
 	cmd->atc.size	= log2_span;
 }
 
-static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
+static int arm_smmu_atc_inv_master(struct arm_smmu_master *master,
+				   struct arm_smmu_cmdq_ent *cmd)
 {
 	int i;
-	struct arm_smmu_cmdq_ent cmd;
 
-	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
+	if (!master->ats_enabled)
+		return 0;
 
 	for (i = 0; i < master->num_sids; i++) {
-		cmd.atc.sid = master->sids[i];
-		arm_smmu_cmdq_issue_cmd(master->smmu, &cmd);
+		cmd->atc.sid = master->sids[i];
+		arm_smmu_cmdq_issue_cmd(master->smmu, cmd);
 	}
 
 	return arm_smmu_cmdq_issue_sync(master->smmu);
@@ -2207,11 +1809,10 @@ static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
 static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
 				   int ssid, unsigned long iova, size_t size)
 {
-	int i;
+	int ret = 0;
 	unsigned long flags;
 	struct arm_smmu_cmdq_ent cmd;
 	struct arm_smmu_master *master;
-	struct arm_smmu_cmdq_batch cmds = {};
 
 	if (!(smmu_domain->smmu->features & ARM_SMMU_FEAT_ATS))
 		return 0;
@@ -2236,18 +1837,11 @@ static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
 	arm_smmu_atc_inv_to_cmd(ssid, iova, size, &cmd);
 
 	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
-		if (!master->ats_enabled)
-			continue;
-
-		for (i = 0; i < master->num_sids; i++) {
-			cmd.atc.sid = master->sids[i];
-			arm_smmu_cmdq_batch_add(smmu_domain->smmu, &cmds, &cmd);
-		}
-	}
+	list_for_each_entry(master, &smmu_domain->devices, domain_head)
+		ret |= arm_smmu_atc_inv_master(master, &cmd);
 	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
 
-	return arm_smmu_cmdq_batch_submit(smmu_domain->smmu, &cmds);
+	return ret ? -ETIMEDOUT : 0;
 }
 
 /* IO_PGTABLE API */
@@ -2269,32 +1863,27 @@ static void arm_smmu_tlb_inv_context(void *cookie)
 	/*
 	 * NOTE: when io-pgtable is in non-strict mode, we may get here with
 	 * PTEs previously cleared by unmaps on the current CPU not yet visible
-	 * to the SMMU. We are relying on the dma_wmb() implicit during cmd
-	 * insertion to guarantee those are observed before the TLBI. Do be
-	 * careful, 007.
+	 * to the SMMU. We are relying on the DSB implicit in
+	 * queue_sync_prod_out() to guarantee those are observed before the
+	 * TLBI. Do be careful, 007.
 	 */
 	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
 	arm_smmu_cmdq_issue_sync(smmu);
 	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
 }
 
-static void arm_smmu_tlb_inv_range(unsigned long iova, size_t size,
-				   size_t granule, bool leaf,
-				   struct arm_smmu_domain *smmu_domain)
+static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
+					  size_t granule, bool leaf, void *cookie)
 {
+	struct arm_smmu_domain *smmu_domain = cookie;
 	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	unsigned long start = iova, end = iova + size, num_pages = 0, tg = 0;
-	size_t inv_range = granule;
-	struct arm_smmu_cmdq_batch cmds = {};
 	struct arm_smmu_cmdq_ent cmd = {
 		.tlbi = {
 			.leaf	= leaf,
+			.addr	= iova,
 		},
 	};
 
-	if (!size)
-		return;
-
 	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
 		cmd.opcode	= CMDQ_OP_TLBI_NH_VA;
 		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
@@ -2303,78 +1892,37 @@ static void arm_smmu_tlb_inv_range(unsigned long iova, size_t size,
 		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
 	}
 
-	if (smmu->features & ARM_SMMU_FEAT_RANGE_INV) {
-		/* Get the leaf page size */
-		tg = __ffs(smmu_domain->domain.pgsize_bitmap);
-
-		/* Convert page size of 12,14,16 (log2) to 1,2,3 */
-		cmd.tlbi.tg = (tg - 10) / 2;
-
-		/* Determine what level the granule is at */
-		cmd.tlbi.ttl = 4 - ((ilog2(granule) - 3) / (tg - 3));
-
-		num_pages = size >> tg;
-	}
-
-	while (iova < end) {
-		if (smmu->features & ARM_SMMU_FEAT_RANGE_INV) {
-			/*
-			 * On each iteration of the loop, the range is 5 bits
-			 * worth of the aligned size remaining.
-			 * The range in pages is:
-			 *
-			 * range = (num_pages & (0x1f << __ffs(num_pages)))
-			 */
-			unsigned long scale, num;
-
-			/* Determine the power of 2 multiple number of pages */
-			scale = __ffs(num_pages);
-			cmd.tlbi.scale = scale;
-
-			/* Determine how many chunks of 2^scale size we have */
-			num = (num_pages >> scale) & CMDQ_TLBI_RANGE_NUM_MAX;
-			cmd.tlbi.num = num - 1;
-
-			/* range is num * 2^scale * pgsize */
-			inv_range = num << (scale + tg);
-
-			/* Clear out the lower order bits for the next iteration */
-			num_pages -= num << scale;
-		}
-
-		cmd.tlbi.addr = iova;
-		arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
-		iova += inv_range;
-	}
-	arm_smmu_cmdq_batch_submit(smmu, &cmds);
-
-	/*
-	 * Unfortunately, this can't be leaf-only since we may have
-	 * zapped an entire table.
-	 */
-	arm_smmu_atc_inv_domain(smmu_domain, 0, start, size);
+	do {
+		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
+		cmd.tlbi.addr += granule;
+	} while (size -= granule);
 }
 
 static void arm_smmu_tlb_inv_page_nosync(struct iommu_iotlb_gather *gather,
 					 unsigned long iova, size_t granule,
 					 void *cookie)
 {
-	struct arm_smmu_domain *smmu_domain = cookie;
-	struct iommu_domain *domain = &smmu_domain->domain;
-
-	iommu_iotlb_gather_add_page(domain, gather, iova, granule);
+	arm_smmu_tlb_inv_range_nosync(iova, granule, granule, true, cookie);
 }
 
 static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,
 				  size_t granule, void *cookie)
 {
-	arm_smmu_tlb_inv_range(iova, size, granule, false, cookie);
+	struct arm_smmu_domain *smmu_domain = cookie;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+
+	arm_smmu_tlb_inv_range_nosync(iova, size, granule, false, cookie);
+	arm_smmu_cmdq_issue_sync(smmu);
 }
 
 static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,
 				  size_t granule, void *cookie)
 {
-	arm_smmu_tlb_inv_range(iova, size, granule, true, cookie);
+	struct arm_smmu_domain *smmu_domain = cookie;
+	struct arm_smmu_device *smmu = smmu_domain->smmu;
+
+	arm_smmu_tlb_inv_range_nosync(iova, size, granule, true, cookie);
+	arm_smmu_cmdq_issue_sync(smmu);
 }
 
 static const struct iommu_flush_ops arm_smmu_flush_ops = {
@@ -2700,6 +2248,7 @@ static void arm_smmu_enable_ats(struct arm_smmu_master *master)
 
 static void arm_smmu_disable_ats(struct arm_smmu_master *master)
 {
+	struct arm_smmu_cmdq_ent cmd;
 	struct arm_smmu_domain *smmu_domain = master->domain;
 
 	if (!master->ats_enabled)
@@ -2711,8 +2260,9 @@ static void arm_smmu_disable_ats(struct arm_smmu_master *master)
 	 * ATC invalidation via the SMMU.
 	 */
 	wmb();
-	arm_smmu_atc_inv_master(master);
-	atomic_dec(&smmu_domain->nr_ats_masters);
+	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
+	arm_smmu_atc_inv_master(master, &cmd);
+    atomic_dec(&smmu_domain->nr_ats_masters);
 }
 
 static int arm_smmu_enable_pasid(struct arm_smmu_master *master)
@@ -2875,10 +2425,10 @@ static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain)
 static void arm_smmu_iotlb_sync(struct iommu_domain *domain,
 				struct iommu_iotlb_gather *gather)
 {
-	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
+	struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
 
-	arm_smmu_tlb_inv_range(gather->start, gather->end - gather->start,
-			       gather->pgsize, true, smmu_domain);
+	if (smmu)
+		arm_smmu_cmdq_issue_sync(smmu);
 }
 
 static phys_addr_t
@@ -3176,49 +2726,18 @@ static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
 	return 0;
 }
 
-static void arm_smmu_cmdq_free_bitmap(void *data)
-{
-	unsigned long *bitmap = data;
-	bitmap_free(bitmap);
-}
-
-static int arm_smmu_cmdq_init(struct arm_smmu_device *smmu)
-{
-	int ret = 0;
-	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
-	unsigned int nents = 1 << cmdq->q.llq.max_n_shift;
-	atomic_long_t *bitmap;
-
-	atomic_set(&cmdq->owner_prod, 0);
-	atomic_set(&cmdq->lock, 0);
-
-	bitmap = (atomic_long_t *)bitmap_zalloc(nents, GFP_KERNEL);
-	if (!bitmap) {
-		dev_err(smmu->dev, "failed to allocate cmdq bitmap\n");
-		ret = -ENOMEM;
-	} else {
-		cmdq->valid_map = bitmap;
-		devm_add_action(smmu->dev, arm_smmu_cmdq_free_bitmap, bitmap);
-	}
-
-	return ret;
-}
-
 static int arm_smmu_init_queues(struct arm_smmu_device *smmu)
 {
 	int ret;
 
 	/* cmdq */
+	spin_lock_init(&smmu->cmdq.lock);
 	ret = arm_smmu_init_one_queue(smmu, &smmu->cmdq.q, ARM_SMMU_CMDQ_PROD,
 				      ARM_SMMU_CMDQ_CONS, CMDQ_ENT_DWORDS,
 				      "cmdq");
 	if (ret)
 		return ret;
 
-	ret = arm_smmu_cmdq_init(smmu);
-	if (ret)
-		return ret;
-
 	/* evtq */
 	ret = arm_smmu_init_one_queue(smmu, &smmu->evtq.q, ARM_SMMU_EVTQ_PROD,
 				      ARM_SMMU_EVTQ_CONS, EVTQ_ENT_DWORDS,
@@ -3799,15 +3318,9 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 	/* Queue sizes, capped to ensure natural alignment */
 	smmu->cmdq.q.llq.max_n_shift = min_t(u32, CMDQ_MAX_SZ_SHIFT,
 					     FIELD_GET(IDR1_CMDQS, reg));
-	if (smmu->cmdq.q.llq.max_n_shift <= ilog2(CMDQ_BATCH_ENTRIES)) {
-		/*
-		 * We don't support splitting up batches, so one batch of
-		 * commands plus an extra sync needs to fit inside the command
-		 * queue. There's also no way we can handle the weird alignment
-		 * restrictions on the base pointer for a unit-length queue.
-		 */
-		dev_err(smmu->dev, "command queue size <= %d entries not supported\n",
-			CMDQ_BATCH_ENTRIES);
+	if (!smmu->cmdq.q.llq.max_n_shift) {
+		/* Odd alignment restrictions on the base, so ignore for now */
+		dev_err(smmu->dev, "unit-length command queue not supported\n");
 		return -ENXIO;
 	}
 
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 53+ messages in thread

* [PATCH v2 3/8] xen/arm: revert patch related to XArray
  2020-11-26 17:01 [PATCH v2 0/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
  2020-11-26 17:02 ` [PATCH v2 1/8] xen/arm: Import the SMMUv3 driver from Linux Rahul Singh
  2020-11-26 17:02 ` [PATCH v2 2/8] xen/arm: revert atomic operation related command-queue insertion patch Rahul Singh
@ 2020-11-26 17:02 ` Rahul Singh
  2020-12-02  0:20   ` Stefano Stabellini
  2020-12-02 13:46   ` Julien Grall
  2020-11-26 17:02 ` [PATCH v2 4/8] xen/arm: Remove support for MSI on SMMUv3 Rahul Singh
                   ` (4 subsequent siblings)
  7 siblings, 2 replies; 53+ messages in thread
From: Rahul Singh @ 2020-11-26 17:02 UTC (permalink / raw)
  To: xen-devel
  Cc: bertrand.marquis, rahul.singh, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

XArray is not implemented in XEN revert the patch that introduce the
XArray code in SMMUv3 driver.

Once XArray is implemented in XEN this patch can be added in XEN.

Reverted the commit 0299a1a81ca056e79c1a7fb751f936ec0d5c7afe

Signed-off-by: Rahul Singh <rahul.singh@arm.com>
---
 xen/drivers/passthrough/arm/smmu-v3.c | 27 +++++++++------------------
 1 file changed, 9 insertions(+), 18 deletions(-)

diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
index 97eac61ea4..cec304e51a 100644
--- a/xen/drivers/passthrough/arm/smmu-v3.c
+++ b/xen/drivers/passthrough/arm/smmu-v3.c
@@ -638,6 +638,7 @@ struct arm_smmu_device {
 
 #define ARM_SMMU_MAX_ASIDS		(1 << 16)
 	unsigned int			asid_bits;
+	DECLARE_BITMAP(asid_map, ARM_SMMU_MAX_ASIDS);
 
 #define ARM_SMMU_MAX_VMIDS		(1 << 16)
 	unsigned int			vmid_bits;
@@ -703,8 +704,6 @@ struct arm_smmu_option_prop {
 	const char *prop;
 };
 
-static DEFINE_XARRAY_ALLOC1(asid_xa);
-
 static struct arm_smmu_option_prop arm_smmu_options[] = {
 	{ ARM_SMMU_OPT_SKIP_PREFETCH, "hisilicon,broken-prefetch-cmd" },
 	{ ARM_SMMU_OPT_PAGE0_REGS_ONLY, "cavium,cn9900-broken-page1-regspace"},
@@ -1366,14 +1365,6 @@ static void arm_smmu_free_cd_tables(struct arm_smmu_domain *smmu_domain)
 	cdcfg->cdtab = NULL;
 }
 
-static void arm_smmu_free_asid(struct arm_smmu_ctx_desc *cd)
-{
-	if (!cd->asid)
-		return;
-
-	xa_erase(&asid_xa, cd->asid);
-}
-
 /* Stream table manipulation functions */
 static void
 arm_smmu_write_strtab_l1_desc(__le64 *dst, struct arm_smmu_strtab_l1_desc *desc)
@@ -2006,9 +1997,10 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
 	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
 		struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
 
-		if (cfg->cdcfg.cdtab)
+		if (cfg->cdcfg.cdtab) {
 			arm_smmu_free_cd_tables(smmu_domain);
-		arm_smmu_free_asid(&cfg->cd);
+			arm_smmu_bitmap_free(smmu->asid_map, cfg->cd.asid);
+		}
 	} else {
 		struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
 		if (cfg->vmid)
@@ -2023,15 +2015,14 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 				       struct io_pgtable_cfg *pgtbl_cfg)
 {
 	int ret;
-	u32 asid;
+	int asid;
 	struct arm_smmu_device *smmu = smmu_domain->smmu;
 	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
 	typeof(&pgtbl_cfg->arm_lpae_s1_cfg.tcr) tcr = &pgtbl_cfg->arm_lpae_s1_cfg.tcr;
 
-	ret = xa_alloc(&asid_xa, &asid, &cfg->cd,
-		       XA_LIMIT(1, (1 << smmu->asid_bits) - 1), GFP_KERNEL);
-	if (ret)
-		return ret;
+	asid = arm_smmu_bitmap_alloc(smmu->asid_map, smmu->asid_bits);
+	if (asid < 0)
+		return asid;
 
 	cfg->s1cdmax = master->ssid_bits;
 
@@ -2064,7 +2055,7 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
 out_free_cd_tables:
 	arm_smmu_free_cd_tables(smmu_domain);
 out_free_asid:
-	arm_smmu_free_asid(&cfg->cd);
+	arm_smmu_bitmap_free(smmu->asid_map, asid);
 	return ret;
 }
 
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 53+ messages in thread

* [PATCH v2 4/8] xen/arm: Remove support for MSI on SMMUv3
  2020-11-26 17:01 [PATCH v2 0/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
                   ` (2 preceding siblings ...)
  2020-11-26 17:02 ` [PATCH v2 3/8] xen/arm: revert patch related to XArray Rahul Singh
@ 2020-11-26 17:02 ` Rahul Singh
  2020-12-02  0:33   ` Stefano Stabellini
  2020-11-26 17:02 ` [PATCH v2 5/8] xen/arm: Remove support for PCI ATS " Rahul Singh
                   ` (3 subsequent siblings)
  7 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-11-26 17:02 UTC (permalink / raw)
  To: xen-devel
  Cc: bertrand.marquis, rahul.singh, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

XEN does not support MSI on ARM platforms, therefore remove the MSI
support from SMMUv3 driver.

Signed-off-by: Rahul Singh <rahul.singh@arm.com>
---
 xen/drivers/passthrough/arm/smmu-v3.c | 176 +-------------------------
 1 file changed, 3 insertions(+), 173 deletions(-)

diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
index cec304e51a..401f7ae006 100644
--- a/xen/drivers/passthrough/arm/smmu-v3.c
+++ b/xen/drivers/passthrough/arm/smmu-v3.c
@@ -416,31 +416,6 @@ enum pri_resp {
 	PRI_RESP_SUCC = 2,
 };
 
-enum arm_smmu_msi_index {
-	EVTQ_MSI_INDEX,
-	GERROR_MSI_INDEX,
-	PRIQ_MSI_INDEX,
-	ARM_SMMU_MAX_MSIS,
-};
-
-static phys_addr_t arm_smmu_msi_cfg[ARM_SMMU_MAX_MSIS][3] = {
-	[EVTQ_MSI_INDEX] = {
-		ARM_SMMU_EVTQ_IRQ_CFG0,
-		ARM_SMMU_EVTQ_IRQ_CFG1,
-		ARM_SMMU_EVTQ_IRQ_CFG2,
-	},
-	[GERROR_MSI_INDEX] = {
-		ARM_SMMU_GERROR_IRQ_CFG0,
-		ARM_SMMU_GERROR_IRQ_CFG1,
-		ARM_SMMU_GERROR_IRQ_CFG2,
-	},
-	[PRIQ_MSI_INDEX] = {
-		ARM_SMMU_PRIQ_IRQ_CFG0,
-		ARM_SMMU_PRIQ_IRQ_CFG1,
-		ARM_SMMU_PRIQ_IRQ_CFG2,
-	},
-};
-
 struct arm_smmu_cmdq_ent {
 	/* Common fields */
 	u8				opcode;
@@ -504,10 +479,6 @@ struct arm_smmu_cmdq_ent {
 		} pri;
 
 		#define CMDQ_OP_CMD_SYNC	0x46
-		struct {
-			u32			msidata;
-			u64			msiaddr;
-		} sync;
 	};
 };
 
@@ -649,12 +620,6 @@ struct arm_smmu_device {
 
 	struct arm_smmu_strtab_cfg	strtab_cfg;
 
-	/* Hi16xx adds an extra 32 bits of goodness to its MSI payload */
-	union {
-		u32			sync_count;
-		u64			padding;
-	};
-
 	/* IOMMU core code handle */
 	struct iommu_device		iommu;
 };
@@ -945,20 +910,7 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
 		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_RESP, ent->pri.resp);
 		break;
 	case CMDQ_OP_CMD_SYNC:
-		if (ent->sync.msiaddr)
-			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_IRQ);
-		else
-			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
-		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSH, ARM_SMMU_SH_ISH);
-		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIATTR, ARM_SMMU_MEMATTR_OIWB);
-		/*
-		 * Commands are written little-endian, but we want the SMMU to
-		 * receive MSIData, and thus write it back to memory, in CPU
-		 * byte order, so big-endian needs an extra byteswap here.
-		 */
-		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIDATA,
-				     cpu_to_le32(ent->sync.msidata));
-		cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
+		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
 		break;
 	default:
 		return -ENOENT;
@@ -1054,50 +1006,6 @@ static void arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
 	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
 }
 
-/*
- * The difference between val and sync_idx is bounded by the maximum size of
- * a queue at 2^20 entries, so 32 bits is plenty for wrap-safe arithmetic.
- */
-static int __arm_smmu_sync_poll_msi(struct arm_smmu_device *smmu, u32 sync_idx)
-{
-	ktime_t timeout;
-	u32 val;
-
-	timeout = ktime_add_us(ktime_get(), ARM_SMMU_CMDQ_SYNC_TIMEOUT_US);
-	val = smp_cond_load_acquire(&smmu->sync_count,
-				    (int)(VAL - sync_idx) >= 0 ||
-				    !ktime_before(ktime_get(), timeout));
-
-	return (int)(val - sync_idx) < 0 ? -ETIMEDOUT : 0;
-}
-
-static int __arm_smmu_cmdq_issue_sync_msi(struct arm_smmu_device *smmu)
-{
-	u64 cmd[CMDQ_ENT_DWORDS];
-	unsigned long flags;
-	struct arm_smmu_cmdq_ent  ent = {
-		.opcode = CMDQ_OP_CMD_SYNC,
-		.sync	= {
-			.msiaddr = virt_to_phys(&smmu->sync_count),
-		},
-	};
-
-	spin_lock_irqsave(&smmu->cmdq.lock, flags);
-
-	/* Piggy-back on the previous command if it's a SYNC */
-	if (smmu->prev_cmd_opcode == CMDQ_OP_CMD_SYNC) {
-		ent.sync.msidata = smmu->sync_nr;
-	} else {
-		ent.sync.msidata = ++smmu->sync_nr;
-		arm_smmu_cmdq_build_cmd(cmd, &ent);
-		arm_smmu_cmdq_insert_cmd(smmu, cmd);
-	}
-
-	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
-
-	return __arm_smmu_sync_poll_msi(smmu, ent.sync.msidata);
-}
-
 static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
 {
 	u64 cmd[CMDQ_ENT_DWORDS];
@@ -1119,12 +1027,9 @@ static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
 static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
 {
 	int ret;
-	bool msi = (smmu->features & ARM_SMMU_FEAT_MSI) &&
-		   (smmu->features & ARM_SMMU_FEAT_COHERENCY);
 
-	ret = msi ? __arm_smmu_cmdq_issue_sync_msi(smmu)
-		  : __arm_smmu_cmdq_issue_sync(smmu);
-	if (ret)
+	ret = __arm_smmu_cmdq_issue_sync(smmu);
+	if ( ret )
 		dev_err_ratelimited(smmu->dev, "CMD_SYNC timeout\n");
 	return ret;
 }
@@ -2898,83 +2803,10 @@ static int arm_smmu_update_gbpa(struct arm_smmu_device *smmu, u32 set, u32 clr)
 	return ret;
 }
 
-static void arm_smmu_free_msis(void *data)
-{
-	struct device *dev = data;
-	platform_msi_domain_free_irqs(dev);
-}
-
-static void arm_smmu_write_msi_msg(struct msi_desc *desc, struct msi_msg *msg)
-{
-	phys_addr_t doorbell;
-	struct device *dev = msi_desc_to_dev(desc);
-	struct arm_smmu_device *smmu = dev_get_drvdata(dev);
-	phys_addr_t *cfg = arm_smmu_msi_cfg[desc->platform.msi_index];
-
-	doorbell = (((u64)msg->address_hi) << 32) | msg->address_lo;
-	doorbell &= MSI_CFG0_ADDR_MASK;
-
-	writeq_relaxed(doorbell, smmu->base + cfg[0]);
-	writel_relaxed(msg->data, smmu->base + cfg[1]);
-	writel_relaxed(ARM_SMMU_MEMATTR_DEVICE_nGnRE, smmu->base + cfg[2]);
-}
-
-static void arm_smmu_setup_msis(struct arm_smmu_device *smmu)
-{
-	struct msi_desc *desc;
-	int ret, nvec = ARM_SMMU_MAX_MSIS;
-	struct device *dev = smmu->dev;
-
-	/* Clear the MSI address regs */
-	writeq_relaxed(0, smmu->base + ARM_SMMU_GERROR_IRQ_CFG0);
-	writeq_relaxed(0, smmu->base + ARM_SMMU_EVTQ_IRQ_CFG0);
-
-	if (smmu->features & ARM_SMMU_FEAT_PRI)
-		writeq_relaxed(0, smmu->base + ARM_SMMU_PRIQ_IRQ_CFG0);
-	else
-		nvec--;
-
-	if (!(smmu->features & ARM_SMMU_FEAT_MSI))
-		return;
-
-	if (!dev->msi_domain) {
-		dev_info(smmu->dev, "msi_domain absent - falling back to wired irqs\n");
-		return;
-	}
-
-	/* Allocate MSIs for evtq, gerror and priq. Ignore cmdq */
-	ret = platform_msi_domain_alloc_irqs(dev, nvec, arm_smmu_write_msi_msg);
-	if (ret) {
-		dev_warn(dev, "failed to allocate MSIs - falling back to wired irqs\n");
-		return;
-	}
-
-	for_each_msi_entry(desc, dev) {
-		switch (desc->platform.msi_index) {
-		case EVTQ_MSI_INDEX:
-			smmu->evtq.q.irq = desc->irq;
-			break;
-		case GERROR_MSI_INDEX:
-			smmu->gerr_irq = desc->irq;
-			break;
-		case PRIQ_MSI_INDEX:
-			smmu->priq.q.irq = desc->irq;
-			break;
-		default:	/* Unknown */
-			continue;
-		}
-	}
-
-	/* Add callback to free MSIs on teardown */
-	devm_add_action(dev, arm_smmu_free_msis, dev);
-}
-
 static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
 {
 	int irq, ret;
 
-	arm_smmu_setup_msis(smmu);
-
 	/* Request interrupt lines */
 	irq = smmu->evtq.q.irq;
 	if (irq) {
@@ -3250,8 +3082,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 	if (reg & IDR0_SEV)
 		smmu->features |= ARM_SMMU_FEAT_SEV;
 
-	if (reg & IDR0_MSI)
-		smmu->features |= ARM_SMMU_FEAT_MSI;
 
 	if (reg & IDR0_HYP)
 		smmu->features |= ARM_SMMU_FEAT_HYP;
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 53+ messages in thread

* [PATCH v2 5/8] xen/arm: Remove support for PCI ATS on SMMUv3
  2020-11-26 17:01 [PATCH v2 0/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
                   ` (3 preceding siblings ...)
  2020-11-26 17:02 ` [PATCH v2 4/8] xen/arm: Remove support for MSI on SMMUv3 Rahul Singh
@ 2020-11-26 17:02 ` Rahul Singh
  2020-12-02  0:39   ` Stefano Stabellini
  2020-12-02 13:57   ` Julien Grall
  2020-11-26 17:02 ` [PATCH v2 6/8] xen/arm: Remove support for Stage-1 translation " Rahul Singh
                   ` (2 subsequent siblings)
  7 siblings, 2 replies; 53+ messages in thread
From: Rahul Singh @ 2020-11-26 17:02 UTC (permalink / raw)
  To: xen-devel
  Cc: bertrand.marquis, rahul.singh, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

PCI ATS functionality is not implemented and tested on ARM. Remove the
PCI ATS support, once PCI ATS support is tested and available this
patch can be added.

Signed-off-by: Rahul Singh <rahul.singh@arm.com>
---
 xen/drivers/passthrough/arm/smmu-v3.c | 273 --------------------------
 1 file changed, 273 deletions(-)

diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
index 401f7ae006..6a33628087 100644
--- a/xen/drivers/passthrough/arm/smmu-v3.c
+++ b/xen/drivers/passthrough/arm/smmu-v3.c
@@ -460,16 +460,6 @@ struct arm_smmu_cmdq_ent {
 			u64			addr;
 		} tlbi;
 
-		#define CMDQ_OP_ATC_INV		0x40
-		#define ATC_INV_SIZE_ALL	52
-		struct {
-			u32			sid;
-			u32			ssid;
-			u64			addr;
-			u8			size;
-			bool			global;
-		} atc;
-
 		#define CMDQ_OP_PRI_RESP	0x41
 		struct {
 			u32			sid;
@@ -632,7 +622,6 @@ struct arm_smmu_master {
 	struct list_head		domain_head;
 	u32				*sids;
 	unsigned int			num_sids;
-	bool				ats_enabled;
 	unsigned int			ssid_bits;
 };
 
@@ -650,7 +639,6 @@ struct arm_smmu_domain {
 
 	struct io_pgtable_ops		*pgtbl_ops;
 	bool				non_strict;
-	atomic_t			nr_ats_masters;
 
 	enum arm_smmu_domain_stage	stage;
 	union {
@@ -886,14 +874,6 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
 	case CMDQ_OP_TLBI_S12_VMALL:
 		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
 		break;
-	case CMDQ_OP_ATC_INV:
-		cmd[0] |= FIELD_PREP(CMDQ_0_SSV, ent->substream_valid);
-		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_GLOBAL, ent->atc.global);
-		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_SSID, ent->atc.ssid);
-		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_SID, ent->atc.sid);
-		cmd[1] |= FIELD_PREP(CMDQ_ATC_1_SIZE, ent->atc.size);
-		cmd[1] |= ent->atc.addr & CMDQ_ATC_1_ADDR_MASK;
-		break;
 	case CMDQ_OP_PRI_RESP:
 		cmd[0] |= FIELD_PREP(CMDQ_0_SSV, ent->substream_valid);
 		cmd[0] |= FIELD_PREP(CMDQ_PRI_0_SSID, ent->pri.ssid);
@@ -925,7 +905,6 @@ static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
 		[CMDQ_ERR_CERROR_NONE_IDX]	= "No error",
 		[CMDQ_ERR_CERROR_ILL_IDX]	= "Illegal command",
 		[CMDQ_ERR_CERROR_ABT_IDX]	= "Abort on command fetch",
-		[CMDQ_ERR_CERROR_ATC_INV_IDX]	= "ATC invalidate timeout",
 	};
 
 	int i;
@@ -945,14 +924,6 @@ static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
 		dev_err(smmu->dev, "retrying command fetch\n");
 	case CMDQ_ERR_CERROR_NONE_IDX:
 		return;
-	case CMDQ_ERR_CERROR_ATC_INV_IDX:
-		/*
-		 * ATC Invalidation Completion timeout. CONS is still pointing
-		 * at the CMD_SYNC. Attempt to complete other pending commands
-		 * by repeating the CMD_SYNC, though we might well end up back
-		 * here since the ATC invalidation may still be pending.
-		 */
-		return;
 	case CMDQ_ERR_CERROR_ILL_IDX:
 	default:
 		break;
@@ -1422,9 +1393,6 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
 		val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_S2_TRANS);
 	}
 
-	if (master->ats_enabled)
-		dst[1] |= cpu_to_le64(FIELD_PREP(STRTAB_STE_1_EATS,
-						 STRTAB_STE_1_EATS_TRANS));
 
 	arm_smmu_sync_ste_for_sid(smmu, sid);
 	/* See comment in arm_smmu_write_ctx_desc() */
@@ -1633,112 +1601,6 @@ static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
 	return IRQ_WAKE_THREAD;
 }
 
-static void
-arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size,
-			struct arm_smmu_cmdq_ent *cmd)
-{
-	size_t log2_span;
-	size_t span_mask;
-	/* ATC invalidates are always on 4096-bytes pages */
-	size_t inval_grain_shift = 12;
-	unsigned long page_start, page_end;
-
-	*cmd = (struct arm_smmu_cmdq_ent) {
-		.opcode			= CMDQ_OP_ATC_INV,
-		.substream_valid	= !!ssid,
-		.atc.ssid		= ssid,
-	};
-
-	if (!size) {
-		cmd->atc.size = ATC_INV_SIZE_ALL;
-		return;
-	}
-
-	page_start	= iova >> inval_grain_shift;
-	page_end	= (iova + size - 1) >> inval_grain_shift;
-
-	/*
-	 * In an ATS Invalidate Request, the address must be aligned on the
-	 * range size, which must be a power of two number of page sizes. We
-	 * thus have to choose between grossly over-invalidating the region, or
-	 * splitting the invalidation into multiple commands. For simplicity
-	 * we'll go with the first solution, but should refine it in the future
-	 * if multiple commands are shown to be more efficient.
-	 *
-	 * Find the smallest power of two that covers the range. The most
-	 * significant differing bit between the start and end addresses,
-	 * fls(start ^ end), indicates the required span. For example:
-	 *
-	 * We want to invalidate pages [8; 11]. This is already the ideal range:
-	 *		x = 0b1000 ^ 0b1011 = 0b11
-	 *		span = 1 << fls(x) = 4
-	 *
-	 * To invalidate pages [7; 10], we need to invalidate [0; 15]:
-	 *		x = 0b0111 ^ 0b1010 = 0b1101
-	 *		span = 1 << fls(x) = 16
-	 */
-	log2_span	= fls_long(page_start ^ page_end);
-	span_mask	= (1ULL << log2_span) - 1;
-
-	page_start	&= ~span_mask;
-
-	cmd->atc.addr	= page_start << inval_grain_shift;
-	cmd->atc.size	= log2_span;
-}
-
-static int arm_smmu_atc_inv_master(struct arm_smmu_master *master,
-				   struct arm_smmu_cmdq_ent *cmd)
-{
-	int i;
-
-	if (!master->ats_enabled)
-		return 0;
-
-	for (i = 0; i < master->num_sids; i++) {
-		cmd->atc.sid = master->sids[i];
-		arm_smmu_cmdq_issue_cmd(master->smmu, cmd);
-	}
-
-	return arm_smmu_cmdq_issue_sync(master->smmu);
-}
-
-static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
-				   int ssid, unsigned long iova, size_t size)
-{
-	int ret = 0;
-	unsigned long flags;
-	struct arm_smmu_cmdq_ent cmd;
-	struct arm_smmu_master *master;
-
-	if (!(smmu_domain->smmu->features & ARM_SMMU_FEAT_ATS))
-		return 0;
-
-	/*
-	 * Ensure that we've completed prior invalidation of the main TLBs
-	 * before we read 'nr_ats_masters' in case of a concurrent call to
-	 * arm_smmu_enable_ats():
-	 *
-	 *	// unmap()			// arm_smmu_enable_ats()
-	 *	TLBI+SYNC			atomic_inc(&nr_ats_masters);
-	 *	smp_mb();			[...]
-	 *	atomic_read(&nr_ats_masters);	pci_enable_ats() // writel()
-	 *
-	 * Ensures that we always see the incremented 'nr_ats_masters' count if
-	 * ATS was enabled at the PCI device before completion of the TLBI.
-	 */
-	smp_mb();
-	if (!atomic_read(&smmu_domain->nr_ats_masters))
-		return 0;
-
-	arm_smmu_atc_inv_to_cmd(ssid, iova, size, &cmd);
-
-	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_for_each_entry(master, &smmu_domain->devices, domain_head)
-		ret |= arm_smmu_atc_inv_master(master, &cmd);
-	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
-
-	return ret ? -ETIMEDOUT : 0;
-}
 
 /* IO_PGTABLE API */
 static void arm_smmu_tlb_inv_context(void *cookie)
@@ -1765,7 +1627,6 @@ static void arm_smmu_tlb_inv_context(void *cookie)
 	 */
 	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
 	arm_smmu_cmdq_issue_sync(smmu);
-	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
 }
 
 static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
@@ -2106,108 +1967,6 @@ static void arm_smmu_install_ste_for_dev(struct arm_smmu_master *master)
 	}
 }
 
-static bool arm_smmu_ats_supported(struct arm_smmu_master *master)
-{
-	struct device *dev = master->dev;
-	struct arm_smmu_device *smmu = master->smmu;
-	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
-
-	if (!(smmu->features & ARM_SMMU_FEAT_ATS))
-		return false;
-
-	if (!(fwspec->flags & IOMMU_FWSPEC_PCI_RC_ATS))
-		return false;
-
-	return dev_is_pci(dev) && pci_ats_supported(to_pci_dev(dev));
-}
-
-static void arm_smmu_enable_ats(struct arm_smmu_master *master)
-{
-	size_t stu;
-	struct pci_dev *pdev;
-	struct arm_smmu_device *smmu = master->smmu;
-	struct arm_smmu_domain *smmu_domain = master->domain;
-
-	/* Don't enable ATS at the endpoint if it's not enabled in the STE */
-	if (!master->ats_enabled)
-		return;
-
-	/* Smallest Translation Unit: log2 of the smallest supported granule */
-	stu = __ffs(smmu->pgsize_bitmap);
-	pdev = to_pci_dev(master->dev);
-
-	atomic_inc(&smmu_domain->nr_ats_masters);
-	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
-	if (pci_enable_ats(pdev, stu))
-		dev_err(master->dev, "Failed to enable ATS (STU %zu)\n", stu);
-}
-
-static void arm_smmu_disable_ats(struct arm_smmu_master *master)
-{
-	struct arm_smmu_cmdq_ent cmd;
-	struct arm_smmu_domain *smmu_domain = master->domain;
-
-	if (!master->ats_enabled)
-		return;
-
-	pci_disable_ats(to_pci_dev(master->dev));
-	/*
-	 * Ensure ATS is disabled at the endpoint before we issue the
-	 * ATC invalidation via the SMMU.
-	 */
-	wmb();
-	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
-	arm_smmu_atc_inv_master(master, &cmd);
-    atomic_dec(&smmu_domain->nr_ats_masters);
-}
-
-static int arm_smmu_enable_pasid(struct arm_smmu_master *master)
-{
-	int ret;
-	int features;
-	int num_pasids;
-	struct pci_dev *pdev;
-
-	if (!dev_is_pci(master->dev))
-		return -ENODEV;
-
-	pdev = to_pci_dev(master->dev);
-
-	features = pci_pasid_features(pdev);
-	if (features < 0)
-		return features;
-
-	num_pasids = pci_max_pasids(pdev);
-	if (num_pasids <= 0)
-		return num_pasids;
-
-	ret = pci_enable_pasid(pdev, features);
-	if (ret) {
-		dev_err(&pdev->dev, "Failed to enable PASID\n");
-		return ret;
-	}
-
-	master->ssid_bits = min_t(u8, ilog2(num_pasids),
-				  master->smmu->ssid_bits);
-	return 0;
-}
-
-static void arm_smmu_disable_pasid(struct arm_smmu_master *master)
-{
-	struct pci_dev *pdev;
-
-	if (!dev_is_pci(master->dev))
-		return;
-
-	pdev = to_pci_dev(master->dev);
-
-	if (!pdev->pasid_enabled)
-		return;
-
-	master->ssid_bits = 0;
-	pci_disable_pasid(pdev);
-}
-
 static void arm_smmu_detach_dev(struct arm_smmu_master *master)
 {
 	unsigned long flags;
@@ -2216,14 +1975,11 @@ static void arm_smmu_detach_dev(struct arm_smmu_master *master)
 	if (!smmu_domain)
 		return;
 
-	arm_smmu_disable_ats(master);
-
 	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
 	list_del(&master->domain_head);
 	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
 
 	master->domain = NULL;
-	master->ats_enabled = false;
 	arm_smmu_install_ste_for_dev(master);
 }
 
@@ -2271,17 +2027,12 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 
 	master->domain = smmu_domain;
 
-	if (smmu_domain->stage != ARM_SMMU_DOMAIN_BYPASS)
-		master->ats_enabled = arm_smmu_ats_supported(master);
-
 	arm_smmu_install_ste_for_dev(master);
 
 	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
 	list_add(&master->domain_head, &smmu_domain->devices);
 	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
 
-	arm_smmu_enable_ats(master);
-
 out_unlock:
 	mutex_unlock(&smmu_domain->init_mutex);
 	return ret;
@@ -2410,16 +2161,6 @@ static struct iommu_device *arm_smmu_probe_device(struct device *dev)
 
 	master->ssid_bits = min(smmu->ssid_bits, fwspec->num_pasid_bits);
 
-	/*
-	 * Note that PASID must be enabled before, and disabled after ATS:
-	 * PCI Express Base 4.0r1.0 - 10.5.1.3 ATS Control Register
-	 *
-	 *   Behavior is undefined if this bit is Set and the value of the PASID
-	 *   Enable, Execute Requested Enable, or Privileged Mode Requested bits
-	 *   are changed.
-	 */
-	arm_smmu_enable_pasid(master);
-
 	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB))
 		master->ssid_bits = min_t(u8, master->ssid_bits,
 					  CTXDESC_LINEAR_CDMAX);
@@ -2442,7 +2183,6 @@ static void arm_smmu_release_device(struct device *dev)
 
 	master = dev_iommu_priv_get(dev);
 	arm_smmu_detach_dev(master);
-	arm_smmu_disable_pasid(master);
 	kfree(master);
 	iommu_fwspec_free(dev);
 }
@@ -2997,15 +2737,6 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
 		}
 	}
 
-	if (smmu->features & ARM_SMMU_FEAT_ATS) {
-		enables |= CR0_ATSCHK;
-		ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
-					      ARM_SMMU_CR0ACK);
-		if (ret) {
-			dev_err(smmu->dev, "failed to enable ATS check\n");
-			return ret;
-		}
-	}
 
 	ret = arm_smmu_setup_irqs(smmu);
 	if (ret) {
@@ -3076,13 +2807,9 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 	if (IS_ENABLED(CONFIG_PCI_PRI) && reg & IDR0_PRI)
 		smmu->features |= ARM_SMMU_FEAT_PRI;
 
-	if (IS_ENABLED(CONFIG_PCI_ATS) && reg & IDR0_ATS)
-		smmu->features |= ARM_SMMU_FEAT_ATS;
-
 	if (reg & IDR0_SEV)
 		smmu->features |= ARM_SMMU_FEAT_SEV;
 
-
 	if (reg & IDR0_HYP)
 		smmu->features |= ARM_SMMU_FEAT_HYP;
 
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 53+ messages in thread

* [PATCH v2 6/8] xen/arm: Remove support for Stage-1 translation on SMMUv3.
  2020-11-26 17:01 [PATCH v2 0/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
                   ` (4 preceding siblings ...)
  2020-11-26 17:02 ` [PATCH v2 5/8] xen/arm: Remove support for PCI ATS " Rahul Singh
@ 2020-11-26 17:02 ` Rahul Singh
  2020-12-02  0:53   ` Stefano Stabellini
  2020-11-26 17:02 ` [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN Rahul Singh
  2020-11-26 17:02 ` [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
  7 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-11-26 17:02 UTC (permalink / raw)
  To: xen-devel
  Cc: bertrand.marquis, rahul.singh, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

Linux SMMUv3 driver supports both Stage-1 and Stage-2 translations.
As of now only Stage-2 translation support has been tested.

Once Stage-1 translation support is tested this patch can be added.

Signed-off-by: Rahul Singh <rahul.singh@arm.com>
---
 xen/drivers/passthrough/arm/smmu-v3.c | 440 +-------------------------
 1 file changed, 10 insertions(+), 430 deletions(-)

diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
index 6a33628087..40e3890a58 100644
--- a/xen/drivers/passthrough/arm/smmu-v3.c
+++ b/xen/drivers/passthrough/arm/smmu-v3.c
@@ -432,19 +432,14 @@ struct arm_smmu_cmdq_ent {
 
 		#define CMDQ_OP_CFGI_STE	0x3
 		#define CMDQ_OP_CFGI_ALL	0x4
-		#define CMDQ_OP_CFGI_CD		0x5
-		#define CMDQ_OP_CFGI_CD_ALL	0x6
 		struct {
 			u32			sid;
-			u32			ssid;
 			union {
 				bool		leaf;
 				u8		span;
 			};
 		} cfgi;
 
-		#define CMDQ_OP_TLBI_NH_ASID	0x11
-		#define CMDQ_OP_TLBI_NH_VA	0x12
 		#define CMDQ_OP_TLBI_EL2_ALL	0x20
 		#define CMDQ_OP_TLBI_S12_VMALL	0x28
 		#define CMDQ_OP_TLBI_S2_IPA	0x2a
@@ -514,32 +509,6 @@ struct arm_smmu_strtab_l1_desc {
 	dma_addr_t			l2ptr_dma;
 };
 
-struct arm_smmu_ctx_desc {
-	u16				asid;
-	u64				ttbr;
-	u64				tcr;
-	u64				mair;
-};
-
-struct arm_smmu_l1_ctx_desc {
-	__le64				*l2ptr;
-	dma_addr_t			l2ptr_dma;
-};
-
-struct arm_smmu_ctx_desc_cfg {
-	__le64				*cdtab;
-	dma_addr_t			cdtab_dma;
-	struct arm_smmu_l1_ctx_desc	*l1_desc;
-	unsigned int			num_l1_ents;
-};
-
-struct arm_smmu_s1_cfg {
-	struct arm_smmu_ctx_desc_cfg	cdcfg;
-	struct arm_smmu_ctx_desc	cd;
-	u8				s1fmt;
-	u8				s1cdmax;
-};
-
 struct arm_smmu_s2_cfg {
 	u16				vmid;
 	u64				vttbr;
@@ -590,22 +559,16 @@ struct arm_smmu_device {
 
 	int				gerr_irq;
 	int				combined_irq;
-	u32				sync_nr;
 	u8				prev_cmd_opcode;
 
 	unsigned long			ias; /* IPA */
 	unsigned long			oas; /* PA */
 	unsigned long			pgsize_bitmap;
 
-#define ARM_SMMU_MAX_ASIDS		(1 << 16)
-	unsigned int			asid_bits;
-	DECLARE_BITMAP(asid_map, ARM_SMMU_MAX_ASIDS);
-
 #define ARM_SMMU_MAX_VMIDS		(1 << 16)
 	unsigned int			vmid_bits;
 	DECLARE_BITMAP(vmid_map, ARM_SMMU_MAX_VMIDS);
 
-	unsigned int			ssid_bits;
 	unsigned int			sid_bits;
 
 	struct arm_smmu_strtab_cfg	strtab_cfg;
@@ -622,7 +585,6 @@ struct arm_smmu_master {
 	struct list_head		domain_head;
 	u32				*sids;
 	unsigned int			num_sids;
-	unsigned int			ssid_bits;
 };
 
 /* SMMU private data for an IOMMU domain */
@@ -642,7 +604,6 @@ struct arm_smmu_domain {
 
 	enum arm_smmu_domain_stage	stage;
 	union {
-		struct arm_smmu_s1_cfg	s1_cfg;
 		struct arm_smmu_s2_cfg	s2_cfg;
 	};
 
@@ -835,30 +796,14 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
 		cmd[1] |= FIELD_PREP(CMDQ_PREFETCH_1_SIZE, ent->prefetch.size);
 		cmd[1] |= ent->prefetch.addr & CMDQ_PREFETCH_1_ADDR_MASK;
 		break;
-	case CMDQ_OP_CFGI_CD:
-		cmd[0] |= FIELD_PREP(CMDQ_CFGI_0_SSID, ent->cfgi.ssid);
-		fallthrough;
 	case CMDQ_OP_CFGI_STE:
 		cmd[0] |= FIELD_PREP(CMDQ_CFGI_0_SID, ent->cfgi.sid);
 		cmd[1] |= FIELD_PREP(CMDQ_CFGI_1_LEAF, ent->cfgi.leaf);
 		break;
-	case CMDQ_OP_CFGI_CD_ALL:
-		cmd[0] |= FIELD_PREP(CMDQ_CFGI_0_SID, ent->cfgi.sid);
-		break;
 	case CMDQ_OP_CFGI_ALL:
 		/* Cover the entire SID range */
 		cmd[1] |= FIELD_PREP(CMDQ_CFGI_1_RANGE, 31);
 		break;
-	case CMDQ_OP_TLBI_NH_VA:
-		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_NUM, ent->tlbi.num);
-		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_SCALE, ent->tlbi.scale);
-		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
-		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_ASID, ent->tlbi.asid);
-		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_LEAF, ent->tlbi.leaf);
-		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TTL, ent->tlbi.ttl);
-		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TG, ent->tlbi.tg);
-		cmd[1] |= ent->tlbi.addr & CMDQ_TLBI_1_VA_MASK;
-		break;
 	case CMDQ_OP_TLBI_S2_IPA:
 		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_NUM, ent->tlbi.num);
 		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_SCALE, ent->tlbi.scale);
@@ -868,9 +813,6 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
 		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TG, ent->tlbi.tg);
 		cmd[1] |= ent->tlbi.addr & CMDQ_TLBI_1_IPA_MASK;
 		break;
-	case CMDQ_OP_TLBI_NH_ASID:
-		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_ASID, ent->tlbi.asid);
-		fallthrough;
 	case CMDQ_OP_TLBI_S12_VMALL:
 		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
 		break;
@@ -1005,242 +947,6 @@ static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
 	return ret;
 }
 
-/* Context descriptor manipulation functions */
-static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
-			     int ssid, bool leaf)
-{
-	size_t i;
-	unsigned long flags;
-	struct arm_smmu_master *master;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	struct arm_smmu_cmdq_ent cmd = {
-		.opcode	= CMDQ_OP_CFGI_CD,
-		.cfgi	= {
-			.ssid	= ssid,
-			.leaf	= leaf,
-		},
-	};
-
-	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
-	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
-		for (i = 0; i < master->num_sids; i++) {
-			cmd.cfgi.sid = master->sids[i];
-			arm_smmu_cmdq_issue_cmd(smmu, &cmd);
-		}
-	}
-	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
-
-	arm_smmu_cmdq_issue_sync(smmu);
-}
-
-static int arm_smmu_alloc_cd_leaf_table(struct arm_smmu_device *smmu,
-					struct arm_smmu_l1_ctx_desc *l1_desc)
-{
-	size_t size = CTXDESC_L2_ENTRIES * (CTXDESC_CD_DWORDS << 3);
-
-	l1_desc->l2ptr = dmam_alloc_coherent(smmu->dev, size,
-					     &l1_desc->l2ptr_dma, GFP_KERNEL);
-	if (!l1_desc->l2ptr) {
-		dev_warn(smmu->dev,
-			 "failed to allocate context descriptor table\n");
-		return -ENOMEM;
-	}
-	return 0;
-}
-
-static void arm_smmu_write_cd_l1_desc(__le64 *dst,
-				      struct arm_smmu_l1_ctx_desc *l1_desc)
-{
-	u64 val = (l1_desc->l2ptr_dma & CTXDESC_L1_DESC_L2PTR_MASK) |
-		  CTXDESC_L1_DESC_V;
-
-	/* See comment in arm_smmu_write_ctx_desc() */
-	WRITE_ONCE(*dst, cpu_to_le64(val));
-}
-
-static __le64 *arm_smmu_get_cd_ptr(struct arm_smmu_domain *smmu_domain,
-				   u32 ssid)
-{
-	__le64 *l1ptr;
-	unsigned int idx;
-	struct arm_smmu_l1_ctx_desc *l1_desc;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	struct arm_smmu_ctx_desc_cfg *cdcfg = &smmu_domain->s1_cfg.cdcfg;
-
-	if (smmu_domain->s1_cfg.s1fmt == STRTAB_STE_0_S1FMT_LINEAR)
-		return cdcfg->cdtab + ssid * CTXDESC_CD_DWORDS;
-
-	idx = ssid >> CTXDESC_SPLIT;
-	l1_desc = &cdcfg->l1_desc[idx];
-	if (!l1_desc->l2ptr) {
-		if (arm_smmu_alloc_cd_leaf_table(smmu, l1_desc))
-			return NULL;
-
-		l1ptr = cdcfg->cdtab + idx * CTXDESC_L1_DESC_DWORDS;
-		arm_smmu_write_cd_l1_desc(l1ptr, l1_desc);
-		/* An invalid L1CD can be cached */
-		arm_smmu_sync_cd(smmu_domain, ssid, false);
-	}
-	idx = ssid & (CTXDESC_L2_ENTRIES - 1);
-	return l1_desc->l2ptr + idx * CTXDESC_CD_DWORDS;
-}
-
-static int arm_smmu_write_ctx_desc(struct arm_smmu_domain *smmu_domain,
-				   int ssid, struct arm_smmu_ctx_desc *cd)
-{
-	/*
-	 * This function handles the following cases:
-	 *
-	 * (1) Install primary CD, for normal DMA traffic (SSID = 0).
-	 * (2) Install a secondary CD, for SID+SSID traffic.
-	 * (3) Update ASID of a CD. Atomically write the first 64 bits of the
-	 *     CD, then invalidate the old entry and mappings.
-	 * (4) Remove a secondary CD.
-	 */
-	u64 val;
-	bool cd_live;
-	__le64 *cdptr;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-
-	if (WARN_ON(ssid >= (1 << smmu_domain->s1_cfg.s1cdmax)))
-		return -E2BIG;
-
-	cdptr = arm_smmu_get_cd_ptr(smmu_domain, ssid);
-	if (!cdptr)
-		return -ENOMEM;
-
-	val = le64_to_cpu(cdptr[0]);
-	cd_live = !!(val & CTXDESC_CD_0_V);
-
-	if (!cd) { /* (4) */
-		val = 0;
-	} else if (cd_live) { /* (3) */
-		val &= ~CTXDESC_CD_0_ASID;
-		val |= FIELD_PREP(CTXDESC_CD_0_ASID, cd->asid);
-		/*
-		 * Until CD+TLB invalidation, both ASIDs may be used for tagging
-		 * this substream's traffic
-		 */
-	} else { /* (1) and (2) */
-		cdptr[1] = cpu_to_le64(cd->ttbr & CTXDESC_CD_1_TTB0_MASK);
-		cdptr[2] = 0;
-		cdptr[3] = cpu_to_le64(cd->mair);
-
-		/*
-		 * STE is live, and the SMMU might read dwords of this CD in any
-		 * order. Ensure that it observes valid values before reading
-		 * V=1.
-		 */
-		arm_smmu_sync_cd(smmu_domain, ssid, true);
-
-		val = cd->tcr |
-#ifdef __BIG_ENDIAN
-			CTXDESC_CD_0_ENDI |
-#endif
-			CTXDESC_CD_0_R | CTXDESC_CD_0_A | CTXDESC_CD_0_ASET |
-			CTXDESC_CD_0_AA64 |
-			FIELD_PREP(CTXDESC_CD_0_ASID, cd->asid) |
-			CTXDESC_CD_0_V;
-
-		/* STALL_MODEL==0b10 && CD.S==0 is ILLEGAL */
-		if (smmu->features & ARM_SMMU_FEAT_STALL_FORCE)
-			val |= CTXDESC_CD_0_S;
-	}
-
-	/*
-	 * The SMMU accesses 64-bit values atomically. See IHI0070Ca 3.21.3
-	 * "Configuration structures and configuration invalidation completion"
-	 *
-	 *   The size of single-copy atomic reads made by the SMMU is
-	 *   IMPLEMENTATION DEFINED but must be at least 64 bits. Any single
-	 *   field within an aligned 64-bit span of a structure can be altered
-	 *   without first making the structure invalid.
-	 */
-	WRITE_ONCE(cdptr[0], cpu_to_le64(val));
-	arm_smmu_sync_cd(smmu_domain, ssid, true);
-	return 0;
-}
-
-static int arm_smmu_alloc_cd_tables(struct arm_smmu_domain *smmu_domain)
-{
-	int ret;
-	size_t l1size;
-	size_t max_contexts;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
-	struct arm_smmu_ctx_desc_cfg *cdcfg = &cfg->cdcfg;
-
-	max_contexts = 1 << cfg->s1cdmax;
-
-	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB) ||
-	    max_contexts <= CTXDESC_L2_ENTRIES) {
-		cfg->s1fmt = STRTAB_STE_0_S1FMT_LINEAR;
-		cdcfg->num_l1_ents = max_contexts;
-
-		l1size = max_contexts * (CTXDESC_CD_DWORDS << 3);
-	} else {
-		cfg->s1fmt = STRTAB_STE_0_S1FMT_64K_L2;
-		cdcfg->num_l1_ents = DIV_ROUND_UP(max_contexts,
-						  CTXDESC_L2_ENTRIES);
-
-		cdcfg->l1_desc = devm_kcalloc(smmu->dev, cdcfg->num_l1_ents,
-					      sizeof(*cdcfg->l1_desc),
-					      GFP_KERNEL);
-		if (!cdcfg->l1_desc)
-			return -ENOMEM;
-
-		l1size = cdcfg->num_l1_ents * (CTXDESC_L1_DESC_DWORDS << 3);
-	}
-
-	cdcfg->cdtab = dmam_alloc_coherent(smmu->dev, l1size, &cdcfg->cdtab_dma,
-					   GFP_KERNEL);
-	if (!cdcfg->cdtab) {
-		dev_warn(smmu->dev, "failed to allocate context descriptor\n");
-		ret = -ENOMEM;
-		goto err_free_l1;
-	}
-
-	return 0;
-
-err_free_l1:
-	if (cdcfg->l1_desc) {
-		devm_kfree(smmu->dev, cdcfg->l1_desc);
-		cdcfg->l1_desc = NULL;
-	}
-	return ret;
-}
-
-static void arm_smmu_free_cd_tables(struct arm_smmu_domain *smmu_domain)
-{
-	int i;
-	size_t size, l1size;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	struct arm_smmu_ctx_desc_cfg *cdcfg = &smmu_domain->s1_cfg.cdcfg;
-
-	if (cdcfg->l1_desc) {
-		size = CTXDESC_L2_ENTRIES * (CTXDESC_CD_DWORDS << 3);
-
-		for (i = 0; i < cdcfg->num_l1_ents; i++) {
-			if (!cdcfg->l1_desc[i].l2ptr)
-				continue;
-
-			dmam_free_coherent(smmu->dev, size,
-					   cdcfg->l1_desc[i].l2ptr,
-					   cdcfg->l1_desc[i].l2ptr_dma);
-		}
-		devm_kfree(smmu->dev, cdcfg->l1_desc);
-		cdcfg->l1_desc = NULL;
-
-		l1size = cdcfg->num_l1_ents * (CTXDESC_L1_DESC_DWORDS << 3);
-	} else {
-		l1size = cdcfg->num_l1_ents * (CTXDESC_CD_DWORDS << 3);
-	}
-
-	dmam_free_coherent(smmu->dev, l1size, cdcfg->cdtab, cdcfg->cdtab_dma);
-	cdcfg->cdtab_dma = 0;
-	cdcfg->cdtab = NULL;
-}
-
 /* Stream table manipulation functions */
 static void
 arm_smmu_write_strtab_l1_desc(__le64 *dst, struct arm_smmu_strtab_l1_desc *desc)
@@ -1290,7 +996,6 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
 	u64 val = le64_to_cpu(dst[0]);
 	bool ste_live = false;
 	struct arm_smmu_device *smmu = NULL;
-	struct arm_smmu_s1_cfg *s1_cfg = NULL;
 	struct arm_smmu_s2_cfg *s2_cfg = NULL;
 	struct arm_smmu_domain *smmu_domain = NULL;
 	struct arm_smmu_cmdq_ent prefetch_cmd = {
@@ -1306,24 +1011,13 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
 	}
 
 	if (smmu_domain) {
-		switch (smmu_domain->stage) {
-		case ARM_SMMU_DOMAIN_S1:
-			s1_cfg = &smmu_domain->s1_cfg;
-			break;
-		case ARM_SMMU_DOMAIN_S2:
-		case ARM_SMMU_DOMAIN_NESTED:
-			s2_cfg = &smmu_domain->s2_cfg;
-			break;
-		default:
-			break;
-		}
+		s2_cfg = &smmu_domain->s2_cfg;
 	}
 
 	if (val & STRTAB_STE_0_V) {
 		switch (FIELD_GET(STRTAB_STE_0_CFG, val)) {
 		case STRTAB_STE_0_CFG_BYPASS:
 			break;
-		case STRTAB_STE_0_CFG_S1_TRANS:
 		case STRTAB_STE_0_CFG_S2_TRANS:
 			ste_live = true;
 			break;
@@ -1339,7 +1033,7 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
 	val = STRTAB_STE_0_V;
 
 	/* Bypass/fault */
-	if (!smmu_domain || !(s1_cfg || s2_cfg)) {
+	if (!smmu_domain || !(s2_cfg)) {
 		if (!smmu_domain && disable_bypass)
 			val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_ABORT);
 		else
@@ -1358,25 +1052,6 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
 		return;
 	}
 
-	if (s1_cfg) {
-		BUG_ON(ste_live);
-		dst[1] = cpu_to_le64(
-			 FIELD_PREP(STRTAB_STE_1_S1DSS, STRTAB_STE_1_S1DSS_SSID0) |
-			 FIELD_PREP(STRTAB_STE_1_S1CIR, STRTAB_STE_1_S1C_CACHE_WBRA) |
-			 FIELD_PREP(STRTAB_STE_1_S1COR, STRTAB_STE_1_S1C_CACHE_WBRA) |
-			 FIELD_PREP(STRTAB_STE_1_S1CSH, ARM_SMMU_SH_ISH) |
-			 FIELD_PREP(STRTAB_STE_1_STRW, STRTAB_STE_1_STRW_NSEL1));
-
-		if (smmu->features & ARM_SMMU_FEAT_STALLS &&
-		   !(smmu->features & ARM_SMMU_FEAT_STALL_FORCE))
-			dst[1] |= cpu_to_le64(STRTAB_STE_1_S1STALLD);
-
-		val |= (s1_cfg->cdcfg.cdtab_dma & STRTAB_STE_0_S1CTXPTR_MASK) |
-			FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_S1_TRANS) |
-			FIELD_PREP(STRTAB_STE_0_S1CDMAX, s1_cfg->s1cdmax) |
-			FIELD_PREP(STRTAB_STE_0_S1FMT, s1_cfg->s1fmt);
-	}
-
 	if (s2_cfg) {
 		BUG_ON(ste_live);
 		dst[2] = cpu_to_le64(
@@ -1395,7 +1070,6 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
 
 
 	arm_smmu_sync_ste_for_sid(smmu, sid);
-	/* See comment in arm_smmu_write_ctx_desc() */
 	WRITE_ONCE(dst[0], cpu_to_le64(val));
 	arm_smmu_sync_ste_for_sid(smmu, sid);
 
@@ -1609,14 +1283,8 @@ static void arm_smmu_tlb_inv_context(void *cookie)
 	struct arm_smmu_device *smmu = smmu_domain->smmu;
 	struct arm_smmu_cmdq_ent cmd;
 
-	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
-		cmd.opcode	= CMDQ_OP_TLBI_NH_ASID;
-		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
-		cmd.tlbi.vmid	= 0;
-	} else {
-		cmd.opcode	= CMDQ_OP_TLBI_S12_VMALL;
-		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
-	}
+	cmd.opcode	= CMDQ_OP_TLBI_S12_VMALL;
+	cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
 
 	/*
 	 * NOTE: when io-pgtable is in non-strict mode, we may get here with
@@ -1641,13 +1309,8 @@ static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
 		},
 	};
 
-	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
-		cmd.opcode	= CMDQ_OP_TLBI_NH_VA;
-		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
-	} else {
-		cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
-		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
-	}
+	cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
+	cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
 
 	do {
 		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
@@ -1755,75 +1418,17 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
 {
 	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
 	struct arm_smmu_device *smmu = smmu_domain->smmu;
+	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
 
 	iommu_put_dma_cookie(domain);
 	free_io_pgtable_ops(smmu_domain->pgtbl_ops);
 
-	/* Free the CD and ASID, if we allocated them */
-	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
-		struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
-
-		if (cfg->cdcfg.cdtab) {
-			arm_smmu_free_cd_tables(smmu_domain);
-			arm_smmu_bitmap_free(smmu->asid_map, cfg->cd.asid);
-		}
-	} else {
-		struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
-		if (cfg->vmid)
-			arm_smmu_bitmap_free(smmu->vmid_map, cfg->vmid);
-	}
+	if (cfg->vmid)
+		arm_smmu_bitmap_free(smmu->vmid_map, cfg->vmid);
 
 	kfree(smmu_domain);
 }
 
-static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
-				       struct arm_smmu_master *master,
-				       struct io_pgtable_cfg *pgtbl_cfg)
-{
-	int ret;
-	int asid;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
-	typeof(&pgtbl_cfg->arm_lpae_s1_cfg.tcr) tcr = &pgtbl_cfg->arm_lpae_s1_cfg.tcr;
-
-	asid = arm_smmu_bitmap_alloc(smmu->asid_map, smmu->asid_bits);
-	if (asid < 0)
-		return asid;
-
-	cfg->s1cdmax = master->ssid_bits;
-
-	ret = arm_smmu_alloc_cd_tables(smmu_domain);
-	if (ret)
-		goto out_free_asid;
-
-	cfg->cd.asid	= (u16)asid;
-	cfg->cd.ttbr	= pgtbl_cfg->arm_lpae_s1_cfg.ttbr;
-	cfg->cd.tcr	= FIELD_PREP(CTXDESC_CD_0_TCR_T0SZ, tcr->tsz) |
-			  FIELD_PREP(CTXDESC_CD_0_TCR_TG0, tcr->tg) |
-			  FIELD_PREP(CTXDESC_CD_0_TCR_IRGN0, tcr->irgn) |
-			  FIELD_PREP(CTXDESC_CD_0_TCR_ORGN0, tcr->orgn) |
-			  FIELD_PREP(CTXDESC_CD_0_TCR_SH0, tcr->sh) |
-			  FIELD_PREP(CTXDESC_CD_0_TCR_IPS, tcr->ips) |
-			  CTXDESC_CD_0_TCR_EPD1 | CTXDESC_CD_0_AA64;
-	cfg->cd.mair	= pgtbl_cfg->arm_lpae_s1_cfg.mair;
-
-	/*
-	 * Note that this will end up calling arm_smmu_sync_cd() before
-	 * the master has been added to the devices list for this domain.
-	 * This isn't an issue because the STE hasn't been installed yet.
-	 */
-	ret = arm_smmu_write_ctx_desc(smmu_domain, 0, &cfg->cd);
-	if (ret)
-		goto out_free_cd_tables;
-
-	return 0;
-
-out_free_cd_tables:
-	arm_smmu_free_cd_tables(smmu_domain);
-out_free_asid:
-	arm_smmu_bitmap_free(smmu->asid_map, asid);
-	return ret;
-}
 
 static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
 				       struct arm_smmu_master *master,
@@ -1871,19 +1476,9 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
 	}
 
 	/* Restrict the stage to what we can actually support */
-	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S1))
-		smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
-	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S2))
-		smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
+	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
 
 	switch (smmu_domain->stage) {
-	case ARM_SMMU_DOMAIN_S1:
-		ias = (smmu->features & ARM_SMMU_FEAT_VAX) ? 52 : 48;
-		ias = min_t(unsigned long, ias, VA_BITS);
-		oas = smmu->ias;
-		fmt = ARM_64_LPAE_S1;
-		finalise_stage_fn = arm_smmu_domain_finalise_s1;
-		break;
 	case ARM_SMMU_DOMAIN_NESTED:
 	case ARM_SMMU_DOMAIN_S2:
 		ias = smmu->ias;
@@ -2016,13 +1611,6 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
 			dev_name(smmu->dev));
 		ret = -ENXIO;
 		goto out_unlock;
-	} else if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1 &&
-		   master->ssid_bits != smmu_domain->s1_cfg.s1cdmax) {
-		dev_err(dev,
-			"cannot attach to incompatible domain (%u SSID bits != %u)\n",
-			smmu_domain->s1_cfg.s1cdmax, master->ssid_bits);
-		ret = -EINVAL;
-		goto out_unlock;
 	}
 
 	master->domain = smmu_domain;
@@ -2159,12 +1747,6 @@ static struct iommu_device *arm_smmu_probe_device(struct device *dev)
 		}
 	}
 
-	master->ssid_bits = min(smmu->ssid_bits, fwspec->num_pasid_bits);
-
-	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB))
-		master->ssid_bits = min_t(u8, master->ssid_bits,
-					  CTXDESC_LINEAR_CDMAX);
-
 	return &smmu->iommu;
 
 err_free_master:
@@ -2853,7 +2435,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 	}
 
 	/* ASID/VMID sizes */
-	smmu->asid_bits = reg & IDR0_ASID16 ? 16 : 8;
 	smmu->vmid_bits = reg & IDR0_VMID16 ? 16 : 8;
 
 	/* IDR1 */
@@ -2878,7 +2459,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 					     FIELD_GET(IDR1_PRIQS, reg));
 
 	/* SID/SSID sizes */
-	smmu->ssid_bits = FIELD_GET(IDR1_SSIDSIZE, reg);
 	smmu->sid_bits = FIELD_GET(IDR1_SIDSIZE, reg);
 
 	/*
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 53+ messages in thread

* [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN
  2020-11-26 17:01 [PATCH v2 0/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
                   ` (5 preceding siblings ...)
  2020-11-26 17:02 ` [PATCH v2 6/8] xen/arm: Remove support for Stage-1 translation " Rahul Singh
@ 2020-11-26 17:02 ` Rahul Singh
  2020-12-02  1:48   ` Stefano Stabellini
  2020-12-02 14:45   ` Julien Grall
  2020-11-26 17:02 ` [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
  7 siblings, 2 replies; 53+ messages in thread
From: Rahul Singh @ 2020-11-26 17:02 UTC (permalink / raw)
  To: xen-devel
  Cc: bertrand.marquis, rahul.singh, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

struct io_pgtable_ops, struct io_pgtable_cfg, struct iommu_flush_ops,
and struct iommu_ops related code are linux specific.

Remove code related to above struct as code is dead code in XEN.

Signed-off-by: Rahul Singh <rahul.singh@arm.com>
---
 xen/drivers/passthrough/arm/smmu-v3.c | 457 --------------------------
 1 file changed, 457 deletions(-)

diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
index 40e3890a58..55d1cba194 100644
--- a/xen/drivers/passthrough/arm/smmu-v3.c
+++ b/xen/drivers/passthrough/arm/smmu-v3.c
@@ -402,13 +402,7 @@
 #define ARM_SMMU_CMDQ_SYNC_TIMEOUT_US	1000000 /* 1s! */
 #define ARM_SMMU_CMDQ_SYNC_SPIN_COUNT	10
 
-#define MSI_IOVA_BASE			0x8000000
-#define MSI_IOVA_LENGTH			0x100000
-
 static bool disable_bypass = 1;
-module_param_named(disable_bypass, disable_bypass, bool, S_IRUGO);
-MODULE_PARM_DESC(disable_bypass,
-	"Disable bypass streams such that incoming transactions from devices that are not attached to an iommu domain will report an abort back to the device and will not be allowed to pass through the SMMU.");
 
 enum pri_resp {
 	PRI_RESP_DENY = 0,
@@ -599,7 +593,6 @@ struct arm_smmu_domain {
 	struct arm_smmu_device		*smmu;
 	struct mutex			init_mutex; /* Protects smmu pointer */
 
-	struct io_pgtable_ops		*pgtbl_ops;
 	bool				non_strict;
 
 	enum arm_smmu_domain_stage	stage;
@@ -1297,74 +1290,6 @@ static void arm_smmu_tlb_inv_context(void *cookie)
 	arm_smmu_cmdq_issue_sync(smmu);
 }
 
-static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
-					  size_t granule, bool leaf, void *cookie)
-{
-	struct arm_smmu_domain *smmu_domain = cookie;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-	struct arm_smmu_cmdq_ent cmd = {
-		.tlbi = {
-			.leaf	= leaf,
-			.addr	= iova,
-		},
-	};
-
-	cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
-	cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
-
-	do {
-		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
-		cmd.tlbi.addr += granule;
-	} while (size -= granule);
-}
-
-static void arm_smmu_tlb_inv_page_nosync(struct iommu_iotlb_gather *gather,
-					 unsigned long iova, size_t granule,
-					 void *cookie)
-{
-	arm_smmu_tlb_inv_range_nosync(iova, granule, granule, true, cookie);
-}
-
-static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,
-				  size_t granule, void *cookie)
-{
-	struct arm_smmu_domain *smmu_domain = cookie;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-
-	arm_smmu_tlb_inv_range_nosync(iova, size, granule, false, cookie);
-	arm_smmu_cmdq_issue_sync(smmu);
-}
-
-static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,
-				  size_t granule, void *cookie)
-{
-	struct arm_smmu_domain *smmu_domain = cookie;
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
-
-	arm_smmu_tlb_inv_range_nosync(iova, size, granule, true, cookie);
-	arm_smmu_cmdq_issue_sync(smmu);
-}
-
-static const struct iommu_flush_ops arm_smmu_flush_ops = {
-	.tlb_flush_all	= arm_smmu_tlb_inv_context,
-	.tlb_flush_walk = arm_smmu_tlb_inv_walk,
-	.tlb_flush_leaf = arm_smmu_tlb_inv_leaf,
-	.tlb_add_page	= arm_smmu_tlb_inv_page_nosync,
-};
-
-/* IOMMU API */
-static bool arm_smmu_capable(enum iommu_cap cap)
-{
-	switch (cap) {
-	case IOMMU_CAP_CACHE_COHERENCY:
-		return true;
-	case IOMMU_CAP_NOEXEC:
-		return true;
-	default:
-		return false;
-	}
-}
-
 static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
 {
 	struct arm_smmu_domain *smmu_domain;
@@ -1421,7 +1346,6 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
 	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
 
 	iommu_put_dma_cookie(domain);
-	free_io_pgtable_ops(smmu_domain->pgtbl_ops);
 
 	if (cfg->vmid)
 		arm_smmu_bitmap_free(smmu->vmid_map, cfg->vmid);
@@ -1429,7 +1353,6 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
 	kfree(smmu_domain);
 }
 
-
 static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
 				       struct arm_smmu_master *master,
 				       struct io_pgtable_cfg *pgtbl_cfg)
@@ -1437,7 +1360,6 @@ static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
 	int vmid;
 	struct arm_smmu_device *smmu = smmu_domain->smmu;
 	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
-	typeof(&pgtbl_cfg->arm_lpae_s2_cfg.vtcr) vtcr;
 
 	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
 	if (vmid < 0)
@@ -1461,20 +1383,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
 {
 	int ret;
 	unsigned long ias, oas;
-	enum io_pgtable_fmt fmt;
-	struct io_pgtable_cfg pgtbl_cfg;
-	struct io_pgtable_ops *pgtbl_ops;
 	int (*finalise_stage_fn)(struct arm_smmu_domain *,
 				 struct arm_smmu_master *,
 				 struct io_pgtable_cfg *);
 	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
 	struct arm_smmu_device *smmu = smmu_domain->smmu;
 
-	if (domain->type == IOMMU_DOMAIN_IDENTITY) {
-		smmu_domain->stage = ARM_SMMU_DOMAIN_BYPASS;
-		return 0;
-	}
-
 	/* Restrict the stage to what we can actually support */
 	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
 
@@ -1483,40 +1397,17 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
 	case ARM_SMMU_DOMAIN_S2:
 		ias = smmu->ias;
 		oas = smmu->oas;
-		fmt = ARM_64_LPAE_S2;
 		finalise_stage_fn = arm_smmu_domain_finalise_s2;
 		break;
 	default:
 		return -EINVAL;
 	}
 
-	pgtbl_cfg = (struct io_pgtable_cfg) {
-		.pgsize_bitmap	= smmu->pgsize_bitmap,
-		.ias		= ias,
-		.oas		= oas,
-		.coherent_walk	= smmu->features & ARM_SMMU_FEAT_COHERENCY,
-		.tlb		= &arm_smmu_flush_ops,
-		.iommu_dev	= smmu->dev,
-	};
-
-	if (smmu_domain->non_strict)
-		pgtbl_cfg.quirks |= IO_PGTABLE_QUIRK_NON_STRICT;
-
-	pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
-	if (!pgtbl_ops)
-		return -ENOMEM;
-
-	domain->pgsize_bitmap = pgtbl_cfg.pgsize_bitmap;
-	domain->geometry.aperture_end = (1UL << pgtbl_cfg.ias) - 1;
-	domain->geometry.force_aperture = true;
-
 	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
 	if (ret < 0) {
-		free_io_pgtable_ops(pgtbl_ops);
 		return ret;
 	}
 
-	smmu_domain->pgtbl_ops = pgtbl_ops;
 	return 0;
 }
 
@@ -1626,71 +1517,6 @@ out_unlock:
 	return ret;
 }
 
-static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
-			phys_addr_t paddr, size_t size, int prot, gfp_t gfp)
-{
-	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
-
-	if (!ops)
-		return -ENODEV;
-
-	return ops->map(ops, iova, paddr, size, prot, gfp);
-}
-
-static size_t arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova,
-			     size_t size, struct iommu_iotlb_gather *gather)
-{
-	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
-
-	if (!ops)
-		return 0;
-
-	return ops->unmap(ops, iova, size, gather);
-}
-
-static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain)
-{
-	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-
-	if (smmu_domain->smmu)
-		arm_smmu_tlb_inv_context(smmu_domain);
-}
-
-static void arm_smmu_iotlb_sync(struct iommu_domain *domain,
-				struct iommu_iotlb_gather *gather)
-{
-	struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
-
-	if (smmu)
-		arm_smmu_cmdq_issue_sync(smmu);
-}
-
-static phys_addr_t
-arm_smmu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
-{
-	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
-
-	if (domain->type == IOMMU_DOMAIN_IDENTITY)
-		return iova;
-
-	if (!ops)
-		return 0;
-
-	return ops->iova_to_phys(ops, iova);
-}
-
-static struct platform_driver arm_smmu_driver;
-
-static
-struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle *fwnode)
-{
-	struct device *dev = driver_find_device_by_fwnode(&arm_smmu_driver.driver,
-							  fwnode);
-	put_device(dev);
-	return dev ? dev_get_drvdata(dev) : NULL;
-}
-
 static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
 {
 	unsigned long limit = smmu->strtab_cfg.num_l1_ents;
@@ -1701,206 +1527,6 @@ static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
 	return sid < limit;
 }
 
-static struct iommu_ops arm_smmu_ops;
-
-static struct iommu_device *arm_smmu_probe_device(struct device *dev)
-{
-	int i, ret;
-	struct arm_smmu_device *smmu;
-	struct arm_smmu_master *master;
-	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
-
-	if (!fwspec || fwspec->ops != &arm_smmu_ops)
-		return ERR_PTR(-ENODEV);
-
-	if (WARN_ON_ONCE(dev_iommu_priv_get(dev)))
-		return ERR_PTR(-EBUSY);
-
-	smmu = arm_smmu_get_by_fwnode(fwspec->iommu_fwnode);
-	if (!smmu)
-		return ERR_PTR(-ENODEV);
-
-	master = kzalloc(sizeof(*master), GFP_KERNEL);
-	if (!master)
-		return ERR_PTR(-ENOMEM);
-
-	master->dev = dev;
-	master->smmu = smmu;
-	master->sids = fwspec->ids;
-	master->num_sids = fwspec->num_ids;
-	dev_iommu_priv_set(dev, master);
-
-	/* Check the SIDs are in range of the SMMU and our stream table */
-	for (i = 0; i < master->num_sids; i++) {
-		u32 sid = master->sids[i];
-
-		if (!arm_smmu_sid_in_range(smmu, sid)) {
-			ret = -ERANGE;
-			goto err_free_master;
-		}
-
-		/* Ensure l2 strtab is initialised */
-		if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
-			ret = arm_smmu_init_l2_strtab(smmu, sid);
-			if (ret)
-				goto err_free_master;
-		}
-	}
-
-	return &smmu->iommu;
-
-err_free_master:
-	kfree(master);
-	dev_iommu_priv_set(dev, NULL);
-	return ERR_PTR(ret);
-}
-
-static void arm_smmu_release_device(struct device *dev)
-{
-	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
-	struct arm_smmu_master *master;
-
-	if (!fwspec || fwspec->ops != &arm_smmu_ops)
-		return;
-
-	master = dev_iommu_priv_get(dev);
-	arm_smmu_detach_dev(master);
-	kfree(master);
-	iommu_fwspec_free(dev);
-}
-
-static struct iommu_group *arm_smmu_device_group(struct device *dev)
-{
-	struct iommu_group *group;
-
-	/*
-	 * We don't support devices sharing stream IDs other than PCI RID
-	 * aliases, since the necessary ID-to-device lookup becomes rather
-	 * impractical given a potential sparse 32-bit stream ID space.
-	 */
-	if (dev_is_pci(dev))
-		group = pci_device_group(dev);
-	else
-		group = generic_device_group(dev);
-
-	return group;
-}
-
-static int arm_smmu_domain_get_attr(struct iommu_domain *domain,
-				    enum iommu_attr attr, void *data)
-{
-	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-
-	switch (domain->type) {
-	case IOMMU_DOMAIN_UNMANAGED:
-		switch (attr) {
-		case DOMAIN_ATTR_NESTING:
-			*(int *)data = (smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED);
-			return 0;
-		default:
-			return -ENODEV;
-		}
-		break;
-	case IOMMU_DOMAIN_DMA:
-		switch (attr) {
-		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
-			*(int *)data = smmu_domain->non_strict;
-			return 0;
-		default:
-			return -ENODEV;
-		}
-		break;
-	default:
-		return -EINVAL;
-	}
-}
-
-static int arm_smmu_domain_set_attr(struct iommu_domain *domain,
-				    enum iommu_attr attr, void *data)
-{
-	int ret = 0;
-	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-
-	mutex_lock(&smmu_domain->init_mutex);
-
-	switch (domain->type) {
-	case IOMMU_DOMAIN_UNMANAGED:
-		switch (attr) {
-		case DOMAIN_ATTR_NESTING:
-			if (smmu_domain->smmu) {
-				ret = -EPERM;
-				goto out_unlock;
-			}
-
-			if (*(int *)data)
-				smmu_domain->stage = ARM_SMMU_DOMAIN_NESTED;
-			else
-				smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
-			break;
-		default:
-			ret = -ENODEV;
-		}
-		break;
-	case IOMMU_DOMAIN_DMA:
-		switch(attr) {
-		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
-			smmu_domain->non_strict = *(int *)data;
-			break;
-		default:
-			ret = -ENODEV;
-		}
-		break;
-	default:
-		ret = -EINVAL;
-	}
-
-out_unlock:
-	mutex_unlock(&smmu_domain->init_mutex);
-	return ret;
-}
-
-static int arm_smmu_of_xlate(struct device *dev, struct of_phandle_args *args)
-{
-	return iommu_fwspec_add_ids(dev, args->args, 1);
-}
-
-static void arm_smmu_get_resv_regions(struct device *dev,
-				      struct list_head *head)
-{
-	struct iommu_resv_region *region;
-	int prot = IOMMU_WRITE | IOMMU_NOEXEC | IOMMU_MMIO;
-
-	region = iommu_alloc_resv_region(MSI_IOVA_BASE, MSI_IOVA_LENGTH,
-					 prot, IOMMU_RESV_SW_MSI);
-	if (!region)
-		return;
-
-	list_add_tail(&region->list, head);
-
-	iommu_dma_get_resv_regions(dev, head);
-}
-
-static struct iommu_ops arm_smmu_ops = {
-	.capable		= arm_smmu_capable,
-	.domain_alloc		= arm_smmu_domain_alloc,
-	.domain_free		= arm_smmu_domain_free,
-	.attach_dev		= arm_smmu_attach_dev,
-	.map			= arm_smmu_map,
-	.unmap			= arm_smmu_unmap,
-	.flush_iotlb_all	= arm_smmu_flush_iotlb_all,
-	.iotlb_sync		= arm_smmu_iotlb_sync,
-	.iova_to_phys		= arm_smmu_iova_to_phys,
-	.probe_device		= arm_smmu_probe_device,
-	.release_device		= arm_smmu_release_device,
-	.device_group		= arm_smmu_device_group,
-	.domain_get_attr	= arm_smmu_domain_get_attr,
-	.domain_set_attr	= arm_smmu_domain_set_attr,
-	.of_xlate		= arm_smmu_of_xlate,
-	.get_resv_regions	= arm_smmu_get_resv_regions,
-	.put_resv_regions	= generic_iommu_put_resv_regions,
-	.pgsize_bitmap		= -1UL, /* Restricted during device attach */
-};
-
 /* Probing and initialisation functions */
 static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
 				   struct arm_smmu_queue *q,
@@ -2406,7 +2032,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 	switch (FIELD_GET(IDR0_STALL_MODEL, reg)) {
 	case IDR0_STALL_MODEL_FORCE:
 		smmu->features |= ARM_SMMU_FEAT_STALL_FORCE;
-		fallthrough;
 	case IDR0_STALL_MODEL_STALL:
 		smmu->features |= ARM_SMMU_FEAT_STALLS;
 	}
@@ -2426,7 +2051,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 	switch (FIELD_GET(IDR0_TTF, reg)) {
 	case IDR0_TTF_AARCH32_64:
 		smmu->ias = 40;
-		fallthrough;
 	case IDR0_TTF_AARCH64:
 		break;
 	default:
@@ -2515,21 +2139,10 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
 	default:
 		dev_info(smmu->dev,
 			"unknown output address size. Truncating to 48-bit\n");
-		fallthrough;
 	case IDR5_OAS_48_BIT:
 		smmu->oas = 48;
 	}
 
-	if (arm_smmu_ops.pgsize_bitmap == -1UL)
-		arm_smmu_ops.pgsize_bitmap = smmu->pgsize_bitmap;
-	else
-		arm_smmu_ops.pgsize_bitmap |= smmu->pgsize_bitmap;
-
-	/* Set the DMA mask for our table walker */
-	if (dma_set_mask_and_coherent(smmu->dev, DMA_BIT_MASK(smmu->oas)))
-		dev_warn(smmu->dev,
-			 "failed to set DMA mask for table walker\n");
-
 	smmu->ias = max(smmu->ias, smmu->oas);
 
 	dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n",
@@ -2595,9 +2208,6 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev,
 
 	parse_driver_options(smmu);
 
-	if (of_dma_is_coherent(dev->of_node))
-		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
-
 	return ret;
 }
 
@@ -2609,55 +2219,6 @@ 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)
-{
-	struct resource res = {
-		.flags = IORESOURCE_MEM,
-		.start = start,
-		.end = start + size - 1,
-	};
-
-	return devm_ioremap_resource(dev, &res);
-}
-
 static int arm_smmu_device_probe(struct platform_device *pdev)
 {
 	int irq, ret;
@@ -2785,21 +2346,3 @@ static const struct of_device_id arm_smmu_of_match[] = {
 	{ .compatible = "arm,smmu-v3", },
 	{ },
 };
-MODULE_DEVICE_TABLE(of, arm_smmu_of_match);
-
-static struct platform_driver arm_smmu_driver = {
-	.driver	= {
-		.name			= "arm-smmu-v3",
-		.of_match_table		= arm_smmu_of_match,
-		.suppress_bind_attrs	= true,
-	},
-	.probe	= arm_smmu_device_probe,
-	.remove	= arm_smmu_device_remove,
-	.shutdown = arm_smmu_device_shutdown,
-};
-module_platform_driver(arm_smmu_driver);
-
-MODULE_DESCRIPTION("IOMMU API for ARM architected SMMUv3 implementations");
-MODULE_AUTHOR("Will Deacon <will@kernel.org>");
-MODULE_ALIAS("platform:arm-smmu-v3");
-MODULE_LICENSE("GPL v2");
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 53+ messages in thread

* [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-11-26 17:01 [PATCH v2 0/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
                   ` (6 preceding siblings ...)
  2020-11-26 17:02 ` [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN Rahul Singh
@ 2020-11-26 17:02 ` Rahul Singh
  2020-12-02  2:51   ` Stefano Stabellini
  2020-12-02 16:22   ` Julien Grall
  7 siblings, 2 replies; 53+ messages in thread
From: Rahul Singh @ 2020-11-26 17:02 UTC (permalink / raw)
  To: xen-devel
  Cc: bertrand.marquis, rahul.singh, Andrew Cooper, George Dunlap,
	Ian Jackson, Jan Beulich, Julien Grall, Stefano Stabellini,
	Wei Liu, Paul Durrant, Volodymyr Babchuk

Add support for ARM architected SMMUv3 implementation. It is based on
the Linux SMMUv3 driver.

Major differences with regard to Linux driver are as follows:
1. Only Stage-2 translation is supported as compared to the Linux driver
   that supports both Stage-1 and Stage-2 translations.
2. Use P2M  page table instead of creating one as SMMUv3 has the
   capability to share the page tables with the CPU.
3. Tasklets are used in place of threaded IRQ's in Linux for event queue
   and priority queue IRQ handling.
4. Latest version of the Linux SMMUv3 code implements the commands queue
   access functions based on atomic operations implemented in Linux.
   Atomic functions used by the commands queue access functions are not
   implemented in XEN therefore we decided to port the earlier version
   of the code. Once the proper atomic operations will be available in
   XEN the driver can be updated.
5. Driver is currently supported as Tech Preview.

Signed-off-by: Rahul Singh <rahul.singh@arm.com>
---
 MAINTAINERS                           |   6 +
 SUPPORT.md                            |   1 +
 xen/drivers/passthrough/Kconfig       |  10 +
 xen/drivers/passthrough/arm/Makefile  |   1 +
 xen/drivers/passthrough/arm/smmu-v3.c | 986 +++++++++++++++++++++-----
 5 files changed, 814 insertions(+), 190 deletions(-)

diff --git a/MAINTAINERS b/MAINTAINERS
index dab38a6a14..1d63489eec 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -249,6 +249,12 @@ F:	xen/include/asm-arm/
 F:	xen/include/public/arch-arm/
 F:	xen/include/public/arch-arm.h
 
+ARM SMMUv3
+M:	Bertrand Marquis <bertrand.marquis@arm.com>
+M:	Rahul Singh <rahul.singh@arm.com>
+S:	Supported
+F:	xen/drivers/passthrough/arm/smmu-v3.c
+
 Change Log
 M:	Paul Durrant <paul@xen.org>
 R:	Community Manager <community.manager@xenproject.org>
diff --git a/SUPPORT.md b/SUPPORT.md
index ab02aca5f4..e402c7202d 100644
--- a/SUPPORT.md
+++ b/SUPPORT.md
@@ -68,6 +68,7 @@ For the Cortex A57 r0p0 - r1p1, see Errata 832075.
     Status, ARM SMMUv1: Supported, not security supported
     Status, ARM SMMUv2: Supported, not security supported
     Status, Renesas IPMMU-VMSA: Supported, not security supported
+    Status, ARM SMMUv3: Tech Preview
 
 ### ARM/GICv3 ITS
 
diff --git a/xen/drivers/passthrough/Kconfig b/xen/drivers/passthrough/Kconfig
index 0036007ec4..5b71c59f47 100644
--- a/xen/drivers/passthrough/Kconfig
+++ b/xen/drivers/passthrough/Kconfig
@@ -13,6 +13,16 @@ config ARM_SMMU
 	  Say Y here if your SoC includes an IOMMU device implementing the
 	  ARM SMMU architecture.
 
+config ARM_SMMU_V3
+	bool "ARM Ltd. System MMU Version 3 (SMMUv3) Support" if EXPERT
+	depends on ARM_64
+	---help---
+	 Support for implementations of the ARM System MMU architecture
+	 version 3.
+
+	 Say Y here if your system includes an IOMMU device implementing
+	 the ARM SMMUv3 architecture.
+
 config IPMMU_VMSA
 	bool "Renesas IPMMU-VMSA found in R-Car Gen3 SoCs"
 	depends on ARM_64
diff --git a/xen/drivers/passthrough/arm/Makefile b/xen/drivers/passthrough/arm/Makefile
index fcd918ea3e..c5fb3b58a5 100644
--- a/xen/drivers/passthrough/arm/Makefile
+++ b/xen/drivers/passthrough/arm/Makefile
@@ -1,3 +1,4 @@
 obj-y += iommu.o iommu_helpers.o iommu_fwspec.o
 obj-$(CONFIG_ARM_SMMU) += smmu.o
 obj-$(CONFIG_IPMMU_VMSA) += ipmmu-vmsa.o
+obj-$(CONFIG_ARM_SMMU_V3) += smmu-v3.o
diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
index 55d1cba194..8f2337e7f2 100644
--- a/xen/drivers/passthrough/arm/smmu-v3.c
+++ b/xen/drivers/passthrough/arm/smmu-v3.c
@@ -2,36 +2,280 @@
 /*
  * IOMMU API for ARM architected SMMUv3 implementations.
  *
- * Copyright (C) 2015 ARM Limited
+ * Based on Linux's SMMUv3 driver:
+ *    drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
+ *    commit: 951cbbc386ff01b50da4f46387e994e81d9ab431
+ * and Xen's SMMU driver:
+ *    xen/drivers/passthrough/arm/smmu.c
  *
- * Author: Will Deacon <will.deacon@arm.com>
+ * Copyright (C) 2015 ARM Limited Will Deacon <will.deacon@arm.com>
  *
- * This driver is powered by bad coffee and bombay mix.
+ * Copyright (C) 2020 Arm Ltd.
+ *
+ * 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, see <http://www.gnu.org/licenses/>.
+ *
+ */
+
+#include <xen/acpi.h>
+#include <xen/config.h>
+#include <xen/delay.h>
+#include <xen/errno.h>
+#include <xen/err.h>
+#include <xen/irq.h>
+#include <xen/lib.h>
+#include <xen/list.h>
+#include <xen/mm.h>
+#include <xen/rbtree.h>
+#include <xen/sched.h>
+#include <xen/sizes.h>
+#include <xen/vmap.h>
+#include <asm/atomic.h>
+#include <asm/device.h>
+#include <asm/io.h>
+#include <asm/platform.h>
+#include <asm/iommu_fwspec.h>
+
+/* Linux compatibility functions. */
+typedef paddr_t dma_addr_t;
+typedef unsigned int gfp_t;
+
+#define platform_device device
+
+#define GFP_KERNEL 0
+
+/* Alias to Xen device tree helpers */
+#define device_node dt_device_node
+#define of_phandle_args dt_phandle_args
+#define of_device_id dt_device_match
+#define of_match_node dt_match_node
+#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, pname, out))
+#define of_property_read_bool dt_property_read_bool
+#define of_parse_phandle_with_args dt_parse_phandle_with_args
+
+/* Alias to Xen lock functions */
+#define mutex spinlock
+#define mutex_init spin_lock_init
+#define mutex_lock spin_lock
+#define mutex_unlock spin_unlock
+
+/* Alias to Xen time functions */
+#define ktime_t s_time_t
+#define ktime_get()             (NOW())
+#define ktime_add_us(t,i)       (t + MICROSECS(i))
+#define ktime_compare(t,i)      (t > (i))
+
+/* Alias to Xen allocation helpers */
+#define kzalloc(size, flags)    _xzalloc(size, sizeof(void *))
+#define kfree xfree
+#define devm_kzalloc(dev, size, flags)  _xzalloc(size, sizeof(void *))
+
+/* Device logger functions */
+#define dev_name(dev) dt_node_full_name(dev->of_node)
+#define dev_dbg(dev, fmt, ...)      \
+    printk(XENLOG_DEBUG "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
+#define dev_notice(dev, fmt, ...)   \
+    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
+#define dev_warn(dev, fmt, ...)     \
+    printk(XENLOG_WARNING "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
+#define dev_err(dev, fmt, ...)      \
+    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
+#define dev_info(dev, fmt, ...)     \
+    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
+#define dev_err_ratelimited(dev, fmt, ...)      \
+    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
+
+/*
+ * Periodically poll an address and wait between reads in us until a
+ * condition is met or a timeout occurs.
+ */
+#define readx_poll_timeout(op, addr, val, cond, sleep_us, timeout_us) \
+({ \
+     s_time_t deadline = NOW() + MICROSECS(timeout_us); \
+     for (;;) { \
+        (val) = op(addr); \
+        if (cond) \
+            break; \
+        if (NOW() > deadline) { \
+            (val) = op(addr); \
+            break; \
+        } \
+        udelay(sleep_us); \
+     } \
+     (cond) ? 0 : -ETIMEDOUT; \
+})
+
+#define readl_relaxed_poll_timeout(addr, val, cond, delay_us, timeout_us) \
+    readx_poll_timeout(readl_relaxed, addr, val, cond, delay_us, timeout_us)
+
+#define FIELD_PREP(_mask, _val)         \
+    (((typeof(_mask))(_val) << (__builtin_ffsll(_mask) - 1)) & (_mask))
+
+#define FIELD_GET(_mask, _reg)          \
+    (typeof(_mask))(((_reg) & (_mask)) >> (__builtin_ffsll(_mask) - 1))
+
+#define WRITE_ONCE(x, val)                  \
+do {                                        \
+    *(volatile typeof(x) *)&(x) = (val);    \
+} while (0)
+
+/* Xen: Stub out DMA domain related functions */
+#define iommu_get_dma_cookie(dom) 0
+#define iommu_put_dma_cookie(dom)
+
+/*
+ * Helpers for DMA allocation. Just the function name is reused for
+ * porting code, these allocation are not managed allocations
  */
+static void *dmam_alloc_coherent(struct device *dev, size_t size,
+                                 paddr_t *dma_handle, gfp_t gfp)
+{
+    void *vaddr;
+    unsigned long alignment = size;
+
+    /*
+     * _xzalloc requires that the (align & (align -1)) = 0. Most of the
+     * allocations in SMMU code should send the right value for size. In
+     * case this is not true print a warning and align to the size of a
+     * (void *)
+     */
+    if ( size & (size - 1) )
+    {
+        printk(XENLOG_WARNING "SMMUv3: Fixing alignment for the DMA buffer\n");
+        alignment = sizeof(void *);
+    }
+
+    vaddr = _xzalloc(size, alignment);
+    if ( !vaddr )
+    {
+        printk(XENLOG_ERR "SMMUv3: DMA allocation failed\n");
+        return NULL;
+    }
+
+    *dma_handle = virt_to_maddr(vaddr);
+
+    return vaddr;
+}
+
+/* Xen: Type definitions for iommu_domain */
+#define IOMMU_DOMAIN_UNMANAGED 0
+#define IOMMU_DOMAIN_DMA 1
+#define IOMMU_DOMAIN_IDENTITY 2
+
+/* Xen specific code. */
+struct iommu_domain {
+    /* Runtime SMMU configuration for this iommu_domain */
+    atomic_t ref;
+    /*
+     * Used to link iommu_domain contexts for a same domain.
+     * There is at least one per-SMMU to used by the domain.
+     */
+    struct list_head    list;
+};
+
+/* Describes information required for a Xen domain */
+struct arm_smmu_xen_domain {
+    spinlock_t      lock;
+
+    /* List of iommu domains associated to this domain */
+    struct list_head    contexts;
+};
+
+/*
+ * Information about each device stored in dev->archdata.iommu
+ * The dev->archdata.iommu stores the iommu_domain (runtime configuration of
+ * the SMMU).
+ */
+struct arm_smmu_xen_device {
+    struct iommu_domain *domain;
+};
+
+/* Keep a list of devices associated with this driver */
+static DEFINE_SPINLOCK(arm_smmu_devices_lock);
+static LIST_HEAD(arm_smmu_devices);
+
+
+static inline void *dev_iommu_priv_get(struct device *dev)
+{
+    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
+
+    return fwspec && fwspec->iommu_priv ? fwspec->iommu_priv : NULL;
+}
+
+static inline void dev_iommu_priv_set(struct device *dev, void *priv)
+{
+    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
+
+    fwspec->iommu_priv = priv;
+}
+
+int dt_property_match_string(const struct dt_device_node *np,
+                             const char *propname, const char *string)
+{
+    const struct dt_property *dtprop = dt_find_property(np, propname, NULL);
+    size_t l;
+    int i;
+    const char *p, *end;
+
+    if ( !dtprop )
+        return -EINVAL;
+
+    if ( !dtprop->value )
+        return -ENODATA;
+
+    p = dtprop->value;
+    end = p + dtprop->length;
+
+    for ( i = 0; p < end; i++, p += l )
+    {
+        l = strnlen(p, end - p) + 1;
+
+        if ( p + l > end )
+            return -EILSEQ;
+
+        if ( strcmp(string, p) == 0 )
+            return i; /* Found it; return index */
+    }
+
+    return -ENODATA;
+}
+
+static int platform_get_irq_byname_optional(struct device *dev,
+                                            const char *name)
+{
+    int index, ret;
+    struct dt_device_node *np  = dev_to_dt(dev);
+
+    if ( unlikely(!name) )
+        return -EINVAL;
+
+    index = dt_property_match_string(np, "interrupt-names", name);
+    if ( index < 0 )
+    {
+        dev_info(dev, "IRQ %s not found\n", name);
+        return index;
+    }
 
-#include <linux/acpi.h>
-#include <linux/acpi_iort.h>
-#include <linux/bitfield.h>
-#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>
-#include <linux/iommu.h>
-#include <linux/iopoll.h>
-#include <linux/module.h>
-#include <linux/msi.h>
-#include <linux/of.h>
-#include <linux/of_address.h>
-#include <linux/of_iommu.h>
-#include <linux/of_platform.h>
-#include <linux/pci.h>
-#include <linux/pci-ats.h>
-#include <linux/platform_device.h>
-
-#include <linux/amba/bus.h>
+    ret = platform_get_irq(np, index);
+    if ( ret < 0 )
+    {
+        dev_err(dev, "failed to get irq index %d\n", index);
+        return -ENODEV;
+    }
+
+    return ret;
+}
+
+/* Start of Linux SMMUv3 code */
 
 /* MMIO registers */
 #define ARM_SMMU_IDR0			0x0
@@ -507,6 +751,7 @@ struct arm_smmu_s2_cfg {
 	u16				vmid;
 	u64				vttbr;
 	u64				vtcr;
+	struct domain		*domain;
 };
 
 struct arm_smmu_strtab_cfg {
@@ -567,8 +812,13 @@ struct arm_smmu_device {
 
 	struct arm_smmu_strtab_cfg	strtab_cfg;
 
-	/* IOMMU core code handle */
-	struct iommu_device		iommu;
+	/* Need to keep a list of SMMU devices */
+	struct list_head		devices;
+
+	/* Tasklets for handling evts/faults and pci page request IRQs*/
+	struct tasklet		evtq_irq_tasklet;
+	struct tasklet		priq_irq_tasklet;
+	struct tasklet		combined_irq_tasklet;
 };
 
 /* SMMU private data for each master */
@@ -1110,7 +1360,7 @@ static int arm_smmu_init_l2_strtab(struct arm_smmu_device *smmu, u32 sid)
 }
 
 /* IRQ and event handlers */
-static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
+static void arm_smmu_evtq_thread(void *dev)
 {
 	int i;
 	struct arm_smmu_device *smmu = dev;
@@ -1140,7 +1390,6 @@ static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
 	/* Sync our overflow flag, as we believe we're up to speed */
 	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
 		    Q_IDX(llq, llq->cons);
-	return IRQ_HANDLED;
 }
 
 static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
@@ -1181,7 +1430,7 @@ static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
 	}
 }
 
-static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
+static void arm_smmu_priq_thread(void *dev)
 {
 	struct arm_smmu_device *smmu = dev;
 	struct arm_smmu_queue *q = &smmu->priq.q;
@@ -1200,12 +1449,12 @@ static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
 	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
 		      Q_IDX(llq, llq->cons);
 	queue_sync_cons_out(q);
-	return IRQ_HANDLED;
 }
 
 static int arm_smmu_device_disable(struct arm_smmu_device *smmu);
 
-static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
+static void arm_smmu_gerror_handler(int irq, void *dev,
+				struct cpu_user_regs *regs)
 {
 	u32 gerror, gerrorn, active;
 	struct arm_smmu_device *smmu = dev;
@@ -1215,7 +1464,7 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
 
 	active = gerror ^ gerrorn;
 	if (!(active & GERROR_ERR_MASK))
-		return IRQ_NONE; /* No errors pending */
+		return; /* No errors pending */
 
 	dev_warn(smmu->dev,
 		 "unexpected global error reported (0x%08x), this could be serious\n",
@@ -1248,26 +1497,42 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
 		arm_smmu_cmdq_skip_err(smmu);
 
 	writel(gerror, smmu->base + ARM_SMMU_GERRORN);
-	return IRQ_HANDLED;
 }
 
-static irqreturn_t arm_smmu_combined_irq_thread(int irq, void *dev)
+static void arm_smmu_combined_irq_handler(int irq, void *dev,
+				struct cpu_user_regs *regs)
+{
+	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
+
+	arm_smmu_gerror_handler(irq, dev, regs);
+
+	tasklet_schedule(&(smmu->combined_irq_tasklet));
+}
+
+static void arm_smmu_combined_irq_thread(void *dev)
 {
 	struct arm_smmu_device *smmu = dev;
 
-	arm_smmu_evtq_thread(irq, dev);
+	arm_smmu_evtq_thread(dev);
 	if (smmu->features & ARM_SMMU_FEAT_PRI)
-		arm_smmu_priq_thread(irq, dev);
-
-	return IRQ_HANDLED;
+		arm_smmu_priq_thread(dev);
 }
 
-static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
+static void arm_smmu_evtq_irq_tasklet(int irq, void *dev,
+				struct cpu_user_regs *regs)
 {
-	arm_smmu_gerror_handler(irq, dev);
-	return IRQ_WAKE_THREAD;
+	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
+
+	tasklet_schedule(&(smmu->evtq_irq_tasklet));
 }
 
+static void arm_smmu_priq_irq_tasklet(int irq, void *dev,
+				struct cpu_user_regs *regs)
+{
+	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
+
+	tasklet_schedule(&(smmu->priq_irq_tasklet));
+}
 
 /* IO_PGTABLE API */
 static void arm_smmu_tlb_inv_context(void *cookie)
@@ -1354,27 +1619,69 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
 }
 
 static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
-				       struct arm_smmu_master *master,
-				       struct io_pgtable_cfg *pgtbl_cfg)
+				       struct arm_smmu_master *master)
 {
 	int vmid;
+	u64 reg;
 	struct arm_smmu_device *smmu = smmu_domain->smmu;
 	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
 
+	/* VTCR */
+	reg = VTCR_RES1 | VTCR_SH0_IS | VTCR_IRGN0_WBWA | VTCR_ORGN0_WBWA;
+
+	switch (PAGE_SIZE) {
+	case SZ_4K:
+		reg |= VTCR_TG0_4K;
+		break;
+	case SZ_16K:
+		reg |= VTCR_TG0_16K;
+		break;
+	case SZ_64K:
+		reg |= VTCR_TG0_4K;
+		break;
+	}
+
+	switch (smmu->oas) {
+	case 32:
+		reg |= VTCR_PS(_AC(0x0,ULL));
+		break;
+	case 36:
+		reg |= VTCR_PS(_AC(0x1,ULL));
+		break;
+	case 40:
+		reg |= VTCR_PS(_AC(0x2,ULL));
+		break;
+	case 42:
+		reg |= VTCR_PS(_AC(0x3,ULL));
+		break;
+	case 44:
+		reg |= VTCR_PS(_AC(0x4,ULL));
+		break;
+		case 48:
+		reg |= VTCR_PS(_AC(0x5,ULL));
+		break;
+	case 52:
+		reg |= VTCR_PS(_AC(0x6,ULL));
+		break;
+	}
+
+	reg |= VTCR_T0SZ(64ULL - smmu->ias);
+	reg |= VTCR_SL0(0x2);
+	reg |= VTCR_VS;
+
+	cfg->vtcr   = reg;
+
 	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
 	if (vmid < 0)
 		return vmid;
+	cfg->vmid  = (u16)vmid;
+
+	cfg->vttbr  = page_to_maddr(cfg->domain->arch.p2m.root);
+
+	printk(XENLOG_DEBUG
+		   "SMMUv3: d%u: vmid 0x%x vtcr 0x%"PRIpaddr" p2maddr 0x%"PRIpaddr"\n",
+		   cfg->domain->domain_id, cfg->vmid, cfg->vtcr, cfg->vttbr);
 
-	vtcr = &pgtbl_cfg->arm_lpae_s2_cfg.vtcr;
-	cfg->vmid	= (u16)vmid;
-	cfg->vttbr	= pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
-	cfg->vtcr	= FIELD_PREP(STRTAB_STE_2_VTCR_S2T0SZ, vtcr->tsz) |
-			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SL0, vtcr->sl) |
-			  FIELD_PREP(STRTAB_STE_2_VTCR_S2IR0, vtcr->irgn) |
-			  FIELD_PREP(STRTAB_STE_2_VTCR_S2OR0, vtcr->orgn) |
-			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SH0, vtcr->sh) |
-			  FIELD_PREP(STRTAB_STE_2_VTCR_S2TG, vtcr->tg) |
-			  FIELD_PREP(STRTAB_STE_2_VTCR_S2PS, vtcr->ps);
 	return 0;
 }
 
@@ -1382,28 +1689,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
 				    struct arm_smmu_master *master)
 {
 	int ret;
-	unsigned long ias, oas;
-	int (*finalise_stage_fn)(struct arm_smmu_domain *,
-				 struct arm_smmu_master *,
-				 struct io_pgtable_cfg *);
 	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
-	struct arm_smmu_device *smmu = smmu_domain->smmu;
 
 	/* Restrict the stage to what we can actually support */
 	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
 
-	switch (smmu_domain->stage) {
-	case ARM_SMMU_DOMAIN_NESTED:
-	case ARM_SMMU_DOMAIN_S2:
-		ias = smmu->ias;
-		oas = smmu->oas;
-		finalise_stage_fn = arm_smmu_domain_finalise_s2;
-		break;
-	default:
-		return -EINVAL;
-	}
-
-	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
+	ret = arm_smmu_domain_finalise_s2(smmu_domain, master);
 	if (ret < 0) {
 		return ret;
 	}
@@ -1553,7 +1844,8 @@ static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
 		return -ENOMEM;
 	}
 
-	if (!WARN_ON(q->base_dma & (qsz - 1))) {
+	WARN_ON(q->base_dma & (qsz - 1));
+	if (unlikely(q->base_dma & (qsz - 1))) {
 		dev_info(smmu->dev, "allocated %u entries for %s\n",
 			 1 << q->llq.max_n_shift, name);
 	}
@@ -1758,9 +2050,7 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
 	/* Request interrupt lines */
 	irq = smmu->evtq.q.irq;
 	if (irq) {
-		ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
-						arm_smmu_evtq_thread,
-						IRQF_ONESHOT,
+		ret = request_irq(irq, 0, arm_smmu_evtq_irq_tasklet,
 						"arm-smmu-v3-evtq", smmu);
 		if (ret < 0)
 			dev_warn(smmu->dev, "failed to enable evtq irq\n");
@@ -1770,8 +2060,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
 
 	irq = smmu->gerr_irq;
 	if (irq) {
-		ret = devm_request_irq(smmu->dev, irq, arm_smmu_gerror_handler,
-				       0, "arm-smmu-v3-gerror", smmu);
+		ret = request_irq(irq, 0, arm_smmu_gerror_handler,
+						"arm-smmu-v3-gerror", smmu);
 		if (ret < 0)
 			dev_warn(smmu->dev, "failed to enable gerror irq\n");
 	} else {
@@ -1781,11 +2071,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
 	if (smmu->features & ARM_SMMU_FEAT_PRI) {
 		irq = smmu->priq.q.irq;
 		if (irq) {
-			ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
-							arm_smmu_priq_thread,
-							IRQF_ONESHOT,
-							"arm-smmu-v3-priq",
-							smmu);
+			ret = request_irq(irq, 0, arm_smmu_priq_irq_tasklet,
+							"arm-smmu-v3-priq", smmu);
 			if (ret < 0)
 				dev_warn(smmu->dev,
 					 "failed to enable priq irq\n");
@@ -1814,11 +2101,8 @@ static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
 		 * Cavium ThunderX2 implementation doesn't support unique irq
 		 * lines. Use a single irq line for all the SMMUv3 interrupts.
 		 */
-		ret = devm_request_threaded_irq(smmu->dev, irq,
-					arm_smmu_combined_irq_handler,
-					arm_smmu_combined_irq_thread,
-					IRQF_ONESHOT,
-					"arm-smmu-v3-combined-irq", smmu);
+		ret = request_irq(irq, 0, arm_smmu_combined_irq_handler,
+						"arm-smmu-v3-combined-irq", smmu);
 		if (ret < 0)
 			dev_warn(smmu->dev, "failed to enable combined irq\n");
 	} else
@@ -1857,7 +2141,7 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
 	reg = readl_relaxed(smmu->base + ARM_SMMU_CR0);
 	if (reg & CR0_SMMUEN) {
 		dev_warn(smmu->dev, "SMMU currently enabled! Resetting...\n");
-		WARN_ON(is_kdump_kernel() && !disable_bypass);
+		WARN_ON(!disable_bypass);
 		arm_smmu_update_gbpa(smmu, GBPA_ABORT, 0);
 	}
 
@@ -1952,8 +2236,11 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
 		return ret;
 	}
 
-	if (is_kdump_kernel())
-		enables &= ~(CR0_EVTQEN | CR0_PRIQEN);
+	/* Initialize tasklets for threaded IRQs*/
+	tasklet_init(&smmu->evtq_irq_tasklet, arm_smmu_evtq_thread, smmu);
+	tasklet_init(&smmu->priq_irq_tasklet, arm_smmu_priq_thread, smmu);
+	tasklet_init(&smmu->combined_irq_tasklet, arm_smmu_combined_irq_thread,
+				 smmu);
 
 	/* Enable the SMMU interface, or ensure bypass */
 	if (!bypass || disable_bypass) {
@@ -2195,7 +2482,7 @@ static inline int arm_smmu_device_acpi_probe(struct platform_device *pdev,
 static int arm_smmu_device_dt_probe(struct platform_device *pdev,
 				    struct arm_smmu_device *smmu)
 {
-	struct device *dev = &pdev->dev;
+	struct device *dev = pdev;
 	u32 cells;
 	int ret = -EINVAL;
 
@@ -2219,130 +2506,449 @@ static unsigned long arm_smmu_resource_size(struct arm_smmu_device *smmu)
 		return SZ_128K;
 }
 
+/* Start of Xen specific code. */
 static int arm_smmu_device_probe(struct platform_device *pdev)
 {
-	int irq, ret;
-	struct resource *res;
-	resource_size_t ioaddr;
-	struct arm_smmu_device *smmu;
-	struct device *dev = &pdev->dev;
-	bool bypass;
+    int irq, ret;
+    paddr_t ioaddr, iosize;
+    struct arm_smmu_device *smmu;
+    bool bypass;
+
+    smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
+    if ( !smmu )
+    {
+        dev_err(pdev, "failed to allocate arm_smmu_device\n");
+        return -ENOMEM;
+    }
+    smmu->dev = pdev;
+
+    if ( pdev->of_node )
+    {
+        ret = arm_smmu_device_dt_probe(pdev, smmu);
+    } else
+    {
+        ret = arm_smmu_device_acpi_probe(pdev, smmu);
+        if ( ret == -ENODEV )
+            return ret;
+    }
+
+    /* Set bypass mode according to firmware probing result */
+    bypass = !!ret;
+
+    /* Base address */
+    ret = dt_device_get_address(dev_to_dt(pdev), 0, &ioaddr, &iosize);
+    if( ret )
+        return -ENODEV;
+
+    if ( iosize < arm_smmu_resource_size(smmu) )
+    {
+        dev_err(pdev, "MMIO region too small (%lx)\n", iosize);
+        return -EINVAL;
+    }
+
+    /*
+     * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
+     * the PMCG registers which are reserved by the PMU driver.
+     */
+    smmu->base = ioremap_nocache(ioaddr, iosize);
+    if ( IS_ERR(smmu->base) )
+        return PTR_ERR(smmu->base);
+
+    if ( iosize > SZ_64K )
+    {
+        smmu->page1 = ioremap_nocache(ioaddr + SZ_64K, ARM_SMMU_REG_SZ);
+        if ( IS_ERR(smmu->page1) )
+            return PTR_ERR(smmu->page1);
+    }
+    else
+    {
+        smmu->page1 = smmu->base;
+    }
+
+    /* Interrupt lines */
+
+    irq = platform_get_irq_byname_optional(pdev, "combined");
+    if ( irq > 0 )
+        smmu->combined_irq = irq;
+    else
+    {
+        irq = platform_get_irq_byname_optional(pdev, "eventq");
+        if ( irq > 0 )
+            smmu->evtq.q.irq = irq;
+
+        irq = platform_get_irq_byname_optional(pdev, "priq");
+        if ( irq > 0 )
+            smmu->priq.q.irq = irq;
+
+        irq = platform_get_irq_byname_optional(pdev, "gerror");
+        if ( irq > 0 )
+            smmu->gerr_irq = irq;
+    }
+    /* Probe the h/w */
+    ret = arm_smmu_device_hw_probe(smmu);
+    if ( ret )
+        return ret;
+
+    /* Initialise in-memory data structures */
+    ret = arm_smmu_init_structures(smmu);
+    if ( ret )
+        return ret;
+
+    /* Reset the device */
+    ret = arm_smmu_device_reset(smmu, bypass);
+    if ( ret )
+        return ret;
+
+    /*
+     * Keep a list of all probed devices. This will be used to query
+     * the smmu devices based on the fwnode.
+     */
+    INIT_LIST_HEAD(&smmu->devices);
+
+    spin_lock(&arm_smmu_devices_lock);
+    list_add(&smmu->devices, &arm_smmu_devices);
+    spin_unlock(&arm_smmu_devices_lock);
+
+    return 0;
+}
+
+static int __must_check arm_smmu_iotlb_flush_all(struct domain *d)
+{
+    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
+    struct iommu_domain *io_domain;
+
+    spin_lock(&xen_domain->lock);
+
+    list_for_each_entry( io_domain, &xen_domain->contexts, list )
+    {
+        /*
+         * Only invalidate the context when SMMU is present.
+         * This is because the context initialization is delayed
+         * until a master has been added.
+         */
+        if ( unlikely(!ACCESS_ONCE(to_smmu_domain(io_domain)->smmu)) )
+            continue;
+
+        arm_smmu_tlb_inv_context(to_smmu_domain(io_domain));
+    }
 
-	smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
-	if (!smmu) {
-		dev_err(dev, "failed to allocate arm_smmu_device\n");
-		return -ENOMEM;
-	}
-	smmu->dev = dev;
+    spin_unlock(&xen_domain->lock);
+
+    return 0;
+}
+
+static int __must_check arm_smmu_iotlb_flush(struct domain *d, dfn_t dfn,
+                                             unsigned long page_count,
+                                             unsigned int flush_flags)
+{
+    return arm_smmu_iotlb_flush_all(d);
+}
 
-	if (dev->of_node) {
-		ret = arm_smmu_device_dt_probe(pdev, smmu);
-	} else {
-		ret = arm_smmu_device_acpi_probe(pdev, smmu);
-		if (ret == -ENODEV)
-			return ret;
-	}
+static struct arm_smmu_device *arm_smmu_get_by_dev(struct device *dev)
+{
+    struct arm_smmu_device *smmu = NULL;
 
-	/* Set bypass mode according to firmware probing result */
-	bypass = !!ret;
+    spin_lock(&arm_smmu_devices_lock);
+    list_for_each_entry( smmu, &arm_smmu_devices, devices )
+    {
+        if ( smmu->dev  == dev )
+        {
+            spin_unlock(&arm_smmu_devices_lock);
+            return smmu;
+        }
+    }
+    spin_unlock(&arm_smmu_devices_lock);
 
-	/* Base address */
-	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
-	if (resource_size(res) < arm_smmu_resource_size(smmu)) {
-		dev_err(dev, "MMIO region too small (%pr)\n", res);
-		return -EINVAL;
-	}
-	ioaddr = res->start;
+    return NULL;
+}
 
-	/*
-	 * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
-	 * the PMCG registers which are reserved by the PMU driver.
-	 */
-	smmu->base = arm_smmu_ioremap(dev, ioaddr, ARM_SMMU_REG_SZ);
-	if (IS_ERR(smmu->base))
-		return PTR_ERR(smmu->base);
-
-	if (arm_smmu_resource_size(smmu) > SZ_64K) {
-		smmu->page1 = arm_smmu_ioremap(dev, ioaddr + SZ_64K,
-					       ARM_SMMU_REG_SZ);
-		if (IS_ERR(smmu->page1))
-			return PTR_ERR(smmu->page1);
-	} else {
-		smmu->page1 = smmu->base;
-	}
+/* Probing and initialisation functions */
+static struct iommu_domain *arm_smmu_get_domain(struct domain *d,
+                                                struct device *dev)
+{
+    struct iommu_domain *io_domain;
+    struct arm_smmu_domain *smmu_domain;
+    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
+    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
+    struct arm_smmu_device *smmu = arm_smmu_get_by_dev(fwspec->iommu_dev);
+
+    if ( !smmu )
+        return NULL;
 
-	/* Interrupt lines */
+    /*
+     * Loop through the &xen_domain->contexts to locate a context
+     * assigned to this SMMU
+     */
+    list_for_each_entry( io_domain, &xen_domain->contexts, list )
+    {
+        smmu_domain = to_smmu_domain(io_domain);
+        if ( smmu_domain->smmu == smmu )
+            return io_domain;
+    }
 
-	irq = platform_get_irq_byname_optional(pdev, "combined");
-	if (irq > 0)
-		smmu->combined_irq = irq;
-	else {
-		irq = platform_get_irq_byname_optional(pdev, "eventq");
-		if (irq > 0)
-			smmu->evtq.q.irq = irq;
+    return NULL;
+}
 
-		irq = platform_get_irq_byname_optional(pdev, "priq");
-		if (irq > 0)
-			smmu->priq.q.irq = irq;
+static void arm_smmu_destroy_iommu_domain(struct iommu_domain *io_domain)
+{
+    list_del(&io_domain->list);
+    arm_smmu_domain_free(io_domain);
+}
 
-		irq = platform_get_irq_byname_optional(pdev, "gerror");
-		if (irq > 0)
-			smmu->gerr_irq = irq;
-	}
-	/* Probe the h/w */
-	ret = arm_smmu_device_hw_probe(smmu);
-	if (ret)
-		return ret;
+static int arm_smmu_assign_dev(struct domain *d, u8 devfn,
+                               struct device *dev, u32 flag)
+{
+    int ret = 0;
+    struct iommu_domain *io_domain;
+    struct arm_smmu_domain *smmu_domain;
+    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
 
-	/* Initialise in-memory data structures */
-	ret = arm_smmu_init_structures(smmu);
-	if (ret)
-		return ret;
+    if ( !dev->archdata.iommu )
+    {
+        dev->archdata.iommu = xzalloc(struct arm_smmu_xen_device);
+        if ( !dev->archdata.iommu )
+            return -ENOMEM;
+    }
 
-	/* Record our private device structure */
-	platform_set_drvdata(pdev, smmu);
+    spin_lock(&xen_domain->lock);
 
-	/* Reset the device */
-	ret = arm_smmu_device_reset(smmu, bypass);
-	if (ret)
-		return ret;
+    /*
+     * Check to see if an iommu_domain already exists for this xen domain
+     * under the same SMMU
+     */
+    io_domain = arm_smmu_get_domain(d, dev);
+    if ( !io_domain )
+    {
+        io_domain = arm_smmu_domain_alloc(IOMMU_DOMAIN_DMA);
+        if ( !io_domain )
+        {
+            ret = -ENOMEM;
+            goto out;
+        }
 
-	/* And we're up. Go go go! */
-	ret = iommu_device_sysfs_add(&smmu->iommu, dev, NULL,
-				     "smmu3.%pa", &ioaddr);
-	if (ret)
-		return ret;
+        smmu_domain = to_smmu_domain(io_domain);
+        smmu_domain->s2_cfg.domain = d;
 
-	iommu_device_set_ops(&smmu->iommu, &arm_smmu_ops);
-	iommu_device_set_fwnode(&smmu->iommu, dev->fwnode);
+        /* Chain the new context to the domain */
+        list_add(&io_domain->list, &xen_domain->contexts);
 
-	ret = iommu_device_register(&smmu->iommu);
-	if (ret) {
-		dev_err(dev, "Failed to register iommu\n");
-		return ret;
-	}
+    }
+
+    ret = arm_smmu_attach_dev(io_domain, dev);
+    if ( ret )
+    {
+        if ( io_domain->ref.counter == 0 )
+            arm_smmu_destroy_iommu_domain(io_domain);
+    }
+    else
+    {
+        atomic_inc(&io_domain->ref);
+    }
 
-	return arm_smmu_set_bus_ops(&arm_smmu_ops);
+out:
+    spin_unlock(&xen_domain->lock);
+    return ret;
 }
 
-static int arm_smmu_device_remove(struct platform_device *pdev)
+static int arm_smmu_deassign_dev(struct domain *d, struct device *dev)
 {
-	struct arm_smmu_device *smmu = platform_get_drvdata(pdev);
+    struct iommu_domain *io_domain = arm_smmu_get_domain(d, dev);
+    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
+    struct arm_smmu_domain *arm_smmu = to_smmu_domain(io_domain);
+    struct arm_smmu_master *master = dev_iommu_priv_get(dev);
 
-	arm_smmu_set_bus_ops(NULL);
-	iommu_device_unregister(&smmu->iommu);
-	iommu_device_sysfs_remove(&smmu->iommu);
-	arm_smmu_device_disable(smmu);
+    if ( !arm_smmu || arm_smmu->s2_cfg.domain != d )
+    {
+        dev_err(dev, " not attached to domain %d\n", d->domain_id);
+        return -ESRCH;
+    }
 
-	return 0;
+    spin_lock(&xen_domain->lock);
+
+    arm_smmu_detach_dev(master);
+    atomic_dec(&io_domain->ref);
+
+    if ( io_domain->ref.counter == 0 )
+        arm_smmu_destroy_iommu_domain(io_domain);
+
+    spin_unlock(&xen_domain->lock);
+
+    return 0;
+}
+
+static int arm_smmu_reassign_dev(struct domain *s, struct domain *t,
+                                 u8 devfn,  struct device *dev)
+{
+    int ret = 0;
+
+    /* Don't allow remapping on other domain than hwdom */
+    if ( t && t != hardware_domain )
+        return -EPERM;
+
+    if ( t == s )
+        return 0;
+
+    ret = arm_smmu_deassign_dev(s, dev);
+    if ( ret )
+        return ret;
+
+    if ( t )
+    {
+        /* No flags are defined for ARM. */
+        ret = arm_smmu_assign_dev(t, devfn, dev, 0);
+        if ( ret )
+            return ret;
+    }
+
+    return 0;
+}
+
+static int arm_smmu_iommu_xen_domain_init(struct domain *d)
+{
+    struct arm_smmu_xen_domain *xen_domain;
+
+    xen_domain = xzalloc(struct arm_smmu_xen_domain);
+    if ( !xen_domain )
+        return -ENOMEM;
+
+    spin_lock_init(&xen_domain->lock);
+    INIT_LIST_HEAD(&xen_domain->contexts);
+
+    dom_iommu(d)->arch.priv = xen_domain;
+
+    return 0;
+}
+
+static void __hwdom_init arm_smmu_iommu_hwdom_init(struct domain *d)
+{
+}
+
+static void arm_smmu_iommu_xen_domain_teardown(struct domain *d)
+{
+    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
+
+    ASSERT(list_empty(&xen_domain->contexts));
+    xfree(xen_domain);
+}
+
+static int arm_smmu_dt_xlate(struct device *dev,
+                             const struct dt_phandle_args *args)
+{
+    int ret;
+    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
+
+    ret = iommu_fwspec_add_ids(dev, args->args, 1);
+    if ( ret )
+        return ret;
+
+    if ( dt_device_is_protected(dev_to_dt(dev)) )
+    {
+        dev_err(dev, "Already added to SMMUv3\n");
+        return -EEXIST;
+    }
+
+    /* Let Xen know that the master device is protected by an IOMMU. */
+    dt_device_set_protected(dev_to_dt(dev));
+
+    dev_info(dev, "Added master device (SMMUv3 %s StreamIds %u)\n",
+            dev_name(fwspec->iommu_dev), fwspec->num_ids);
+
+    return 0;
 }
 
-static void arm_smmu_device_shutdown(struct platform_device *pdev)
+static int arm_smmu_add_device(u8 devfn, struct device *dev)
 {
-	arm_smmu_device_remove(pdev);
+    int i, ret;
+    struct arm_smmu_device *smmu;
+    struct arm_smmu_master *master;
+    struct iommu_fwspec *fwspec;
+
+    fwspec = dev_iommu_fwspec_get(dev);
+    if ( !fwspec )
+        return -ENODEV;
+
+    smmu = arm_smmu_get_by_dev(fwspec->iommu_dev);
+    if ( !smmu )
+        return -ENODEV;
+
+    master = xzalloc(struct arm_smmu_master);
+    if ( !master )
+        return -ENOMEM;
+
+    master->dev = dev;
+    master->smmu = smmu;
+    master->sids = fwspec->ids;
+    master->num_sids = fwspec->num_ids;
+
+    dev_iommu_priv_set(dev, master);
+
+    /* Check the SIDs are in range of the SMMU and our stream table */
+    for ( i = 0; i < master->num_sids; i++ )
+    {
+        u32 sid = master->sids[i];
+
+        if ( !arm_smmu_sid_in_range(smmu, sid) )
+        {
+            ret = -ERANGE;
+            goto err_free_master;
+        }
+
+        /* Ensure l2 strtab is initialised */
+        if ( smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB )
+        {
+            ret = arm_smmu_init_l2_strtab(smmu, sid);
+            if ( ret )
+                goto err_free_master;
+        }
+    }
+
+    return 0;
+
+err_free_master:
+    xfree(master);
+    dev_iommu_priv_set(dev, NULL);
+    return ret;
 }
 
-static const struct of_device_id arm_smmu_of_match[] = {
-	{ .compatible = "arm,smmu-v3", },
-	{ },
+static const struct iommu_ops arm_smmu_iommu_ops = {
+    .init = arm_smmu_iommu_xen_domain_init,
+    .hwdom_init = arm_smmu_iommu_hwdom_init,
+    .teardown = arm_smmu_iommu_xen_domain_teardown,
+    .iotlb_flush = arm_smmu_iotlb_flush,
+    .iotlb_flush_all = arm_smmu_iotlb_flush_all,
+    .assign_device = arm_smmu_assign_dev,
+    .reassign_device = arm_smmu_reassign_dev,
+    .map_page = arm_iommu_map_page,
+    .unmap_page = arm_iommu_unmap_page,
+    .dt_xlate = arm_smmu_dt_xlate,
+    .add_device = arm_smmu_add_device,
+};
+
+static const struct dt_device_match arm_smmu_of_match[] = {
+    { .compatible = "arm,smmu-v3", },
+    { },
 };
+
+static __init int arm_smmu_dt_init(struct dt_device_node *dev,
+                                   const void *data)
+{
+    int rc;
+
+    /*
+     * Even if the device can't be initialized, we don't want to
+     * give the SMMU device to dom0.
+     */
+    dt_device_set_used_by(dev, DOMID_XEN);
+
+    rc = arm_smmu_device_probe(dt_to_dev(dev));
+    if ( rc )
+        return rc;
+
+    iommu_set_ops(&arm_smmu_iommu_ops);
+    return 0;
+}
+
+DT_DEVICE_START(smmuv3, "ARM SMMU V3", DEVICE_IOMMU)
+    .dt_match = arm_smmu_of_match,
+    .init = arm_smmu_dt_init,
+DT_DEVICE_END
-- 
2.17.1



^ permalink raw reply related	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 1/8] xen/arm: Import the SMMUv3 driver from Linux
  2020-11-26 17:02 ` [PATCH v2 1/8] xen/arm: Import the SMMUv3 driver from Linux Rahul Singh
@ 2020-12-01 22:01   ` Stefano Stabellini
  0 siblings, 0 replies; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-01 22:01 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, bertrand.marquis, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

On Thu, 26 Nov 2020, Rahul Singh wrote:
> Based on tag Linux 5.9.8 commit 951cbbc386ff01b50da4f46387e994e81d9ab431
> 
> It's a copy of the Linux SMMUv3 driver. Xen specific code has not
> been added yet and code has not been compiled.
> 
> Signed-off-by: Rahul Singh <rahul.singh@arm.com>

Acked-by: Stefano Stabellini <sstabellini@kernel.org>


> ---
>  xen/drivers/passthrough/arm/smmu-v3.c | 4164 +++++++++++++++++++++++++
>  1 file changed, 4164 insertions(+)
>  create mode 100644 xen/drivers/passthrough/arm/smmu-v3.c
> 
> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
> new file mode 100644
> index 0000000000..c192544e87
> --- /dev/null
> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
> @@ -0,0 +1,4164 @@
> +// SPDX-License-Identifier: GPL-2.0
> +/*
> + * IOMMU API for ARM architected SMMUv3 implementations.
> + *
> + * Copyright (C) 2015 ARM Limited
> + *
> + * Author: Will Deacon <will.deacon@arm.com>
> + *
> + * This driver is powered by bad coffee and bombay mix.
> + */
> +
> +#include <linux/acpi.h>
> +#include <linux/acpi_iort.h>
> +#include <linux/bitfield.h>
> +#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>
> +#include <linux/iommu.h>
> +#include <linux/iopoll.h>
> +#include <linux/module.h>
> +#include <linux/msi.h>
> +#include <linux/of.h>
> +#include <linux/of_address.h>
> +#include <linux/of_iommu.h>
> +#include <linux/of_platform.h>
> +#include <linux/pci.h>
> +#include <linux/pci-ats.h>
> +#include <linux/platform_device.h>
> +
> +#include <linux/amba/bus.h>
> +
> +/* MMIO registers */
> +#define ARM_SMMU_IDR0			0x0
> +#define IDR0_ST_LVL			GENMASK(28, 27)
> +#define IDR0_ST_LVL_2LVL		1
> +#define IDR0_STALL_MODEL		GENMASK(25, 24)
> +#define IDR0_STALL_MODEL_STALL		0
> +#define IDR0_STALL_MODEL_FORCE		2
> +#define IDR0_TTENDIAN			GENMASK(22, 21)
> +#define IDR0_TTENDIAN_MIXED		0
> +#define IDR0_TTENDIAN_LE		2
> +#define IDR0_TTENDIAN_BE		3
> +#define IDR0_CD2L			(1 << 19)
> +#define IDR0_VMID16			(1 << 18)
> +#define IDR0_PRI			(1 << 16)
> +#define IDR0_SEV			(1 << 14)
> +#define IDR0_MSI			(1 << 13)
> +#define IDR0_ASID16			(1 << 12)
> +#define IDR0_ATS			(1 << 10)
> +#define IDR0_HYP			(1 << 9)
> +#define IDR0_COHACC			(1 << 4)
> +#define IDR0_TTF			GENMASK(3, 2)
> +#define IDR0_TTF_AARCH64		2
> +#define IDR0_TTF_AARCH32_64		3
> +#define IDR0_S1P			(1 << 1)
> +#define IDR0_S2P			(1 << 0)
> +
> +#define ARM_SMMU_IDR1			0x4
> +#define IDR1_TABLES_PRESET		(1 << 30)
> +#define IDR1_QUEUES_PRESET		(1 << 29)
> +#define IDR1_REL			(1 << 28)
> +#define IDR1_CMDQS			GENMASK(25, 21)
> +#define IDR1_EVTQS			GENMASK(20, 16)
> +#define IDR1_PRIQS			GENMASK(15, 11)
> +#define IDR1_SSIDSIZE			GENMASK(10, 6)
> +#define IDR1_SIDSIZE			GENMASK(5, 0)
> +
> +#define ARM_SMMU_IDR3			0xc
> +#define IDR3_RIL			(1 << 10)
> +
> +#define ARM_SMMU_IDR5			0x14
> +#define IDR5_STALL_MAX			GENMASK(31, 16)
> +#define IDR5_GRAN64K			(1 << 6)
> +#define IDR5_GRAN16K			(1 << 5)
> +#define IDR5_GRAN4K			(1 << 4)
> +#define IDR5_OAS			GENMASK(2, 0)
> +#define IDR5_OAS_32_BIT			0
> +#define IDR5_OAS_36_BIT			1
> +#define IDR5_OAS_40_BIT			2
> +#define IDR5_OAS_42_BIT			3
> +#define IDR5_OAS_44_BIT			4
> +#define IDR5_OAS_48_BIT			5
> +#define IDR5_OAS_52_BIT			6
> +#define IDR5_VAX			GENMASK(11, 10)
> +#define IDR5_VAX_52_BIT			1
> +
> +#define ARM_SMMU_CR0			0x20
> +#define CR0_ATSCHK			(1 << 4)
> +#define CR0_CMDQEN			(1 << 3)
> +#define CR0_EVTQEN			(1 << 2)
> +#define CR0_PRIQEN			(1 << 1)
> +#define CR0_SMMUEN			(1 << 0)
> +
> +#define ARM_SMMU_CR0ACK			0x24
> +
> +#define ARM_SMMU_CR1			0x28
> +#define CR1_TABLE_SH			GENMASK(11, 10)
> +#define CR1_TABLE_OC			GENMASK(9, 8)
> +#define CR1_TABLE_IC			GENMASK(7, 6)
> +#define CR1_QUEUE_SH			GENMASK(5, 4)
> +#define CR1_QUEUE_OC			GENMASK(3, 2)
> +#define CR1_QUEUE_IC			GENMASK(1, 0)
> +/* CR1 cacheability fields don't quite follow the usual TCR-style encoding */
> +#define CR1_CACHE_NC			0
> +#define CR1_CACHE_WB			1
> +#define CR1_CACHE_WT			2
> +
> +#define ARM_SMMU_CR2			0x2c
> +#define CR2_PTM				(1 << 2)
> +#define CR2_RECINVSID			(1 << 1)
> +#define CR2_E2H				(1 << 0)
> +
> +#define ARM_SMMU_GBPA			0x44
> +#define GBPA_UPDATE			(1 << 31)
> +#define GBPA_ABORT			(1 << 20)
> +
> +#define ARM_SMMU_IRQ_CTRL		0x50
> +#define IRQ_CTRL_EVTQ_IRQEN		(1 << 2)
> +#define IRQ_CTRL_PRIQ_IRQEN		(1 << 1)
> +#define IRQ_CTRL_GERROR_IRQEN		(1 << 0)
> +
> +#define ARM_SMMU_IRQ_CTRLACK		0x54
> +
> +#define ARM_SMMU_GERROR			0x60
> +#define GERROR_SFM_ERR			(1 << 8)
> +#define GERROR_MSI_GERROR_ABT_ERR	(1 << 7)
> +#define GERROR_MSI_PRIQ_ABT_ERR		(1 << 6)
> +#define GERROR_MSI_EVTQ_ABT_ERR		(1 << 5)
> +#define GERROR_MSI_CMDQ_ABT_ERR		(1 << 4)
> +#define GERROR_PRIQ_ABT_ERR		(1 << 3)
> +#define GERROR_EVTQ_ABT_ERR		(1 << 2)
> +#define GERROR_CMDQ_ERR			(1 << 0)
> +#define GERROR_ERR_MASK			0xfd
> +
> +#define ARM_SMMU_GERRORN		0x64
> +
> +#define ARM_SMMU_GERROR_IRQ_CFG0	0x68
> +#define ARM_SMMU_GERROR_IRQ_CFG1	0x70
> +#define ARM_SMMU_GERROR_IRQ_CFG2	0x74
> +
> +#define ARM_SMMU_STRTAB_BASE		0x80
> +#define STRTAB_BASE_RA			(1UL << 62)
> +#define STRTAB_BASE_ADDR_MASK		GENMASK_ULL(51, 6)
> +
> +#define ARM_SMMU_STRTAB_BASE_CFG	0x88
> +#define STRTAB_BASE_CFG_FMT		GENMASK(17, 16)
> +#define STRTAB_BASE_CFG_FMT_LINEAR	0
> +#define STRTAB_BASE_CFG_FMT_2LVL	1
> +#define STRTAB_BASE_CFG_SPLIT		GENMASK(10, 6)
> +#define STRTAB_BASE_CFG_LOG2SIZE	GENMASK(5, 0)
> +
> +#define ARM_SMMU_CMDQ_BASE		0x90
> +#define ARM_SMMU_CMDQ_PROD		0x98
> +#define ARM_SMMU_CMDQ_CONS		0x9c
> +
> +#define ARM_SMMU_EVTQ_BASE		0xa0
> +#define ARM_SMMU_EVTQ_PROD		0x100a8
> +#define ARM_SMMU_EVTQ_CONS		0x100ac
> +#define ARM_SMMU_EVTQ_IRQ_CFG0		0xb0
> +#define ARM_SMMU_EVTQ_IRQ_CFG1		0xb8
> +#define ARM_SMMU_EVTQ_IRQ_CFG2		0xbc
> +
> +#define ARM_SMMU_PRIQ_BASE		0xc0
> +#define ARM_SMMU_PRIQ_PROD		0x100c8
> +#define ARM_SMMU_PRIQ_CONS		0x100cc
> +#define ARM_SMMU_PRIQ_IRQ_CFG0		0xd0
> +#define ARM_SMMU_PRIQ_IRQ_CFG1		0xd8
> +#define ARM_SMMU_PRIQ_IRQ_CFG2		0xdc
> +
> +#define ARM_SMMU_REG_SZ			0xe00
> +
> +/* Common MSI config fields */
> +#define MSI_CFG0_ADDR_MASK		GENMASK_ULL(51, 2)
> +#define MSI_CFG2_SH			GENMASK(5, 4)
> +#define MSI_CFG2_MEMATTR		GENMASK(3, 0)
> +
> +/* Common memory attribute values */
> +#define ARM_SMMU_SH_NSH			0
> +#define ARM_SMMU_SH_OSH			2
> +#define ARM_SMMU_SH_ISH			3
> +#define ARM_SMMU_MEMATTR_DEVICE_nGnRE	0x1
> +#define ARM_SMMU_MEMATTR_OIWB		0xf
> +
> +#define Q_IDX(llq, p)			((p) & ((1 << (llq)->max_n_shift) - 1))
> +#define Q_WRP(llq, p)			((p) & (1 << (llq)->max_n_shift))
> +#define Q_OVERFLOW_FLAG			(1U << 31)
> +#define Q_OVF(p)			((p) & Q_OVERFLOW_FLAG)
> +#define Q_ENT(q, p)			((q)->base +			\
> +					 Q_IDX(&((q)->llq), p) *	\
> +					 (q)->ent_dwords)
> +
> +#define Q_BASE_RWA			(1UL << 62)
> +#define Q_BASE_ADDR_MASK		GENMASK_ULL(51, 5)
> +#define Q_BASE_LOG2SIZE			GENMASK(4, 0)
> +
> +/* Ensure DMA allocations are naturally aligned */
> +#ifdef CONFIG_CMA_ALIGNMENT
> +#define Q_MAX_SZ_SHIFT			(PAGE_SHIFT + CONFIG_CMA_ALIGNMENT)
> +#else
> +#define Q_MAX_SZ_SHIFT			(PAGE_SHIFT + MAX_ORDER - 1)
> +#endif
> +
> +/*
> + * Stream table.
> + *
> + * Linear: Enough to cover 1 << IDR1.SIDSIZE entries
> + * 2lvl: 128k L1 entries,
> + *       256 lazy entries per table (each table covers a PCI bus)
> + */
> +#define STRTAB_L1_SZ_SHIFT		20
> +#define STRTAB_SPLIT			8
> +
> +#define STRTAB_L1_DESC_DWORDS		1
> +#define STRTAB_L1_DESC_SPAN		GENMASK_ULL(4, 0)
> +#define STRTAB_L1_DESC_L2PTR_MASK	GENMASK_ULL(51, 6)
> +
> +#define STRTAB_STE_DWORDS		8
> +#define STRTAB_STE_0_V			(1UL << 0)
> +#define STRTAB_STE_0_CFG		GENMASK_ULL(3, 1)
> +#define STRTAB_STE_0_CFG_ABORT		0
> +#define STRTAB_STE_0_CFG_BYPASS		4
> +#define STRTAB_STE_0_CFG_S1_TRANS	5
> +#define STRTAB_STE_0_CFG_S2_TRANS	6
> +
> +#define STRTAB_STE_0_S1FMT		GENMASK_ULL(5, 4)
> +#define STRTAB_STE_0_S1FMT_LINEAR	0
> +#define STRTAB_STE_0_S1FMT_64K_L2	2
> +#define STRTAB_STE_0_S1CTXPTR_MASK	GENMASK_ULL(51, 6)
> +#define STRTAB_STE_0_S1CDMAX		GENMASK_ULL(63, 59)
> +
> +#define STRTAB_STE_1_S1DSS		GENMASK_ULL(1, 0)
> +#define STRTAB_STE_1_S1DSS_TERMINATE	0x0
> +#define STRTAB_STE_1_S1DSS_BYPASS	0x1
> +#define STRTAB_STE_1_S1DSS_SSID0	0x2
> +
> +#define STRTAB_STE_1_S1C_CACHE_NC	0UL
> +#define STRTAB_STE_1_S1C_CACHE_WBRA	1UL
> +#define STRTAB_STE_1_S1C_CACHE_WT	2UL
> +#define STRTAB_STE_1_S1C_CACHE_WB	3UL
> +#define STRTAB_STE_1_S1CIR		GENMASK_ULL(3, 2)
> +#define STRTAB_STE_1_S1COR		GENMASK_ULL(5, 4)
> +#define STRTAB_STE_1_S1CSH		GENMASK_ULL(7, 6)
> +
> +#define STRTAB_STE_1_S1STALLD		(1UL << 27)
> +
> +#define STRTAB_STE_1_EATS		GENMASK_ULL(29, 28)
> +#define STRTAB_STE_1_EATS_ABT		0UL
> +#define STRTAB_STE_1_EATS_TRANS		1UL
> +#define STRTAB_STE_1_EATS_S1CHK		2UL
> +
> +#define STRTAB_STE_1_STRW		GENMASK_ULL(31, 30)
> +#define STRTAB_STE_1_STRW_NSEL1		0UL
> +#define STRTAB_STE_1_STRW_EL2		2UL
> +
> +#define STRTAB_STE_1_SHCFG		GENMASK_ULL(45, 44)
> +#define STRTAB_STE_1_SHCFG_INCOMING	1UL
> +
> +#define STRTAB_STE_2_S2VMID		GENMASK_ULL(15, 0)
> +#define STRTAB_STE_2_VTCR		GENMASK_ULL(50, 32)
> +#define STRTAB_STE_2_VTCR_S2T0SZ	GENMASK_ULL(5, 0)
> +#define STRTAB_STE_2_VTCR_S2SL0		GENMASK_ULL(7, 6)
> +#define STRTAB_STE_2_VTCR_S2IR0		GENMASK_ULL(9, 8)
> +#define STRTAB_STE_2_VTCR_S2OR0		GENMASK_ULL(11, 10)
> +#define STRTAB_STE_2_VTCR_S2SH0		GENMASK_ULL(13, 12)
> +#define STRTAB_STE_2_VTCR_S2TG		GENMASK_ULL(15, 14)
> +#define STRTAB_STE_2_VTCR_S2PS		GENMASK_ULL(18, 16)
> +#define STRTAB_STE_2_S2AA64		(1UL << 51)
> +#define STRTAB_STE_2_S2ENDI		(1UL << 52)
> +#define STRTAB_STE_2_S2PTW		(1UL << 54)
> +#define STRTAB_STE_2_S2R		(1UL << 58)
> +
> +#define STRTAB_STE_3_S2TTB_MASK		GENMASK_ULL(51, 4)
> +
> +/*
> + * Context descriptors.
> + *
> + * Linear: when less than 1024 SSIDs are supported
> + * 2lvl: at most 1024 L1 entries,
> + *       1024 lazy entries per table.
> + */
> +#define CTXDESC_SPLIT			10
> +#define CTXDESC_L2_ENTRIES		(1 << CTXDESC_SPLIT)
> +
> +#define CTXDESC_L1_DESC_DWORDS		1
> +#define CTXDESC_L1_DESC_V		(1UL << 0)
> +#define CTXDESC_L1_DESC_L2PTR_MASK	GENMASK_ULL(51, 12)
> +
> +#define CTXDESC_CD_DWORDS		8
> +#define CTXDESC_CD_0_TCR_T0SZ		GENMASK_ULL(5, 0)
> +#define CTXDESC_CD_0_TCR_TG0		GENMASK_ULL(7, 6)
> +#define CTXDESC_CD_0_TCR_IRGN0		GENMASK_ULL(9, 8)
> +#define CTXDESC_CD_0_TCR_ORGN0		GENMASK_ULL(11, 10)
> +#define CTXDESC_CD_0_TCR_SH0		GENMASK_ULL(13, 12)
> +#define CTXDESC_CD_0_TCR_EPD0		(1ULL << 14)
> +#define CTXDESC_CD_0_TCR_EPD1		(1ULL << 30)
> +
> +#define CTXDESC_CD_0_ENDI		(1UL << 15)
> +#define CTXDESC_CD_0_V			(1UL << 31)
> +
> +#define CTXDESC_CD_0_TCR_IPS		GENMASK_ULL(34, 32)
> +#define CTXDESC_CD_0_TCR_TBI0		(1ULL << 38)
> +
> +#define CTXDESC_CD_0_AA64		(1UL << 41)
> +#define CTXDESC_CD_0_S			(1UL << 44)
> +#define CTXDESC_CD_0_R			(1UL << 45)
> +#define CTXDESC_CD_0_A			(1UL << 46)
> +#define CTXDESC_CD_0_ASET		(1UL << 47)
> +#define CTXDESC_CD_0_ASID		GENMASK_ULL(63, 48)
> +
> +#define CTXDESC_CD_1_TTB0_MASK		GENMASK_ULL(51, 4)
> +
> +/*
> + * When the SMMU only supports linear context descriptor tables, pick a
> + * reasonable size limit (64kB).
> + */
> +#define CTXDESC_LINEAR_CDMAX		ilog2(SZ_64K / (CTXDESC_CD_DWORDS << 3))
> +
> +/* Command queue */
> +#define CMDQ_ENT_SZ_SHIFT		4
> +#define CMDQ_ENT_DWORDS			((1 << CMDQ_ENT_SZ_SHIFT) >> 3)
> +#define CMDQ_MAX_SZ_SHIFT		(Q_MAX_SZ_SHIFT - CMDQ_ENT_SZ_SHIFT)
> +
> +#define CMDQ_CONS_ERR			GENMASK(30, 24)
> +#define CMDQ_ERR_CERROR_NONE_IDX	0
> +#define CMDQ_ERR_CERROR_ILL_IDX		1
> +#define CMDQ_ERR_CERROR_ABT_IDX		2
> +#define CMDQ_ERR_CERROR_ATC_INV_IDX	3
> +
> +#define CMDQ_PROD_OWNED_FLAG		Q_OVERFLOW_FLAG
> +
> +/*
> + * This is used to size the command queue and therefore must be at least
> + * BITS_PER_LONG so that the valid_map works correctly (it relies on the
> + * total number of queue entries being a multiple of BITS_PER_LONG).
> + */
> +#define CMDQ_BATCH_ENTRIES		BITS_PER_LONG
> +
> +#define CMDQ_0_OP			GENMASK_ULL(7, 0)
> +#define CMDQ_0_SSV			(1UL << 11)
> +
> +#define CMDQ_PREFETCH_0_SID		GENMASK_ULL(63, 32)
> +#define CMDQ_PREFETCH_1_SIZE		GENMASK_ULL(4, 0)
> +#define CMDQ_PREFETCH_1_ADDR_MASK	GENMASK_ULL(63, 12)
> +
> +#define CMDQ_CFGI_0_SSID		GENMASK_ULL(31, 12)
> +#define CMDQ_CFGI_0_SID			GENMASK_ULL(63, 32)
> +#define CMDQ_CFGI_1_LEAF		(1UL << 0)
> +#define CMDQ_CFGI_1_RANGE		GENMASK_ULL(4, 0)
> +
> +#define CMDQ_TLBI_0_NUM			GENMASK_ULL(16, 12)
> +#define CMDQ_TLBI_RANGE_NUM_MAX		31
> +#define CMDQ_TLBI_0_SCALE		GENMASK_ULL(24, 20)
> +#define CMDQ_TLBI_0_VMID		GENMASK_ULL(47, 32)
> +#define CMDQ_TLBI_0_ASID		GENMASK_ULL(63, 48)
> +#define CMDQ_TLBI_1_LEAF		(1UL << 0)
> +#define CMDQ_TLBI_1_TTL			GENMASK_ULL(9, 8)
> +#define CMDQ_TLBI_1_TG			GENMASK_ULL(11, 10)
> +#define CMDQ_TLBI_1_VA_MASK		GENMASK_ULL(63, 12)
> +#define CMDQ_TLBI_1_IPA_MASK		GENMASK_ULL(51, 12)
> +
> +#define CMDQ_ATC_0_SSID			GENMASK_ULL(31, 12)
> +#define CMDQ_ATC_0_SID			GENMASK_ULL(63, 32)
> +#define CMDQ_ATC_0_GLOBAL		(1UL << 9)
> +#define CMDQ_ATC_1_SIZE			GENMASK_ULL(5, 0)
> +#define CMDQ_ATC_1_ADDR_MASK		GENMASK_ULL(63, 12)
> +
> +#define CMDQ_PRI_0_SSID			GENMASK_ULL(31, 12)
> +#define CMDQ_PRI_0_SID			GENMASK_ULL(63, 32)
> +#define CMDQ_PRI_1_GRPID		GENMASK_ULL(8, 0)
> +#define CMDQ_PRI_1_RESP			GENMASK_ULL(13, 12)
> +
> +#define CMDQ_SYNC_0_CS			GENMASK_ULL(13, 12)
> +#define CMDQ_SYNC_0_CS_NONE		0
> +#define CMDQ_SYNC_0_CS_IRQ		1
> +#define CMDQ_SYNC_0_CS_SEV		2
> +#define CMDQ_SYNC_0_MSH			GENMASK_ULL(23, 22)
> +#define CMDQ_SYNC_0_MSIATTR		GENMASK_ULL(27, 24)
> +#define CMDQ_SYNC_0_MSIDATA		GENMASK_ULL(63, 32)
> +#define CMDQ_SYNC_1_MSIADDR_MASK	GENMASK_ULL(51, 2)
> +
> +/* Event queue */
> +#define EVTQ_ENT_SZ_SHIFT		5
> +#define EVTQ_ENT_DWORDS			((1 << EVTQ_ENT_SZ_SHIFT) >> 3)
> +#define EVTQ_MAX_SZ_SHIFT		(Q_MAX_SZ_SHIFT - EVTQ_ENT_SZ_SHIFT)
> +
> +#define EVTQ_0_ID			GENMASK_ULL(7, 0)
> +
> +/* PRI queue */
> +#define PRIQ_ENT_SZ_SHIFT		4
> +#define PRIQ_ENT_DWORDS			((1 << PRIQ_ENT_SZ_SHIFT) >> 3)
> +#define PRIQ_MAX_SZ_SHIFT		(Q_MAX_SZ_SHIFT - PRIQ_ENT_SZ_SHIFT)
> +
> +#define PRIQ_0_SID			GENMASK_ULL(31, 0)
> +#define PRIQ_0_SSID			GENMASK_ULL(51, 32)
> +#define PRIQ_0_PERM_PRIV		(1UL << 58)
> +#define PRIQ_0_PERM_EXEC		(1UL << 59)
> +#define PRIQ_0_PERM_READ		(1UL << 60)
> +#define PRIQ_0_PERM_WRITE		(1UL << 61)
> +#define PRIQ_0_PRG_LAST			(1UL << 62)
> +#define PRIQ_0_SSID_V			(1UL << 63)
> +
> +#define PRIQ_1_PRG_IDX			GENMASK_ULL(8, 0)
> +#define PRIQ_1_ADDR_MASK		GENMASK_ULL(63, 12)
> +
> +/* High-level queue structures */
> +#define ARM_SMMU_POLL_TIMEOUT_US	1000000 /* 1s! */
> +#define ARM_SMMU_POLL_SPIN_COUNT	10
> +
> +#define MSI_IOVA_BASE			0x8000000
> +#define MSI_IOVA_LENGTH			0x100000
> +
> +static bool disable_bypass = 1;
> +module_param_named(disable_bypass, disable_bypass, bool, S_IRUGO);
> +MODULE_PARM_DESC(disable_bypass,
> +	"Disable bypass streams such that incoming transactions from devices that are not attached to an iommu domain will report an abort back to the device and will not be allowed to pass through the SMMU.");
> +
> +enum pri_resp {
> +	PRI_RESP_DENY = 0,
> +	PRI_RESP_FAIL = 1,
> +	PRI_RESP_SUCC = 2,
> +};
> +
> +enum arm_smmu_msi_index {
> +	EVTQ_MSI_INDEX,
> +	GERROR_MSI_INDEX,
> +	PRIQ_MSI_INDEX,
> +	ARM_SMMU_MAX_MSIS,
> +};
> +
> +static phys_addr_t arm_smmu_msi_cfg[ARM_SMMU_MAX_MSIS][3] = {
> +	[EVTQ_MSI_INDEX] = {
> +		ARM_SMMU_EVTQ_IRQ_CFG0,
> +		ARM_SMMU_EVTQ_IRQ_CFG1,
> +		ARM_SMMU_EVTQ_IRQ_CFG2,
> +	},
> +	[GERROR_MSI_INDEX] = {
> +		ARM_SMMU_GERROR_IRQ_CFG0,
> +		ARM_SMMU_GERROR_IRQ_CFG1,
> +		ARM_SMMU_GERROR_IRQ_CFG2,
> +	},
> +	[PRIQ_MSI_INDEX] = {
> +		ARM_SMMU_PRIQ_IRQ_CFG0,
> +		ARM_SMMU_PRIQ_IRQ_CFG1,
> +		ARM_SMMU_PRIQ_IRQ_CFG2,
> +	},
> +};
> +
> +struct arm_smmu_cmdq_ent {
> +	/* Common fields */
> +	u8				opcode;
> +	bool				substream_valid;
> +
> +	/* Command-specific fields */
> +	union {
> +		#define CMDQ_OP_PREFETCH_CFG	0x1
> +		struct {
> +			u32			sid;
> +			u8			size;
> +			u64			addr;
> +		} prefetch;
> +
> +		#define CMDQ_OP_CFGI_STE	0x3
> +		#define CMDQ_OP_CFGI_ALL	0x4
> +		#define CMDQ_OP_CFGI_CD		0x5
> +		#define CMDQ_OP_CFGI_CD_ALL	0x6
> +		struct {
> +			u32			sid;
> +			u32			ssid;
> +			union {
> +				bool		leaf;
> +				u8		span;
> +			};
> +		} cfgi;
> +
> +		#define CMDQ_OP_TLBI_NH_ASID	0x11
> +		#define CMDQ_OP_TLBI_NH_VA	0x12
> +		#define CMDQ_OP_TLBI_EL2_ALL	0x20
> +		#define CMDQ_OP_TLBI_S12_VMALL	0x28
> +		#define CMDQ_OP_TLBI_S2_IPA	0x2a
> +		#define CMDQ_OP_TLBI_NSNH_ALL	0x30
> +		struct {
> +			u8			num;
> +			u8			scale;
> +			u16			asid;
> +			u16			vmid;
> +			bool			leaf;
> +			u8			ttl;
> +			u8			tg;
> +			u64			addr;
> +		} tlbi;
> +
> +		#define CMDQ_OP_ATC_INV		0x40
> +		#define ATC_INV_SIZE_ALL	52
> +		struct {
> +			u32			sid;
> +			u32			ssid;
> +			u64			addr;
> +			u8			size;
> +			bool			global;
> +		} atc;
> +
> +		#define CMDQ_OP_PRI_RESP	0x41
> +		struct {
> +			u32			sid;
> +			u32			ssid;
> +			u16			grpid;
> +			enum pri_resp		resp;
> +		} pri;
> +
> +		#define CMDQ_OP_CMD_SYNC	0x46
> +		struct {
> +			u64			msiaddr;
> +		} sync;
> +	};
> +};
> +
> +struct arm_smmu_ll_queue {
> +	union {
> +		u64			val;
> +		struct {
> +			u32		prod;
> +			u32		cons;
> +		};
> +		struct {
> +			atomic_t	prod;
> +			atomic_t	cons;
> +		} atomic;
> +		u8			__pad[SMP_CACHE_BYTES];
> +	} ____cacheline_aligned_in_smp;
> +	u32				max_n_shift;
> +};
> +
> +struct arm_smmu_queue {
> +	struct arm_smmu_ll_queue	llq;
> +	int				irq; /* Wired interrupt */
> +
> +	__le64				*base;
> +	dma_addr_t			base_dma;
> +	u64				q_base;
> +
> +	size_t				ent_dwords;
> +
> +	u32 __iomem			*prod_reg;
> +	u32 __iomem			*cons_reg;
> +};
> +
> +struct arm_smmu_queue_poll {
> +	ktime_t				timeout;
> +	unsigned int			delay;
> +	unsigned int			spin_cnt;
> +	bool				wfe;
> +};
> +
> +struct arm_smmu_cmdq {
> +	struct arm_smmu_queue		q;
> +	atomic_long_t			*valid_map;
> +	atomic_t			owner_prod;
> +	atomic_t			lock;
> +};
> +
> +struct arm_smmu_cmdq_batch {
> +	u64				cmds[CMDQ_BATCH_ENTRIES * CMDQ_ENT_DWORDS];
> +	int				num;
> +};
> +
> +struct arm_smmu_evtq {
> +	struct arm_smmu_queue		q;
> +	u32				max_stalls;
> +};
> +
> +struct arm_smmu_priq {
> +	struct arm_smmu_queue		q;
> +};
> +
> +/* High-level stream table and context descriptor structures */
> +struct arm_smmu_strtab_l1_desc {
> +	u8				span;
> +
> +	__le64				*l2ptr;
> +	dma_addr_t			l2ptr_dma;
> +};
> +
> +struct arm_smmu_ctx_desc {
> +	u16				asid;
> +	u64				ttbr;
> +	u64				tcr;
> +	u64				mair;
> +};
> +
> +struct arm_smmu_l1_ctx_desc {
> +	__le64				*l2ptr;
> +	dma_addr_t			l2ptr_dma;
> +};
> +
> +struct arm_smmu_ctx_desc_cfg {
> +	__le64				*cdtab;
> +	dma_addr_t			cdtab_dma;
> +	struct arm_smmu_l1_ctx_desc	*l1_desc;
> +	unsigned int			num_l1_ents;
> +};
> +
> +struct arm_smmu_s1_cfg {
> +	struct arm_smmu_ctx_desc_cfg	cdcfg;
> +	struct arm_smmu_ctx_desc	cd;
> +	u8				s1fmt;
> +	u8				s1cdmax;
> +};
> +
> +struct arm_smmu_s2_cfg {
> +	u16				vmid;
> +	u64				vttbr;
> +	u64				vtcr;
> +};
> +
> +struct arm_smmu_strtab_cfg {
> +	__le64				*strtab;
> +	dma_addr_t			strtab_dma;
> +	struct arm_smmu_strtab_l1_desc	*l1_desc;
> +	unsigned int			num_l1_ents;
> +
> +	u64				strtab_base;
> +	u32				strtab_base_cfg;
> +};
> +
> +/* An SMMUv3 instance */
> +struct arm_smmu_device {
> +	struct device			*dev;
> +	void __iomem			*base;
> +	void __iomem			*page1;
> +
> +#define ARM_SMMU_FEAT_2_LVL_STRTAB	(1 << 0)
> +#define ARM_SMMU_FEAT_2_LVL_CDTAB	(1 << 1)
> +#define ARM_SMMU_FEAT_TT_LE		(1 << 2)
> +#define ARM_SMMU_FEAT_TT_BE		(1 << 3)
> +#define ARM_SMMU_FEAT_PRI		(1 << 4)
> +#define ARM_SMMU_FEAT_ATS		(1 << 5)
> +#define ARM_SMMU_FEAT_SEV		(1 << 6)
> +#define ARM_SMMU_FEAT_MSI		(1 << 7)
> +#define ARM_SMMU_FEAT_COHERENCY		(1 << 8)
> +#define ARM_SMMU_FEAT_TRANS_S1		(1 << 9)
> +#define ARM_SMMU_FEAT_TRANS_S2		(1 << 10)
> +#define ARM_SMMU_FEAT_STALLS		(1 << 11)
> +#define ARM_SMMU_FEAT_HYP		(1 << 12)
> +#define ARM_SMMU_FEAT_STALL_FORCE	(1 << 13)
> +#define ARM_SMMU_FEAT_VAX		(1 << 14)
> +#define ARM_SMMU_FEAT_RANGE_INV		(1 << 15)
> +	u32				features;
> +
> +#define ARM_SMMU_OPT_SKIP_PREFETCH	(1 << 0)
> +#define ARM_SMMU_OPT_PAGE0_REGS_ONLY	(1 << 1)
> +	u32				options;
> +
> +	struct arm_smmu_cmdq		cmdq;
> +	struct arm_smmu_evtq		evtq;
> +	struct arm_smmu_priq		priq;
> +
> +	int				gerr_irq;
> +	int				combined_irq;
> +
> +	unsigned long			ias; /* IPA */
> +	unsigned long			oas; /* PA */
> +	unsigned long			pgsize_bitmap;
> +
> +#define ARM_SMMU_MAX_ASIDS		(1 << 16)
> +	unsigned int			asid_bits;
> +
> +#define ARM_SMMU_MAX_VMIDS		(1 << 16)
> +	unsigned int			vmid_bits;
> +	DECLARE_BITMAP(vmid_map, ARM_SMMU_MAX_VMIDS);
> +
> +	unsigned int			ssid_bits;
> +	unsigned int			sid_bits;
> +
> +	struct arm_smmu_strtab_cfg	strtab_cfg;
> +
> +	/* IOMMU core code handle */
> +	struct iommu_device		iommu;
> +};
> +
> +/* SMMU private data for each master */
> +struct arm_smmu_master {
> +	struct arm_smmu_device		*smmu;
> +	struct device			*dev;
> +	struct arm_smmu_domain		*domain;
> +	struct list_head		domain_head;
> +	u32				*sids;
> +	unsigned int			num_sids;
> +	bool				ats_enabled;
> +	unsigned int			ssid_bits;
> +};
> +
> +/* SMMU private data for an IOMMU domain */
> +enum arm_smmu_domain_stage {
> +	ARM_SMMU_DOMAIN_S1 = 0,
> +	ARM_SMMU_DOMAIN_S2,
> +	ARM_SMMU_DOMAIN_NESTED,
> +	ARM_SMMU_DOMAIN_BYPASS,
> +};
> +
> +struct arm_smmu_domain {
> +	struct arm_smmu_device		*smmu;
> +	struct mutex			init_mutex; /* Protects smmu pointer */
> +
> +	struct io_pgtable_ops		*pgtbl_ops;
> +	bool				non_strict;
> +	atomic_t			nr_ats_masters;
> +
> +	enum arm_smmu_domain_stage	stage;
> +	union {
> +		struct arm_smmu_s1_cfg	s1_cfg;
> +		struct arm_smmu_s2_cfg	s2_cfg;
> +	};
> +
> +	struct iommu_domain		domain;
> +
> +	struct list_head		devices;
> +	spinlock_t			devices_lock;
> +};
> +
> +struct arm_smmu_option_prop {
> +	u32 opt;
> +	const char *prop;
> +};
> +
> +static DEFINE_XARRAY_ALLOC1(asid_xa);
> +
> +static struct arm_smmu_option_prop arm_smmu_options[] = {
> +	{ ARM_SMMU_OPT_SKIP_PREFETCH, "hisilicon,broken-prefetch-cmd" },
> +	{ ARM_SMMU_OPT_PAGE0_REGS_ONLY, "cavium,cn9900-broken-page1-regspace"},
> +	{ 0, NULL},
> +};
> +
> +static inline void __iomem *arm_smmu_page1_fixup(unsigned long offset,
> +						 struct arm_smmu_device *smmu)
> +{
> +	if (offset > SZ_64K)
> +		return smmu->page1 + offset - SZ_64K;
> +
> +	return smmu->base + offset;
> +}
> +
> +static struct arm_smmu_domain *to_smmu_domain(struct iommu_domain *dom)
> +{
> +	return container_of(dom, struct arm_smmu_domain, domain);
> +}
> +
> +static void parse_driver_options(struct arm_smmu_device *smmu)
> +{
> +	int i = 0;
> +
> +	do {
> +		if (of_property_read_bool(smmu->dev->of_node,
> +						arm_smmu_options[i].prop)) {
> +			smmu->options |= arm_smmu_options[i].opt;
> +			dev_notice(smmu->dev, "option %s\n",
> +				arm_smmu_options[i].prop);
> +		}
> +	} while (arm_smmu_options[++i].opt);
> +}
> +
> +/* Low-level queue manipulation functions */
> +static bool queue_has_space(struct arm_smmu_ll_queue *q, u32 n)
> +{
> +	u32 space, prod, cons;
> +
> +	prod = Q_IDX(q, q->prod);
> +	cons = Q_IDX(q, q->cons);
> +
> +	if (Q_WRP(q, q->prod) == Q_WRP(q, q->cons))
> +		space = (1 << q->max_n_shift) - (prod - cons);
> +	else
> +		space = cons - prod;
> +
> +	return space >= n;
> +}
> +
> +static bool queue_full(struct arm_smmu_ll_queue *q)
> +{
> +	return Q_IDX(q, q->prod) == Q_IDX(q, q->cons) &&
> +	       Q_WRP(q, q->prod) != Q_WRP(q, q->cons);
> +}
> +
> +static bool queue_empty(struct arm_smmu_ll_queue *q)
> +{
> +	return Q_IDX(q, q->prod) == Q_IDX(q, q->cons) &&
> +	       Q_WRP(q, q->prod) == Q_WRP(q, q->cons);
> +}
> +
> +static bool queue_consumed(struct arm_smmu_ll_queue *q, u32 prod)
> +{
> +	return ((Q_WRP(q, q->cons) == Q_WRP(q, prod)) &&
> +		(Q_IDX(q, q->cons) > Q_IDX(q, prod))) ||
> +	       ((Q_WRP(q, q->cons) != Q_WRP(q, prod)) &&
> +		(Q_IDX(q, q->cons) <= Q_IDX(q, prod)));
> +}
> +
> +static void queue_sync_cons_out(struct arm_smmu_queue *q)
> +{
> +	/*
> +	 * Ensure that all CPU accesses (reads and writes) to the queue
> +	 * are complete before we update the cons pointer.
> +	 */
> +	mb();
> +	writel_relaxed(q->llq.cons, q->cons_reg);
> +}
> +
> +static void queue_inc_cons(struct arm_smmu_ll_queue *q)
> +{
> +	u32 cons = (Q_WRP(q, q->cons) | Q_IDX(q, q->cons)) + 1;
> +	q->cons = Q_OVF(q->cons) | Q_WRP(q, cons) | Q_IDX(q, cons);
> +}
> +
> +static int queue_sync_prod_in(struct arm_smmu_queue *q)
> +{
> +	int ret = 0;
> +	u32 prod = readl_relaxed(q->prod_reg);
> +
> +	if (Q_OVF(prod) != Q_OVF(q->llq.prod))
> +		ret = -EOVERFLOW;
> +
> +	q->llq.prod = prod;
> +	return ret;
> +}
> +
> +static u32 queue_inc_prod_n(struct arm_smmu_ll_queue *q, int n)
> +{
> +	u32 prod = (Q_WRP(q, q->prod) | Q_IDX(q, q->prod)) + n;
> +	return Q_OVF(q->prod) | Q_WRP(q, prod) | Q_IDX(q, prod);
> +}
> +
> +static void queue_poll_init(struct arm_smmu_device *smmu,
> +			    struct arm_smmu_queue_poll *qp)
> +{
> +	qp->delay = 1;
> +	qp->spin_cnt = 0;
> +	qp->wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
> +	qp->timeout = ktime_add_us(ktime_get(), ARM_SMMU_POLL_TIMEOUT_US);
> +}
> +
> +static int queue_poll(struct arm_smmu_queue_poll *qp)
> +{
> +	if (ktime_compare(ktime_get(), qp->timeout) > 0)
> +		return -ETIMEDOUT;
> +
> +	if (qp->wfe) {
> +		wfe();
> +	} else if (++qp->spin_cnt < ARM_SMMU_POLL_SPIN_COUNT) {
> +		cpu_relax();
> +	} else {
> +		udelay(qp->delay);
> +		qp->delay *= 2;
> +		qp->spin_cnt = 0;
> +	}
> +
> +	return 0;
> +}
> +
> +static void queue_write(__le64 *dst, u64 *src, size_t n_dwords)
> +{
> +	int i;
> +
> +	for (i = 0; i < n_dwords; ++i)
> +		*dst++ = cpu_to_le64(*src++);
> +}
> +
> +static void queue_read(__le64 *dst, u64 *src, size_t n_dwords)
> +{
> +	int i;
> +
> +	for (i = 0; i < n_dwords; ++i)
> +		*dst++ = le64_to_cpu(*src++);
> +}
> +
> +static int queue_remove_raw(struct arm_smmu_queue *q, u64 *ent)
> +{
> +	if (queue_empty(&q->llq))
> +		return -EAGAIN;
> +
> +	queue_read(ent, Q_ENT(q, q->llq.cons), q->ent_dwords);
> +	queue_inc_cons(&q->llq);
> +	queue_sync_cons_out(q);
> +	return 0;
> +}
> +
> +/* High-level queue accessors */
> +static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
> +{
> +	memset(cmd, 0, 1 << CMDQ_ENT_SZ_SHIFT);
> +	cmd[0] |= FIELD_PREP(CMDQ_0_OP, ent->opcode);
> +
> +	switch (ent->opcode) {
> +	case CMDQ_OP_TLBI_EL2_ALL:
> +	case CMDQ_OP_TLBI_NSNH_ALL:
> +		break;
> +	case CMDQ_OP_PREFETCH_CFG:
> +		cmd[0] |= FIELD_PREP(CMDQ_PREFETCH_0_SID, ent->prefetch.sid);
> +		cmd[1] |= FIELD_PREP(CMDQ_PREFETCH_1_SIZE, ent->prefetch.size);
> +		cmd[1] |= ent->prefetch.addr & CMDQ_PREFETCH_1_ADDR_MASK;
> +		break;
> +	case CMDQ_OP_CFGI_CD:
> +		cmd[0] |= FIELD_PREP(CMDQ_CFGI_0_SSID, ent->cfgi.ssid);
> +		fallthrough;
> +	case CMDQ_OP_CFGI_STE:
> +		cmd[0] |= FIELD_PREP(CMDQ_CFGI_0_SID, ent->cfgi.sid);
> +		cmd[1] |= FIELD_PREP(CMDQ_CFGI_1_LEAF, ent->cfgi.leaf);
> +		break;
> +	case CMDQ_OP_CFGI_CD_ALL:
> +		cmd[0] |= FIELD_PREP(CMDQ_CFGI_0_SID, ent->cfgi.sid);
> +		break;
> +	case CMDQ_OP_CFGI_ALL:
> +		/* Cover the entire SID range */
> +		cmd[1] |= FIELD_PREP(CMDQ_CFGI_1_RANGE, 31);
> +		break;
> +	case CMDQ_OP_TLBI_NH_VA:
> +		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_NUM, ent->tlbi.num);
> +		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_SCALE, ent->tlbi.scale);
> +		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
> +		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_ASID, ent->tlbi.asid);
> +		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_LEAF, ent->tlbi.leaf);
> +		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TTL, ent->tlbi.ttl);
> +		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TG, ent->tlbi.tg);
> +		cmd[1] |= ent->tlbi.addr & CMDQ_TLBI_1_VA_MASK;
> +		break;
> +	case CMDQ_OP_TLBI_S2_IPA:
> +		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_NUM, ent->tlbi.num);
> +		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_SCALE, ent->tlbi.scale);
> +		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
> +		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_LEAF, ent->tlbi.leaf);
> +		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TTL, ent->tlbi.ttl);
> +		cmd[1] |= FIELD_PREP(CMDQ_TLBI_1_TG, ent->tlbi.tg);
> +		cmd[1] |= ent->tlbi.addr & CMDQ_TLBI_1_IPA_MASK;
> +		break;
> +	case CMDQ_OP_TLBI_NH_ASID:
> +		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_ASID, ent->tlbi.asid);
> +		fallthrough;
> +	case CMDQ_OP_TLBI_S12_VMALL:
> +		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
> +		break;
> +	case CMDQ_OP_ATC_INV:
> +		cmd[0] |= FIELD_PREP(CMDQ_0_SSV, ent->substream_valid);
> +		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_GLOBAL, ent->atc.global);
> +		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_SSID, ent->atc.ssid);
> +		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_SID, ent->atc.sid);
> +		cmd[1] |= FIELD_PREP(CMDQ_ATC_1_SIZE, ent->atc.size);
> +		cmd[1] |= ent->atc.addr & CMDQ_ATC_1_ADDR_MASK;
> +		break;
> +	case CMDQ_OP_PRI_RESP:
> +		cmd[0] |= FIELD_PREP(CMDQ_0_SSV, ent->substream_valid);
> +		cmd[0] |= FIELD_PREP(CMDQ_PRI_0_SSID, ent->pri.ssid);
> +		cmd[0] |= FIELD_PREP(CMDQ_PRI_0_SID, ent->pri.sid);
> +		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_GRPID, ent->pri.grpid);
> +		switch (ent->pri.resp) {
> +		case PRI_RESP_DENY:
> +		case PRI_RESP_FAIL:
> +		case PRI_RESP_SUCC:
> +			break;
> +		default:
> +			return -EINVAL;
> +		}
> +		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_RESP, ent->pri.resp);
> +		break;
> +	case CMDQ_OP_CMD_SYNC:
> +		if (ent->sync.msiaddr) {
> +			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_IRQ);
> +			cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
> +		} else {
> +			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
> +		}
> +		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSH, ARM_SMMU_SH_ISH);
> +		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIATTR, ARM_SMMU_MEMATTR_OIWB);
> +		break;
> +	default:
> +		return -ENOENT;
> +	}
> +
> +	return 0;
> +}
> +
> +static void arm_smmu_cmdq_build_sync_cmd(u64 *cmd, struct arm_smmu_device *smmu,
> +					 u32 prod)
> +{
> +	struct arm_smmu_queue *q = &smmu->cmdq.q;
> +	struct arm_smmu_cmdq_ent ent = {
> +		.opcode = CMDQ_OP_CMD_SYNC,
> +	};
> +
> +	/*
> +	 * Beware that Hi16xx adds an extra 32 bits of goodness to its MSI
> +	 * payload, so the write will zero the entire command on that platform.
> +	 */
> +	if (smmu->features & ARM_SMMU_FEAT_MSI &&
> +	    smmu->features & ARM_SMMU_FEAT_COHERENCY) {
> +		ent.sync.msiaddr = q->base_dma + Q_IDX(&q->llq, prod) *
> +				   q->ent_dwords * 8;
> +	}
> +
> +	arm_smmu_cmdq_build_cmd(cmd, &ent);
> +}
> +
> +static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
> +{
> +	static const char *cerror_str[] = {
> +		[CMDQ_ERR_CERROR_NONE_IDX]	= "No error",
> +		[CMDQ_ERR_CERROR_ILL_IDX]	= "Illegal command",
> +		[CMDQ_ERR_CERROR_ABT_IDX]	= "Abort on command fetch",
> +		[CMDQ_ERR_CERROR_ATC_INV_IDX]	= "ATC invalidate timeout",
> +	};
> +
> +	int i;
> +	u64 cmd[CMDQ_ENT_DWORDS];
> +	struct arm_smmu_queue *q = &smmu->cmdq.q;
> +	u32 cons = readl_relaxed(q->cons_reg);
> +	u32 idx = FIELD_GET(CMDQ_CONS_ERR, cons);
> +	struct arm_smmu_cmdq_ent cmd_sync = {
> +		.opcode = CMDQ_OP_CMD_SYNC,
> +	};
> +
> +	dev_err(smmu->dev, "CMDQ error (cons 0x%08x): %s\n", cons,
> +		idx < ARRAY_SIZE(cerror_str) ?  cerror_str[idx] : "Unknown");
> +
> +	switch (idx) {
> +	case CMDQ_ERR_CERROR_ABT_IDX:
> +		dev_err(smmu->dev, "retrying command fetch\n");
> +	case CMDQ_ERR_CERROR_NONE_IDX:
> +		return;
> +	case CMDQ_ERR_CERROR_ATC_INV_IDX:
> +		/*
> +		 * ATC Invalidation Completion timeout. CONS is still pointing
> +		 * at the CMD_SYNC. Attempt to complete other pending commands
> +		 * by repeating the CMD_SYNC, though we might well end up back
> +		 * here since the ATC invalidation may still be pending.
> +		 */
> +		return;
> +	case CMDQ_ERR_CERROR_ILL_IDX:
> +	default:
> +		break;
> +	}
> +
> +	/*
> +	 * We may have concurrent producers, so we need to be careful
> +	 * not to touch any of the shadow cmdq state.
> +	 */
> +	queue_read(cmd, Q_ENT(q, cons), q->ent_dwords);
> +	dev_err(smmu->dev, "skipping command in error state:\n");
> +	for (i = 0; i < ARRAY_SIZE(cmd); ++i)
> +		dev_err(smmu->dev, "\t0x%016llx\n", (unsigned long long)cmd[i]);
> +
> +	/* Convert the erroneous command into a CMD_SYNC */
> +	if (arm_smmu_cmdq_build_cmd(cmd, &cmd_sync)) {
> +		dev_err(smmu->dev, "failed to convert to CMD_SYNC\n");
> +		return;
> +	}
> +
> +	queue_write(Q_ENT(q, cons), cmd, q->ent_dwords);
> +}
> +
> +/*
> + * Command queue locking.
> + * This is a form of bastardised rwlock with the following major changes:
> + *
> + * - The only LOCK routines are exclusive_trylock() and shared_lock().
> + *   Neither have barrier semantics, and instead provide only a control
> + *   dependency.
> + *
> + * - The UNLOCK routines are supplemented with shared_tryunlock(), which
> + *   fails if the caller appears to be the last lock holder (yes, this is
> + *   racy). All successful UNLOCK routines have RELEASE semantics.
> + */
> +static void arm_smmu_cmdq_shared_lock(struct arm_smmu_cmdq *cmdq)
> +{
> +	int val;
> +
> +	/*
> +	 * We can try to avoid the cmpxchg() loop by simply incrementing the
> +	 * lock counter. When held in exclusive state, the lock counter is set
> +	 * to INT_MIN so these increments won't hurt as the value will remain
> +	 * negative.
> +	 */
> +	if (atomic_fetch_inc_relaxed(&cmdq->lock) >= 0)
> +		return;
> +
> +	do {
> +		val = atomic_cond_read_relaxed(&cmdq->lock, VAL >= 0);
> +	} while (atomic_cmpxchg_relaxed(&cmdq->lock, val, val + 1) != val);
> +}
> +
> +static void arm_smmu_cmdq_shared_unlock(struct arm_smmu_cmdq *cmdq)
> +{
> +	(void)atomic_dec_return_release(&cmdq->lock);
> +}
> +
> +static bool arm_smmu_cmdq_shared_tryunlock(struct arm_smmu_cmdq *cmdq)
> +{
> +	if (atomic_read(&cmdq->lock) == 1)
> +		return false;
> +
> +	arm_smmu_cmdq_shared_unlock(cmdq);
> +	return true;
> +}
> +
> +#define arm_smmu_cmdq_exclusive_trylock_irqsave(cmdq, flags)		\
> +({									\
> +	bool __ret;							\
> +	local_irq_save(flags);						\
> +	__ret = !atomic_cmpxchg_relaxed(&cmdq->lock, 0, INT_MIN);	\
> +	if (!__ret)							\
> +		local_irq_restore(flags);				\
> +	__ret;								\
> +})
> +
> +#define arm_smmu_cmdq_exclusive_unlock_irqrestore(cmdq, flags)		\
> +({									\
> +	atomic_set_release(&cmdq->lock, 0);				\
> +	local_irq_restore(flags);					\
> +})
> +
> +
> +/*
> + * Command queue insertion.
> + * This is made fiddly by our attempts to achieve some sort of scalability
> + * since there is one queue shared amongst all of the CPUs in the system.  If
> + * you like mixed-size concurrency, dependency ordering and relaxed atomics,
> + * then you'll *love* this monstrosity.
> + *
> + * The basic idea is to split the queue up into ranges of commands that are
> + * owned by a given CPU; the owner may not have written all of the commands
> + * itself, but is responsible for advancing the hardware prod pointer when
> + * the time comes. The algorithm is roughly:
> + *
> + * 	1. Allocate some space in the queue. At this point we also discover
> + *	   whether the head of the queue is currently owned by another CPU,
> + *	   or whether we are the owner.
> + *
> + *	2. Write our commands into our allocated slots in the queue.
> + *
> + *	3. Mark our slots as valid in arm_smmu_cmdq.valid_map.
> + *
> + *	4. If we are an owner:
> + *		a. Wait for the previous owner to finish.
> + *		b. Mark the queue head as unowned, which tells us the range
> + *		   that we are responsible for publishing.
> + *		c. Wait for all commands in our owned range to become valid.
> + *		d. Advance the hardware prod pointer.
> + *		e. Tell the next owner we've finished.
> + *
> + *	5. If we are inserting a CMD_SYNC (we may or may not have been an
> + *	   owner), then we need to stick around until it has completed:
> + *		a. If we have MSIs, the SMMU can write back into the CMD_SYNC
> + *		   to clear the first 4 bytes.
> + *		b. Otherwise, we spin waiting for the hardware cons pointer to
> + *		   advance past our command.
> + *
> + * The devil is in the details, particularly the use of locking for handling
> + * SYNC completion and freeing up space in the queue before we think that it is
> + * full.
> + */
> +static void __arm_smmu_cmdq_poll_set_valid_map(struct arm_smmu_cmdq *cmdq,
> +					       u32 sprod, u32 eprod, bool set)
> +{
> +	u32 swidx, sbidx, ewidx, ebidx;
> +	struct arm_smmu_ll_queue llq = {
> +		.max_n_shift	= cmdq->q.llq.max_n_shift,
> +		.prod		= sprod,
> +	};
> +
> +	ewidx = BIT_WORD(Q_IDX(&llq, eprod));
> +	ebidx = Q_IDX(&llq, eprod) % BITS_PER_LONG;
> +
> +	while (llq.prod != eprod) {
> +		unsigned long mask;
> +		atomic_long_t *ptr;
> +		u32 limit = BITS_PER_LONG;
> +
> +		swidx = BIT_WORD(Q_IDX(&llq, llq.prod));
> +		sbidx = Q_IDX(&llq, llq.prod) % BITS_PER_LONG;
> +
> +		ptr = &cmdq->valid_map[swidx];
> +
> +		if ((swidx == ewidx) && (sbidx < ebidx))
> +			limit = ebidx;
> +
> +		mask = GENMASK(limit - 1, sbidx);
> +
> +		/*
> +		 * The valid bit is the inverse of the wrap bit. This means
> +		 * that a zero-initialised queue is invalid and, after marking
> +		 * all entries as valid, they become invalid again when we
> +		 * wrap.
> +		 */
> +		if (set) {
> +			atomic_long_xor(mask, ptr);
> +		} else { /* Poll */
> +			unsigned long valid;
> +
> +			valid = (ULONG_MAX + !!Q_WRP(&llq, llq.prod)) & mask;
> +			atomic_long_cond_read_relaxed(ptr, (VAL & mask) == valid);
> +		}
> +
> +		llq.prod = queue_inc_prod_n(&llq, limit - sbidx);
> +	}
> +}
> +
> +/* Mark all entries in the range [sprod, eprod) as valid */
> +static void arm_smmu_cmdq_set_valid_map(struct arm_smmu_cmdq *cmdq,
> +					u32 sprod, u32 eprod)
> +{
> +	__arm_smmu_cmdq_poll_set_valid_map(cmdq, sprod, eprod, true);
> +}
> +
> +/* Wait for all entries in the range [sprod, eprod) to become valid */
> +static void arm_smmu_cmdq_poll_valid_map(struct arm_smmu_cmdq *cmdq,
> +					 u32 sprod, u32 eprod)
> +{
> +	__arm_smmu_cmdq_poll_set_valid_map(cmdq, sprod, eprod, false);
> +}
> +
> +/* Wait for the command queue to become non-full */
> +static int arm_smmu_cmdq_poll_until_not_full(struct arm_smmu_device *smmu,
> +					     struct arm_smmu_ll_queue *llq)
> +{
> +	unsigned long flags;
> +	struct arm_smmu_queue_poll qp;
> +	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
> +	int ret = 0;
> +
> +	/*
> +	 * Try to update our copy of cons by grabbing exclusive cmdq access. If
> +	 * that fails, spin until somebody else updates it for us.
> +	 */
> +	if (arm_smmu_cmdq_exclusive_trylock_irqsave(cmdq, flags)) {
> +		WRITE_ONCE(cmdq->q.llq.cons, readl_relaxed(cmdq->q.cons_reg));
> +		arm_smmu_cmdq_exclusive_unlock_irqrestore(cmdq, flags);
> +		llq->val = READ_ONCE(cmdq->q.llq.val);
> +		return 0;
> +	}
> +
> +	queue_poll_init(smmu, &qp);
> +	do {
> +		llq->val = READ_ONCE(smmu->cmdq.q.llq.val);
> +		if (!queue_full(llq))
> +			break;
> +
> +		ret = queue_poll(&qp);
> +	} while (!ret);
> +
> +	return ret;
> +}
> +
> +/*
> + * Wait until the SMMU signals a CMD_SYNC completion MSI.
> + * Must be called with the cmdq lock held in some capacity.
> + */
> +static int __arm_smmu_cmdq_poll_until_msi(struct arm_smmu_device *smmu,
> +					  struct arm_smmu_ll_queue *llq)
> +{
> +	int ret = 0;
> +	struct arm_smmu_queue_poll qp;
> +	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
> +	u32 *cmd = (u32 *)(Q_ENT(&cmdq->q, llq->prod));
> +
> +	queue_poll_init(smmu, &qp);
> +
> +	/*
> +	 * The MSI won't generate an event, since it's being written back
> +	 * into the command queue.
> +	 */
> +	qp.wfe = false;
> +	smp_cond_load_relaxed(cmd, !VAL || (ret = queue_poll(&qp)));
> +	llq->cons = ret ? llq->prod : queue_inc_prod_n(llq, 1);
> +	return ret;
> +}
> +
> +/*
> + * Wait until the SMMU cons index passes llq->prod.
> + * Must be called with the cmdq lock held in some capacity.
> + */
> +static int __arm_smmu_cmdq_poll_until_consumed(struct arm_smmu_device *smmu,
> +					       struct arm_smmu_ll_queue *llq)
> +{
> +	struct arm_smmu_queue_poll qp;
> +	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
> +	u32 prod = llq->prod;
> +	int ret = 0;
> +
> +	queue_poll_init(smmu, &qp);
> +	llq->val = READ_ONCE(smmu->cmdq.q.llq.val);
> +	do {
> +		if (queue_consumed(llq, prod))
> +			break;
> +
> +		ret = queue_poll(&qp);
> +
> +		/*
> +		 * This needs to be a readl() so that our subsequent call
> +		 * to arm_smmu_cmdq_shared_tryunlock() can fail accurately.
> +		 *
> +		 * Specifically, we need to ensure that we observe all
> +		 * shared_lock()s by other CMD_SYNCs that share our owner,
> +		 * so that a failing call to tryunlock() means that we're
> +		 * the last one out and therefore we can safely advance
> +		 * cmdq->q.llq.cons. Roughly speaking:
> +		 *
> +		 * CPU 0		CPU1			CPU2 (us)
> +		 *
> +		 * if (sync)
> +		 * 	shared_lock();
> +		 *
> +		 * dma_wmb();
> +		 * set_valid_map();
> +		 *
> +		 * 			if (owner) {
> +		 *				poll_valid_map();
> +		 *				<control dependency>
> +		 *				writel(prod_reg);
> +		 *
> +		 *						readl(cons_reg);
> +		 *						tryunlock();
> +		 *
> +		 * Requires us to see CPU 0's shared_lock() acquisition.
> +		 */
> +		llq->cons = readl(cmdq->q.cons_reg);
> +	} while (!ret);
> +
> +	return ret;
> +}
> +
> +static int arm_smmu_cmdq_poll_until_sync(struct arm_smmu_device *smmu,
> +					 struct arm_smmu_ll_queue *llq)
> +{
> +	if (smmu->features & ARM_SMMU_FEAT_MSI &&
> +	    smmu->features & ARM_SMMU_FEAT_COHERENCY)
> +		return __arm_smmu_cmdq_poll_until_msi(smmu, llq);
> +
> +	return __arm_smmu_cmdq_poll_until_consumed(smmu, llq);
> +}
> +
> +static void arm_smmu_cmdq_write_entries(struct arm_smmu_cmdq *cmdq, u64 *cmds,
> +					u32 prod, int n)
> +{
> +	int i;
> +	struct arm_smmu_ll_queue llq = {
> +		.max_n_shift	= cmdq->q.llq.max_n_shift,
> +		.prod		= prod,
> +	};
> +
> +	for (i = 0; i < n; ++i) {
> +		u64 *cmd = &cmds[i * CMDQ_ENT_DWORDS];
> +
> +		prod = queue_inc_prod_n(&llq, i);
> +		queue_write(Q_ENT(&cmdq->q, prod), cmd, CMDQ_ENT_DWORDS);
> +	}
> +}
> +
> +/*
> + * This is the actual insertion function, and provides the following
> + * ordering guarantees to callers:
> + *
> + * - There is a dma_wmb() before publishing any commands to the queue.
> + *   This can be relied upon to order prior writes to data structures
> + *   in memory (such as a CD or an STE) before the command.
> + *
> + * - On completion of a CMD_SYNC, there is a control dependency.
> + *   This can be relied upon to order subsequent writes to memory (e.g.
> + *   freeing an IOVA) after completion of the CMD_SYNC.
> + *
> + * - Command insertion is totally ordered, so if two CPUs each race to
> + *   insert their own list of commands then all of the commands from one
> + *   CPU will appear before any of the commands from the other CPU.
> + */
> +static int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
> +				       u64 *cmds, int n, bool sync)
> +{
> +	u64 cmd_sync[CMDQ_ENT_DWORDS];
> +	u32 prod;
> +	unsigned long flags;
> +	bool owner;
> +	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
> +	struct arm_smmu_ll_queue llq = {
> +		.max_n_shift = cmdq->q.llq.max_n_shift,
> +	}, head = llq;
> +	int ret = 0;
> +
> +	/* 1. Allocate some space in the queue */
> +	local_irq_save(flags);
> +	llq.val = READ_ONCE(cmdq->q.llq.val);
> +	do {
> +		u64 old;
> +
> +		while (!queue_has_space(&llq, n + sync)) {
> +			local_irq_restore(flags);
> +			if (arm_smmu_cmdq_poll_until_not_full(smmu, &llq))
> +				dev_err_ratelimited(smmu->dev, "CMDQ timeout\n");
> +			local_irq_save(flags);
> +		}
> +
> +		head.cons = llq.cons;
> +		head.prod = queue_inc_prod_n(&llq, n + sync) |
> +					     CMDQ_PROD_OWNED_FLAG;
> +
> +		old = cmpxchg_relaxed(&cmdq->q.llq.val, llq.val, head.val);
> +		if (old == llq.val)
> +			break;
> +
> +		llq.val = old;
> +	} while (1);
> +	owner = !(llq.prod & CMDQ_PROD_OWNED_FLAG);
> +	head.prod &= ~CMDQ_PROD_OWNED_FLAG;
> +	llq.prod &= ~CMDQ_PROD_OWNED_FLAG;
> +
> +	/*
> +	 * 2. Write our commands into the queue
> +	 * Dependency ordering from the cmpxchg() loop above.
> +	 */
> +	arm_smmu_cmdq_write_entries(cmdq, cmds, llq.prod, n);
> +	if (sync) {
> +		prod = queue_inc_prod_n(&llq, n);
> +		arm_smmu_cmdq_build_sync_cmd(cmd_sync, smmu, prod);
> +		queue_write(Q_ENT(&cmdq->q, prod), cmd_sync, CMDQ_ENT_DWORDS);
> +
> +		/*
> +		 * In order to determine completion of our CMD_SYNC, we must
> +		 * ensure that the queue can't wrap twice without us noticing.
> +		 * We achieve that by taking the cmdq lock as shared before
> +		 * marking our slot as valid.
> +		 */
> +		arm_smmu_cmdq_shared_lock(cmdq);
> +	}
> +
> +	/* 3. Mark our slots as valid, ensuring commands are visible first */
> +	dma_wmb();
> +	arm_smmu_cmdq_set_valid_map(cmdq, llq.prod, head.prod);
> +
> +	/* 4. If we are the owner, take control of the SMMU hardware */
> +	if (owner) {
> +		/* a. Wait for previous owner to finish */
> +		atomic_cond_read_relaxed(&cmdq->owner_prod, VAL == llq.prod);
> +
> +		/* b. Stop gathering work by clearing the owned flag */
> +		prod = atomic_fetch_andnot_relaxed(CMDQ_PROD_OWNED_FLAG,
> +						   &cmdq->q.llq.atomic.prod);
> +		prod &= ~CMDQ_PROD_OWNED_FLAG;
> +
> +		/*
> +		 * c. Wait for any gathered work to be written to the queue.
> +		 * Note that we read our own entries so that we have the control
> +		 * dependency required by (d).
> +		 */
> +		arm_smmu_cmdq_poll_valid_map(cmdq, llq.prod, prod);
> +
> +		/*
> +		 * d. Advance the hardware prod pointer
> +		 * Control dependency ordering from the entries becoming valid.
> +		 */
> +		writel_relaxed(prod, cmdq->q.prod_reg);
> +
> +		/*
> +		 * e. Tell the next owner we're done
> +		 * Make sure we've updated the hardware first, so that we don't
> +		 * race to update prod and potentially move it backwards.
> +		 */
> +		atomic_set_release(&cmdq->owner_prod, prod);
> +	}
> +
> +	/* 5. If we are inserting a CMD_SYNC, we must wait for it to complete */
> +	if (sync) {
> +		llq.prod = queue_inc_prod_n(&llq, n);
> +		ret = arm_smmu_cmdq_poll_until_sync(smmu, &llq);
> +		if (ret) {
> +			dev_err_ratelimited(smmu->dev,
> +					    "CMD_SYNC timeout at 0x%08x [hwprod 0x%08x, hwcons 0x%08x]\n",
> +					    llq.prod,
> +					    readl_relaxed(cmdq->q.prod_reg),
> +					    readl_relaxed(cmdq->q.cons_reg));
> +		}
> +
> +		/*
> +		 * Try to unlock the cmdq lock. This will fail if we're the last
> +		 * reader, in which case we can safely update cmdq->q.llq.cons
> +		 */
> +		if (!arm_smmu_cmdq_shared_tryunlock(cmdq)) {
> +			WRITE_ONCE(cmdq->q.llq.cons, llq.cons);
> +			arm_smmu_cmdq_shared_unlock(cmdq);
> +		}
> +	}
> +
> +	local_irq_restore(flags);
> +	return ret;
> +}
> +
> +static int arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
> +				   struct arm_smmu_cmdq_ent *ent)
> +{
> +	u64 cmd[CMDQ_ENT_DWORDS];
> +
> +	if (arm_smmu_cmdq_build_cmd(cmd, ent)) {
> +		dev_warn(smmu->dev, "ignoring unknown CMDQ opcode 0x%x\n",
> +			 ent->opcode);
> +		return -EINVAL;
> +	}
> +
> +	return arm_smmu_cmdq_issue_cmdlist(smmu, cmd, 1, false);
> +}
> +
> +static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
> +{
> +	return arm_smmu_cmdq_issue_cmdlist(smmu, NULL, 0, true);
> +}
> +
> +static void arm_smmu_cmdq_batch_add(struct arm_smmu_device *smmu,
> +				    struct arm_smmu_cmdq_batch *cmds,
> +				    struct arm_smmu_cmdq_ent *cmd)
> +{
> +	if (cmds->num == CMDQ_BATCH_ENTRIES) {
> +		arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, false);
> +		cmds->num = 0;
> +	}
> +	arm_smmu_cmdq_build_cmd(&cmds->cmds[cmds->num * CMDQ_ENT_DWORDS], cmd);
> +	cmds->num++;
> +}
> +
> +static int arm_smmu_cmdq_batch_submit(struct arm_smmu_device *smmu,
> +				      struct arm_smmu_cmdq_batch *cmds)
> +{
> +	return arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, true);
> +}
> +
> +/* Context descriptor manipulation functions */
> +static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
> +			     int ssid, bool leaf)
> +{
> +	size_t i;
> +	unsigned long flags;
> +	struct arm_smmu_master *master;
> +	struct arm_smmu_cmdq_batch cmds = {};
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +	struct arm_smmu_cmdq_ent cmd = {
> +		.opcode	= CMDQ_OP_CFGI_CD,
> +		.cfgi	= {
> +			.ssid	= ssid,
> +			.leaf	= leaf,
> +		},
> +	};
> +
> +	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
> +	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
> +		for (i = 0; i < master->num_sids; i++) {
> +			cmd.cfgi.sid = master->sids[i];
> +			arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
> +		}
> +	}
> +	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
> +
> +	arm_smmu_cmdq_batch_submit(smmu, &cmds);
> +}
> +
> +static int arm_smmu_alloc_cd_leaf_table(struct arm_smmu_device *smmu,
> +					struct arm_smmu_l1_ctx_desc *l1_desc)
> +{
> +	size_t size = CTXDESC_L2_ENTRIES * (CTXDESC_CD_DWORDS << 3);
> +
> +	l1_desc->l2ptr = dmam_alloc_coherent(smmu->dev, size,
> +					     &l1_desc->l2ptr_dma, GFP_KERNEL);
> +	if (!l1_desc->l2ptr) {
> +		dev_warn(smmu->dev,
> +			 "failed to allocate context descriptor table\n");
> +		return -ENOMEM;
> +	}
> +	return 0;
> +}
> +
> +static void arm_smmu_write_cd_l1_desc(__le64 *dst,
> +				      struct arm_smmu_l1_ctx_desc *l1_desc)
> +{
> +	u64 val = (l1_desc->l2ptr_dma & CTXDESC_L1_DESC_L2PTR_MASK) |
> +		  CTXDESC_L1_DESC_V;
> +
> +	/* See comment in arm_smmu_write_ctx_desc() */
> +	WRITE_ONCE(*dst, cpu_to_le64(val));
> +}
> +
> +static __le64 *arm_smmu_get_cd_ptr(struct arm_smmu_domain *smmu_domain,
> +				   u32 ssid)
> +{
> +	__le64 *l1ptr;
> +	unsigned int idx;
> +	struct arm_smmu_l1_ctx_desc *l1_desc;
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +	struct arm_smmu_ctx_desc_cfg *cdcfg = &smmu_domain->s1_cfg.cdcfg;
> +
> +	if (smmu_domain->s1_cfg.s1fmt == STRTAB_STE_0_S1FMT_LINEAR)
> +		return cdcfg->cdtab + ssid * CTXDESC_CD_DWORDS;
> +
> +	idx = ssid >> CTXDESC_SPLIT;
> +	l1_desc = &cdcfg->l1_desc[idx];
> +	if (!l1_desc->l2ptr) {
> +		if (arm_smmu_alloc_cd_leaf_table(smmu, l1_desc))
> +			return NULL;
> +
> +		l1ptr = cdcfg->cdtab + idx * CTXDESC_L1_DESC_DWORDS;
> +		arm_smmu_write_cd_l1_desc(l1ptr, l1_desc);
> +		/* An invalid L1CD can be cached */
> +		arm_smmu_sync_cd(smmu_domain, ssid, false);
> +	}
> +	idx = ssid & (CTXDESC_L2_ENTRIES - 1);
> +	return l1_desc->l2ptr + idx * CTXDESC_CD_DWORDS;
> +}
> +
> +static int arm_smmu_write_ctx_desc(struct arm_smmu_domain *smmu_domain,
> +				   int ssid, struct arm_smmu_ctx_desc *cd)
> +{
> +	/*
> +	 * This function handles the following cases:
> +	 *
> +	 * (1) Install primary CD, for normal DMA traffic (SSID = 0).
> +	 * (2) Install a secondary CD, for SID+SSID traffic.
> +	 * (3) Update ASID of a CD. Atomically write the first 64 bits of the
> +	 *     CD, then invalidate the old entry and mappings.
> +	 * (4) Remove a secondary CD.
> +	 */
> +	u64 val;
> +	bool cd_live;
> +	__le64 *cdptr;
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +
> +	if (WARN_ON(ssid >= (1 << smmu_domain->s1_cfg.s1cdmax)))
> +		return -E2BIG;
> +
> +	cdptr = arm_smmu_get_cd_ptr(smmu_domain, ssid);
> +	if (!cdptr)
> +		return -ENOMEM;
> +
> +	val = le64_to_cpu(cdptr[0]);
> +	cd_live = !!(val & CTXDESC_CD_0_V);
> +
> +	if (!cd) { /* (4) */
> +		val = 0;
> +	} else if (cd_live) { /* (3) */
> +		val &= ~CTXDESC_CD_0_ASID;
> +		val |= FIELD_PREP(CTXDESC_CD_0_ASID, cd->asid);
> +		/*
> +		 * Until CD+TLB invalidation, both ASIDs may be used for tagging
> +		 * this substream's traffic
> +		 */
> +	} else { /* (1) and (2) */
> +		cdptr[1] = cpu_to_le64(cd->ttbr & CTXDESC_CD_1_TTB0_MASK);
> +		cdptr[2] = 0;
> +		cdptr[3] = cpu_to_le64(cd->mair);
> +
> +		/*
> +		 * STE is live, and the SMMU might read dwords of this CD in any
> +		 * order. Ensure that it observes valid values before reading
> +		 * V=1.
> +		 */
> +		arm_smmu_sync_cd(smmu_domain, ssid, true);
> +
> +		val = cd->tcr |
> +#ifdef __BIG_ENDIAN
> +			CTXDESC_CD_0_ENDI |
> +#endif
> +			CTXDESC_CD_0_R | CTXDESC_CD_0_A | CTXDESC_CD_0_ASET |
> +			CTXDESC_CD_0_AA64 |
> +			FIELD_PREP(CTXDESC_CD_0_ASID, cd->asid) |
> +			CTXDESC_CD_0_V;
> +
> +		/* STALL_MODEL==0b10 && CD.S==0 is ILLEGAL */
> +		if (smmu->features & ARM_SMMU_FEAT_STALL_FORCE)
> +			val |= CTXDESC_CD_0_S;
> +	}
> +
> +	/*
> +	 * The SMMU accesses 64-bit values atomically. See IHI0070Ca 3.21.3
> +	 * "Configuration structures and configuration invalidation completion"
> +	 *
> +	 *   The size of single-copy atomic reads made by the SMMU is
> +	 *   IMPLEMENTATION DEFINED but must be at least 64 bits. Any single
> +	 *   field within an aligned 64-bit span of a structure can be altered
> +	 *   without first making the structure invalid.
> +	 */
> +	WRITE_ONCE(cdptr[0], cpu_to_le64(val));
> +	arm_smmu_sync_cd(smmu_domain, ssid, true);
> +	return 0;
> +}
> +
> +static int arm_smmu_alloc_cd_tables(struct arm_smmu_domain *smmu_domain)
> +{
> +	int ret;
> +	size_t l1size;
> +	size_t max_contexts;
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
> +	struct arm_smmu_ctx_desc_cfg *cdcfg = &cfg->cdcfg;
> +
> +	max_contexts = 1 << cfg->s1cdmax;
> +
> +	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB) ||
> +	    max_contexts <= CTXDESC_L2_ENTRIES) {
> +		cfg->s1fmt = STRTAB_STE_0_S1FMT_LINEAR;
> +		cdcfg->num_l1_ents = max_contexts;
> +
> +		l1size = max_contexts * (CTXDESC_CD_DWORDS << 3);
> +	} else {
> +		cfg->s1fmt = STRTAB_STE_0_S1FMT_64K_L2;
> +		cdcfg->num_l1_ents = DIV_ROUND_UP(max_contexts,
> +						  CTXDESC_L2_ENTRIES);
> +
> +		cdcfg->l1_desc = devm_kcalloc(smmu->dev, cdcfg->num_l1_ents,
> +					      sizeof(*cdcfg->l1_desc),
> +					      GFP_KERNEL);
> +		if (!cdcfg->l1_desc)
> +			return -ENOMEM;
> +
> +		l1size = cdcfg->num_l1_ents * (CTXDESC_L1_DESC_DWORDS << 3);
> +	}
> +
> +	cdcfg->cdtab = dmam_alloc_coherent(smmu->dev, l1size, &cdcfg->cdtab_dma,
> +					   GFP_KERNEL);
> +	if (!cdcfg->cdtab) {
> +		dev_warn(smmu->dev, "failed to allocate context descriptor\n");
> +		ret = -ENOMEM;
> +		goto err_free_l1;
> +	}
> +
> +	return 0;
> +
> +err_free_l1:
> +	if (cdcfg->l1_desc) {
> +		devm_kfree(smmu->dev, cdcfg->l1_desc);
> +		cdcfg->l1_desc = NULL;
> +	}
> +	return ret;
> +}
> +
> +static void arm_smmu_free_cd_tables(struct arm_smmu_domain *smmu_domain)
> +{
> +	int i;
> +	size_t size, l1size;
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +	struct arm_smmu_ctx_desc_cfg *cdcfg = &smmu_domain->s1_cfg.cdcfg;
> +
> +	if (cdcfg->l1_desc) {
> +		size = CTXDESC_L2_ENTRIES * (CTXDESC_CD_DWORDS << 3);
> +
> +		for (i = 0; i < cdcfg->num_l1_ents; i++) {
> +			if (!cdcfg->l1_desc[i].l2ptr)
> +				continue;
> +
> +			dmam_free_coherent(smmu->dev, size,
> +					   cdcfg->l1_desc[i].l2ptr,
> +					   cdcfg->l1_desc[i].l2ptr_dma);
> +		}
> +		devm_kfree(smmu->dev, cdcfg->l1_desc);
> +		cdcfg->l1_desc = NULL;
> +
> +		l1size = cdcfg->num_l1_ents * (CTXDESC_L1_DESC_DWORDS << 3);
> +	} else {
> +		l1size = cdcfg->num_l1_ents * (CTXDESC_CD_DWORDS << 3);
> +	}
> +
> +	dmam_free_coherent(smmu->dev, l1size, cdcfg->cdtab, cdcfg->cdtab_dma);
> +	cdcfg->cdtab_dma = 0;
> +	cdcfg->cdtab = NULL;
> +}
> +
> +static void arm_smmu_free_asid(struct arm_smmu_ctx_desc *cd)
> +{
> +	if (!cd->asid)
> +		return;
> +
> +	xa_erase(&asid_xa, cd->asid);
> +}
> +
> +/* Stream table manipulation functions */
> +static void
> +arm_smmu_write_strtab_l1_desc(__le64 *dst, struct arm_smmu_strtab_l1_desc *desc)
> +{
> +	u64 val = 0;
> +
> +	val |= FIELD_PREP(STRTAB_L1_DESC_SPAN, desc->span);
> +	val |= desc->l2ptr_dma & STRTAB_L1_DESC_L2PTR_MASK;
> +
> +	/* See comment in arm_smmu_write_ctx_desc() */
> +	WRITE_ONCE(*dst, cpu_to_le64(val));
> +}
> +
> +static void arm_smmu_sync_ste_for_sid(struct arm_smmu_device *smmu, u32 sid)
> +{
> +	struct arm_smmu_cmdq_ent cmd = {
> +		.opcode	= CMDQ_OP_CFGI_STE,
> +		.cfgi	= {
> +			.sid	= sid,
> +			.leaf	= true,
> +		},
> +	};
> +
> +	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
> +	arm_smmu_cmdq_issue_sync(smmu);
> +}
> +
> +static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
> +				      __le64 *dst)
> +{
> +	/*
> +	 * This is hideously complicated, but we only really care about
> +	 * three cases at the moment:
> +	 *
> +	 * 1. Invalid (all zero) -> bypass/fault (init)
> +	 * 2. Bypass/fault -> translation/bypass (attach)
> +	 * 3. Translation/bypass -> bypass/fault (detach)
> +	 *
> +	 * Given that we can't update the STE atomically and the SMMU
> +	 * doesn't read the thing in a defined order, that leaves us
> +	 * with the following maintenance requirements:
> +	 *
> +	 * 1. Update Config, return (init time STEs aren't live)
> +	 * 2. Write everything apart from dword 0, sync, write dword 0, sync
> +	 * 3. Update Config, sync
> +	 */
> +	u64 val = le64_to_cpu(dst[0]);
> +	bool ste_live = false;
> +	struct arm_smmu_device *smmu = NULL;
> +	struct arm_smmu_s1_cfg *s1_cfg = NULL;
> +	struct arm_smmu_s2_cfg *s2_cfg = NULL;
> +	struct arm_smmu_domain *smmu_domain = NULL;
> +	struct arm_smmu_cmdq_ent prefetch_cmd = {
> +		.opcode		= CMDQ_OP_PREFETCH_CFG,
> +		.prefetch	= {
> +			.sid	= sid,
> +		},
> +	};
> +
> +	if (master) {
> +		smmu_domain = master->domain;
> +		smmu = master->smmu;
> +	}
> +
> +	if (smmu_domain) {
> +		switch (smmu_domain->stage) {
> +		case ARM_SMMU_DOMAIN_S1:
> +			s1_cfg = &smmu_domain->s1_cfg;
> +			break;
> +		case ARM_SMMU_DOMAIN_S2:
> +		case ARM_SMMU_DOMAIN_NESTED:
> +			s2_cfg = &smmu_domain->s2_cfg;
> +			break;
> +		default:
> +			break;
> +		}
> +	}
> +
> +	if (val & STRTAB_STE_0_V) {
> +		switch (FIELD_GET(STRTAB_STE_0_CFG, val)) {
> +		case STRTAB_STE_0_CFG_BYPASS:
> +			break;
> +		case STRTAB_STE_0_CFG_S1_TRANS:
> +		case STRTAB_STE_0_CFG_S2_TRANS:
> +			ste_live = true;
> +			break;
> +		case STRTAB_STE_0_CFG_ABORT:
> +			BUG_ON(!disable_bypass);
> +			break;
> +		default:
> +			BUG(); /* STE corruption */
> +		}
> +	}
> +
> +	/* Nuke the existing STE_0 value, as we're going to rewrite it */
> +	val = STRTAB_STE_0_V;
> +
> +	/* Bypass/fault */
> +	if (!smmu_domain || !(s1_cfg || s2_cfg)) {
> +		if (!smmu_domain && disable_bypass)
> +			val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_ABORT);
> +		else
> +			val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_BYPASS);
> +
> +		dst[0] = cpu_to_le64(val);
> +		dst[1] = cpu_to_le64(FIELD_PREP(STRTAB_STE_1_SHCFG,
> +						STRTAB_STE_1_SHCFG_INCOMING));
> +		dst[2] = 0; /* Nuke the VMID */
> +		/*
> +		 * The SMMU can perform negative caching, so we must sync
> +		 * the STE regardless of whether the old value was live.
> +		 */
> +		if (smmu)
> +			arm_smmu_sync_ste_for_sid(smmu, sid);
> +		return;
> +	}
> +
> +	if (s1_cfg) {
> +		BUG_ON(ste_live);
> +		dst[1] = cpu_to_le64(
> +			 FIELD_PREP(STRTAB_STE_1_S1DSS, STRTAB_STE_1_S1DSS_SSID0) |
> +			 FIELD_PREP(STRTAB_STE_1_S1CIR, STRTAB_STE_1_S1C_CACHE_WBRA) |
> +			 FIELD_PREP(STRTAB_STE_1_S1COR, STRTAB_STE_1_S1C_CACHE_WBRA) |
> +			 FIELD_PREP(STRTAB_STE_1_S1CSH, ARM_SMMU_SH_ISH) |
> +			 FIELD_PREP(STRTAB_STE_1_STRW, STRTAB_STE_1_STRW_NSEL1));
> +
> +		if (smmu->features & ARM_SMMU_FEAT_STALLS &&
> +		   !(smmu->features & ARM_SMMU_FEAT_STALL_FORCE))
> +			dst[1] |= cpu_to_le64(STRTAB_STE_1_S1STALLD);
> +
> +		val |= (s1_cfg->cdcfg.cdtab_dma & STRTAB_STE_0_S1CTXPTR_MASK) |
> +			FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_S1_TRANS) |
> +			FIELD_PREP(STRTAB_STE_0_S1CDMAX, s1_cfg->s1cdmax) |
> +			FIELD_PREP(STRTAB_STE_0_S1FMT, s1_cfg->s1fmt);
> +	}
> +
> +	if (s2_cfg) {
> +		BUG_ON(ste_live);
> +		dst[2] = cpu_to_le64(
> +			 FIELD_PREP(STRTAB_STE_2_S2VMID, s2_cfg->vmid) |
> +			 FIELD_PREP(STRTAB_STE_2_VTCR, s2_cfg->vtcr) |
> +#ifdef __BIG_ENDIAN
> +			 STRTAB_STE_2_S2ENDI |
> +#endif
> +			 STRTAB_STE_2_S2PTW | STRTAB_STE_2_S2AA64 |
> +			 STRTAB_STE_2_S2R);
> +
> +		dst[3] = cpu_to_le64(s2_cfg->vttbr & STRTAB_STE_3_S2TTB_MASK);
> +
> +		val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_S2_TRANS);
> +	}
> +
> +	if (master->ats_enabled)
> +		dst[1] |= cpu_to_le64(FIELD_PREP(STRTAB_STE_1_EATS,
> +						 STRTAB_STE_1_EATS_TRANS));
> +
> +	arm_smmu_sync_ste_for_sid(smmu, sid);
> +	/* See comment in arm_smmu_write_ctx_desc() */
> +	WRITE_ONCE(dst[0], cpu_to_le64(val));
> +	arm_smmu_sync_ste_for_sid(smmu, sid);
> +
> +	/* It's likely that we'll want to use the new STE soon */
> +	if (!(smmu->options & ARM_SMMU_OPT_SKIP_PREFETCH))
> +		arm_smmu_cmdq_issue_cmd(smmu, &prefetch_cmd);
> +}
> +
> +static void arm_smmu_init_bypass_stes(u64 *strtab, unsigned int nent)
> +{
> +	unsigned int i;
> +
> +	for (i = 0; i < nent; ++i) {
> +		arm_smmu_write_strtab_ent(NULL, -1, strtab);
> +		strtab += STRTAB_STE_DWORDS;
> +	}
> +}
> +
> +static int arm_smmu_init_l2_strtab(struct arm_smmu_device *smmu, u32 sid)
> +{
> +	size_t size;
> +	void *strtab;
> +	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
> +	struct arm_smmu_strtab_l1_desc *desc = &cfg->l1_desc[sid >> STRTAB_SPLIT];
> +
> +	if (desc->l2ptr)
> +		return 0;
> +
> +	size = 1 << (STRTAB_SPLIT + ilog2(STRTAB_STE_DWORDS) + 3);
> +	strtab = &cfg->strtab[(sid >> STRTAB_SPLIT) * STRTAB_L1_DESC_DWORDS];
> +
> +	desc->span = STRTAB_SPLIT + 1;
> +	desc->l2ptr = dmam_alloc_coherent(smmu->dev, size, &desc->l2ptr_dma,
> +					  GFP_KERNEL);
> +	if (!desc->l2ptr) {
> +		dev_err(smmu->dev,
> +			"failed to allocate l2 stream table for SID %u\n",
> +			sid);
> +		return -ENOMEM;
> +	}
> +
> +	arm_smmu_init_bypass_stes(desc->l2ptr, 1 << STRTAB_SPLIT);
> +	arm_smmu_write_strtab_l1_desc(strtab, desc);
> +	return 0;
> +}
> +
> +/* IRQ and event handlers */
> +static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
> +{
> +	int i;
> +	struct arm_smmu_device *smmu = dev;
> +	struct arm_smmu_queue *q = &smmu->evtq.q;
> +	struct arm_smmu_ll_queue *llq = &q->llq;
> +	u64 evt[EVTQ_ENT_DWORDS];
> +
> +	do {
> +		while (!queue_remove_raw(q, evt)) {
> +			u8 id = FIELD_GET(EVTQ_0_ID, evt[0]);
> +
> +			dev_info(smmu->dev, "event 0x%02x received:\n", id);
> +			for (i = 0; i < ARRAY_SIZE(evt); ++i)
> +				dev_info(smmu->dev, "\t0x%016llx\n",
> +					 (unsigned long long)evt[i]);
> +
> +		}
> +
> +		/*
> +		 * Not much we can do on overflow, so scream and pretend we're
> +		 * trying harder.
> +		 */
> +		if (queue_sync_prod_in(q) == -EOVERFLOW)
> +			dev_err(smmu->dev, "EVTQ overflow detected -- events lost\n");
> +	} while (!queue_empty(llq));
> +
> +	/* Sync our overflow flag, as we believe we're up to speed */
> +	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
> +		    Q_IDX(llq, llq->cons);
> +	return IRQ_HANDLED;
> +}
> +
> +static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
> +{
> +	u32 sid, ssid;
> +	u16 grpid;
> +	bool ssv, last;
> +
> +	sid = FIELD_GET(PRIQ_0_SID, evt[0]);
> +	ssv = FIELD_GET(PRIQ_0_SSID_V, evt[0]);
> +	ssid = ssv ? FIELD_GET(PRIQ_0_SSID, evt[0]) : 0;
> +	last = FIELD_GET(PRIQ_0_PRG_LAST, evt[0]);
> +	grpid = FIELD_GET(PRIQ_1_PRG_IDX, evt[1]);
> +
> +	dev_info(smmu->dev, "unexpected PRI request received:\n");
> +	dev_info(smmu->dev,
> +		 "\tsid 0x%08x.0x%05x: [%u%s] %sprivileged %s%s%s access at iova 0x%016llx\n",
> +		 sid, ssid, grpid, last ? "L" : "",
> +		 evt[0] & PRIQ_0_PERM_PRIV ? "" : "un",
> +		 evt[0] & PRIQ_0_PERM_READ ? "R" : "",
> +		 evt[0] & PRIQ_0_PERM_WRITE ? "W" : "",
> +		 evt[0] & PRIQ_0_PERM_EXEC ? "X" : "",
> +		 evt[1] & PRIQ_1_ADDR_MASK);
> +
> +	if (last) {
> +		struct arm_smmu_cmdq_ent cmd = {
> +			.opcode			= CMDQ_OP_PRI_RESP,
> +			.substream_valid	= ssv,
> +			.pri			= {
> +				.sid	= sid,
> +				.ssid	= ssid,
> +				.grpid	= grpid,
> +				.resp	= PRI_RESP_DENY,
> +			},
> +		};
> +
> +		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
> +	}
> +}
> +
> +static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
> +{
> +	struct arm_smmu_device *smmu = dev;
> +	struct arm_smmu_queue *q = &smmu->priq.q;
> +	struct arm_smmu_ll_queue *llq = &q->llq;
> +	u64 evt[PRIQ_ENT_DWORDS];
> +
> +	do {
> +		while (!queue_remove_raw(q, evt))
> +			arm_smmu_handle_ppr(smmu, evt);
> +
> +		if (queue_sync_prod_in(q) == -EOVERFLOW)
> +			dev_err(smmu->dev, "PRIQ overflow detected -- requests lost\n");
> +	} while (!queue_empty(llq));
> +
> +	/* Sync our overflow flag, as we believe we're up to speed */
> +	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
> +		      Q_IDX(llq, llq->cons);
> +	queue_sync_cons_out(q);
> +	return IRQ_HANDLED;
> +}
> +
> +static int arm_smmu_device_disable(struct arm_smmu_device *smmu);
> +
> +static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
> +{
> +	u32 gerror, gerrorn, active;
> +	struct arm_smmu_device *smmu = dev;
> +
> +	gerror = readl_relaxed(smmu->base + ARM_SMMU_GERROR);
> +	gerrorn = readl_relaxed(smmu->base + ARM_SMMU_GERRORN);
> +
> +	active = gerror ^ gerrorn;
> +	if (!(active & GERROR_ERR_MASK))
> +		return IRQ_NONE; /* No errors pending */
> +
> +	dev_warn(smmu->dev,
> +		 "unexpected global error reported (0x%08x), this could be serious\n",
> +		 active);
> +
> +	if (active & GERROR_SFM_ERR) {
> +		dev_err(smmu->dev, "device has entered Service Failure Mode!\n");
> +		arm_smmu_device_disable(smmu);
> +	}
> +
> +	if (active & GERROR_MSI_GERROR_ABT_ERR)
> +		dev_warn(smmu->dev, "GERROR MSI write aborted\n");
> +
> +	if (active & GERROR_MSI_PRIQ_ABT_ERR)
> +		dev_warn(smmu->dev, "PRIQ MSI write aborted\n");
> +
> +	if (active & GERROR_MSI_EVTQ_ABT_ERR)
> +		dev_warn(smmu->dev, "EVTQ MSI write aborted\n");
> +
> +	if (active & GERROR_MSI_CMDQ_ABT_ERR)
> +		dev_warn(smmu->dev, "CMDQ MSI write aborted\n");
> +
> +	if (active & GERROR_PRIQ_ABT_ERR)
> +		dev_err(smmu->dev, "PRIQ write aborted -- events may have been lost\n");
> +
> +	if (active & GERROR_EVTQ_ABT_ERR)
> +		dev_err(smmu->dev, "EVTQ write aborted -- events may have been lost\n");
> +
> +	if (active & GERROR_CMDQ_ERR)
> +		arm_smmu_cmdq_skip_err(smmu);
> +
> +	writel(gerror, smmu->base + ARM_SMMU_GERRORN);
> +	return IRQ_HANDLED;
> +}
> +
> +static irqreturn_t arm_smmu_combined_irq_thread(int irq, void *dev)
> +{
> +	struct arm_smmu_device *smmu = dev;
> +
> +	arm_smmu_evtq_thread(irq, dev);
> +	if (smmu->features & ARM_SMMU_FEAT_PRI)
> +		arm_smmu_priq_thread(irq, dev);
> +
> +	return IRQ_HANDLED;
> +}
> +
> +static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
> +{
> +	arm_smmu_gerror_handler(irq, dev);
> +	return IRQ_WAKE_THREAD;
> +}
> +
> +static void
> +arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size,
> +			struct arm_smmu_cmdq_ent *cmd)
> +{
> +	size_t log2_span;
> +	size_t span_mask;
> +	/* ATC invalidates are always on 4096-bytes pages */
> +	size_t inval_grain_shift = 12;
> +	unsigned long page_start, page_end;
> +
> +	*cmd = (struct arm_smmu_cmdq_ent) {
> +		.opcode			= CMDQ_OP_ATC_INV,
> +		.substream_valid	= !!ssid,
> +		.atc.ssid		= ssid,
> +	};
> +
> +	if (!size) {
> +		cmd->atc.size = ATC_INV_SIZE_ALL;
> +		return;
> +	}
> +
> +	page_start	= iova >> inval_grain_shift;
> +	page_end	= (iova + size - 1) >> inval_grain_shift;
> +
> +	/*
> +	 * In an ATS Invalidate Request, the address must be aligned on the
> +	 * range size, which must be a power of two number of page sizes. We
> +	 * thus have to choose between grossly over-invalidating the region, or
> +	 * splitting the invalidation into multiple commands. For simplicity
> +	 * we'll go with the first solution, but should refine it in the future
> +	 * if multiple commands are shown to be more efficient.
> +	 *
> +	 * Find the smallest power of two that covers the range. The most
> +	 * significant differing bit between the start and end addresses,
> +	 * fls(start ^ end), indicates the required span. For example:
> +	 *
> +	 * We want to invalidate pages [8; 11]. This is already the ideal range:
> +	 *		x = 0b1000 ^ 0b1011 = 0b11
> +	 *		span = 1 << fls(x) = 4
> +	 *
> +	 * To invalidate pages [7; 10], we need to invalidate [0; 15]:
> +	 *		x = 0b0111 ^ 0b1010 = 0b1101
> +	 *		span = 1 << fls(x) = 16
> +	 */
> +	log2_span	= fls_long(page_start ^ page_end);
> +	span_mask	= (1ULL << log2_span) - 1;
> +
> +	page_start	&= ~span_mask;
> +
> +	cmd->atc.addr	= page_start << inval_grain_shift;
> +	cmd->atc.size	= log2_span;
> +}
> +
> +static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
> +{
> +	int i;
> +	struct arm_smmu_cmdq_ent cmd;
> +
> +	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
> +
> +	for (i = 0; i < master->num_sids; i++) {
> +		cmd.atc.sid = master->sids[i];
> +		arm_smmu_cmdq_issue_cmd(master->smmu, &cmd);
> +	}
> +
> +	return arm_smmu_cmdq_issue_sync(master->smmu);
> +}
> +
> +static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
> +				   int ssid, unsigned long iova, size_t size)
> +{
> +	int i;
> +	unsigned long flags;
> +	struct arm_smmu_cmdq_ent cmd;
> +	struct arm_smmu_master *master;
> +	struct arm_smmu_cmdq_batch cmds = {};
> +
> +	if (!(smmu_domain->smmu->features & ARM_SMMU_FEAT_ATS))
> +		return 0;
> +
> +	/*
> +	 * Ensure that we've completed prior invalidation of the main TLBs
> +	 * before we read 'nr_ats_masters' in case of a concurrent call to
> +	 * arm_smmu_enable_ats():
> +	 *
> +	 *	// unmap()			// arm_smmu_enable_ats()
> +	 *	TLBI+SYNC			atomic_inc(&nr_ats_masters);
> +	 *	smp_mb();			[...]
> +	 *	atomic_read(&nr_ats_masters);	pci_enable_ats() // writel()
> +	 *
> +	 * Ensures that we always see the incremented 'nr_ats_masters' count if
> +	 * ATS was enabled at the PCI device before completion of the TLBI.
> +	 */
> +	smp_mb();
> +	if (!atomic_read(&smmu_domain->nr_ats_masters))
> +		return 0;
> +
> +	arm_smmu_atc_inv_to_cmd(ssid, iova, size, &cmd);
> +
> +	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
> +	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
> +		if (!master->ats_enabled)
> +			continue;
> +
> +		for (i = 0; i < master->num_sids; i++) {
> +			cmd.atc.sid = master->sids[i];
> +			arm_smmu_cmdq_batch_add(smmu_domain->smmu, &cmds, &cmd);
> +		}
> +	}
> +	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
> +
> +	return arm_smmu_cmdq_batch_submit(smmu_domain->smmu, &cmds);
> +}
> +
> +/* IO_PGTABLE API */
> +static void arm_smmu_tlb_inv_context(void *cookie)
> +{
> +	struct arm_smmu_domain *smmu_domain = cookie;
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +	struct arm_smmu_cmdq_ent cmd;
> +
> +	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
> +		cmd.opcode	= CMDQ_OP_TLBI_NH_ASID;
> +		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
> +		cmd.tlbi.vmid	= 0;
> +	} else {
> +		cmd.opcode	= CMDQ_OP_TLBI_S12_VMALL;
> +		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
> +	}
> +
> +	/*
> +	 * NOTE: when io-pgtable is in non-strict mode, we may get here with
> +	 * PTEs previously cleared by unmaps on the current CPU not yet visible
> +	 * to the SMMU. We are relying on the dma_wmb() implicit during cmd
> +	 * insertion to guarantee those are observed before the TLBI. Do be
> +	 * careful, 007.
> +	 */
> +	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
> +	arm_smmu_cmdq_issue_sync(smmu);
> +	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
> +}
> +
> +static void arm_smmu_tlb_inv_range(unsigned long iova, size_t size,
> +				   size_t granule, bool leaf,
> +				   struct arm_smmu_domain *smmu_domain)
> +{
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +	unsigned long start = iova, end = iova + size, num_pages = 0, tg = 0;
> +	size_t inv_range = granule;
> +	struct arm_smmu_cmdq_batch cmds = {};
> +	struct arm_smmu_cmdq_ent cmd = {
> +		.tlbi = {
> +			.leaf	= leaf,
> +		},
> +	};
> +
> +	if (!size)
> +		return;
> +
> +	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
> +		cmd.opcode	= CMDQ_OP_TLBI_NH_VA;
> +		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
> +	} else {
> +		cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
> +		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
> +	}
> +
> +	if (smmu->features & ARM_SMMU_FEAT_RANGE_INV) {
> +		/* Get the leaf page size */
> +		tg = __ffs(smmu_domain->domain.pgsize_bitmap);
> +
> +		/* Convert page size of 12,14,16 (log2) to 1,2,3 */
> +		cmd.tlbi.tg = (tg - 10) / 2;
> +
> +		/* Determine what level the granule is at */
> +		cmd.tlbi.ttl = 4 - ((ilog2(granule) - 3) / (tg - 3));
> +
> +		num_pages = size >> tg;
> +	}
> +
> +	while (iova < end) {
> +		if (smmu->features & ARM_SMMU_FEAT_RANGE_INV) {
> +			/*
> +			 * On each iteration of the loop, the range is 5 bits
> +			 * worth of the aligned size remaining.
> +			 * The range in pages is:
> +			 *
> +			 * range = (num_pages & (0x1f << __ffs(num_pages)))
> +			 */
> +			unsigned long scale, num;
> +
> +			/* Determine the power of 2 multiple number of pages */
> +			scale = __ffs(num_pages);
> +			cmd.tlbi.scale = scale;
> +
> +			/* Determine how many chunks of 2^scale size we have */
> +			num = (num_pages >> scale) & CMDQ_TLBI_RANGE_NUM_MAX;
> +			cmd.tlbi.num = num - 1;
> +
> +			/* range is num * 2^scale * pgsize */
> +			inv_range = num << (scale + tg);
> +
> +			/* Clear out the lower order bits for the next iteration */
> +			num_pages -= num << scale;
> +		}
> +
> +		cmd.tlbi.addr = iova;
> +		arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
> +		iova += inv_range;
> +	}
> +	arm_smmu_cmdq_batch_submit(smmu, &cmds);
> +
> +	/*
> +	 * Unfortunately, this can't be leaf-only since we may have
> +	 * zapped an entire table.
> +	 */
> +	arm_smmu_atc_inv_domain(smmu_domain, 0, start, size);
> +}
> +
> +static void arm_smmu_tlb_inv_page_nosync(struct iommu_iotlb_gather *gather,
> +					 unsigned long iova, size_t granule,
> +					 void *cookie)
> +{
> +	struct arm_smmu_domain *smmu_domain = cookie;
> +	struct iommu_domain *domain = &smmu_domain->domain;
> +
> +	iommu_iotlb_gather_add_page(domain, gather, iova, granule);
> +}
> +
> +static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,
> +				  size_t granule, void *cookie)
> +{
> +	arm_smmu_tlb_inv_range(iova, size, granule, false, cookie);
> +}
> +
> +static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,
> +				  size_t granule, void *cookie)
> +{
> +	arm_smmu_tlb_inv_range(iova, size, granule, true, cookie);
> +}
> +
> +static const struct iommu_flush_ops arm_smmu_flush_ops = {
> +	.tlb_flush_all	= arm_smmu_tlb_inv_context,
> +	.tlb_flush_walk = arm_smmu_tlb_inv_walk,
> +	.tlb_flush_leaf = arm_smmu_tlb_inv_leaf,
> +	.tlb_add_page	= arm_smmu_tlb_inv_page_nosync,
> +};
> +
> +/* IOMMU API */
> +static bool arm_smmu_capable(enum iommu_cap cap)
> +{
> +	switch (cap) {
> +	case IOMMU_CAP_CACHE_COHERENCY:
> +		return true;
> +	case IOMMU_CAP_NOEXEC:
> +		return true;
> +	default:
> +		return false;
> +	}
> +}
> +
> +static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
> +{
> +	struct arm_smmu_domain *smmu_domain;
> +
> +	if (type != IOMMU_DOMAIN_UNMANAGED &&
> +	    type != IOMMU_DOMAIN_DMA &&
> +	    type != IOMMU_DOMAIN_IDENTITY)
> +		return NULL;
> +
> +	/*
> +	 * Allocate the domain and initialise some of its data structures.
> +	 * We can't really do anything meaningful until we've added a
> +	 * master.
> +	 */
> +	smmu_domain = kzalloc(sizeof(*smmu_domain), GFP_KERNEL);
> +	if (!smmu_domain)
> +		return NULL;
> +
> +	if (type == IOMMU_DOMAIN_DMA &&
> +	    iommu_get_dma_cookie(&smmu_domain->domain)) {
> +		kfree(smmu_domain);
> +		return NULL;
> +	}
> +
> +	mutex_init(&smmu_domain->init_mutex);
> +	INIT_LIST_HEAD(&smmu_domain->devices);
> +	spin_lock_init(&smmu_domain->devices_lock);
> +
> +	return &smmu_domain->domain;
> +}
> +
> +static int arm_smmu_bitmap_alloc(unsigned long *map, int span)
> +{
> +	int idx, size = 1 << span;
> +
> +	do {
> +		idx = find_first_zero_bit(map, size);
> +		if (idx == size)
> +			return -ENOSPC;
> +	} while (test_and_set_bit(idx, map));
> +
> +	return idx;
> +}
> +
> +static void arm_smmu_bitmap_free(unsigned long *map, int idx)
> +{
> +	clear_bit(idx, map);
> +}
> +
> +static void arm_smmu_domain_free(struct iommu_domain *domain)
> +{
> +	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +
> +	iommu_put_dma_cookie(domain);
> +	free_io_pgtable_ops(smmu_domain->pgtbl_ops);
> +
> +	/* Free the CD and ASID, if we allocated them */
> +	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
> +		struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
> +
> +		if (cfg->cdcfg.cdtab)
> +			arm_smmu_free_cd_tables(smmu_domain);
> +		arm_smmu_free_asid(&cfg->cd);
> +	} else {
> +		struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
> +		if (cfg->vmid)
> +			arm_smmu_bitmap_free(smmu->vmid_map, cfg->vmid);
> +	}
> +
> +	kfree(smmu_domain);
> +}
> +
> +static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
> +				       struct arm_smmu_master *master,
> +				       struct io_pgtable_cfg *pgtbl_cfg)
> +{
> +	int ret;
> +	u32 asid;
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
> +	typeof(&pgtbl_cfg->arm_lpae_s1_cfg.tcr) tcr = &pgtbl_cfg->arm_lpae_s1_cfg.tcr;
> +
> +	ret = xa_alloc(&asid_xa, &asid, &cfg->cd,
> +		       XA_LIMIT(1, (1 << smmu->asid_bits) - 1), GFP_KERNEL);
> +	if (ret)
> +		return ret;
> +
> +	cfg->s1cdmax = master->ssid_bits;
> +
> +	ret = arm_smmu_alloc_cd_tables(smmu_domain);
> +	if (ret)
> +		goto out_free_asid;
> +
> +	cfg->cd.asid	= (u16)asid;
> +	cfg->cd.ttbr	= pgtbl_cfg->arm_lpae_s1_cfg.ttbr;
> +	cfg->cd.tcr	= FIELD_PREP(CTXDESC_CD_0_TCR_T0SZ, tcr->tsz) |
> +			  FIELD_PREP(CTXDESC_CD_0_TCR_TG0, tcr->tg) |
> +			  FIELD_PREP(CTXDESC_CD_0_TCR_IRGN0, tcr->irgn) |
> +			  FIELD_PREP(CTXDESC_CD_0_TCR_ORGN0, tcr->orgn) |
> +			  FIELD_PREP(CTXDESC_CD_0_TCR_SH0, tcr->sh) |
> +			  FIELD_PREP(CTXDESC_CD_0_TCR_IPS, tcr->ips) |
> +			  CTXDESC_CD_0_TCR_EPD1 | CTXDESC_CD_0_AA64;
> +	cfg->cd.mair	= pgtbl_cfg->arm_lpae_s1_cfg.mair;
> +
> +	/*
> +	 * Note that this will end up calling arm_smmu_sync_cd() before
> +	 * the master has been added to the devices list for this domain.
> +	 * This isn't an issue because the STE hasn't been installed yet.
> +	 */
> +	ret = arm_smmu_write_ctx_desc(smmu_domain, 0, &cfg->cd);
> +	if (ret)
> +		goto out_free_cd_tables;
> +
> +	return 0;
> +
> +out_free_cd_tables:
> +	arm_smmu_free_cd_tables(smmu_domain);
> +out_free_asid:
> +	arm_smmu_free_asid(&cfg->cd);
> +	return ret;
> +}
> +
> +static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
> +				       struct arm_smmu_master *master,
> +				       struct io_pgtable_cfg *pgtbl_cfg)
> +{
> +	int vmid;
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
> +	typeof(&pgtbl_cfg->arm_lpae_s2_cfg.vtcr) vtcr;
> +
> +	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
> +	if (vmid < 0)
> +		return vmid;
> +
> +	vtcr = &pgtbl_cfg->arm_lpae_s2_cfg.vtcr;
> +	cfg->vmid	= (u16)vmid;
> +	cfg->vttbr	= pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
> +	cfg->vtcr	= FIELD_PREP(STRTAB_STE_2_VTCR_S2T0SZ, vtcr->tsz) |
> +			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SL0, vtcr->sl) |
> +			  FIELD_PREP(STRTAB_STE_2_VTCR_S2IR0, vtcr->irgn) |
> +			  FIELD_PREP(STRTAB_STE_2_VTCR_S2OR0, vtcr->orgn) |
> +			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SH0, vtcr->sh) |
> +			  FIELD_PREP(STRTAB_STE_2_VTCR_S2TG, vtcr->tg) |
> +			  FIELD_PREP(STRTAB_STE_2_VTCR_S2PS, vtcr->ps);
> +	return 0;
> +}
> +
> +static int arm_smmu_domain_finalise(struct iommu_domain *domain,
> +				    struct arm_smmu_master *master)
> +{
> +	int ret;
> +	unsigned long ias, oas;
> +	enum io_pgtable_fmt fmt;
> +	struct io_pgtable_cfg pgtbl_cfg;
> +	struct io_pgtable_ops *pgtbl_ops;
> +	int (*finalise_stage_fn)(struct arm_smmu_domain *,
> +				 struct arm_smmu_master *,
> +				 struct io_pgtable_cfg *);
> +	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +
> +	if (domain->type == IOMMU_DOMAIN_IDENTITY) {
> +		smmu_domain->stage = ARM_SMMU_DOMAIN_BYPASS;
> +		return 0;
> +	}
> +
> +	/* Restrict the stage to what we can actually support */
> +	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S1))
> +		smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
> +	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S2))
> +		smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
> +
> +	switch (smmu_domain->stage) {
> +	case ARM_SMMU_DOMAIN_S1:
> +		ias = (smmu->features & ARM_SMMU_FEAT_VAX) ? 52 : 48;
> +		ias = min_t(unsigned long, ias, VA_BITS);
> +		oas = smmu->ias;
> +		fmt = ARM_64_LPAE_S1;
> +		finalise_stage_fn = arm_smmu_domain_finalise_s1;
> +		break;
> +	case ARM_SMMU_DOMAIN_NESTED:
> +	case ARM_SMMU_DOMAIN_S2:
> +		ias = smmu->ias;
> +		oas = smmu->oas;
> +		fmt = ARM_64_LPAE_S2;
> +		finalise_stage_fn = arm_smmu_domain_finalise_s2;
> +		break;
> +	default:
> +		return -EINVAL;
> +	}
> +
> +	pgtbl_cfg = (struct io_pgtable_cfg) {
> +		.pgsize_bitmap	= smmu->pgsize_bitmap,
> +		.ias		= ias,
> +		.oas		= oas,
> +		.coherent_walk	= smmu->features & ARM_SMMU_FEAT_COHERENCY,
> +		.tlb		= &arm_smmu_flush_ops,
> +		.iommu_dev	= smmu->dev,
> +	};
> +
> +	if (smmu_domain->non_strict)
> +		pgtbl_cfg.quirks |= IO_PGTABLE_QUIRK_NON_STRICT;
> +
> +	pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
> +	if (!pgtbl_ops)
> +		return -ENOMEM;
> +
> +	domain->pgsize_bitmap = pgtbl_cfg.pgsize_bitmap;
> +	domain->geometry.aperture_end = (1UL << pgtbl_cfg.ias) - 1;
> +	domain->geometry.force_aperture = true;
> +
> +	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
> +	if (ret < 0) {
> +		free_io_pgtable_ops(pgtbl_ops);
> +		return ret;
> +	}
> +
> +	smmu_domain->pgtbl_ops = pgtbl_ops;
> +	return 0;
> +}
> +
> +static __le64 *arm_smmu_get_step_for_sid(struct arm_smmu_device *smmu, u32 sid)
> +{
> +	__le64 *step;
> +	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
> +
> +	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
> +		struct arm_smmu_strtab_l1_desc *l1_desc;
> +		int idx;
> +
> +		/* Two-level walk */
> +		idx = (sid >> STRTAB_SPLIT) * STRTAB_L1_DESC_DWORDS;
> +		l1_desc = &cfg->l1_desc[idx];
> +		idx = (sid & ((1 << STRTAB_SPLIT) - 1)) * STRTAB_STE_DWORDS;
> +		step = &l1_desc->l2ptr[idx];
> +	} else {
> +		/* Simple linear lookup */
> +		step = &cfg->strtab[sid * STRTAB_STE_DWORDS];
> +	}
> +
> +	return step;
> +}
> +
> +static void arm_smmu_install_ste_for_dev(struct arm_smmu_master *master)
> +{
> +	int i, j;
> +	struct arm_smmu_device *smmu = master->smmu;
> +
> +	for (i = 0; i < master->num_sids; ++i) {
> +		u32 sid = master->sids[i];
> +		__le64 *step = arm_smmu_get_step_for_sid(smmu, sid);
> +
> +		/* Bridged PCI devices may end up with duplicated IDs */
> +		for (j = 0; j < i; j++)
> +			if (master->sids[j] == sid)
> +				break;
> +		if (j < i)
> +			continue;
> +
> +		arm_smmu_write_strtab_ent(master, sid, step);
> +	}
> +}
> +
> +static bool arm_smmu_ats_supported(struct arm_smmu_master *master)
> +{
> +	struct device *dev = master->dev;
> +	struct arm_smmu_device *smmu = master->smmu;
> +	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> +
> +	if (!(smmu->features & ARM_SMMU_FEAT_ATS))
> +		return false;
> +
> +	if (!(fwspec->flags & IOMMU_FWSPEC_PCI_RC_ATS))
> +		return false;
> +
> +	return dev_is_pci(dev) && pci_ats_supported(to_pci_dev(dev));
> +}
> +
> +static void arm_smmu_enable_ats(struct arm_smmu_master *master)
> +{
> +	size_t stu;
> +	struct pci_dev *pdev;
> +	struct arm_smmu_device *smmu = master->smmu;
> +	struct arm_smmu_domain *smmu_domain = master->domain;
> +
> +	/* Don't enable ATS at the endpoint if it's not enabled in the STE */
> +	if (!master->ats_enabled)
> +		return;
> +
> +	/* Smallest Translation Unit: log2 of the smallest supported granule */
> +	stu = __ffs(smmu->pgsize_bitmap);
> +	pdev = to_pci_dev(master->dev);
> +
> +	atomic_inc(&smmu_domain->nr_ats_masters);
> +	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
> +	if (pci_enable_ats(pdev, stu))
> +		dev_err(master->dev, "Failed to enable ATS (STU %zu)\n", stu);
> +}
> +
> +static void arm_smmu_disable_ats(struct arm_smmu_master *master)
> +{
> +	struct arm_smmu_domain *smmu_domain = master->domain;
> +
> +	if (!master->ats_enabled)
> +		return;
> +
> +	pci_disable_ats(to_pci_dev(master->dev));
> +	/*
> +	 * Ensure ATS is disabled at the endpoint before we issue the
> +	 * ATC invalidation via the SMMU.
> +	 */
> +	wmb();
> +	arm_smmu_atc_inv_master(master);
> +	atomic_dec(&smmu_domain->nr_ats_masters);
> +}
> +
> +static int arm_smmu_enable_pasid(struct arm_smmu_master *master)
> +{
> +	int ret;
> +	int features;
> +	int num_pasids;
> +	struct pci_dev *pdev;
> +
> +	if (!dev_is_pci(master->dev))
> +		return -ENODEV;
> +
> +	pdev = to_pci_dev(master->dev);
> +
> +	features = pci_pasid_features(pdev);
> +	if (features < 0)
> +		return features;
> +
> +	num_pasids = pci_max_pasids(pdev);
> +	if (num_pasids <= 0)
> +		return num_pasids;
> +
> +	ret = pci_enable_pasid(pdev, features);
> +	if (ret) {
> +		dev_err(&pdev->dev, "Failed to enable PASID\n");
> +		return ret;
> +	}
> +
> +	master->ssid_bits = min_t(u8, ilog2(num_pasids),
> +				  master->smmu->ssid_bits);
> +	return 0;
> +}
> +
> +static void arm_smmu_disable_pasid(struct arm_smmu_master *master)
> +{
> +	struct pci_dev *pdev;
> +
> +	if (!dev_is_pci(master->dev))
> +		return;
> +
> +	pdev = to_pci_dev(master->dev);
> +
> +	if (!pdev->pasid_enabled)
> +		return;
> +
> +	master->ssid_bits = 0;
> +	pci_disable_pasid(pdev);
> +}
> +
> +static void arm_smmu_detach_dev(struct arm_smmu_master *master)
> +{
> +	unsigned long flags;
> +	struct arm_smmu_domain *smmu_domain = master->domain;
> +
> +	if (!smmu_domain)
> +		return;
> +
> +	arm_smmu_disable_ats(master);
> +
> +	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
> +	list_del(&master->domain_head);
> +	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
> +
> +	master->domain = NULL;
> +	master->ats_enabled = false;
> +	arm_smmu_install_ste_for_dev(master);
> +}
> +
> +static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
> +{
> +	int ret = 0;
> +	unsigned long flags;
> +	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> +	struct arm_smmu_device *smmu;
> +	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> +	struct arm_smmu_master *master;
> +
> +	if (!fwspec)
> +		return -ENOENT;
> +
> +	master = dev_iommu_priv_get(dev);
> +	smmu = master->smmu;
> +
> +	arm_smmu_detach_dev(master);
> +
> +	mutex_lock(&smmu_domain->init_mutex);
> +
> +	if (!smmu_domain->smmu) {
> +		smmu_domain->smmu = smmu;
> +		ret = arm_smmu_domain_finalise(domain, master);
> +		if (ret) {
> +			smmu_domain->smmu = NULL;
> +			goto out_unlock;
> +		}
> +	} else if (smmu_domain->smmu != smmu) {
> +		dev_err(dev,
> +			"cannot attach to SMMU %s (upstream of %s)\n",
> +			dev_name(smmu_domain->smmu->dev),
> +			dev_name(smmu->dev));
> +		ret = -ENXIO;
> +		goto out_unlock;
> +	} else if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1 &&
> +		   master->ssid_bits != smmu_domain->s1_cfg.s1cdmax) {
> +		dev_err(dev,
> +			"cannot attach to incompatible domain (%u SSID bits != %u)\n",
> +			smmu_domain->s1_cfg.s1cdmax, master->ssid_bits);
> +		ret = -EINVAL;
> +		goto out_unlock;
> +	}
> +
> +	master->domain = smmu_domain;
> +
> +	if (smmu_domain->stage != ARM_SMMU_DOMAIN_BYPASS)
> +		master->ats_enabled = arm_smmu_ats_supported(master);
> +
> +	arm_smmu_install_ste_for_dev(master);
> +
> +	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
> +	list_add(&master->domain_head, &smmu_domain->devices);
> +	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
> +
> +	arm_smmu_enable_ats(master);
> +
> +out_unlock:
> +	mutex_unlock(&smmu_domain->init_mutex);
> +	return ret;
> +}
> +
> +static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
> +			phys_addr_t paddr, size_t size, int prot, gfp_t gfp)
> +{
> +	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
> +
> +	if (!ops)
> +		return -ENODEV;
> +
> +	return ops->map(ops, iova, paddr, size, prot, gfp);
> +}
> +
> +static size_t arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova,
> +			     size_t size, struct iommu_iotlb_gather *gather)
> +{
> +	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> +	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
> +
> +	if (!ops)
> +		return 0;
> +
> +	return ops->unmap(ops, iova, size, gather);
> +}
> +
> +static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain)
> +{
> +	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> +
> +	if (smmu_domain->smmu)
> +		arm_smmu_tlb_inv_context(smmu_domain);
> +}
> +
> +static void arm_smmu_iotlb_sync(struct iommu_domain *domain,
> +				struct iommu_iotlb_gather *gather)
> +{
> +	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> +
> +	arm_smmu_tlb_inv_range(gather->start, gather->end - gather->start,
> +			       gather->pgsize, true, smmu_domain);
> +}
> +
> +static phys_addr_t
> +arm_smmu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
> +{
> +	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
> +
> +	if (domain->type == IOMMU_DOMAIN_IDENTITY)
> +		return iova;
> +
> +	if (!ops)
> +		return 0;
> +
> +	return ops->iova_to_phys(ops, iova);
> +}
> +
> +static struct platform_driver arm_smmu_driver;
> +
> +static
> +struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle *fwnode)
> +{
> +	struct device *dev = driver_find_device_by_fwnode(&arm_smmu_driver.driver,
> +							  fwnode);
> +	put_device(dev);
> +	return dev ? dev_get_drvdata(dev) : NULL;
> +}
> +
> +static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
> +{
> +	unsigned long limit = smmu->strtab_cfg.num_l1_ents;
> +
> +	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB)
> +		limit *= 1UL << STRTAB_SPLIT;
> +
> +	return sid < limit;
> +}
> +
> +static struct iommu_ops arm_smmu_ops;
> +
> +static struct iommu_device *arm_smmu_probe_device(struct device *dev)
> +{
> +	int i, ret;
> +	struct arm_smmu_device *smmu;
> +	struct arm_smmu_master *master;
> +	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> +
> +	if (!fwspec || fwspec->ops != &arm_smmu_ops)
> +		return ERR_PTR(-ENODEV);
> +
> +	if (WARN_ON_ONCE(dev_iommu_priv_get(dev)))
> +		return ERR_PTR(-EBUSY);
> +
> +	smmu = arm_smmu_get_by_fwnode(fwspec->iommu_fwnode);
> +	if (!smmu)
> +		return ERR_PTR(-ENODEV);
> +
> +	master = kzalloc(sizeof(*master), GFP_KERNEL);
> +	if (!master)
> +		return ERR_PTR(-ENOMEM);
> +
> +	master->dev = dev;
> +	master->smmu = smmu;
> +	master->sids = fwspec->ids;
> +	master->num_sids = fwspec->num_ids;
> +	dev_iommu_priv_set(dev, master);
> +
> +	/* Check the SIDs are in range of the SMMU and our stream table */
> +	for (i = 0; i < master->num_sids; i++) {
> +		u32 sid = master->sids[i];
> +
> +		if (!arm_smmu_sid_in_range(smmu, sid)) {
> +			ret = -ERANGE;
> +			goto err_free_master;
> +		}
> +
> +		/* Ensure l2 strtab is initialised */
> +		if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
> +			ret = arm_smmu_init_l2_strtab(smmu, sid);
> +			if (ret)
> +				goto err_free_master;
> +		}
> +	}
> +
> +	master->ssid_bits = min(smmu->ssid_bits, fwspec->num_pasid_bits);
> +
> +	/*
> +	 * Note that PASID must be enabled before, and disabled after ATS:
> +	 * PCI Express Base 4.0r1.0 - 10.5.1.3 ATS Control Register
> +	 *
> +	 *   Behavior is undefined if this bit is Set and the value of the PASID
> +	 *   Enable, Execute Requested Enable, or Privileged Mode Requested bits
> +	 *   are changed.
> +	 */
> +	arm_smmu_enable_pasid(master);
> +
> +	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB))
> +		master->ssid_bits = min_t(u8, master->ssid_bits,
> +					  CTXDESC_LINEAR_CDMAX);
> +
> +	return &smmu->iommu;
> +
> +err_free_master:
> +	kfree(master);
> +	dev_iommu_priv_set(dev, NULL);
> +	return ERR_PTR(ret);
> +}
> +
> +static void arm_smmu_release_device(struct device *dev)
> +{
> +	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> +	struct arm_smmu_master *master;
> +
> +	if (!fwspec || fwspec->ops != &arm_smmu_ops)
> +		return;
> +
> +	master = dev_iommu_priv_get(dev);
> +	arm_smmu_detach_dev(master);
> +	arm_smmu_disable_pasid(master);
> +	kfree(master);
> +	iommu_fwspec_free(dev);
> +}
> +
> +static struct iommu_group *arm_smmu_device_group(struct device *dev)
> +{
> +	struct iommu_group *group;
> +
> +	/*
> +	 * We don't support devices sharing stream IDs other than PCI RID
> +	 * aliases, since the necessary ID-to-device lookup becomes rather
> +	 * impractical given a potential sparse 32-bit stream ID space.
> +	 */
> +	if (dev_is_pci(dev))
> +		group = pci_device_group(dev);
> +	else
> +		group = generic_device_group(dev);
> +
> +	return group;
> +}
> +
> +static int arm_smmu_domain_get_attr(struct iommu_domain *domain,
> +				    enum iommu_attr attr, void *data)
> +{
> +	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> +
> +	switch (domain->type) {
> +	case IOMMU_DOMAIN_UNMANAGED:
> +		switch (attr) {
> +		case DOMAIN_ATTR_NESTING:
> +			*(int *)data = (smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED);
> +			return 0;
> +		default:
> +			return -ENODEV;
> +		}
> +		break;
> +	case IOMMU_DOMAIN_DMA:
> +		switch (attr) {
> +		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
> +			*(int *)data = smmu_domain->non_strict;
> +			return 0;
> +		default:
> +			return -ENODEV;
> +		}
> +		break;
> +	default:
> +		return -EINVAL;
> +	}
> +}
> +
> +static int arm_smmu_domain_set_attr(struct iommu_domain *domain,
> +				    enum iommu_attr attr, void *data)
> +{
> +	int ret = 0;
> +	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> +
> +	mutex_lock(&smmu_domain->init_mutex);
> +
> +	switch (domain->type) {
> +	case IOMMU_DOMAIN_UNMANAGED:
> +		switch (attr) {
> +		case DOMAIN_ATTR_NESTING:
> +			if (smmu_domain->smmu) {
> +				ret = -EPERM;
> +				goto out_unlock;
> +			}
> +
> +			if (*(int *)data)
> +				smmu_domain->stage = ARM_SMMU_DOMAIN_NESTED;
> +			else
> +				smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
> +			break;
> +		default:
> +			ret = -ENODEV;
> +		}
> +		break;
> +	case IOMMU_DOMAIN_DMA:
> +		switch(attr) {
> +		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
> +			smmu_domain->non_strict = *(int *)data;
> +			break;
> +		default:
> +			ret = -ENODEV;
> +		}
> +		break;
> +	default:
> +		ret = -EINVAL;
> +	}
> +
> +out_unlock:
> +	mutex_unlock(&smmu_domain->init_mutex);
> +	return ret;
> +}
> +
> +static int arm_smmu_of_xlate(struct device *dev, struct of_phandle_args *args)
> +{
> +	return iommu_fwspec_add_ids(dev, args->args, 1);
> +}
> +
> +static void arm_smmu_get_resv_regions(struct device *dev,
> +				      struct list_head *head)
> +{
> +	struct iommu_resv_region *region;
> +	int prot = IOMMU_WRITE | IOMMU_NOEXEC | IOMMU_MMIO;
> +
> +	region = iommu_alloc_resv_region(MSI_IOVA_BASE, MSI_IOVA_LENGTH,
> +					 prot, IOMMU_RESV_SW_MSI);
> +	if (!region)
> +		return;
> +
> +	list_add_tail(&region->list, head);
> +
> +	iommu_dma_get_resv_regions(dev, head);
> +}
> +
> +static struct iommu_ops arm_smmu_ops = {
> +	.capable		= arm_smmu_capable,
> +	.domain_alloc		= arm_smmu_domain_alloc,
> +	.domain_free		= arm_smmu_domain_free,
> +	.attach_dev		= arm_smmu_attach_dev,
> +	.map			= arm_smmu_map,
> +	.unmap			= arm_smmu_unmap,
> +	.flush_iotlb_all	= arm_smmu_flush_iotlb_all,
> +	.iotlb_sync		= arm_smmu_iotlb_sync,
> +	.iova_to_phys		= arm_smmu_iova_to_phys,
> +	.probe_device		= arm_smmu_probe_device,
> +	.release_device		= arm_smmu_release_device,
> +	.device_group		= arm_smmu_device_group,
> +	.domain_get_attr	= arm_smmu_domain_get_attr,
> +	.domain_set_attr	= arm_smmu_domain_set_attr,
> +	.of_xlate		= arm_smmu_of_xlate,
> +	.get_resv_regions	= arm_smmu_get_resv_regions,
> +	.put_resv_regions	= generic_iommu_put_resv_regions,
> +	.pgsize_bitmap		= -1UL, /* Restricted during device attach */
> +};
> +
> +/* Probing and initialisation functions */
> +static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
> +				   struct arm_smmu_queue *q,
> +				   unsigned long prod_off,
> +				   unsigned long cons_off,
> +				   size_t dwords, const char *name)
> +{
> +	size_t qsz;
> +
> +	do {
> +		qsz = ((1 << q->llq.max_n_shift) * dwords) << 3;
> +		q->base = dmam_alloc_coherent(smmu->dev, qsz, &q->base_dma,
> +					      GFP_KERNEL);
> +		if (q->base || qsz < PAGE_SIZE)
> +			break;
> +
> +		q->llq.max_n_shift--;
> +	} while (1);
> +
> +	if (!q->base) {
> +		dev_err(smmu->dev,
> +			"failed to allocate queue (0x%zx bytes) for %s\n",
> +			qsz, name);
> +		return -ENOMEM;
> +	}
> +
> +	if (!WARN_ON(q->base_dma & (qsz - 1))) {
> +		dev_info(smmu->dev, "allocated %u entries for %s\n",
> +			 1 << q->llq.max_n_shift, name);
> +	}
> +
> +	q->prod_reg	= arm_smmu_page1_fixup(prod_off, smmu);
> +	q->cons_reg	= arm_smmu_page1_fixup(cons_off, smmu);
> +	q->ent_dwords	= dwords;
> +
> +	q->q_base  = Q_BASE_RWA;
> +	q->q_base |= q->base_dma & Q_BASE_ADDR_MASK;
> +	q->q_base |= FIELD_PREP(Q_BASE_LOG2SIZE, q->llq.max_n_shift);
> +
> +	q->llq.prod = q->llq.cons = 0;
> +	return 0;
> +}
> +
> +static void arm_smmu_cmdq_free_bitmap(void *data)
> +{
> +	unsigned long *bitmap = data;
> +	bitmap_free(bitmap);
> +}
> +
> +static int arm_smmu_cmdq_init(struct arm_smmu_device *smmu)
> +{
> +	int ret = 0;
> +	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
> +	unsigned int nents = 1 << cmdq->q.llq.max_n_shift;
> +	atomic_long_t *bitmap;
> +
> +	atomic_set(&cmdq->owner_prod, 0);
> +	atomic_set(&cmdq->lock, 0);
> +
> +	bitmap = (atomic_long_t *)bitmap_zalloc(nents, GFP_KERNEL);
> +	if (!bitmap) {
> +		dev_err(smmu->dev, "failed to allocate cmdq bitmap\n");
> +		ret = -ENOMEM;
> +	} else {
> +		cmdq->valid_map = bitmap;
> +		devm_add_action(smmu->dev, arm_smmu_cmdq_free_bitmap, bitmap);
> +	}
> +
> +	return ret;
> +}
> +
> +static int arm_smmu_init_queues(struct arm_smmu_device *smmu)
> +{
> +	int ret;
> +
> +	/* cmdq */
> +	ret = arm_smmu_init_one_queue(smmu, &smmu->cmdq.q, ARM_SMMU_CMDQ_PROD,
> +				      ARM_SMMU_CMDQ_CONS, CMDQ_ENT_DWORDS,
> +				      "cmdq");
> +	if (ret)
> +		return ret;
> +
> +	ret = arm_smmu_cmdq_init(smmu);
> +	if (ret)
> +		return ret;
> +
> +	/* evtq */
> +	ret = arm_smmu_init_one_queue(smmu, &smmu->evtq.q, ARM_SMMU_EVTQ_PROD,
> +				      ARM_SMMU_EVTQ_CONS, EVTQ_ENT_DWORDS,
> +				      "evtq");
> +	if (ret)
> +		return ret;
> +
> +	/* priq */
> +	if (!(smmu->features & ARM_SMMU_FEAT_PRI))
> +		return 0;
> +
> +	return arm_smmu_init_one_queue(smmu, &smmu->priq.q, ARM_SMMU_PRIQ_PROD,
> +				       ARM_SMMU_PRIQ_CONS, PRIQ_ENT_DWORDS,
> +				       "priq");
> +}
> +
> +static int arm_smmu_init_l1_strtab(struct arm_smmu_device *smmu)
> +{
> +	unsigned int i;
> +	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
> +	size_t size = sizeof(*cfg->l1_desc) * cfg->num_l1_ents;
> +	void *strtab = smmu->strtab_cfg.strtab;
> +
> +	cfg->l1_desc = devm_kzalloc(smmu->dev, size, GFP_KERNEL);
> +	if (!cfg->l1_desc) {
> +		dev_err(smmu->dev, "failed to allocate l1 stream table desc\n");
> +		return -ENOMEM;
> +	}
> +
> +	for (i = 0; i < cfg->num_l1_ents; ++i) {
> +		arm_smmu_write_strtab_l1_desc(strtab, &cfg->l1_desc[i]);
> +		strtab += STRTAB_L1_DESC_DWORDS << 3;
> +	}
> +
> +	return 0;
> +}
> +
> +static int arm_smmu_init_strtab_2lvl(struct arm_smmu_device *smmu)
> +{
> +	void *strtab;
> +	u64 reg;
> +	u32 size, l1size;
> +	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
> +
> +	/* Calculate the L1 size, capped to the SIDSIZE. */
> +	size = STRTAB_L1_SZ_SHIFT - (ilog2(STRTAB_L1_DESC_DWORDS) + 3);
> +	size = min(size, smmu->sid_bits - STRTAB_SPLIT);
> +	cfg->num_l1_ents = 1 << size;
> +
> +	size += STRTAB_SPLIT;
> +	if (size < smmu->sid_bits)
> +		dev_warn(smmu->dev,
> +			 "2-level strtab only covers %u/%u bits of SID\n",
> +			 size, smmu->sid_bits);
> +
> +	l1size = cfg->num_l1_ents * (STRTAB_L1_DESC_DWORDS << 3);
> +	strtab = dmam_alloc_coherent(smmu->dev, l1size, &cfg->strtab_dma,
> +				     GFP_KERNEL);
> +	if (!strtab) {
> +		dev_err(smmu->dev,
> +			"failed to allocate l1 stream table (%u bytes)\n",
> +			size);
> +		return -ENOMEM;
> +	}
> +	cfg->strtab = strtab;
> +
> +	/* Configure strtab_base_cfg for 2 levels */
> +	reg  = FIELD_PREP(STRTAB_BASE_CFG_FMT, STRTAB_BASE_CFG_FMT_2LVL);
> +	reg |= FIELD_PREP(STRTAB_BASE_CFG_LOG2SIZE, size);
> +	reg |= FIELD_PREP(STRTAB_BASE_CFG_SPLIT, STRTAB_SPLIT);
> +	cfg->strtab_base_cfg = reg;
> +
> +	return arm_smmu_init_l1_strtab(smmu);
> +}
> +
> +static int arm_smmu_init_strtab_linear(struct arm_smmu_device *smmu)
> +{
> +	void *strtab;
> +	u64 reg;
> +	u32 size;
> +	struct arm_smmu_strtab_cfg *cfg = &smmu->strtab_cfg;
> +
> +	size = (1 << smmu->sid_bits) * (STRTAB_STE_DWORDS << 3);
> +	strtab = dmam_alloc_coherent(smmu->dev, size, &cfg->strtab_dma,
> +				     GFP_KERNEL);
> +	if (!strtab) {
> +		dev_err(smmu->dev,
> +			"failed to allocate linear stream table (%u bytes)\n",
> +			size);
> +		return -ENOMEM;
> +	}
> +	cfg->strtab = strtab;
> +	cfg->num_l1_ents = 1 << smmu->sid_bits;
> +
> +	/* Configure strtab_base_cfg for a linear table covering all SIDs */
> +	reg  = FIELD_PREP(STRTAB_BASE_CFG_FMT, STRTAB_BASE_CFG_FMT_LINEAR);
> +	reg |= FIELD_PREP(STRTAB_BASE_CFG_LOG2SIZE, smmu->sid_bits);
> +	cfg->strtab_base_cfg = reg;
> +
> +	arm_smmu_init_bypass_stes(strtab, cfg->num_l1_ents);
> +	return 0;
> +}
> +
> +static int arm_smmu_init_strtab(struct arm_smmu_device *smmu)
> +{
> +	u64 reg;
> +	int ret;
> +
> +	if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB)
> +		ret = arm_smmu_init_strtab_2lvl(smmu);
> +	else
> +		ret = arm_smmu_init_strtab_linear(smmu);
> +
> +	if (ret)
> +		return ret;
> +
> +	/* Set the strtab base address */
> +	reg  = smmu->strtab_cfg.strtab_dma & STRTAB_BASE_ADDR_MASK;
> +	reg |= STRTAB_BASE_RA;
> +	smmu->strtab_cfg.strtab_base = reg;
> +
> +	/* Allocate the first VMID for stage-2 bypass STEs */
> +	set_bit(0, smmu->vmid_map);
> +	return 0;
> +}
> +
> +static int arm_smmu_init_structures(struct arm_smmu_device *smmu)
> +{
> +	int ret;
> +
> +	ret = arm_smmu_init_queues(smmu);
> +	if (ret)
> +		return ret;
> +
> +	return arm_smmu_init_strtab(smmu);
> +}
> +
> +static int arm_smmu_write_reg_sync(struct arm_smmu_device *smmu, u32 val,
> +				   unsigned int reg_off, unsigned int ack_off)
> +{
> +	u32 reg;
> +
> +	writel_relaxed(val, smmu->base + reg_off);
> +	return readl_relaxed_poll_timeout(smmu->base + ack_off, reg, reg == val,
> +					  1, ARM_SMMU_POLL_TIMEOUT_US);
> +}
> +
> +/* GBPA is "special" */
> +static int arm_smmu_update_gbpa(struct arm_smmu_device *smmu, u32 set, u32 clr)
> +{
> +	int ret;
> +	u32 reg, __iomem *gbpa = smmu->base + ARM_SMMU_GBPA;
> +
> +	ret = readl_relaxed_poll_timeout(gbpa, reg, !(reg & GBPA_UPDATE),
> +					 1, ARM_SMMU_POLL_TIMEOUT_US);
> +	if (ret)
> +		return ret;
> +
> +	reg &= ~clr;
> +	reg |= set;
> +	writel_relaxed(reg | GBPA_UPDATE, gbpa);
> +	ret = readl_relaxed_poll_timeout(gbpa, reg, !(reg & GBPA_UPDATE),
> +					 1, ARM_SMMU_POLL_TIMEOUT_US);
> +
> +	if (ret)
> +		dev_err(smmu->dev, "GBPA not responding to update\n");
> +	return ret;
> +}
> +
> +static void arm_smmu_free_msis(void *data)
> +{
> +	struct device *dev = data;
> +	platform_msi_domain_free_irqs(dev);
> +}
> +
> +static void arm_smmu_write_msi_msg(struct msi_desc *desc, struct msi_msg *msg)
> +{
> +	phys_addr_t doorbell;
> +	struct device *dev = msi_desc_to_dev(desc);
> +	struct arm_smmu_device *smmu = dev_get_drvdata(dev);
> +	phys_addr_t *cfg = arm_smmu_msi_cfg[desc->platform.msi_index];
> +
> +	doorbell = (((u64)msg->address_hi) << 32) | msg->address_lo;
> +	doorbell &= MSI_CFG0_ADDR_MASK;
> +
> +	writeq_relaxed(doorbell, smmu->base + cfg[0]);
> +	writel_relaxed(msg->data, smmu->base + cfg[1]);
> +	writel_relaxed(ARM_SMMU_MEMATTR_DEVICE_nGnRE, smmu->base + cfg[2]);
> +}
> +
> +static void arm_smmu_setup_msis(struct arm_smmu_device *smmu)
> +{
> +	struct msi_desc *desc;
> +	int ret, nvec = ARM_SMMU_MAX_MSIS;
> +	struct device *dev = smmu->dev;
> +
> +	/* Clear the MSI address regs */
> +	writeq_relaxed(0, smmu->base + ARM_SMMU_GERROR_IRQ_CFG0);
> +	writeq_relaxed(0, smmu->base + ARM_SMMU_EVTQ_IRQ_CFG0);
> +
> +	if (smmu->features & ARM_SMMU_FEAT_PRI)
> +		writeq_relaxed(0, smmu->base + ARM_SMMU_PRIQ_IRQ_CFG0);
> +	else
> +		nvec--;
> +
> +	if (!(smmu->features & ARM_SMMU_FEAT_MSI))
> +		return;
> +
> +	if (!dev->msi_domain) {
> +		dev_info(smmu->dev, "msi_domain absent - falling back to wired irqs\n");
> +		return;
> +	}
> +
> +	/* Allocate MSIs for evtq, gerror and priq. Ignore cmdq */
> +	ret = platform_msi_domain_alloc_irqs(dev, nvec, arm_smmu_write_msi_msg);
> +	if (ret) {
> +		dev_warn(dev, "failed to allocate MSIs - falling back to wired irqs\n");
> +		return;
> +	}
> +
> +	for_each_msi_entry(desc, dev) {
> +		switch (desc->platform.msi_index) {
> +		case EVTQ_MSI_INDEX:
> +			smmu->evtq.q.irq = desc->irq;
> +			break;
> +		case GERROR_MSI_INDEX:
> +			smmu->gerr_irq = desc->irq;
> +			break;
> +		case PRIQ_MSI_INDEX:
> +			smmu->priq.q.irq = desc->irq;
> +			break;
> +		default:	/* Unknown */
> +			continue;
> +		}
> +	}
> +
> +	/* Add callback to free MSIs on teardown */
> +	devm_add_action(dev, arm_smmu_free_msis, dev);
> +}
> +
> +static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
> +{
> +	int irq, ret;
> +
> +	arm_smmu_setup_msis(smmu);
> +
> +	/* Request interrupt lines */
> +	irq = smmu->evtq.q.irq;
> +	if (irq) {
> +		ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
> +						arm_smmu_evtq_thread,
> +						IRQF_ONESHOT,
> +						"arm-smmu-v3-evtq", smmu);
> +		if (ret < 0)
> +			dev_warn(smmu->dev, "failed to enable evtq irq\n");
> +	} else {
> +		dev_warn(smmu->dev, "no evtq irq - events will not be reported!\n");
> +	}
> +
> +	irq = smmu->gerr_irq;
> +	if (irq) {
> +		ret = devm_request_irq(smmu->dev, irq, arm_smmu_gerror_handler,
> +				       0, "arm-smmu-v3-gerror", smmu);
> +		if (ret < 0)
> +			dev_warn(smmu->dev, "failed to enable gerror irq\n");
> +	} else {
> +		dev_warn(smmu->dev, "no gerr irq - errors will not be reported!\n");
> +	}
> +
> +	if (smmu->features & ARM_SMMU_FEAT_PRI) {
> +		irq = smmu->priq.q.irq;
> +		if (irq) {
> +			ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
> +							arm_smmu_priq_thread,
> +							IRQF_ONESHOT,
> +							"arm-smmu-v3-priq",
> +							smmu);
> +			if (ret < 0)
> +				dev_warn(smmu->dev,
> +					 "failed to enable priq irq\n");
> +		} else {
> +			dev_warn(smmu->dev, "no priq irq - PRI will be broken\n");
> +		}
> +	}
> +}
> +
> +static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
> +{
> +	int ret, irq;
> +	u32 irqen_flags = IRQ_CTRL_EVTQ_IRQEN | IRQ_CTRL_GERROR_IRQEN;
> +
> +	/* Disable IRQs first */
> +	ret = arm_smmu_write_reg_sync(smmu, 0, ARM_SMMU_IRQ_CTRL,
> +				      ARM_SMMU_IRQ_CTRLACK);
> +	if (ret) {
> +		dev_err(smmu->dev, "failed to disable irqs\n");
> +		return ret;
> +	}
> +
> +	irq = smmu->combined_irq;
> +	if (irq) {
> +		/*
> +		 * Cavium ThunderX2 implementation doesn't support unique irq
> +		 * lines. Use a single irq line for all the SMMUv3 interrupts.
> +		 */
> +		ret = devm_request_threaded_irq(smmu->dev, irq,
> +					arm_smmu_combined_irq_handler,
> +					arm_smmu_combined_irq_thread,
> +					IRQF_ONESHOT,
> +					"arm-smmu-v3-combined-irq", smmu);
> +		if (ret < 0)
> +			dev_warn(smmu->dev, "failed to enable combined irq\n");
> +	} else
> +		arm_smmu_setup_unique_irqs(smmu);
> +
> +	if (smmu->features & ARM_SMMU_FEAT_PRI)
> +		irqen_flags |= IRQ_CTRL_PRIQ_IRQEN;
> +
> +	/* Enable interrupt generation on the SMMU */
> +	ret = arm_smmu_write_reg_sync(smmu, irqen_flags,
> +				      ARM_SMMU_IRQ_CTRL, ARM_SMMU_IRQ_CTRLACK);
> +	if (ret)
> +		dev_warn(smmu->dev, "failed to enable irqs\n");
> +
> +	return 0;
> +}
> +
> +static int arm_smmu_device_disable(struct arm_smmu_device *smmu)
> +{
> +	int ret;
> +
> +	ret = arm_smmu_write_reg_sync(smmu, 0, ARM_SMMU_CR0, ARM_SMMU_CR0ACK);
> +	if (ret)
> +		dev_err(smmu->dev, "failed to clear cr0\n");
> +
> +	return ret;
> +}
> +
> +static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
> +{
> +	int ret;
> +	u32 reg, enables;
> +	struct arm_smmu_cmdq_ent cmd;
> +
> +	/* Clear CR0 and sync (disables SMMU and queue processing) */
> +	reg = readl_relaxed(smmu->base + ARM_SMMU_CR0);
> +	if (reg & CR0_SMMUEN) {
> +		dev_warn(smmu->dev, "SMMU currently enabled! Resetting...\n");
> +		WARN_ON(is_kdump_kernel() && !disable_bypass);
> +		arm_smmu_update_gbpa(smmu, GBPA_ABORT, 0);
> +	}
> +
> +	ret = arm_smmu_device_disable(smmu);
> +	if (ret)
> +		return ret;
> +
> +	/* CR1 (table and queue memory attributes) */
> +	reg = FIELD_PREP(CR1_TABLE_SH, ARM_SMMU_SH_ISH) |
> +	      FIELD_PREP(CR1_TABLE_OC, CR1_CACHE_WB) |
> +	      FIELD_PREP(CR1_TABLE_IC, CR1_CACHE_WB) |
> +	      FIELD_PREP(CR1_QUEUE_SH, ARM_SMMU_SH_ISH) |
> +	      FIELD_PREP(CR1_QUEUE_OC, CR1_CACHE_WB) |
> +	      FIELD_PREP(CR1_QUEUE_IC, CR1_CACHE_WB);
> +	writel_relaxed(reg, smmu->base + ARM_SMMU_CR1);
> +
> +	/* CR2 (random crap) */
> +	reg = CR2_PTM | CR2_RECINVSID | CR2_E2H;
> +	writel_relaxed(reg, smmu->base + ARM_SMMU_CR2);
> +
> +	/* Stream table */
> +	writeq_relaxed(smmu->strtab_cfg.strtab_base,
> +		       smmu->base + ARM_SMMU_STRTAB_BASE);
> +	writel_relaxed(smmu->strtab_cfg.strtab_base_cfg,
> +		       smmu->base + ARM_SMMU_STRTAB_BASE_CFG);
> +
> +	/* Command queue */
> +	writeq_relaxed(smmu->cmdq.q.q_base, smmu->base + ARM_SMMU_CMDQ_BASE);
> +	writel_relaxed(smmu->cmdq.q.llq.prod, smmu->base + ARM_SMMU_CMDQ_PROD);
> +	writel_relaxed(smmu->cmdq.q.llq.cons, smmu->base + ARM_SMMU_CMDQ_CONS);
> +
> +	enables = CR0_CMDQEN;
> +	ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
> +				      ARM_SMMU_CR0ACK);
> +	if (ret) {
> +		dev_err(smmu->dev, "failed to enable command queue\n");
> +		return ret;
> +	}
> +
> +	/* Invalidate any cached configuration */
> +	cmd.opcode = CMDQ_OP_CFGI_ALL;
> +	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
> +	arm_smmu_cmdq_issue_sync(smmu);
> +
> +	/* Invalidate any stale TLB entries */
> +	if (smmu->features & ARM_SMMU_FEAT_HYP) {
> +		cmd.opcode = CMDQ_OP_TLBI_EL2_ALL;
> +		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
> +	}
> +
> +	cmd.opcode = CMDQ_OP_TLBI_NSNH_ALL;
> +	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
> +	arm_smmu_cmdq_issue_sync(smmu);
> +
> +	/* Event queue */
> +	writeq_relaxed(smmu->evtq.q.q_base, smmu->base + ARM_SMMU_EVTQ_BASE);
> +	writel_relaxed(smmu->evtq.q.llq.prod,
> +		       arm_smmu_page1_fixup(ARM_SMMU_EVTQ_PROD, smmu));
> +	writel_relaxed(smmu->evtq.q.llq.cons,
> +		       arm_smmu_page1_fixup(ARM_SMMU_EVTQ_CONS, smmu));
> +
> +	enables |= CR0_EVTQEN;
> +	ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
> +				      ARM_SMMU_CR0ACK);
> +	if (ret) {
> +		dev_err(smmu->dev, "failed to enable event queue\n");
> +		return ret;
> +	}
> +
> +	/* PRI queue */
> +	if (smmu->features & ARM_SMMU_FEAT_PRI) {
> +		writeq_relaxed(smmu->priq.q.q_base,
> +			       smmu->base + ARM_SMMU_PRIQ_BASE);
> +		writel_relaxed(smmu->priq.q.llq.prod,
> +			       arm_smmu_page1_fixup(ARM_SMMU_PRIQ_PROD, smmu));
> +		writel_relaxed(smmu->priq.q.llq.cons,
> +			       arm_smmu_page1_fixup(ARM_SMMU_PRIQ_CONS, smmu));
> +
> +		enables |= CR0_PRIQEN;
> +		ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
> +					      ARM_SMMU_CR0ACK);
> +		if (ret) {
> +			dev_err(smmu->dev, "failed to enable PRI queue\n");
> +			return ret;
> +		}
> +	}
> +
> +	if (smmu->features & ARM_SMMU_FEAT_ATS) {
> +		enables |= CR0_ATSCHK;
> +		ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
> +					      ARM_SMMU_CR0ACK);
> +		if (ret) {
> +			dev_err(smmu->dev, "failed to enable ATS check\n");
> +			return ret;
> +		}
> +	}
> +
> +	ret = arm_smmu_setup_irqs(smmu);
> +	if (ret) {
> +		dev_err(smmu->dev, "failed to setup irqs\n");
> +		return ret;
> +	}
> +
> +	if (is_kdump_kernel())
> +		enables &= ~(CR0_EVTQEN | CR0_PRIQEN);
> +
> +	/* Enable the SMMU interface, or ensure bypass */
> +	if (!bypass || disable_bypass) {
> +		enables |= CR0_SMMUEN;
> +	} else {
> +		ret = arm_smmu_update_gbpa(smmu, 0, GBPA_ABORT);
> +		if (ret)
> +			return ret;
> +	}
> +	ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
> +				      ARM_SMMU_CR0ACK);
> +	if (ret) {
> +		dev_err(smmu->dev, "failed to enable SMMU interface\n");
> +		return ret;
> +	}
> +
> +	return 0;
> +}
> +
> +static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
> +{
> +	u32 reg;
> +	bool coherent = smmu->features & ARM_SMMU_FEAT_COHERENCY;
> +
> +	/* IDR0 */
> +	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR0);
> +
> +	/* 2-level structures */
> +	if (FIELD_GET(IDR0_ST_LVL, reg) == IDR0_ST_LVL_2LVL)
> +		smmu->features |= ARM_SMMU_FEAT_2_LVL_STRTAB;
> +
> +	if (reg & IDR0_CD2L)
> +		smmu->features |= ARM_SMMU_FEAT_2_LVL_CDTAB;
> +
> +	/*
> +	 * Translation table endianness.
> +	 * We currently require the same endianness as the CPU, but this
> +	 * could be changed later by adding a new IO_PGTABLE_QUIRK.
> +	 */
> +	switch (FIELD_GET(IDR0_TTENDIAN, reg)) {
> +	case IDR0_TTENDIAN_MIXED:
> +		smmu->features |= ARM_SMMU_FEAT_TT_LE | ARM_SMMU_FEAT_TT_BE;
> +		break;
> +#ifdef __BIG_ENDIAN
> +	case IDR0_TTENDIAN_BE:
> +		smmu->features |= ARM_SMMU_FEAT_TT_BE;
> +		break;
> +#else
> +	case IDR0_TTENDIAN_LE:
> +		smmu->features |= ARM_SMMU_FEAT_TT_LE;
> +		break;
> +#endif
> +	default:
> +		dev_err(smmu->dev, "unknown/unsupported TT endianness!\n");
> +		return -ENXIO;
> +	}
> +
> +	/* Boolean feature flags */
> +	if (IS_ENABLED(CONFIG_PCI_PRI) && reg & IDR0_PRI)
> +		smmu->features |= ARM_SMMU_FEAT_PRI;
> +
> +	if (IS_ENABLED(CONFIG_PCI_ATS) && reg & IDR0_ATS)
> +		smmu->features |= ARM_SMMU_FEAT_ATS;
> +
> +	if (reg & IDR0_SEV)
> +		smmu->features |= ARM_SMMU_FEAT_SEV;
> +
> +	if (reg & IDR0_MSI)
> +		smmu->features |= ARM_SMMU_FEAT_MSI;
> +
> +	if (reg & IDR0_HYP)
> +		smmu->features |= ARM_SMMU_FEAT_HYP;
> +
> +	/*
> +	 * The coherency feature as set by FW is used in preference to the ID
> +	 * register, but warn on mismatch.
> +	 */
> +	if (!!(reg & IDR0_COHACC) != coherent)
> +		dev_warn(smmu->dev, "IDR0.COHACC overridden by FW configuration (%s)\n",
> +			 coherent ? "true" : "false");
> +
> +	switch (FIELD_GET(IDR0_STALL_MODEL, reg)) {
> +	case IDR0_STALL_MODEL_FORCE:
> +		smmu->features |= ARM_SMMU_FEAT_STALL_FORCE;
> +		fallthrough;
> +	case IDR0_STALL_MODEL_STALL:
> +		smmu->features |= ARM_SMMU_FEAT_STALLS;
> +	}
> +
> +	if (reg & IDR0_S1P)
> +		smmu->features |= ARM_SMMU_FEAT_TRANS_S1;
> +
> +	if (reg & IDR0_S2P)
> +		smmu->features |= ARM_SMMU_FEAT_TRANS_S2;
> +
> +	if (!(reg & (IDR0_S1P | IDR0_S2P))) {
> +		dev_err(smmu->dev, "no translation support!\n");
> +		return -ENXIO;
> +	}
> +
> +	/* We only support the AArch64 table format at present */
> +	switch (FIELD_GET(IDR0_TTF, reg)) {
> +	case IDR0_TTF_AARCH32_64:
> +		smmu->ias = 40;
> +		fallthrough;
> +	case IDR0_TTF_AARCH64:
> +		break;
> +	default:
> +		dev_err(smmu->dev, "AArch64 table format not supported!\n");
> +		return -ENXIO;
> +	}
> +
> +	/* ASID/VMID sizes */
> +	smmu->asid_bits = reg & IDR0_ASID16 ? 16 : 8;
> +	smmu->vmid_bits = reg & IDR0_VMID16 ? 16 : 8;
> +
> +	/* IDR1 */
> +	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR1);
> +	if (reg & (IDR1_TABLES_PRESET | IDR1_QUEUES_PRESET | IDR1_REL)) {
> +		dev_err(smmu->dev, "embedded implementation not supported\n");
> +		return -ENXIO;
> +	}
> +
> +	/* Queue sizes, capped to ensure natural alignment */
> +	smmu->cmdq.q.llq.max_n_shift = min_t(u32, CMDQ_MAX_SZ_SHIFT,
> +					     FIELD_GET(IDR1_CMDQS, reg));
> +	if (smmu->cmdq.q.llq.max_n_shift <= ilog2(CMDQ_BATCH_ENTRIES)) {
> +		/*
> +		 * We don't support splitting up batches, so one batch of
> +		 * commands plus an extra sync needs to fit inside the command
> +		 * queue. There's also no way we can handle the weird alignment
> +		 * restrictions on the base pointer for a unit-length queue.
> +		 */
> +		dev_err(smmu->dev, "command queue size <= %d entries not supported\n",
> +			CMDQ_BATCH_ENTRIES);
> +		return -ENXIO;
> +	}
> +
> +	smmu->evtq.q.llq.max_n_shift = min_t(u32, EVTQ_MAX_SZ_SHIFT,
> +					     FIELD_GET(IDR1_EVTQS, reg));
> +	smmu->priq.q.llq.max_n_shift = min_t(u32, PRIQ_MAX_SZ_SHIFT,
> +					     FIELD_GET(IDR1_PRIQS, reg));
> +
> +	/* SID/SSID sizes */
> +	smmu->ssid_bits = FIELD_GET(IDR1_SSIDSIZE, reg);
> +	smmu->sid_bits = FIELD_GET(IDR1_SIDSIZE, reg);
> +
> +	/*
> +	 * If the SMMU supports fewer bits than would fill a single L2 stream
> +	 * table, use a linear table instead.
> +	 */
> +	if (smmu->sid_bits <= STRTAB_SPLIT)
> +		smmu->features &= ~ARM_SMMU_FEAT_2_LVL_STRTAB;
> +
> +	/* IDR3 */
> +	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR3);
> +	if (FIELD_GET(IDR3_RIL, reg))
> +		smmu->features |= ARM_SMMU_FEAT_RANGE_INV;
> +
> +	/* IDR5 */
> +	reg = readl_relaxed(smmu->base + ARM_SMMU_IDR5);
> +
> +	/* Maximum number of outstanding stalls */
> +	smmu->evtq.max_stalls = FIELD_GET(IDR5_STALL_MAX, reg);
> +
> +	/* Page sizes */
> +	if (reg & IDR5_GRAN64K)
> +		smmu->pgsize_bitmap |= SZ_64K | SZ_512M;
> +	if (reg & IDR5_GRAN16K)
> +		smmu->pgsize_bitmap |= SZ_16K | SZ_32M;
> +	if (reg & IDR5_GRAN4K)
> +		smmu->pgsize_bitmap |= SZ_4K | SZ_2M | SZ_1G;
> +
> +	/* Input address size */
> +	if (FIELD_GET(IDR5_VAX, reg) == IDR5_VAX_52_BIT)
> +		smmu->features |= ARM_SMMU_FEAT_VAX;
> +
> +	/* Output address size */
> +	switch (FIELD_GET(IDR5_OAS, reg)) {
> +	case IDR5_OAS_32_BIT:
> +		smmu->oas = 32;
> +		break;
> +	case IDR5_OAS_36_BIT:
> +		smmu->oas = 36;
> +		break;
> +	case IDR5_OAS_40_BIT:
> +		smmu->oas = 40;
> +		break;
> +	case IDR5_OAS_42_BIT:
> +		smmu->oas = 42;
> +		break;
> +	case IDR5_OAS_44_BIT:
> +		smmu->oas = 44;
> +		break;
> +	case IDR5_OAS_52_BIT:
> +		smmu->oas = 52;
> +		smmu->pgsize_bitmap |= 1ULL << 42; /* 4TB */
> +		break;
> +	default:
> +		dev_info(smmu->dev,
> +			"unknown output address size. Truncating to 48-bit\n");
> +		fallthrough;
> +	case IDR5_OAS_48_BIT:
> +		smmu->oas = 48;
> +	}
> +
> +	if (arm_smmu_ops.pgsize_bitmap == -1UL)
> +		arm_smmu_ops.pgsize_bitmap = smmu->pgsize_bitmap;
> +	else
> +		arm_smmu_ops.pgsize_bitmap |= smmu->pgsize_bitmap;
> +
> +	/* Set the DMA mask for our table walker */
> +	if (dma_set_mask_and_coherent(smmu->dev, DMA_BIT_MASK(smmu->oas)))
> +		dev_warn(smmu->dev,
> +			 "failed to set DMA mask for table walker\n");
> +
> +	smmu->ias = max(smmu->ias, smmu->oas);
> +
> +	dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n",
> +		 smmu->ias, smmu->oas, smmu->features);
> +	return 0;
> +}
> +
> +#ifdef CONFIG_ACPI
> +static void acpi_smmu_get_options(u32 model, struct arm_smmu_device *smmu)
> +{
> +	switch (model) {
> +	case ACPI_IORT_SMMU_V3_CAVIUM_CN99XX:
> +		smmu->options |= ARM_SMMU_OPT_PAGE0_REGS_ONLY;
> +		break;
> +	case ACPI_IORT_SMMU_V3_HISILICON_HI161X:
> +		smmu->options |= ARM_SMMU_OPT_SKIP_PREFETCH;
> +		break;
> +	}
> +
> +	dev_notice(smmu->dev, "option mask 0x%x\n", smmu->options);
> +}
> +
> +static int arm_smmu_device_acpi_probe(struct platform_device *pdev,
> +				      struct arm_smmu_device *smmu)
> +{
> +	struct acpi_iort_smmu_v3 *iort_smmu;
> +	struct device *dev = smmu->dev;
> +	struct acpi_iort_node *node;
> +
> +	node = *(struct acpi_iort_node **)dev_get_platdata(dev);
> +
> +	/* Retrieve SMMUv3 specific data */
> +	iort_smmu = (struct acpi_iort_smmu_v3 *)node->node_data;
> +
> +	acpi_smmu_get_options(iort_smmu->model, smmu);
> +
> +	if (iort_smmu->flags & ACPI_IORT_SMMU_V3_COHACC_OVERRIDE)
> +		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
> +
> +	return 0;
> +}
> +#else
> +static inline int arm_smmu_device_acpi_probe(struct platform_device *pdev,
> +					     struct arm_smmu_device *smmu)
> +{
> +	return -ENODEV;
> +}
> +#endif
> +
> +static int arm_smmu_device_dt_probe(struct platform_device *pdev,
> +				    struct arm_smmu_device *smmu)
> +{
> +	struct device *dev = &pdev->dev;
> +	u32 cells;
> +	int ret = -EINVAL;
> +
> +	if (of_property_read_u32(dev->of_node, "#iommu-cells", &cells))
> +		dev_err(dev, "missing #iommu-cells property\n");
> +	else if (cells != 1)
> +		dev_err(dev, "invalid #iommu-cells value (%d)\n", cells);
> +	else
> +		ret = 0;
> +
> +	parse_driver_options(smmu);
> +
> +	if (of_dma_is_coherent(dev->of_node))
> +		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
> +
> +	return ret;
> +}
> +
> +static unsigned long arm_smmu_resource_size(struct arm_smmu_device *smmu)
> +{
> +	if (smmu->options & ARM_SMMU_OPT_PAGE0_REGS_ONLY)
> +		return SZ_64K;
> +	else
> +		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)
> +{
> +	struct resource res = {
> +		.flags = IORESOURCE_MEM,
> +		.start = start,
> +		.end = start + size - 1,
> +	};
> +
> +	return devm_ioremap_resource(dev, &res);
> +}
> +
> +static int arm_smmu_device_probe(struct platform_device *pdev)
> +{
> +	int irq, ret;
> +	struct resource *res;
> +	resource_size_t ioaddr;
> +	struct arm_smmu_device *smmu;
> +	struct device *dev = &pdev->dev;
> +	bool bypass;
> +
> +	smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
> +	if (!smmu) {
> +		dev_err(dev, "failed to allocate arm_smmu_device\n");
> +		return -ENOMEM;
> +	}
> +	smmu->dev = dev;
> +
> +	if (dev->of_node) {
> +		ret = arm_smmu_device_dt_probe(pdev, smmu);
> +	} else {
> +		ret = arm_smmu_device_acpi_probe(pdev, smmu);
> +		if (ret == -ENODEV)
> +			return ret;
> +	}
> +
> +	/* Set bypass mode according to firmware probing result */
> +	bypass = !!ret;
> +
> +	/* Base address */
> +	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> +	if (resource_size(res) < arm_smmu_resource_size(smmu)) {
> +		dev_err(dev, "MMIO region too small (%pr)\n", res);
> +		return -EINVAL;
> +	}
> +	ioaddr = res->start;
> +
> +	/*
> +	 * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
> +	 * the PMCG registers which are reserved by the PMU driver.
> +	 */
> +	smmu->base = arm_smmu_ioremap(dev, ioaddr, ARM_SMMU_REG_SZ);
> +	if (IS_ERR(smmu->base))
> +		return PTR_ERR(smmu->base);
> +
> +	if (arm_smmu_resource_size(smmu) > SZ_64K) {
> +		smmu->page1 = arm_smmu_ioremap(dev, ioaddr + SZ_64K,
> +					       ARM_SMMU_REG_SZ);
> +		if (IS_ERR(smmu->page1))
> +			return PTR_ERR(smmu->page1);
> +	} else {
> +		smmu->page1 = smmu->base;
> +	}
> +
> +	/* Interrupt lines */
> +
> +	irq = platform_get_irq_byname_optional(pdev, "combined");
> +	if (irq > 0)
> +		smmu->combined_irq = irq;
> +	else {
> +		irq = platform_get_irq_byname_optional(pdev, "eventq");
> +		if (irq > 0)
> +			smmu->evtq.q.irq = irq;
> +
> +		irq = platform_get_irq_byname_optional(pdev, "priq");
> +		if (irq > 0)
> +			smmu->priq.q.irq = irq;
> +
> +		irq = platform_get_irq_byname_optional(pdev, "gerror");
> +		if (irq > 0)
> +			smmu->gerr_irq = irq;
> +	}
> +	/* Probe the h/w */
> +	ret = arm_smmu_device_hw_probe(smmu);
> +	if (ret)
> +		return ret;
> +
> +	/* Initialise in-memory data structures */
> +	ret = arm_smmu_init_structures(smmu);
> +	if (ret)
> +		return ret;
> +
> +	/* Record our private device structure */
> +	platform_set_drvdata(pdev, smmu);
> +
> +	/* Reset the device */
> +	ret = arm_smmu_device_reset(smmu, bypass);
> +	if (ret)
> +		return ret;
> +
> +	/* And we're up. Go go go! */
> +	ret = iommu_device_sysfs_add(&smmu->iommu, dev, NULL,
> +				     "smmu3.%pa", &ioaddr);
> +	if (ret)
> +		return ret;
> +
> +	iommu_device_set_ops(&smmu->iommu, &arm_smmu_ops);
> +	iommu_device_set_fwnode(&smmu->iommu, dev->fwnode);
> +
> +	ret = iommu_device_register(&smmu->iommu);
> +	if (ret) {
> +		dev_err(dev, "Failed to register iommu\n");
> +		return ret;
> +	}
> +
> +	return arm_smmu_set_bus_ops(&arm_smmu_ops);
> +}
> +
> +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);
> +
> +	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", },
> +	{ },
> +};
> +MODULE_DEVICE_TABLE(of, arm_smmu_of_match);
> +
> +static struct platform_driver arm_smmu_driver = {
> +	.driver	= {
> +		.name			= "arm-smmu-v3",
> +		.of_match_table		= arm_smmu_of_match,
> +		.suppress_bind_attrs	= true,
> +	},
> +	.probe	= arm_smmu_device_probe,
> +	.remove	= arm_smmu_device_remove,
> +	.shutdown = arm_smmu_device_shutdown,
> +};
> +module_platform_driver(arm_smmu_driver);
> +
> +MODULE_DESCRIPTION("IOMMU API for ARM architected SMMUv3 implementations");
> +MODULE_AUTHOR("Will Deacon <will@kernel.org>");
> +MODULE_ALIAS("platform:arm-smmu-v3");
> +MODULE_LICENSE("GPL v2");
> -- 
> 2.17.1
> 


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 2/8] xen/arm: revert atomic operation related command-queue insertion patch
  2020-11-26 17:02 ` [PATCH v2 2/8] xen/arm: revert atomic operation related command-queue insertion patch Rahul Singh
@ 2020-12-01 22:23   ` Stefano Stabellini
  2020-12-02 13:05     ` Rahul Singh
  2020-12-02 13:44   ` Julien Grall
  1 sibling, 1 reply; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-01 22:23 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, bertrand.marquis, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

On Thu, 26 Nov 2020, Rahul Singh wrote:
> Linux SMMUv3 code implements the commands-queue insertion based on
> atomic operations implemented in Linux. Atomic functions used by the
> commands-queue insertion is not implemented in XEN therefore revert the
> patch that implemented the commands-queue insertion based on atomic
> operations.
> 
> Once the proper atomic operations will be available in XEN the driver
> can be updated.
> 
> Reverted the commit 587e6c10a7ce89a5924fdbeff2ec524fbd6a124b
> iommu/arm-smmu-v3: Reduce contention during command-queue insertion

I checked 587e6c10a7ce89a5924fdbeff2ec524fbd6a124b: this patch does more
than just reverting 587e6c10a7ce89a5924fdbeff2ec524fbd6a124b. It looks
like it is also reverting edd0351e7bc49555d8b5ad8438a65a7ca262c9f0 and
some other commits.

Please can you provide a complete list of reverted commits? I would like
to be able to do the reverts myself on the linux tree and see that the
driver textually matches the one on the xen tree with this patch
applied.



> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
> ---
>  xen/drivers/passthrough/arm/smmu-v3.c | 847 ++++++--------------------
>  1 file changed, 180 insertions(+), 667 deletions(-)
> 
> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
> index c192544e87..97eac61ea4 100644
> --- a/xen/drivers/passthrough/arm/smmu-v3.c
> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
> @@ -330,15 +330,6 @@
>  #define CMDQ_ERR_CERROR_ABT_IDX		2
>  #define CMDQ_ERR_CERROR_ATC_INV_IDX	3
>  
> -#define CMDQ_PROD_OWNED_FLAG		Q_OVERFLOW_FLAG
> -
> -/*
> - * This is used to size the command queue and therefore must be at least
> - * BITS_PER_LONG so that the valid_map works correctly (it relies on the
> - * total number of queue entries being a multiple of BITS_PER_LONG).
> - */
> -#define CMDQ_BATCH_ENTRIES		BITS_PER_LONG
> -
>  #define CMDQ_0_OP			GENMASK_ULL(7, 0)
>  #define CMDQ_0_SSV			(1UL << 11)
>  
> @@ -407,8 +398,9 @@
>  #define PRIQ_1_ADDR_MASK		GENMASK_ULL(63, 12)
>  
>  /* High-level queue structures */
> -#define ARM_SMMU_POLL_TIMEOUT_US	1000000 /* 1s! */
> -#define ARM_SMMU_POLL_SPIN_COUNT	10
> +#define ARM_SMMU_POLL_TIMEOUT_US	100
> +#define ARM_SMMU_CMDQ_SYNC_TIMEOUT_US	1000000 /* 1s! */
> +#define ARM_SMMU_CMDQ_SYNC_SPIN_COUNT	10
>  
>  #define MSI_IOVA_BASE			0x8000000
>  #define MSI_IOVA_LENGTH			0x100000
> @@ -513,24 +505,15 @@ struct arm_smmu_cmdq_ent {
>  
>  		#define CMDQ_OP_CMD_SYNC	0x46
>  		struct {
> +			u32			msidata;
>  			u64			msiaddr;
>  		} sync;
>  	};
>  };
>  
>  struct arm_smmu_ll_queue {
> -	union {
> -		u64			val;
> -		struct {
> -			u32		prod;
> -			u32		cons;
> -		};
> -		struct {
> -			atomic_t	prod;
> -			atomic_t	cons;
> -		} atomic;
> -		u8			__pad[SMP_CACHE_BYTES];
> -	} ____cacheline_aligned_in_smp;
> +	u32				prod;
> +	u32				cons;
>  	u32				max_n_shift;
>  };
>  
> @@ -548,23 +531,9 @@ struct arm_smmu_queue {
>  	u32 __iomem			*cons_reg;
>  };
>  
> -struct arm_smmu_queue_poll {
> -	ktime_t				timeout;
> -	unsigned int			delay;
> -	unsigned int			spin_cnt;
> -	bool				wfe;
> -};
> -
>  struct arm_smmu_cmdq {
>  	struct arm_smmu_queue		q;
> -	atomic_long_t			*valid_map;
> -	atomic_t			owner_prod;
> -	atomic_t			lock;
> -};
> -
> -struct arm_smmu_cmdq_batch {
> -	u64				cmds[CMDQ_BATCH_ENTRIES * CMDQ_ENT_DWORDS];
> -	int				num;
> +	spinlock_t			lock;
>  };
>  
>  struct arm_smmu_evtq {
> @@ -660,6 +629,8 @@ struct arm_smmu_device {
>  
>  	int				gerr_irq;
>  	int				combined_irq;
> +	u32				sync_nr;
> +	u8				prev_cmd_opcode;
>  
>  	unsigned long			ias; /* IPA */
>  	unsigned long			oas; /* PA */
> @@ -677,6 +648,12 @@ struct arm_smmu_device {
>  
>  	struct arm_smmu_strtab_cfg	strtab_cfg;
>  
> +	/* Hi16xx adds an extra 32 bits of goodness to its MSI payload */
> +	union {
> +		u32			sync_count;
> +		u64			padding;
> +	};
> +
>  	/* IOMMU core code handle */
>  	struct iommu_device		iommu;
>  };
> @@ -763,21 +740,6 @@ static void parse_driver_options(struct arm_smmu_device *smmu)
>  }
>  
>  /* Low-level queue manipulation functions */
> -static bool queue_has_space(struct arm_smmu_ll_queue *q, u32 n)
> -{
> -	u32 space, prod, cons;
> -
> -	prod = Q_IDX(q, q->prod);
> -	cons = Q_IDX(q, q->cons);
> -
> -	if (Q_WRP(q, q->prod) == Q_WRP(q, q->cons))
> -		space = (1 << q->max_n_shift) - (prod - cons);
> -	else
> -		space = cons - prod;
> -
> -	return space >= n;
> -}
> -
>  static bool queue_full(struct arm_smmu_ll_queue *q)
>  {
>  	return Q_IDX(q, q->prod) == Q_IDX(q, q->cons) &&
> @@ -790,12 +752,9 @@ static bool queue_empty(struct arm_smmu_ll_queue *q)
>  	       Q_WRP(q, q->prod) == Q_WRP(q, q->cons);
>  }
>  
> -static bool queue_consumed(struct arm_smmu_ll_queue *q, u32 prod)
> +static void queue_sync_cons_in(struct arm_smmu_queue *q)
>  {
> -	return ((Q_WRP(q, q->cons) == Q_WRP(q, prod)) &&
> -		(Q_IDX(q, q->cons) > Q_IDX(q, prod))) ||
> -	       ((Q_WRP(q, q->cons) != Q_WRP(q, prod)) &&
> -		(Q_IDX(q, q->cons) <= Q_IDX(q, prod)));
> +	q->llq.cons = readl_relaxed(q->cons_reg);
>  }
>  
>  static void queue_sync_cons_out(struct arm_smmu_queue *q)
> @@ -826,34 +785,46 @@ static int queue_sync_prod_in(struct arm_smmu_queue *q)
>  	return ret;
>  }
>  
> -static u32 queue_inc_prod_n(struct arm_smmu_ll_queue *q, int n)
> +static void queue_sync_prod_out(struct arm_smmu_queue *q)
>  {
> -	u32 prod = (Q_WRP(q, q->prod) | Q_IDX(q, q->prod)) + n;
> -	return Q_OVF(q->prod) | Q_WRP(q, prod) | Q_IDX(q, prod);
> +	writel(q->llq.prod, q->prod_reg);
>  }
>  
> -static void queue_poll_init(struct arm_smmu_device *smmu,
> -			    struct arm_smmu_queue_poll *qp)
> +static void queue_inc_prod(struct arm_smmu_ll_queue *q)
>  {
> -	qp->delay = 1;
> -	qp->spin_cnt = 0;
> -	qp->wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
> -	qp->timeout = ktime_add_us(ktime_get(), ARM_SMMU_POLL_TIMEOUT_US);
> +	u32 prod = (Q_WRP(q, q->prod) | Q_IDX(q, q->prod)) + 1;
> +	q->prod = Q_OVF(q->prod) | Q_WRP(q, prod) | Q_IDX(q, prod);
>  }
>  
> -static int queue_poll(struct arm_smmu_queue_poll *qp)
> +/*
> + * Wait for the SMMU to consume items. If sync is true, wait until the queue
> + * is empty. Otherwise, wait until there is at least one free slot.
> + */
> +static int queue_poll_cons(struct arm_smmu_queue *q, bool sync, bool wfe)
>  {
> -	if (ktime_compare(ktime_get(), qp->timeout) > 0)
> -		return -ETIMEDOUT;
> +	ktime_t timeout;
> +	unsigned int delay = 1, spin_cnt = 0;
>  
> -	if (qp->wfe) {
> -		wfe();
> -	} else if (++qp->spin_cnt < ARM_SMMU_POLL_SPIN_COUNT) {
> -		cpu_relax();
> -	} else {
> -		udelay(qp->delay);
> -		qp->delay *= 2;
> -		qp->spin_cnt = 0;
> +	/* Wait longer if it's a CMD_SYNC */
> +	timeout = ktime_add_us(ktime_get(), sync ?
> +					    ARM_SMMU_CMDQ_SYNC_TIMEOUT_US :
> +					    ARM_SMMU_POLL_TIMEOUT_US);
> +
> +	while (queue_sync_cons_in(q),
> +	      (sync ? !queue_empty(&q->llq) : queue_full(&q->llq))) {
> +		if (ktime_compare(ktime_get(), timeout) > 0)
> +			return -ETIMEDOUT;
> +
> +		if (wfe) {
> +			wfe();
> +		} else if (++spin_cnt < ARM_SMMU_CMDQ_SYNC_SPIN_COUNT) {
> +			cpu_relax();
> +			continue;
> +		} else {
> +			udelay(delay);
> +			delay *= 2;
> +			spin_cnt = 0;
> +		}
>  	}
>  
>  	return 0;
> @@ -867,6 +838,17 @@ static void queue_write(__le64 *dst, u64 *src, size_t n_dwords)
>  		*dst++ = cpu_to_le64(*src++);
>  }
>  
> +static int queue_insert_raw(struct arm_smmu_queue *q, u64 *ent)
> +{
> +	if (queue_full(&q->llq))
> +		return -ENOSPC;
> +
> +	queue_write(Q_ENT(q, q->llq.prod), ent, q->ent_dwords);
> +	queue_inc_prod(&q->llq);
> +	queue_sync_prod_out(q);
> +	return 0;
> +}
> +
>  static void queue_read(__le64 *dst, u64 *src, size_t n_dwords)
>  {
>  	int i;
> @@ -964,14 +946,20 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
>  		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_RESP, ent->pri.resp);
>  		break;
>  	case CMDQ_OP_CMD_SYNC:
> -		if (ent->sync.msiaddr) {
> +		if (ent->sync.msiaddr)
>  			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_IRQ);
> -			cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
> -		} else {
> +		else
>  			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
> -		}
>  		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSH, ARM_SMMU_SH_ISH);
>  		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIATTR, ARM_SMMU_MEMATTR_OIWB);
> +		/*
> +		 * Commands are written little-endian, but we want the SMMU to
> +		 * receive MSIData, and thus write it back to memory, in CPU
> +		 * byte order, so big-endian needs an extra byteswap here.
> +		 */
> +		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIDATA,
> +				     cpu_to_le32(ent->sync.msidata));
> +		cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
>  		break;
>  	default:
>  		return -ENOENT;
> @@ -980,27 +968,6 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
>  	return 0;
>  }
>  
> -static void arm_smmu_cmdq_build_sync_cmd(u64 *cmd, struct arm_smmu_device *smmu,
> -					 u32 prod)
> -{
> -	struct arm_smmu_queue *q = &smmu->cmdq.q;
> -	struct arm_smmu_cmdq_ent ent = {
> -		.opcode = CMDQ_OP_CMD_SYNC,
> -	};
> -
> -	/*
> -	 * Beware that Hi16xx adds an extra 32 bits of goodness to its MSI
> -	 * payload, so the write will zero the entire command on that platform.
> -	 */
> -	if (smmu->features & ARM_SMMU_FEAT_MSI &&
> -	    smmu->features & ARM_SMMU_FEAT_COHERENCY) {
> -		ent.sync.msiaddr = q->base_dma + Q_IDX(&q->llq, prod) *
> -				   q->ent_dwords * 8;
> -	}
> -
> -	arm_smmu_cmdq_build_cmd(cmd, &ent);
> -}
> -
>  static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
>  {
>  	static const char *cerror_str[] = {
> @@ -1058,474 +1025,109 @@ static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
>  	queue_write(Q_ENT(q, cons), cmd, q->ent_dwords);
>  }
>  
> -/*
> - * Command queue locking.
> - * This is a form of bastardised rwlock with the following major changes:
> - *
> - * - The only LOCK routines are exclusive_trylock() and shared_lock().
> - *   Neither have barrier semantics, and instead provide only a control
> - *   dependency.
> - *
> - * - The UNLOCK routines are supplemented with shared_tryunlock(), which
> - *   fails if the caller appears to be the last lock holder (yes, this is
> - *   racy). All successful UNLOCK routines have RELEASE semantics.
> - */
> -static void arm_smmu_cmdq_shared_lock(struct arm_smmu_cmdq *cmdq)
> +static void arm_smmu_cmdq_insert_cmd(struct arm_smmu_device *smmu, u64 *cmd)
>  {
> -	int val;
> -
> -	/*
> -	 * We can try to avoid the cmpxchg() loop by simply incrementing the
> -	 * lock counter. When held in exclusive state, the lock counter is set
> -	 * to INT_MIN so these increments won't hurt as the value will remain
> -	 * negative.
> -	 */
> -	if (atomic_fetch_inc_relaxed(&cmdq->lock) >= 0)
> -		return;
> -
> -	do {
> -		val = atomic_cond_read_relaxed(&cmdq->lock, VAL >= 0);
> -	} while (atomic_cmpxchg_relaxed(&cmdq->lock, val, val + 1) != val);
> -}
> -
> -static void arm_smmu_cmdq_shared_unlock(struct arm_smmu_cmdq *cmdq)
> -{
> -	(void)atomic_dec_return_release(&cmdq->lock);
> -}
> -
> -static bool arm_smmu_cmdq_shared_tryunlock(struct arm_smmu_cmdq *cmdq)
> -{
> -	if (atomic_read(&cmdq->lock) == 1)
> -		return false;
> -
> -	arm_smmu_cmdq_shared_unlock(cmdq);
> -	return true;
> -}
> -
> -#define arm_smmu_cmdq_exclusive_trylock_irqsave(cmdq, flags)		\
> -({									\
> -	bool __ret;							\
> -	local_irq_save(flags);						\
> -	__ret = !atomic_cmpxchg_relaxed(&cmdq->lock, 0, INT_MIN);	\
> -	if (!__ret)							\
> -		local_irq_restore(flags);				\
> -	__ret;								\
> -})
> -
> -#define arm_smmu_cmdq_exclusive_unlock_irqrestore(cmdq, flags)		\
> -({									\
> -	atomic_set_release(&cmdq->lock, 0);				\
> -	local_irq_restore(flags);					\
> -})
> -
> -
> -/*
> - * Command queue insertion.
> - * This is made fiddly by our attempts to achieve some sort of scalability
> - * since there is one queue shared amongst all of the CPUs in the system.  If
> - * you like mixed-size concurrency, dependency ordering and relaxed atomics,
> - * then you'll *love* this monstrosity.
> - *
> - * The basic idea is to split the queue up into ranges of commands that are
> - * owned by a given CPU; the owner may not have written all of the commands
> - * itself, but is responsible for advancing the hardware prod pointer when
> - * the time comes. The algorithm is roughly:
> - *
> - * 	1. Allocate some space in the queue. At this point we also discover
> - *	   whether the head of the queue is currently owned by another CPU,
> - *	   or whether we are the owner.
> - *
> - *	2. Write our commands into our allocated slots in the queue.
> - *
> - *	3. Mark our slots as valid in arm_smmu_cmdq.valid_map.
> - *
> - *	4. If we are an owner:
> - *		a. Wait for the previous owner to finish.
> - *		b. Mark the queue head as unowned, which tells us the range
> - *		   that we are responsible for publishing.
> - *		c. Wait for all commands in our owned range to become valid.
> - *		d. Advance the hardware prod pointer.
> - *		e. Tell the next owner we've finished.
> - *
> - *	5. If we are inserting a CMD_SYNC (we may or may not have been an
> - *	   owner), then we need to stick around until it has completed:
> - *		a. If we have MSIs, the SMMU can write back into the CMD_SYNC
> - *		   to clear the first 4 bytes.
> - *		b. Otherwise, we spin waiting for the hardware cons pointer to
> - *		   advance past our command.
> - *
> - * The devil is in the details, particularly the use of locking for handling
> - * SYNC completion and freeing up space in the queue before we think that it is
> - * full.
> - */
> -static void __arm_smmu_cmdq_poll_set_valid_map(struct arm_smmu_cmdq *cmdq,
> -					       u32 sprod, u32 eprod, bool set)
> -{
> -	u32 swidx, sbidx, ewidx, ebidx;
> -	struct arm_smmu_ll_queue llq = {
> -		.max_n_shift	= cmdq->q.llq.max_n_shift,
> -		.prod		= sprod,
> -	};
> -
> -	ewidx = BIT_WORD(Q_IDX(&llq, eprod));
> -	ebidx = Q_IDX(&llq, eprod) % BITS_PER_LONG;
> -
> -	while (llq.prod != eprod) {
> -		unsigned long mask;
> -		atomic_long_t *ptr;
> -		u32 limit = BITS_PER_LONG;
> -
> -		swidx = BIT_WORD(Q_IDX(&llq, llq.prod));
> -		sbidx = Q_IDX(&llq, llq.prod) % BITS_PER_LONG;
> -
> -		ptr = &cmdq->valid_map[swidx];
> -
> -		if ((swidx == ewidx) && (sbidx < ebidx))
> -			limit = ebidx;
> -
> -		mask = GENMASK(limit - 1, sbidx);
> -
> -		/*
> -		 * The valid bit is the inverse of the wrap bit. This means
> -		 * that a zero-initialised queue is invalid and, after marking
> -		 * all entries as valid, they become invalid again when we
> -		 * wrap.
> -		 */
> -		if (set) {
> -			atomic_long_xor(mask, ptr);
> -		} else { /* Poll */
> -			unsigned long valid;
> +	struct arm_smmu_queue *q = &smmu->cmdq.q;
> +	bool wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
>  
> -			valid = (ULONG_MAX + !!Q_WRP(&llq, llq.prod)) & mask;
> -			atomic_long_cond_read_relaxed(ptr, (VAL & mask) == valid);
> -		}
> +	smmu->prev_cmd_opcode = FIELD_GET(CMDQ_0_OP, cmd[0]);
>  
> -		llq.prod = queue_inc_prod_n(&llq, limit - sbidx);
> +	while (queue_insert_raw(q, cmd) == -ENOSPC) {
> +		if (queue_poll_cons(q, false, wfe))
> +			dev_err_ratelimited(smmu->dev, "CMDQ timeout\n");
>  	}
>  }
>  
> -/* Mark all entries in the range [sprod, eprod) as valid */
> -static void arm_smmu_cmdq_set_valid_map(struct arm_smmu_cmdq *cmdq,
> -					u32 sprod, u32 eprod)
> -{
> -	__arm_smmu_cmdq_poll_set_valid_map(cmdq, sprod, eprod, true);
> -}
> -
> -/* Wait for all entries in the range [sprod, eprod) to become valid */
> -static void arm_smmu_cmdq_poll_valid_map(struct arm_smmu_cmdq *cmdq,
> -					 u32 sprod, u32 eprod)
> -{
> -	__arm_smmu_cmdq_poll_set_valid_map(cmdq, sprod, eprod, false);
> -}
> -
> -/* Wait for the command queue to become non-full */
> -static int arm_smmu_cmdq_poll_until_not_full(struct arm_smmu_device *smmu,
> -					     struct arm_smmu_ll_queue *llq)
> +static void arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
> +				    struct arm_smmu_cmdq_ent *ent)
>  {
> +	u64 cmd[CMDQ_ENT_DWORDS];
>  	unsigned long flags;
> -	struct arm_smmu_queue_poll qp;
> -	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
> -	int ret = 0;
>  
> -	/*
> -	 * Try to update our copy of cons by grabbing exclusive cmdq access. If
> -	 * that fails, spin until somebody else updates it for us.
> -	 */
> -	if (arm_smmu_cmdq_exclusive_trylock_irqsave(cmdq, flags)) {
> -		WRITE_ONCE(cmdq->q.llq.cons, readl_relaxed(cmdq->q.cons_reg));
> -		arm_smmu_cmdq_exclusive_unlock_irqrestore(cmdq, flags);
> -		llq->val = READ_ONCE(cmdq->q.llq.val);
> -		return 0;
> +	if (arm_smmu_cmdq_build_cmd(cmd, ent)) {
> +		dev_warn(smmu->dev, "ignoring unknown CMDQ opcode 0x%x\n",
> +			 ent->opcode);
> +		return;
>  	}
>  
> -	queue_poll_init(smmu, &qp);
> -	do {
> -		llq->val = READ_ONCE(smmu->cmdq.q.llq.val);
> -		if (!queue_full(llq))
> -			break;
> -
> -		ret = queue_poll(&qp);
> -	} while (!ret);
> -
> -	return ret;
> -}
> -
> -/*
> - * Wait until the SMMU signals a CMD_SYNC completion MSI.
> - * Must be called with the cmdq lock held in some capacity.
> - */
> -static int __arm_smmu_cmdq_poll_until_msi(struct arm_smmu_device *smmu,
> -					  struct arm_smmu_ll_queue *llq)
> -{
> -	int ret = 0;
> -	struct arm_smmu_queue_poll qp;
> -	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
> -	u32 *cmd = (u32 *)(Q_ENT(&cmdq->q, llq->prod));
> -
> -	queue_poll_init(smmu, &qp);
> -
> -	/*
> -	 * The MSI won't generate an event, since it's being written back
> -	 * into the command queue.
> -	 */
> -	qp.wfe = false;
> -	smp_cond_load_relaxed(cmd, !VAL || (ret = queue_poll(&qp)));
> -	llq->cons = ret ? llq->prod : queue_inc_prod_n(llq, 1);
> -	return ret;
> +	spin_lock_irqsave(&smmu->cmdq.lock, flags);
> +	arm_smmu_cmdq_insert_cmd(smmu, cmd);
> +	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
>  }
>  
>  /*
> - * Wait until the SMMU cons index passes llq->prod.
> - * Must be called with the cmdq lock held in some capacity.
> + * The difference between val and sync_idx is bounded by the maximum size of
> + * a queue at 2^20 entries, so 32 bits is plenty for wrap-safe arithmetic.
>   */
> -static int __arm_smmu_cmdq_poll_until_consumed(struct arm_smmu_device *smmu,
> -					       struct arm_smmu_ll_queue *llq)
> +static int __arm_smmu_sync_poll_msi(struct arm_smmu_device *smmu, u32 sync_idx)
>  {
> -	struct arm_smmu_queue_poll qp;
> -	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
> -	u32 prod = llq->prod;
> -	int ret = 0;
> +	ktime_t timeout;
> +	u32 val;
>  
> -	queue_poll_init(smmu, &qp);
> -	llq->val = READ_ONCE(smmu->cmdq.q.llq.val);
> -	do {
> -		if (queue_consumed(llq, prod))
> -			break;
> -
> -		ret = queue_poll(&qp);
> -
> -		/*
> -		 * This needs to be a readl() so that our subsequent call
> -		 * to arm_smmu_cmdq_shared_tryunlock() can fail accurately.
> -		 *
> -		 * Specifically, we need to ensure that we observe all
> -		 * shared_lock()s by other CMD_SYNCs that share our owner,
> -		 * so that a failing call to tryunlock() means that we're
> -		 * the last one out and therefore we can safely advance
> -		 * cmdq->q.llq.cons. Roughly speaking:
> -		 *
> -		 * CPU 0		CPU1			CPU2 (us)
> -		 *
> -		 * if (sync)
> -		 * 	shared_lock();
> -		 *
> -		 * dma_wmb();
> -		 * set_valid_map();
> -		 *
> -		 * 			if (owner) {
> -		 *				poll_valid_map();
> -		 *				<control dependency>
> -		 *				writel(prod_reg);
> -		 *
> -		 *						readl(cons_reg);
> -		 *						tryunlock();
> -		 *
> -		 * Requires us to see CPU 0's shared_lock() acquisition.
> -		 */
> -		llq->cons = readl(cmdq->q.cons_reg);
> -	} while (!ret);
> +	timeout = ktime_add_us(ktime_get(), ARM_SMMU_CMDQ_SYNC_TIMEOUT_US);
> +	val = smp_cond_load_acquire(&smmu->sync_count,
> +				    (int)(VAL - sync_idx) >= 0 ||
> +				    !ktime_before(ktime_get(), timeout));
>  
> -	return ret;
> +	return (int)(val - sync_idx) < 0 ? -ETIMEDOUT : 0;
>  }
>  
> -static int arm_smmu_cmdq_poll_until_sync(struct arm_smmu_device *smmu,
> -					 struct arm_smmu_ll_queue *llq)
> +static int __arm_smmu_cmdq_issue_sync_msi(struct arm_smmu_device *smmu)
>  {
> -	if (smmu->features & ARM_SMMU_FEAT_MSI &&
> -	    smmu->features & ARM_SMMU_FEAT_COHERENCY)
> -		return __arm_smmu_cmdq_poll_until_msi(smmu, llq);
> -
> -	return __arm_smmu_cmdq_poll_until_consumed(smmu, llq);
> -}
> -
> -static void arm_smmu_cmdq_write_entries(struct arm_smmu_cmdq *cmdq, u64 *cmds,
> -					u32 prod, int n)
> -{
> -	int i;
> -	struct arm_smmu_ll_queue llq = {
> -		.max_n_shift	= cmdq->q.llq.max_n_shift,
> -		.prod		= prod,
> -	};
> -
> -	for (i = 0; i < n; ++i) {
> -		u64 *cmd = &cmds[i * CMDQ_ENT_DWORDS];
> -
> -		prod = queue_inc_prod_n(&llq, i);
> -		queue_write(Q_ENT(&cmdq->q, prod), cmd, CMDQ_ENT_DWORDS);
> -	}
> -}
> -
> -/*
> - * This is the actual insertion function, and provides the following
> - * ordering guarantees to callers:
> - *
> - * - There is a dma_wmb() before publishing any commands to the queue.
> - *   This can be relied upon to order prior writes to data structures
> - *   in memory (such as a CD or an STE) before the command.
> - *
> - * - On completion of a CMD_SYNC, there is a control dependency.
> - *   This can be relied upon to order subsequent writes to memory (e.g.
> - *   freeing an IOVA) after completion of the CMD_SYNC.
> - *
> - * - Command insertion is totally ordered, so if two CPUs each race to
> - *   insert their own list of commands then all of the commands from one
> - *   CPU will appear before any of the commands from the other CPU.
> - */
> -static int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
> -				       u64 *cmds, int n, bool sync)
> -{
> -	u64 cmd_sync[CMDQ_ENT_DWORDS];
> -	u32 prod;
> +	u64 cmd[CMDQ_ENT_DWORDS];
>  	unsigned long flags;
> -	bool owner;
> -	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
> -	struct arm_smmu_ll_queue llq = {
> -		.max_n_shift = cmdq->q.llq.max_n_shift,
> -	}, head = llq;
> -	int ret = 0;
> -
> -	/* 1. Allocate some space in the queue */
> -	local_irq_save(flags);
> -	llq.val = READ_ONCE(cmdq->q.llq.val);
> -	do {
> -		u64 old;
> -
> -		while (!queue_has_space(&llq, n + sync)) {
> -			local_irq_restore(flags);
> -			if (arm_smmu_cmdq_poll_until_not_full(smmu, &llq))
> -				dev_err_ratelimited(smmu->dev, "CMDQ timeout\n");
> -			local_irq_save(flags);
> -		}
> -
> -		head.cons = llq.cons;
> -		head.prod = queue_inc_prod_n(&llq, n + sync) |
> -					     CMDQ_PROD_OWNED_FLAG;
> -
> -		old = cmpxchg_relaxed(&cmdq->q.llq.val, llq.val, head.val);
> -		if (old == llq.val)
> -			break;
> -
> -		llq.val = old;
> -	} while (1);
> -	owner = !(llq.prod & CMDQ_PROD_OWNED_FLAG);
> -	head.prod &= ~CMDQ_PROD_OWNED_FLAG;
> -	llq.prod &= ~CMDQ_PROD_OWNED_FLAG;
> -
> -	/*
> -	 * 2. Write our commands into the queue
> -	 * Dependency ordering from the cmpxchg() loop above.
> -	 */
> -	arm_smmu_cmdq_write_entries(cmdq, cmds, llq.prod, n);
> -	if (sync) {
> -		prod = queue_inc_prod_n(&llq, n);
> -		arm_smmu_cmdq_build_sync_cmd(cmd_sync, smmu, prod);
> -		queue_write(Q_ENT(&cmdq->q, prod), cmd_sync, CMDQ_ENT_DWORDS);
> -
> -		/*
> -		 * In order to determine completion of our CMD_SYNC, we must
> -		 * ensure that the queue can't wrap twice without us noticing.
> -		 * We achieve that by taking the cmdq lock as shared before
> -		 * marking our slot as valid.
> -		 */
> -		arm_smmu_cmdq_shared_lock(cmdq);
> -	}
> -
> -	/* 3. Mark our slots as valid, ensuring commands are visible first */
> -	dma_wmb();
> -	arm_smmu_cmdq_set_valid_map(cmdq, llq.prod, head.prod);
> -
> -	/* 4. If we are the owner, take control of the SMMU hardware */
> -	if (owner) {
> -		/* a. Wait for previous owner to finish */
> -		atomic_cond_read_relaxed(&cmdq->owner_prod, VAL == llq.prod);
> -
> -		/* b. Stop gathering work by clearing the owned flag */
> -		prod = atomic_fetch_andnot_relaxed(CMDQ_PROD_OWNED_FLAG,
> -						   &cmdq->q.llq.atomic.prod);
> -		prod &= ~CMDQ_PROD_OWNED_FLAG;
> +	struct arm_smmu_cmdq_ent  ent = {
> +		.opcode = CMDQ_OP_CMD_SYNC,
> +		.sync	= {
> +			.msiaddr = virt_to_phys(&smmu->sync_count),
> +		},
> +	};
>  
> -		/*
> -		 * c. Wait for any gathered work to be written to the queue.
> -		 * Note that we read our own entries so that we have the control
> -		 * dependency required by (d).
> -		 */
> -		arm_smmu_cmdq_poll_valid_map(cmdq, llq.prod, prod);
> +	spin_lock_irqsave(&smmu->cmdq.lock, flags);
>  
> -		/*
> -		 * d. Advance the hardware prod pointer
> -		 * Control dependency ordering from the entries becoming valid.
> -		 */
> -		writel_relaxed(prod, cmdq->q.prod_reg);
> -
> -		/*
> -		 * e. Tell the next owner we're done
> -		 * Make sure we've updated the hardware first, so that we don't
> -		 * race to update prod and potentially move it backwards.
> -		 */
> -		atomic_set_release(&cmdq->owner_prod, prod);
> +	/* Piggy-back on the previous command if it's a SYNC */
> +	if (smmu->prev_cmd_opcode == CMDQ_OP_CMD_SYNC) {
> +		ent.sync.msidata = smmu->sync_nr;
> +	} else {
> +		ent.sync.msidata = ++smmu->sync_nr;
> +		arm_smmu_cmdq_build_cmd(cmd, &ent);
> +		arm_smmu_cmdq_insert_cmd(smmu, cmd);
>  	}
>  
> -	/* 5. If we are inserting a CMD_SYNC, we must wait for it to complete */
> -	if (sync) {
> -		llq.prod = queue_inc_prod_n(&llq, n);
> -		ret = arm_smmu_cmdq_poll_until_sync(smmu, &llq);
> -		if (ret) {
> -			dev_err_ratelimited(smmu->dev,
> -					    "CMD_SYNC timeout at 0x%08x [hwprod 0x%08x, hwcons 0x%08x]\n",
> -					    llq.prod,
> -					    readl_relaxed(cmdq->q.prod_reg),
> -					    readl_relaxed(cmdq->q.cons_reg));
> -		}
> -
> -		/*
> -		 * Try to unlock the cmdq lock. This will fail if we're the last
> -		 * reader, in which case we can safely update cmdq->q.llq.cons
> -		 */
> -		if (!arm_smmu_cmdq_shared_tryunlock(cmdq)) {
> -			WRITE_ONCE(cmdq->q.llq.cons, llq.cons);
> -			arm_smmu_cmdq_shared_unlock(cmdq);
> -		}
> -	}
> +	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
>  
> -	local_irq_restore(flags);
> -	return ret;
> +	return __arm_smmu_sync_poll_msi(smmu, ent.sync.msidata);
>  }
>  
> -static int arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
> -				   struct arm_smmu_cmdq_ent *ent)
> +static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>  {
>  	u64 cmd[CMDQ_ENT_DWORDS];
> +	unsigned long flags;
> +	bool wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
> +	struct arm_smmu_cmdq_ent ent = { .opcode = CMDQ_OP_CMD_SYNC };
> +	int ret;
>  
> -	if (arm_smmu_cmdq_build_cmd(cmd, ent)) {
> -		dev_warn(smmu->dev, "ignoring unknown CMDQ opcode 0x%x\n",
> -			 ent->opcode);
> -		return -EINVAL;
> -	}
> +	arm_smmu_cmdq_build_cmd(cmd, &ent);
>  
> -	return arm_smmu_cmdq_issue_cmdlist(smmu, cmd, 1, false);
> -}
> +	spin_lock_irqsave(&smmu->cmdq.lock, flags);
> +	arm_smmu_cmdq_insert_cmd(smmu, cmd);
> +	ret = queue_poll_cons(&smmu->cmdq.q, true, wfe);
> +	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
>  
> -static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
> -{
> -	return arm_smmu_cmdq_issue_cmdlist(smmu, NULL, 0, true);
> +	return ret;
>  }
>  
> -static void arm_smmu_cmdq_batch_add(struct arm_smmu_device *smmu,
> -				    struct arm_smmu_cmdq_batch *cmds,
> -				    struct arm_smmu_cmdq_ent *cmd)
> +static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>  {
> -	if (cmds->num == CMDQ_BATCH_ENTRIES) {
> -		arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, false);
> -		cmds->num = 0;
> -	}
> -	arm_smmu_cmdq_build_cmd(&cmds->cmds[cmds->num * CMDQ_ENT_DWORDS], cmd);
> -	cmds->num++;
> -}
> +	int ret;
> +	bool msi = (smmu->features & ARM_SMMU_FEAT_MSI) &&
> +		   (smmu->features & ARM_SMMU_FEAT_COHERENCY);
>  
> -static int arm_smmu_cmdq_batch_submit(struct arm_smmu_device *smmu,
> -				      struct arm_smmu_cmdq_batch *cmds)
> -{
> -	return arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, true);
> +	ret = msi ? __arm_smmu_cmdq_issue_sync_msi(smmu)
> +		  : __arm_smmu_cmdq_issue_sync(smmu);
> +	if (ret)
> +		dev_err_ratelimited(smmu->dev, "CMD_SYNC timeout\n");
> +	return ret;
>  }
>  
>  /* Context descriptor manipulation functions */
> @@ -1535,7 +1137,6 @@ static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
>  	size_t i;
>  	unsigned long flags;
>  	struct arm_smmu_master *master;
> -	struct arm_smmu_cmdq_batch cmds = {};
>  	struct arm_smmu_device *smmu = smmu_domain->smmu;
>  	struct arm_smmu_cmdq_ent cmd = {
>  		.opcode	= CMDQ_OP_CFGI_CD,
> @@ -1549,12 +1150,12 @@ static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
>  	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
>  		for (i = 0; i < master->num_sids; i++) {
>  			cmd.cfgi.sid = master->sids[i];
> -			arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
> +			arm_smmu_cmdq_issue_cmd(smmu, &cmd);
>  		}
>  	}
>  	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
>  
> -	arm_smmu_cmdq_batch_submit(smmu, &cmds);
> +	arm_smmu_cmdq_issue_sync(smmu);
>  }
>  
>  static int arm_smmu_alloc_cd_leaf_table(struct arm_smmu_device *smmu,
> @@ -2189,16 +1790,17 @@ arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size,
>  	cmd->atc.size	= log2_span;
>  }
>  
> -static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
> +static int arm_smmu_atc_inv_master(struct arm_smmu_master *master,
> +				   struct arm_smmu_cmdq_ent *cmd)
>  {
>  	int i;
> -	struct arm_smmu_cmdq_ent cmd;
>  
> -	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
> +	if (!master->ats_enabled)
> +		return 0;
>  
>  	for (i = 0; i < master->num_sids; i++) {
> -		cmd.atc.sid = master->sids[i];
> -		arm_smmu_cmdq_issue_cmd(master->smmu, &cmd);
> +		cmd->atc.sid = master->sids[i];
> +		arm_smmu_cmdq_issue_cmd(master->smmu, cmd);
>  	}
>  
>  	return arm_smmu_cmdq_issue_sync(master->smmu);
> @@ -2207,11 +1809,10 @@ static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
>  static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
>  				   int ssid, unsigned long iova, size_t size)
>  {
> -	int i;
> +	int ret = 0;
>  	unsigned long flags;
>  	struct arm_smmu_cmdq_ent cmd;
>  	struct arm_smmu_master *master;
> -	struct arm_smmu_cmdq_batch cmds = {};
>  
>  	if (!(smmu_domain->smmu->features & ARM_SMMU_FEAT_ATS))
>  		return 0;
> @@ -2236,18 +1837,11 @@ static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
>  	arm_smmu_atc_inv_to_cmd(ssid, iova, size, &cmd);
>  
>  	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
> -	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
> -		if (!master->ats_enabled)
> -			continue;
> -
> -		for (i = 0; i < master->num_sids; i++) {
> -			cmd.atc.sid = master->sids[i];
> -			arm_smmu_cmdq_batch_add(smmu_domain->smmu, &cmds, &cmd);
> -		}
> -	}
> +	list_for_each_entry(master, &smmu_domain->devices, domain_head)
> +		ret |= arm_smmu_atc_inv_master(master, &cmd);
>  	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
>  
> -	return arm_smmu_cmdq_batch_submit(smmu_domain->smmu, &cmds);
> +	return ret ? -ETIMEDOUT : 0;
>  }
>  
>  /* IO_PGTABLE API */
> @@ -2269,32 +1863,27 @@ static void arm_smmu_tlb_inv_context(void *cookie)
>  	/*
>  	 * NOTE: when io-pgtable is in non-strict mode, we may get here with
>  	 * PTEs previously cleared by unmaps on the current CPU not yet visible
> -	 * to the SMMU. We are relying on the dma_wmb() implicit during cmd
> -	 * insertion to guarantee those are observed before the TLBI. Do be
> -	 * careful, 007.
> +	 * to the SMMU. We are relying on the DSB implicit in
> +	 * queue_sync_prod_out() to guarantee those are observed before the
> +	 * TLBI. Do be careful, 007.
>  	 */
>  	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
>  	arm_smmu_cmdq_issue_sync(smmu);
>  	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
>  }
>  
> -static void arm_smmu_tlb_inv_range(unsigned long iova, size_t size,
> -				   size_t granule, bool leaf,
> -				   struct arm_smmu_domain *smmu_domain)
> +static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
> +					  size_t granule, bool leaf, void *cookie)
>  {
> +	struct arm_smmu_domain *smmu_domain = cookie;
>  	struct arm_smmu_device *smmu = smmu_domain->smmu;
> -	unsigned long start = iova, end = iova + size, num_pages = 0, tg = 0;
> -	size_t inv_range = granule;
> -	struct arm_smmu_cmdq_batch cmds = {};
>  	struct arm_smmu_cmdq_ent cmd = {
>  		.tlbi = {
>  			.leaf	= leaf,
> +			.addr	= iova,
>  		},
>  	};
>  
> -	if (!size)
> -		return;
> -
>  	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
>  		cmd.opcode	= CMDQ_OP_TLBI_NH_VA;
>  		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
> @@ -2303,78 +1892,37 @@ static void arm_smmu_tlb_inv_range(unsigned long iova, size_t size,
>  		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
>  	}
>  
> -	if (smmu->features & ARM_SMMU_FEAT_RANGE_INV) {
> -		/* Get the leaf page size */
> -		tg = __ffs(smmu_domain->domain.pgsize_bitmap);
> -
> -		/* Convert page size of 12,14,16 (log2) to 1,2,3 */
> -		cmd.tlbi.tg = (tg - 10) / 2;
> -
> -		/* Determine what level the granule is at */
> -		cmd.tlbi.ttl = 4 - ((ilog2(granule) - 3) / (tg - 3));
> -
> -		num_pages = size >> tg;
> -	}
> -
> -	while (iova < end) {
> -		if (smmu->features & ARM_SMMU_FEAT_RANGE_INV) {
> -			/*
> -			 * On each iteration of the loop, the range is 5 bits
> -			 * worth of the aligned size remaining.
> -			 * The range in pages is:
> -			 *
> -			 * range = (num_pages & (0x1f << __ffs(num_pages)))
> -			 */
> -			unsigned long scale, num;
> -
> -			/* Determine the power of 2 multiple number of pages */
> -			scale = __ffs(num_pages);
> -			cmd.tlbi.scale = scale;
> -
> -			/* Determine how many chunks of 2^scale size we have */
> -			num = (num_pages >> scale) & CMDQ_TLBI_RANGE_NUM_MAX;
> -			cmd.tlbi.num = num - 1;
> -
> -			/* range is num * 2^scale * pgsize */
> -			inv_range = num << (scale + tg);
> -
> -			/* Clear out the lower order bits for the next iteration */
> -			num_pages -= num << scale;
> -		}
> -
> -		cmd.tlbi.addr = iova;
> -		arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
> -		iova += inv_range;
> -	}
> -	arm_smmu_cmdq_batch_submit(smmu, &cmds);
> -
> -	/*
> -	 * Unfortunately, this can't be leaf-only since we may have
> -	 * zapped an entire table.
> -	 */
> -	arm_smmu_atc_inv_domain(smmu_domain, 0, start, size);
> +	do {
> +		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
> +		cmd.tlbi.addr += granule;
> +	} while (size -= granule);
>  }
>  
>  static void arm_smmu_tlb_inv_page_nosync(struct iommu_iotlb_gather *gather,
>  					 unsigned long iova, size_t granule,
>  					 void *cookie)
>  {
> -	struct arm_smmu_domain *smmu_domain = cookie;
> -	struct iommu_domain *domain = &smmu_domain->domain;
> -
> -	iommu_iotlb_gather_add_page(domain, gather, iova, granule);
> +	arm_smmu_tlb_inv_range_nosync(iova, granule, granule, true, cookie);
>  }
>  
>  static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,
>  				  size_t granule, void *cookie)
>  {
> -	arm_smmu_tlb_inv_range(iova, size, granule, false, cookie);
> +	struct arm_smmu_domain *smmu_domain = cookie;
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +
> +	arm_smmu_tlb_inv_range_nosync(iova, size, granule, false, cookie);
> +	arm_smmu_cmdq_issue_sync(smmu);
>  }
>  
>  static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,
>  				  size_t granule, void *cookie)
>  {
> -	arm_smmu_tlb_inv_range(iova, size, granule, true, cookie);
> +	struct arm_smmu_domain *smmu_domain = cookie;
> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
> +
> +	arm_smmu_tlb_inv_range_nosync(iova, size, granule, true, cookie);
> +	arm_smmu_cmdq_issue_sync(smmu);
>  }
>  
>  static const struct iommu_flush_ops arm_smmu_flush_ops = {
> @@ -2700,6 +2248,7 @@ static void arm_smmu_enable_ats(struct arm_smmu_master *master)
>  
>  static void arm_smmu_disable_ats(struct arm_smmu_master *master)
>  {
> +	struct arm_smmu_cmdq_ent cmd;
>  	struct arm_smmu_domain *smmu_domain = master->domain;
>  
>  	if (!master->ats_enabled)
> @@ -2711,8 +2260,9 @@ static void arm_smmu_disable_ats(struct arm_smmu_master *master)
>  	 * ATC invalidation via the SMMU.
>  	 */
>  	wmb();
> -	arm_smmu_atc_inv_master(master);
> -	atomic_dec(&smmu_domain->nr_ats_masters);
> +	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
> +	arm_smmu_atc_inv_master(master, &cmd);
> +    atomic_dec(&smmu_domain->nr_ats_masters);
>  }
>  
>  static int arm_smmu_enable_pasid(struct arm_smmu_master *master)
> @@ -2875,10 +2425,10 @@ static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain)
>  static void arm_smmu_iotlb_sync(struct iommu_domain *domain,
>  				struct iommu_iotlb_gather *gather)
>  {
> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> +	struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
>  
> -	arm_smmu_tlb_inv_range(gather->start, gather->end - gather->start,
> -			       gather->pgsize, true, smmu_domain);
> +	if (smmu)
> +		arm_smmu_cmdq_issue_sync(smmu);
>  }
>  
>  static phys_addr_t
> @@ -3176,49 +2726,18 @@ static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>  	return 0;
>  }
>  
> -static void arm_smmu_cmdq_free_bitmap(void *data)
> -{
> -	unsigned long *bitmap = data;
> -	bitmap_free(bitmap);
> -}
> -
> -static int arm_smmu_cmdq_init(struct arm_smmu_device *smmu)
> -{
> -	int ret = 0;
> -	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
> -	unsigned int nents = 1 << cmdq->q.llq.max_n_shift;
> -	atomic_long_t *bitmap;
> -
> -	atomic_set(&cmdq->owner_prod, 0);
> -	atomic_set(&cmdq->lock, 0);
> -
> -	bitmap = (atomic_long_t *)bitmap_zalloc(nents, GFP_KERNEL);
> -	if (!bitmap) {
> -		dev_err(smmu->dev, "failed to allocate cmdq bitmap\n");
> -		ret = -ENOMEM;
> -	} else {
> -		cmdq->valid_map = bitmap;
> -		devm_add_action(smmu->dev, arm_smmu_cmdq_free_bitmap, bitmap);
> -	}
> -
> -	return ret;
> -}
> -
>  static int arm_smmu_init_queues(struct arm_smmu_device *smmu)
>  {
>  	int ret;
>  
>  	/* cmdq */
> +	spin_lock_init(&smmu->cmdq.lock);
>  	ret = arm_smmu_init_one_queue(smmu, &smmu->cmdq.q, ARM_SMMU_CMDQ_PROD,
>  				      ARM_SMMU_CMDQ_CONS, CMDQ_ENT_DWORDS,
>  				      "cmdq");
>  	if (ret)
>  		return ret;
>  
> -	ret = arm_smmu_cmdq_init(smmu);
> -	if (ret)
> -		return ret;
> -
>  	/* evtq */
>  	ret = arm_smmu_init_one_queue(smmu, &smmu->evtq.q, ARM_SMMU_EVTQ_PROD,
>  				      ARM_SMMU_EVTQ_CONS, EVTQ_ENT_DWORDS,
> @@ -3799,15 +3318,9 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>  	/* Queue sizes, capped to ensure natural alignment */
>  	smmu->cmdq.q.llq.max_n_shift = min_t(u32, CMDQ_MAX_SZ_SHIFT,
>  					     FIELD_GET(IDR1_CMDQS, reg));
> -	if (smmu->cmdq.q.llq.max_n_shift <= ilog2(CMDQ_BATCH_ENTRIES)) {
> -		/*
> -		 * We don't support splitting up batches, so one batch of
> -		 * commands plus an extra sync needs to fit inside the command
> -		 * queue. There's also no way we can handle the weird alignment
> -		 * restrictions on the base pointer for a unit-length queue.
> -		 */
> -		dev_err(smmu->dev, "command queue size <= %d entries not supported\n",
> -			CMDQ_BATCH_ENTRIES);
> +	if (!smmu->cmdq.q.llq.max_n_shift) {
> +		/* Odd alignment restrictions on the base, so ignore for now */
> +		dev_err(smmu->dev, "unit-length command queue not supported\n");
>  		return -ENXIO;
>  	}
>  
> -- 
> 2.17.1
> 


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 3/8] xen/arm: revert patch related to XArray
  2020-11-26 17:02 ` [PATCH v2 3/8] xen/arm: revert patch related to XArray Rahul Singh
@ 2020-12-02  0:20   ` Stefano Stabellini
  2020-12-02 13:46   ` Julien Grall
  1 sibling, 0 replies; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-02  0:20 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, bertrand.marquis, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

On Thu, 26 Nov 2020, Rahul Singh wrote:
> XArray is not implemented in XEN revert the patch that introduce the
> XArray code in SMMUv3 driver.
> 
> Once XArray is implemented in XEN this patch can be added in XEN.
> 
> Reverted the commit 0299a1a81ca056e79c1a7fb751f936ec0d5c7afe
> 
> Signed-off-by: Rahul Singh <rahul.singh@arm.com>

Acked-by: Stefano Stabellini <sstabellini@kernel.org>


> ---
>  xen/drivers/passthrough/arm/smmu-v3.c | 27 +++++++++------------------
>  1 file changed, 9 insertions(+), 18 deletions(-)
> 
> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
> index 97eac61ea4..cec304e51a 100644
> --- a/xen/drivers/passthrough/arm/smmu-v3.c
> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
> @@ -638,6 +638,7 @@ struct arm_smmu_device {
>  
>  #define ARM_SMMU_MAX_ASIDS		(1 << 16)
>  	unsigned int			asid_bits;
> +	DECLARE_BITMAP(asid_map, ARM_SMMU_MAX_ASIDS);
>  
>  #define ARM_SMMU_MAX_VMIDS		(1 << 16)
>  	unsigned int			vmid_bits;
> @@ -703,8 +704,6 @@ struct arm_smmu_option_prop {
>  	const char *prop;
>  };
>  
> -static DEFINE_XARRAY_ALLOC1(asid_xa);
> -
>  static struct arm_smmu_option_prop arm_smmu_options[] = {
>  	{ ARM_SMMU_OPT_SKIP_PREFETCH, "hisilicon,broken-prefetch-cmd" },
>  	{ ARM_SMMU_OPT_PAGE0_REGS_ONLY, "cavium,cn9900-broken-page1-regspace"},
> @@ -1366,14 +1365,6 @@ static void arm_smmu_free_cd_tables(struct arm_smmu_domain *smmu_domain)
>  	cdcfg->cdtab = NULL;
>  }
>  
> -static void arm_smmu_free_asid(struct arm_smmu_ctx_desc *cd)
> -{
> -	if (!cd->asid)
> -		return;
> -
> -	xa_erase(&asid_xa, cd->asid);
> -}
> -
>  /* Stream table manipulation functions */
>  static void
>  arm_smmu_write_strtab_l1_desc(__le64 *dst, struct arm_smmu_strtab_l1_desc *desc)
> @@ -2006,9 +1997,10 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>  	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
>  		struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
>  
> -		if (cfg->cdcfg.cdtab)
> +		if (cfg->cdcfg.cdtab) {
>  			arm_smmu_free_cd_tables(smmu_domain);
> -		arm_smmu_free_asid(&cfg->cd);
> +			arm_smmu_bitmap_free(smmu->asid_map, cfg->cd.asid);
> +		}
>  	} else {
>  		struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>  		if (cfg->vmid)
> @@ -2023,15 +2015,14 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
>  				       struct io_pgtable_cfg *pgtbl_cfg)
>  {
>  	int ret;
> -	u32 asid;
> +	int asid;
>  	struct arm_smmu_device *smmu = smmu_domain->smmu;
>  	struct arm_smmu_s1_cfg *cfg = &smmu_domain->s1_cfg;
>  	typeof(&pgtbl_cfg->arm_lpae_s1_cfg.tcr) tcr = &pgtbl_cfg->arm_lpae_s1_cfg.tcr;
>  
> -	ret = xa_alloc(&asid_xa, &asid, &cfg->cd,
> -		       XA_LIMIT(1, (1 << smmu->asid_bits) - 1), GFP_KERNEL);
> -	if (ret)
> -		return ret;
> +	asid = arm_smmu_bitmap_alloc(smmu->asid_map, smmu->asid_bits);
> +	if (asid < 0)
> +		return asid;
>  
>  	cfg->s1cdmax = master->ssid_bits;
>  
> @@ -2064,7 +2055,7 @@ static int arm_smmu_domain_finalise_s1(struct arm_smmu_domain *smmu_domain,
>  out_free_cd_tables:
>  	arm_smmu_free_cd_tables(smmu_domain);
>  out_free_asid:
> -	arm_smmu_free_asid(&cfg->cd);
> +	arm_smmu_bitmap_free(smmu->asid_map, asid);
>  	return ret;
>  }
>  
> -- 
> 2.17.1
> 


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 4/8] xen/arm: Remove support for MSI on SMMUv3
  2020-11-26 17:02 ` [PATCH v2 4/8] xen/arm: Remove support for MSI on SMMUv3 Rahul Singh
@ 2020-12-02  0:33   ` Stefano Stabellini
  2020-12-02  0:40     ` Stefano Stabellini
  0 siblings, 1 reply; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-02  0:33 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, bertrand.marquis, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

On Thu, 26 Nov 2020, Rahul Singh wrote:
> XEN does not support MSI on ARM platforms, therefore remove the MSI
> support from SMMUv3 driver.
> 
> Signed-off-by: Rahul Singh <rahul.singh@arm.com>

I wonder if it makes sense to #ifdef CONFIG_MSI this code instead of
removing it completely.


In the past, we tried to keep the entire file as textually similar to
the original Linux driver as possible to make it easier to backport
features and fixes. So, in this case we would probably not even used an
#ifdef but maybe something like:

  if (/* msi_enabled */ 0)
      return;

at the beginning of arm_smmu_setup_msis.


However, that strategy didn't actually work very well because backports
have proven difficult to do anyway. So at that point we might as well at
least have clean code in Xen and do the changes properly.

So that's my reasoning for accepting this patch :-)

Julien, are you happy with this too?


> ---
>  xen/drivers/passthrough/arm/smmu-v3.c | 176 +-------------------------
>  1 file changed, 3 insertions(+), 173 deletions(-)
> 
> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
> index cec304e51a..401f7ae006 100644
> --- a/xen/drivers/passthrough/arm/smmu-v3.c
> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
> @@ -416,31 +416,6 @@ enum pri_resp {
>  	PRI_RESP_SUCC = 2,
>  };
>  
> -enum arm_smmu_msi_index {
> -	EVTQ_MSI_INDEX,
> -	GERROR_MSI_INDEX,
> -	PRIQ_MSI_INDEX,
> -	ARM_SMMU_MAX_MSIS,
> -};
> -
> -static phys_addr_t arm_smmu_msi_cfg[ARM_SMMU_MAX_MSIS][3] = {
> -	[EVTQ_MSI_INDEX] = {
> -		ARM_SMMU_EVTQ_IRQ_CFG0,
> -		ARM_SMMU_EVTQ_IRQ_CFG1,
> -		ARM_SMMU_EVTQ_IRQ_CFG2,
> -	},
> -	[GERROR_MSI_INDEX] = {
> -		ARM_SMMU_GERROR_IRQ_CFG0,
> -		ARM_SMMU_GERROR_IRQ_CFG1,
> -		ARM_SMMU_GERROR_IRQ_CFG2,
> -	},
> -	[PRIQ_MSI_INDEX] = {
> -		ARM_SMMU_PRIQ_IRQ_CFG0,
> -		ARM_SMMU_PRIQ_IRQ_CFG1,
> -		ARM_SMMU_PRIQ_IRQ_CFG2,
> -	},
> -};
> -
>  struct arm_smmu_cmdq_ent {
>  	/* Common fields */
>  	u8				opcode;
> @@ -504,10 +479,6 @@ struct arm_smmu_cmdq_ent {
>  		} pri;
>  
>  		#define CMDQ_OP_CMD_SYNC	0x46
> -		struct {
> -			u32			msidata;
> -			u64			msiaddr;
> -		} sync;
>  	};
>  };
>  
> @@ -649,12 +620,6 @@ struct arm_smmu_device {
>  
>  	struct arm_smmu_strtab_cfg	strtab_cfg;
>  
> -	/* Hi16xx adds an extra 32 bits of goodness to its MSI payload */
> -	union {
> -		u32			sync_count;
> -		u64			padding;
> -	};
> -
>  	/* IOMMU core code handle */
>  	struct iommu_device		iommu;
>  };
> @@ -945,20 +910,7 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
>  		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_RESP, ent->pri.resp);
>  		break;
>  	case CMDQ_OP_CMD_SYNC:
> -		if (ent->sync.msiaddr)
> -			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_IRQ);
> -		else
> -			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
> -		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSH, ARM_SMMU_SH_ISH);
> -		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIATTR, ARM_SMMU_MEMATTR_OIWB);
> -		/*
> -		 * Commands are written little-endian, but we want the SMMU to
> -		 * receive MSIData, and thus write it back to memory, in CPU
> -		 * byte order, so big-endian needs an extra byteswap here.
> -		 */
> -		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIDATA,
> -				     cpu_to_le32(ent->sync.msidata));
> -		cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
> +		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
>  		break;
>  	default:
>  		return -ENOENT;
> @@ -1054,50 +1006,6 @@ static void arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
>  	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
>  }
>  
> -/*
> - * The difference between val and sync_idx is bounded by the maximum size of
> - * a queue at 2^20 entries, so 32 bits is plenty for wrap-safe arithmetic.
> - */
> -static int __arm_smmu_sync_poll_msi(struct arm_smmu_device *smmu, u32 sync_idx)
> -{
> -	ktime_t timeout;
> -	u32 val;
> -
> -	timeout = ktime_add_us(ktime_get(), ARM_SMMU_CMDQ_SYNC_TIMEOUT_US);
> -	val = smp_cond_load_acquire(&smmu->sync_count,
> -				    (int)(VAL - sync_idx) >= 0 ||
> -				    !ktime_before(ktime_get(), timeout));
> -
> -	return (int)(val - sync_idx) < 0 ? -ETIMEDOUT : 0;
> -}
> -
> -static int __arm_smmu_cmdq_issue_sync_msi(struct arm_smmu_device *smmu)
> -{
> -	u64 cmd[CMDQ_ENT_DWORDS];
> -	unsigned long flags;
> -	struct arm_smmu_cmdq_ent  ent = {
> -		.opcode = CMDQ_OP_CMD_SYNC,
> -		.sync	= {
> -			.msiaddr = virt_to_phys(&smmu->sync_count),
> -		},
> -	};
> -
> -	spin_lock_irqsave(&smmu->cmdq.lock, flags);
> -
> -	/* Piggy-back on the previous command if it's a SYNC */
> -	if (smmu->prev_cmd_opcode == CMDQ_OP_CMD_SYNC) {
> -		ent.sync.msidata = smmu->sync_nr;
> -	} else {
> -		ent.sync.msidata = ++smmu->sync_nr;
> -		arm_smmu_cmdq_build_cmd(cmd, &ent);
> -		arm_smmu_cmdq_insert_cmd(smmu, cmd);
> -	}
> -
> -	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
> -
> -	return __arm_smmu_sync_poll_msi(smmu, ent.sync.msidata);
> -}
> -
>  static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>  {
>  	u64 cmd[CMDQ_ENT_DWORDS];
> @@ -1119,12 +1027,9 @@ static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>  static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>  {
>  	int ret;
> -	bool msi = (smmu->features & ARM_SMMU_FEAT_MSI) &&
> -		   (smmu->features & ARM_SMMU_FEAT_COHERENCY);
>  
> -	ret = msi ? __arm_smmu_cmdq_issue_sync_msi(smmu)
> -		  : __arm_smmu_cmdq_issue_sync(smmu);
> -	if (ret)
> +	ret = __arm_smmu_cmdq_issue_sync(smmu);
> +	if ( ret )
>  		dev_err_ratelimited(smmu->dev, "CMD_SYNC timeout\n");
>  	return ret;
>  }
> @@ -2898,83 +2803,10 @@ static int arm_smmu_update_gbpa(struct arm_smmu_device *smmu, u32 set, u32 clr)
>  	return ret;
>  }
>  
> -static void arm_smmu_free_msis(void *data)
> -{
> -	struct device *dev = data;
> -	platform_msi_domain_free_irqs(dev);
> -}
> -
> -static void arm_smmu_write_msi_msg(struct msi_desc *desc, struct msi_msg *msg)
> -{
> -	phys_addr_t doorbell;
> -	struct device *dev = msi_desc_to_dev(desc);
> -	struct arm_smmu_device *smmu = dev_get_drvdata(dev);
> -	phys_addr_t *cfg = arm_smmu_msi_cfg[desc->platform.msi_index];
> -
> -	doorbell = (((u64)msg->address_hi) << 32) | msg->address_lo;
> -	doorbell &= MSI_CFG0_ADDR_MASK;
> -
> -	writeq_relaxed(doorbell, smmu->base + cfg[0]);
> -	writel_relaxed(msg->data, smmu->base + cfg[1]);
> -	writel_relaxed(ARM_SMMU_MEMATTR_DEVICE_nGnRE, smmu->base + cfg[2]);
> -}
> -
> -static void arm_smmu_setup_msis(struct arm_smmu_device *smmu)
> -{
> -	struct msi_desc *desc;
> -	int ret, nvec = ARM_SMMU_MAX_MSIS;
> -	struct device *dev = smmu->dev;
> -
> -	/* Clear the MSI address regs */
> -	writeq_relaxed(0, smmu->base + ARM_SMMU_GERROR_IRQ_CFG0);
> -	writeq_relaxed(0, smmu->base + ARM_SMMU_EVTQ_IRQ_CFG0);
> -
> -	if (smmu->features & ARM_SMMU_FEAT_PRI)
> -		writeq_relaxed(0, smmu->base + ARM_SMMU_PRIQ_IRQ_CFG0);
> -	else
> -		nvec--;
> -
> -	if (!(smmu->features & ARM_SMMU_FEAT_MSI))
> -		return;
> -
> -	if (!dev->msi_domain) {
> -		dev_info(smmu->dev, "msi_domain absent - falling back to wired irqs\n");
> -		return;
> -	}
> -
> -	/* Allocate MSIs for evtq, gerror and priq. Ignore cmdq */
> -	ret = platform_msi_domain_alloc_irqs(dev, nvec, arm_smmu_write_msi_msg);
> -	if (ret) {
> -		dev_warn(dev, "failed to allocate MSIs - falling back to wired irqs\n");
> -		return;
> -	}
> -
> -	for_each_msi_entry(desc, dev) {
> -		switch (desc->platform.msi_index) {
> -		case EVTQ_MSI_INDEX:
> -			smmu->evtq.q.irq = desc->irq;
> -			break;
> -		case GERROR_MSI_INDEX:
> -			smmu->gerr_irq = desc->irq;
> -			break;
> -		case PRIQ_MSI_INDEX:
> -			smmu->priq.q.irq = desc->irq;
> -			break;
> -		default:	/* Unknown */
> -			continue;
> -		}
> -	}
> -
> -	/* Add callback to free MSIs on teardown */
> -	devm_add_action(dev, arm_smmu_free_msis, dev);
> -}
> -
>  static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>  {
>  	int irq, ret;
>  
> -	arm_smmu_setup_msis(smmu);
> -
>  	/* Request interrupt lines */
>  	irq = smmu->evtq.q.irq;
>  	if (irq) {
> @@ -3250,8 +3082,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>  	if (reg & IDR0_SEV)
>  		smmu->features |= ARM_SMMU_FEAT_SEV;
>  
> -	if (reg & IDR0_MSI)
> -		smmu->features |= ARM_SMMU_FEAT_MSI;
>  
>  	if (reg & IDR0_HYP)
>  		smmu->features |= ARM_SMMU_FEAT_HYP;
> -- 
> 2.17.1
> 


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 5/8] xen/arm: Remove support for PCI ATS on SMMUv3
  2020-11-26 17:02 ` [PATCH v2 5/8] xen/arm: Remove support for PCI ATS " Rahul Singh
@ 2020-12-02  0:39   ` Stefano Stabellini
  2020-12-02 13:07     ` Rahul Singh
  2020-12-02 13:57   ` Julien Grall
  1 sibling, 1 reply; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-02  0:39 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, bertrand.marquis, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

On Thu, 26 Nov 2020, Rahul Singh wrote:
> PCI ATS functionality is not implemented and tested on ARM. Remove the
> PCI ATS support, once PCI ATS support is tested and available this
> patch can be added.
> 
> Signed-off-by: Rahul Singh <rahul.singh@arm.com>

This looks like a revert of 9ce27afc0830f. Can we add that as a note to
the commit message?

One very minor comment at the bottom


> ---
>  xen/drivers/passthrough/arm/smmu-v3.c | 273 --------------------------
>  1 file changed, 273 deletions(-)
> 
> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
> index 401f7ae006..6a33628087 100644
> --- a/xen/drivers/passthrough/arm/smmu-v3.c
> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
> @@ -460,16 +460,6 @@ struct arm_smmu_cmdq_ent {
>  			u64			addr;
>  		} tlbi;
>  
> -		#define CMDQ_OP_ATC_INV		0x40
> -		#define ATC_INV_SIZE_ALL	52
> -		struct {
> -			u32			sid;
> -			u32			ssid;
> -			u64			addr;
> -			u8			size;
> -			bool			global;
> -		} atc;
> -
>  		#define CMDQ_OP_PRI_RESP	0x41
>  		struct {
>  			u32			sid;
> @@ -632,7 +622,6 @@ struct arm_smmu_master {
>  	struct list_head		domain_head;
>  	u32				*sids;
>  	unsigned int			num_sids;
> -	bool				ats_enabled;
>  	unsigned int			ssid_bits;
>  };
>  
> @@ -650,7 +639,6 @@ struct arm_smmu_domain {
>  
>  	struct io_pgtable_ops		*pgtbl_ops;
>  	bool				non_strict;
> -	atomic_t			nr_ats_masters;
>  
>  	enum arm_smmu_domain_stage	stage;
>  	union {
> @@ -886,14 +874,6 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
>  	case CMDQ_OP_TLBI_S12_VMALL:
>  		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
>  		break;
> -	case CMDQ_OP_ATC_INV:
> -		cmd[0] |= FIELD_PREP(CMDQ_0_SSV, ent->substream_valid);
> -		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_GLOBAL, ent->atc.global);
> -		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_SSID, ent->atc.ssid);
> -		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_SID, ent->atc.sid);
> -		cmd[1] |= FIELD_PREP(CMDQ_ATC_1_SIZE, ent->atc.size);
> -		cmd[1] |= ent->atc.addr & CMDQ_ATC_1_ADDR_MASK;
> -		break;
>  	case CMDQ_OP_PRI_RESP:
>  		cmd[0] |= FIELD_PREP(CMDQ_0_SSV, ent->substream_valid);
>  		cmd[0] |= FIELD_PREP(CMDQ_PRI_0_SSID, ent->pri.ssid);
> @@ -925,7 +905,6 @@ static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
>  		[CMDQ_ERR_CERROR_NONE_IDX]	= "No error",
>  		[CMDQ_ERR_CERROR_ILL_IDX]	= "Illegal command",
>  		[CMDQ_ERR_CERROR_ABT_IDX]	= "Abort on command fetch",
> -		[CMDQ_ERR_CERROR_ATC_INV_IDX]	= "ATC invalidate timeout",
>  	};
>  
>  	int i;
> @@ -945,14 +924,6 @@ static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
>  		dev_err(smmu->dev, "retrying command fetch\n");
>  	case CMDQ_ERR_CERROR_NONE_IDX:
>  		return;
> -	case CMDQ_ERR_CERROR_ATC_INV_IDX:
> -		/*
> -		 * ATC Invalidation Completion timeout. CONS is still pointing
> -		 * at the CMD_SYNC. Attempt to complete other pending commands
> -		 * by repeating the CMD_SYNC, though we might well end up back
> -		 * here since the ATC invalidation may still be pending.
> -		 */
> -		return;
>  	case CMDQ_ERR_CERROR_ILL_IDX:
>  	default:
>  		break;
> @@ -1422,9 +1393,6 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
>  		val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_S2_TRANS);
>  	}
>  
> -	if (master->ats_enabled)
> -		dst[1] |= cpu_to_le64(FIELD_PREP(STRTAB_STE_1_EATS,
> -						 STRTAB_STE_1_EATS_TRANS));
>  
>  	arm_smmu_sync_ste_for_sid(smmu, sid);
>  	/* See comment in arm_smmu_write_ctx_desc() */
> @@ -1633,112 +1601,6 @@ static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
>  	return IRQ_WAKE_THREAD;
>  }
>  
> -static void
> -arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size,
> -			struct arm_smmu_cmdq_ent *cmd)
> -{
> -	size_t log2_span;
> -	size_t span_mask;
> -	/* ATC invalidates are always on 4096-bytes pages */
> -	size_t inval_grain_shift = 12;
> -	unsigned long page_start, page_end;
> -
> -	*cmd = (struct arm_smmu_cmdq_ent) {
> -		.opcode			= CMDQ_OP_ATC_INV,
> -		.substream_valid	= !!ssid,
> -		.atc.ssid		= ssid,
> -	};
> -
> -	if (!size) {
> -		cmd->atc.size = ATC_INV_SIZE_ALL;
> -		return;
> -	}
> -
> -	page_start	= iova >> inval_grain_shift;
> -	page_end	= (iova + size - 1) >> inval_grain_shift;
> -
> -	/*
> -	 * In an ATS Invalidate Request, the address must be aligned on the
> -	 * range size, which must be a power of two number of page sizes. We
> -	 * thus have to choose between grossly over-invalidating the region, or
> -	 * splitting the invalidation into multiple commands. For simplicity
> -	 * we'll go with the first solution, but should refine it in the future
> -	 * if multiple commands are shown to be more efficient.
> -	 *
> -	 * Find the smallest power of two that covers the range. The most
> -	 * significant differing bit between the start and end addresses,
> -	 * fls(start ^ end), indicates the required span. For example:
> -	 *
> -	 * We want to invalidate pages [8; 11]. This is already the ideal range:
> -	 *		x = 0b1000 ^ 0b1011 = 0b11
> -	 *		span = 1 << fls(x) = 4
> -	 *
> -	 * To invalidate pages [7; 10], we need to invalidate [0; 15]:
> -	 *		x = 0b0111 ^ 0b1010 = 0b1101
> -	 *		span = 1 << fls(x) = 16
> -	 */
> -	log2_span	= fls_long(page_start ^ page_end);
> -	span_mask	= (1ULL << log2_span) - 1;
> -
> -	page_start	&= ~span_mask;
> -
> -	cmd->atc.addr	= page_start << inval_grain_shift;
> -	cmd->atc.size	= log2_span;
> -}
> -
> -static int arm_smmu_atc_inv_master(struct arm_smmu_master *master,
> -				   struct arm_smmu_cmdq_ent *cmd)
> -{
> -	int i;
> -
> -	if (!master->ats_enabled)
> -		return 0;
> -
> -	for (i = 0; i < master->num_sids; i++) {
> -		cmd->atc.sid = master->sids[i];
> -		arm_smmu_cmdq_issue_cmd(master->smmu, cmd);
> -	}
> -
> -	return arm_smmu_cmdq_issue_sync(master->smmu);
> -}
> -
> -static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
> -				   int ssid, unsigned long iova, size_t size)
> -{
> -	int ret = 0;
> -	unsigned long flags;
> -	struct arm_smmu_cmdq_ent cmd;
> -	struct arm_smmu_master *master;
> -
> -	if (!(smmu_domain->smmu->features & ARM_SMMU_FEAT_ATS))
> -		return 0;
> -
> -	/*
> -	 * Ensure that we've completed prior invalidation of the main TLBs
> -	 * before we read 'nr_ats_masters' in case of a concurrent call to
> -	 * arm_smmu_enable_ats():
> -	 *
> -	 *	// unmap()			// arm_smmu_enable_ats()
> -	 *	TLBI+SYNC			atomic_inc(&nr_ats_masters);
> -	 *	smp_mb();			[...]
> -	 *	atomic_read(&nr_ats_masters);	pci_enable_ats() // writel()
> -	 *
> -	 * Ensures that we always see the incremented 'nr_ats_masters' count if
> -	 * ATS was enabled at the PCI device before completion of the TLBI.
> -	 */
> -	smp_mb();
> -	if (!atomic_read(&smmu_domain->nr_ats_masters))
> -		return 0;
> -
> -	arm_smmu_atc_inv_to_cmd(ssid, iova, size, &cmd);
> -
> -	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
> -	list_for_each_entry(master, &smmu_domain->devices, domain_head)
> -		ret |= arm_smmu_atc_inv_master(master, &cmd);
> -	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
> -
> -	return ret ? -ETIMEDOUT : 0;
> -}
>  
>  /* IO_PGTABLE API */
>  static void arm_smmu_tlb_inv_context(void *cookie)
> @@ -1765,7 +1627,6 @@ static void arm_smmu_tlb_inv_context(void *cookie)
>  	 */
>  	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
>  	arm_smmu_cmdq_issue_sync(smmu);
> -	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
>  }
>  
>  static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
> @@ -2106,108 +1967,6 @@ static void arm_smmu_install_ste_for_dev(struct arm_smmu_master *master)
>  	}
>  }
>  
> -static bool arm_smmu_ats_supported(struct arm_smmu_master *master)
> -{
> -	struct device *dev = master->dev;
> -	struct arm_smmu_device *smmu = master->smmu;
> -	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> -
> -	if (!(smmu->features & ARM_SMMU_FEAT_ATS))
> -		return false;
> -
> -	if (!(fwspec->flags & IOMMU_FWSPEC_PCI_RC_ATS))
> -		return false;
> -
> -	return dev_is_pci(dev) && pci_ats_supported(to_pci_dev(dev));
> -}
> -
> -static void arm_smmu_enable_ats(struct arm_smmu_master *master)
> -{
> -	size_t stu;
> -	struct pci_dev *pdev;
> -	struct arm_smmu_device *smmu = master->smmu;
> -	struct arm_smmu_domain *smmu_domain = master->domain;
> -
> -	/* Don't enable ATS at the endpoint if it's not enabled in the STE */
> -	if (!master->ats_enabled)
> -		return;
> -
> -	/* Smallest Translation Unit: log2 of the smallest supported granule */
> -	stu = __ffs(smmu->pgsize_bitmap);
> -	pdev = to_pci_dev(master->dev);
> -
> -	atomic_inc(&smmu_domain->nr_ats_masters);
> -	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
> -	if (pci_enable_ats(pdev, stu))
> -		dev_err(master->dev, "Failed to enable ATS (STU %zu)\n", stu);
> -}
> -
> -static void arm_smmu_disable_ats(struct arm_smmu_master *master)
> -{
> -	struct arm_smmu_cmdq_ent cmd;
> -	struct arm_smmu_domain *smmu_domain = master->domain;
> -
> -	if (!master->ats_enabled)
> -		return;
> -
> -	pci_disable_ats(to_pci_dev(master->dev));
> -	/*
> -	 * Ensure ATS is disabled at the endpoint before we issue the
> -	 * ATC invalidation via the SMMU.
> -	 */
> -	wmb();
> -	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
> -	arm_smmu_atc_inv_master(master, &cmd);
> -    atomic_dec(&smmu_domain->nr_ats_masters);
> -}
> -
> -static int arm_smmu_enable_pasid(struct arm_smmu_master *master)
> -{
> -	int ret;
> -	int features;
> -	int num_pasids;
> -	struct pci_dev *pdev;
> -
> -	if (!dev_is_pci(master->dev))
> -		return -ENODEV;
> -
> -	pdev = to_pci_dev(master->dev);
> -
> -	features = pci_pasid_features(pdev);
> -	if (features < 0)
> -		return features;
> -
> -	num_pasids = pci_max_pasids(pdev);
> -	if (num_pasids <= 0)
> -		return num_pasids;
> -
> -	ret = pci_enable_pasid(pdev, features);
> -	if (ret) {
> -		dev_err(&pdev->dev, "Failed to enable PASID\n");
> -		return ret;
> -	}
> -
> -	master->ssid_bits = min_t(u8, ilog2(num_pasids),
> -				  master->smmu->ssid_bits);
> -	return 0;
> -}
> -
> -static void arm_smmu_disable_pasid(struct arm_smmu_master *master)
> -{
> -	struct pci_dev *pdev;
> -
> -	if (!dev_is_pci(master->dev))
> -		return;
> -
> -	pdev = to_pci_dev(master->dev);
> -
> -	if (!pdev->pasid_enabled)
> -		return;
> -
> -	master->ssid_bits = 0;
> -	pci_disable_pasid(pdev);
> -}
> -
>  static void arm_smmu_detach_dev(struct arm_smmu_master *master)
>  {
>  	unsigned long flags;
> @@ -2216,14 +1975,11 @@ static void arm_smmu_detach_dev(struct arm_smmu_master *master)
>  	if (!smmu_domain)
>  		return;
>  
> -	arm_smmu_disable_ats(master);
> -
>  	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
>  	list_del(&master->domain_head);
>  	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
>  
>  	master->domain = NULL;
> -	master->ats_enabled = false;
>  	arm_smmu_install_ste_for_dev(master);
>  }
>  
> @@ -2271,17 +2027,12 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
>  
>  	master->domain = smmu_domain;
>  
> -	if (smmu_domain->stage != ARM_SMMU_DOMAIN_BYPASS)
> -		master->ats_enabled = arm_smmu_ats_supported(master);
> -
>  	arm_smmu_install_ste_for_dev(master);
>  
>  	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
>  	list_add(&master->domain_head, &smmu_domain->devices);
>  	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
>  
> -	arm_smmu_enable_ats(master);
> -
>  out_unlock:
>  	mutex_unlock(&smmu_domain->init_mutex);
>  	return ret;
> @@ -2410,16 +2161,6 @@ static struct iommu_device *arm_smmu_probe_device(struct device *dev)
>  
>  	master->ssid_bits = min(smmu->ssid_bits, fwspec->num_pasid_bits);
>  
> -	/*
> -	 * Note that PASID must be enabled before, and disabled after ATS:
> -	 * PCI Express Base 4.0r1.0 - 10.5.1.3 ATS Control Register
> -	 *
> -	 *   Behavior is undefined if this bit is Set and the value of the PASID
> -	 *   Enable, Execute Requested Enable, or Privileged Mode Requested bits
> -	 *   are changed.
> -	 */
> -	arm_smmu_enable_pasid(master);
> -
>  	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB))
>  		master->ssid_bits = min_t(u8, master->ssid_bits,
>  					  CTXDESC_LINEAR_CDMAX);
> @@ -2442,7 +2183,6 @@ static void arm_smmu_release_device(struct device *dev)
>  
>  	master = dev_iommu_priv_get(dev);
>  	arm_smmu_detach_dev(master);
> -	arm_smmu_disable_pasid(master);
>  	kfree(master);
>  	iommu_fwspec_free(dev);
>  }
> @@ -2997,15 +2737,6 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>  		}
>  	}
>  
> -	if (smmu->features & ARM_SMMU_FEAT_ATS) {
> -		enables |= CR0_ATSCHK;
> -		ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
> -					      ARM_SMMU_CR0ACK);
> -		if (ret) {
> -			dev_err(smmu->dev, "failed to enable ATS check\n");
> -			return ret;
> -		}
> -	}
>  
>  	ret = arm_smmu_setup_irqs(smmu);
>  	if (ret) {
> @@ -3076,13 +2807,9 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>  	if (IS_ENABLED(CONFIG_PCI_PRI) && reg & IDR0_PRI)
>  		smmu->features |= ARM_SMMU_FEAT_PRI;
>  
> -	if (IS_ENABLED(CONFIG_PCI_ATS) && reg & IDR0_ATS)
> -		smmu->features |= ARM_SMMU_FEAT_ATS;
> -
>  	if (reg & IDR0_SEV)
>  		smmu->features |= ARM_SMMU_FEAT_SEV;
>  
> -
>  	if (reg & IDR0_HYP)

spurious change


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 4/8] xen/arm: Remove support for MSI on SMMUv3
  2020-12-02  0:33   ` Stefano Stabellini
@ 2020-12-02  0:40     ` Stefano Stabellini
  2020-12-02 13:12       ` Rahul Singh
  0 siblings, 1 reply; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-02  0:40 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: Rahul Singh, xen-devel, bertrand.marquis, Julien Grall,
	Volodymyr Babchuk

On Tue, 1 Dec 2020, Stefano Stabellini wrote:
> On Thu, 26 Nov 2020, Rahul Singh wrote:
> > XEN does not support MSI on ARM platforms, therefore remove the MSI
> > support from SMMUv3 driver.
> > 
> > Signed-off-by: Rahul Singh <rahul.singh@arm.com>
> 
> I wonder if it makes sense to #ifdef CONFIG_MSI this code instead of
> removing it completely.

One more thought: could this patch be achieved by reverting
166bdbd23161160f2abcea70621adba179050bee ? If this patch could be done
by a couple of revert, it would be great to say it in the commit
message.


> In the past, we tried to keep the entire file as textually similar to
> the original Linux driver as possible to make it easier to backport
> features and fixes. So, in this case we would probably not even used an
> #ifdef but maybe something like:
> 
>   if (/* msi_enabled */ 0)
>       return;
> 
> at the beginning of arm_smmu_setup_msis.
> 
> 
> However, that strategy didn't actually work very well because backports
> have proven difficult to do anyway. So at that point we might as well at
> least have clean code in Xen and do the changes properly.
> 
> So that's my reasoning for accepting this patch :-)
> 
> Julien, are you happy with this too?
> 
> 
> > ---
> >  xen/drivers/passthrough/arm/smmu-v3.c | 176 +-------------------------
> >  1 file changed, 3 insertions(+), 173 deletions(-)
> > 
> > diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
> > index cec304e51a..401f7ae006 100644
> > --- a/xen/drivers/passthrough/arm/smmu-v3.c
> > +++ b/xen/drivers/passthrough/arm/smmu-v3.c
> > @@ -416,31 +416,6 @@ enum pri_resp {
> >  	PRI_RESP_SUCC = 2,
> >  };
> >  
> > -enum arm_smmu_msi_index {
> > -	EVTQ_MSI_INDEX,
> > -	GERROR_MSI_INDEX,
> > -	PRIQ_MSI_INDEX,
> > -	ARM_SMMU_MAX_MSIS,
> > -};
> > -
> > -static phys_addr_t arm_smmu_msi_cfg[ARM_SMMU_MAX_MSIS][3] = {
> > -	[EVTQ_MSI_INDEX] = {
> > -		ARM_SMMU_EVTQ_IRQ_CFG0,
> > -		ARM_SMMU_EVTQ_IRQ_CFG1,
> > -		ARM_SMMU_EVTQ_IRQ_CFG2,
> > -	},
> > -	[GERROR_MSI_INDEX] = {
> > -		ARM_SMMU_GERROR_IRQ_CFG0,
> > -		ARM_SMMU_GERROR_IRQ_CFG1,
> > -		ARM_SMMU_GERROR_IRQ_CFG2,
> > -	},
> > -	[PRIQ_MSI_INDEX] = {
> > -		ARM_SMMU_PRIQ_IRQ_CFG0,
> > -		ARM_SMMU_PRIQ_IRQ_CFG1,
> > -		ARM_SMMU_PRIQ_IRQ_CFG2,
> > -	},
> > -};
> > -
> >  struct arm_smmu_cmdq_ent {
> >  	/* Common fields */
> >  	u8				opcode;
> > @@ -504,10 +479,6 @@ struct arm_smmu_cmdq_ent {
> >  		} pri;
> >  
> >  		#define CMDQ_OP_CMD_SYNC	0x46
> > -		struct {
> > -			u32			msidata;
> > -			u64			msiaddr;
> > -		} sync;
> >  	};
> >  };
> >  
> > @@ -649,12 +620,6 @@ struct arm_smmu_device {
> >  
> >  	struct arm_smmu_strtab_cfg	strtab_cfg;
> >  
> > -	/* Hi16xx adds an extra 32 bits of goodness to its MSI payload */
> > -	union {
> > -		u32			sync_count;
> > -		u64			padding;
> > -	};
> > -
> >  	/* IOMMU core code handle */
> >  	struct iommu_device		iommu;
> >  };
> > @@ -945,20 +910,7 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
> >  		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_RESP, ent->pri.resp);
> >  		break;
> >  	case CMDQ_OP_CMD_SYNC:
> > -		if (ent->sync.msiaddr)
> > -			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_IRQ);
> > -		else
> > -			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
> > -		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSH, ARM_SMMU_SH_ISH);
> > -		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIATTR, ARM_SMMU_MEMATTR_OIWB);
> > -		/*
> > -		 * Commands are written little-endian, but we want the SMMU to
> > -		 * receive MSIData, and thus write it back to memory, in CPU
> > -		 * byte order, so big-endian needs an extra byteswap here.
> > -		 */
> > -		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIDATA,
> > -				     cpu_to_le32(ent->sync.msidata));
> > -		cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
> > +		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
> >  		break;
> >  	default:
> >  		return -ENOENT;
> > @@ -1054,50 +1006,6 @@ static void arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
> >  	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
> >  }
> >  
> > -/*
> > - * The difference between val and sync_idx is bounded by the maximum size of
> > - * a queue at 2^20 entries, so 32 bits is plenty for wrap-safe arithmetic.
> > - */
> > -static int __arm_smmu_sync_poll_msi(struct arm_smmu_device *smmu, u32 sync_idx)
> > -{
> > -	ktime_t timeout;
> > -	u32 val;
> > -
> > -	timeout = ktime_add_us(ktime_get(), ARM_SMMU_CMDQ_SYNC_TIMEOUT_US);
> > -	val = smp_cond_load_acquire(&smmu->sync_count,
> > -				    (int)(VAL - sync_idx) >= 0 ||
> > -				    !ktime_before(ktime_get(), timeout));
> > -
> > -	return (int)(val - sync_idx) < 0 ? -ETIMEDOUT : 0;
> > -}
> > -
> > -static int __arm_smmu_cmdq_issue_sync_msi(struct arm_smmu_device *smmu)
> > -{
> > -	u64 cmd[CMDQ_ENT_DWORDS];
> > -	unsigned long flags;
> > -	struct arm_smmu_cmdq_ent  ent = {
> > -		.opcode = CMDQ_OP_CMD_SYNC,
> > -		.sync	= {
> > -			.msiaddr = virt_to_phys(&smmu->sync_count),
> > -		},
> > -	};
> > -
> > -	spin_lock_irqsave(&smmu->cmdq.lock, flags);
> > -
> > -	/* Piggy-back on the previous command if it's a SYNC */
> > -	if (smmu->prev_cmd_opcode == CMDQ_OP_CMD_SYNC) {
> > -		ent.sync.msidata = smmu->sync_nr;
> > -	} else {
> > -		ent.sync.msidata = ++smmu->sync_nr;
> > -		arm_smmu_cmdq_build_cmd(cmd, &ent);
> > -		arm_smmu_cmdq_insert_cmd(smmu, cmd);
> > -	}
> > -
> > -	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
> > -
> > -	return __arm_smmu_sync_poll_msi(smmu, ent.sync.msidata);
> > -}
> > -
> >  static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
> >  {
> >  	u64 cmd[CMDQ_ENT_DWORDS];
> > @@ -1119,12 +1027,9 @@ static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
> >  static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
> >  {
> >  	int ret;
> > -	bool msi = (smmu->features & ARM_SMMU_FEAT_MSI) &&
> > -		   (smmu->features & ARM_SMMU_FEAT_COHERENCY);
> >  
> > -	ret = msi ? __arm_smmu_cmdq_issue_sync_msi(smmu)
> > -		  : __arm_smmu_cmdq_issue_sync(smmu);
> > -	if (ret)
> > +	ret = __arm_smmu_cmdq_issue_sync(smmu);
> > +	if ( ret )
> >  		dev_err_ratelimited(smmu->dev, "CMD_SYNC timeout\n");
> >  	return ret;
> >  }
> > @@ -2898,83 +2803,10 @@ static int arm_smmu_update_gbpa(struct arm_smmu_device *smmu, u32 set, u32 clr)
> >  	return ret;
> >  }
> >  
> > -static void arm_smmu_free_msis(void *data)
> > -{
> > -	struct device *dev = data;
> > -	platform_msi_domain_free_irqs(dev);
> > -}
> > -
> > -static void arm_smmu_write_msi_msg(struct msi_desc *desc, struct msi_msg *msg)
> > -{
> > -	phys_addr_t doorbell;
> > -	struct device *dev = msi_desc_to_dev(desc);
> > -	struct arm_smmu_device *smmu = dev_get_drvdata(dev);
> > -	phys_addr_t *cfg = arm_smmu_msi_cfg[desc->platform.msi_index];
> > -
> > -	doorbell = (((u64)msg->address_hi) << 32) | msg->address_lo;
> > -	doorbell &= MSI_CFG0_ADDR_MASK;
> > -
> > -	writeq_relaxed(doorbell, smmu->base + cfg[0]);
> > -	writel_relaxed(msg->data, smmu->base + cfg[1]);
> > -	writel_relaxed(ARM_SMMU_MEMATTR_DEVICE_nGnRE, smmu->base + cfg[2]);
> > -}
> > -
> > -static void arm_smmu_setup_msis(struct arm_smmu_device *smmu)
> > -{
> > -	struct msi_desc *desc;
> > -	int ret, nvec = ARM_SMMU_MAX_MSIS;
> > -	struct device *dev = smmu->dev;
> > -
> > -	/* Clear the MSI address regs */
> > -	writeq_relaxed(0, smmu->base + ARM_SMMU_GERROR_IRQ_CFG0);
> > -	writeq_relaxed(0, smmu->base + ARM_SMMU_EVTQ_IRQ_CFG0);
> > -
> > -	if (smmu->features & ARM_SMMU_FEAT_PRI)
> > -		writeq_relaxed(0, smmu->base + ARM_SMMU_PRIQ_IRQ_CFG0);
> > -	else
> > -		nvec--;
> > -
> > -	if (!(smmu->features & ARM_SMMU_FEAT_MSI))
> > -		return;
> > -
> > -	if (!dev->msi_domain) {
> > -		dev_info(smmu->dev, "msi_domain absent - falling back to wired irqs\n");
> > -		return;
> > -	}
> > -
> > -	/* Allocate MSIs for evtq, gerror and priq. Ignore cmdq */
> > -	ret = platform_msi_domain_alloc_irqs(dev, nvec, arm_smmu_write_msi_msg);
> > -	if (ret) {
> > -		dev_warn(dev, "failed to allocate MSIs - falling back to wired irqs\n");
> > -		return;
> > -	}
> > -
> > -	for_each_msi_entry(desc, dev) {
> > -		switch (desc->platform.msi_index) {
> > -		case EVTQ_MSI_INDEX:
> > -			smmu->evtq.q.irq = desc->irq;
> > -			break;
> > -		case GERROR_MSI_INDEX:
> > -			smmu->gerr_irq = desc->irq;
> > -			break;
> > -		case PRIQ_MSI_INDEX:
> > -			smmu->priq.q.irq = desc->irq;
> > -			break;
> > -		default:	/* Unknown */
> > -			continue;
> > -		}
> > -	}
> > -
> > -	/* Add callback to free MSIs on teardown */
> > -	devm_add_action(dev, arm_smmu_free_msis, dev);
> > -}
> > -
> >  static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
> >  {
> >  	int irq, ret;
> >  
> > -	arm_smmu_setup_msis(smmu);
> > -
> >  	/* Request interrupt lines */
> >  	irq = smmu->evtq.q.irq;
> >  	if (irq) {
> > @@ -3250,8 +3082,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
> >  	if (reg & IDR0_SEV)
> >  		smmu->features |= ARM_SMMU_FEAT_SEV;
> >  
> > -	if (reg & IDR0_MSI)
> > -		smmu->features |= ARM_SMMU_FEAT_MSI;
> >  
> >  	if (reg & IDR0_HYP)
> >  		smmu->features |= ARM_SMMU_FEAT_HYP;
> > -- 
> > 2.17.1
> > 
> 


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 6/8] xen/arm: Remove support for Stage-1 translation on SMMUv3.
  2020-11-26 17:02 ` [PATCH v2 6/8] xen/arm: Remove support for Stage-1 translation " Rahul Singh
@ 2020-12-02  0:53   ` Stefano Stabellini
  2020-12-02 13:13     ` Rahul Singh
  0 siblings, 1 reply; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-02  0:53 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, bertrand.marquis, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

On Thu, 26 Nov 2020, Rahul Singh wrote:
> Linux SMMUv3 driver supports both Stage-1 and Stage-2 translations.
> As of now only Stage-2 translation support has been tested.
> 
> Once Stage-1 translation support is tested this patch can be added.
> 
> Signed-off-by: Rahul Singh <rahul.singh@arm.com>

[...]


> @@ -1871,19 +1476,9 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>  	}
>  
>  	/* Restrict the stage to what we can actually support */
> -	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S1))
> -		smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
> -	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S2))
> -		smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
> +	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;

It would be good to add an helpful error message if
ARM_SMMU_FEAT_TRANS_S2 is missing.


>  	switch (smmu_domain->stage) {
> -	case ARM_SMMU_DOMAIN_S1:
> -		ias = (smmu->features & ARM_SMMU_FEAT_VAX) ? 52 : 48;
> -		ias = min_t(unsigned long, ias, VA_BITS);
> -		oas = smmu->ias;
> -		fmt = ARM_64_LPAE_S1;
> -		finalise_stage_fn = arm_smmu_domain_finalise_s1;
> -		break;
>  	case ARM_SMMU_DOMAIN_NESTED:
>  	case ARM_SMMU_DOMAIN_S2:
>  		ias = smmu->ias;


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN
  2020-11-26 17:02 ` [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN Rahul Singh
@ 2020-12-02  1:48   ` Stefano Stabellini
  2020-12-02 14:34     ` Rahul Singh
  2020-12-02 14:45   ` Julien Grall
  1 sibling, 1 reply; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-02  1:48 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, bertrand.marquis, Stefano Stabellini, Julien Grall,
	Volodymyr Babchuk

On Thu, 26 Nov 2020, Rahul Singh wrote:
> struct io_pgtable_ops, struct io_pgtable_cfg, struct iommu_flush_ops,
> and struct iommu_ops related code are linux specific.
> 
> Remove code related to above struct as code is dead code in XEN.

There are still instances of struct io_pgtable_cfg after applying this
patch in the following functions:
- arm_smmu_domain_finalise_s2
- arm_smmu_domain_finalise



> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
> ---
>  xen/drivers/passthrough/arm/smmu-v3.c | 457 --------------------------
>  1 file changed, 457 deletions(-)
> 
> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
> index 40e3890a58..55d1cba194 100644
> --- a/xen/drivers/passthrough/arm/smmu-v3.c
> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
> @@ -599,7 +593,6 @@ struct arm_smmu_domain {
>  	struct arm_smmu_device		*smmu;
>  	struct mutex			init_mutex; /* Protects smmu pointer */
>  
> -	struct io_pgtable_ops		*pgtbl_ops;
>  	bool				non_strict;
>  
>  	enum arm_smmu_domain_stage	stage;
> @@ -1297,74 +1290,6 @@ static void arm_smmu_tlb_inv_context(void *cookie)
>  	arm_smmu_cmdq_issue_sync(smmu);
>  }
>  
> -static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
> -					  size_t granule, bool leaf, void *cookie)
> -{
> -	struct arm_smmu_domain *smmu_domain = cookie;
> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
> -	struct arm_smmu_cmdq_ent cmd = {
> -		.tlbi = {
> -			.leaf	= leaf,
> -			.addr	= iova,
> -		},
> -	};
> -
> -	cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
> -	cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
> -
> -	do {
> -		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
> -		cmd.tlbi.addr += granule;
> -	} while (size -= granule);
> -}
> -
> -static void arm_smmu_tlb_inv_page_nosync(struct iommu_iotlb_gather *gather,
> -					 unsigned long iova, size_t granule,
> -					 void *cookie)
> -{
> -	arm_smmu_tlb_inv_range_nosync(iova, granule, granule, true, cookie);
> -}
> -
> -static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,
> -				  size_t granule, void *cookie)
> -{
> -	struct arm_smmu_domain *smmu_domain = cookie;
> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
> -
> -	arm_smmu_tlb_inv_range_nosync(iova, size, granule, false, cookie);
> -	arm_smmu_cmdq_issue_sync(smmu);
> -}
> -
> -static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,
> -				  size_t granule, void *cookie)
> -{
> -	struct arm_smmu_domain *smmu_domain = cookie;
> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
> -
> -	arm_smmu_tlb_inv_range_nosync(iova, size, granule, true, cookie);
> -	arm_smmu_cmdq_issue_sync(smmu);
> -}
> -
> -static const struct iommu_flush_ops arm_smmu_flush_ops = {
> -	.tlb_flush_all	= arm_smmu_tlb_inv_context,
> -	.tlb_flush_walk = arm_smmu_tlb_inv_walk,
> -	.tlb_flush_leaf = arm_smmu_tlb_inv_leaf,
> -	.tlb_add_page	= arm_smmu_tlb_inv_page_nosync,
> -};
> -
> -/* IOMMU API */
> -static bool arm_smmu_capable(enum iommu_cap cap)
> -{
> -	switch (cap) {
> -	case IOMMU_CAP_CACHE_COHERENCY:
> -		return true;
> -	case IOMMU_CAP_NOEXEC:
> -		return true;
> -	default:
> -		return false;
> -	}
> -}
> -
>  static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
>  {
>  	struct arm_smmu_domain *smmu_domain;
> @@ -1421,7 +1346,6 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>  	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>  
>  	iommu_put_dma_cookie(domain);
> -	free_io_pgtable_ops(smmu_domain->pgtbl_ops);
>  
>  	if (cfg->vmid)
>  		arm_smmu_bitmap_free(smmu->vmid_map, cfg->vmid);
> @@ -1429,7 +1353,6 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>  	kfree(smmu_domain);
>  }
>  
> -
>  static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>  				       struct arm_smmu_master *master,
>  				       struct io_pgtable_cfg *pgtbl_cfg)
> @@ -1437,7 +1360,6 @@ static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>  	int vmid;
>  	struct arm_smmu_device *smmu = smmu_domain->smmu;
>  	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
> -	typeof(&pgtbl_cfg->arm_lpae_s2_cfg.vtcr) vtcr;
>  
>  	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
>  	if (vmid < 0)
> @@ -1461,20 +1383,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>  {
>  	int ret;
>  	unsigned long ias, oas;
> -	enum io_pgtable_fmt fmt;
> -	struct io_pgtable_cfg pgtbl_cfg;
> -	struct io_pgtable_ops *pgtbl_ops;
>  	int (*finalise_stage_fn)(struct arm_smmu_domain *,
>  				 struct arm_smmu_master *,
>  				 struct io_pgtable_cfg *);
>  	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>  	struct arm_smmu_device *smmu = smmu_domain->smmu;
>  
> -	if (domain->type == IOMMU_DOMAIN_IDENTITY) {
> -		smmu_domain->stage = ARM_SMMU_DOMAIN_BYPASS;
> -		return 0;
> -	}
> -
>  	/* Restrict the stage to what we can actually support */
>  	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
>  
> @@ -1483,40 +1397,17 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>  	case ARM_SMMU_DOMAIN_S2:
>  		ias = smmu->ias;
>  		oas = smmu->oas;
> -		fmt = ARM_64_LPAE_S2;
>  		finalise_stage_fn = arm_smmu_domain_finalise_s2;
>  		break;
>  	default:
>  		return -EINVAL;
>  	}
>  
> -	pgtbl_cfg = (struct io_pgtable_cfg) {
> -		.pgsize_bitmap	= smmu->pgsize_bitmap,
> -		.ias		= ias,
> -		.oas		= oas,
> -		.coherent_walk	= smmu->features & ARM_SMMU_FEAT_COHERENCY,
> -		.tlb		= &arm_smmu_flush_ops,
> -		.iommu_dev	= smmu->dev,
> -	};
> -
> -	if (smmu_domain->non_strict)
> -		pgtbl_cfg.quirks |= IO_PGTABLE_QUIRK_NON_STRICT;
> -
> -	pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
> -	if (!pgtbl_ops)
> -		return -ENOMEM;
> -
> -	domain->pgsize_bitmap = pgtbl_cfg.pgsize_bitmap;
> -	domain->geometry.aperture_end = (1UL << pgtbl_cfg.ias) - 1;
> -	domain->geometry.force_aperture = true;
> -
>  	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
>  	if (ret < 0) {
> -		free_io_pgtable_ops(pgtbl_ops);
>  		return ret;
>  	}
>  
> -	smmu_domain->pgtbl_ops = pgtbl_ops;
>  	return 0;
>  }
>  
> @@ -1626,71 +1517,6 @@ out_unlock:
>  	return ret;
>  }
>  
> -static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
> -			phys_addr_t paddr, size_t size, int prot, gfp_t gfp)
> -{
> -	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
> -
> -	if (!ops)
> -		return -ENODEV;
> -
> -	return ops->map(ops, iova, paddr, size, prot, gfp);
> -}
> -
> -static size_t arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova,
> -			     size_t size, struct iommu_iotlb_gather *gather)
> -{
> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> -	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
> -
> -	if (!ops)
> -		return 0;
> -
> -	return ops->unmap(ops, iova, size, gather);
> -}
> -
> -static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain)
> -{
> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> -
> -	if (smmu_domain->smmu)
> -		arm_smmu_tlb_inv_context(smmu_domain);
> -}
> -
> -static void arm_smmu_iotlb_sync(struct iommu_domain *domain,
> -				struct iommu_iotlb_gather *gather)
> -{
> -	struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
> -
> -	if (smmu)
> -		arm_smmu_cmdq_issue_sync(smmu);
> -}
> -
> -static phys_addr_t
> -arm_smmu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
> -{
> -	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
> -
> -	if (domain->type == IOMMU_DOMAIN_IDENTITY)
> -		return iova;
> -
> -	if (!ops)
> -		return 0;
> -
> -	return ops->iova_to_phys(ops, iova);
> -}
> -
> -static struct platform_driver arm_smmu_driver;
> -
> -static
> -struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle *fwnode)
> -{
> -	struct device *dev = driver_find_device_by_fwnode(&arm_smmu_driver.driver,
> -							  fwnode);
> -	put_device(dev);
> -	return dev ? dev_get_drvdata(dev) : NULL;
> -}
> -
>  static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
>  {
>  	unsigned long limit = smmu->strtab_cfg.num_l1_ents;
> @@ -1701,206 +1527,6 @@ static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
>  	return sid < limit;
>  }
>  
> -static struct iommu_ops arm_smmu_ops;
> -
> -static struct iommu_device *arm_smmu_probe_device(struct device *dev)
> -{
> -	int i, ret;
> -	struct arm_smmu_device *smmu;
> -	struct arm_smmu_master *master;
> -	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> -
> -	if (!fwspec || fwspec->ops != &arm_smmu_ops)
> -		return ERR_PTR(-ENODEV);
> -
> -	if (WARN_ON_ONCE(dev_iommu_priv_get(dev)))
> -		return ERR_PTR(-EBUSY);
> -
> -	smmu = arm_smmu_get_by_fwnode(fwspec->iommu_fwnode);
> -	if (!smmu)
> -		return ERR_PTR(-ENODEV);
> -
> -	master = kzalloc(sizeof(*master), GFP_KERNEL);
> -	if (!master)
> -		return ERR_PTR(-ENOMEM);
> -
> -	master->dev = dev;
> -	master->smmu = smmu;
> -	master->sids = fwspec->ids;
> -	master->num_sids = fwspec->num_ids;
> -	dev_iommu_priv_set(dev, master);
> -
> -	/* Check the SIDs are in range of the SMMU and our stream table */
> -	for (i = 0; i < master->num_sids; i++) {
> -		u32 sid = master->sids[i];
> -
> -		if (!arm_smmu_sid_in_range(smmu, sid)) {
> -			ret = -ERANGE;
> -			goto err_free_master;
> -		}
> -
> -		/* Ensure l2 strtab is initialised */
> -		if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
> -			ret = arm_smmu_init_l2_strtab(smmu, sid);
> -			if (ret)
> -				goto err_free_master;
> -		}
> -	}
> -
> -	return &smmu->iommu;
> -
> -err_free_master:
> -	kfree(master);
> -	dev_iommu_priv_set(dev, NULL);
> -	return ERR_PTR(ret);
> -}
> -
> -static void arm_smmu_release_device(struct device *dev)
> -{
> -	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> -	struct arm_smmu_master *master;
> -
> -	if (!fwspec || fwspec->ops != &arm_smmu_ops)
> -		return;
> -
> -	master = dev_iommu_priv_get(dev);
> -	arm_smmu_detach_dev(master);
> -	kfree(master);
> -	iommu_fwspec_free(dev);
> -}
> -
> -static struct iommu_group *arm_smmu_device_group(struct device *dev)
> -{
> -	struct iommu_group *group;
> -
> -	/*
> -	 * We don't support devices sharing stream IDs other than PCI RID
> -	 * aliases, since the necessary ID-to-device lookup becomes rather
> -	 * impractical given a potential sparse 32-bit stream ID space.
> -	 */
> -	if (dev_is_pci(dev))
> -		group = pci_device_group(dev);
> -	else
> -		group = generic_device_group(dev);
> -
> -	return group;
> -}
> -
> -static int arm_smmu_domain_get_attr(struct iommu_domain *domain,
> -				    enum iommu_attr attr, void *data)
> -{
> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> -
> -	switch (domain->type) {
> -	case IOMMU_DOMAIN_UNMANAGED:
> -		switch (attr) {
> -		case DOMAIN_ATTR_NESTING:
> -			*(int *)data = (smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED);
> -			return 0;
> -		default:
> -			return -ENODEV;
> -		}
> -		break;
> -	case IOMMU_DOMAIN_DMA:
> -		switch (attr) {
> -		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
> -			*(int *)data = smmu_domain->non_strict;
> -			return 0;
> -		default:
> -			return -ENODEV;
> -		}
> -		break;
> -	default:
> -		return -EINVAL;
> -	}
> -}
> -
> -static int arm_smmu_domain_set_attr(struct iommu_domain *domain,
> -				    enum iommu_attr attr, void *data)
> -{
> -	int ret = 0;
> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> -
> -	mutex_lock(&smmu_domain->init_mutex);
> -
> -	switch (domain->type) {
> -	case IOMMU_DOMAIN_UNMANAGED:
> -		switch (attr) {
> -		case DOMAIN_ATTR_NESTING:
> -			if (smmu_domain->smmu) {
> -				ret = -EPERM;
> -				goto out_unlock;
> -			}
> -
> -			if (*(int *)data)
> -				smmu_domain->stage = ARM_SMMU_DOMAIN_NESTED;
> -			else
> -				smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
> -			break;
> -		default:
> -			ret = -ENODEV;
> -		}
> -		break;
> -	case IOMMU_DOMAIN_DMA:
> -		switch(attr) {
> -		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
> -			smmu_domain->non_strict = *(int *)data;
> -			break;
> -		default:
> -			ret = -ENODEV;
> -		}
> -		break;
> -	default:
> -		ret = -EINVAL;
> -	}
> -
> -out_unlock:
> -	mutex_unlock(&smmu_domain->init_mutex);
> -	return ret;
> -}
> -
> -static int arm_smmu_of_xlate(struct device *dev, struct of_phandle_args *args)
> -{
> -	return iommu_fwspec_add_ids(dev, args->args, 1);
> -}
> -
> -static void arm_smmu_get_resv_regions(struct device *dev,
> -				      struct list_head *head)
> -{
> -	struct iommu_resv_region *region;
> -	int prot = IOMMU_WRITE | IOMMU_NOEXEC | IOMMU_MMIO;
> -
> -	region = iommu_alloc_resv_region(MSI_IOVA_BASE, MSI_IOVA_LENGTH,
> -					 prot, IOMMU_RESV_SW_MSI);
> -	if (!region)
> -		return;
> -
> -	list_add_tail(&region->list, head);
> -
> -	iommu_dma_get_resv_regions(dev, head);
> -}

Arguably this could have been removed previously as part of the MSI
patch, but that's OK either way.


> -static struct iommu_ops arm_smmu_ops = {
> -	.capable		= arm_smmu_capable,
> -	.domain_alloc		= arm_smmu_domain_alloc,
> -	.domain_free		= arm_smmu_domain_free,
> -	.attach_dev		= arm_smmu_attach_dev,
> -	.map			= arm_smmu_map,
> -	.unmap			= arm_smmu_unmap,
> -	.flush_iotlb_all	= arm_smmu_flush_iotlb_all,
> -	.iotlb_sync		= arm_smmu_iotlb_sync,
> -	.iova_to_phys		= arm_smmu_iova_to_phys,
> -	.probe_device		= arm_smmu_probe_device,
> -	.release_device		= arm_smmu_release_device,
> -	.device_group		= arm_smmu_device_group,
> -	.domain_get_attr	= arm_smmu_domain_get_attr,
> -	.domain_set_attr	= arm_smmu_domain_set_attr,
> -	.of_xlate		= arm_smmu_of_xlate,
> -	.get_resv_regions	= arm_smmu_get_resv_regions,
> -	.put_resv_regions	= generic_iommu_put_resv_regions,
> -	.pgsize_bitmap		= -1UL, /* Restricted during device attach */
> -};
> -
>  /* Probing and initialisation functions */
>  static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>  				   struct arm_smmu_queue *q,
> @@ -2515,21 +2139,10 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>  	default:
>  		dev_info(smmu->dev,
>  			"unknown output address size. Truncating to 48-bit\n");
> -		fallthrough;
>  	case IDR5_OAS_48_BIT:
>  		smmu->oas = 48;
>  	}
>  
> -	if (arm_smmu_ops.pgsize_bitmap == -1UL)
> -		arm_smmu_ops.pgsize_bitmap = smmu->pgsize_bitmap;
> -	else
> -		arm_smmu_ops.pgsize_bitmap |= smmu->pgsize_bitmap;
> -
> -	/* Set the DMA mask for our table walker */
> -	if (dma_set_mask_and_coherent(smmu->dev, DMA_BIT_MASK(smmu->oas)))
> -		dev_warn(smmu->dev,
> -			 "failed to set DMA mask for table walker\n");
> -
>  	smmu->ias = max(smmu->ias, smmu->oas);
>  
>  	dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n",
> @@ -2595,9 +2208,6 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev,
>  
>  	parse_driver_options(smmu);
>  
> -	if (of_dma_is_coherent(dev->of_node))
> -		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
> -

Why this change? The ARM_SMMU_FEAT_COHERENCY flag is still used in
arm_smmu_device_hw_probe.


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-11-26 17:02 ` [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
@ 2020-12-02  2:51   ` Stefano Stabellini
  2020-12-02 16:27     ` Rahul Singh
  2020-12-02 16:47     ` Julien Grall
  2020-12-02 16:22   ` Julien Grall
  1 sibling, 2 replies; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-02  2:51 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, bertrand.marquis, Andrew Cooper, George Dunlap,
	Ian Jackson, Jan Beulich, Julien Grall, Stefano Stabellini,
	Wei Liu, Paul Durrant, Volodymyr Babchuk

On Thu, 26 Nov 2020, Rahul Singh wrote:
> Add support for ARM architected SMMUv3 implementation. It is based on
> the Linux SMMUv3 driver.
> 
> Major differences with regard to Linux driver are as follows:
> 1. Only Stage-2 translation is supported as compared to the Linux driver
>    that supports both Stage-1 and Stage-2 translations.
> 2. Use P2M  page table instead of creating one as SMMUv3 has the
>    capability to share the page tables with the CPU.
> 3. Tasklets are used in place of threaded IRQ's in Linux for event queue
>    and priority queue IRQ handling.
> 4. Latest version of the Linux SMMUv3 code implements the commands queue
>    access functions based on atomic operations implemented in Linux.
>    Atomic functions used by the commands queue access functions are not
>    implemented in XEN therefore we decided to port the earlier version
>    of the code. Once the proper atomic operations will be available in
>    XEN the driver can be updated.
> 5. Driver is currently supported as Tech Preview.

This patch is big and was difficult to review, nonetheless I tried :-)

That said, the code is self-contained, marked as Tech Preview, and not
at risk of making other things unstable, so it is low risk to accept it.

Comments below.


> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
> ---
>  MAINTAINERS                           |   6 +
>  SUPPORT.md                            |   1 +
>  xen/drivers/passthrough/Kconfig       |  10 +
>  xen/drivers/passthrough/arm/Makefile  |   1 +
>  xen/drivers/passthrough/arm/smmu-v3.c | 986 +++++++++++++++++++++-----
>  5 files changed, 814 insertions(+), 190 deletions(-)
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index dab38a6a14..1d63489eec 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -249,6 +249,12 @@ F:	xen/include/asm-arm/
>  F:	xen/include/public/arch-arm/
>  F:	xen/include/public/arch-arm.h
>  
> +ARM SMMUv3
> +M:	Bertrand Marquis <bertrand.marquis@arm.com>
> +M:	Rahul Singh <rahul.singh@arm.com>
> +S:	Supported
> +F:	xen/drivers/passthrough/arm/smmu-v3.c
> +
>  Change Log
>  M:	Paul Durrant <paul@xen.org>
>  R:	Community Manager <community.manager@xenproject.org>
> diff --git a/SUPPORT.md b/SUPPORT.md
> index ab02aca5f4..e402c7202d 100644
> --- a/SUPPORT.md
> +++ b/SUPPORT.md
> @@ -68,6 +68,7 @@ For the Cortex A57 r0p0 - r1p1, see Errata 832075.
>      Status, ARM SMMUv1: Supported, not security supported
>      Status, ARM SMMUv2: Supported, not security supported
>      Status, Renesas IPMMU-VMSA: Supported, not security supported
> +    Status, ARM SMMUv3: Tech Preview
>  
>  ### ARM/GICv3 ITS
>  
> diff --git a/xen/drivers/passthrough/Kconfig b/xen/drivers/passthrough/Kconfig
> index 0036007ec4..5b71c59f47 100644
> --- a/xen/drivers/passthrough/Kconfig
> +++ b/xen/drivers/passthrough/Kconfig
> @@ -13,6 +13,16 @@ config ARM_SMMU
>  	  Say Y here if your SoC includes an IOMMU device implementing the
>  	  ARM SMMU architecture.
>  
> +config ARM_SMMU_V3
> +	bool "ARM Ltd. System MMU Version 3 (SMMUv3) Support" if EXPERT
> +	depends on ARM_64
> +	---help---
> +	 Support for implementations of the ARM System MMU architecture
> +	 version 3.
> +
> +	 Say Y here if your system includes an IOMMU device implementing
> +	 the ARM SMMUv3 architecture.
> +
>  config IPMMU_VMSA
>  	bool "Renesas IPMMU-VMSA found in R-Car Gen3 SoCs"
>  	depends on ARM_64
> diff --git a/xen/drivers/passthrough/arm/Makefile b/xen/drivers/passthrough/arm/Makefile
> index fcd918ea3e..c5fb3b58a5 100644
> --- a/xen/drivers/passthrough/arm/Makefile
> +++ b/xen/drivers/passthrough/arm/Makefile
> @@ -1,3 +1,4 @@
>  obj-y += iommu.o iommu_helpers.o iommu_fwspec.o
>  obj-$(CONFIG_ARM_SMMU) += smmu.o
>  obj-$(CONFIG_IPMMU_VMSA) += ipmmu-vmsa.o
> +obj-$(CONFIG_ARM_SMMU_V3) += smmu-v3.o
> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
> index 55d1cba194..8f2337e7f2 100644
> --- a/xen/drivers/passthrough/arm/smmu-v3.c
> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
> @@ -2,36 +2,280 @@
>  /*
>   * IOMMU API for ARM architected SMMUv3 implementations.
>   *
> - * Copyright (C) 2015 ARM Limited
> + * Based on Linux's SMMUv3 driver:
> + *    drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
> + *    commit: 951cbbc386ff01b50da4f46387e994e81d9ab431
> + * and Xen's SMMU driver:
> + *    xen/drivers/passthrough/arm/smmu.c
>   *
> - * Author: Will Deacon <will.deacon@arm.com>
> + * Copyright (C) 2015 ARM Limited Will Deacon <will.deacon@arm.com>
>   *
> - * This driver is powered by bad coffee and bombay mix.
> + * Copyright (C) 2020 Arm Ltd.
> + *
> + * 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, see <http://www.gnu.org/licenses/>.
> + *
> + */
> +
> +#include <xen/acpi.h>
> +#include <xen/config.h>
> +#include <xen/delay.h>
> +#include <xen/errno.h>
> +#include <xen/err.h>
> +#include <xen/irq.h>
> +#include <xen/lib.h>
> +#include <xen/list.h>
> +#include <xen/mm.h>
> +#include <xen/rbtree.h>
> +#include <xen/sched.h>
> +#include <xen/sizes.h>
> +#include <xen/vmap.h>
> +#include <asm/atomic.h>
> +#include <asm/device.h>
> +#include <asm/io.h>
> +#include <asm/platform.h>
> +#include <asm/iommu_fwspec.h>
> +
> +/* Linux compatibility functions. */
> +typedef paddr_t dma_addr_t;
> +typedef unsigned int gfp_t;
> +
> +#define platform_device device
> +
> +#define GFP_KERNEL 0
> +
> +/* Alias to Xen device tree helpers */
> +#define device_node dt_device_node
> +#define of_phandle_args dt_phandle_args
> +#define of_device_id dt_device_match
> +#define of_match_node dt_match_node
> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, pname, out))
> +#define of_property_read_bool dt_property_read_bool
> +#define of_parse_phandle_with_args dt_parse_phandle_with_args

Given all the changes to the file by the previous patches we are
basically fully (or almost fully) adapting this code to Xen.

So at that point I wonder if we should just as well make these changes
(e.g. s/of_phandle_args/dt_phandle_args/g) to the code too.

Julien, Rahul, what do you guys think?


> +/* Alias to Xen lock functions */

I think this deserves at least one statement to explain why it is OK.

Also, a similar comment to the one above: maybe we should add a couple
of more preparation patches to s/mutex/spinlock/g and change the time
and allocation functions too.


> +#define mutex spinlock
> +#define mutex_init spin_lock_init
> +#define mutex_lock spin_lock
> +#define mutex_unlock spin_unlock
> +
> +/* Alias to Xen time functions */
> +#define ktime_t s_time_t
> +#define ktime_get()             (NOW())
> +#define ktime_add_us(t,i)       (t + MICROSECS(i))
> +#define ktime_compare(t,i)      (t > (i))
> +
> +/* Alias to Xen allocation helpers */
> +#define kzalloc(size, flags)    _xzalloc(size, sizeof(void *))
> +#define kfree xfree
> +#define devm_kzalloc(dev, size, flags)  _xzalloc(size, sizeof(void *))
> +
> +/* Device logger functions */
> +#define dev_name(dev) dt_node_full_name(dev->of_node)
> +#define dev_dbg(dev, fmt, ...)      \
> +    printk(XENLOG_DEBUG "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +#define dev_notice(dev, fmt, ...)   \
> +    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +#define dev_warn(dev, fmt, ...)     \
> +    printk(XENLOG_WARNING "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +#define dev_err(dev, fmt, ...)      \
> +    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +#define dev_info(dev, fmt, ...)     \
> +    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +#define dev_err_ratelimited(dev, fmt, ...)      \
> +    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +
> +/*
> + * Periodically poll an address and wait between reads in us until a
> + * condition is met or a timeout occurs.

It would be good to add a statement to point out what the return value
is going to be.


> + */
> +#define readx_poll_timeout(op, addr, val, cond, sleep_us, timeout_us) \
> +({ \
> +     s_time_t deadline = NOW() + MICROSECS(timeout_us); \
> +     for (;;) { \
> +        (val) = op(addr); \
> +        if (cond) \
> +            break; \
> +        if (NOW() > deadline) { \
> +            (val) = op(addr); \
> +            break; \
> +        } \
> +        udelay(sleep_us); \
> +     } \
> +     (cond) ? 0 : -ETIMEDOUT; \
> +})
> +
> +#define readl_relaxed_poll_timeout(addr, val, cond, delay_us, timeout_us) \
> +    readx_poll_timeout(readl_relaxed, addr, val, cond, delay_us, timeout_us)
> +
> +#define FIELD_PREP(_mask, _val)         \
> +    (((typeof(_mask))(_val) << (__builtin_ffsll(_mask) - 1)) & (_mask))

Let's add the definition of ffsll to bitops.h


> +#define FIELD_GET(_mask, _reg)          \
> +    (typeof(_mask))(((_reg) & (_mask)) >> (__builtin_ffsll(_mask) - 1))
> +
> +#define WRITE_ONCE(x, val)                  \
> +do {                                        \
> +    *(volatile typeof(x) *)&(x) = (val);    \
> +} while (0)

maybe we should define this in xen/include/xen/lib.h


> +
> +/* Xen: Stub out DMA domain related functions */
> +#define iommu_get_dma_cookie(dom) 0
> +#define iommu_put_dma_cookie(dom)

Shouldn't we remove any call to iommu_get_dma_cookie and
iommu_put_dma_cookie in one of the previous patches?


> +/*
> + * Helpers for DMA allocation. Just the function name is reused for
> + * porting code, these allocation are not managed allocations
>   */
> +static void *dmam_alloc_coherent(struct device *dev, size_t size,
> +                                 paddr_t *dma_handle, gfp_t gfp)
> +{
> +    void *vaddr;
> +    unsigned long alignment = size;
> +
> +    /*
> +     * _xzalloc requires that the (align & (align -1)) = 0. Most of the
> +     * allocations in SMMU code should send the right value for size. In
> +     * case this is not true print a warning and align to the size of a
> +     * (void *)
> +     */
> +    if ( size & (size - 1) )
> +    {
> +        printk(XENLOG_WARNING "SMMUv3: Fixing alignment for the DMA buffer\n");
> +        alignment = sizeof(void *);
> +    }
> +
> +    vaddr = _xzalloc(size, alignment);
> +    if ( !vaddr )
> +    {
> +        printk(XENLOG_ERR "SMMUv3: DMA allocation failed\n");
> +        return NULL;
> +    }
> +
> +    *dma_handle = virt_to_maddr(vaddr);
> +
> +    return vaddr;
> +}
> +
> +/* Xen: Type definitions for iommu_domain */
> +#define IOMMU_DOMAIN_UNMANAGED 0
> +#define IOMMU_DOMAIN_DMA 1
> +#define IOMMU_DOMAIN_IDENTITY 2
> +
> +/* Xen specific code. */
> +struct iommu_domain {
> +    /* Runtime SMMU configuration for this iommu_domain */
> +    atomic_t ref;
> +    /*
> +     * Used to link iommu_domain contexts for a same domain.
> +     * There is at least one per-SMMU to used by the domain.
> +     */
> +    struct list_head    list;
> +};
> +
> +/* Describes information required for a Xen domain */
> +struct arm_smmu_xen_domain {
> +    spinlock_t      lock;
> +
> +    /* List of iommu domains associated to this domain */
> +    struct list_head    contexts;
> +};
> +
> +/*
> + * Information about each device stored in dev->archdata.iommu
> + * The dev->archdata.iommu stores the iommu_domain (runtime configuration of
> + * the SMMU).
> + */
> +struct arm_smmu_xen_device {
> +    struct iommu_domain *domain;
> +};

Do we need both struct arm_smmu_xen_device and struct iommu_domain?


> +/* Keep a list of devices associated with this driver */
> +static DEFINE_SPINLOCK(arm_smmu_devices_lock);
> +static LIST_HEAD(arm_smmu_devices);
> +
> +
> +static inline void *dev_iommu_priv_get(struct device *dev)
> +{
> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> +
> +    return fwspec && fwspec->iommu_priv ? fwspec->iommu_priv : NULL;
> +}
> +
> +static inline void dev_iommu_priv_set(struct device *dev, void *priv)
> +{
> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> +
> +    fwspec->iommu_priv = priv;
> +}
> +
> +int dt_property_match_string(const struct dt_device_node *np,
> +                             const char *propname, const char *string)
> +{
> +    const struct dt_property *dtprop = dt_find_property(np, propname, NULL);
> +    size_t l;
> +    int i;
> +    const char *p, *end;
> +
> +    if ( !dtprop )
> +        return -EINVAL;
> +
> +    if ( !dtprop->value )
> +        return -ENODATA;
> +
> +    p = dtprop->value;
> +    end = p + dtprop->length;
> +
> +    for ( i = 0; p < end; i++, p += l )
> +    {
> +        l = strnlen(p, end - p) + 1;
> +
> +        if ( p + l > end )
> +            return -EILSEQ;
> +
> +        if ( strcmp(string, p) == 0 )
> +            return i; /* Found it; return index */
> +    }
> +
> +    return -ENODATA;
> +}

I think you should either use dt_property_read_string or move the
implementation of dt_property_match_string to xen/common/device_tree.c
(in which case I suggest to do it in a separate patch.)

I'd just use dt_property_read_string.



> +static int platform_get_irq_byname_optional(struct device *dev,
> +                                            const char *name)
> +{
> +    int index, ret;
> +    struct dt_device_node *np  = dev_to_dt(dev);
> +
> +    if ( unlikely(!name) )
> +        return -EINVAL;
> +
> +    index = dt_property_match_string(np, "interrupt-names", name);
> +    if ( index < 0 )
> +    {
> +        dev_info(dev, "IRQ %s not found\n", name);
> +        return index;
> +    }
>  
> -#include <linux/acpi.h>
> -#include <linux/acpi_iort.h>
> -#include <linux/bitfield.h>
> -#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>
> -#include <linux/iommu.h>
> -#include <linux/iopoll.h>
> -#include <linux/module.h>
> -#include <linux/msi.h>
> -#include <linux/of.h>
> -#include <linux/of_address.h>
> -#include <linux/of_iommu.h>
> -#include <linux/of_platform.h>
> -#include <linux/pci.h>
> -#include <linux/pci-ats.h>
> -#include <linux/platform_device.h>
> -
> -#include <linux/amba/bus.h>
> +    ret = platform_get_irq(np, index);
> +    if ( ret < 0 )
> +    {
> +        dev_err(dev, "failed to get irq index %d\n", index);
> +        return -ENODEV;
> +    }
> +
> +    return ret;
> +}
> +
> +/* Start of Linux SMMUv3 code */
>  
>  /* MMIO registers */
>  #define ARM_SMMU_IDR0			0x0
> @@ -507,6 +751,7 @@ struct arm_smmu_s2_cfg {
>  	u16				vmid;
>  	u64				vttbr;
>  	u64				vtcr;
> +	struct domain		*domain;
>  };
>  
>  struct arm_smmu_strtab_cfg {
> @@ -567,8 +812,13 @@ struct arm_smmu_device {
>  
>  	struct arm_smmu_strtab_cfg	strtab_cfg;
>  
> -	/* IOMMU core code handle */
> -	struct iommu_device		iommu;
> +	/* Need to keep a list of SMMU devices */
> +	struct list_head		devices;
> +
> +	/* Tasklets for handling evts/faults and pci page request IRQs*/
> +	struct tasklet		evtq_irq_tasklet;
> +	struct tasklet		priq_irq_tasklet;
> +	struct tasklet		combined_irq_tasklet;
>  };
>  
>  /* SMMU private data for each master */
> @@ -1110,7 +1360,7 @@ static int arm_smmu_init_l2_strtab(struct arm_smmu_device *smmu, u32 sid)
>  }
>  
>  /* IRQ and event handlers */
> -static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
> +static void arm_smmu_evtq_thread(void *dev)

I think we shouldn't call it a thread given that's a tasklet. We should
rename the function or it will be confusing.


>  {
>  	int i;
>  	struct arm_smmu_device *smmu = dev;
> @@ -1140,7 +1390,6 @@ static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
>  	/* Sync our overflow flag, as we believe we're up to speed */
>  	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
>  		    Q_IDX(llq, llq->cons);
> -	return IRQ_HANDLED;
>  }
>  
>  static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
> @@ -1181,7 +1430,7 @@ static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
>  	}
>  }
>  
> -static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
> +static void arm_smmu_priq_thread(void *dev)

same here


>  {
>  	struct arm_smmu_device *smmu = dev;
>  	struct arm_smmu_queue *q = &smmu->priq.q;
> @@ -1200,12 +1449,12 @@ static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
>  	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
>  		      Q_IDX(llq, llq->cons);
>  	queue_sync_cons_out(q);
> -	return IRQ_HANDLED;
>  }
>  
>  static int arm_smmu_device_disable(struct arm_smmu_device *smmu);
>  
> -static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
> +static void arm_smmu_gerror_handler(int irq, void *dev,
> +				struct cpu_user_regs *regs)
>  {
>  	u32 gerror, gerrorn, active;
>  	struct arm_smmu_device *smmu = dev;
> @@ -1215,7 +1464,7 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>  
>  	active = gerror ^ gerrorn;
>  	if (!(active & GERROR_ERR_MASK))
> -		return IRQ_NONE; /* No errors pending */
> +		return; /* No errors pending */
>  
>  	dev_warn(smmu->dev,
>  		 "unexpected global error reported (0x%08x), this could be serious\n",
> @@ -1248,26 +1497,42 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>  		arm_smmu_cmdq_skip_err(smmu);
>  
>  	writel(gerror, smmu->base + ARM_SMMU_GERRORN);
> -	return IRQ_HANDLED;
>  }
>  
> -static irqreturn_t arm_smmu_combined_irq_thread(int irq, void *dev)
> +static void arm_smmu_combined_irq_handler(int irq, void *dev,
> +				struct cpu_user_regs *regs)
> +{
> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
> +
> +	arm_smmu_gerror_handler(irq, dev, regs);
> +
> +	tasklet_schedule(&(smmu->combined_irq_tasklet));
> +}
> +
> +static void arm_smmu_combined_irq_thread(void *dev)
>  {
>  	struct arm_smmu_device *smmu = dev;
>  
> -	arm_smmu_evtq_thread(irq, dev);
> +	arm_smmu_evtq_thread(dev);
>  	if (smmu->features & ARM_SMMU_FEAT_PRI)
> -		arm_smmu_priq_thread(irq, dev);
> -
> -	return IRQ_HANDLED;
> +		arm_smmu_priq_thread(dev);
>  }
>  
> -static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
> +static void arm_smmu_evtq_irq_tasklet(int irq, void *dev,
> +				struct cpu_user_regs *regs)
>  {
> -	arm_smmu_gerror_handler(irq, dev);
> -	return IRQ_WAKE_THREAD;
> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
> +
> +	tasklet_schedule(&(smmu->evtq_irq_tasklet));
>  }
>  
> +static void arm_smmu_priq_irq_tasklet(int irq, void *dev,
> +				struct cpu_user_regs *regs)
> +{
> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
> +
> +	tasklet_schedule(&(smmu->priq_irq_tasklet));
> +}
>  
>  /* IO_PGTABLE API */
>  static void arm_smmu_tlb_inv_context(void *cookie)
> @@ -1354,27 +1619,69 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>  }
>  
>  static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
> -				       struct arm_smmu_master *master,
> -				       struct io_pgtable_cfg *pgtbl_cfg)
> +				       struct arm_smmu_master *master)
>  {
>  	int vmid;
> +	u64 reg;
>  	struct arm_smmu_device *smmu = smmu_domain->smmu;
>  	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>  
> +	/* VTCR */
> +	reg = VTCR_RES1 | VTCR_SH0_IS | VTCR_IRGN0_WBWA | VTCR_ORGN0_WBWA;
> +
> +	switch (PAGE_SIZE) {
> +	case SZ_4K:
> +		reg |= VTCR_TG0_4K;
> +		break;
> +	case SZ_16K:
> +		reg |= VTCR_TG0_16K;
> +		break;
> +	case SZ_64K:
> +		reg |= VTCR_TG0_4K;
> +		break;
> +	}
> +
> +	switch (smmu->oas) {
> +	case 32:
> +		reg |= VTCR_PS(_AC(0x0,ULL));
> +		break;
> +	case 36:
> +		reg |= VTCR_PS(_AC(0x1,ULL));
> +		break;
> +	case 40:
> +		reg |= VTCR_PS(_AC(0x2,ULL));
> +		break;
> +	case 42:
> +		reg |= VTCR_PS(_AC(0x3,ULL));
> +		break;
> +	case 44:
> +		reg |= VTCR_PS(_AC(0x4,ULL));
> +		break;
> +		case 48:
> +		reg |= VTCR_PS(_AC(0x5,ULL));
> +		break;
> +	case 52:
> +		reg |= VTCR_PS(_AC(0x6,ULL));
> +		break;
> +	}
> +
> +	reg |= VTCR_T0SZ(64ULL - smmu->ias);
> +	reg |= VTCR_SL0(0x2);
> +	reg |= VTCR_VS;
> +
> +	cfg->vtcr   = reg;
> +
>  	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
>  	if (vmid < 0)
>  		return vmid;
> +	cfg->vmid  = (u16)vmid;
> +
> +	cfg->vttbr  = page_to_maddr(cfg->domain->arch.p2m.root);
> +
> +	printk(XENLOG_DEBUG
> +		   "SMMUv3: d%u: vmid 0x%x vtcr 0x%"PRIpaddr" p2maddr 0x%"PRIpaddr"\n",
> +		   cfg->domain->domain_id, cfg->vmid, cfg->vtcr, cfg->vttbr);
>  
> -	vtcr = &pgtbl_cfg->arm_lpae_s2_cfg.vtcr;
> -	cfg->vmid	= (u16)vmid;
> -	cfg->vttbr	= pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
> -	cfg->vtcr	= FIELD_PREP(STRTAB_STE_2_VTCR_S2T0SZ, vtcr->tsz) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SL0, vtcr->sl) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2IR0, vtcr->irgn) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2OR0, vtcr->orgn) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SH0, vtcr->sh) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2TG, vtcr->tg) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2PS, vtcr->ps);
>  	return 0;
>  }
>  
> @@ -1382,28 +1689,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>  				    struct arm_smmu_master *master)
>  {
>  	int ret;
> -	unsigned long ias, oas;
> -	int (*finalise_stage_fn)(struct arm_smmu_domain *,
> -				 struct arm_smmu_master *,
> -				 struct io_pgtable_cfg *);
>  	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>  
>  	/* Restrict the stage to what we can actually support */
>  	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
>  
> -	switch (smmu_domain->stage) {
> -	case ARM_SMMU_DOMAIN_NESTED:
> -	case ARM_SMMU_DOMAIN_S2:
> -		ias = smmu->ias;
> -		oas = smmu->oas;
> -		finalise_stage_fn = arm_smmu_domain_finalise_s2;
> -		break;
> -	default:
> -		return -EINVAL;
> -	}
> -
> -	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
> +	ret = arm_smmu_domain_finalise_s2(smmu_domain, master);
>  	if (ret < 0) {
>  		return ret;
>  	}
> @@ -1553,7 +1844,8 @@ static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>  		return -ENOMEM;
>  	}
>  
> -	if (!WARN_ON(q->base_dma & (qsz - 1))) {
> +	WARN_ON(q->base_dma & (qsz - 1));
> +	if (unlikely(q->base_dma & (qsz - 1))) {
>  		dev_info(smmu->dev, "allocated %u entries for %s\n",
>  			 1 << q->llq.max_n_shift, name);

We don't need both the WARNING and the dev_info. you could turn the
dev_warn / XENLOG_WARNING.


>  	}
> @@ -1758,9 +2050,7 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>  	/* Request interrupt lines */
>  	irq = smmu->evtq.q.irq;
>  	if (irq) {
> -		ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
> -						arm_smmu_evtq_thread,
> -						IRQF_ONESHOT,
> +		ret = request_irq(irq, 0, arm_smmu_evtq_irq_tasklet,
>  						"arm-smmu-v3-evtq", smmu);
>  		if (ret < 0)
>  			dev_warn(smmu->dev, "failed to enable evtq irq\n");
> @@ -1770,8 +2060,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>  
>  	irq = smmu->gerr_irq;
>  	if (irq) {
> -		ret = devm_request_irq(smmu->dev, irq, arm_smmu_gerror_handler,
> -				       0, "arm-smmu-v3-gerror", smmu);
> +		ret = request_irq(irq, 0, arm_smmu_gerror_handler,
> +						"arm-smmu-v3-gerror", smmu);
>  		if (ret < 0)
>  			dev_warn(smmu->dev, "failed to enable gerror irq\n");
>  	} else {
> @@ -1781,11 +2071,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>  	if (smmu->features & ARM_SMMU_FEAT_PRI) {
>  		irq = smmu->priq.q.irq;
>  		if (irq) {
> -			ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
> -							arm_smmu_priq_thread,
> -							IRQF_ONESHOT,
> -							"arm-smmu-v3-priq",
> -							smmu);
> +			ret = request_irq(irq, 0, arm_smmu_priq_irq_tasklet,
> +							"arm-smmu-v3-priq", smmu);
>  			if (ret < 0)
>  				dev_warn(smmu->dev,
>  					 "failed to enable priq irq\n");
> @@ -1814,11 +2101,8 @@ static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
>  		 * Cavium ThunderX2 implementation doesn't support unique irq
>  		 * lines. Use a single irq line for all the SMMUv3 interrupts.
>  		 */
> -		ret = devm_request_threaded_irq(smmu->dev, irq,
> -					arm_smmu_combined_irq_handler,
> -					arm_smmu_combined_irq_thread,
> -					IRQF_ONESHOT,
> -					"arm-smmu-v3-combined-irq", smmu);
> +		ret = request_irq(irq, 0, arm_smmu_combined_irq_handler,
> +						"arm-smmu-v3-combined-irq", smmu);
>  		if (ret < 0)
>  			dev_warn(smmu->dev, "failed to enable combined irq\n");
>  	} else
> @@ -1857,7 +2141,7 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>  	reg = readl_relaxed(smmu->base + ARM_SMMU_CR0);
>  	if (reg & CR0_SMMUEN) {
>  		dev_warn(smmu->dev, "SMMU currently enabled! Resetting...\n");
> -		WARN_ON(is_kdump_kernel() && !disable_bypass);
> +		WARN_ON(!disable_bypass);
>  		arm_smmu_update_gbpa(smmu, GBPA_ABORT, 0);
>  	}
>  
> @@ -1952,8 +2236,11 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>  		return ret;
>  	}
>  
> -	if (is_kdump_kernel())
> -		enables &= ~(CR0_EVTQEN | CR0_PRIQEN);
> +	/* Initialize tasklets for threaded IRQs*/
> +	tasklet_init(&smmu->evtq_irq_tasklet, arm_smmu_evtq_thread, smmu);
> +	tasklet_init(&smmu->priq_irq_tasklet, arm_smmu_priq_thread, smmu);
> +	tasklet_init(&smmu->combined_irq_tasklet, arm_smmu_combined_irq_thread,
> +				 smmu);
>  
>  	/* Enable the SMMU interface, or ensure bypass */
>  	if (!bypass || disable_bypass) {
> @@ -2195,7 +2482,7 @@ static inline int arm_smmu_device_acpi_probe(struct platform_device *pdev,
>  static int arm_smmu_device_dt_probe(struct platform_device *pdev,
>  				    struct arm_smmu_device *smmu)
>  {
> -	struct device *dev = &pdev->dev;
> +	struct device *dev = pdev;
>  	u32 cells;
>  	int ret = -EINVAL;
>  
> @@ -2219,130 +2506,449 @@ static unsigned long arm_smmu_resource_size(struct arm_smmu_device *smmu)
>  		return SZ_128K;
>  }
>  
> +/* Start of Xen specific code. */
>  static int arm_smmu_device_probe(struct platform_device *pdev)
>  {
> -	int irq, ret;
> -	struct resource *res;
> -	resource_size_t ioaddr;
> -	struct arm_smmu_device *smmu;
> -	struct device *dev = &pdev->dev;
> -	bool bypass;
> +    int irq, ret;
> +    paddr_t ioaddr, iosize;
> +    struct arm_smmu_device *smmu;
> +    bool bypass;
> +
> +    smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
> +    if ( !smmu )
> +    {
> +        dev_err(pdev, "failed to allocate arm_smmu_device\n");
> +        return -ENOMEM;
> +    }
> +    smmu->dev = pdev;
> +
> +    if ( pdev->of_node )
> +    {
> +        ret = arm_smmu_device_dt_probe(pdev, smmu);
> +    } else
> +    {

coding style


> +        ret = arm_smmu_device_acpi_probe(pdev, smmu);
> +        if ( ret == -ENODEV )
> +            return ret;
> +    }
> +
> +    /* Set bypass mode according to firmware probing result */
> +    bypass = !!ret;
> +
> +    /* Base address */
> +    ret = dt_device_get_address(dev_to_dt(pdev), 0, &ioaddr, &iosize);
> +    if( ret )
> +        return -ENODEV;
> +
> +    if ( iosize < arm_smmu_resource_size(smmu) )
> +    {
> +        dev_err(pdev, "MMIO region too small (%lx)\n", iosize);
> +        return -EINVAL;
> +    }
> +
> +    /*
> +     * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
> +     * the PMCG registers which are reserved by the PMU driver.

Does this apply to Xen too?


> +     */
> +    smmu->base = ioremap_nocache(ioaddr, iosize);
> +    if ( IS_ERR(smmu->base) )
> +        return PTR_ERR(smmu->base);
> +
> +    if ( iosize > SZ_64K )
> +    {
> +        smmu->page1 = ioremap_nocache(ioaddr + SZ_64K, ARM_SMMU_REG_SZ);
> +        if ( IS_ERR(smmu->page1) )
> +            return PTR_ERR(smmu->page1);
> +    }
> +    else
> +    {
> +        smmu->page1 = smmu->base;
> +    }
> +
> +    /* Interrupt lines */
> +
> +    irq = platform_get_irq_byname_optional(pdev, "combined");
> +    if ( irq > 0 )
> +        smmu->combined_irq = irq;
> +    else
> +    {
> +        irq = platform_get_irq_byname_optional(pdev, "eventq");
> +        if ( irq > 0 )
> +            smmu->evtq.q.irq = irq;
> +
> +        irq = platform_get_irq_byname_optional(pdev, "priq");
> +        if ( irq > 0 )
> +            smmu->priq.q.irq = irq;
> +
> +        irq = platform_get_irq_byname_optional(pdev, "gerror");
> +        if ( irq > 0 )
> +            smmu->gerr_irq = irq;
> +    }
> +    /* Probe the h/w */
> +    ret = arm_smmu_device_hw_probe(smmu);
> +    if ( ret )
> +        return ret;
> +
> +    /* Initialise in-memory data structures */
> +    ret = arm_smmu_init_structures(smmu);
> +    if ( ret )
> +        return ret;
> +
> +    /* Reset the device */
> +    ret = arm_smmu_device_reset(smmu, bypass);
> +    if ( ret )
> +        return ret;
> +
> +    /*
> +     * Keep a list of all probed devices. This will be used to query
> +     * the smmu devices based on the fwnode.
> +     */
> +    INIT_LIST_HEAD(&smmu->devices);
> +
> +    spin_lock(&arm_smmu_devices_lock);
> +    list_add(&smmu->devices, &arm_smmu_devices);
> +    spin_unlock(&arm_smmu_devices_lock);
> +
> +    return 0;
> +}


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 2/8] xen/arm: revert atomic operation related command-queue insertion patch
  2020-12-01 22:23   ` Stefano Stabellini
@ 2020-12-02 13:05     ` Rahul Singh
  0 siblings, 0 replies; 53+ messages in thread
From: Rahul Singh @ 2020-12-02 13:05 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: xen-devel, Bertrand Marquis, Julien Grall, Volodymyr Babchuk

Hello Stefano,

Thanks for reviewing the code.

> On 1 Dec 2020, at 10:23 pm, Stefano Stabellini <sstabellini@kernel.org> wrote:
> 
> On Thu, 26 Nov 2020, Rahul Singh wrote:
>> Linux SMMUv3 code implements the commands-queue insertion based on
>> atomic operations implemented in Linux. Atomic functions used by the
>> commands-queue insertion is not implemented in XEN therefore revert the
>> patch that implemented the commands-queue insertion based on atomic
>> operations.
>> 
>> Once the proper atomic operations will be available in XEN the driver
>> can be updated.
>> 
>> Reverted the commit 587e6c10a7ce89a5924fdbeff2ec524fbd6a124b
>> iommu/arm-smmu-v3: Reduce contention during command-queue insertion
> 
> I checked 587e6c10a7ce89a5924fdbeff2ec524fbd6a124b: this patch does more
> than just reverting 587e6c10a7ce89a5924fdbeff2ec524fbd6a124b. It looks
> like it is also reverting edd0351e7bc49555d8b5ad8438a65a7ca262c9f0 and
> some other commits.
> 
> Please can you provide a complete list of reverted commits? I would like
> to be able to do the reverts myself on the linux tree and see that the
> driver textually matches the one on the xen tree with this patch
> applied.
> 
> 

Yes this patch is also reverting the commits that is based on the code that introduced the atomic-operations. I will add all the commit id in next version of the patch in commit msg. 

Patches that are reverted in this patch are as follows:

9e773aee8c3e1b3ba019c5c7f8435aaa836c6130  iommu/arm-smmu-v3: Batch ATC invalidation commands
edd0351e7bc49555d8b5ad8438a65a7ca262c9f0  iommu/arm-smmu-v3: Batch context descriptor invalidation
4ce8da453640147101bda418640394637c1a7cfc  iommu/arm-smmu-v3: Add command queue batching helpers
2af2e72b18b499fa36d3f7379fd010ff25d2a984.      iommu/arm-smmu-v3: Defer TLB invalidation until ->iotlb_sync() 
587e6c10a7ce89a5924fdbeff2ec524fbd6a124b      iommu/arm-smmu-v3: Reduce contention during command-queue insertion


Regards,
Rahul

> 
>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
>> ---
>> xen/drivers/passthrough/arm/smmu-v3.c | 847 ++++++--------------------
>> 1 file changed, 180 insertions(+), 667 deletions(-)
>> 
>> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
>> index c192544e87..97eac61ea4 100644
>> --- a/xen/drivers/passthrough/arm/smmu-v3.c
>> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
>> @@ -330,15 +330,6 @@
>> #define CMDQ_ERR_CERROR_ABT_IDX		2
>> #define CMDQ_ERR_CERROR_ATC_INV_IDX	3
>> 
>> -#define CMDQ_PROD_OWNED_FLAG		Q_OVERFLOW_FLAG
>> -
>> -/*
>> - * This is used to size the command queue and therefore must be at least
>> - * BITS_PER_LONG so that the valid_map works correctly (it relies on the
>> - * total number of queue entries being a multiple of BITS_PER_LONG).
>> - */
>> -#define CMDQ_BATCH_ENTRIES		BITS_PER_LONG
>> -
>> #define CMDQ_0_OP			GENMASK_ULL(7, 0)
>> #define CMDQ_0_SSV			(1UL << 11)
>> 
>> @@ -407,8 +398,9 @@
>> #define PRIQ_1_ADDR_MASK		GENMASK_ULL(63, 12)
>> 
>> /* High-level queue structures */
>> -#define ARM_SMMU_POLL_TIMEOUT_US	1000000 /* 1s! */
>> -#define ARM_SMMU_POLL_SPIN_COUNT	10
>> +#define ARM_SMMU_POLL_TIMEOUT_US	100
>> +#define ARM_SMMU_CMDQ_SYNC_TIMEOUT_US	1000000 /* 1s! */
>> +#define ARM_SMMU_CMDQ_SYNC_SPIN_COUNT	10
>> 
>> #define MSI_IOVA_BASE			0x8000000
>> #define MSI_IOVA_LENGTH			0x100000
>> @@ -513,24 +505,15 @@ struct arm_smmu_cmdq_ent {
>> 
>> 		#define CMDQ_OP_CMD_SYNC	0x46
>> 		struct {
>> +			u32			msidata;
>> 			u64			msiaddr;
>> 		} sync;
>> 	};
>> };
>> 
>> struct arm_smmu_ll_queue {
>> -	union {
>> -		u64			val;
>> -		struct {
>> -			u32		prod;
>> -			u32		cons;
>> -		};
>> -		struct {
>> -			atomic_t	prod;
>> -			atomic_t	cons;
>> -		} atomic;
>> -		u8			__pad[SMP_CACHE_BYTES];
>> -	} ____cacheline_aligned_in_smp;
>> +	u32				prod;
>> +	u32				cons;
>> 	u32				max_n_shift;
>> };
>> 
>> @@ -548,23 +531,9 @@ struct arm_smmu_queue {
>> 	u32 __iomem			*cons_reg;
>> };
>> 
>> -struct arm_smmu_queue_poll {
>> -	ktime_t				timeout;
>> -	unsigned int			delay;
>> -	unsigned int			spin_cnt;
>> -	bool				wfe;
>> -};
>> -
>> struct arm_smmu_cmdq {
>> 	struct arm_smmu_queue		q;
>> -	atomic_long_t			*valid_map;
>> -	atomic_t			owner_prod;
>> -	atomic_t			lock;
>> -};
>> -
>> -struct arm_smmu_cmdq_batch {
>> -	u64				cmds[CMDQ_BATCH_ENTRIES * CMDQ_ENT_DWORDS];
>> -	int				num;
>> +	spinlock_t			lock;
>> };
>> 
>> struct arm_smmu_evtq {
>> @@ -660,6 +629,8 @@ struct arm_smmu_device {
>> 
>> 	int				gerr_irq;
>> 	int				combined_irq;
>> +	u32				sync_nr;
>> +	u8				prev_cmd_opcode;
>> 
>> 	unsigned long			ias; /* IPA */
>> 	unsigned long			oas; /* PA */
>> @@ -677,6 +648,12 @@ struct arm_smmu_device {
>> 
>> 	struct arm_smmu_strtab_cfg	strtab_cfg;
>> 
>> +	/* Hi16xx adds an extra 32 bits of goodness to its MSI payload */
>> +	union {
>> +		u32			sync_count;
>> +		u64			padding;
>> +	};
>> +
>> 	/* IOMMU core code handle */
>> 	struct iommu_device		iommu;
>> };
>> @@ -763,21 +740,6 @@ static void parse_driver_options(struct arm_smmu_device *smmu)
>> }
>> 
>> /* Low-level queue manipulation functions */
>> -static bool queue_has_space(struct arm_smmu_ll_queue *q, u32 n)
>> -{
>> -	u32 space, prod, cons;
>> -
>> -	prod = Q_IDX(q, q->prod);
>> -	cons = Q_IDX(q, q->cons);
>> -
>> -	if (Q_WRP(q, q->prod) == Q_WRP(q, q->cons))
>> -		space = (1 << q->max_n_shift) - (prod - cons);
>> -	else
>> -		space = cons - prod;
>> -
>> -	return space >= n;
>> -}
>> -
>> static bool queue_full(struct arm_smmu_ll_queue *q)
>> {
>> 	return Q_IDX(q, q->prod) == Q_IDX(q, q->cons) &&
>> @@ -790,12 +752,9 @@ static bool queue_empty(struct arm_smmu_ll_queue *q)
>> 	       Q_WRP(q, q->prod) == Q_WRP(q, q->cons);
>> }
>> 
>> -static bool queue_consumed(struct arm_smmu_ll_queue *q, u32 prod)
>> +static void queue_sync_cons_in(struct arm_smmu_queue *q)
>> {
>> -	return ((Q_WRP(q, q->cons) == Q_WRP(q, prod)) &&
>> -		(Q_IDX(q, q->cons) > Q_IDX(q, prod))) ||
>> -	       ((Q_WRP(q, q->cons) != Q_WRP(q, prod)) &&
>> -		(Q_IDX(q, q->cons) <= Q_IDX(q, prod)));
>> +	q->llq.cons = readl_relaxed(q->cons_reg);
>> }
>> 
>> static void queue_sync_cons_out(struct arm_smmu_queue *q)
>> @@ -826,34 +785,46 @@ static int queue_sync_prod_in(struct arm_smmu_queue *q)
>> 	return ret;
>> }
>> 
>> -static u32 queue_inc_prod_n(struct arm_smmu_ll_queue *q, int n)
>> +static void queue_sync_prod_out(struct arm_smmu_queue *q)
>> {
>> -	u32 prod = (Q_WRP(q, q->prod) | Q_IDX(q, q->prod)) + n;
>> -	return Q_OVF(q->prod) | Q_WRP(q, prod) | Q_IDX(q, prod);
>> +	writel(q->llq.prod, q->prod_reg);
>> }
>> 
>> -static void queue_poll_init(struct arm_smmu_device *smmu,
>> -			    struct arm_smmu_queue_poll *qp)
>> +static void queue_inc_prod(struct arm_smmu_ll_queue *q)
>> {
>> -	qp->delay = 1;
>> -	qp->spin_cnt = 0;
>> -	qp->wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
>> -	qp->timeout = ktime_add_us(ktime_get(), ARM_SMMU_POLL_TIMEOUT_US);
>> +	u32 prod = (Q_WRP(q, q->prod) | Q_IDX(q, q->prod)) + 1;
>> +	q->prod = Q_OVF(q->prod) | Q_WRP(q, prod) | Q_IDX(q, prod);
>> }
>> 
>> -static int queue_poll(struct arm_smmu_queue_poll *qp)
>> +/*
>> + * Wait for the SMMU to consume items. If sync is true, wait until the queue
>> + * is empty. Otherwise, wait until there is at least one free slot.
>> + */
>> +static int queue_poll_cons(struct arm_smmu_queue *q, bool sync, bool wfe)
>> {
>> -	if (ktime_compare(ktime_get(), qp->timeout) > 0)
>> -		return -ETIMEDOUT;
>> +	ktime_t timeout;
>> +	unsigned int delay = 1, spin_cnt = 0;
>> 
>> -	if (qp->wfe) {
>> -		wfe();
>> -	} else if (++qp->spin_cnt < ARM_SMMU_POLL_SPIN_COUNT) {
>> -		cpu_relax();
>> -	} else {
>> -		udelay(qp->delay);
>> -		qp->delay *= 2;
>> -		qp->spin_cnt = 0;
>> +	/* Wait longer if it's a CMD_SYNC */
>> +	timeout = ktime_add_us(ktime_get(), sync ?
>> +					    ARM_SMMU_CMDQ_SYNC_TIMEOUT_US :
>> +					    ARM_SMMU_POLL_TIMEOUT_US);
>> +
>> +	while (queue_sync_cons_in(q),
>> +	      (sync ? !queue_empty(&q->llq) : queue_full(&q->llq))) {
>> +		if (ktime_compare(ktime_get(), timeout) > 0)
>> +			return -ETIMEDOUT;
>> +
>> +		if (wfe) {
>> +			wfe();
>> +		} else if (++spin_cnt < ARM_SMMU_CMDQ_SYNC_SPIN_COUNT) {
>> +			cpu_relax();
>> +			continue;
>> +		} else {
>> +			udelay(delay);
>> +			delay *= 2;
>> +			spin_cnt = 0;
>> +		}
>> 	}
>> 
>> 	return 0;
>> @@ -867,6 +838,17 @@ static void queue_write(__le64 *dst, u64 *src, size_t n_dwords)
>> 		*dst++ = cpu_to_le64(*src++);
>> }
>> 
>> +static int queue_insert_raw(struct arm_smmu_queue *q, u64 *ent)
>> +{
>> +	if (queue_full(&q->llq))
>> +		return -ENOSPC;
>> +
>> +	queue_write(Q_ENT(q, q->llq.prod), ent, q->ent_dwords);
>> +	queue_inc_prod(&q->llq);
>> +	queue_sync_prod_out(q);
>> +	return 0;
>> +}
>> +
>> static void queue_read(__le64 *dst, u64 *src, size_t n_dwords)
>> {
>> 	int i;
>> @@ -964,14 +946,20 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
>> 		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_RESP, ent->pri.resp);
>> 		break;
>> 	case CMDQ_OP_CMD_SYNC:
>> -		if (ent->sync.msiaddr) {
>> +		if (ent->sync.msiaddr)
>> 			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_IRQ);
>> -			cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
>> -		} else {
>> +		else
>> 			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
>> -		}
>> 		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSH, ARM_SMMU_SH_ISH);
>> 		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIATTR, ARM_SMMU_MEMATTR_OIWB);
>> +		/*
>> +		 * Commands are written little-endian, but we want the SMMU to
>> +		 * receive MSIData, and thus write it back to memory, in CPU
>> +		 * byte order, so big-endian needs an extra byteswap here.
>> +		 */
>> +		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIDATA,
>> +				     cpu_to_le32(ent->sync.msidata));
>> +		cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
>> 		break;
>> 	default:
>> 		return -ENOENT;
>> @@ -980,27 +968,6 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
>> 	return 0;
>> }
>> 
>> -static void arm_smmu_cmdq_build_sync_cmd(u64 *cmd, struct arm_smmu_device *smmu,
>> -					 u32 prod)
>> -{
>> -	struct arm_smmu_queue *q = &smmu->cmdq.q;
>> -	struct arm_smmu_cmdq_ent ent = {
>> -		.opcode = CMDQ_OP_CMD_SYNC,
>> -	};
>> -
>> -	/*
>> -	 * Beware that Hi16xx adds an extra 32 bits of goodness to its MSI
>> -	 * payload, so the write will zero the entire command on that platform.
>> -	 */
>> -	if (smmu->features & ARM_SMMU_FEAT_MSI &&
>> -	    smmu->features & ARM_SMMU_FEAT_COHERENCY) {
>> -		ent.sync.msiaddr = q->base_dma + Q_IDX(&q->llq, prod) *
>> -				   q->ent_dwords * 8;
>> -	}
>> -
>> -	arm_smmu_cmdq_build_cmd(cmd, &ent);
>> -}
>> -
>> static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
>> {
>> 	static const char *cerror_str[] = {
>> @@ -1058,474 +1025,109 @@ static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
>> 	queue_write(Q_ENT(q, cons), cmd, q->ent_dwords);
>> }
>> 
>> -/*
>> - * Command queue locking.
>> - * This is a form of bastardised rwlock with the following major changes:
>> - *
>> - * - The only LOCK routines are exclusive_trylock() and shared_lock().
>> - *   Neither have barrier semantics, and instead provide only a control
>> - *   dependency.
>> - *
>> - * - The UNLOCK routines are supplemented with shared_tryunlock(), which
>> - *   fails if the caller appears to be the last lock holder (yes, this is
>> - *   racy). All successful UNLOCK routines have RELEASE semantics.
>> - */
>> -static void arm_smmu_cmdq_shared_lock(struct arm_smmu_cmdq *cmdq)
>> +static void arm_smmu_cmdq_insert_cmd(struct arm_smmu_device *smmu, u64 *cmd)
>> {
>> -	int val;
>> -
>> -	/*
>> -	 * We can try to avoid the cmpxchg() loop by simply incrementing the
>> -	 * lock counter. When held in exclusive state, the lock counter is set
>> -	 * to INT_MIN so these increments won't hurt as the value will remain
>> -	 * negative.
>> -	 */
>> -	if (atomic_fetch_inc_relaxed(&cmdq->lock) >= 0)
>> -		return;
>> -
>> -	do {
>> -		val = atomic_cond_read_relaxed(&cmdq->lock, VAL >= 0);
>> -	} while (atomic_cmpxchg_relaxed(&cmdq->lock, val, val + 1) != val);
>> -}
>> -
>> -static void arm_smmu_cmdq_shared_unlock(struct arm_smmu_cmdq *cmdq)
>> -{
>> -	(void)atomic_dec_return_release(&cmdq->lock);
>> -}
>> -
>> -static bool arm_smmu_cmdq_shared_tryunlock(struct arm_smmu_cmdq *cmdq)
>> -{
>> -	if (atomic_read(&cmdq->lock) == 1)
>> -		return false;
>> -
>> -	arm_smmu_cmdq_shared_unlock(cmdq);
>> -	return true;
>> -}
>> -
>> -#define arm_smmu_cmdq_exclusive_trylock_irqsave(cmdq, flags)		\
>> -({									\
>> -	bool __ret;							\
>> -	local_irq_save(flags);						\
>> -	__ret = !atomic_cmpxchg_relaxed(&cmdq->lock, 0, INT_MIN);	\
>> -	if (!__ret)							\
>> -		local_irq_restore(flags);				\
>> -	__ret;								\
>> -})
>> -
>> -#define arm_smmu_cmdq_exclusive_unlock_irqrestore(cmdq, flags)		\
>> -({									\
>> -	atomic_set_release(&cmdq->lock, 0);				\
>> -	local_irq_restore(flags);					\
>> -})
>> -
>> -
>> -/*
>> - * Command queue insertion.
>> - * This is made fiddly by our attempts to achieve some sort of scalability
>> - * since there is one queue shared amongst all of the CPUs in the system.  If
>> - * you like mixed-size concurrency, dependency ordering and relaxed atomics,
>> - * then you'll *love* this monstrosity.
>> - *
>> - * The basic idea is to split the queue up into ranges of commands that are
>> - * owned by a given CPU; the owner may not have written all of the commands
>> - * itself, but is responsible for advancing the hardware prod pointer when
>> - * the time comes. The algorithm is roughly:
>> - *
>> - * 	1. Allocate some space in the queue. At this point we also discover
>> - *	   whether the head of the queue is currently owned by another CPU,
>> - *	   or whether we are the owner.
>> - *
>> - *	2. Write our commands into our allocated slots in the queue.
>> - *
>> - *	3. Mark our slots as valid in arm_smmu_cmdq.valid_map.
>> - *
>> - *	4. If we are an owner:
>> - *		a. Wait for the previous owner to finish.
>> - *		b. Mark the queue head as unowned, which tells us the range
>> - *		   that we are responsible for publishing.
>> - *		c. Wait for all commands in our owned range to become valid.
>> - *		d. Advance the hardware prod pointer.
>> - *		e. Tell the next owner we've finished.
>> - *
>> - *	5. If we are inserting a CMD_SYNC (we may or may not have been an
>> - *	   owner), then we need to stick around until it has completed:
>> - *		a. If we have MSIs, the SMMU can write back into the CMD_SYNC
>> - *		   to clear the first 4 bytes.
>> - *		b. Otherwise, we spin waiting for the hardware cons pointer to
>> - *		   advance past our command.
>> - *
>> - * The devil is in the details, particularly the use of locking for handling
>> - * SYNC completion and freeing up space in the queue before we think that it is
>> - * full.
>> - */
>> -static void __arm_smmu_cmdq_poll_set_valid_map(struct arm_smmu_cmdq *cmdq,
>> -					       u32 sprod, u32 eprod, bool set)
>> -{
>> -	u32 swidx, sbidx, ewidx, ebidx;
>> -	struct arm_smmu_ll_queue llq = {
>> -		.max_n_shift	= cmdq->q.llq.max_n_shift,
>> -		.prod		= sprod,
>> -	};
>> -
>> -	ewidx = BIT_WORD(Q_IDX(&llq, eprod));
>> -	ebidx = Q_IDX(&llq, eprod) % BITS_PER_LONG;
>> -
>> -	while (llq.prod != eprod) {
>> -		unsigned long mask;
>> -		atomic_long_t *ptr;
>> -		u32 limit = BITS_PER_LONG;
>> -
>> -		swidx = BIT_WORD(Q_IDX(&llq, llq.prod));
>> -		sbidx = Q_IDX(&llq, llq.prod) % BITS_PER_LONG;
>> -
>> -		ptr = &cmdq->valid_map[swidx];
>> -
>> -		if ((swidx == ewidx) && (sbidx < ebidx))
>> -			limit = ebidx;
>> -
>> -		mask = GENMASK(limit - 1, sbidx);
>> -
>> -		/*
>> -		 * The valid bit is the inverse of the wrap bit. This means
>> -		 * that a zero-initialised queue is invalid and, after marking
>> -		 * all entries as valid, they become invalid again when we
>> -		 * wrap.
>> -		 */
>> -		if (set) {
>> -			atomic_long_xor(mask, ptr);
>> -		} else { /* Poll */
>> -			unsigned long valid;
>> +	struct arm_smmu_queue *q = &smmu->cmdq.q;
>> +	bool wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
>> 
>> -			valid = (ULONG_MAX + !!Q_WRP(&llq, llq.prod)) & mask;
>> -			atomic_long_cond_read_relaxed(ptr, (VAL & mask) == valid);
>> -		}
>> +	smmu->prev_cmd_opcode = FIELD_GET(CMDQ_0_OP, cmd[0]);
>> 
>> -		llq.prod = queue_inc_prod_n(&llq, limit - sbidx);
>> +	while (queue_insert_raw(q, cmd) == -ENOSPC) {
>> +		if (queue_poll_cons(q, false, wfe))
>> +			dev_err_ratelimited(smmu->dev, "CMDQ timeout\n");
>> 	}
>> }
>> 
>> -/* Mark all entries in the range [sprod, eprod) as valid */
>> -static void arm_smmu_cmdq_set_valid_map(struct arm_smmu_cmdq *cmdq,
>> -					u32 sprod, u32 eprod)
>> -{
>> -	__arm_smmu_cmdq_poll_set_valid_map(cmdq, sprod, eprod, true);
>> -}
>> -
>> -/* Wait for all entries in the range [sprod, eprod) to become valid */
>> -static void arm_smmu_cmdq_poll_valid_map(struct arm_smmu_cmdq *cmdq,
>> -					 u32 sprod, u32 eprod)
>> -{
>> -	__arm_smmu_cmdq_poll_set_valid_map(cmdq, sprod, eprod, false);
>> -}
>> -
>> -/* Wait for the command queue to become non-full */
>> -static int arm_smmu_cmdq_poll_until_not_full(struct arm_smmu_device *smmu,
>> -					     struct arm_smmu_ll_queue *llq)
>> +static void arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
>> +				    struct arm_smmu_cmdq_ent *ent)
>> {
>> +	u64 cmd[CMDQ_ENT_DWORDS];
>> 	unsigned long flags;
>> -	struct arm_smmu_queue_poll qp;
>> -	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
>> -	int ret = 0;
>> 
>> -	/*
>> -	 * Try to update our copy of cons by grabbing exclusive cmdq access. If
>> -	 * that fails, spin until somebody else updates it for us.
>> -	 */
>> -	if (arm_smmu_cmdq_exclusive_trylock_irqsave(cmdq, flags)) {
>> -		WRITE_ONCE(cmdq->q.llq.cons, readl_relaxed(cmdq->q.cons_reg));
>> -		arm_smmu_cmdq_exclusive_unlock_irqrestore(cmdq, flags);
>> -		llq->val = READ_ONCE(cmdq->q.llq.val);
>> -		return 0;
>> +	if (arm_smmu_cmdq_build_cmd(cmd, ent)) {
>> +		dev_warn(smmu->dev, "ignoring unknown CMDQ opcode 0x%x\n",
>> +			 ent->opcode);
>> +		return;
>> 	}
>> 
>> -	queue_poll_init(smmu, &qp);
>> -	do {
>> -		llq->val = READ_ONCE(smmu->cmdq.q.llq.val);
>> -		if (!queue_full(llq))
>> -			break;
>> -
>> -		ret = queue_poll(&qp);
>> -	} while (!ret);
>> -
>> -	return ret;
>> -}
>> -
>> -/*
>> - * Wait until the SMMU signals a CMD_SYNC completion MSI.
>> - * Must be called with the cmdq lock held in some capacity.
>> - */
>> -static int __arm_smmu_cmdq_poll_until_msi(struct arm_smmu_device *smmu,
>> -					  struct arm_smmu_ll_queue *llq)
>> -{
>> -	int ret = 0;
>> -	struct arm_smmu_queue_poll qp;
>> -	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
>> -	u32 *cmd = (u32 *)(Q_ENT(&cmdq->q, llq->prod));
>> -
>> -	queue_poll_init(smmu, &qp);
>> -
>> -	/*
>> -	 * The MSI won't generate an event, since it's being written back
>> -	 * into the command queue.
>> -	 */
>> -	qp.wfe = false;
>> -	smp_cond_load_relaxed(cmd, !VAL || (ret = queue_poll(&qp)));
>> -	llq->cons = ret ? llq->prod : queue_inc_prod_n(llq, 1);
>> -	return ret;
>> +	spin_lock_irqsave(&smmu->cmdq.lock, flags);
>> +	arm_smmu_cmdq_insert_cmd(smmu, cmd);
>> +	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
>> }
>> 
>> /*
>> - * Wait until the SMMU cons index passes llq->prod.
>> - * Must be called with the cmdq lock held in some capacity.
>> + * The difference between val and sync_idx is bounded by the maximum size of
>> + * a queue at 2^20 entries, so 32 bits is plenty for wrap-safe arithmetic.
>>  */
>> -static int __arm_smmu_cmdq_poll_until_consumed(struct arm_smmu_device *smmu,
>> -					       struct arm_smmu_ll_queue *llq)
>> +static int __arm_smmu_sync_poll_msi(struct arm_smmu_device *smmu, u32 sync_idx)
>> {
>> -	struct arm_smmu_queue_poll qp;
>> -	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
>> -	u32 prod = llq->prod;
>> -	int ret = 0;
>> +	ktime_t timeout;
>> +	u32 val;
>> 
>> -	queue_poll_init(smmu, &qp);
>> -	llq->val = READ_ONCE(smmu->cmdq.q.llq.val);
>> -	do {
>> -		if (queue_consumed(llq, prod))
>> -			break;
>> -
>> -		ret = queue_poll(&qp);
>> -
>> -		/*
>> -		 * This needs to be a readl() so that our subsequent call
>> -		 * to arm_smmu_cmdq_shared_tryunlock() can fail accurately.
>> -		 *
>> -		 * Specifically, we need to ensure that we observe all
>> -		 * shared_lock()s by other CMD_SYNCs that share our owner,
>> -		 * so that a failing call to tryunlock() means that we're
>> -		 * the last one out and therefore we can safely advance
>> -		 * cmdq->q.llq.cons. Roughly speaking:
>> -		 *
>> -		 * CPU 0		CPU1			CPU2 (us)
>> -		 *
>> -		 * if (sync)
>> -		 * 	shared_lock();
>> -		 *
>> -		 * dma_wmb();
>> -		 * set_valid_map();
>> -		 *
>> -		 * 			if (owner) {
>> -		 *				poll_valid_map();
>> -		 *				<control dependency>
>> -		 *				writel(prod_reg);
>> -		 *
>> -		 *						readl(cons_reg);
>> -		 *						tryunlock();
>> -		 *
>> -		 * Requires us to see CPU 0's shared_lock() acquisition.
>> -		 */
>> -		llq->cons = readl(cmdq->q.cons_reg);
>> -	} while (!ret);
>> +	timeout = ktime_add_us(ktime_get(), ARM_SMMU_CMDQ_SYNC_TIMEOUT_US);
>> +	val = smp_cond_load_acquire(&smmu->sync_count,
>> +				    (int)(VAL - sync_idx) >= 0 ||
>> +				    !ktime_before(ktime_get(), timeout));
>> 
>> -	return ret;
>> +	return (int)(val - sync_idx) < 0 ? -ETIMEDOUT : 0;
>> }
>> 
>> -static int arm_smmu_cmdq_poll_until_sync(struct arm_smmu_device *smmu,
>> -					 struct arm_smmu_ll_queue *llq)
>> +static int __arm_smmu_cmdq_issue_sync_msi(struct arm_smmu_device *smmu)
>> {
>> -	if (smmu->features & ARM_SMMU_FEAT_MSI &&
>> -	    smmu->features & ARM_SMMU_FEAT_COHERENCY)
>> -		return __arm_smmu_cmdq_poll_until_msi(smmu, llq);
>> -
>> -	return __arm_smmu_cmdq_poll_until_consumed(smmu, llq);
>> -}
>> -
>> -static void arm_smmu_cmdq_write_entries(struct arm_smmu_cmdq *cmdq, u64 *cmds,
>> -					u32 prod, int n)
>> -{
>> -	int i;
>> -	struct arm_smmu_ll_queue llq = {
>> -		.max_n_shift	= cmdq->q.llq.max_n_shift,
>> -		.prod		= prod,
>> -	};
>> -
>> -	for (i = 0; i < n; ++i) {
>> -		u64 *cmd = &cmds[i * CMDQ_ENT_DWORDS];
>> -
>> -		prod = queue_inc_prod_n(&llq, i);
>> -		queue_write(Q_ENT(&cmdq->q, prod), cmd, CMDQ_ENT_DWORDS);
>> -	}
>> -}
>> -
>> -/*
>> - * This is the actual insertion function, and provides the following
>> - * ordering guarantees to callers:
>> - *
>> - * - There is a dma_wmb() before publishing any commands to the queue.
>> - *   This can be relied upon to order prior writes to data structures
>> - *   in memory (such as a CD or an STE) before the command.
>> - *
>> - * - On completion of a CMD_SYNC, there is a control dependency.
>> - *   This can be relied upon to order subsequent writes to memory (e.g.
>> - *   freeing an IOVA) after completion of the CMD_SYNC.
>> - *
>> - * - Command insertion is totally ordered, so if two CPUs each race to
>> - *   insert their own list of commands then all of the commands from one
>> - *   CPU will appear before any of the commands from the other CPU.
>> - */
>> -static int arm_smmu_cmdq_issue_cmdlist(struct arm_smmu_device *smmu,
>> -				       u64 *cmds, int n, bool sync)
>> -{
>> -	u64 cmd_sync[CMDQ_ENT_DWORDS];
>> -	u32 prod;
>> +	u64 cmd[CMDQ_ENT_DWORDS];
>> 	unsigned long flags;
>> -	bool owner;
>> -	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
>> -	struct arm_smmu_ll_queue llq = {
>> -		.max_n_shift = cmdq->q.llq.max_n_shift,
>> -	}, head = llq;
>> -	int ret = 0;
>> -
>> -	/* 1. Allocate some space in the queue */
>> -	local_irq_save(flags);
>> -	llq.val = READ_ONCE(cmdq->q.llq.val);
>> -	do {
>> -		u64 old;
>> -
>> -		while (!queue_has_space(&llq, n + sync)) {
>> -			local_irq_restore(flags);
>> -			if (arm_smmu_cmdq_poll_until_not_full(smmu, &llq))
>> -				dev_err_ratelimited(smmu->dev, "CMDQ timeout\n");
>> -			local_irq_save(flags);
>> -		}
>> -
>> -		head.cons = llq.cons;
>> -		head.prod = queue_inc_prod_n(&llq, n + sync) |
>> -					     CMDQ_PROD_OWNED_FLAG;
>> -
>> -		old = cmpxchg_relaxed(&cmdq->q.llq.val, llq.val, head.val);
>> -		if (old == llq.val)
>> -			break;
>> -
>> -		llq.val = old;
>> -	} while (1);
>> -	owner = !(llq.prod & CMDQ_PROD_OWNED_FLAG);
>> -	head.prod &= ~CMDQ_PROD_OWNED_FLAG;
>> -	llq.prod &= ~CMDQ_PROD_OWNED_FLAG;
>> -
>> -	/*
>> -	 * 2. Write our commands into the queue
>> -	 * Dependency ordering from the cmpxchg() loop above.
>> -	 */
>> -	arm_smmu_cmdq_write_entries(cmdq, cmds, llq.prod, n);
>> -	if (sync) {
>> -		prod = queue_inc_prod_n(&llq, n);
>> -		arm_smmu_cmdq_build_sync_cmd(cmd_sync, smmu, prod);
>> -		queue_write(Q_ENT(&cmdq->q, prod), cmd_sync, CMDQ_ENT_DWORDS);
>> -
>> -		/*
>> -		 * In order to determine completion of our CMD_SYNC, we must
>> -		 * ensure that the queue can't wrap twice without us noticing.
>> -		 * We achieve that by taking the cmdq lock as shared before
>> -		 * marking our slot as valid.
>> -		 */
>> -		arm_smmu_cmdq_shared_lock(cmdq);
>> -	}
>> -
>> -	/* 3. Mark our slots as valid, ensuring commands are visible first */
>> -	dma_wmb();
>> -	arm_smmu_cmdq_set_valid_map(cmdq, llq.prod, head.prod);
>> -
>> -	/* 4. If we are the owner, take control of the SMMU hardware */
>> -	if (owner) {
>> -		/* a. Wait for previous owner to finish */
>> -		atomic_cond_read_relaxed(&cmdq->owner_prod, VAL == llq.prod);
>> -
>> -		/* b. Stop gathering work by clearing the owned flag */
>> -		prod = atomic_fetch_andnot_relaxed(CMDQ_PROD_OWNED_FLAG,
>> -						   &cmdq->q.llq.atomic.prod);
>> -		prod &= ~CMDQ_PROD_OWNED_FLAG;
>> +	struct arm_smmu_cmdq_ent  ent = {
>> +		.opcode = CMDQ_OP_CMD_SYNC,
>> +		.sync	= {
>> +			.msiaddr = virt_to_phys(&smmu->sync_count),
>> +		},
>> +	};
>> 
>> -		/*
>> -		 * c. Wait for any gathered work to be written to the queue.
>> -		 * Note that we read our own entries so that we have the control
>> -		 * dependency required by (d).
>> -		 */
>> -		arm_smmu_cmdq_poll_valid_map(cmdq, llq.prod, prod);
>> +	spin_lock_irqsave(&smmu->cmdq.lock, flags);
>> 
>> -		/*
>> -		 * d. Advance the hardware prod pointer
>> -		 * Control dependency ordering from the entries becoming valid.
>> -		 */
>> -		writel_relaxed(prod, cmdq->q.prod_reg);
>> -
>> -		/*
>> -		 * e. Tell the next owner we're done
>> -		 * Make sure we've updated the hardware first, so that we don't
>> -		 * race to update prod and potentially move it backwards.
>> -		 */
>> -		atomic_set_release(&cmdq->owner_prod, prod);
>> +	/* Piggy-back on the previous command if it's a SYNC */
>> +	if (smmu->prev_cmd_opcode == CMDQ_OP_CMD_SYNC) {
>> +		ent.sync.msidata = smmu->sync_nr;
>> +	} else {
>> +		ent.sync.msidata = ++smmu->sync_nr;
>> +		arm_smmu_cmdq_build_cmd(cmd, &ent);
>> +		arm_smmu_cmdq_insert_cmd(smmu, cmd);
>> 	}
>> 
>> -	/* 5. If we are inserting a CMD_SYNC, we must wait for it to complete */
>> -	if (sync) {
>> -		llq.prod = queue_inc_prod_n(&llq, n);
>> -		ret = arm_smmu_cmdq_poll_until_sync(smmu, &llq);
>> -		if (ret) {
>> -			dev_err_ratelimited(smmu->dev,
>> -					    "CMD_SYNC timeout at 0x%08x [hwprod 0x%08x, hwcons 0x%08x]\n",
>> -					    llq.prod,
>> -					    readl_relaxed(cmdq->q.prod_reg),
>> -					    readl_relaxed(cmdq->q.cons_reg));
>> -		}
>> -
>> -		/*
>> -		 * Try to unlock the cmdq lock. This will fail if we're the last
>> -		 * reader, in which case we can safely update cmdq->q.llq.cons
>> -		 */
>> -		if (!arm_smmu_cmdq_shared_tryunlock(cmdq)) {
>> -			WRITE_ONCE(cmdq->q.llq.cons, llq.cons);
>> -			arm_smmu_cmdq_shared_unlock(cmdq);
>> -		}
>> -	}
>> +	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
>> 
>> -	local_irq_restore(flags);
>> -	return ret;
>> +	return __arm_smmu_sync_poll_msi(smmu, ent.sync.msidata);
>> }
>> 
>> -static int arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
>> -				   struct arm_smmu_cmdq_ent *ent)
>> +static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>> {
>> 	u64 cmd[CMDQ_ENT_DWORDS];
>> +	unsigned long flags;
>> +	bool wfe = !!(smmu->features & ARM_SMMU_FEAT_SEV);
>> +	struct arm_smmu_cmdq_ent ent = { .opcode = CMDQ_OP_CMD_SYNC };
>> +	int ret;
>> 
>> -	if (arm_smmu_cmdq_build_cmd(cmd, ent)) {
>> -		dev_warn(smmu->dev, "ignoring unknown CMDQ opcode 0x%x\n",
>> -			 ent->opcode);
>> -		return -EINVAL;
>> -	}
>> +	arm_smmu_cmdq_build_cmd(cmd, &ent);
>> 
>> -	return arm_smmu_cmdq_issue_cmdlist(smmu, cmd, 1, false);
>> -}
>> +	spin_lock_irqsave(&smmu->cmdq.lock, flags);
>> +	arm_smmu_cmdq_insert_cmd(smmu, cmd);
>> +	ret = queue_poll_cons(&smmu->cmdq.q, true, wfe);
>> +	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
>> 
>> -static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>> -{
>> -	return arm_smmu_cmdq_issue_cmdlist(smmu, NULL, 0, true);
>> +	return ret;
>> }
>> 
>> -static void arm_smmu_cmdq_batch_add(struct arm_smmu_device *smmu,
>> -				    struct arm_smmu_cmdq_batch *cmds,
>> -				    struct arm_smmu_cmdq_ent *cmd)
>> +static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>> {
>> -	if (cmds->num == CMDQ_BATCH_ENTRIES) {
>> -		arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, false);
>> -		cmds->num = 0;
>> -	}
>> -	arm_smmu_cmdq_build_cmd(&cmds->cmds[cmds->num * CMDQ_ENT_DWORDS], cmd);
>> -	cmds->num++;
>> -}
>> +	int ret;
>> +	bool msi = (smmu->features & ARM_SMMU_FEAT_MSI) &&
>> +		   (smmu->features & ARM_SMMU_FEAT_COHERENCY);
>> 
>> -static int arm_smmu_cmdq_batch_submit(struct arm_smmu_device *smmu,
>> -				      struct arm_smmu_cmdq_batch *cmds)
>> -{
>> -	return arm_smmu_cmdq_issue_cmdlist(smmu, cmds->cmds, cmds->num, true);
>> +	ret = msi ? __arm_smmu_cmdq_issue_sync_msi(smmu)
>> +		  : __arm_smmu_cmdq_issue_sync(smmu);
>> +	if (ret)
>> +		dev_err_ratelimited(smmu->dev, "CMD_SYNC timeout\n");
>> +	return ret;
>> }
>> 
>> /* Context descriptor manipulation functions */
>> @@ -1535,7 +1137,6 @@ static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
>> 	size_t i;
>> 	unsigned long flags;
>> 	struct arm_smmu_master *master;
>> -	struct arm_smmu_cmdq_batch cmds = {};
>> 	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> 	struct arm_smmu_cmdq_ent cmd = {
>> 		.opcode	= CMDQ_OP_CFGI_CD,
>> @@ -1549,12 +1150,12 @@ static void arm_smmu_sync_cd(struct arm_smmu_domain *smmu_domain,
>> 	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
>> 		for (i = 0; i < master->num_sids; i++) {
>> 			cmd.cfgi.sid = master->sids[i];
>> -			arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
>> +			arm_smmu_cmdq_issue_cmd(smmu, &cmd);
>> 		}
>> 	}
>> 	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
>> 
>> -	arm_smmu_cmdq_batch_submit(smmu, &cmds);
>> +	arm_smmu_cmdq_issue_sync(smmu);
>> }
>> 
>> static int arm_smmu_alloc_cd_leaf_table(struct arm_smmu_device *smmu,
>> @@ -2189,16 +1790,17 @@ arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size,
>> 	cmd->atc.size	= log2_span;
>> }
>> 
>> -static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
>> +static int arm_smmu_atc_inv_master(struct arm_smmu_master *master,
>> +				   struct arm_smmu_cmdq_ent *cmd)
>> {
>> 	int i;
>> -	struct arm_smmu_cmdq_ent cmd;
>> 
>> -	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
>> +	if (!master->ats_enabled)
>> +		return 0;
>> 
>> 	for (i = 0; i < master->num_sids; i++) {
>> -		cmd.atc.sid = master->sids[i];
>> -		arm_smmu_cmdq_issue_cmd(master->smmu, &cmd);
>> +		cmd->atc.sid = master->sids[i];
>> +		arm_smmu_cmdq_issue_cmd(master->smmu, cmd);
>> 	}
>> 
>> 	return arm_smmu_cmdq_issue_sync(master->smmu);
>> @@ -2207,11 +1809,10 @@ static int arm_smmu_atc_inv_master(struct arm_smmu_master *master)
>> static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
>> 				   int ssid, unsigned long iova, size_t size)
>> {
>> -	int i;
>> +	int ret = 0;
>> 	unsigned long flags;
>> 	struct arm_smmu_cmdq_ent cmd;
>> 	struct arm_smmu_master *master;
>> -	struct arm_smmu_cmdq_batch cmds = {};
>> 
>> 	if (!(smmu_domain->smmu->features & ARM_SMMU_FEAT_ATS))
>> 		return 0;
>> @@ -2236,18 +1837,11 @@ static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
>> 	arm_smmu_atc_inv_to_cmd(ssid, iova, size, &cmd);
>> 
>> 	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
>> -	list_for_each_entry(master, &smmu_domain->devices, domain_head) {
>> -		if (!master->ats_enabled)
>> -			continue;
>> -
>> -		for (i = 0; i < master->num_sids; i++) {
>> -			cmd.atc.sid = master->sids[i];
>> -			arm_smmu_cmdq_batch_add(smmu_domain->smmu, &cmds, &cmd);
>> -		}
>> -	}
>> +	list_for_each_entry(master, &smmu_domain->devices, domain_head)
>> +		ret |= arm_smmu_atc_inv_master(master, &cmd);
>> 	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
>> 
>> -	return arm_smmu_cmdq_batch_submit(smmu_domain->smmu, &cmds);
>> +	return ret ? -ETIMEDOUT : 0;
>> }
>> 
>> /* IO_PGTABLE API */
>> @@ -2269,32 +1863,27 @@ static void arm_smmu_tlb_inv_context(void *cookie)
>> 	/*
>> 	 * NOTE: when io-pgtable is in non-strict mode, we may get here with
>> 	 * PTEs previously cleared by unmaps on the current CPU not yet visible
>> -	 * to the SMMU. We are relying on the dma_wmb() implicit during cmd
>> -	 * insertion to guarantee those are observed before the TLBI. Do be
>> -	 * careful, 007.
>> +	 * to the SMMU. We are relying on the DSB implicit in
>> +	 * queue_sync_prod_out() to guarantee those are observed before the
>> +	 * TLBI. Do be careful, 007.
>> 	 */
>> 	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
>> 	arm_smmu_cmdq_issue_sync(smmu);
>> 	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
>> }
>> 
>> -static void arm_smmu_tlb_inv_range(unsigned long iova, size_t size,
>> -				   size_t granule, bool leaf,
>> -				   struct arm_smmu_domain *smmu_domain)
>> +static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
>> +					  size_t granule, bool leaf, void *cookie)
>> {
>> +	struct arm_smmu_domain *smmu_domain = cookie;
>> 	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> -	unsigned long start = iova, end = iova + size, num_pages = 0, tg = 0;
>> -	size_t inv_range = granule;
>> -	struct arm_smmu_cmdq_batch cmds = {};
>> 	struct arm_smmu_cmdq_ent cmd = {
>> 		.tlbi = {
>> 			.leaf	= leaf,
>> +			.addr	= iova,
>> 		},
>> 	};
>> 
>> -	if (!size)
>> -		return;
>> -
>> 	if (smmu_domain->stage == ARM_SMMU_DOMAIN_S1) {
>> 		cmd.opcode	= CMDQ_OP_TLBI_NH_VA;
>> 		cmd.tlbi.asid	= smmu_domain->s1_cfg.cd.asid;
>> @@ -2303,78 +1892,37 @@ static void arm_smmu_tlb_inv_range(unsigned long iova, size_t size,
>> 		cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
>> 	}
>> 
>> -	if (smmu->features & ARM_SMMU_FEAT_RANGE_INV) {
>> -		/* Get the leaf page size */
>> -		tg = __ffs(smmu_domain->domain.pgsize_bitmap);
>> -
>> -		/* Convert page size of 12,14,16 (log2) to 1,2,3 */
>> -		cmd.tlbi.tg = (tg - 10) / 2;
>> -
>> -		/* Determine what level the granule is at */
>> -		cmd.tlbi.ttl = 4 - ((ilog2(granule) - 3) / (tg - 3));
>> -
>> -		num_pages = size >> tg;
>> -	}
>> -
>> -	while (iova < end) {
>> -		if (smmu->features & ARM_SMMU_FEAT_RANGE_INV) {
>> -			/*
>> -			 * On each iteration of the loop, the range is 5 bits
>> -			 * worth of the aligned size remaining.
>> -			 * The range in pages is:
>> -			 *
>> -			 * range = (num_pages & (0x1f << __ffs(num_pages)))
>> -			 */
>> -			unsigned long scale, num;
>> -
>> -			/* Determine the power of 2 multiple number of pages */
>> -			scale = __ffs(num_pages);
>> -			cmd.tlbi.scale = scale;
>> -
>> -			/* Determine how many chunks of 2^scale size we have */
>> -			num = (num_pages >> scale) & CMDQ_TLBI_RANGE_NUM_MAX;
>> -			cmd.tlbi.num = num - 1;
>> -
>> -			/* range is num * 2^scale * pgsize */
>> -			inv_range = num << (scale + tg);
>> -
>> -			/* Clear out the lower order bits for the next iteration */
>> -			num_pages -= num << scale;
>> -		}
>> -
>> -		cmd.tlbi.addr = iova;
>> -		arm_smmu_cmdq_batch_add(smmu, &cmds, &cmd);
>> -		iova += inv_range;
>> -	}
>> -	arm_smmu_cmdq_batch_submit(smmu, &cmds);
>> -
>> -	/*
>> -	 * Unfortunately, this can't be leaf-only since we may have
>> -	 * zapped an entire table.
>> -	 */
>> -	arm_smmu_atc_inv_domain(smmu_domain, 0, start, size);
>> +	do {
>> +		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
>> +		cmd.tlbi.addr += granule;
>> +	} while (size -= granule);
>> }
>> 
>> static void arm_smmu_tlb_inv_page_nosync(struct iommu_iotlb_gather *gather,
>> 					 unsigned long iova, size_t granule,
>> 					 void *cookie)
>> {
>> -	struct arm_smmu_domain *smmu_domain = cookie;
>> -	struct iommu_domain *domain = &smmu_domain->domain;
>> -
>> -	iommu_iotlb_gather_add_page(domain, gather, iova, granule);
>> +	arm_smmu_tlb_inv_range_nosync(iova, granule, granule, true, cookie);
>> }
>> 
>> static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,
>> 				  size_t granule, void *cookie)
>> {
>> -	arm_smmu_tlb_inv_range(iova, size, granule, false, cookie);
>> +	struct arm_smmu_domain *smmu_domain = cookie;
>> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> +
>> +	arm_smmu_tlb_inv_range_nosync(iova, size, granule, false, cookie);
>> +	arm_smmu_cmdq_issue_sync(smmu);
>> }
>> 
>> static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,
>> 				  size_t granule, void *cookie)
>> {
>> -	arm_smmu_tlb_inv_range(iova, size, granule, true, cookie);
>> +	struct arm_smmu_domain *smmu_domain = cookie;
>> +	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> +
>> +	arm_smmu_tlb_inv_range_nosync(iova, size, granule, true, cookie);
>> +	arm_smmu_cmdq_issue_sync(smmu);
>> }
>> 
>> static const struct iommu_flush_ops arm_smmu_flush_ops = {
>> @@ -2700,6 +2248,7 @@ static void arm_smmu_enable_ats(struct arm_smmu_master *master)
>> 
>> static void arm_smmu_disable_ats(struct arm_smmu_master *master)
>> {
>> +	struct arm_smmu_cmdq_ent cmd;
>> 	struct arm_smmu_domain *smmu_domain = master->domain;
>> 
>> 	if (!master->ats_enabled)
>> @@ -2711,8 +2260,9 @@ static void arm_smmu_disable_ats(struct arm_smmu_master *master)
>> 	 * ATC invalidation via the SMMU.
>> 	 */
>> 	wmb();
>> -	arm_smmu_atc_inv_master(master);
>> -	atomic_dec(&smmu_domain->nr_ats_masters);
>> +	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
>> +	arm_smmu_atc_inv_master(master, &cmd);
>> +    atomic_dec(&smmu_domain->nr_ats_masters);
>> }
>> 
>> static int arm_smmu_enable_pasid(struct arm_smmu_master *master)
>> @@ -2875,10 +2425,10 @@ static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain)
>> static void arm_smmu_iotlb_sync(struct iommu_domain *domain,
>> 				struct iommu_iotlb_gather *gather)
>> {
>> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> +	struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
>> 
>> -	arm_smmu_tlb_inv_range(gather->start, gather->end - gather->start,
>> -			       gather->pgsize, true, smmu_domain);
>> +	if (smmu)
>> +		arm_smmu_cmdq_issue_sync(smmu);
>> }
>> 
>> static phys_addr_t
>> @@ -3176,49 +2726,18 @@ static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>> 	return 0;
>> }
>> 
>> -static void arm_smmu_cmdq_free_bitmap(void *data)
>> -{
>> -	unsigned long *bitmap = data;
>> -	bitmap_free(bitmap);
>> -}
>> -
>> -static int arm_smmu_cmdq_init(struct arm_smmu_device *smmu)
>> -{
>> -	int ret = 0;
>> -	struct arm_smmu_cmdq *cmdq = &smmu->cmdq;
>> -	unsigned int nents = 1 << cmdq->q.llq.max_n_shift;
>> -	atomic_long_t *bitmap;
>> -
>> -	atomic_set(&cmdq->owner_prod, 0);
>> -	atomic_set(&cmdq->lock, 0);
>> -
>> -	bitmap = (atomic_long_t *)bitmap_zalloc(nents, GFP_KERNEL);
>> -	if (!bitmap) {
>> -		dev_err(smmu->dev, "failed to allocate cmdq bitmap\n");
>> -		ret = -ENOMEM;
>> -	} else {
>> -		cmdq->valid_map = bitmap;
>> -		devm_add_action(smmu->dev, arm_smmu_cmdq_free_bitmap, bitmap);
>> -	}
>> -
>> -	return ret;
>> -}
>> -
>> static int arm_smmu_init_queues(struct arm_smmu_device *smmu)
>> {
>> 	int ret;
>> 
>> 	/* cmdq */
>> +	spin_lock_init(&smmu->cmdq.lock);
>> 	ret = arm_smmu_init_one_queue(smmu, &smmu->cmdq.q, ARM_SMMU_CMDQ_PROD,
>> 				      ARM_SMMU_CMDQ_CONS, CMDQ_ENT_DWORDS,
>> 				      "cmdq");
>> 	if (ret)
>> 		return ret;
>> 
>> -	ret = arm_smmu_cmdq_init(smmu);
>> -	if (ret)
>> -		return ret;
>> -
>> 	/* evtq */
>> 	ret = arm_smmu_init_one_queue(smmu, &smmu->evtq.q, ARM_SMMU_EVTQ_PROD,
>> 				      ARM_SMMU_EVTQ_CONS, EVTQ_ENT_DWORDS,
>> @@ -3799,15 +3318,9 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>> 	/* Queue sizes, capped to ensure natural alignment */
>> 	smmu->cmdq.q.llq.max_n_shift = min_t(u32, CMDQ_MAX_SZ_SHIFT,
>> 					     FIELD_GET(IDR1_CMDQS, reg));
>> -	if (smmu->cmdq.q.llq.max_n_shift <= ilog2(CMDQ_BATCH_ENTRIES)) {
>> -		/*
>> -		 * We don't support splitting up batches, so one batch of
>> -		 * commands plus an extra sync needs to fit inside the command
>> -		 * queue. There's also no way we can handle the weird alignment
>> -		 * restrictions on the base pointer for a unit-length queue.
>> -		 */
>> -		dev_err(smmu->dev, "command queue size <= %d entries not supported\n",
>> -			CMDQ_BATCH_ENTRIES);
>> +	if (!smmu->cmdq.q.llq.max_n_shift) {
>> +		/* Odd alignment restrictions on the base, so ignore for now */
>> +		dev_err(smmu->dev, "unit-length command queue not supported\n");
>> 		return -ENXIO;
>> 	}
>> 
>> -- 
>> 2.17.1
>> 



^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 5/8] xen/arm: Remove support for PCI ATS on SMMUv3
  2020-12-02  0:39   ` Stefano Stabellini
@ 2020-12-02 13:07     ` Rahul Singh
  0 siblings, 0 replies; 53+ messages in thread
From: Rahul Singh @ 2020-12-02 13:07 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: xen-devel, Bertrand Marquis, Julien Grall, Volodymyr Babchuk

Hello Stefano,

> On 2 Dec 2020, at 12:39 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
> 
> On Thu, 26 Nov 2020, Rahul Singh wrote:
>> PCI ATS functionality is not implemented and tested on ARM. Remove the
>> PCI ATS support, once PCI ATS support is tested and available this
>> patch can be added.
>> 
>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
> 
> This looks like a revert of 9ce27afc0830f. Can we add that as a note to
> the commit message?
 
Ok I will add in next version.

> 
> One very minor comment at the bottom

Ack.

Regards,
Rahul
> 
> 
>> ---
>> xen/drivers/passthrough/arm/smmu-v3.c | 273 --------------------------
>> 1 file changed, 273 deletions(-)
>> 
>> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
>> index 401f7ae006..6a33628087 100644
>> --- a/xen/drivers/passthrough/arm/smmu-v3.c
>> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
>> @@ -460,16 +460,6 @@ struct arm_smmu_cmdq_ent {
>> 			u64			addr;
>> 		} tlbi;
>> 
>> -		#define CMDQ_OP_ATC_INV		0x40
>> -		#define ATC_INV_SIZE_ALL	52
>> -		struct {
>> -			u32			sid;
>> -			u32			ssid;
>> -			u64			addr;
>> -			u8			size;
>> -			bool			global;
>> -		} atc;
>> -
>> 		#define CMDQ_OP_PRI_RESP	0x41
>> 		struct {
>> 			u32			sid;
>> @@ -632,7 +622,6 @@ struct arm_smmu_master {
>> 	struct list_head		domain_head;
>> 	u32				*sids;
>> 	unsigned int			num_sids;
>> -	bool				ats_enabled;
>> 	unsigned int			ssid_bits;
>> };
>> 
>> @@ -650,7 +639,6 @@ struct arm_smmu_domain {
>> 
>> 	struct io_pgtable_ops		*pgtbl_ops;
>> 	bool				non_strict;
>> -	atomic_t			nr_ats_masters;
>> 
>> 	enum arm_smmu_domain_stage	stage;
>> 	union {
>> @@ -886,14 +874,6 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
>> 	case CMDQ_OP_TLBI_S12_VMALL:
>> 		cmd[0] |= FIELD_PREP(CMDQ_TLBI_0_VMID, ent->tlbi.vmid);
>> 		break;
>> -	case CMDQ_OP_ATC_INV:
>> -		cmd[0] |= FIELD_PREP(CMDQ_0_SSV, ent->substream_valid);
>> -		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_GLOBAL, ent->atc.global);
>> -		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_SSID, ent->atc.ssid);
>> -		cmd[0] |= FIELD_PREP(CMDQ_ATC_0_SID, ent->atc.sid);
>> -		cmd[1] |= FIELD_PREP(CMDQ_ATC_1_SIZE, ent->atc.size);
>> -		cmd[1] |= ent->atc.addr & CMDQ_ATC_1_ADDR_MASK;
>> -		break;
>> 	case CMDQ_OP_PRI_RESP:
>> 		cmd[0] |= FIELD_PREP(CMDQ_0_SSV, ent->substream_valid);
>> 		cmd[0] |= FIELD_PREP(CMDQ_PRI_0_SSID, ent->pri.ssid);
>> @@ -925,7 +905,6 @@ static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
>> 		[CMDQ_ERR_CERROR_NONE_IDX]	= "No error",
>> 		[CMDQ_ERR_CERROR_ILL_IDX]	= "Illegal command",
>> 		[CMDQ_ERR_CERROR_ABT_IDX]	= "Abort on command fetch",
>> -		[CMDQ_ERR_CERROR_ATC_INV_IDX]	= "ATC invalidate timeout",
>> 	};
>> 
>> 	int i;
>> @@ -945,14 +924,6 @@ static void arm_smmu_cmdq_skip_err(struct arm_smmu_device *smmu)
>> 		dev_err(smmu->dev, "retrying command fetch\n");
>> 	case CMDQ_ERR_CERROR_NONE_IDX:
>> 		return;
>> -	case CMDQ_ERR_CERROR_ATC_INV_IDX:
>> -		/*
>> -		 * ATC Invalidation Completion timeout. CONS is still pointing
>> -		 * at the CMD_SYNC. Attempt to complete other pending commands
>> -		 * by repeating the CMD_SYNC, though we might well end up back
>> -		 * here since the ATC invalidation may still be pending.
>> -		 */
>> -		return;
>> 	case CMDQ_ERR_CERROR_ILL_IDX:
>> 	default:
>> 		break;
>> @@ -1422,9 +1393,6 @@ static void arm_smmu_write_strtab_ent(struct arm_smmu_master *master, u32 sid,
>> 		val |= FIELD_PREP(STRTAB_STE_0_CFG, STRTAB_STE_0_CFG_S2_TRANS);
>> 	}
>> 
>> -	if (master->ats_enabled)
>> -		dst[1] |= cpu_to_le64(FIELD_PREP(STRTAB_STE_1_EATS,
>> -						 STRTAB_STE_1_EATS_TRANS));
>> 
>> 	arm_smmu_sync_ste_for_sid(smmu, sid);
>> 	/* See comment in arm_smmu_write_ctx_desc() */
>> @@ -1633,112 +1601,6 @@ static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
>> 	return IRQ_WAKE_THREAD;
>> }
>> 
>> -static void
>> -arm_smmu_atc_inv_to_cmd(int ssid, unsigned long iova, size_t size,
>> -			struct arm_smmu_cmdq_ent *cmd)
>> -{
>> -	size_t log2_span;
>> -	size_t span_mask;
>> -	/* ATC invalidates are always on 4096-bytes pages */
>> -	size_t inval_grain_shift = 12;
>> -	unsigned long page_start, page_end;
>> -
>> -	*cmd = (struct arm_smmu_cmdq_ent) {
>> -		.opcode			= CMDQ_OP_ATC_INV,
>> -		.substream_valid	= !!ssid,
>> -		.atc.ssid		= ssid,
>> -	};
>> -
>> -	if (!size) {
>> -		cmd->atc.size = ATC_INV_SIZE_ALL;
>> -		return;
>> -	}
>> -
>> -	page_start	= iova >> inval_grain_shift;
>> -	page_end	= (iova + size - 1) >> inval_grain_shift;
>> -
>> -	/*
>> -	 * In an ATS Invalidate Request, the address must be aligned on the
>> -	 * range size, which must be a power of two number of page sizes. We
>> -	 * thus have to choose between grossly over-invalidating the region, or
>> -	 * splitting the invalidation into multiple commands. For simplicity
>> -	 * we'll go with the first solution, but should refine it in the future
>> -	 * if multiple commands are shown to be more efficient.
>> -	 *
>> -	 * Find the smallest power of two that covers the range. The most
>> -	 * significant differing bit between the start and end addresses,
>> -	 * fls(start ^ end), indicates the required span. For example:
>> -	 *
>> -	 * We want to invalidate pages [8; 11]. This is already the ideal range:
>> -	 *		x = 0b1000 ^ 0b1011 = 0b11
>> -	 *		span = 1 << fls(x) = 4
>> -	 *
>> -	 * To invalidate pages [7; 10], we need to invalidate [0; 15]:
>> -	 *		x = 0b0111 ^ 0b1010 = 0b1101
>> -	 *		span = 1 << fls(x) = 16
>> -	 */
>> -	log2_span	= fls_long(page_start ^ page_end);
>> -	span_mask	= (1ULL << log2_span) - 1;
>> -
>> -	page_start	&= ~span_mask;
>> -
>> -	cmd->atc.addr	= page_start << inval_grain_shift;
>> -	cmd->atc.size	= log2_span;
>> -}
>> -
>> -static int arm_smmu_atc_inv_master(struct arm_smmu_master *master,
>> -				   struct arm_smmu_cmdq_ent *cmd)
>> -{
>> -	int i;
>> -
>> -	if (!master->ats_enabled)
>> -		return 0;
>> -
>> -	for (i = 0; i < master->num_sids; i++) {
>> -		cmd->atc.sid = master->sids[i];
>> -		arm_smmu_cmdq_issue_cmd(master->smmu, cmd);
>> -	}
>> -
>> -	return arm_smmu_cmdq_issue_sync(master->smmu);
>> -}
>> -
>> -static int arm_smmu_atc_inv_domain(struct arm_smmu_domain *smmu_domain,
>> -				   int ssid, unsigned long iova, size_t size)
>> -{
>> -	int ret = 0;
>> -	unsigned long flags;
>> -	struct arm_smmu_cmdq_ent cmd;
>> -	struct arm_smmu_master *master;
>> -
>> -	if (!(smmu_domain->smmu->features & ARM_SMMU_FEAT_ATS))
>> -		return 0;
>> -
>> -	/*
>> -	 * Ensure that we've completed prior invalidation of the main TLBs
>> -	 * before we read 'nr_ats_masters' in case of a concurrent call to
>> -	 * arm_smmu_enable_ats():
>> -	 *
>> -	 *	// unmap()			// arm_smmu_enable_ats()
>> -	 *	TLBI+SYNC			atomic_inc(&nr_ats_masters);
>> -	 *	smp_mb();			[...]
>> -	 *	atomic_read(&nr_ats_masters);	pci_enable_ats() // writel()
>> -	 *
>> -	 * Ensures that we always see the incremented 'nr_ats_masters' count if
>> -	 * ATS was enabled at the PCI device before completion of the TLBI.
>> -	 */
>> -	smp_mb();
>> -	if (!atomic_read(&smmu_domain->nr_ats_masters))
>> -		return 0;
>> -
>> -	arm_smmu_atc_inv_to_cmd(ssid, iova, size, &cmd);
>> -
>> -	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
>> -	list_for_each_entry(master, &smmu_domain->devices, domain_head)
>> -		ret |= arm_smmu_atc_inv_master(master, &cmd);
>> -	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
>> -
>> -	return ret ? -ETIMEDOUT : 0;
>> -}
>> 
>> /* IO_PGTABLE API */
>> static void arm_smmu_tlb_inv_context(void *cookie)
>> @@ -1765,7 +1627,6 @@ static void arm_smmu_tlb_inv_context(void *cookie)
>> 	 */
>> 	arm_smmu_cmdq_issue_cmd(smmu, &cmd);
>> 	arm_smmu_cmdq_issue_sync(smmu);
>> -	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
>> }
>> 
>> static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
>> @@ -2106,108 +1967,6 @@ static void arm_smmu_install_ste_for_dev(struct arm_smmu_master *master)
>> 	}
>> }
>> 
>> -static bool arm_smmu_ats_supported(struct arm_smmu_master *master)
>> -{
>> -	struct device *dev = master->dev;
>> -	struct arm_smmu_device *smmu = master->smmu;
>> -	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> -
>> -	if (!(smmu->features & ARM_SMMU_FEAT_ATS))
>> -		return false;
>> -
>> -	if (!(fwspec->flags & IOMMU_FWSPEC_PCI_RC_ATS))
>> -		return false;
>> -
>> -	return dev_is_pci(dev) && pci_ats_supported(to_pci_dev(dev));
>> -}
>> -
>> -static void arm_smmu_enable_ats(struct arm_smmu_master *master)
>> -{
>> -	size_t stu;
>> -	struct pci_dev *pdev;
>> -	struct arm_smmu_device *smmu = master->smmu;
>> -	struct arm_smmu_domain *smmu_domain = master->domain;
>> -
>> -	/* Don't enable ATS at the endpoint if it's not enabled in the STE */
>> -	if (!master->ats_enabled)
>> -		return;
>> -
>> -	/* Smallest Translation Unit: log2 of the smallest supported granule */
>> -	stu = __ffs(smmu->pgsize_bitmap);
>> -	pdev = to_pci_dev(master->dev);
>> -
>> -	atomic_inc(&smmu_domain->nr_ats_masters);
>> -	arm_smmu_atc_inv_domain(smmu_domain, 0, 0, 0);
>> -	if (pci_enable_ats(pdev, stu))
>> -		dev_err(master->dev, "Failed to enable ATS (STU %zu)\n", stu);
>> -}
>> -
>> -static void arm_smmu_disable_ats(struct arm_smmu_master *master)
>> -{
>> -	struct arm_smmu_cmdq_ent cmd;
>> -	struct arm_smmu_domain *smmu_domain = master->domain;
>> -
>> -	if (!master->ats_enabled)
>> -		return;
>> -
>> -	pci_disable_ats(to_pci_dev(master->dev));
>> -	/*
>> -	 * Ensure ATS is disabled at the endpoint before we issue the
>> -	 * ATC invalidation via the SMMU.
>> -	 */
>> -	wmb();
>> -	arm_smmu_atc_inv_to_cmd(0, 0, 0, &cmd);
>> -	arm_smmu_atc_inv_master(master, &cmd);
>> -    atomic_dec(&smmu_domain->nr_ats_masters);
>> -}
>> -
>> -static int arm_smmu_enable_pasid(struct arm_smmu_master *master)
>> -{
>> -	int ret;
>> -	int features;
>> -	int num_pasids;
>> -	struct pci_dev *pdev;
>> -
>> -	if (!dev_is_pci(master->dev))
>> -		return -ENODEV;
>> -
>> -	pdev = to_pci_dev(master->dev);
>> -
>> -	features = pci_pasid_features(pdev);
>> -	if (features < 0)
>> -		return features;
>> -
>> -	num_pasids = pci_max_pasids(pdev);
>> -	if (num_pasids <= 0)
>> -		return num_pasids;
>> -
>> -	ret = pci_enable_pasid(pdev, features);
>> -	if (ret) {
>> -		dev_err(&pdev->dev, "Failed to enable PASID\n");
>> -		return ret;
>> -	}
>> -
>> -	master->ssid_bits = min_t(u8, ilog2(num_pasids),
>> -				  master->smmu->ssid_bits);
>> -	return 0;
>> -}
>> -
>> -static void arm_smmu_disable_pasid(struct arm_smmu_master *master)
>> -{
>> -	struct pci_dev *pdev;
>> -
>> -	if (!dev_is_pci(master->dev))
>> -		return;
>> -
>> -	pdev = to_pci_dev(master->dev);
>> -
>> -	if (!pdev->pasid_enabled)
>> -		return;
>> -
>> -	master->ssid_bits = 0;
>> -	pci_disable_pasid(pdev);
>> -}
>> -
>> static void arm_smmu_detach_dev(struct arm_smmu_master *master)
>> {
>> 	unsigned long flags;
>> @@ -2216,14 +1975,11 @@ static void arm_smmu_detach_dev(struct arm_smmu_master *master)
>> 	if (!smmu_domain)
>> 		return;
>> 
>> -	arm_smmu_disable_ats(master);
>> -
>> 	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
>> 	list_del(&master->domain_head);
>> 	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
>> 
>> 	master->domain = NULL;
>> -	master->ats_enabled = false;
>> 	arm_smmu_install_ste_for_dev(master);
>> }
>> 
>> @@ -2271,17 +2027,12 @@ static int arm_smmu_attach_dev(struct iommu_domain *domain, struct device *dev)
>> 
>> 	master->domain = smmu_domain;
>> 
>> -	if (smmu_domain->stage != ARM_SMMU_DOMAIN_BYPASS)
>> -		master->ats_enabled = arm_smmu_ats_supported(master);
>> -
>> 	arm_smmu_install_ste_for_dev(master);
>> 
>> 	spin_lock_irqsave(&smmu_domain->devices_lock, flags);
>> 	list_add(&master->domain_head, &smmu_domain->devices);
>> 	spin_unlock_irqrestore(&smmu_domain->devices_lock, flags);
>> 
>> -	arm_smmu_enable_ats(master);
>> -
>> out_unlock:
>> 	mutex_unlock(&smmu_domain->init_mutex);
>> 	return ret;
>> @@ -2410,16 +2161,6 @@ static struct iommu_device *arm_smmu_probe_device(struct device *dev)
>> 
>> 	master->ssid_bits = min(smmu->ssid_bits, fwspec->num_pasid_bits);
>> 
>> -	/*
>> -	 * Note that PASID must be enabled before, and disabled after ATS:
>> -	 * PCI Express Base 4.0r1.0 - 10.5.1.3 ATS Control Register
>> -	 *
>> -	 *   Behavior is undefined if this bit is Set and the value of the PASID
>> -	 *   Enable, Execute Requested Enable, or Privileged Mode Requested bits
>> -	 *   are changed.
>> -	 */
>> -	arm_smmu_enable_pasid(master);
>> -
>> 	if (!(smmu->features & ARM_SMMU_FEAT_2_LVL_CDTAB))
>> 		master->ssid_bits = min_t(u8, master->ssid_bits,
>> 					  CTXDESC_LINEAR_CDMAX);
>> @@ -2442,7 +2183,6 @@ static void arm_smmu_release_device(struct device *dev)
>> 
>> 	master = dev_iommu_priv_get(dev);
>> 	arm_smmu_detach_dev(master);
>> -	arm_smmu_disable_pasid(master);
>> 	kfree(master);
>> 	iommu_fwspec_free(dev);
>> }
>> @@ -2997,15 +2737,6 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>> 		}
>> 	}
>> 
>> -	if (smmu->features & ARM_SMMU_FEAT_ATS) {
>> -		enables |= CR0_ATSCHK;
>> -		ret = arm_smmu_write_reg_sync(smmu, enables, ARM_SMMU_CR0,
>> -					      ARM_SMMU_CR0ACK);
>> -		if (ret) {
>> -			dev_err(smmu->dev, "failed to enable ATS check\n");
>> -			return ret;
>> -		}
>> -	}
>> 
>> 	ret = arm_smmu_setup_irqs(smmu);
>> 	if (ret) {
>> @@ -3076,13 +2807,9 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>> 	if (IS_ENABLED(CONFIG_PCI_PRI) && reg & IDR0_PRI)
>> 		smmu->features |= ARM_SMMU_FEAT_PRI;
>> 
>> -	if (IS_ENABLED(CONFIG_PCI_ATS) && reg & IDR0_ATS)
>> -		smmu->features |= ARM_SMMU_FEAT_ATS;
>> -
>> 	if (reg & IDR0_SEV)
>> 		smmu->features |= ARM_SMMU_FEAT_SEV;
>> 
>> -
>> 	if (reg & IDR0_HYP)
> 
> spurious change



^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 4/8] xen/arm: Remove support for MSI on SMMUv3
  2020-12-02  0:40     ` Stefano Stabellini
@ 2020-12-02 13:12       ` Rahul Singh
  2020-12-02 14:11         ` Julien Grall
  0 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-12-02 13:12 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: xen-devel, Bertrand Marquis, Julien Grall, Volodymyr Babchuk

Hello Stefano,

> On 2 Dec 2020, at 12:40 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
> 
> On Tue, 1 Dec 2020, Stefano Stabellini wrote:
>> On Thu, 26 Nov 2020, Rahul Singh wrote:
>>> XEN does not support MSI on ARM platforms, therefore remove the MSI
>>> support from SMMUv3 driver.
>>> 
>>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
>> 
>> I wonder if it makes sense to #ifdef CONFIG_MSI this code instead of
>> removing it completely.
> 
> One more thought: could this patch be achieved by reverting
> 166bdbd23161160f2abcea70621adba179050bee ? If this patch could be done
> by a couple of revert, it would be great to say it in the commit
> message.
> 
 Ok will add in next version.

> 
>> In the past, we tried to keep the entire file as textually similar to
>> the original Linux driver as possible to make it easier to backport
>> features and fixes. So, in this case we would probably not even used an
>> #ifdef but maybe something like:
>> 
>>  if (/* msi_enabled */ 0)
>>      return;
>> 
>> at the beginning of arm_smmu_setup_msis.
>> 
>> 
>> However, that strategy didn't actually work very well because backports
>> have proven difficult to do anyway. So at that point we might as well at
>> least have clean code in Xen and do the changes properly.

Main reason to remove the feature/code that is not usable in XEN is to have a clean code.

Regards,
Rahul

>> 
>> So that's my reasoning for accepting this patch :-)
>> 
>> Julien, are you happy with this too?
>> 
>> 
>>> ---
>>> xen/drivers/passthrough/arm/smmu-v3.c | 176 +-------------------------
>>> 1 file changed, 3 insertions(+), 173 deletions(-)
>>> 
>>> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
>>> index cec304e51a..401f7ae006 100644
>>> --- a/xen/drivers/passthrough/arm/smmu-v3.c
>>> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
>>> @@ -416,31 +416,6 @@ enum pri_resp {
>>> 	PRI_RESP_SUCC = 2,
>>> };
>>> 
>>> -enum arm_smmu_msi_index {
>>> -	EVTQ_MSI_INDEX,
>>> -	GERROR_MSI_INDEX,
>>> -	PRIQ_MSI_INDEX,
>>> -	ARM_SMMU_MAX_MSIS,
>>> -};
>>> -
>>> -static phys_addr_t arm_smmu_msi_cfg[ARM_SMMU_MAX_MSIS][3] = {
>>> -	[EVTQ_MSI_INDEX] = {
>>> -		ARM_SMMU_EVTQ_IRQ_CFG0,
>>> -		ARM_SMMU_EVTQ_IRQ_CFG1,
>>> -		ARM_SMMU_EVTQ_IRQ_CFG2,
>>> -	},
>>> -	[GERROR_MSI_INDEX] = {
>>> -		ARM_SMMU_GERROR_IRQ_CFG0,
>>> -		ARM_SMMU_GERROR_IRQ_CFG1,
>>> -		ARM_SMMU_GERROR_IRQ_CFG2,
>>> -	},
>>> -	[PRIQ_MSI_INDEX] = {
>>> -		ARM_SMMU_PRIQ_IRQ_CFG0,
>>> -		ARM_SMMU_PRIQ_IRQ_CFG1,
>>> -		ARM_SMMU_PRIQ_IRQ_CFG2,
>>> -	},
>>> -};
>>> -
>>> struct arm_smmu_cmdq_ent {
>>> 	/* Common fields */
>>> 	u8				opcode;
>>> @@ -504,10 +479,6 @@ struct arm_smmu_cmdq_ent {
>>> 		} pri;
>>> 
>>> 		#define CMDQ_OP_CMD_SYNC	0x46
>>> -		struct {
>>> -			u32			msidata;
>>> -			u64			msiaddr;
>>> -		} sync;
>>> 	};
>>> };
>>> 
>>> @@ -649,12 +620,6 @@ struct arm_smmu_device {
>>> 
>>> 	struct arm_smmu_strtab_cfg	strtab_cfg;
>>> 
>>> -	/* Hi16xx adds an extra 32 bits of goodness to its MSI payload */
>>> -	union {
>>> -		u32			sync_count;
>>> -		u64			padding;
>>> -	};
>>> -
>>> 	/* IOMMU core code handle */
>>> 	struct iommu_device		iommu;
>>> };
>>> @@ -945,20 +910,7 @@ static int arm_smmu_cmdq_build_cmd(u64 *cmd, struct arm_smmu_cmdq_ent *ent)
>>> 		cmd[1] |= FIELD_PREP(CMDQ_PRI_1_RESP, ent->pri.resp);
>>> 		break;
>>> 	case CMDQ_OP_CMD_SYNC:
>>> -		if (ent->sync.msiaddr)
>>> -			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_IRQ);
>>> -		else
>>> -			cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
>>> -		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSH, ARM_SMMU_SH_ISH);
>>> -		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIATTR, ARM_SMMU_MEMATTR_OIWB);
>>> -		/*
>>> -		 * Commands are written little-endian, but we want the SMMU to
>>> -		 * receive MSIData, and thus write it back to memory, in CPU
>>> -		 * byte order, so big-endian needs an extra byteswap here.
>>> -		 */
>>> -		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_MSIDATA,
>>> -				     cpu_to_le32(ent->sync.msidata));
>>> -		cmd[1] |= ent->sync.msiaddr & CMDQ_SYNC_1_MSIADDR_MASK;
>>> +		cmd[0] |= FIELD_PREP(CMDQ_SYNC_0_CS, CMDQ_SYNC_0_CS_SEV);
>>> 		break;
>>> 	default:
>>> 		return -ENOENT;
>>> @@ -1054,50 +1006,6 @@ static void arm_smmu_cmdq_issue_cmd(struct arm_smmu_device *smmu,
>>> 	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
>>> }
>>> 
>>> -/*
>>> - * The difference between val and sync_idx is bounded by the maximum size of
>>> - * a queue at 2^20 entries, so 32 bits is plenty for wrap-safe arithmetic.
>>> - */
>>> -static int __arm_smmu_sync_poll_msi(struct arm_smmu_device *smmu, u32 sync_idx)
>>> -{
>>> -	ktime_t timeout;
>>> -	u32 val;
>>> -
>>> -	timeout = ktime_add_us(ktime_get(), ARM_SMMU_CMDQ_SYNC_TIMEOUT_US);
>>> -	val = smp_cond_load_acquire(&smmu->sync_count,
>>> -				    (int)(VAL - sync_idx) >= 0 ||
>>> -				    !ktime_before(ktime_get(), timeout));
>>> -
>>> -	return (int)(val - sync_idx) < 0 ? -ETIMEDOUT : 0;
>>> -}
>>> -
>>> -static int __arm_smmu_cmdq_issue_sync_msi(struct arm_smmu_device *smmu)
>>> -{
>>> -	u64 cmd[CMDQ_ENT_DWORDS];
>>> -	unsigned long flags;
>>> -	struct arm_smmu_cmdq_ent  ent = {
>>> -		.opcode = CMDQ_OP_CMD_SYNC,
>>> -		.sync	= {
>>> -			.msiaddr = virt_to_phys(&smmu->sync_count),
>>> -		},
>>> -	};
>>> -
>>> -	spin_lock_irqsave(&smmu->cmdq.lock, flags);
>>> -
>>> -	/* Piggy-back on the previous command if it's a SYNC */
>>> -	if (smmu->prev_cmd_opcode == CMDQ_OP_CMD_SYNC) {
>>> -		ent.sync.msidata = smmu->sync_nr;
>>> -	} else {
>>> -		ent.sync.msidata = ++smmu->sync_nr;
>>> -		arm_smmu_cmdq_build_cmd(cmd, &ent);
>>> -		arm_smmu_cmdq_insert_cmd(smmu, cmd);
>>> -	}
>>> -
>>> -	spin_unlock_irqrestore(&smmu->cmdq.lock, flags);
>>> -
>>> -	return __arm_smmu_sync_poll_msi(smmu, ent.sync.msidata);
>>> -}
>>> -
>>> static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>>> {
>>> 	u64 cmd[CMDQ_ENT_DWORDS];
>>> @@ -1119,12 +1027,9 @@ static int __arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>>> static int arm_smmu_cmdq_issue_sync(struct arm_smmu_device *smmu)
>>> {
>>> 	int ret;
>>> -	bool msi = (smmu->features & ARM_SMMU_FEAT_MSI) &&
>>> -		   (smmu->features & ARM_SMMU_FEAT_COHERENCY);
>>> 
>>> -	ret = msi ? __arm_smmu_cmdq_issue_sync_msi(smmu)
>>> -		  : __arm_smmu_cmdq_issue_sync(smmu);
>>> -	if (ret)
>>> +	ret = __arm_smmu_cmdq_issue_sync(smmu);
>>> +	if ( ret )
>>> 		dev_err_ratelimited(smmu->dev, "CMD_SYNC timeout\n");
>>> 	return ret;
>>> }
>>> @@ -2898,83 +2803,10 @@ static int arm_smmu_update_gbpa(struct arm_smmu_device *smmu, u32 set, u32 clr)
>>> 	return ret;
>>> }
>>> 
>>> -static void arm_smmu_free_msis(void *data)
>>> -{
>>> -	struct device *dev = data;
>>> -	platform_msi_domain_free_irqs(dev);
>>> -}
>>> -
>>> -static void arm_smmu_write_msi_msg(struct msi_desc *desc, struct msi_msg *msg)
>>> -{
>>> -	phys_addr_t doorbell;
>>> -	struct device *dev = msi_desc_to_dev(desc);
>>> -	struct arm_smmu_device *smmu = dev_get_drvdata(dev);
>>> -	phys_addr_t *cfg = arm_smmu_msi_cfg[desc->platform.msi_index];
>>> -
>>> -	doorbell = (((u64)msg->address_hi) << 32) | msg->address_lo;
>>> -	doorbell &= MSI_CFG0_ADDR_MASK;
>>> -
>>> -	writeq_relaxed(doorbell, smmu->base + cfg[0]);
>>> -	writel_relaxed(msg->data, smmu->base + cfg[1]);
>>> -	writel_relaxed(ARM_SMMU_MEMATTR_DEVICE_nGnRE, smmu->base + cfg[2]);
>>> -}
>>> -
>>> -static void arm_smmu_setup_msis(struct arm_smmu_device *smmu)
>>> -{
>>> -	struct msi_desc *desc;
>>> -	int ret, nvec = ARM_SMMU_MAX_MSIS;
>>> -	struct device *dev = smmu->dev;
>>> -
>>> -	/* Clear the MSI address regs */
>>> -	writeq_relaxed(0, smmu->base + ARM_SMMU_GERROR_IRQ_CFG0);
>>> -	writeq_relaxed(0, smmu->base + ARM_SMMU_EVTQ_IRQ_CFG0);
>>> -
>>> -	if (smmu->features & ARM_SMMU_FEAT_PRI)
>>> -		writeq_relaxed(0, smmu->base + ARM_SMMU_PRIQ_IRQ_CFG0);
>>> -	else
>>> -		nvec--;
>>> -
>>> -	if (!(smmu->features & ARM_SMMU_FEAT_MSI))
>>> -		return;
>>> -
>>> -	if (!dev->msi_domain) {
>>> -		dev_info(smmu->dev, "msi_domain absent - falling back to wired irqs\n");
>>> -		return;
>>> -	}
>>> -
>>> -	/* Allocate MSIs for evtq, gerror and priq. Ignore cmdq */
>>> -	ret = platform_msi_domain_alloc_irqs(dev, nvec, arm_smmu_write_msi_msg);
>>> -	if (ret) {
>>> -		dev_warn(dev, "failed to allocate MSIs - falling back to wired irqs\n");
>>> -		return;
>>> -	}
>>> -
>>> -	for_each_msi_entry(desc, dev) {
>>> -		switch (desc->platform.msi_index) {
>>> -		case EVTQ_MSI_INDEX:
>>> -			smmu->evtq.q.irq = desc->irq;
>>> -			break;
>>> -		case GERROR_MSI_INDEX:
>>> -			smmu->gerr_irq = desc->irq;
>>> -			break;
>>> -		case PRIQ_MSI_INDEX:
>>> -			smmu->priq.q.irq = desc->irq;
>>> -			break;
>>> -		default:	/* Unknown */
>>> -			continue;
>>> -		}
>>> -	}
>>> -
>>> -	/* Add callback to free MSIs on teardown */
>>> -	devm_add_action(dev, arm_smmu_free_msis, dev);
>>> -}
>>> -
>>> static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>>> {
>>> 	int irq, ret;
>>> 
>>> -	arm_smmu_setup_msis(smmu);
>>> -
>>> 	/* Request interrupt lines */
>>> 	irq = smmu->evtq.q.irq;
>>> 	if (irq) {
>>> @@ -3250,8 +3082,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>>> 	if (reg & IDR0_SEV)
>>> 		smmu->features |= ARM_SMMU_FEAT_SEV;
>>> 
>>> -	if (reg & IDR0_MSI)
>>> -		smmu->features |= ARM_SMMU_FEAT_MSI;
>>> 
>>> 	if (reg & IDR0_HYP)
>>> 		smmu->features |= ARM_SMMU_FEAT_HYP;
>>> -- 
>>> 2.17.1
>>> 
>> 



^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 6/8] xen/arm: Remove support for Stage-1 translation on SMMUv3.
  2020-12-02  0:53   ` Stefano Stabellini
@ 2020-12-02 13:13     ` Rahul Singh
  0 siblings, 0 replies; 53+ messages in thread
From: Rahul Singh @ 2020-12-02 13:13 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: xen-devel, Bertrand Marquis, Julien Grall, Volodymyr Babchuk

Hello Stefano,

> On 2 Dec 2020, at 12:53 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
> 
> On Thu, 26 Nov 2020, Rahul Singh wrote:
>> Linux SMMUv3 driver supports both Stage-1 and Stage-2 translations.
>> As of now only Stage-2 translation support has been tested.
>> 
>> Once Stage-1 translation support is tested this patch can be added.
>> 
>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
> 
> [...]
> 
> 
>> @@ -1871,19 +1476,9 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>> 	}
>> 
>> 	/* Restrict the stage to what we can actually support */
>> -	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S1))
>> -		smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
>> -	if (!(smmu->features & ARM_SMMU_FEAT_TRANS_S2))
>> -		smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
>> +	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
> 
> It would be good to add an helpful error message if
> ARM_SMMU_FEAT_TRANS_S2 is missing.
> 

Ack. I will add error message.

Regards,
Rahul
> 
>> 	switch (smmu_domain->stage) {
>> -	case ARM_SMMU_DOMAIN_S1:
>> -		ias = (smmu->features & ARM_SMMU_FEAT_VAX) ? 52 : 48;
>> -		ias = min_t(unsigned long, ias, VA_BITS);
>> -		oas = smmu->ias;
>> -		fmt = ARM_64_LPAE_S1;
>> -		finalise_stage_fn = arm_smmu_domain_finalise_s1;
>> -		break;
>> 	case ARM_SMMU_DOMAIN_NESTED:
>> 	case ARM_SMMU_DOMAIN_S2:
>> 		ias = smmu->ias;



^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 2/8] xen/arm: revert atomic operation related command-queue insertion patch
  2020-11-26 17:02 ` [PATCH v2 2/8] xen/arm: revert atomic operation related command-queue insertion patch Rahul Singh
  2020-12-01 22:23   ` Stefano Stabellini
@ 2020-12-02 13:44   ` Julien Grall
  2020-12-03 11:49     ` Rahul Singh
  1 sibling, 1 reply; 53+ messages in thread
From: Julien Grall @ 2020-12-02 13:44 UTC (permalink / raw)
  To: Rahul Singh, xen-devel
  Cc: bertrand.marquis, Stefano Stabellini, Volodymyr Babchuk

Hi Rahul,

On 26/11/2020 17:02, Rahul Singh wrote:
> Linux SMMUv3 code implements the commands-queue insertion based on
> atomic operations implemented in Linux. Atomic functions used by the
> commands-queue insertion is not implemented in XEN therefore revert the
> patch that implemented the commands-queue insertion based on atomic
> operations.

This commit message explains why we revert but not the consequence of 
the revert. Can outline if there are any and why they are fine?

I am also interested to have a list of *must* have for the driver to be 
out of the tech preview.

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 3/8] xen/arm: revert patch related to XArray
  2020-11-26 17:02 ` [PATCH v2 3/8] xen/arm: revert patch related to XArray Rahul Singh
  2020-12-02  0:20   ` Stefano Stabellini
@ 2020-12-02 13:46   ` Julien Grall
  2020-12-03 12:57     ` Rahul Singh
  1 sibling, 1 reply; 53+ messages in thread
From: Julien Grall @ 2020-12-02 13:46 UTC (permalink / raw)
  To: Rahul Singh, xen-devel
  Cc: bertrand.marquis, Stefano Stabellini, Volodymyr Babchuk

Hi Rahul,

On 26/11/2020 17:02, Rahul Singh wrote:
> XArray is not implemented in XEN revert the patch that introduce the
> XArray code in SMMUv3 driver.

Similar to the atomic revert, you are explaining why the revert but not 
the consequences. I think this is quite important to have them outline 
in the commit message as it looks like it means the SMMU driver would 
not scale.

> 
> Once XArray is implemented in XEN this patch can be added in XEN.

What's the plan for that?

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 5/8] xen/arm: Remove support for PCI ATS on SMMUv3
  2020-11-26 17:02 ` [PATCH v2 5/8] xen/arm: Remove support for PCI ATS " Rahul Singh
  2020-12-02  0:39   ` Stefano Stabellini
@ 2020-12-02 13:57   ` Julien Grall
  1 sibling, 0 replies; 53+ messages in thread
From: Julien Grall @ 2020-12-02 13:57 UTC (permalink / raw)
  To: Rahul Singh, xen-devel
  Cc: bertrand.marquis, Stefano Stabellini, Volodymyr Babchuk

Hi Rahul,

On 26/11/2020 17:02, Rahul Singh wrote:
> PCI ATS functionality is not implemented and tested on ARM. 

I agree that short term, this is not going to be implemented. However, I 
do expect this feature to be used medium term (mostly likely before the 
driver is out of tech preview).

So I think it would be best to keep the code around.

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 4/8] xen/arm: Remove support for MSI on SMMUv3
  2020-12-02 13:12       ` Rahul Singh
@ 2020-12-02 14:11         ` Julien Grall
  2020-12-03 12:59           ` Rahul Singh
  0 siblings, 1 reply; 53+ messages in thread
From: Julien Grall @ 2020-12-02 14:11 UTC (permalink / raw)
  To: Rahul Singh, Stefano Stabellini
  Cc: xen-devel, Bertrand Marquis, Volodymyr Babchuk

Hi Rahul,

On 02/12/2020 13:12, Rahul Singh wrote:
> Hello Stefano,
> 
>> On 2 Dec 2020, at 12:40 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
>>
>> On Tue, 1 Dec 2020, Stefano Stabellini wrote:
>>> On Thu, 26 Nov 2020, Rahul Singh wrote:
>>>> XEN does not support MSI on ARM platforms, therefore remove the MSI
>>>> support from SMMUv3 driver.
>>>>
>>>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
>>>
>>> I wonder if it makes sense to #ifdef CONFIG_MSI this code instead of
>>> removing it completely.
>>
>> One more thought: could this patch be achieved by reverting
>> 166bdbd23161160f2abcea70621adba179050bee ? If this patch could be done
>> by a couple of revert, it would be great to say it in the commit
>> message.
>>
>   Ok will add in next version.
> 
>>
>>> In the past, we tried to keep the entire file as textually similar to
>>> the original Linux driver as possible to make it easier to backport
>>> features and fixes. So, in this case we would probably not even used an
>>> #ifdef but maybe something like:
>>>
>>>   if (/* msi_enabled */ 0)
>>>       return;
>>>
>>> at the beginning of arm_smmu_setup_msis.
>>>
>>>
>>> However, that strategy didn't actually work very well because backports
>>> have proven difficult to do anyway. So at that point we might as well at
>>> least have clean code in Xen and do the changes properly.

It was difficult because Linux decided to rework how IOMMU drivers 
works. I agree the risk is still there and therefore clean code would be 
better with some caveats (see below).

> 
> Main reason to remove the feature/code that is not usable in XEN is to have a clean code.

I agree that short term this feature will not be usable. However, I 
think we need to think about {medium, long}-term plan to avoid extra 
effort in the future because the driver evolve in a way making the 
revert of revert impossible.

Therefore I would prefer to keep both the MSI and PCI ATS present as 
they are going to be useful/necessary on some platforms. It doesn't 
matter that they don't work because the driver will be in tech preview.

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN
  2020-12-02  1:48   ` Stefano Stabellini
@ 2020-12-02 14:34     ` Rahul Singh
  2020-12-02 14:39       ` Julien Grall
  0 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-12-02 14:34 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: xen-devel, Bertrand Marquis, Julien Grall, Volodymyr Babchuk

Hello Stefano,

> On 2 Dec 2020, at 1:48 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
> 
> On Thu, 26 Nov 2020, Rahul Singh wrote:
>> struct io_pgtable_ops, struct io_pgtable_cfg, struct iommu_flush_ops,
>> and struct iommu_ops related code are linux specific.
>> 
>> Remove code related to above struct as code is dead code in XEN.
> 
> There are still instances of struct io_pgtable_cfg after applying this
> patch in the following functions:
> - arm_smmu_domain_finalise_s2
> - arm_smmu_domain_finalise
> 

Ok. I will remove the instances.
> 
> 
>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
>> ---
>> xen/drivers/passthrough/arm/smmu-v3.c | 457 --------------------------
>> 1 file changed, 457 deletions(-)
>> 
>> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
>> index 40e3890a58..55d1cba194 100644
>> --- a/xen/drivers/passthrough/arm/smmu-v3.c
>> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
>> @@ -599,7 +593,6 @@ struct arm_smmu_domain {
>> 	struct arm_smmu_device		*smmu;
>> 	struct mutex			init_mutex; /* Protects smmu pointer */
>> 
>> -	struct io_pgtable_ops		*pgtbl_ops;
>> 	bool				non_strict;
>> 
>> 	enum arm_smmu_domain_stage	stage;
>> @@ -1297,74 +1290,6 @@ static void arm_smmu_tlb_inv_context(void *cookie)
>> 	arm_smmu_cmdq_issue_sync(smmu);
>> }
>> 
>> -static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
>> -					  size_t granule, bool leaf, void *cookie)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = cookie;
>> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> -	struct arm_smmu_cmdq_ent cmd = {
>> -		.tlbi = {
>> -			.leaf	= leaf,
>> -			.addr	= iova,
>> -		},
>> -	};
>> -
>> -	cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
>> -	cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
>> -
>> -	do {
>> -		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
>> -		cmd.tlbi.addr += granule;
>> -	} while (size -= granule);
>> -}
>> -
>> -static void arm_smmu_tlb_inv_page_nosync(struct iommu_iotlb_gather *gather,
>> -					 unsigned long iova, size_t granule,
>> -					 void *cookie)
>> -{
>> -	arm_smmu_tlb_inv_range_nosync(iova, granule, granule, true, cookie);
>> -}
>> -
>> -static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,
>> -				  size_t granule, void *cookie)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = cookie;
>> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> -
>> -	arm_smmu_tlb_inv_range_nosync(iova, size, granule, false, cookie);
>> -	arm_smmu_cmdq_issue_sync(smmu);
>> -}
>> -
>> -static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,
>> -				  size_t granule, void *cookie)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = cookie;
>> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> -
>> -	arm_smmu_tlb_inv_range_nosync(iova, size, granule, true, cookie);
>> -	arm_smmu_cmdq_issue_sync(smmu);
>> -}
>> -
>> -static const struct iommu_flush_ops arm_smmu_flush_ops = {
>> -	.tlb_flush_all	= arm_smmu_tlb_inv_context,
>> -	.tlb_flush_walk = arm_smmu_tlb_inv_walk,
>> -	.tlb_flush_leaf = arm_smmu_tlb_inv_leaf,
>> -	.tlb_add_page	= arm_smmu_tlb_inv_page_nosync,
>> -};
>> -
>> -/* IOMMU API */
>> -static bool arm_smmu_capable(enum iommu_cap cap)
>> -{
>> -	switch (cap) {
>> -	case IOMMU_CAP_CACHE_COHERENCY:
>> -		return true;
>> -	case IOMMU_CAP_NOEXEC:
>> -		return true;
>> -	default:
>> -		return false;
>> -	}
>> -}
>> -
>> static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
>> {
>> 	struct arm_smmu_domain *smmu_domain;
>> @@ -1421,7 +1346,6 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>> 	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>> 
>> 	iommu_put_dma_cookie(domain);
>> -	free_io_pgtable_ops(smmu_domain->pgtbl_ops);
>> 
>> 	if (cfg->vmid)
>> 		arm_smmu_bitmap_free(smmu->vmid_map, cfg->vmid);
>> @@ -1429,7 +1353,6 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>> 	kfree(smmu_domain);
>> }
>> 
>> -
>> static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>> 				       struct arm_smmu_master *master,
>> 				       struct io_pgtable_cfg *pgtbl_cfg)
>> @@ -1437,7 +1360,6 @@ static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>> 	int vmid;
>> 	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> 	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>> -	typeof(&pgtbl_cfg->arm_lpae_s2_cfg.vtcr) vtcr;
>> 
>> 	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
>> 	if (vmid < 0)
>> @@ -1461,20 +1383,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>> {
>> 	int ret;
>> 	unsigned long ias, oas;
>> -	enum io_pgtable_fmt fmt;
>> -	struct io_pgtable_cfg pgtbl_cfg;
>> -	struct io_pgtable_ops *pgtbl_ops;
>> 	int (*finalise_stage_fn)(struct arm_smmu_domain *,
>> 				 struct arm_smmu_master *,
>> 				 struct io_pgtable_cfg *);
>> 	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> 	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> 
>> -	if (domain->type == IOMMU_DOMAIN_IDENTITY) {
>> -		smmu_domain->stage = ARM_SMMU_DOMAIN_BYPASS;
>> -		return 0;
>> -	}
>> -
>> 	/* Restrict the stage to what we can actually support */
>> 	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
>> 
>> @@ -1483,40 +1397,17 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>> 	case ARM_SMMU_DOMAIN_S2:
>> 		ias = smmu->ias;
>> 		oas = smmu->oas;
>> -		fmt = ARM_64_LPAE_S2;
>> 		finalise_stage_fn = arm_smmu_domain_finalise_s2;
>> 		break;
>> 	default:
>> 		return -EINVAL;
>> 	}
>> 
>> -	pgtbl_cfg = (struct io_pgtable_cfg) {
>> -		.pgsize_bitmap	= smmu->pgsize_bitmap,
>> -		.ias		= ias,
>> -		.oas		= oas,
>> -		.coherent_walk	= smmu->features & ARM_SMMU_FEAT_COHERENCY,
>> -		.tlb		= &arm_smmu_flush_ops,
>> -		.iommu_dev	= smmu->dev,
>> -	};
>> -
>> -	if (smmu_domain->non_strict)
>> -		pgtbl_cfg.quirks |= IO_PGTABLE_QUIRK_NON_STRICT;
>> -
>> -	pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
>> -	if (!pgtbl_ops)
>> -		return -ENOMEM;
>> -
>> -	domain->pgsize_bitmap = pgtbl_cfg.pgsize_bitmap;
>> -	domain->geometry.aperture_end = (1UL << pgtbl_cfg.ias) - 1;
>> -	domain->geometry.force_aperture = true;
>> -
>> 	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
>> 	if (ret < 0) {
>> -		free_io_pgtable_ops(pgtbl_ops);
>> 		return ret;
>> 	}
>> 
>> -	smmu_domain->pgtbl_ops = pgtbl_ops;
>> 	return 0;
>> }
>> 
>> @@ -1626,71 +1517,6 @@ out_unlock:
>> 	return ret;
>> }
>> 
>> -static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
>> -			phys_addr_t paddr, size_t size, int prot, gfp_t gfp)
>> -{
>> -	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
>> -
>> -	if (!ops)
>> -		return -ENODEV;
>> -
>> -	return ops->map(ops, iova, paddr, size, prot, gfp);
>> -}
>> -
>> -static size_t arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova,
>> -			     size_t size, struct iommu_iotlb_gather *gather)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> -	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
>> -
>> -	if (!ops)
>> -		return 0;
>> -
>> -	return ops->unmap(ops, iova, size, gather);
>> -}
>> -
>> -static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> -
>> -	if (smmu_domain->smmu)
>> -		arm_smmu_tlb_inv_context(smmu_domain);
>> -}
>> -
>> -static void arm_smmu_iotlb_sync(struct iommu_domain *domain,
>> -				struct iommu_iotlb_gather *gather)
>> -{
>> -	struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
>> -
>> -	if (smmu)
>> -		arm_smmu_cmdq_issue_sync(smmu);
>> -}
>> -
>> -static phys_addr_t
>> -arm_smmu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
>> -{
>> -	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
>> -
>> -	if (domain->type == IOMMU_DOMAIN_IDENTITY)
>> -		return iova;
>> -
>> -	if (!ops)
>> -		return 0;
>> -
>> -	return ops->iova_to_phys(ops, iova);
>> -}
>> -
>> -static struct platform_driver arm_smmu_driver;
>> -
>> -static
>> -struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle *fwnode)
>> -{
>> -	struct device *dev = driver_find_device_by_fwnode(&arm_smmu_driver.driver,
>> -							  fwnode);
>> -	put_device(dev);
>> -	return dev ? dev_get_drvdata(dev) : NULL;
>> -}
>> -
>> static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
>> {
>> 	unsigned long limit = smmu->strtab_cfg.num_l1_ents;
>> @@ -1701,206 +1527,6 @@ static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
>> 	return sid < limit;
>> }
>> 
>> -static struct iommu_ops arm_smmu_ops;
>> -
>> -static struct iommu_device *arm_smmu_probe_device(struct device *dev)
>> -{
>> -	int i, ret;
>> -	struct arm_smmu_device *smmu;
>> -	struct arm_smmu_master *master;
>> -	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> -
>> -	if (!fwspec || fwspec->ops != &arm_smmu_ops)
>> -		return ERR_PTR(-ENODEV);
>> -
>> -	if (WARN_ON_ONCE(dev_iommu_priv_get(dev)))
>> -		return ERR_PTR(-EBUSY);
>> -
>> -	smmu = arm_smmu_get_by_fwnode(fwspec->iommu_fwnode);
>> -	if (!smmu)
>> -		return ERR_PTR(-ENODEV);
>> -
>> -	master = kzalloc(sizeof(*master), GFP_KERNEL);
>> -	if (!master)
>> -		return ERR_PTR(-ENOMEM);
>> -
>> -	master->dev = dev;
>> -	master->smmu = smmu;
>> -	master->sids = fwspec->ids;
>> -	master->num_sids = fwspec->num_ids;
>> -	dev_iommu_priv_set(dev, master);
>> -
>> -	/* Check the SIDs are in range of the SMMU and our stream table */
>> -	for (i = 0; i < master->num_sids; i++) {
>> -		u32 sid = master->sids[i];
>> -
>> -		if (!arm_smmu_sid_in_range(smmu, sid)) {
>> -			ret = -ERANGE;
>> -			goto err_free_master;
>> -		}
>> -
>> -		/* Ensure l2 strtab is initialised */
>> -		if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
>> -			ret = arm_smmu_init_l2_strtab(smmu, sid);
>> -			if (ret)
>> -				goto err_free_master;
>> -		}
>> -	}
>> -
>> -	return &smmu->iommu;
>> -
>> -err_free_master:
>> -	kfree(master);
>> -	dev_iommu_priv_set(dev, NULL);
>> -	return ERR_PTR(ret);
>> -}
>> -
>> -static void arm_smmu_release_device(struct device *dev)
>> -{
>> -	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> -	struct arm_smmu_master *master;
>> -
>> -	if (!fwspec || fwspec->ops != &arm_smmu_ops)
>> -		return;
>> -
>> -	master = dev_iommu_priv_get(dev);
>> -	arm_smmu_detach_dev(master);
>> -	kfree(master);
>> -	iommu_fwspec_free(dev);
>> -}
>> -
>> -static struct iommu_group *arm_smmu_device_group(struct device *dev)
>> -{
>> -	struct iommu_group *group;
>> -
>> -	/*
>> -	 * We don't support devices sharing stream IDs other than PCI RID
>> -	 * aliases, since the necessary ID-to-device lookup becomes rather
>> -	 * impractical given a potential sparse 32-bit stream ID space.
>> -	 */
>> -	if (dev_is_pci(dev))
>> -		group = pci_device_group(dev);
>> -	else
>> -		group = generic_device_group(dev);
>> -
>> -	return group;
>> -}
>> -
>> -static int arm_smmu_domain_get_attr(struct iommu_domain *domain,
>> -				    enum iommu_attr attr, void *data)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> -
>> -	switch (domain->type) {
>> -	case IOMMU_DOMAIN_UNMANAGED:
>> -		switch (attr) {
>> -		case DOMAIN_ATTR_NESTING:
>> -			*(int *)data = (smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED);
>> -			return 0;
>> -		default:
>> -			return -ENODEV;
>> -		}
>> -		break;
>> -	case IOMMU_DOMAIN_DMA:
>> -		switch (attr) {
>> -		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
>> -			*(int *)data = smmu_domain->non_strict;
>> -			return 0;
>> -		default:
>> -			return -ENODEV;
>> -		}
>> -		break;
>> -	default:
>> -		return -EINVAL;
>> -	}
>> -}
>> -
>> -static int arm_smmu_domain_set_attr(struct iommu_domain *domain,
>> -				    enum iommu_attr attr, void *data)
>> -{
>> -	int ret = 0;
>> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> -
>> -	mutex_lock(&smmu_domain->init_mutex);
>> -
>> -	switch (domain->type) {
>> -	case IOMMU_DOMAIN_UNMANAGED:
>> -		switch (attr) {
>> -		case DOMAIN_ATTR_NESTING:
>> -			if (smmu_domain->smmu) {
>> -				ret = -EPERM;
>> -				goto out_unlock;
>> -			}
>> -
>> -			if (*(int *)data)
>> -				smmu_domain->stage = ARM_SMMU_DOMAIN_NESTED;
>> -			else
>> -				smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
>> -			break;
>> -		default:
>> -			ret = -ENODEV;
>> -		}
>> -		break;
>> -	case IOMMU_DOMAIN_DMA:
>> -		switch(attr) {
>> -		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
>> -			smmu_domain->non_strict = *(int *)data;
>> -			break;
>> -		default:
>> -			ret = -ENODEV;
>> -		}
>> -		break;
>> -	default:
>> -		ret = -EINVAL;
>> -	}
>> -
>> -out_unlock:
>> -	mutex_unlock(&smmu_domain->init_mutex);
>> -	return ret;
>> -}
>> -
>> -static int arm_smmu_of_xlate(struct device *dev, struct of_phandle_args *args)
>> -{
>> -	return iommu_fwspec_add_ids(dev, args->args, 1);
>> -}
>> -
>> -static void arm_smmu_get_resv_regions(struct device *dev,
>> -				      struct list_head *head)
>> -{
>> -	struct iommu_resv_region *region;
>> -	int prot = IOMMU_WRITE | IOMMU_NOEXEC | IOMMU_MMIO;
>> -
>> -	region = iommu_alloc_resv_region(MSI_IOVA_BASE, MSI_IOVA_LENGTH,
>> -					 prot, IOMMU_RESV_SW_MSI);
>> -	if (!region)
>> -		return;
>> -
>> -	list_add_tail(&region->list, head);
>> -
>> -	iommu_dma_get_resv_regions(dev, head);
>> -}
> 
> Arguably this could have been removed previously as part of the MSI
> patch, but that's OK either way.

Ack. 
> 
> 
>> -static struct iommu_ops arm_smmu_ops = {
>> -	.capable		= arm_smmu_capable,
>> -	.domain_alloc		= arm_smmu_domain_alloc,
>> -	.domain_free		= arm_smmu_domain_free,
>> -	.attach_dev		= arm_smmu_attach_dev,
>> -	.map			= arm_smmu_map,
>> -	.unmap			= arm_smmu_unmap,
>> -	.flush_iotlb_all	= arm_smmu_flush_iotlb_all,
>> -	.iotlb_sync		= arm_smmu_iotlb_sync,
>> -	.iova_to_phys		= arm_smmu_iova_to_phys,
>> -	.probe_device		= arm_smmu_probe_device,
>> -	.release_device		= arm_smmu_release_device,
>> -	.device_group		= arm_smmu_device_group,
>> -	.domain_get_attr	= arm_smmu_domain_get_attr,
>> -	.domain_set_attr	= arm_smmu_domain_set_attr,
>> -	.of_xlate		= arm_smmu_of_xlate,
>> -	.get_resv_regions	= arm_smmu_get_resv_regions,
>> -	.put_resv_regions	= generic_iommu_put_resv_regions,
>> -	.pgsize_bitmap		= -1UL, /* Restricted during device attach */
>> -};
>> -
>> /* Probing and initialisation functions */
>> static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>> 				   struct arm_smmu_queue *q,
>> @@ -2515,21 +2139,10 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>> 	default:
>> 		dev_info(smmu->dev,
>> 			"unknown output address size. Truncating to 48-bit\n");
>> -		fallthrough;
>> 	case IDR5_OAS_48_BIT:
>> 		smmu->oas = 48;
>> 	}
>> 
>> -	if (arm_smmu_ops.pgsize_bitmap == -1UL)
>> -		arm_smmu_ops.pgsize_bitmap = smmu->pgsize_bitmap;
>> -	else
>> -		arm_smmu_ops.pgsize_bitmap |= smmu->pgsize_bitmap;
>> -
>> -	/* Set the DMA mask for our table walker */
>> -	if (dma_set_mask_and_coherent(smmu->dev, DMA_BIT_MASK(smmu->oas)))
>> -		dev_warn(smmu->dev,
>> -			 "failed to set DMA mask for table walker\n");
>> -
>> 	smmu->ias = max(smmu->ias, smmu->oas);
>> 
>> 	dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n",
>> @@ -2595,9 +2208,6 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev,
>> 
>> 	parse_driver_options(smmu);
>> 
>> -	if (of_dma_is_coherent(dev->of_node))
>> -		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
>> -
> 
> Why this change? The ARM_SMMU_FEAT_COHERENCY flag is still used in
> arm_smmu_device_hw_probe.

I remove this as this is linux specific.  I will remove ARM_SMMU_FEAT_COHERENCY flag  used in arm_smmu_device_hw_probe

Regards,
Rahul




^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN
  2020-12-02 14:34     ` Rahul Singh
@ 2020-12-02 14:39       ` Julien Grall
  0 siblings, 0 replies; 53+ messages in thread
From: Julien Grall @ 2020-12-02 14:39 UTC (permalink / raw)
  To: Rahul Singh, Stefano Stabellini
  Cc: xen-devel, Bertrand Marquis, Volodymyr Babchuk

Hi Rahul,

On 02/12/2020 14:34, Rahul Singh wrote:
>>> 	dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n",
>>> @@ -2595,9 +2208,6 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev,
>>>
>>> 	parse_driver_options(smmu);
>>>
>>> -	if (of_dma_is_coherent(dev->of_node))
>>> -		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
>>> -
>>
>> Why this change? The ARM_SMMU_FEAT_COHERENCY flag is still used in
>> arm_smmu_device_hw_probe.
> 
> I remove this as this is linux specific.  I will remove ARM_SMMU_FEAT_COHERENCY flag  used in arm_smmu_device_hw_probe

 From my understanding, ARM_SMMU_FEAT_COHERENCY indicate whether the 
SMMU page table walker will snoop the cache. If the flag is not set, 
then Xen will have to clean to PoC every entry updated in the p2m.

Therefore, I think we need to keep this code.

In the case we don't need to keep the code, then I think the reason 
should be explained in the commit message.

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN
  2020-11-26 17:02 ` [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN Rahul Singh
  2020-12-02  1:48   ` Stefano Stabellini
@ 2020-12-02 14:45   ` Julien Grall
  2020-12-03 14:33     ` Rahul Singh
  1 sibling, 1 reply; 53+ messages in thread
From: Julien Grall @ 2020-12-02 14:45 UTC (permalink / raw)
  To: Rahul Singh, xen-devel
  Cc: bertrand.marquis, Stefano Stabellini, Volodymyr Babchuk



On 26/11/2020 17:02, Rahul Singh wrote:
> struct io_pgtable_ops, struct io_pgtable_cfg, struct iommu_flush_ops,
> and struct iommu_ops related code are linux specific.

So the assumption is we are going to support only sharing with the P2M 
and the IOMMU. That's probably fine short term, but long-term we are 
going to need unsharing the page-tables (there are issues on some 
platforms if the ITS doorbell is accessed by the processors).


I am ok with removing anything related to the unsharing code. But I 
think it should be clarified here.

> 
> Remove code related to above struct as code is dead code in XEN.
> 
> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
> ---
>   xen/drivers/passthrough/arm/smmu-v3.c | 457 --------------------------
>   1 file changed, 457 deletions(-)
> 
> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
> index 40e3890a58..55d1cba194 100644
> --- a/xen/drivers/passthrough/arm/smmu-v3.c
> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
> @@ -402,13 +402,7 @@
>   #define ARM_SMMU_CMDQ_SYNC_TIMEOUT_US	1000000 /* 1s! */
>   #define ARM_SMMU_CMDQ_SYNC_SPIN_COUNT	10
>   
> -#define MSI_IOVA_BASE			0x8000000
> -#define MSI_IOVA_LENGTH			0x100000
> -
>   static bool disable_bypass = 1;
> -module_param_named(disable_bypass, disable_bypass, bool, S_IRUGO);
> -MODULE_PARM_DESC(disable_bypass,
> -	"Disable bypass streams such that incoming transactions from devices that are not attached to an iommu domain will report an abort back to the device and will not be allowed to pass through the SMMU.");

Per your commit message, this doesn't look related to what you intend to 
remove. Maybe your commit message should be updated?

>   
>   enum pri_resp {
>   	PRI_RESP_DENY = 0,
> @@ -599,7 +593,6 @@ struct arm_smmu_domain {
>   	struct arm_smmu_device		*smmu;
>   	struct mutex			init_mutex; /* Protects smmu pointer */
>   
> -	struct io_pgtable_ops		*pgtbl_ops;
>   	bool				non_strict;
>   
>   	enum arm_smmu_domain_stage	stage;
> @@ -1297,74 +1290,6 @@ static void arm_smmu_tlb_inv_context(void *cookie)
>   	arm_smmu_cmdq_issue_sync(smmu);
>   }
>   
> -static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
> -					  size_t granule, bool leaf, void *cookie)
> -{
> -	struct arm_smmu_domain *smmu_domain = cookie;
> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
> -	struct arm_smmu_cmdq_ent cmd = {
> -		.tlbi = {
> -			.leaf	= leaf,
> -			.addr	= iova,
> -		},
> -	};
> -
> -	cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
> -	cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
> -
> -	do {
> -		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
> -		cmd.tlbi.addr += granule;
> -	} while (size -= granule);
> -}
> -
> -static void arm_smmu_tlb_inv_page_nosync(struct iommu_iotlb_gather *gather,
> -					 unsigned long iova, size_t granule,
> -					 void *cookie)
> -{
> -	arm_smmu_tlb_inv_range_nosync(iova, granule, granule, true, cookie);
> -}
> -
> -static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,
> -				  size_t granule, void *cookie)
> -{
> -	struct arm_smmu_domain *smmu_domain = cookie;
> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
> -
> -	arm_smmu_tlb_inv_range_nosync(iova, size, granule, false, cookie);
> -	arm_smmu_cmdq_issue_sync(smmu);
> -}
> -
> -static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,
> -				  size_t granule, void *cookie)
> -{
> -	struct arm_smmu_domain *smmu_domain = cookie;
> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
> -
> -	arm_smmu_tlb_inv_range_nosync(iova, size, granule, true, cookie);
> -	arm_smmu_cmdq_issue_sync(smmu);
> -}
> -
> -static const struct iommu_flush_ops arm_smmu_flush_ops = {
> -	.tlb_flush_all	= arm_smmu_tlb_inv_context,
> -	.tlb_flush_walk = arm_smmu_tlb_inv_walk,
> -	.tlb_flush_leaf = arm_smmu_tlb_inv_leaf,
> -	.tlb_add_page	= arm_smmu_tlb_inv_page_nosync,
> -};
> -
> -/* IOMMU API */
> -static bool arm_smmu_capable(enum iommu_cap cap)
> -{
> -	switch (cap) {
> -	case IOMMU_CAP_CACHE_COHERENCY:
> -		return true;
> -	case IOMMU_CAP_NOEXEC:
> -		return true;
> -	default:
> -		return false;
> -	}
> -}
> -
>   static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
>   {
>   	struct arm_smmu_domain *smmu_domain;
> @@ -1421,7 +1346,6 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>   	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>   
>   	iommu_put_dma_cookie(domain);
> -	free_io_pgtable_ops(smmu_domain->pgtbl_ops);
>   
>   	if (cfg->vmid)
>   		arm_smmu_bitmap_free(smmu->vmid_map, cfg->vmid);
> @@ -1429,7 +1353,6 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>   	kfree(smmu_domain);
>   }
>   
> -

Looks like a spurious change.

>   static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>   				       struct arm_smmu_master *master,
>   				       struct io_pgtable_cfg *pgtbl_cfg)

You commit message leads to think that all the use of struct 
io_pgtable_cfg will be removed.

> @@ -1437,7 +1360,6 @@ static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>   	int vmid;
>   	struct arm_smmu_device *smmu = smmu_domain->smmu;
>   	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
> -	typeof(&pgtbl_cfg->arm_lpae_s2_cfg.vtcr) vtcr;

It feels a bit odd that the definition of 'vtcr' is removed but there 
are still users.

>   
>   	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
>   	if (vmid < 0)
> @@ -1461,20 +1383,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>   {
>   	int ret;
>   	unsigned long ias, oas;
> -	enum io_pgtable_fmt fmt;
> -	struct io_pgtable_cfg pgtbl_cfg;
> -	struct io_pgtable_ops *pgtbl_ops;
>   	int (*finalise_stage_fn)(struct arm_smmu_domain *,
>   				 struct arm_smmu_master *,
>   				 struct io_pgtable_cfg *);
>   	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>   	struct arm_smmu_device *smmu = smmu_domain->smmu;
>   
> -	if (domain->type == IOMMU_DOMAIN_IDENTITY) {
> -		smmu_domain->stage = ARM_SMMU_DOMAIN_BYPASS;
> -		return 0;
> -	}
> -

Per your commit message, this doesn't look related to what you intend to 
remove. Maybe your commit message should be updated?


>   	/* Restrict the stage to what we can actually support */
>   	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
>   
> @@ -1483,40 +1397,17 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>   	case ARM_SMMU_DOMAIN_S2:
>   		ias = smmu->ias;
>   		oas = smmu->oas;
> -		fmt = ARM_64_LPAE_S2;
>   		finalise_stage_fn = arm_smmu_domain_finalise_s2;
>   		break;
>   	default:
>   		return -EINVAL;
>   	}
>   
> -	pgtbl_cfg = (struct io_pgtable_cfg) {
> -		.pgsize_bitmap	= smmu->pgsize_bitmap,
> -		.ias		= ias,
> -		.oas		= oas,
> -		.coherent_walk	= smmu->features & ARM_SMMU_FEAT_COHERENCY,
> -		.tlb		= &arm_smmu_flush_ops,
> -		.iommu_dev	= smmu->dev,
> -	};
> -
> -	if (smmu_domain->non_strict)
> -		pgtbl_cfg.quirks |= IO_PGTABLE_QUIRK_NON_STRICT;
> -
> -	pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
> -	if (!pgtbl_ops)
> -		return -ENOMEM;
> -
> -	domain->pgsize_bitmap = pgtbl_cfg.pgsize_bitmap;
> -	domain->geometry.aperture_end = (1UL << pgtbl_cfg.ias) - 1;
> -	domain->geometry.force_aperture = true;
> -
>   	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
>   	if (ret < 0) {
> -		free_io_pgtable_ops(pgtbl_ops);
>   		return ret;
>   	}
>   
> -	smmu_domain->pgtbl_ops = pgtbl_ops;
>   	return 0;
>   }
>   
> @@ -1626,71 +1517,6 @@ out_unlock:
>   	return ret;
>   }
>   
> -static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
> -			phys_addr_t paddr, size_t size, int prot, gfp_t gfp)
> -{
> -	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
> -
> -	if (!ops)
> -		return -ENODEV;
> -
> -	return ops->map(ops, iova, paddr, size, prot, gfp);
> -}
> -
> -static size_t arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova,
> -			     size_t size, struct iommu_iotlb_gather *gather)
> -{
> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> -	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
> -
> -	if (!ops)
> -		return 0;
> -
> -	return ops->unmap(ops, iova, size, gather);
> -}
> -
> -static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain)
> -{
> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> -
> -	if (smmu_domain->smmu)
> -		arm_smmu_tlb_inv_context(smmu_domain);
> -}
> -
> -static void arm_smmu_iotlb_sync(struct iommu_domain *domain,
> -				struct iommu_iotlb_gather *gather)
> -{
> -	struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
> -
> -	if (smmu)
> -		arm_smmu_cmdq_issue_sync(smmu);
> -}
> -
> -static phys_addr_t
> -arm_smmu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
> -{
> -	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
> -
> -	if (domain->type == IOMMU_DOMAIN_IDENTITY)
> -		return iova;
> -
> -	if (!ops)
> -		return 0;
> -
> -	return ops->iova_to_phys(ops, iova);
> -}
> -
> -static struct platform_driver arm_smmu_driver;
> -
> -static
> -struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle *fwnode)
> -{
> -	struct device *dev = driver_find_device_by_fwnode(&arm_smmu_driver.driver,
> -							  fwnode);
> -	put_device(dev);
> -	return dev ? dev_get_drvdata(dev) : NULL;
> -}
> -
>   static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
>   {
>   	unsigned long limit = smmu->strtab_cfg.num_l1_ents;
> @@ -1701,206 +1527,6 @@ static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
>   	return sid < limit;
>   }
>   
> -static struct iommu_ops arm_smmu_ops;
> -
> -static struct iommu_device *arm_smmu_probe_device(struct device *dev)
> -{

Most of the code here looks useful to Xen. I think you want to keep the 
code and re-use it afterwards.

> -	int i, ret;
> -	struct arm_smmu_device *smmu;
> -	struct arm_smmu_master *master;
> -	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> -
> -	if (!fwspec || fwspec->ops != &arm_smmu_ops)
> -		return ERR_PTR(-ENODEV);
> -
> -	if (WARN_ON_ONCE(dev_iommu_priv_get(dev)))
> -		return ERR_PTR(-EBUSY);
> -
> -	smmu = arm_smmu_get_by_fwnode(fwspec->iommu_fwnode);
> -	if (!smmu)
> -		return ERR_PTR(-ENODEV);
> -
> -	master = kzalloc(sizeof(*master), GFP_KERNEL);
> -	if (!master)
> -		return ERR_PTR(-ENOMEM);
> -
> -	master->dev = dev;
> -	master->smmu = smmu;
> -	master->sids = fwspec->ids;
> -	master->num_sids = fwspec->num_ids;
> -	dev_iommu_priv_set(dev, master);
> -
> -	/* Check the SIDs are in range of the SMMU and our stream table */
> -	for (i = 0; i < master->num_sids; i++) {
> -		u32 sid = master->sids[i];
> -
> -		if (!arm_smmu_sid_in_range(smmu, sid)) {
> -			ret = -ERANGE;
> -			goto err_free_master;
> -		}
> -
> -		/* Ensure l2 strtab is initialised */
> -		if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
> -			ret = arm_smmu_init_l2_strtab(smmu, sid);
> -			if (ret)
> -				goto err_free_master;
> -		}
> -	}
> -
> -	return &smmu->iommu;
> -
> -err_free_master:
> -	kfree(master);
> -	dev_iommu_priv_set(dev, NULL);
> -	return ERR_PTR(ret);
> -}
> -
> -static void arm_smmu_release_device(struct device *dev)
> -{
> -	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> -	struct arm_smmu_master *master;
> -
> -	if (!fwspec || fwspec->ops != &arm_smmu_ops)
> -		return;
> -
> -	master = dev_iommu_priv_get(dev);
> -	arm_smmu_detach_dev(master);
> -	kfree(master);
> -	iommu_fwspec_free(dev);
> -}
> -
> -static struct iommu_group *arm_smmu_device_group(struct device *dev)
> -{
> -	struct iommu_group *group;
> -
> -	/*
> -	 * We don't support devices sharing stream IDs other than PCI RID
> -	 * aliases, since the necessary ID-to-device lookup becomes rather
> -	 * impractical given a potential sparse 32-bit stream ID space.
> -	 */
> -	if (dev_is_pci(dev))
> -		group = pci_device_group(dev);
> -	else
> -		group = generic_device_group(dev);
> -
> -	return group;
> -}
> -
> -static int arm_smmu_domain_get_attr(struct iommu_domain *domain,
> -				    enum iommu_attr attr, void *data)
> -{
> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> -
> -	switch (domain->type) {
> -	case IOMMU_DOMAIN_UNMANAGED:
> -		switch (attr) {
> -		case DOMAIN_ATTR_NESTING:
> -			*(int *)data = (smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED);
> -			return 0;
> -		default:
> -			return -ENODEV;
> -		}
> -		break;
> -	case IOMMU_DOMAIN_DMA:
> -		switch (attr) {
> -		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
> -			*(int *)data = smmu_domain->non_strict;
> -			return 0;
> -		default:
> -			return -ENODEV;
> -		}
> -		break;
> -	default:
> -		return -EINVAL;
> -	}
> -}
> -
> -static int arm_smmu_domain_set_attr(struct iommu_domain *domain,
> -				    enum iommu_attr attr, void *data)
> -{
> -	int ret = 0;
> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> -
> -	mutex_lock(&smmu_domain->init_mutex);
> -
> -	switch (domain->type) {
> -	case IOMMU_DOMAIN_UNMANAGED:
> -		switch (attr) {
> -		case DOMAIN_ATTR_NESTING:
> -			if (smmu_domain->smmu) {
> -				ret = -EPERM;
> -				goto out_unlock;
> -			}
> -
> -			if (*(int *)data)
> -				smmu_domain->stage = ARM_SMMU_DOMAIN_NESTED;
> -			else
> -				smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
> -			break;
> -		default:
> -			ret = -ENODEV;
> -		}
> -		break;
> -	case IOMMU_DOMAIN_DMA:
> -		switch(attr) {
> -		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
> -			smmu_domain->non_strict = *(int *)data;
> -			break;
> -		default:
> -			ret = -ENODEV;
> -		}
> -		break;
> -	default:
> -		ret = -EINVAL;
> -	}
> -
> -out_unlock:
> -	mutex_unlock(&smmu_domain->init_mutex);
> -	return ret;
> -}
> -
> -static int arm_smmu_of_xlate(struct device *dev, struct of_phandle_args *args)
> -{
> -	return iommu_fwspec_add_ids(dev, args->args, 1);
> -}
> -
> -static void arm_smmu_get_resv_regions(struct device *dev,
> -				      struct list_head *head)
> -{
> -	struct iommu_resv_region *region;
> -	int prot = IOMMU_WRITE | IOMMU_NOEXEC | IOMMU_MMIO;
> -
> -	region = iommu_alloc_resv_region(MSI_IOVA_BASE, MSI_IOVA_LENGTH,
> -					 prot, IOMMU_RESV_SW_MSI);
> -	if (!region)
> -		return;
> -
> -	list_add_tail(&region->list, head);
> -
> -	iommu_dma_get_resv_regions(dev, head);
> -}
> -
> -static struct iommu_ops arm_smmu_ops = {
> -	.capable		= arm_smmu_capable,
> -	.domain_alloc		= arm_smmu_domain_alloc,
> -	.domain_free		= arm_smmu_domain_free,
> -	.attach_dev		= arm_smmu_attach_dev,
> -	.map			= arm_smmu_map,
> -	.unmap			= arm_smmu_unmap,
> -	.flush_iotlb_all	= arm_smmu_flush_iotlb_all,
> -	.iotlb_sync		= arm_smmu_iotlb_sync,
> -	.iova_to_phys		= arm_smmu_iova_to_phys,
> -	.probe_device		= arm_smmu_probe_device,
> -	.release_device		= arm_smmu_release_device,
> -	.device_group		= arm_smmu_device_group,
> -	.domain_get_attr	= arm_smmu_domain_get_attr,
> -	.domain_set_attr	= arm_smmu_domain_set_attr,
> -	.of_xlate		= arm_smmu_of_xlate,
> -	.get_resv_regions	= arm_smmu_get_resv_regions,
> -	.put_resv_regions	= generic_iommu_put_resv_regions,
> -	.pgsize_bitmap		= -1UL, /* Restricted during device attach */
> -};
> -
>   /* Probing and initialisation functions */
>   static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>   				   struct arm_smmu_queue *q,
> @@ -2406,7 +2032,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>   	switch (FIELD_GET(IDR0_STALL_MODEL, reg)) {
>   	case IDR0_STALL_MODEL_FORCE:
>   		smmu->features |= ARM_SMMU_FEAT_STALL_FORCE;
> -		fallthrough;

We should keep all the fallthrough documented. So I think we want to 
introduce the fallthrough in Xen as well.

>   	case IDR0_STALL_MODEL_STALL:
>   		smmu->features |= ARM_SMMU_FEAT_STALLS;
>   	}
> @@ -2426,7 +2051,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>   	switch (FIELD_GET(IDR0_TTF, reg)) {
>   	case IDR0_TTF_AARCH32_64:
>   		smmu->ias = 40;
> -		fallthrough;
>   	case IDR0_TTF_AARCH64:
>   		break;
>   	default:
> @@ -2515,21 +2139,10 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>   	default:
>   		dev_info(smmu->dev,
>   			"unknown output address size. Truncating to 48-bit\n");
> -		fallthrough;
>   	case IDR5_OAS_48_BIT:
>   		smmu->oas = 48;
>   	}
>   
> -	if (arm_smmu_ops.pgsize_bitmap == -1UL)
> -		arm_smmu_ops.pgsize_bitmap = smmu->pgsize_bitmap;
> -	else
> -		arm_smmu_ops.pgsize_bitmap |= smmu->pgsize_bitmap;
> -
> -	/* Set the DMA mask for our table walker */
> -	if (dma_set_mask_and_coherent(smmu->dev, DMA_BIT_MASK(smmu->oas)))
> -		dev_warn(smmu->dev,
> -			 "failed to set DMA mask for table walker\n");
> -
>   	smmu->ias = max(smmu->ias, smmu->oas);
>   
>   	dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n",
> @@ -2595,9 +2208,6 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev,
>   
>   	parse_driver_options(smmu);
>   
> -	if (of_dma_is_coherent(dev->of_node))
> -		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
> -
>   	return ret;
>   }
>   
> @@ -2609,55 +2219,6 @@ 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)

This seems a bit odd that you are removing the function but not the 
callers. Shouldn't you keep this function until next patch?

> -{
> -	struct resource res = {
> -		.flags = IORESOURCE_MEM,
> -		.start = start,
> -		.end = start + size - 1,
> -	};
> -
> -	return devm_ioremap_resource(dev, &res);
> -}
> -
>   static int arm_smmu_device_probe(struct platform_device *pdev)
>   {
>   	int irq, ret;
> @@ -2785,21 +2346,3 @@ static const struct of_device_id arm_smmu_of_match[] = {
>   	{ .compatible = "arm,smmu-v3", },
>   	{ },
>   };
> -MODULE_DEVICE_TABLE(of, arm_smmu_of_match);
> -
> -static struct platform_driver arm_smmu_driver = {
> -	.driver	= {
> -		.name			= "arm-smmu-v3",
> -		.of_match_table		= arm_smmu_of_match,
> -		.suppress_bind_attrs	= true,
> -	},
> -	.probe	= arm_smmu_device_probe,
> -	.remove	= arm_smmu_device_remove,
> -	.shutdown = arm_smmu_device_shutdown,
> -};
> -module_platform_driver(arm_smmu_driver);
> -
> -MODULE_DESCRIPTION("IOMMU API for ARM architected SMMUv3 implementations");
> -MODULE_AUTHOR("Will Deacon <will@kernel.org>");
> -MODULE_ALIAS("platform:arm-smmu-v3");
> -MODULE_LICENSE("GPL v2");
> 

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-11-26 17:02 ` [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
  2020-12-02  2:51   ` Stefano Stabellini
@ 2020-12-02 16:22   ` Julien Grall
  2020-12-07 12:12     ` Rahul Singh
  1 sibling, 1 reply; 53+ messages in thread
From: Julien Grall @ 2020-12-02 16:22 UTC (permalink / raw)
  To: Rahul Singh, xen-devel
  Cc: bertrand.marquis, Andrew Cooper, George Dunlap, Ian Jackson,
	Jan Beulich, Stefano Stabellini, Wei Liu, Paul Durrant,
	Volodymyr Babchuk

Hi Rahul,

On 26/11/2020 17:02, Rahul Singh wrote:
> Add support for ARM architected SMMUv3 implementation. It is based on
> the Linux SMMUv3 driver.
> 
> Major differences with regard to Linux driver are as follows:
> 1. Only Stage-2 translation is supported as compared to the Linux driver
>     that supports both Stage-1 and Stage-2 translations.
> 2. Use P2M  page table instead of creating one as SMMUv3 has the
>     capability to share the page tables with the CPU.
> 3. Tasklets are used in place of threaded IRQ's in Linux for event queue
>     and priority queue IRQ handling.

On the previous version, we discussed that using tasklets is not a 
suitable replacement for threaded IRQs. What's the plan to address it?

> 4. Latest version of the Linux SMMUv3 code implements the commands queue
>     access functions based on atomic operations implemented in Linux.
>     Atomic functions used by the commands queue access functions are not
>     implemented in XEN therefore we decided to port the earlier version
>     of the code. Once the proper atomic operations will be available in
>     XEN the driver can be updated.
> 5. Driver is currently supported as Tech Preview.
> 
> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
> ---
>   MAINTAINERS                           |   6 +
>   SUPPORT.md                            |   1 +
>   xen/drivers/passthrough/Kconfig       |  10 +
>   xen/drivers/passthrough/arm/Makefile  |   1 +
>   xen/drivers/passthrough/arm/smmu-v3.c | 986 +++++++++++++++++++++-----
>   5 files changed, 814 insertions(+), 190 deletions(-)
> 
> diff --git a/MAINTAINERS b/MAINTAINERS
> index dab38a6a14..1d63489eec 100644
> --- a/MAINTAINERS
> +++ b/MAINTAINERS
> @@ -249,6 +249,12 @@ F:	xen/include/asm-arm/
>   F:	xen/include/public/arch-arm/
>   F:	xen/include/public/arch-arm.h
>   
> +ARM SMMUv3
> +M:	Bertrand Marquis <bertrand.marquis@arm.com>
> +M:	Rahul Singh <rahul.singh@arm.com>
> +S:	Supported
> +F:	xen/drivers/passthrough/arm/smmu-v3.c
> +
>   Change Log
>   M:	Paul Durrant <paul@xen.org>
>   R:	Community Manager <community.manager@xenproject.org>
> diff --git a/SUPPORT.md b/SUPPORT.md
> index ab02aca5f4..e402c7202d 100644
> --- a/SUPPORT.md
> +++ b/SUPPORT.md
> @@ -68,6 +68,7 @@ For the Cortex A57 r0p0 - r1p1, see Errata 832075.
>       Status, ARM SMMUv1: Supported, not security supported
>       Status, ARM SMMUv2: Supported, not security supported
>       Status, Renesas IPMMU-VMSA: Supported, not security supported
> +    Status, ARM SMMUv3: Tech Preview

Please move this right after "ARM SMMUv2".

>   
>   ### ARM/GICv3 ITS
>   
> diff --git a/xen/drivers/passthrough/Kconfig b/xen/drivers/passthrough/Kconfig
> index 0036007ec4..5b71c59f47 100644
> --- a/xen/drivers/passthrough/Kconfig
> +++ b/xen/drivers/passthrough/Kconfig
> @@ -13,6 +13,16 @@ config ARM_SMMU
>   	  Say Y here if your SoC includes an IOMMU device implementing the
>   	  ARM SMMU architecture.
>   
> +config ARM_SMMU_V3
> +	bool "ARM Ltd. System MMU Version 3 (SMMUv3) Support" if EXPERT
> +	depends on ARM_64
> +	---help---
> +	 Support for implementations of the ARM System MMU architecture
> +	 version 3.
> +
> +	 Say Y here if your system includes an IOMMU device implementing
> +	 the ARM SMMUv3 architecture.
> +
>   config IPMMU_VMSA
>   	bool "Renesas IPMMU-VMSA found in R-Car Gen3 SoCs"
>   	depends on ARM_64
> diff --git a/xen/drivers/passthrough/arm/Makefile b/xen/drivers/passthrough/arm/Makefile
> index fcd918ea3e..c5fb3b58a5 100644
> --- a/xen/drivers/passthrough/arm/Makefile
> +++ b/xen/drivers/passthrough/arm/Makefile
> @@ -1,3 +1,4 @@
>   obj-y += iommu.o iommu_helpers.o iommu_fwspec.o
>   obj-$(CONFIG_ARM_SMMU) += smmu.o
>   obj-$(CONFIG_IPMMU_VMSA) += ipmmu-vmsa.o
> +obj-$(CONFIG_ARM_SMMU_V3) += smmu-v3.o
> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
> index 55d1cba194..8f2337e7f2 100644
> --- a/xen/drivers/passthrough/arm/smmu-v3.c
> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
> @@ -2,36 +2,280 @@
>   /*
>    * IOMMU API for ARM architected SMMUv3 implementations.
>    *
> - * Copyright (C) 2015 ARM Limited
> + * Based on Linux's SMMUv3 driver:
> + *    drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
> + *    commit: 951cbbc386ff01b50da4f46387e994e81d9ab431
> + * and Xen's SMMU driver:
> + *    xen/drivers/passthrough/arm/smmu.c

I would suggest to list the major differences here as well.

>    *
> - * Author: Will Deacon <will.deacon@arm.com>
> + * Copyright (C) 2015 ARM Limited Will Deacon <will.deacon@arm.com>

Why did you merge the Author and copyright line?

>    *
> - * This driver is powered by bad coffee and bombay mix.
> + * Copyright (C) 2020 Arm Ltd.
> + *
> + * 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, see <http://www.gnu.org/licenses/>.
> + *
> + */
> +
> +#include <xen/acpi.h>
> +#include <xen/config.h>
> +#include <xen/delay.h>
> +#include <xen/errno.h>
> +#include <xen/err.h>
> +#include <xen/irq.h>
> +#include <xen/lib.h>
> +#include <xen/list.h>
> +#include <xen/mm.h>
> +#include <xen/rbtree.h>
> +#include <xen/sched.h>
> +#include <xen/sizes.h>
> +#include <xen/vmap.h>
> +#include <asm/atomic.h>
> +#include <asm/device.h>
> +#include <asm/io.h>
> +#include <asm/platform.h>
> +#include <asm/iommu_fwspec.h>

All the headers seem to be alphabetically ordered but this one.

> +
> +/* Linux compatibility functions. */

Some of the helpers here seem to be similar to the SMMU driver. Can we 
have an header that can be shared between the two?

> +typedef paddr_t dma_addr_t;
> +typedef unsigned int gfp_t;
> +
> +#define platform_device device
> +
> +#define GFP_KERNEL 0
> +
> +/* Alias to Xen device tree helpers */
> +#define device_node dt_device_node
> +#define of_phandle_args dt_phandle_args
> +#define of_device_id dt_device_match
> +#define of_match_node dt_match_node
> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, pname, out))
> +#define of_property_read_bool dt_property_read_bool
> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
> +
> +/* Alias to Xen lock functions */
> +#define mutex spinlock
> +#define mutex_init spin_lock_init
> +#define mutex_lock spin_lock
> +#define mutex_unlock spin_unlock

Hmm... mutex are not spinlock. Can you explain why this is fine to 
switch to spinlock?

> +
> +/* Alias to Xen time functions */
> +#define ktime_t s_time_t
> +#define ktime_get()             (NOW())
> +#define ktime_add_us(t,i)       (t + MICROSECS(i))
> +#define ktime_compare(t,i)      (t > (i))
> +
> +/* Alias to Xen allocation helpers */
> +#define kzalloc(size, flags)    _xzalloc(size, sizeof(void *))
> +#define kfree xfree
> +#define devm_kzalloc(dev, size, flags)  _xzalloc(size, sizeof(void *))
> +
> +/* Device logger functions */
> +#define dev_name(dev) dt_node_full_name(dev->of_node)
> +#define dev_dbg(dev, fmt, ...)      \
> +    printk(XENLOG_DEBUG "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +#define dev_notice(dev, fmt, ...)   \
> +    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +#define dev_warn(dev, fmt, ...)     \
> +    printk(XENLOG_WARNING "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +#define dev_err(dev, fmt, ...)      \
> +    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +#define dev_info(dev, fmt, ...)     \
> +    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +#define dev_err_ratelimited(dev, fmt, ...)      \
> +    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
> +
> +/*
> + * Periodically poll an address and wait between reads in us until a
> + * condition is met or a timeout occurs.
> + */
> +#define readx_poll_timeout(op, addr, val, cond, sleep_us, timeout_us) \
> +({ \
> +     s_time_t deadline = NOW() + MICROSECS(timeout_us); \
> +     for (;;) { \
> +        (val) = op(addr); \
> +        if (cond) \
> +            break; \
> +        if (NOW() > deadline) { \
> +            (val) = op(addr); \
> +            break; \
> +        } \
> +        udelay(sleep_us); \
> +     } \
> +     (cond) ? 0 : -ETIMEDOUT; \
> +})
> +
> +#define readl_relaxed_poll_timeout(addr, val, cond, delay_us, timeout_us) \
> +    readx_poll_timeout(readl_relaxed, addr, val, cond, delay_us, timeout_us)
> +
> +#define FIELD_PREP(_mask, _val)         \
> +    (((typeof(_mask))(_val) << (__builtin_ffsll(_mask) - 1)) & (_mask))
> +
> +#define FIELD_GET(_mask, _reg)          \
> +    (typeof(_mask))(((_reg) & (_mask)) >> (__builtin_ffsll(_mask) - 1))
> +
> +#define WRITE_ONCE(x, val)                  \
> +do {                                        \
> +    *(volatile typeof(x) *)&(x) = (val);    \
> +} while (0)

Please implement it with write_atomic() or ACCESS_ONCE().

> +
> +/* Xen: Stub out DMA domain related functions */
> +#define iommu_get_dma_cookie(dom) 0
> +#define iommu_put_dma_cookie(dom)
> +
> +/*
> + * Helpers for DMA allocation. Just the function name is reused for
> + * porting code, these allocation are not managed allocations
>    */
> +static void *dmam_alloc_coherent(struct device *dev, size_t size,
> +                                 paddr_t *dma_handle, gfp_t gfp)
> +{
> +    void *vaddr;
> +    unsigned long alignment = size;
> +
> +    /*
> +     * _xzalloc requires that the (align & (align -1)) = 0. Most of the
> +     * allocations in SMMU code should send the right value for size. In
> +     * case this is not true print a warning and align to the size of a
> +     * (void *)
> +     */
> +    if ( size & (size - 1) )

We should use the same coding style within the file. As the file is 
imported from Linux, new code should follow Linux coding style.

> +    {
> +        printk(XENLOG_WARNING "SMMUv3: Fixing alignment for the DMA buffer\n");
> +        alignment = sizeof(void *);
> +    }
> +
> +    vaddr = _xzalloc(size, alignment);
> +    if ( !vaddr )
> +    {
> +        printk(XENLOG_ERR "SMMUv3: DMA allocation failed\n");
> +        return NULL;
> +    }
> +
> +    *dma_handle = virt_to_maddr(vaddr);
> +
> +    return vaddr;
> +}
> +
> +/* Xen: Type definitions for iommu_domain */
> +#define IOMMU_DOMAIN_UNMANAGED 0
> +#define IOMMU_DOMAIN_DMA 1
> +#define IOMMU_DOMAIN_IDENTITY 2
> +
> +/* Xen specific code. */
> +struct iommu_domain {
> +    /* Runtime SMMU configuration for this iommu_domain */
> +    atomic_t ref;
> +    /*
> +     * Used to link iommu_domain contexts for a same domain.
> +     * There is at least one per-SMMU to used by the domain.
> +     */
> +    struct list_head    list;
> +};
> +
> +/* Describes information required for a Xen domain */
> +struct arm_smmu_xen_domain {
> +    spinlock_t      lock;
> +
> +    /* List of iommu domains associated to this domain */
> +    struct list_head    contexts;
> +};
> +
> +/*
> + * Information about each device stored in dev->archdata.iommu
> + * The dev->archdata.iommu stores the iommu_domain (runtime configuration of
> + * the SMMU).
> + */
> +struct arm_smmu_xen_device {
> +    struct iommu_domain *domain;
> +};
> +
> +/* Keep a list of devices associated with this driver */
> +static DEFINE_SPINLOCK(arm_smmu_devices_lock);
> +static LIST_HEAD(arm_smmu_devices);
> +
> +
> +static inline void *dev_iommu_priv_get(struct device *dev)
> +{
> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> +
> +    return fwspec && fwspec->iommu_priv ? fwspec->iommu_priv : NULL;
> +}
> +
> +static inline void dev_iommu_priv_set(struct device *dev, void *priv)
> +{
> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> +
> +    fwspec->iommu_priv = priv;
> +}
> +
> +int dt_property_match_string(const struct dt_device_node *np,
> +                             const char *propname, const char *string)

I think this should be implemented in device_tree.c

> +{
> +    const struct dt_property *dtprop = dt_find_property(np, propname, NULL);
> +    size_t l;
> +    int i;
> +    const char *p, *end;
> +
> +    if ( !dtprop )
> +        return -EINVAL;
> +
> +    if ( !dtprop->value )
> +        return -ENODATA;
> +
> +    p = dtprop->value;
> +    end = p + dtprop->length;
> +
> +    for ( i = 0; p < end; i++, p += l )
> +    {
> +        l = strnlen(p, end - p) + 1;
> +
> +        if ( p + l > end )
> +            return -EILSEQ;
> +
> +        if ( strcmp(string, p) == 0 )
> +            return i; /* Found it; return index */
> +    }
> +
> +    return -ENODATA;
> +}
> +
> +static int platform_get_irq_byname_optional(struct device *dev,
> +                                            const char *name)
> +{
> +    int index, ret;
> +    struct dt_device_node *np  = dev_to_dt(dev);
> +
> +    if ( unlikely(!name) )
> +        return -EINVAL;
> +
> +    index = dt_property_match_string(np, "interrupt-names", name);
> +    if ( index < 0 )
> +    {
> +        dev_info(dev, "IRQ %s not found\n", name);
> +        return index;
> +    }
>   
> -#include <linux/acpi.h>
> -#include <linux/acpi_iort.h>
> -#include <linux/bitfield.h>
> -#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>
> -#include <linux/iommu.h>
> -#include <linux/iopoll.h>
> -#include <linux/module.h>
> -#include <linux/msi.h>
> -#include <linux/of.h>
> -#include <linux/of_address.h>
> -#include <linux/of_iommu.h>
> -#include <linux/of_platform.h>
> -#include <linux/pci.h>
> -#include <linux/pci-ats.h>
> -#include <linux/platform_device.h>
> -
> -#include <linux/amba/bus.h>
> +    ret = platform_get_irq(np, index);
> +    if ( ret < 0 )
> +    {
> +        dev_err(dev, "failed to get irq index %d\n", index);
> +        return -ENODEV;
> +    }
> +
> +    return ret;
> +}
> +
> +/* Start of Linux SMMUv3 code */
>   
>   /* MMIO registers */
>   #define ARM_SMMU_IDR0			0x0
> @@ -507,6 +751,7 @@ struct arm_smmu_s2_cfg {
>   	u16				vmid;
>   	u64				vttbr;
>   	u64				vtcr;
> +	struct domain		*domain;
>   };
>   
>   struct arm_smmu_strtab_cfg {
> @@ -567,8 +812,13 @@ struct arm_smmu_device {
>   
>   	struct arm_smmu_strtab_cfg	strtab_cfg;
>   
> -	/* IOMMU core code handle */
> -	struct iommu_device		iommu;
> +	/* Need to keep a list of SMMU devices */
> +	struct list_head		devices;
> +
> +	/* Tasklets for handling evts/faults and pci page request IRQs*/
> +	struct tasklet		evtq_irq_tasklet;
> +	struct tasklet		priq_irq_tasklet;
> +	struct tasklet		combined_irq_tasklet;
>   };
>   
>   /* SMMU private data for each master */
> @@ -1110,7 +1360,7 @@ static int arm_smmu_init_l2_strtab(struct arm_smmu_device *smmu, u32 sid)
>   }
>   
>   /* IRQ and event handlers */
> -static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
> +static void arm_smmu_evtq_thread(void *dev)
>   {
>   	int i;
>   	struct arm_smmu_device *smmu = dev;
> @@ -1140,7 +1390,6 @@ static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
>   	/* Sync our overflow flag, as we believe we're up to speed */
>   	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
>   		    Q_IDX(llq, llq->cons);
> -	return IRQ_HANDLED;
>   }
>   
>   static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
> @@ -1181,7 +1430,7 @@ static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
>   	}
>   }
>   
> -static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
> +static void arm_smmu_priq_thread(void *dev)
>   {
>   	struct arm_smmu_device *smmu = dev;
>   	struct arm_smmu_queue *q = &smmu->priq.q;
> @@ -1200,12 +1449,12 @@ static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
>   	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
>   		      Q_IDX(llq, llq->cons);
>   	queue_sync_cons_out(q);
> -	return IRQ_HANDLED;
>   }
>   
>   static int arm_smmu_device_disable(struct arm_smmu_device *smmu);
>   
> -static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
> +static void arm_smmu_gerror_handler(int irq, void *dev,
> +				struct cpu_user_regs *regs)
>   {
>   	u32 gerror, gerrorn, active;
>   	struct arm_smmu_device *smmu = dev;
> @@ -1215,7 +1464,7 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>   
>   	active = gerror ^ gerrorn;
>   	if (!(active & GERROR_ERR_MASK))
> -		return IRQ_NONE; /* No errors pending */
> +		return; /* No errors pending */
>   
>   	dev_warn(smmu->dev,
>   		 "unexpected global error reported (0x%08x), this could be serious\n",
> @@ -1248,26 +1497,42 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>   		arm_smmu_cmdq_skip_err(smmu);
>   
>   	writel(gerror, smmu->base + ARM_SMMU_GERRORN);
> -	return IRQ_HANDLED;
>   }
>   
> -static irqreturn_t arm_smmu_combined_irq_thread(int irq, void *dev)
> +static void arm_smmu_combined_irq_handler(int irq, void *dev,
> +				struct cpu_user_regs *regs)
> +{
> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;

The cast is not necessary.

> +
> +	arm_smmu_gerror_handler(irq, dev, regs);
> +
> +	tasklet_schedule(&(smmu->combined_irq_tasklet));
> +}
> +
> +static void arm_smmu_combined_irq_thread(void *dev)
>   {
>   	struct arm_smmu_device *smmu = dev;
>   
> -	arm_smmu_evtq_thread(irq, dev);
> +	arm_smmu_evtq_thread(dev);
>   	if (smmu->features & ARM_SMMU_FEAT_PRI)
> -		arm_smmu_priq_thread(irq, dev);
> -
> -	return IRQ_HANDLED;
> +		arm_smmu_priq_thread(dev);
>   }
>   
> -static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
> +static void arm_smmu_evtq_irq_tasklet(int irq, void *dev,
> +				struct cpu_user_regs *regs)
>   {
> -	arm_smmu_gerror_handler(irq, dev);
> -	return IRQ_WAKE_THREAD;
> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;

Ditto.

> +
> +	tasklet_schedule(&(smmu->evtq_irq_tasklet));
>   }
>   
> +static void arm_smmu_priq_irq_tasklet(int irq, void *dev,
> +				struct cpu_user_regs *regs)
> +{
> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;

Ditto.

> +
> +	tasklet_schedule(&(smmu->priq_irq_tasklet));
> +}
>   
>   /* IO_PGTABLE API */
>   static void arm_smmu_tlb_inv_context(void *cookie)
> @@ -1354,27 +1619,69 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>   }
>   
>   static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
> -				       struct arm_smmu_master *master,
> -				       struct io_pgtable_cfg *pgtbl_cfg)
> +				       struct arm_smmu_master *master)
>   {
>   	int vmid;
> +	u64 reg;

I think uint32_t is sufficient here.

>   	struct arm_smmu_device *smmu = smmu_domain->smmu;
>   	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>   
> +	/* VTCR */
> +	reg = VTCR_RES1 | VTCR_SH0_IS | VTCR_IRGN0_WBWA | VTCR_ORGN0_WBWA;

VTCR_RES1 will set bit 31 to 1. However, from the spec it looks like the 
equivalent bit in the entry (see 5.2 in ARM IHI 0070C.a) will be RES0.

> +
> +	switch (PAGE_SIZE) {
> +	case SZ_4K:
> +		reg |= VTCR_TG0_4K;
> +		break;
> +	case SZ_16K:
> +		reg |= VTCR_TG0_16K;
> +		break;
> +	case SZ_64K:
> +		reg |= VTCR_TG0_4K;
> +		break;
> +	}

I would just handle 4K here and add a BUILD_BUG_ON(PAGE_SIZE != SZ_4K).

> + > +	switch (smmu->oas) {

AFAICT smmu->oas and ...

> +	case 32:
> +		reg |= VTCR_PS(_AC(0x0,ULL));
> +		break;
> +	case 36:
> +		reg |= VTCR_PS(_AC(0x1,ULL));
> +		break;
> +	case 40:
> +		reg |= VTCR_PS(_AC(0x2,ULL));
> +		break;
> +	case 42:
> +		reg |= VTCR_PS(_AC(0x3,ULL));
> +		break;
> +	case 44:
> +		reg |= VTCR_PS(_AC(0x4,ULL));
> +		break;
> +		case 48:
> +		reg |= VTCR_PS(_AC(0x5,ULL));
> +		break;
> +	case 52:
> +		reg |= VTCR_PS(_AC(0x6,ULL));
> +		break;
> +	}
> +
> +	reg |= VTCR_T0SZ(64ULL - smmu->ias);

... are directly taken from the SMMU configuration. However, as we share 
the P2M, we need to make sure the value match what the CPU is using.

For the IAS, you will want to use p2m_ipa_bits and for the output, we 
will want to cap to PADDR_BITS.

> +	reg |= VTCR_SL0(0x2);

Similar to above, the starting level will depend on how the P2M was 
configured.

> +	reg |= VTCR_VS;

AFAICT, the bit 179 (bit 19 in the word) is indicating whether AArch32 
or AArch64 translation table is used. However, bit 19 in VTCR_EL2 
indicates whether we are using 8-bit or 16-bit VMID.

> +
> +	cfg->vtcr   = reg;

It would be better to initialize vtcr exactly at the same place as Linux 
does. This would make easier to match the code.
> +
>   	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
>   	if (vmid < 0)
>   		return vmid;
> +	cfg->vmid  = (u16)vmid;
> +
> +	cfg->vttbr  = page_to_maddr(cfg->domain->arch.p2m.root);
> +
> +	printk(XENLOG_DEBUG
> +		   "SMMUv3: d%u: vmid 0x%x vtcr 0x%"PRIpaddr" p2maddr 0x%"PRIpaddr"\n",
> +		   cfg->domain->domain_id, cfg->vmid, cfg->vtcr, cfg->vttbr);
>   
> -	vtcr = &pgtbl_cfg->arm_lpae_s2_cfg.vtcr;
> -	cfg->vmid	= (u16)vmid;
> -	cfg->vttbr	= pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
> -	cfg->vtcr	= FIELD_PREP(STRTAB_STE_2_VTCR_S2T0SZ, vtcr->tsz) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SL0, vtcr->sl) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2IR0, vtcr->irgn) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2OR0, vtcr->orgn) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SH0, vtcr->sh) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2TG, vtcr->tg) |
> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2PS, vtcr->ps);
>   	return 0;
>   }
>   
> @@ -1382,28 +1689,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>   				    struct arm_smmu_master *master)
>   {
>   	int ret;
> -	unsigned long ias, oas;
> -	int (*finalise_stage_fn)(struct arm_smmu_domain *,
> -				 struct arm_smmu_master *,
> -				 struct io_pgtable_cfg *);
>   	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>   
>   	/* Restrict the stage to what we can actually support */
>   	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
>   
> -	switch (smmu_domain->stage) {
> -	case ARM_SMMU_DOMAIN_NESTED:
> -	case ARM_SMMU_DOMAIN_S2:
> -		ias = smmu->ias;
> -		oas = smmu->oas;
> -		finalise_stage_fn = arm_smmu_domain_finalise_s2;
> -		break;
> -	default:
> -		return -EINVAL;
> -	}
> -
> -	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);

It is not entirely clear why this code is removed here and not the 
previous patch?

> +	ret = arm_smmu_domain_finalise_s2(smmu_domain, master);
>   	if (ret < 0) {
>   		return ret;
>   	}
> @@ -1553,7 +1844,8 @@ static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>   		return -ENOMEM;
>   	}
>   
> -	if (!WARN_ON(q->base_dma & (qsz - 1))) {
> +	WARN_ON(q->base_dma & (qsz - 1));

This is a call to rework how our WARN_ON() in Xen.

> +	if (unlikely(q->base_dma & (qsz - 1))) {
>   		dev_info(smmu->dev, "allocated %u entries for %s\n",
>   			 1 << q->llq.max_n_shift, name);
>   	}
> @@ -1758,9 +2050,7 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>   	/* Request interrupt lines */
>   	irq = smmu->evtq.q.irq;
>   	if (irq) {
> -		ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
> -						arm_smmu_evtq_thread,
> -						IRQF_ONESHOT,
> +		ret = request_irq(irq, 0, arm_smmu_evtq_irq_tasklet,
>   						"arm-smmu-v3-evtq", smmu);
>   		if (ret < 0)
>   			dev_warn(smmu->dev, "failed to enable evtq irq\n");
> @@ -1770,8 +2060,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>   
>   	irq = smmu->gerr_irq;
>   	if (irq) {
> -		ret = devm_request_irq(smmu->dev, irq, arm_smmu_gerror_handler,
> -				       0, "arm-smmu-v3-gerror", smmu);
> +		ret = request_irq(irq, 0, arm_smmu_gerror_handler,
> +						"arm-smmu-v3-gerror", smmu);
>   		if (ret < 0)
>   			dev_warn(smmu->dev, "failed to enable gerror irq\n");
>   	} else {
> @@ -1781,11 +2071,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>   	if (smmu->features & ARM_SMMU_FEAT_PRI) {
>   		irq = smmu->priq.q.irq;
>   		if (irq) {
> -			ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
> -							arm_smmu_priq_thread,
> -							IRQF_ONESHOT,
> -							"arm-smmu-v3-priq",
> -							smmu);
> +			ret = request_irq(irq, 0, arm_smmu_priq_irq_tasklet,
> +							"arm-smmu-v3-priq", smmu);
>   			if (ret < 0)
>   				dev_warn(smmu->dev,
>   					 "failed to enable priq irq\n");
> @@ -1814,11 +2101,8 @@ static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
>   		 * Cavium ThunderX2 implementation doesn't support unique irq
>   		 * lines. Use a single irq line for all the SMMUv3 interrupts.
>   		 */
> -		ret = devm_request_threaded_irq(smmu->dev, irq,
> -					arm_smmu_combined_irq_handler,
> -					arm_smmu_combined_irq_thread,
> -					IRQF_ONESHOT,
> -					"arm-smmu-v3-combined-irq", smmu);
> +		ret = request_irq(irq, 0, arm_smmu_combined_irq_handler,
> +						"arm-smmu-v3-combined-irq", smmu);
>   		if (ret < 0)
>   			dev_warn(smmu->dev, "failed to enable combined irq\n");
>   	} else
> @@ -1857,7 +2141,7 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>   	reg = readl_relaxed(smmu->base + ARM_SMMU_CR0);
>   	if (reg & CR0_SMMUEN) {
>   		dev_warn(smmu->dev, "SMMU currently enabled! Resetting...\n");
> -		WARN_ON(is_kdump_kernel() && !disable_bypass);
> +		WARN_ON(!disable_bypass);
>   		arm_smmu_update_gbpa(smmu, GBPA_ABORT, 0);
>   	}
>   
> @@ -1952,8 +2236,11 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>   		return ret;
>   	}
>   
> -	if (is_kdump_kernel())
> -		enables &= ~(CR0_EVTQEN | CR0_PRIQEN);
> +	/* Initialize tasklets for threaded IRQs*/
> +	tasklet_init(&smmu->evtq_irq_tasklet, arm_smmu_evtq_thread, smmu);
> +	tasklet_init(&smmu->priq_irq_tasklet, arm_smmu_priq_thread, smmu);
> +	tasklet_init(&smmu->combined_irq_tasklet, arm_smmu_combined_irq_thread,
> +				 smmu);
>   
>   	/* Enable the SMMU interface, or ensure bypass */
>   	if (!bypass || disable_bypass) {
> @@ -2195,7 +2482,7 @@ static inline int arm_smmu_device_acpi_probe(struct platform_device *pdev,
>   static int arm_smmu_device_dt_probe(struct platform_device *pdev,
>   				    struct arm_smmu_device *smmu)
>   {
> -	struct device *dev = &pdev->dev;
> +	struct device *dev = pdev;
>   	u32 cells;
>   	int ret = -EINVAL;
>   
> @@ -2219,130 +2506,449 @@ static unsigned long arm_smmu_resource_size(struct arm_smmu_device *smmu)
>   		return SZ_128K;
>   }
>   
> +/* Start of Xen specific code. */
>   static int arm_smmu_device_probe(struct platform_device *pdev)
>   {
> -	int irq, ret;
> -	struct resource *res;
> -	resource_size_t ioaddr;
> -	struct arm_smmu_device *smmu;
> -	struct device *dev = &pdev->dev;
> -	bool bypass;
> +    int irq, ret;
> +    paddr_t ioaddr, iosize;
> +    struct arm_smmu_device *smmu;
> +    bool bypass;
> +
> +    smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
> +    if ( !smmu )
> +    {
> +        dev_err(pdev, "failed to allocate arm_smmu_device\n");
> +        return -ENOMEM;
> +    }
> +    smmu->dev = pdev;
> +
> +    if ( pdev->of_node )
> +    {
> +        ret = arm_smmu_device_dt_probe(pdev, smmu);
> +    } else
> +    {
> +        ret = arm_smmu_device_acpi_probe(pdev, smmu);
> +        if ( ret == -ENODEV )
> +            return ret;
> +    }
> +
> +    /* Set bypass mode according to firmware probing result */
> +    bypass = !!ret;

AFAICT, bypass would be set if the device-tree is buggy. For Xen, I 
think it would be saner to not continue as this would break isolation.

> +
> +    /* Base address */
> +    ret = dt_device_get_address(dev_to_dt(pdev), 0, &ioaddr, &iosize);
> +    if( ret )
> +        return -ENODEV;
> +
> +    if ( iosize < arm_smmu_resource_size(smmu) )
> +    {
> +        dev_err(pdev, "MMIO region too small (%lx)\n", iosize);
> +        return -EINVAL;
> +    }
> +
> +    /*
> +     * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
> +     * the PMCG registers which are reserved by the PMU driver.
> +     */

This comment doesn't seem to match the code. Did you intend to...


> +    smmu->base = ioremap_nocache(ioaddr, iosize);

... use ARM_SMMU_REG_SZ which would only map the necessary region?

> +    if ( IS_ERR(smmu->base) )
> +        return PTR_ERR(smmu->base);
> +
> +    if ( iosize > SZ_64K )
> +    {
> +        smmu->page1 = ioremap_nocache(ioaddr + SZ_64K, ARM_SMMU_REG_SZ);
> +        if ( IS_ERR(smmu->page1) )
> +            return PTR_ERR(smmu->page1);
> +    }
> +    else
> +    {
> +        smmu->page1 = smmu->base;
> +    }
> +
> +    /* Interrupt lines */
> +
> +    irq = platform_get_irq_byname_optional(pdev, "combined");
> +    if ( irq > 0 )
> +        smmu->combined_irq = irq;
> +    else
> +    {
> +        irq = platform_get_irq_byname_optional(pdev, "eventq");
> +        if ( irq > 0 )
> +            smmu->evtq.q.irq = irq;
> +
> +        irq = platform_get_irq_byname_optional(pdev, "priq");
> +        if ( irq > 0 )
> +            smmu->priq.q.irq = irq;
> +
> +        irq = platform_get_irq_byname_optional(pdev, "gerror");
> +        if ( irq > 0 )
> +            smmu->gerr_irq = irq;
> +    }
> +    /* Probe the h/w */
> +    ret = arm_smmu_device_hw_probe(smmu);
> +    if ( ret )
> +        return ret;
> +
> +    /* Initialise in-memory data structures */
> +    ret = arm_smmu_init_structures(smmu);
> +    if ( ret )
> +        return ret;
> +
> +    /* Reset the device */
> +    ret = arm_smmu_device_reset(smmu, bypass);
> +    if ( ret )
> +        return ret;
> +
> +    /*
> +     * Keep a list of all probed devices. This will be used to query
> +     * the smmu devices based on the fwnode.
> +     */
> +    INIT_LIST_HEAD(&smmu->devices);
> +
> +    spin_lock(&arm_smmu_devices_lock);
> +    list_add(&smmu->devices, &arm_smmu_devices);
> +    spin_unlock(&arm_smmu_devices_lock);
> +
> +    return 0;
> +}
> +
> +static int __must_check arm_smmu_iotlb_flush_all(struct domain *d)
> +{
> +    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
> +    struct iommu_domain *io_domain;
> +
> +    spin_lock(&xen_domain->lock);
> +
> +    list_for_each_entry( io_domain, &xen_domain->contexts, list )
> +    {
> +        /*
> +         * Only invalidate the context when SMMU is present.
> +         * This is because the context initialization is delayed
> +         * until a master has been added.
> +         */
> +        if ( unlikely(!ACCESS_ONCE(to_smmu_domain(io_domain)->smmu)) )
> +            continue;
> +
> +        arm_smmu_tlb_inv_context(to_smmu_domain(io_domain));
> +    }
>   
> -	smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
> -	if (!smmu) {
> -		dev_err(dev, "failed to allocate arm_smmu_device\n");
> -		return -ENOMEM;
> -	}
> -	smmu->dev = dev;
> +    spin_unlock(&xen_domain->lock);
> +
> +    return 0;
> +}
> +
> +static int __must_check arm_smmu_iotlb_flush(struct domain *d, dfn_t dfn,
> +                                             unsigned long page_count,
> +                                             unsigned int flush_flags)
> +{
> +    return arm_smmu_iotlb_flush_all(d);
> +}
>   
> -	if (dev->of_node) {
> -		ret = arm_smmu_device_dt_probe(pdev, smmu);
> -	} else {
> -		ret = arm_smmu_device_acpi_probe(pdev, smmu);
> -		if (ret == -ENODEV)
> -			return ret;
> -	}
> +static struct arm_smmu_device *arm_smmu_get_by_dev(struct device *dev)
> +{
> +    struct arm_smmu_device *smmu = NULL;
>   
> -	/* Set bypass mode according to firmware probing result */
> -	bypass = !!ret;
> +    spin_lock(&arm_smmu_devices_lock);
> +    list_for_each_entry( smmu, &arm_smmu_devices, devices )
> +    {
> +        if ( smmu->dev  == dev )
> +        {
> +            spin_unlock(&arm_smmu_devices_lock);
> +            return smmu;
> +        }
> +    }
> +    spin_unlock(&arm_smmu_devices_lock);
>   
> -	/* Base address */
> -	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> -	if (resource_size(res) < arm_smmu_resource_size(smmu)) {
> -		dev_err(dev, "MMIO region too small (%pr)\n", res);
> -		return -EINVAL;
> -	}
> -	ioaddr = res->start;

The code removed is basically the same as the one you added except the 
coding style change. This patch is already quite long to review, so can 
we please keep the change to the strict minimum?

If you want to do to clean-up then they should be done before/after.

> +    return NULL;
> +}
>   
> -	/*
> -	 * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
> -	 * the PMCG registers which are reserved by the PMU driver.
> -	 */
> -	smmu->base = arm_smmu_ioremap(dev, ioaddr, ARM_SMMU_REG_SZ);
> -	if (IS_ERR(smmu->base))
> -		return PTR_ERR(smmu->base);
> -
> -	if (arm_smmu_resource_size(smmu) > SZ_64K) {
> -		smmu->page1 = arm_smmu_ioremap(dev, ioaddr + SZ_64K,
> -					       ARM_SMMU_REG_SZ);
> -		if (IS_ERR(smmu->page1))
> -			return PTR_ERR(smmu->page1);
> -	} else {
> -		smmu->page1 = smmu->base;
> -	}
> +/* Probing and initialisation functions */
> +static struct iommu_domain *arm_smmu_get_domain(struct domain *d,
> +                                                struct device *dev)
> +{
> +    struct iommu_domain *io_domain;
> +    struct arm_smmu_domain *smmu_domain;
> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> +    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
> +    struct arm_smmu_device *smmu = arm_smmu_get_by_dev(fwspec->iommu_dev);
> +
> +    if ( !smmu )
> +        return NULL;
>   
> -	/* Interrupt lines */
> +    /*
> +     * Loop through the &xen_domain->contexts to locate a context
> +     * assigned to this SMMU
> +     */
> +    list_for_each_entry( io_domain, &xen_domain->contexts, list )
> +    {
> +        smmu_domain = to_smmu_domain(io_domain);
> +        if ( smmu_domain->smmu == smmu )
> +            return io_domain;
> +    }
>   
> -	irq = platform_get_irq_byname_optional(pdev, "combined");
> -	if (irq > 0)
> -		smmu->combined_irq = irq;
> -	else {
> -		irq = platform_get_irq_byname_optional(pdev, "eventq");
> -		if (irq > 0)
> -			smmu->evtq.q.irq = irq;
> +    return NULL;
> +}
>   
> -		irq = platform_get_irq_byname_optional(pdev, "priq");
> -		if (irq > 0)
> -			smmu->priq.q.irq = irq;
> +static void arm_smmu_destroy_iommu_domain(struct iommu_domain *io_domain)
> +{
> +    list_del(&io_domain->list);
> +    arm_smmu_domain_free(io_domain);
> +}
>   
> -		irq = platform_get_irq_byname_optional(pdev, "gerror");
> -		if (irq > 0)
> -			smmu->gerr_irq = irq;
> -	}
> -	/* Probe the h/w */
> -	ret = arm_smmu_device_hw_probe(smmu);
> -	if (ret)
> -		return ret;
> +static int arm_smmu_assign_dev(struct domain *d, u8 devfn,
> +                               struct device *dev, u32 flag)
> +{
> +    int ret = 0;
> +    struct iommu_domain *io_domain;
> +    struct arm_smmu_domain *smmu_domain;
> +    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
>   
> -	/* Initialise in-memory data structures */
> -	ret = arm_smmu_init_structures(smmu);
> -	if (ret)
> -		return ret;
> +    if ( !dev->archdata.iommu )
> +    {
> +        dev->archdata.iommu = xzalloc(struct arm_smmu_xen_device);
> +        if ( !dev->archdata.iommu )
> +            return -ENOMEM;
> +    }
>   
> -	/* Record our private device structure */
> -	platform_set_drvdata(pdev, smmu);
> +    spin_lock(&xen_domain->lock);
>   
> -	/* Reset the device */
> -	ret = arm_smmu_device_reset(smmu, bypass);
> -	if (ret)
> -		return ret;
> +    /*
> +     * Check to see if an iommu_domain already exists for this xen domain
> +     * under the same SMMU
> +     */
> +    io_domain = arm_smmu_get_domain(d, dev);
> +    if ( !io_domain )
> +    {
> +        io_domain = arm_smmu_domain_alloc(IOMMU_DOMAIN_DMA);
> +        if ( !io_domain )
> +        {
> +            ret = -ENOMEM;
> +            goto out;
> +        }
>   
> -	/* And we're up. Go go go! */
> -	ret = iommu_device_sysfs_add(&smmu->iommu, dev, NULL,
> -				     "smmu3.%pa", &ioaddr);
> -	if (ret)
> -		return ret;
> +        smmu_domain = to_smmu_domain(io_domain);
> +        smmu_domain->s2_cfg.domain = d;
>   
> -	iommu_device_set_ops(&smmu->iommu, &arm_smmu_ops);
> -	iommu_device_set_fwnode(&smmu->iommu, dev->fwnode);
> +        /* Chain the new context to the domain */
> +        list_add(&io_domain->list, &xen_domain->contexts);
>   
> -	ret = iommu_device_register(&smmu->iommu);
> -	if (ret) {
> -		dev_err(dev, "Failed to register iommu\n");
> -		return ret;
> -	}
> +    }
> +
> +    ret = arm_smmu_attach_dev(io_domain, dev);
> +    if ( ret )
> +    {
> +        if ( io_domain->ref.counter == 0 )
> +            arm_smmu_destroy_iommu_domain(io_domain);
> +    }
> +    else
> +    {
> +        atomic_inc(&io_domain->ref);
> +    }
>   
> -	return arm_smmu_set_bus_ops(&arm_smmu_ops);
> +out:
> +    spin_unlock(&xen_domain->lock);
> +    return ret;
>   }
>   
> -static int arm_smmu_device_remove(struct platform_device *pdev)
> +static int arm_smmu_deassign_dev(struct domain *d, struct device *dev)
>   {
> -	struct arm_smmu_device *smmu = platform_get_drvdata(pdev);
> +    struct iommu_domain *io_domain = arm_smmu_get_domain(d, dev);
> +    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
> +    struct arm_smmu_domain *arm_smmu = to_smmu_domain(io_domain);
> +    struct arm_smmu_master *master = dev_iommu_priv_get(dev);
>   
> -	arm_smmu_set_bus_ops(NULL);
> -	iommu_device_unregister(&smmu->iommu);
> -	iommu_device_sysfs_remove(&smmu->iommu);
> -	arm_smmu_device_disable(smmu);
> +    if ( !arm_smmu || arm_smmu->s2_cfg.domain != d )
> +    {
> +        dev_err(dev, " not attached to domain %d\n", d->domain_id);
> +        return -ESRCH;
> +    }
>   
> -	return 0;
> +    spin_lock(&xen_domain->lock);
> +
> +    arm_smmu_detach_dev(master);
> +    atomic_dec(&io_domain->ref);
> +
> +    if ( io_domain->ref.counter == 0 )
> +        arm_smmu_destroy_iommu_domain(io_domain);
> +
> +    spin_unlock(&xen_domain->lock);
> +
> +    return 0;
> +}
> +
> +static int arm_smmu_reassign_dev(struct domain *s, struct domain *t,
> +                                 u8 devfn,  struct device *dev)
> +{
> +    int ret = 0;
> +
> +    /* Don't allow remapping on other domain than hwdom */
> +    if ( t && t != hardware_domain )
> +        return -EPERM;
> +
> +    if ( t == s )
> +        return 0;
> +
> +    ret = arm_smmu_deassign_dev(s, dev);
> +    if ( ret )
> +        return ret;
> +
> +    if ( t )
> +    {
> +        /* No flags are defined for ARM. */
> +        ret = arm_smmu_assign_dev(t, devfn, dev, 0);
> +        if ( ret )
> +            return ret;
> +    }
> +
> +    return 0;
> +}
> +
> +static int arm_smmu_iommu_xen_domain_init(struct domain *d)
> +{
> +    struct arm_smmu_xen_domain *xen_domain;
> +
> +    xen_domain = xzalloc(struct arm_smmu_xen_domain);
> +    if ( !xen_domain )
> +        return -ENOMEM;
> +
> +    spin_lock_init(&xen_domain->lock);
> +    INIT_LIST_HEAD(&xen_domain->contexts);
> +
> +    dom_iommu(d)->arch.priv = xen_domain;
> +
> +    return 0;
> +}
> +
> +static void __hwdom_init arm_smmu_iommu_hwdom_init(struct domain *d)
> +{
> +}
> +
> +static void arm_smmu_iommu_xen_domain_teardown(struct domain *d)
> +{
> +    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
> +
> +    ASSERT(list_empty(&xen_domain->contexts));
> +    xfree(xen_domain);
> +}
> +
> +static int arm_smmu_dt_xlate(struct device *dev,
> +                             const struct dt_phandle_args *args)
> +{
> +    int ret;
> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
> +
> +    ret = iommu_fwspec_add_ids(dev, args->args, 1);
> +    if ( ret )
> +        return ret;
> +
> +    if ( dt_device_is_protected(dev_to_dt(dev)) )
> +    {
> +        dev_err(dev, "Already added to SMMUv3\n");
> +        return -EEXIST;
> +    }
> +
> +    /* Let Xen know that the master device is protected by an IOMMU. */
> +    dt_device_set_protected(dev_to_dt(dev));
> +
> +    dev_info(dev, "Added master device (SMMUv3 %s StreamIds %u)\n",
> +            dev_name(fwspec->iommu_dev), fwspec->num_ids);
> +
> +    return 0;
>   }
>   
> -static void arm_smmu_device_shutdown(struct platform_device *pdev)

I think this function should have been dropped in the previous patch.

> +static int arm_smmu_add_device(u8 devfn, struct device *dev)
>   {
> -	arm_smmu_device_remove(pdev);
> +    int i, ret;
> +    struct arm_smmu_device *smmu;
> +    struct arm_smmu_master *master;
> +    struct iommu_fwspec *fwspec;
> +
> +    fwspec = dev_iommu_fwspec_get(dev);
> +    if ( !fwspec )
> +        return -ENODEV;
> +
> +    smmu = arm_smmu_get_by_dev(fwspec->iommu_dev);
> +    if ( !smmu )
> +        return -ENODEV;
> +
> +    master = xzalloc(struct arm_smmu_master);
> +    if ( !master )
> +        return -ENOMEM;
> +
> +    master->dev = dev;
> +    master->smmu = smmu;
> +    master->sids = fwspec->ids;
> +    master->num_sids = fwspec->num_ids;
> +
> +    dev_iommu_priv_set(dev, master);
> +
> +    /* Check the SIDs are in range of the SMMU and our stream table */
> +    for ( i = 0; i < master->num_sids; i++ )
> +    {
> +        u32 sid = master->sids[i];
> +
> +        if ( !arm_smmu_sid_in_range(smmu, sid) )
> +        {
> +            ret = -ERANGE;
> +            goto err_free_master;
> +        }
> +
> +        /* Ensure l2 strtab is initialised */
> +        if ( smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB )
> +        {
> +            ret = arm_smmu_init_l2_strtab(smmu, sid);
> +            if ( ret )
> +                goto err_free_master;
> +        }
> +    }
> +
> +    return 0;
> +
> +err_free_master:
> +    xfree(master);
> +    dev_iommu_priv_set(dev, NULL);
> +    return ret;
>   }
>   
> -static const struct of_device_id arm_smmu_of_match[] = {
> -	{ .compatible = "arm,smmu-v3", },
> -	{ },
> +static const struct iommu_ops arm_smmu_iommu_ops = {
> +    .init = arm_smmu_iommu_xen_domain_init,
> +    .hwdom_init = arm_smmu_iommu_hwdom_init,
> +    .teardown = arm_smmu_iommu_xen_domain_teardown,
> +    .iotlb_flush = arm_smmu_iotlb_flush,
> +    .iotlb_flush_all = arm_smmu_iotlb_flush_all,
> +    .assign_device = arm_smmu_assign_dev,
> +    .reassign_device = arm_smmu_reassign_dev,
> +    .map_page = arm_iommu_map_page,
> +    .unmap_page = arm_iommu_unmap_page,
> +    .dt_xlate = arm_smmu_dt_xlate,
> +    .add_device = arm_smmu_add_device,
> +};
> +
> +static const struct dt_device_match arm_smmu_of_match[] = {
> +    { .compatible = "arm,smmu-v3", },
> +    { },
>   };
> +
> +static __init int arm_smmu_dt_init(struct dt_device_node *dev,
> +                                   const void *data)
> +{
> +    int rc;
> +
> +    /*
> +     * Even if the device can't be initialized, we don't want to
> +     * give the SMMU device to dom0.
> +     */
> +    dt_device_set_used_by(dev, DOMID_XEN);
> +
> +    rc = arm_smmu_device_probe(dt_to_dev(dev));
> +    if ( rc )
> +        return rc;
> +
> +    iommu_set_ops(&arm_smmu_iommu_ops);
> +    return 0;
> +}
> +
> +DT_DEVICE_START(smmuv3, "ARM SMMU V3", DEVICE_IOMMU)
> +    .dt_match = arm_smmu_of_match,
> +    .init = arm_smmu_dt_init,
> +DT_DEVICE_END
> 

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-02  2:51   ` Stefano Stabellini
@ 2020-12-02 16:27     ` Rahul Singh
  2020-12-02 19:26       ` Rahul Singh
  2020-12-02 16:47     ` Julien Grall
  1 sibling, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-12-02 16:27 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: xen-devel, Bertrand Marquis, Andrew Cooper, George Dunlap,
	Ian Jackson, Jan Beulich, Julien Grall, Wei Liu, Paul Durrant,
	Volodymyr Babchuk

Hello Stefano,

> On 2 Dec 2020, at 2:51 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
> 
> On Thu, 26 Nov 2020, Rahul Singh wrote:
>> Add support for ARM architected SMMUv3 implementation. It is based on
>> the Linux SMMUv3 driver.
>> 
>> Major differences with regard to Linux driver are as follows:
>> 1. Only Stage-2 translation is supported as compared to the Linux driver
>>   that supports both Stage-1 and Stage-2 translations.
>> 2. Use P2M  page table instead of creating one as SMMUv3 has the
>>   capability to share the page tables with the CPU.
>> 3. Tasklets are used in place of threaded IRQ's in Linux for event queue
>>   and priority queue IRQ handling.
>> 4. Latest version of the Linux SMMUv3 code implements the commands queue
>>   access functions based on atomic operations implemented in Linux.
>>   Atomic functions used by the commands queue access functions are not
>>   implemented in XEN therefore we decided to port the earlier version
>>   of the code. Once the proper atomic operations will be available in
>>   XEN the driver can be updated.
>> 5. Driver is currently supported as Tech Preview.
> 
> This patch is big and was difficult to review, nonetheless I tried :-)

Thanks again for reviewing the code.
> 
> That said, the code is self-contained, marked as Tech Preview, and not
> at risk of making other things unstable, so it is low risk to accept it.
> 
> Comments below.
> 
> 
>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
>> ---
>> MAINTAINERS                           |   6 +
>> SUPPORT.md                            |   1 +
>> xen/drivers/passthrough/Kconfig       |  10 +
>> xen/drivers/passthrough/arm/Makefile  |   1 +
>> xen/drivers/passthrough/arm/smmu-v3.c | 986 +++++++++++++++++++++-----
>> 5 files changed, 814 insertions(+), 190 deletions(-)
>> 
>> diff --git a/MAINTAINERS b/MAINTAINERS
>> index dab38a6a14..1d63489eec 100644
>> --- a/MAINTAINERS
>> +++ b/MAINTAINERS
>> @@ -249,6 +249,12 @@ F:	xen/include/asm-arm/
>> F:	xen/include/public/arch-arm/
>> F:	xen/include/public/arch-arm.h
>> 
>> +ARM SMMUv3
>> +M:	Bertrand Marquis <bertrand.marquis@arm.com>
>> +M:	Rahul Singh <rahul.singh@arm.com>
>> +S:	Supported
>> +F:	xen/drivers/passthrough/arm/smmu-v3.c
>> +
>> Change Log
>> M:	Paul Durrant <paul@xen.org>
>> R:	Community Manager <community.manager@xenproject.org>
>> diff --git a/SUPPORT.md b/SUPPORT.md
>> index ab02aca5f4..e402c7202d 100644
>> --- a/SUPPORT.md
>> +++ b/SUPPORT.md
>> @@ -68,6 +68,7 @@ For the Cortex A57 r0p0 - r1p1, see Errata 832075.
>>     Status, ARM SMMUv1: Supported, not security supported
>>     Status, ARM SMMUv2: Supported, not security supported
>>     Status, Renesas IPMMU-VMSA: Supported, not security supported
>> +    Status, ARM SMMUv3: Tech Preview
>> 
>> ### ARM/GICv3 ITS
>> 
>> diff --git a/xen/drivers/passthrough/Kconfig b/xen/drivers/passthrough/Kconfig
>> index 0036007ec4..5b71c59f47 100644
>> --- a/xen/drivers/passthrough/Kconfig
>> +++ b/xen/drivers/passthrough/Kconfig
>> @@ -13,6 +13,16 @@ config ARM_SMMU
>> 	  Say Y here if your SoC includes an IOMMU device implementing the
>> 	  ARM SMMU architecture.
>> 
>> +config ARM_SMMU_V3
>> +	bool "ARM Ltd. System MMU Version 3 (SMMUv3) Support" if EXPERT
>> +	depends on ARM_64
>> +	---help---
>> +	 Support for implementations of the ARM System MMU architecture
>> +	 version 3.
>> +
>> +	 Say Y here if your system includes an IOMMU device implementing
>> +	 the ARM SMMUv3 architecture.
>> +
>> config IPMMU_VMSA
>> 	bool "Renesas IPMMU-VMSA found in R-Car Gen3 SoCs"
>> 	depends on ARM_64
>> diff --git a/xen/drivers/passthrough/arm/Makefile b/xen/drivers/passthrough/arm/Makefile
>> index fcd918ea3e..c5fb3b58a5 100644
>> --- a/xen/drivers/passthrough/arm/Makefile
>> +++ b/xen/drivers/passthrough/arm/Makefile
>> @@ -1,3 +1,4 @@
>> obj-y += iommu.o iommu_helpers.o iommu_fwspec.o
>> obj-$(CONFIG_ARM_SMMU) += smmu.o
>> obj-$(CONFIG_IPMMU_VMSA) += ipmmu-vmsa.o
>> +obj-$(CONFIG_ARM_SMMU_V3) += smmu-v3.o
>> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
>> index 55d1cba194..8f2337e7f2 100644
>> --- a/xen/drivers/passthrough/arm/smmu-v3.c
>> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
>> @@ -2,36 +2,280 @@
>> /*
>>  * IOMMU API for ARM architected SMMUv3 implementations.
>>  *
>> - * Copyright (C) 2015 ARM Limited
>> + * Based on Linux's SMMUv3 driver:
>> + *    drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
>> + *    commit: 951cbbc386ff01b50da4f46387e994e81d9ab431
>> + * and Xen's SMMU driver:
>> + *    xen/drivers/passthrough/arm/smmu.c
>>  *
>> - * Author: Will Deacon <will.deacon@arm.com>
>> + * Copyright (C) 2015 ARM Limited Will Deacon <will.deacon@arm.com>
>>  *
>> - * This driver is powered by bad coffee and bombay mix.
>> + * Copyright (C) 2020 Arm Ltd.
>> + *
>> + * 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, see <http://www.gnu.org/licenses/>.
>> + *
>> + */
>> +
>> +#include <xen/acpi.h>
>> +#include <xen/config.h>
>> +#include <xen/delay.h>
>> +#include <xen/errno.h>
>> +#include <xen/err.h>
>> +#include <xen/irq.h>
>> +#include <xen/lib.h>
>> +#include <xen/list.h>
>> +#include <xen/mm.h>
>> +#include <xen/rbtree.h>
>> +#include <xen/sched.h>
>> +#include <xen/sizes.h>
>> +#include <xen/vmap.h>
>> +#include <asm/atomic.h>
>> +#include <asm/device.h>
>> +#include <asm/io.h>
>> +#include <asm/platform.h>
>> +#include <asm/iommu_fwspec.h>
>> +
>> +/* Linux compatibility functions. */
>> +typedef paddr_t dma_addr_t;
>> +typedef unsigned int gfp_t;
>> +
>> +#define platform_device device
>> +
>> +#define GFP_KERNEL 0
>> +
>> +/* Alias to Xen device tree helpers */
>> +#define device_node dt_device_node
>> +#define of_phandle_args dt_phandle_args
>> +#define of_device_id dt_device_match
>> +#define of_match_node dt_match_node
>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, pname, out))
>> +#define of_property_read_bool dt_property_read_bool
>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
> 
> Given all the changes to the file by the previous patches we are
> basically fully (or almost fully) adapting this code to Xen.
> 
> So at that point I wonder if we should just as well make these changes
> (e.g. s/of_phandle_args/dt_phandle_args/g) to the code too.

Yes, you are right that is why in the first version of the patch I modifying the code to make it fully XEN compatible. But in this patch series I use the Linux compatibility function to have Linux code unmodified.I also prefer to make changes (s/of_phandle_args/dt_phandle_args/g).

> 
> Julien, Rahul, what do you guys think? 
> 
> 
>> +/* Alias to Xen lock functions */
> 
> I think this deserves at least one statement to explain why it is OK.

Ack. I Will add the comment.
> 
> Also, a similar comment to the one above: maybe we should add a couple
> of more preparation patches to s/mutex/spinlock/g and change the time
> and allocation functions too.
> 
> 
>> +#define mutex spinlock
>> +#define mutex_init spin_lock_init
>> +#define mutex_lock spin_lock
>> +#define mutex_unlock spin_unlock
>> +
>> +/* Alias to Xen time functions */
>> +#define ktime_t s_time_t
>> +#define ktime_get()             (NOW())
>> +#define ktime_add_us(t,i)       (t + MICROSECS(i))
>> +#define ktime_compare(t,i)      (t > (i))
>> +
>> +/* Alias to Xen allocation helpers */
>> +#define kzalloc(size, flags)    _xzalloc(size, sizeof(void *))
>> +#define kfree xfree
>> +#define devm_kzalloc(dev, size, flags)  _xzalloc(size, sizeof(void *))
>> +
>> +/* Device logger functions */
>> +#define dev_name(dev) dt_node_full_name(dev->of_node)
>> +#define dev_dbg(dev, fmt, ...)      \
>> +    printk(XENLOG_DEBUG "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +#define dev_notice(dev, fmt, ...)   \
>> +    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +#define dev_warn(dev, fmt, ...)     \
>> +    printk(XENLOG_WARNING "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +#define dev_err(dev, fmt, ...)      \
>> +    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +#define dev_info(dev, fmt, ...)     \
>> +    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +#define dev_err_ratelimited(dev, fmt, ...)      \
>> +    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +
>> +/*
>> + * Periodically poll an address and wait between reads in us until a
>> + * condition is met or a timeout occurs.
> 
> It would be good to add a statement to point out what the return value
> is going to be.

Ack. 
> 
> 
>> + */
>> +#define readx_poll_timeout(op, addr, val, cond, sleep_us, timeout_us) \
>> +({ \
>> +     s_time_t deadline = NOW() + MICROSECS(timeout_us); \
>> +     for (;;) { \
>> +        (val) = op(addr); \
>> +        if (cond) \
>> +            break; \
>> +        if (NOW() > deadline) { \
>> +            (val) = op(addr); \
>> +            break; \
>> +        } \
>> +        udelay(sleep_us); \
>> +     } \
>> +     (cond) ? 0 : -ETIMEDOUT; \
>> +})
>> +
>> +#define readl_relaxed_poll_timeout(addr, val, cond, delay_us, timeout_us) \
>> +    readx_poll_timeout(readl_relaxed, addr, val, cond, delay_us, timeout_us)
>> +
>> +#define FIELD_PREP(_mask, _val)         \
>> +    (((typeof(_mask))(_val) << (__builtin_ffsll(_mask) - 1)) & (_mask))
> 
> Let's add the definition of ffsll to bitops.h

Ok. 
> 
> 
>> +#define FIELD_GET(_mask, _reg)          \
>> +    (typeof(_mask))(((_reg) & (_mask)) >> (__builtin_ffsll(_mask) - 1))
>> +
>> +#define WRITE_ONCE(x, val)                  \
>> +do {                                        \
>> +    *(volatile typeof(x) *)&(x) = (val);    \
>> +} while (0)
> 
> maybe we should define this in xen/include/xen/lib.h

Ok. 
> 
> 
>> +
>> +/* Xen: Stub out DMA domain related functions */
>> +#define iommu_get_dma_cookie(dom) 0
>> +#define iommu_put_dma_cookie(dom)
> 
> Shouldn't we remove any call to iommu_get_dma_cookie and
> iommu_put_dma_cookie in one of the previous patches?

Ack. 
> 
> 
>> +/*
>> + * Helpers for DMA allocation. Just the function name is reused for
>> + * porting code, these allocation are not managed allocations
>>  */
>> +static void *dmam_alloc_coherent(struct device *dev, size_t size,
>> +                                 paddr_t *dma_handle, gfp_t gfp)
>> +{
>> +    void *vaddr;
>> +    unsigned long alignment = size;
>> +
>> +    /*
>> +     * _xzalloc requires that the (align & (align -1)) = 0. Most of the
>> +     * allocations in SMMU code should send the right value for size. In
>> +     * case this is not true print a warning and align to the size of a
>> +     * (void *)
>> +     */
>> +    if ( size & (size - 1) )
>> +    {
>> +        printk(XENLOG_WARNING "SMMUv3: Fixing alignment for the DMA buffer\n");
>> +        alignment = sizeof(void *);
>> +    }
>> +
>> +    vaddr = _xzalloc(size, alignment);
>> +    if ( !vaddr )
>> +    {
>> +        printk(XENLOG_ERR "SMMUv3: DMA allocation failed\n");
>> +        return NULL;
>> +    }
>> +
>> +    *dma_handle = virt_to_maddr(vaddr);
>> +
>> +    return vaddr;
>> +}
>> +
>> +/* Xen: Type definitions for iommu_domain */
>> +#define IOMMU_DOMAIN_UNMANAGED 0
>> +#define IOMMU_DOMAIN_DMA 1
>> +#define IOMMU_DOMAIN_IDENTITY 2
>> +
>> +/* Xen specific code. */
>> +struct iommu_domain {
>> +    /* Runtime SMMU configuration for this iommu_domain */
>> +    atomic_t ref;
>> +    /*
>> +     * Used to link iommu_domain contexts for a same domain.
>> +     * There is at least one per-SMMU to used by the domain.
>> +     */
>> +    struct list_head    list;
>> +};
>> +
>> +/* Describes information required for a Xen domain */
>> +struct arm_smmu_xen_domain {
>> +    spinlock_t      lock;
>> +
>> +    /* List of iommu domains associated to this domain */
>> +    struct list_head    contexts;
>> +};
>> +
>> +/*
>> + * Information about each device stored in dev->archdata.iommu
>> + * The dev->archdata.iommu stores the iommu_domain (runtime configuration of
>> + * the SMMU).
>> + */
>> +struct arm_smmu_xen_device {
>> +    struct iommu_domain *domain;
>> +};
> 
> Do we need both struct arm_smmu_xen_device and struct iommu_domain?
> 

No we don’t need both. I will remove the struct arm_smmu_xen_device. 

> 
>> +/* Keep a list of devices associated with this driver */
>> +static DEFINE_SPINLOCK(arm_smmu_devices_lock);
>> +static LIST_HEAD(arm_smmu_devices);
>> +
>> +
>> +static inline void *dev_iommu_priv_get(struct device *dev)
>> +{
>> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> +
>> +    return fwspec && fwspec->iommu_priv ? fwspec->iommu_priv : NULL;
>> +}
>> +
>> +static inline void dev_iommu_priv_set(struct device *dev, void *priv)
>> +{
>> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> +
>> +    fwspec->iommu_priv = priv;
>> +}
>> +
>> +int dt_property_match_string(const struct dt_device_node *np,
>> +                             const char *propname, const char *string)
>> +{
>> +    const struct dt_property *dtprop = dt_find_property(np, propname, NULL);
>> +    size_t l;
>> +    int i;
>> +    const char *p, *end;
>> +
>> +    if ( !dtprop )
>> +        return -EINVAL;
>> +
>> +    if ( !dtprop->value )
>> +        return -ENODATA;
>> +
>> +    p = dtprop->value;
>> +    end = p + dtprop->length;
>> +
>> +    for ( i = 0; p < end; i++, p += l )
>> +    {
>> +        l = strnlen(p, end - p) + 1;
>> +
>> +        if ( p + l > end )
>> +            return -EILSEQ;
>> +
>> +        if ( strcmp(string, p) == 0 )
>> +            return i; /* Found it; return index */
>> +    }
>> +
>> +    return -ENODATA;
>> +}
> 
> I think you should either use dt_property_read_string or move the
> implementation of dt_property_match_string to xen/common/device_tree.c
> (in which case I suggest to do it in a separate patch.)

I will prefer to move the code to xen/common/device_tree.c as it might help in future to use.
> 
> 
> I'd just use dt_property_read_string.
> 
> 
> 
>> +static int platform_get_irq_byname_optional(struct device *dev,
>> +                                            const char *name)
>> +{
>> +    int index, ret;
>> +    struct dt_device_node *np  = dev_to_dt(dev);
>> +
>> +    if ( unlikely(!name) )
>> +        return -EINVAL;
>> +
>> +    index = dt_property_match_string(np, "interrupt-names", name);
>> +    if ( index < 0 )
>> +    {
>> +        dev_info(dev, "IRQ %s not found\n", name);
>> +        return index;
>> +    }
>> 
>> -#include <linux/acpi.h>
>> -#include <linux/acpi_iort.h>
>> -#include <linux/bitfield.h>
>> -#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>
>> -#include <linux/iommu.h>
>> -#include <linux/iopoll.h>
>> -#include <linux/module.h>
>> -#include <linux/msi.h>
>> -#include <linux/of.h>
>> -#include <linux/of_address.h>
>> -#include <linux/of_iommu.h>
>> -#include <linux/of_platform.h>
>> -#include <linux/pci.h>
>> -#include <linux/pci-ats.h>
>> -#include <linux/platform_device.h>
>> -
>> -#include <linux/amba/bus.h>
>> +    ret = platform_get_irq(np, index);
>> +    if ( ret < 0 )
>> +    {
>> +        dev_err(dev, "failed to get irq index %d\n", index);
>> +        return -ENODEV;
>> +    }
>> +
>> +    return ret;
>> +}
>> +
>> +/* Start of Linux SMMUv3 code */
>> 
>> /* MMIO registers */
>> #define ARM_SMMU_IDR0			0x0
>> @@ -507,6 +751,7 @@ struct arm_smmu_s2_cfg {
>> 	u16				vmid;
>> 	u64				vttbr;
>> 	u64				vtcr;
>> +	struct domain		*domain;
>> };
>> 
>> struct arm_smmu_strtab_cfg {
>> @@ -567,8 +812,13 @@ struct arm_smmu_device {
>> 
>> 	struct arm_smmu_strtab_cfg	strtab_cfg;
>> 
>> -	/* IOMMU core code handle */
>> -	struct iommu_device		iommu;
>> +	/* Need to keep a list of SMMU devices */
>> +	struct list_head		devices;
>> +
>> +	/* Tasklets for handling evts/faults and pci page request IRQs*/
>> +	struct tasklet		evtq_irq_tasklet;
>> +	struct tasklet		priq_irq_tasklet;
>> +	struct tasklet		combined_irq_tasklet;
>> };
>> 
>> /* SMMU private data for each master */
>> @@ -1110,7 +1360,7 @@ static int arm_smmu_init_l2_strtab(struct arm_smmu_device *smmu, u32 sid)
>> }
>> 
>> /* IRQ and event handlers */
>> -static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
>> +static void arm_smmu_evtq_thread(void *dev)
> 
> I think we shouldn't call it a thread given that's a tasklet. We should
> rename the function or it will be confusing

Ok. 
> 
>> {
>> 	int i;
>> 	struct arm_smmu_device *smmu = dev;
>> @@ -1140,7 +1390,6 @@ static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
>> 	/* Sync our overflow flag, as we believe we're up to speed */
>> 	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
>> 		    Q_IDX(llq, llq->cons);
>> -	return IRQ_HANDLED;
>> }
>> 
>> static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
>> @@ -1181,7 +1430,7 @@ static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
>> 	}
>> }
>> 
>> -static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
>> +static void arm_smmu_priq_thread(void *dev)
> 
> same here

Ok. 
> 
> 
>> {
>> 	struct arm_smmu_device *smmu = dev;
>> 	struct arm_smmu_queue *q = &smmu->priq.q;
>> @@ -1200,12 +1449,12 @@ static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
>> 	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
>> 		      Q_IDX(llq, llq->cons);
>> 	queue_sync_cons_out(q);
>> -	return IRQ_HANDLED;
>> }
>> 
>> static int arm_smmu_device_disable(struct arm_smmu_device *smmu);
>> 
>> -static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>> +static void arm_smmu_gerror_handler(int irq, void *dev,
>> +				struct cpu_user_regs *regs)
>> {
>> 	u32 gerror, gerrorn, active;
>> 	struct arm_smmu_device *smmu = dev;
>> @@ -1215,7 +1464,7 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>> 
>> 	active = gerror ^ gerrorn;
>> 	if (!(active & GERROR_ERR_MASK))
>> -		return IRQ_NONE; /* No errors pending */
>> +		return; /* No errors pending */
>> 
>> 	dev_warn(smmu->dev,
>> 		 "unexpected global error reported (0x%08x), this could be serious\n",
>> @@ -1248,26 +1497,42 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>> 		arm_smmu_cmdq_skip_err(smmu);
>> 
>> 	writel(gerror, smmu->base + ARM_SMMU_GERRORN);
>> -	return IRQ_HANDLED;
>> }
>> 
>> -static irqreturn_t arm_smmu_combined_irq_thread(int irq, void *dev)
>> +static void arm_smmu_combined_irq_handler(int irq, void *dev,
>> +				struct cpu_user_regs *regs)
>> +{
>> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
>> +
>> +	arm_smmu_gerror_handler(irq, dev, regs);
>> +
>> +	tasklet_schedule(&(smmu->combined_irq_tasklet));
>> +}
>> +
>> +static void arm_smmu_combined_irq_thread(void *dev)
>> {
>> 	struct arm_smmu_device *smmu = dev;
>> 
>> -	arm_smmu_evtq_thread(irq, dev);
>> +	arm_smmu_evtq_thread(dev);
>> 	if (smmu->features & ARM_SMMU_FEAT_PRI)
>> -		arm_smmu_priq_thread(irq, dev);
>> -
>> -	return IRQ_HANDLED;
>> +		arm_smmu_priq_thread(dev);
>> }
>> 
>> -static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
>> +static void arm_smmu_evtq_irq_tasklet(int irq, void *dev,
>> +				struct cpu_user_regs *regs)
>> {
>> -	arm_smmu_gerror_handler(irq, dev);
>> -	return IRQ_WAKE_THREAD;
>> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
>> +
>> +	tasklet_schedule(&(smmu->evtq_irq_tasklet));
>> }
>> 
>> +static void arm_smmu_priq_irq_tasklet(int irq, void *dev,
>> +				struct cpu_user_regs *regs)
>> +{
>> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
>> +
>> +	tasklet_schedule(&(smmu->priq_irq_tasklet));
>> +}
>> 
>> /* IO_PGTABLE API */
>> static void arm_smmu_tlb_inv_context(void *cookie)
>> @@ -1354,27 +1619,69 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>> }
>> 
>> static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>> -				       struct arm_smmu_master *master,
>> -				       struct io_pgtable_cfg *pgtbl_cfg)
>> +				       struct arm_smmu_master *master)
>> {
>> 	int vmid;
>> +	u64 reg;
>> 	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> 	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>> 
>> +	/* VTCR */
>> +	reg = VTCR_RES1 | VTCR_SH0_IS | VTCR_IRGN0_WBWA | VTCR_ORGN0_WBWA;
>> +
>> +	switch (PAGE_SIZE) {
>> +	case SZ_4K:
>> +		reg |= VTCR_TG0_4K;
>> +		break;
>> +	case SZ_16K:
>> +		reg |= VTCR_TG0_16K;
>> +		break;
>> +	case SZ_64K:
>> +		reg |= VTCR_TG0_4K;
>> +		break;
>> +	}
>> +
>> +	switch (smmu->oas) {
>> +	case 32:
>> +		reg |= VTCR_PS(_AC(0x0,ULL));
>> +		break;
>> +	case 36:
>> +		reg |= VTCR_PS(_AC(0x1,ULL));
>> +		break;
>> +	case 40:
>> +		reg |= VTCR_PS(_AC(0x2,ULL));
>> +		break;
>> +	case 42:
>> +		reg |= VTCR_PS(_AC(0x3,ULL));
>> +		break;
>> +	case 44:
>> +		reg |= VTCR_PS(_AC(0x4,ULL));
>> +		break;
>> +		case 48:
>> +		reg |= VTCR_PS(_AC(0x5,ULL));
>> +		break;
>> +	case 52:
>> +		reg |= VTCR_PS(_AC(0x6,ULL));
>> +		break;
>> +	}
>> +
>> +	reg |= VTCR_T0SZ(64ULL - smmu->ias);
>> +	reg |= VTCR_SL0(0x2);
>> +	reg |= VTCR_VS;
>> +
>> +	cfg->vtcr   = reg;
>> +
>> 	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
>> 	if (vmid < 0)
>> 		return vmid;
>> +	cfg->vmid  = (u16)vmid;
>> +
>> +	cfg->vttbr  = page_to_maddr(cfg->domain->arch.p2m.root);
>> +
>> +	printk(XENLOG_DEBUG
>> +		   "SMMUv3: d%u: vmid 0x%x vtcr 0x%"PRIpaddr" p2maddr 0x%"PRIpaddr"\n",
>> +		   cfg->domain->domain_id, cfg->vmid, cfg->vtcr, cfg->vttbr);
>> 
>> -	vtcr = &pgtbl_cfg->arm_lpae_s2_cfg.vtcr;
>> -	cfg->vmid	= (u16)vmid;
>> -	cfg->vttbr	= pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
>> -	cfg->vtcr	= FIELD_PREP(STRTAB_STE_2_VTCR_S2T0SZ, vtcr->tsz) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SL0, vtcr->sl) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2IR0, vtcr->irgn) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2OR0, vtcr->orgn) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SH0, vtcr->sh) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2TG, vtcr->tg) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2PS, vtcr->ps);
>> 	return 0;
>> }
>> 
>> @@ -1382,28 +1689,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>> 				    struct arm_smmu_master *master)
>> {
>> 	int ret;
>> -	unsigned long ias, oas;
>> -	int (*finalise_stage_fn)(struct arm_smmu_domain *,
>> -				 struct arm_smmu_master *,
>> -				 struct io_pgtable_cfg *);
>> 	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> 
>> 	/* Restrict the stage to what we can actually support */
>> 	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
>> 
>> -	switch (smmu_domain->stage) {
>> -	case ARM_SMMU_DOMAIN_NESTED:
>> -	case ARM_SMMU_DOMAIN_S2:
>> -		ias = smmu->ias;
>> -		oas = smmu->oas;
>> -		finalise_stage_fn = arm_smmu_domain_finalise_s2;
>> -		break;
>> -	default:
>> -		return -EINVAL;
>> -	}
>> -
>> -	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
>> +	ret = arm_smmu_domain_finalise_s2(smmu_domain, master);
>> 	if (ret < 0) {
>> 		return ret;
>> 	}
>> @@ -1553,7 +1844,8 @@ static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>> 		return -ENOMEM;
>> 	}
>> 
>> -	if (!WARN_ON(q->base_dma & (qsz - 1))) {
>> +	WARN_ON(q->base_dma & (qsz - 1));
>> +	if (unlikely(q->base_dma & (qsz - 1))) {
>> 		dev_info(smmu->dev, "allocated %u entries for %s\n",
>> 			 1 << q->llq.max_n_shift, name);
> 
> We don't need both the WARNING and the dev_info. you could turn the
> dev_warn / XENLOG_WARNING.
> 
Ack.

> 
>> 	}
>> @@ -1758,9 +2050,7 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>> 	/* Request interrupt lines */
>> 	irq = smmu->evtq.q.irq;
>> 	if (irq) {
>> -		ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
>> -						arm_smmu_evtq_thread,
>> -						IRQF_ONESHOT,
>> +		ret = request_irq(irq, 0, arm_smmu_evtq_irq_tasklet,
>> 						"arm-smmu-v3-evtq", smmu);
>> 		if (ret < 0)
>> 			dev_warn(smmu->dev, "failed to enable evtq irq\n");
>> @@ -1770,8 +2060,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>> 
>> 	irq = smmu->gerr_irq;
>> 	if (irq) {
>> -		ret = devm_request_irq(smmu->dev, irq, arm_smmu_gerror_handler,
>> -				       0, "arm-smmu-v3-gerror", smmu);
>> +		ret = request_irq(irq, 0, arm_smmu_gerror_handler,
>> +						"arm-smmu-v3-gerror", smmu);
>> 		if (ret < 0)
>> 			dev_warn(smmu->dev, "failed to enable gerror irq\n");
>> 	} else {
>> @@ -1781,11 +2071,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>> 	if (smmu->features & ARM_SMMU_FEAT_PRI) {
>> 		irq = smmu->priq.q.irq;
>> 		if (irq) {
>> -			ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
>> -							arm_smmu_priq_thread,
>> -							IRQF_ONESHOT,
>> -							"arm-smmu-v3-priq",
>> -							smmu);
>> +			ret = request_irq(irq, 0, arm_smmu_priq_irq_tasklet,
>> +							"arm-smmu-v3-priq", smmu);
>> 			if (ret < 0)
>> 				dev_warn(smmu->dev,
>> 					 "failed to enable priq irq\n");
>> @@ -1814,11 +2101,8 @@ static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
>> 		 * Cavium ThunderX2 implementation doesn't support unique irq
>> 		 * lines. Use a single irq line for all the SMMUv3 interrupts.
>> 		 */
>> -		ret = devm_request_threaded_irq(smmu->dev, irq,
>> -					arm_smmu_combined_irq_handler,
>> -					arm_smmu_combined_irq_thread,
>> -					IRQF_ONESHOT,
>> -					"arm-smmu-v3-combined-irq", smmu);
>> +		ret = request_irq(irq, 0, arm_smmu_combined_irq_handler,
>> +						"arm-smmu-v3-combined-irq", smmu);
>> 		if (ret < 0)
>> 			dev_warn(smmu->dev, "failed to enable combined irq\n");
>> 	} else
>> @@ -1857,7 +2141,7 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>> 	reg = readl_relaxed(smmu->base + ARM_SMMU_CR0);
>> 	if (reg & CR0_SMMUEN) {
>> 		dev_warn(smmu->dev, "SMMU currently enabled! Resetting...\n");
>> -		WARN_ON(is_kdump_kernel() && !disable_bypass);
>> +		WARN_ON(!disable_bypass);
>> 		arm_smmu_update_gbpa(smmu, GBPA_ABORT, 0);
>> 	}
>> 
>> @@ -1952,8 +2236,11 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>> 		return ret;
>> 	}
>> 
>> -	if (is_kdump_kernel())
>> -		enables &= ~(CR0_EVTQEN | CR0_PRIQEN);
>> +	/* Initialize tasklets for threaded IRQs*/
>> +	tasklet_init(&smmu->evtq_irq_tasklet, arm_smmu_evtq_thread, smmu);
>> +	tasklet_init(&smmu->priq_irq_tasklet, arm_smmu_priq_thread, smmu);
>> +	tasklet_init(&smmu->combined_irq_tasklet, arm_smmu_combined_irq_thread,
>> +				 smmu);
>> 
>> 	/* Enable the SMMU interface, or ensure bypass */
>> 	if (!bypass || disable_bypass) {
>> @@ -2195,7 +2482,7 @@ static inline int arm_smmu_device_acpi_probe(struct platform_device *pdev,
>> static int arm_smmu_device_dt_probe(struct platform_device *pdev,
>> 				    struct arm_smmu_device *smmu)
>> {
>> -	struct device *dev = &pdev->dev;
>> +	struct device *dev = pdev;
>> 	u32 cells;
>> 	int ret = -EINVAL;
>> 
>> @@ -2219,130 +2506,449 @@ static unsigned long arm_smmu_resource_size(struct arm_smmu_device *smmu)
>> 		return SZ_128K;
>> }
>> 
>> +/* Start of Xen specific code. */
>> static int arm_smmu_device_probe(struct platform_device *pdev)
>> {
>> -	int irq, ret;
>> -	struct resource *res;
>> -	resource_size_t ioaddr;
>> -	struct arm_smmu_device *smmu;
>> -	struct device *dev = &pdev->dev;
>> -	bool bypass;
>> +    int irq, ret;
>> +    paddr_t ioaddr, iosize;
>> +    struct arm_smmu_device *smmu;
>> +    bool bypass;
>> +
>> +    smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
>> +    if ( !smmu )
>> +    {
>> +        dev_err(pdev, "failed to allocate arm_smmu_device\n");
>> +        return -ENOMEM;
>> +    }
>> +    smmu->dev = pdev;
>> +
>> +    if ( pdev->of_node )
>> +    {
>> +        ret = arm_smmu_device_dt_probe(pdev, smmu);
>> +    } else
>> +    {
> 
> coding style

Ack. 
> 
> 
>> +        ret = arm_smmu_device_acpi_probe(pdev, smmu);
>> +        if ( ret == -ENODEV )
>> +            return ret;
>> +    }
>> +
>> +    /* Set bypass mode according to firmware probing result */
>> +    bypass = !!ret;
>> +
>> +    /* Base address */
>> +    ret = dt_device_get_address(dev_to_dt(pdev), 0, &ioaddr, &iosize);
>> +    if( ret )
>> +        return -ENODEV;
>> +
>> +    if ( iosize < arm_smmu_resource_size(smmu) )
>> +    {
>> +        dev_err(pdev, "MMIO region too small (%lx)\n", iosize);
>> +        return -EINVAL;
>> +    }
>> +
>> +    /*
>> +     * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
>> +     * the PMCG registers which are reserved by the PMU driver.
> 
> Does this apply to Xen too?

Yes. Performance monitoring facilities are optional. Currently there is no plan to support also in XEN.
> 
> 
>> +     */
>> +    smmu->base = ioremap_nocache(ioaddr, iosize);
>> +    if ( IS_ERR(smmu->base) )
>> +        return PTR_ERR(smmu->base);
>> +
>> +    if ( iosize > SZ_64K )
>> +    {
>> +        smmu->page1 = ioremap_nocache(ioaddr + SZ_64K, ARM_SMMU_REG_SZ);
>> +        if ( IS_ERR(smmu->page1) )
>> +            return PTR_ERR(smmu->page1);
>> +    }
>> +    else
>> +    {
>> +        smmu->page1 = smmu->base;
>> +    }
>> +
>> +    /* Interrupt lines */
>> +
>> +    irq = platform_get_irq_byname_optional(pdev, "combined");
>> +    if ( irq > 0 )
>> +        smmu->combined_irq = irq;
>> +    else
>> +    {
>> +        irq = platform_get_irq_byname_optional(pdev, "eventq");
>> +        if ( irq > 0 )
>> +            smmu->evtq.q.irq = irq;
>> +
>> +        irq = platform_get_irq_byname_optional(pdev, "priq");
>> +        if ( irq > 0 )
>> +            smmu->priq.q.irq = irq;
>> +
>> +        irq = platform_get_irq_byname_optional(pdev, "gerror");
>> +        if ( irq > 0 )
>> +            smmu->gerr_irq = irq;
>> +    }
>> +    /* Probe the h/w */
>> +    ret = arm_smmu_device_hw_probe(smmu);
>> +    if ( ret )
>> +        return ret;
>> +
>> +    /* Initialise in-memory data structures */
>> +    ret = arm_smmu_init_structures(smmu);
>> +    if ( ret )
>> +        return ret;
>> +
>> +    /* Reset the device */
>> +    ret = arm_smmu_device_reset(smmu, bypass);
>> +    if ( ret )
>> +        return ret;
>> +
>> +    /*
>> +     * Keep a list of all probed devices. This will be used to query
>> +     * the smmu devices based on the fwnode.
>> +     */
>> +    INIT_LIST_HEAD(&smmu->devices);
>> +
>> +    spin_lock(&arm_smmu_devices_lock);
>> +    list_add(&smmu->devices, &arm_smmu_devices);
>> +    spin_unlock(&arm_smmu_devices_lock);
>> +
>> +    return 0;
>> +}


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-02  2:51   ` Stefano Stabellini
  2020-12-02 16:27     ` Rahul Singh
@ 2020-12-02 16:47     ` Julien Grall
  2020-12-03  4:13       ` Stefano Stabellini
  1 sibling, 1 reply; 53+ messages in thread
From: Julien Grall @ 2020-12-02 16:47 UTC (permalink / raw)
  To: Stefano Stabellini, Rahul Singh
  Cc: xen-devel, bertrand.marquis, Andrew Cooper, George Dunlap,
	Ian Jackson, Jan Beulich, Wei Liu, Paul Durrant,
	Volodymyr Babchuk

Hi Stefano,

On 02/12/2020 02:51, Stefano Stabellini wrote:
> On Thu, 26 Nov 2020, Rahul Singh wrote:

>> +/* Alias to Xen device tree helpers */
>> +#define device_node dt_device_node
>> +#define of_phandle_args dt_phandle_args
>> +#define of_device_id dt_device_match
>> +#define of_match_node dt_match_node
>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, pname, out))
>> +#define of_property_read_bool dt_property_read_bool
>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
> 
> Given all the changes to the file by the previous patches we are
> basically fully (or almost fully) adapting this code to Xen.
> 
> So at that point I wonder if we should just as well make these changes
> (e.g. s/of_phandle_args/dt_phandle_args/g) to the code too.

I have already accepted the fact that keeping Linux code as-is is nearly 
impossible without much workaround :). The benefits tends to also 
limited as we noticed for the SMMU driver.

I would like to point out that this may make quite difficult (if not 
impossible) to revert the previous patches which remove support for some 
features (e.g. atomic, MSI, ATS).

If we are going to adapt the code to Xen (I'd like to keep Linux code 
style though), then I think we should consider to keep code that may be 
useful in the near future (at least MSI, ATS).

> 
>> +#define FIELD_GET(_mask, _reg)          \
>> +    (typeof(_mask))(((_reg) & (_mask)) >> (__builtin_ffsll(_mask) - 1))
>> +
>> +#define WRITE_ONCE(x, val)                  \
>> +do {                                        \
>> +    *(volatile typeof(x) *)&(x) = (val);    \
>> +} while (0)
> 
> maybe we should define this in xen/include/xen/lib.h

I have attempted such discussion in the past and this resulted to more 
bikeshed than it is worth it. So I would suggest to re-implement 
WRITE_ONCE() as write_atomic() for now.

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-02 16:27     ` Rahul Singh
@ 2020-12-02 19:26       ` Rahul Singh
  0 siblings, 0 replies; 53+ messages in thread
From: Rahul Singh @ 2020-12-02 19:26 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: xen-devel, Bertrand Marquis, Andrew Cooper, George Dunlap,
	Ian Jackson, Jan Beulich, Julien Grall, Wei Liu, Paul Durrant,
	Volodymyr Babchuk

Hi Stefano,

> On 2 Dec 2020, at 4:27 pm, Rahul Singh <Rahul.Singh@arm.com> wrote:
> 
> Hello Stefano,
> 
>> On 2 Dec 2020, at 2:51 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
>> 
>> On Thu, 26 Nov 2020, Rahul Singh wrote:
>>> Add support for ARM architected SMMUv3 implementation. It is based on
>>> the Linux SMMUv3 driver.
>>> 
>>> Major differences with regard to Linux driver are as follows:
>>> 1. Only Stage-2 translation is supported as compared to the Linux driver
>>>  that supports both Stage-1 and Stage-2 translations.
>>> 2. Use P2M  page table instead of creating one as SMMUv3 has the
>>>  capability to share the page tables with the CPU.
>>> 3. Tasklets are used in place of threaded IRQ's in Linux for event queue
>>>  and priority queue IRQ handling.
>>> 4. Latest version of the Linux SMMUv3 code implements the commands queue
>>>  access functions based on atomic operations implemented in Linux.
>>>  Atomic functions used by the commands queue access functions are not
>>>  implemented in XEN therefore we decided to port the earlier version
>>>  of the code. Once the proper atomic operations will be available in
>>>  XEN the driver can be updated.
>>> 5. Driver is currently supported as Tech Preview.
>> 
>> This patch is big and was difficult to review, nonetheless I tried :-)
> 
> Thanks again for reviewing the code.
>> 
>> That said, the code is self-contained, marked as Tech Preview, and not
>> at risk of making other things unstable, so it is low risk to accept it.
>> 
>> Comments below.
>> 
>> 
>>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
>>> ---
>>> MAINTAINERS                           |   6 +
>>> SUPPORT.md                            |   1 +
>>> xen/drivers/passthrough/Kconfig       |  10 +
>>> xen/drivers/passthrough/arm/Makefile  |   1 +
>>> xen/drivers/passthrough/arm/smmu-v3.c | 986 +++++++++++++++++++++-----
>>> 5 files changed, 814 insertions(+), 190 deletions(-)
>>> 
>>> diff --git a/MAINTAINERS b/MAINTAINERS
>>> index dab38a6a14..1d63489eec 100644
>>> --- a/MAINTAINERS
>>> +++ b/MAINTAINERS
>>> @@ -249,6 +249,12 @@ F:	xen/include/asm-arm/
>>> F:	xen/include/public/arch-arm/
>>> F:	xen/include/public/arch-arm.h
>>> 
>>> +ARM SMMUv3
>>> +M:	Bertrand Marquis <bertrand.marquis@arm.com>
>>> +M:	Rahul Singh <rahul.singh@arm.com>
>>> +S:	Supported
>>> +F:	xen/drivers/passthrough/arm/smmu-v3.c
>>> +
>>> Change Log
>>> M:	Paul Durrant <paul@xen.org>
>>> R:	Community Manager <community.manager@xenproject.org>
>>> diff --git a/SUPPORT.md b/SUPPORT.md
>>> index ab02aca5f4..e402c7202d 100644
>>> --- a/SUPPORT.md
>>> +++ b/SUPPORT.md
>>> @@ -68,6 +68,7 @@ For the Cortex A57 r0p0 - r1p1, see Errata 832075.
>>>    Status, ARM SMMUv1: Supported, not security supported
>>>    Status, ARM SMMUv2: Supported, not security supported
>>>    Status, Renesas IPMMU-VMSA: Supported, not security supported
>>> +    Status, ARM SMMUv3: Tech Preview
>>> 
>>> ### ARM/GICv3 ITS
>>> 
>>> diff --git a/xen/drivers/passthrough/Kconfig b/xen/drivers/passthrough/Kconfig
>>> index 0036007ec4..5b71c59f47 100644
>>> --- a/xen/drivers/passthrough/Kconfig
>>> +++ b/xen/drivers/passthrough/Kconfig
>>> @@ -13,6 +13,16 @@ config ARM_SMMU
>>> 	  Say Y here if your SoC includes an IOMMU device implementing the
>>> 	  ARM SMMU architecture.
>>> 
>>> +config ARM_SMMU_V3
>>> +	bool "ARM Ltd. System MMU Version 3 (SMMUv3) Support" if EXPERT
>>> +	depends on ARM_64
>>> +	---help---
>>> +	 Support for implementations of the ARM System MMU architecture
>>> +	 version 3.
>>> +
>>> +	 Say Y here if your system includes an IOMMU device implementing
>>> +	 the ARM SMMUv3 architecture.
>>> +
>>> config IPMMU_VMSA
>>> 	bool "Renesas IPMMU-VMSA found in R-Car Gen3 SoCs"
>>> 	depends on ARM_64
>>> diff --git a/xen/drivers/passthrough/arm/Makefile b/xen/drivers/passthrough/arm/Makefile
>>> index fcd918ea3e..c5fb3b58a5 100644
>>> --- a/xen/drivers/passthrough/arm/Makefile
>>> +++ b/xen/drivers/passthrough/arm/Makefile
>>> @@ -1,3 +1,4 @@
>>> obj-y += iommu.o iommu_helpers.o iommu_fwspec.o
>>> obj-$(CONFIG_ARM_SMMU) += smmu.o
>>> obj-$(CONFIG_IPMMU_VMSA) += ipmmu-vmsa.o
>>> +obj-$(CONFIG_ARM_SMMU_V3) += smmu-v3.o
>>> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
>>> index 55d1cba194..8f2337e7f2 100644
>>> --- a/xen/drivers/passthrough/arm/smmu-v3.c
>>> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
>>> @@ -2,36 +2,280 @@
>>> /*
>>> * IOMMU API for ARM architected SMMUv3 implementations.
>>> *
>>> - * Copyright (C) 2015 ARM Limited
>>> + * Based on Linux's SMMUv3 driver:
>>> + *    drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
>>> + *    commit: 951cbbc386ff01b50da4f46387e994e81d9ab431
>>> + * and Xen's SMMU driver:
>>> + *    xen/drivers/passthrough/arm/smmu.c
>>> *
>>> - * Author: Will Deacon <will.deacon@arm.com>
>>> + * Copyright (C) 2015 ARM Limited Will Deacon <will.deacon@arm.com>
>>> *
>>> - * This driver is powered by bad coffee and bombay mix.
>>> + * Copyright (C) 2020 Arm Ltd.
>>> + *
>>> + * 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, see <http://www.gnu.org/licenses/>.
>>> + *
>>> + */
>>> +
>>> +#include <xen/acpi.h>
>>> +#include <xen/config.h>
>>> +#include <xen/delay.h>
>>> +#include <xen/errno.h>
>>> +#include <xen/err.h>
>>> +#include <xen/irq.h>
>>> +#include <xen/lib.h>
>>> +#include <xen/list.h>
>>> +#include <xen/mm.h>
>>> +#include <xen/rbtree.h>
>>> +#include <xen/sched.h>
>>> +#include <xen/sizes.h>
>>> +#include <xen/vmap.h>
>>> +#include <asm/atomic.h>
>>> +#include <asm/device.h>
>>> +#include <asm/io.h>
>>> +#include <asm/platform.h>
>>> +#include <asm/iommu_fwspec.h>
>>> +
>>> +/* Linux compatibility functions. */
>>> +typedef paddr_t dma_addr_t;
>>> +typedef unsigned int gfp_t;
>>> +
>>> +#define platform_device device
>>> +
>>> +#define GFP_KERNEL 0
>>> +
>>> +/* Alias to Xen device tree helpers */
>>> +#define device_node dt_device_node
>>> +#define of_phandle_args dt_phandle_args
>>> +#define of_device_id dt_device_match
>>> +#define of_match_node dt_match_node
>>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, pname, out))
>>> +#define of_property_read_bool dt_property_read_bool
>>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
>> 
>> Given all the changes to the file by the previous patches we are
>> basically fully (or almost fully) adapting this code to Xen.
>> 
>> So at that point I wonder if we should just as well make these changes
>> (e.g. s/of_phandle_args/dt_phandle_args/g) to the code too.
> 
> Yes, you are right that is why in the first version of the patch I modifying the code to make it fully XEN compatible. But in this patch series I use the Linux compatibility function to have Linux code unmodified.I also prefer to make changes (s/of_phandle_args/dt_phandle_args/g).
> 
>> 
>> Julien, Rahul, what do you guys think? 
>> 
>> 
>>> +/* Alias to Xen lock functions */
>> 
>> I think this deserves at least one statement to explain why it is OK.
> 
> Ack. I Will add the comment.
>> 
>> Also, a similar comment to the one above: maybe we should add a couple
>> of more preparation patches to s/mutex/spinlock/g and change the time
>> and allocation functions too.
>> 
>> 
>>> +#define mutex spinlock
>>> +#define mutex_init spin_lock_init
>>> +#define mutex_lock spin_lock
>>> +#define mutex_unlock spin_unlock
>>> +
>>> +/* Alias to Xen time functions */
>>> +#define ktime_t s_time_t
>>> +#define ktime_get()             (NOW())
>>> +#define ktime_add_us(t,i)       (t + MICROSECS(i))
>>> +#define ktime_compare(t,i)      (t > (i))
>>> +
>>> +/* Alias to Xen allocation helpers */
>>> +#define kzalloc(size, flags)    _xzalloc(size, sizeof(void *))
>>> +#define kfree xfree
>>> +#define devm_kzalloc(dev, size, flags)  _xzalloc(size, sizeof(void *))
>>> +
>>> +/* Device logger functions */
>>> +#define dev_name(dev) dt_node_full_name(dev->of_node)
>>> +#define dev_dbg(dev, fmt, ...)      \
>>> +    printk(XENLOG_DEBUG "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>>> +#define dev_notice(dev, fmt, ...)   \
>>> +    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>>> +#define dev_warn(dev, fmt, ...)     \
>>> +    printk(XENLOG_WARNING "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>>> +#define dev_err(dev, fmt, ...)      \
>>> +    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>>> +#define dev_info(dev, fmt, ...)     \
>>> +    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>>> +#define dev_err_ratelimited(dev, fmt, ...)      \
>>> +    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>>> +
>>> +/*
>>> + * Periodically poll an address and wait between reads in us until a
>>> + * condition is met or a timeout occurs.
>> 
>> It would be good to add a statement to point out what the return value
>> is going to be.
> 
> Ack. 
>> 
>> 
>>> + */
>>> +#define readx_poll_timeout(op, addr, val, cond, sleep_us, timeout_us) \
>>> +({ \
>>> +     s_time_t deadline = NOW() + MICROSECS(timeout_us); \
>>> +     for (;;) { \
>>> +        (val) = op(addr); \
>>> +        if (cond) \
>>> +            break; \
>>> +        if (NOW() > deadline) { \
>>> +            (val) = op(addr); \
>>> +            break; \
>>> +        } \
>>> +        udelay(sleep_us); \
>>> +     } \
>>> +     (cond) ? 0 : -ETIMEDOUT; \
>>> +})
>>> +
>>> +#define readl_relaxed_poll_timeout(addr, val, cond, delay_us, timeout_us) \
>>> +    readx_poll_timeout(readl_relaxed, addr, val, cond, delay_us, timeout_us)
>>> +
>>> +#define FIELD_PREP(_mask, _val)         \
>>> +    (((typeof(_mask))(_val) << (__builtin_ffsll(_mask) - 1)) & (_mask))
>> 
>> Let's add the definition of ffsll to bitops.h
> 
> Ok. 
>> 
>> 
>>> +#define FIELD_GET(_mask, _reg)          \
>>> +    (typeof(_mask))(((_reg) & (_mask)) >> (__builtin_ffsll(_mask) - 1))
>>> +
>>> +#define WRITE_ONCE(x, val)                  \
>>> +do {                                        \
>>> +    *(volatile typeof(x) *)&(x) = (val);    \
>>> +} while (0)
>> 
>> maybe we should define this in xen/include/xen/lib.h
> 
> Ok. 
>> 
>> 
>>> +
>>> +/* Xen: Stub out DMA domain related functions */
>>> +#define iommu_get_dma_cookie(dom) 0
>>> +#define iommu_put_dma_cookie(dom)
>> 
>> Shouldn't we remove any call to iommu_get_dma_cookie and
>> iommu_put_dma_cookie in one of the previous patches?
> 
> Ack. 
>> 
>> 
>>> +/*
>>> + * Helpers for DMA allocation. Just the function name is reused for
>>> + * porting code, these allocation are not managed allocations
>>> */
>>> +static void *dmam_alloc_coherent(struct device *dev, size_t size,
>>> +                                 paddr_t *dma_handle, gfp_t gfp)
>>> +{
>>> +    void *vaddr;
>>> +    unsigned long alignment = size;
>>> +
>>> +    /*
>>> +     * _xzalloc requires that the (align & (align -1)) = 0. Most of the
>>> +     * allocations in SMMU code should send the right value for size. In
>>> +     * case this is not true print a warning and align to the size of a
>>> +     * (void *)
>>> +     */
>>> +    if ( size & (size - 1) )
>>> +    {
>>> +        printk(XENLOG_WARNING "SMMUv3: Fixing alignment for the DMA buffer\n");
>>> +        alignment = sizeof(void *);
>>> +    }
>>> +
>>> +    vaddr = _xzalloc(size, alignment);
>>> +    if ( !vaddr )
>>> +    {
>>> +        printk(XENLOG_ERR "SMMUv3: DMA allocation failed\n");
>>> +        return NULL;
>>> +    }
>>> +
>>> +    *dma_handle = virt_to_maddr(vaddr);
>>> +
>>> +    return vaddr;
>>> +}
>>> +
>>> +/* Xen: Type definitions for iommu_domain */
>>> +#define IOMMU_DOMAIN_UNMANAGED 0
>>> +#define IOMMU_DOMAIN_DMA 1
>>> +#define IOMMU_DOMAIN_IDENTITY 2
>>> +
>>> +/* Xen specific code. */
>>> +struct iommu_domain {
>>> +    /* Runtime SMMU configuration for this iommu_domain */
>>> +    atomic_t ref;
>>> +    /*
>>> +     * Used to link iommu_domain contexts for a same domain.
>>> +     * There is at least one per-SMMU to used by the domain.
>>> +     */
>>> +    struct list_head    list;
>>> +};
>>> +
>>> +/* Describes information required for a Xen domain */
>>> +struct arm_smmu_xen_domain {
>>> +    spinlock_t      lock;
>>> +
>>> +    /* List of iommu domains associated to this domain */
>>> +    struct list_head    contexts;
>>> +};
>>> +
>>> +/*
>>> + * Information about each device stored in dev->archdata.iommu
>>> + * The dev->archdata.iommu stores the iommu_domain (runtime configuration of
>>> + * the SMMU).
>>> + */
>>> +struct arm_smmu_xen_device {
>>> +    struct iommu_domain *domain;
>>> +};
>> 
>> Do we need both struct arm_smmu_xen_device and struct iommu_domain?
>> 
> 
> No we don’t need both. I will remove the struct arm_smmu_xen_device. 
> 
>> 
>>> +/* Keep a list of devices associated with this driver */
>>> +static DEFINE_SPINLOCK(arm_smmu_devices_lock);
>>> +static LIST_HEAD(arm_smmu_devices);
>>> +
>>> +
>>> +static inline void *dev_iommu_priv_get(struct device *dev)
>>> +{
>>> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>>> +
>>> +    return fwspec && fwspec->iommu_priv ? fwspec->iommu_priv : NULL;
>>> +}
>>> +
>>> +static inline void dev_iommu_priv_set(struct device *dev, void *priv)
>>> +{
>>> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>>> +
>>> +    fwspec->iommu_priv = priv;
>>> +}
>>> +
>>> +int dt_property_match_string(const struct dt_device_node *np,
>>> +                             const char *propname, const char *string)
>>> +{
>>> +    const struct dt_property *dtprop = dt_find_property(np, propname, NULL);
>>> +    size_t l;
>>> +    int i;
>>> +    const char *p, *end;
>>> +
>>> +    if ( !dtprop )
>>> +        return -EINVAL;
>>> +
>>> +    if ( !dtprop->value )
>>> +        return -ENODATA;
>>> +
>>> +    p = dtprop->value;
>>> +    end = p + dtprop->length;
>>> +
>>> +    for ( i = 0; p < end; i++, p += l )
>>> +    {
>>> +        l = strnlen(p, end - p) + 1;
>>> +
>>> +        if ( p + l > end )
>>> +            return -EILSEQ;
>>> +
>>> +        if ( strcmp(string, p) == 0 )
>>> +            return i; /* Found it; return index */
>>> +    }
>>> +
>>> +    return -ENODATA;
>>> +}
>> 
>> I think you should either use dt_property_read_string or move the
>> implementation of dt_property_match_string to xen/common/device_tree.c
>> (in which case I suggest to do it in a separate patch.)
> 
> I will prefer to move the code to xen/common/device_tree.c as it might help in future to use.
>> 
>> 
>> I'd just use dt_property_read_string.
>> 
>> 
>> 
>>> +static int platform_get_irq_byname_optional(struct device *dev,
>>> +                                            const char *name)
>>> +{
>>> +    int index, ret;
>>> +    struct dt_device_node *np  = dev_to_dt(dev);
>>> +
>>> +    if ( unlikely(!name) )
>>> +        return -EINVAL;
>>> +
>>> +    index = dt_property_match_string(np, "interrupt-names", name);
>>> +    if ( index < 0 )
>>> +    {
>>> +        dev_info(dev, "IRQ %s not found\n", name);
>>> +        return index;
>>> +    }
>>> 
>>> -#include <linux/acpi.h>
>>> -#include <linux/acpi_iort.h>
>>> -#include <linux/bitfield.h>
>>> -#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>
>>> -#include <linux/iommu.h>
>>> -#include <linux/iopoll.h>
>>> -#include <linux/module.h>
>>> -#include <linux/msi.h>
>>> -#include <linux/of.h>
>>> -#include <linux/of_address.h>
>>> -#include <linux/of_iommu.h>
>>> -#include <linux/of_platform.h>
>>> -#include <linux/pci.h>
>>> -#include <linux/pci-ats.h>
>>> -#include <linux/platform_device.h>
>>> -
>>> -#include <linux/amba/bus.h>
>>> +    ret = platform_get_irq(np, index);
>>> +    if ( ret < 0 )
>>> +    {
>>> +        dev_err(dev, "failed to get irq index %d\n", index);
>>> +        return -ENODEV;
>>> +    }
>>> +
>>> +    return ret;
>>> +}
>>> +
>>> +/* Start of Linux SMMUv3 code */
>>> 
>>> /* MMIO registers */
>>> #define ARM_SMMU_IDR0			0x0
>>> @@ -507,6 +751,7 @@ struct arm_smmu_s2_cfg {
>>> 	u16				vmid;
>>> 	u64				vttbr;
>>> 	u64				vtcr;
>>> +	struct domain		*domain;
>>> };
>>> 
>>> struct arm_smmu_strtab_cfg {
>>> @@ -567,8 +812,13 @@ struct arm_smmu_device {
>>> 
>>> 	struct arm_smmu_strtab_cfg	strtab_cfg;
>>> 
>>> -	/* IOMMU core code handle */
>>> -	struct iommu_device		iommu;
>>> +	/* Need to keep a list of SMMU devices */
>>> +	struct list_head		devices;
>>> +
>>> +	/* Tasklets for handling evts/faults and pci page request IRQs*/
>>> +	struct tasklet		evtq_irq_tasklet;
>>> +	struct tasklet		priq_irq_tasklet;
>>> +	struct tasklet		combined_irq_tasklet;
>>> };
>>> 
>>> /* SMMU private data for each master */
>>> @@ -1110,7 +1360,7 @@ static int arm_smmu_init_l2_strtab(struct arm_smmu_device *smmu, u32 sid)
>>> }
>>> 
>>> /* IRQ and event handlers */
>>> -static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
>>> +static void arm_smmu_evtq_thread(void *dev)
>> 
>> I think we shouldn't call it a thread given that's a tasklet. We should
>> rename the function or it will be confusing
> 
> Ok. 
>> 
>>> {
>>> 	int i;
>>> 	struct arm_smmu_device *smmu = dev;
>>> @@ -1140,7 +1390,6 @@ static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
>>> 	/* Sync our overflow flag, as we believe we're up to speed */
>>> 	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
>>> 		    Q_IDX(llq, llq->cons);
>>> -	return IRQ_HANDLED;
>>> }
>>> 
>>> static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
>>> @@ -1181,7 +1430,7 @@ static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
>>> 	}
>>> }
>>> 
>>> -static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
>>> +static void arm_smmu_priq_thread(void *dev)
>> 
>> same here
> 
> Ok. 
>> 
>> 
>>> {
>>> 	struct arm_smmu_device *smmu = dev;
>>> 	struct arm_smmu_queue *q = &smmu->priq.q;
>>> @@ -1200,12 +1449,12 @@ static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
>>> 	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
>>> 		      Q_IDX(llq, llq->cons);
>>> 	queue_sync_cons_out(q);
>>> -	return IRQ_HANDLED;
>>> }
>>> 
>>> static int arm_smmu_device_disable(struct arm_smmu_device *smmu);
>>> 
>>> -static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>>> +static void arm_smmu_gerror_handler(int irq, void *dev,
>>> +				struct cpu_user_regs *regs)
>>> {
>>> 	u32 gerror, gerrorn, active;
>>> 	struct arm_smmu_device *smmu = dev;
>>> @@ -1215,7 +1464,7 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>>> 
>>> 	active = gerror ^ gerrorn;
>>> 	if (!(active & GERROR_ERR_MASK))
>>> -		return IRQ_NONE; /* No errors pending */
>>> +		return; /* No errors pending */
>>> 
>>> 	dev_warn(smmu->dev,
>>> 		 "unexpected global error reported (0x%08x), this could be serious\n",
>>> @@ -1248,26 +1497,42 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>>> 		arm_smmu_cmdq_skip_err(smmu);
>>> 
>>> 	writel(gerror, smmu->base + ARM_SMMU_GERRORN);
>>> -	return IRQ_HANDLED;
>>> }
>>> 
>>> -static irqreturn_t arm_smmu_combined_irq_thread(int irq, void *dev)
>>> +static void arm_smmu_combined_irq_handler(int irq, void *dev,
>>> +				struct cpu_user_regs *regs)
>>> +{
>>> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
>>> +
>>> +	arm_smmu_gerror_handler(irq, dev, regs);
>>> +
>>> +	tasklet_schedule(&(smmu->combined_irq_tasklet));
>>> +}
>>> +
>>> +static void arm_smmu_combined_irq_thread(void *dev)
>>> {
>>> 	struct arm_smmu_device *smmu = dev;
>>> 
>>> -	arm_smmu_evtq_thread(irq, dev);
>>> +	arm_smmu_evtq_thread(dev);
>>> 	if (smmu->features & ARM_SMMU_FEAT_PRI)
>>> -		arm_smmu_priq_thread(irq, dev);
>>> -
>>> -	return IRQ_HANDLED;
>>> +		arm_smmu_priq_thread(dev);
>>> }
>>> 
>>> -static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
>>> +static void arm_smmu_evtq_irq_tasklet(int irq, void *dev,
>>> +				struct cpu_user_regs *regs)
>>> {
>>> -	arm_smmu_gerror_handler(irq, dev);
>>> -	return IRQ_WAKE_THREAD;
>>> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
>>> +
>>> +	tasklet_schedule(&(smmu->evtq_irq_tasklet));
>>> }
>>> 
>>> +static void arm_smmu_priq_irq_tasklet(int irq, void *dev,
>>> +				struct cpu_user_regs *regs)
>>> +{
>>> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
>>> +
>>> +	tasklet_schedule(&(smmu->priq_irq_tasklet));
>>> +}
>>> 
>>> /* IO_PGTABLE API */
>>> static void arm_smmu_tlb_inv_context(void *cookie)
>>> @@ -1354,27 +1619,69 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>>> }
>>> 
>>> static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>>> -				       struct arm_smmu_master *master,
>>> -				       struct io_pgtable_cfg *pgtbl_cfg)
>>> +				       struct arm_smmu_master *master)
>>> {
>>> 	int vmid;
>>> +	u64 reg;
>>> 	struct arm_smmu_device *smmu = smmu_domain->smmu;
>>> 	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>>> 
>>> +	/* VTCR */
>>> +	reg = VTCR_RES1 | VTCR_SH0_IS | VTCR_IRGN0_WBWA | VTCR_ORGN0_WBWA;
>>> +
>>> +	switch (PAGE_SIZE) {
>>> +	case SZ_4K:
>>> +		reg |= VTCR_TG0_4K;
>>> +		break;
>>> +	case SZ_16K:
>>> +		reg |= VTCR_TG0_16K;
>>> +		break;
>>> +	case SZ_64K:
>>> +		reg |= VTCR_TG0_4K;
>>> +		break;
>>> +	}
>>> +
>>> +	switch (smmu->oas) {
>>> +	case 32:
>>> +		reg |= VTCR_PS(_AC(0x0,ULL));
>>> +		break;
>>> +	case 36:
>>> +		reg |= VTCR_PS(_AC(0x1,ULL));
>>> +		break;
>>> +	case 40:
>>> +		reg |= VTCR_PS(_AC(0x2,ULL));
>>> +		break;
>>> +	case 42:
>>> +		reg |= VTCR_PS(_AC(0x3,ULL));
>>> +		break;
>>> +	case 44:
>>> +		reg |= VTCR_PS(_AC(0x4,ULL));
>>> +		break;
>>> +		case 48:
>>> +		reg |= VTCR_PS(_AC(0x5,ULL));
>>> +		break;
>>> +	case 52:
>>> +		reg |= VTCR_PS(_AC(0x6,ULL));
>>> +		break;
>>> +	}
>>> +
>>> +	reg |= VTCR_T0SZ(64ULL - smmu->ias);
>>> +	reg |= VTCR_SL0(0x2);
>>> +	reg |= VTCR_VS;
>>> +
>>> +	cfg->vtcr   = reg;
>>> +
>>> 	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
>>> 	if (vmid < 0)
>>> 		return vmid;
>>> +	cfg->vmid  = (u16)vmid;
>>> +
>>> +	cfg->vttbr  = page_to_maddr(cfg->domain->arch.p2m.root);
>>> +
>>> +	printk(XENLOG_DEBUG
>>> +		   "SMMUv3: d%u: vmid 0x%x vtcr 0x%"PRIpaddr" p2maddr 0x%"PRIpaddr"\n",
>>> +		   cfg->domain->domain_id, cfg->vmid, cfg->vtcr, cfg->vttbr);
>>> 
>>> -	vtcr = &pgtbl_cfg->arm_lpae_s2_cfg.vtcr;
>>> -	cfg->vmid	= (u16)vmid;
>>> -	cfg->vttbr	= pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
>>> -	cfg->vtcr	= FIELD_PREP(STRTAB_STE_2_VTCR_S2T0SZ, vtcr->tsz) |
>>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SL0, vtcr->sl) |
>>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2IR0, vtcr->irgn) |
>>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2OR0, vtcr->orgn) |
>>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SH0, vtcr->sh) |
>>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2TG, vtcr->tg) |
>>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2PS, vtcr->ps);
>>> 	return 0;
>>> }
>>> 
>>> @@ -1382,28 +1689,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>>> 				    struct arm_smmu_master *master)
>>> {
>>> 	int ret;
>>> -	unsigned long ias, oas;
>>> -	int (*finalise_stage_fn)(struct arm_smmu_domain *,
>>> -				 struct arm_smmu_master *,
>>> -				 struct io_pgtable_cfg *);
>>> 	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>>> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>>> 
>>> 	/* Restrict the stage to what we can actually support */
>>> 	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
>>> 
>>> -	switch (smmu_domain->stage) {
>>> -	case ARM_SMMU_DOMAIN_NESTED:
>>> -	case ARM_SMMU_DOMAIN_S2:
>>> -		ias = smmu->ias;
>>> -		oas = smmu->oas;
>>> -		finalise_stage_fn = arm_smmu_domain_finalise_s2;
>>> -		break;
>>> -	default:
>>> -		return -EINVAL;
>>> -	}
>>> -
>>> -	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
>>> +	ret = arm_smmu_domain_finalise_s2(smmu_domain, master);
>>> 	if (ret < 0) {
>>> 		return ret;
>>> 	}
>>> @@ -1553,7 +1844,8 @@ static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>>> 		return -ENOMEM;
>>> 	}
>>> 
>>> -	if (!WARN_ON(q->base_dma & (qsz - 1))) {
>>> +	WARN_ON(q->base_dma & (qsz - 1));
>>> +	if (unlikely(q->base_dma & (qsz - 1))) {
>>> 		dev_info(smmu->dev, "allocated %u entries for %s\n",
>>> 			 1 << q->llq.max_n_shift, name);
>> 
>> We don't need both the WARNING and the dev_info. you could turn the
>> dev_warn / XENLOG_WARNING.
>> 
> Ack.
> 
>> 
>>> 	}
>>> @@ -1758,9 +2050,7 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>>> 	/* Request interrupt lines */
>>> 	irq = smmu->evtq.q.irq;
>>> 	if (irq) {
>>> -		ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
>>> -						arm_smmu_evtq_thread,
>>> -						IRQF_ONESHOT,
>>> +		ret = request_irq(irq, 0, arm_smmu_evtq_irq_tasklet,
>>> 						"arm-smmu-v3-evtq", smmu);
>>> 		if (ret < 0)
>>> 			dev_warn(smmu->dev, "failed to enable evtq irq\n");
>>> @@ -1770,8 +2060,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>>> 
>>> 	irq = smmu->gerr_irq;
>>> 	if (irq) {
>>> -		ret = devm_request_irq(smmu->dev, irq, arm_smmu_gerror_handler,
>>> -				       0, "arm-smmu-v3-gerror", smmu);
>>> +		ret = request_irq(irq, 0, arm_smmu_gerror_handler,
>>> +						"arm-smmu-v3-gerror", smmu);
>>> 		if (ret < 0)
>>> 			dev_warn(smmu->dev, "failed to enable gerror irq\n");
>>> 	} else {
>>> @@ -1781,11 +2071,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>>> 	if (smmu->features & ARM_SMMU_FEAT_PRI) {
>>> 		irq = smmu->priq.q.irq;
>>> 		if (irq) {
>>> -			ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
>>> -							arm_smmu_priq_thread,
>>> -							IRQF_ONESHOT,
>>> -							"arm-smmu-v3-priq",
>>> -							smmu);
>>> +			ret = request_irq(irq, 0, arm_smmu_priq_irq_tasklet,
>>> +							"arm-smmu-v3-priq", smmu);
>>> 			if (ret < 0)
>>> 				dev_warn(smmu->dev,
>>> 					 "failed to enable priq irq\n");
>>> @@ -1814,11 +2101,8 @@ static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
>>> 		 * Cavium ThunderX2 implementation doesn't support unique irq
>>> 		 * lines. Use a single irq line for all the SMMUv3 interrupts.
>>> 		 */
>>> -		ret = devm_request_threaded_irq(smmu->dev, irq,
>>> -					arm_smmu_combined_irq_handler,
>>> -					arm_smmu_combined_irq_thread,
>>> -					IRQF_ONESHOT,
>>> -					"arm-smmu-v3-combined-irq", smmu);
>>> +		ret = request_irq(irq, 0, arm_smmu_combined_irq_handler,
>>> +						"arm-smmu-v3-combined-irq", smmu);
>>> 		if (ret < 0)
>>> 			dev_warn(smmu->dev, "failed to enable combined irq\n");
>>> 	} else
>>> @@ -1857,7 +2141,7 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>>> 	reg = readl_relaxed(smmu->base + ARM_SMMU_CR0);
>>> 	if (reg & CR0_SMMUEN) {
>>> 		dev_warn(smmu->dev, "SMMU currently enabled! Resetting...\n");
>>> -		WARN_ON(is_kdump_kernel() && !disable_bypass);
>>> +		WARN_ON(!disable_bypass);
>>> 		arm_smmu_update_gbpa(smmu, GBPA_ABORT, 0);
>>> 	}
>>> 
>>> @@ -1952,8 +2236,11 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>>> 		return ret;
>>> 	}
>>> 
>>> -	if (is_kdump_kernel())
>>> -		enables &= ~(CR0_EVTQEN | CR0_PRIQEN);
>>> +	/* Initialize tasklets for threaded IRQs*/
>>> +	tasklet_init(&smmu->evtq_irq_tasklet, arm_smmu_evtq_thread, smmu);
>>> +	tasklet_init(&smmu->priq_irq_tasklet, arm_smmu_priq_thread, smmu);
>>> +	tasklet_init(&smmu->combined_irq_tasklet, arm_smmu_combined_irq_thread,
>>> +				 smmu);
>>> 
>>> 	/* Enable the SMMU interface, or ensure bypass */
>>> 	if (!bypass || disable_bypass) {
>>> @@ -2195,7 +2482,7 @@ static inline int arm_smmu_device_acpi_probe(struct platform_device *pdev,
>>> static int arm_smmu_device_dt_probe(struct platform_device *pdev,
>>> 				    struct arm_smmu_device *smmu)
>>> {
>>> -	struct device *dev = &pdev->dev;
>>> +	struct device *dev = pdev;
>>> 	u32 cells;
>>> 	int ret = -EINVAL;
>>> 
>>> @@ -2219,130 +2506,449 @@ static unsigned long arm_smmu_resource_size(struct arm_smmu_device *smmu)
>>> 		return SZ_128K;
>>> }
>>> 
>>> +/* Start of Xen specific code. */
>>> static int arm_smmu_device_probe(struct platform_device *pdev)
>>> {
>>> -	int irq, ret;
>>> -	struct resource *res;
>>> -	resource_size_t ioaddr;
>>> -	struct arm_smmu_device *smmu;
>>> -	struct device *dev = &pdev->dev;
>>> -	bool bypass;
>>> +    int irq, ret;
>>> +    paddr_t ioaddr, iosize;
>>> +    struct arm_smmu_device *smmu;
>>> +    bool bypass;
>>> +
>>> +    smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
>>> +    if ( !smmu )
>>> +    {
>>> +        dev_err(pdev, "failed to allocate arm_smmu_device\n");
>>> +        return -ENOMEM;
>>> +    }
>>> +    smmu->dev = pdev;
>>> +
>>> +    if ( pdev->of_node )
>>> +    {
>>> +        ret = arm_smmu_device_dt_probe(pdev, smmu);
>>> +    } else
>>> +    {
>> 
>> coding style
> 
> Ack. 
>> 
>> 
>>> +        ret = arm_smmu_device_acpi_probe(pdev, smmu);
>>> +        if ( ret == -ENODEV )
>>> +            return ret;
>>> +    }
>>> +
>>> +    /* Set bypass mode according to firmware probing result */
>>> +    bypass = !!ret;
>>> +
>>> +    /* Base address */
>>> +    ret = dt_device_get_address(dev_to_dt(pdev), 0, &ioaddr, &iosize);
>>> +    if( ret )
>>> +        return -ENODEV;
>>> +
>>> +    if ( iosize < arm_smmu_resource_size(smmu) )
>>> +    {
>>> +        dev_err(pdev, "MMIO region too small (%lx)\n", iosize);
>>> +        return -EINVAL;
>>> +    }
>>> +
>>> +    /*
>>> +     * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
>>> +     * the PMCG registers which are reserved by the PMU driver.
>> 
>> Does this apply to Xen too?
> 
> Yes. Performance monitoring facilities are optional. Currently there is no plan to support also in XEN.

I missed to add that in next version of the patch I will use the ARM_SMMU_REG_SZ  as size as also suggested by Julien.

Regards,
Rahul
>> 
>> 
>>> +     */
>>> +    smmu->base = ioremap_nocache(ioaddr, iosize);
>>> +    if ( IS_ERR(smmu->base) )
>>> +        return PTR_ERR(smmu->base);
>>> +
>>> +    if ( iosize > SZ_64K )
>>> +    {
>>> +        smmu->page1 = ioremap_nocache(ioaddr + SZ_64K, ARM_SMMU_REG_SZ);
>>> +        if ( IS_ERR(smmu->page1) )
>>> +            return PTR_ERR(smmu->page1);
>>> +    }
>>> +    else
>>> +    {
>>> +        smmu->page1 = smmu->base;
>>> +    }
>>> +
>>> +    /* Interrupt lines */
>>> +
>>> +    irq = platform_get_irq_byname_optional(pdev, "combined");
>>> +    if ( irq > 0 )
>>> +        smmu->combined_irq = irq;
>>> +    else
>>> +    {
>>> +        irq = platform_get_irq_byname_optional(pdev, "eventq");
>>> +        if ( irq > 0 )
>>> +            smmu->evtq.q.irq = irq;
>>> +
>>> +        irq = platform_get_irq_byname_optional(pdev, "priq");
>>> +        if ( irq > 0 )
>>> +            smmu->priq.q.irq = irq;
>>> +
>>> +        irq = platform_get_irq_byname_optional(pdev, "gerror");
>>> +        if ( irq > 0 )
>>> +            smmu->gerr_irq = irq;
>>> +    }
>>> +    /* Probe the h/w */
>>> +    ret = arm_smmu_device_hw_probe(smmu);
>>> +    if ( ret )
>>> +        return ret;
>>> +
>>> +    /* Initialise in-memory data structures */
>>> +    ret = arm_smmu_init_structures(smmu);
>>> +    if ( ret )
>>> +        return ret;
>>> +
>>> +    /* Reset the device */
>>> +    ret = arm_smmu_device_reset(smmu, bypass);
>>> +    if ( ret )
>>> +        return ret;
>>> +
>>> +    /*
>>> +     * Keep a list of all probed devices. This will be used to query
>>> +     * the smmu devices based on the fwnode.
>>> +     */
>>> +    INIT_LIST_HEAD(&smmu->devices);
>>> +
>>> +    spin_lock(&arm_smmu_devices_lock);
>>> +    list_add(&smmu->devices, &arm_smmu_devices);
>>> +    spin_unlock(&arm_smmu_devices_lock);
>>> +
>>> +    return 0;
>>> +}


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-02 16:47     ` Julien Grall
@ 2020-12-03  4:13       ` Stefano Stabellini
  2020-12-03 14:40         ` Rahul Singh
  0 siblings, 1 reply; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-03  4:13 UTC (permalink / raw)
  To: Julien Grall
  Cc: Stefano Stabellini, Rahul Singh, xen-devel, bertrand.marquis,
	Andrew Cooper, George Dunlap, Ian Jackson, Jan Beulich, Wei Liu,
	Paul Durrant, Volodymyr Babchuk

On Wed, 2 Dec 2020, Julien Grall wrote:
> On 02/12/2020 02:51, Stefano Stabellini wrote:
> > On Thu, 26 Nov 2020, Rahul Singh wrote:
> > > +/* Alias to Xen device tree helpers */
> > > +#define device_node dt_device_node
> > > +#define of_phandle_args dt_phandle_args
> > > +#define of_device_id dt_device_match
> > > +#define of_match_node dt_match_node
> > > +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np,
> > > pname, out))
> > > +#define of_property_read_bool dt_property_read_bool
> > > +#define of_parse_phandle_with_args dt_parse_phandle_with_args
> > 
> > Given all the changes to the file by the previous patches we are
> > basically fully (or almost fully) adapting this code to Xen.
> > 
> > So at that point I wonder if we should just as well make these changes
> > (e.g. s/of_phandle_args/dt_phandle_args/g) to the code too.
> 
> I have already accepted the fact that keeping Linux code as-is is nearly
> impossible without much workaround :). The benefits tends to also limited as
> we noticed for the SMMU driver.
> 
> I would like to point out that this may make quite difficult (if not
> impossible) to revert the previous patches which remove support for some
> features (e.g. atomic, MSI, ATS).
> 
> If we are going to adapt the code to Xen (I'd like to keep Linux code style
> though), then I think we should consider to keep code that may be useful in
> the near future (at least MSI, ATS).

(I am fine with keeping the Linux code style.)

We could try to keep the code as similar to Linux as possible. This
didn't work out in the past.

Otherwise, we could fully adapt the driver to Xen. If we fully adapt the
driver to Xen (code style aside) it is better to be consistent and also
do substitutions like s/of_phandle_args/dt_phandle_args/g. Then the
policy becomes clear: the code comes from Linux but it is 100% adapted
to Xen.


Now the question about what to do about the MSI and ATS code is a good
one. We know that we are going to want that code at some point in the
next 2 years. Like you wrote, if we fully adapt the code to Xen and
remove MSI and ATS code, then it is going to be harder to add it back.

So maybe keeping the MSI and ATS code for now, even if it cannot work,
would be better. I think this strategy works well if the MSI and ATS
code can be disabled easily, i.e. with a couple of lines of code in the
init function rather than #ifdef everywhere. It doesn't work well if we
have to add #ifdef everywhere.

It looks like MSI could be disabled adding a couple of lines to
arm_smmu_setup_msis.

Similarly ATS seems to be easy to disable by forcing ats_enabled to
false.

So yes, this looks like a good way forward. Rahul, what do you think?


 
> > > +#define FIELD_GET(_mask, _reg)          \
> > > +    (typeof(_mask))(((_reg) & (_mask)) >> (__builtin_ffsll(_mask) - 1))
> > > +
> > > +#define WRITE_ONCE(x, val)                  \
> > > +do {                                        \
> > > +    *(volatile typeof(x) *)&(x) = (val);    \
> > > +} while (0)
> > 
> > maybe we should define this in xen/include/xen/lib.h
> 
> I have attempted such discussion in the past and this resulted to more
> bikeshed than it is worth it. So I would suggest to re-implement WRITE_ONCE()
> as write_atomic() for now.

Good suggestion, less discussions more code :-)


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 2/8] xen/arm: revert atomic operation related command-queue insertion patch
  2020-12-02 13:44   ` Julien Grall
@ 2020-12-03 11:49     ` Rahul Singh
  0 siblings, 0 replies; 53+ messages in thread
From: Rahul Singh @ 2020-12-03 11:49 UTC (permalink / raw)
  To: Julien Grall
  Cc: xen-devel, Bertrand Marquis, Stefano Stabellini, Volodymyr Babchuk

Hello Julien,

Thanks for reviewing the code.

> On 2 Dec 2020, at 1:44 pm, Julien Grall <julien@xen.org> wrote:
> 
> Hi Rahul,
> 
> On 26/11/2020 17:02, Rahul Singh wrote:
>> Linux SMMUv3 code implements the commands-queue insertion based on
>> atomic operations implemented in Linux. Atomic functions used by the
>> commands-queue insertion is not implemented in XEN therefore revert the
>> patch that implemented the commands-queue insertion based on atomic
>> operations.
> 
> This commit message explains why we revert but not the consequence of the revert. Can outline if there are any and why they are fine?

Ok let me try to add more detail.
> 
> I am also interested to have a list of *must* have for the driver to be out of the tech preview.

Ok let me add more informing in the commit message in next version of the patch.

Regards,
Rahul

> 
> Cheers,
> 
> -- 
> Julien Grall



^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 3/8] xen/arm: revert patch related to XArray
  2020-12-02 13:46   ` Julien Grall
@ 2020-12-03 12:57     ` Rahul Singh
  2020-12-04  8:52       ` Julien Grall
  0 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-12-03 12:57 UTC (permalink / raw)
  To: Julien Grall
  Cc: xen-devel, Bertrand Marquis, Stefano Stabellini, Volodymyr Babchuk

Hello Julien,

> On 2 Dec 2020, at 1:46 pm, Julien Grall <julien@xen.org> wrote:
> 
> Hi Rahul,
> 
> On 26/11/2020 17:02, Rahul Singh wrote:
>> XArray is not implemented in XEN revert the patch that introduce the
>> XArray code in SMMUv3 driver.
> 
> Similar to the atomic revert, you are explaining why the revert but not the consequences. I think this is quite important to have them outline in the commit message as it looks like it means the SMMU driver would not scale.

Ok I will add.
> 
>> Once XArray is implemented in XEN this patch can be added in XEN.
> 
> What's the plan for that?

As of now there is no plan for Xarray from our side.  
Do we have requirement for Xarray implementation in XEN ?

Regards,
Rahul

> 
> Cheers,
> 
> -- 
> Julien Grall



^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 4/8] xen/arm: Remove support for MSI on SMMUv3
  2020-12-02 14:11         ` Julien Grall
@ 2020-12-03 12:59           ` Rahul Singh
  0 siblings, 0 replies; 53+ messages in thread
From: Rahul Singh @ 2020-12-03 12:59 UTC (permalink / raw)
  To: Julien Grall
  Cc: Stefano Stabellini, xen-devel, Bertrand Marquis, Volodymyr Babchuk

Hello Julien,

> On 2 Dec 2020, at 2:11 pm, Julien Grall <julien@xen.org> wrote:
> 
> Hi Rahul,
> 
> On 02/12/2020 13:12, Rahul Singh wrote:
>> Hello Stefano,
>>> On 2 Dec 2020, at 12:40 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
>>> 
>>> On Tue, 1 Dec 2020, Stefano Stabellini wrote:
>>>> On Thu, 26 Nov 2020, Rahul Singh wrote:
>>>>> XEN does not support MSI on ARM platforms, therefore remove the MSI
>>>>> support from SMMUv3 driver.
>>>>> 
>>>>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
>>>> 
>>>> I wonder if it makes sense to #ifdef CONFIG_MSI this code instead of
>>>> removing it completely.
>>> 
>>> One more thought: could this patch be achieved by reverting
>>> 166bdbd23161160f2abcea70621adba179050bee ? If this patch could be done
>>> by a couple of revert, it would be great to say it in the commit
>>> message.
>>> 
>>  Ok will add in next version.
>>> 
>>>> In the past, we tried to keep the entire file as textually similar to
>>>> the original Linux driver as possible to make it easier to backport
>>>> features and fixes. So, in this case we would probably not even used an
>>>> #ifdef but maybe something like:
>>>> 
>>>>  if (/* msi_enabled */ 0)
>>>>      return;
>>>> 
>>>> at the beginning of arm_smmu_setup_msis.
>>>> 
>>>> 
>>>> However, that strategy didn't actually work very well because backports
>>>> have proven difficult to do anyway. So at that point we might as well at
>>>> least have clean code in Xen and do the changes properly.
> 
> It was difficult because Linux decided to rework how IOMMU drivers works. I agree the risk is still there and therefore clean code would be better with some caveats (see below).
> 
>> Main reason to remove the feature/code that is not usable in XEN is to have a clean code.
> 
> I agree that short term this feature will not be usable. However, I think we need to think about {medium, long}-term plan to avoid extra effort in the future because the driver evolve in a way making the revert of revert impossible.
> 
> Therefore I would prefer to keep both the MSI and PCI ATS present as they are going to be useful/necessary on some platforms. It doesn't matter that they don't work because the driver will be in tech preview.

Ok. As Stefano also agreed the same I will keep the PC ATS and MSI functionality in next version.

Regards,
Rahul

> 
> Cheers,
> 
> -- 
> Julien Grall



^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN
  2020-12-02 14:45   ` Julien Grall
@ 2020-12-03 14:33     ` Rahul Singh
  2020-12-04  9:05       ` Julien Grall
  0 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-12-03 14:33 UTC (permalink / raw)
  To: Julien Grall
  Cc: xen-devel, Bertrand Marquis, Stefano Stabellini, Volodymyr Babchuk

Hello Julien,

> On 2 Dec 2020, at 2:45 pm, Julien Grall <julien@xen.org> wrote:
> 
> 
> 
> On 26/11/2020 17:02, Rahul Singh wrote:
>> struct io_pgtable_ops, struct io_pgtable_cfg, struct iommu_flush_ops,
>> and struct iommu_ops related code are linux specific.
> 
> So the assumption is we are going to support only sharing with the P2M and the IOMMU. That's probably fine short term, but long-term we are going to need unsharing the page-tables (there are issues on some platforms if the ITS doorbell is accessed by the processors).
> 
> I am ok with removing anything related to the unsharing code. But I think it should be clarified here.

Ok I will add in commit message what is removed from the code.
> 
>> Remove code related to above struct as code is dead code in XEN.
>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
>> ---
>>  xen/drivers/passthrough/arm/smmu-v3.c | 457 --------------------------
>>  1 file changed, 457 deletions(-)
>> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
>> index 40e3890a58..55d1cba194 100644
>> --- a/xen/drivers/passthrough/arm/smmu-v3.c
>> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
>> @@ -402,13 +402,7 @@
>>  #define ARM_SMMU_CMDQ_SYNC_TIMEOUT_US	1000000 /* 1s! */
>>  #define ARM_SMMU_CMDQ_SYNC_SPIN_COUNT	10
>>  -#define MSI_IOVA_BASE			0x8000000
>> -#define MSI_IOVA_LENGTH			0x100000
>> -
>>  static bool disable_bypass = 1;
>> -module_param_named(disable_bypass, disable_bypass, bool, S_IRUGO);
>> -MODULE_PARM_DESC(disable_bypass,
>> -	"Disable bypass streams such that incoming transactions from devices that are not attached to an iommu domain will report an abort back to the device and will not be allowed to pass through the SMMU.");
> 
> Per your commit message, this doesn't look related to what you intend to remove. Maybe your commit message should be updated?

Ok.
> 
>>    enum pri_resp {
>>  	PRI_RESP_DENY = 0,
>> @@ -599,7 +593,6 @@ struct arm_smmu_domain {
>>  	struct arm_smmu_device		*smmu;
>>  	struct mutex			init_mutex; /* Protects smmu pointer */
>>  -	struct io_pgtable_ops		*pgtbl_ops;
>>  	bool				non_strict;
>>    	enum arm_smmu_domain_stage	stage;
>> @@ -1297,74 +1290,6 @@ static void arm_smmu_tlb_inv_context(void *cookie)
>>  	arm_smmu_cmdq_issue_sync(smmu);
>>  }
>>  -static void arm_smmu_tlb_inv_range_nosync(unsigned long iova, size_t size,
>> -					  size_t granule, bool leaf, void *cookie)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = cookie;
>> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> -	struct arm_smmu_cmdq_ent cmd = {
>> -		.tlbi = {
>> -			.leaf	= leaf,
>> -			.addr	= iova,
>> -		},
>> -	};
>> -
>> -	cmd.opcode	= CMDQ_OP_TLBI_S2_IPA;
>> -	cmd.tlbi.vmid	= smmu_domain->s2_cfg.vmid;
>> -
>> -	do {
>> -		arm_smmu_cmdq_issue_cmd(smmu, &cmd);
>> -		cmd.tlbi.addr += granule;
>> -	} while (size -= granule);
>> -}
>> -
>> -static void arm_smmu_tlb_inv_page_nosync(struct iommu_iotlb_gather *gather,
>> -					 unsigned long iova, size_t granule,
>> -					 void *cookie)
>> -{
>> -	arm_smmu_tlb_inv_range_nosync(iova, granule, granule, true, cookie);
>> -}
>> -
>> -static void arm_smmu_tlb_inv_walk(unsigned long iova, size_t size,
>> -				  size_t granule, void *cookie)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = cookie;
>> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> -
>> -	arm_smmu_tlb_inv_range_nosync(iova, size, granule, false, cookie);
>> -	arm_smmu_cmdq_issue_sync(smmu);
>> -}
>> -
>> -static void arm_smmu_tlb_inv_leaf(unsigned long iova, size_t size,
>> -				  size_t granule, void *cookie)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = cookie;
>> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>> -
>> -	arm_smmu_tlb_inv_range_nosync(iova, size, granule, true, cookie);
>> -	arm_smmu_cmdq_issue_sync(smmu);
>> -}
>> -
>> -static const struct iommu_flush_ops arm_smmu_flush_ops = {
>> -	.tlb_flush_all	= arm_smmu_tlb_inv_context,
>> -	.tlb_flush_walk = arm_smmu_tlb_inv_walk,
>> -	.tlb_flush_leaf = arm_smmu_tlb_inv_leaf,
>> -	.tlb_add_page	= arm_smmu_tlb_inv_page_nosync,
>> -};
>> -
>> -/* IOMMU API */
>> -static bool arm_smmu_capable(enum iommu_cap cap)
>> -{
>> -	switch (cap) {
>> -	case IOMMU_CAP_CACHE_COHERENCY:
>> -		return true;
>> -	case IOMMU_CAP_NOEXEC:
>> -		return true;
>> -	default:
>> -		return false;
>> -	}
>> -}
>> -
>>  static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
>>  {
>>  	struct arm_smmu_domain *smmu_domain;
>> @@ -1421,7 +1346,6 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>>  	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>>    	iommu_put_dma_cookie(domain);
>> -	free_io_pgtable_ops(smmu_domain->pgtbl_ops);
>>    	if (cfg->vmid)
>>  		arm_smmu_bitmap_free(smmu->vmid_map, cfg->vmid);
>> @@ -1429,7 +1353,6 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>>  	kfree(smmu_domain);
>>  }
>>  -
> 
> Looks like a spurious change.

Ack. 
> 
>>  static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>>  				       struct arm_smmu_master *master,
>>  				       struct io_pgtable_cfg *pgtbl_cfg)
> 
> You commit message leads to think that all the use of struct io_pgtable_cfg will be removed.

Ok I will remove it.
> 
>> @@ -1437,7 +1360,6 @@ static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>>  	int vmid;
>>  	struct arm_smmu_device *smmu = smmu_domain->smmu;
>>  	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>> -	typeof(&pgtbl_cfg->arm_lpae_s2_cfg.vtcr) vtcr;
> 
> It feels a bit odd that the definition of 'vtcr' is removed but there are still users.

Ack. Will fix that.
> 
>>    	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
>>  	if (vmid < 0)
>> @@ -1461,20 +1383,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>>  {
>>  	int ret;
>>  	unsigned long ias, oas;
>> -	enum io_pgtable_fmt fmt;
>> -	struct io_pgtable_cfg pgtbl_cfg;
>> -	struct io_pgtable_ops *pgtbl_ops;
>>  	int (*finalise_stage_fn)(struct arm_smmu_domain *,
>>  				 struct arm_smmu_master *,
>>  				 struct io_pgtable_cfg *);
>>  	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>>  	struct arm_smmu_device *smmu = smmu_domain->smmu;
>>  -	if (domain->type == IOMMU_DOMAIN_IDENTITY) {
>> -		smmu_domain->stage = ARM_SMMU_DOMAIN_BYPASS;
>> -		return 0;
>> -	}
>> -
> 
> Per your commit message, this doesn't look related to what you intend to remove. Maybe your commit message should be updated?

Ok I will update the commit message.
> 
>>  	/* Restrict the stage to what we can actually support */
>>  	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
>>  @@ -1483,40 +1397,17 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>>  	case ARM_SMMU_DOMAIN_S2:
>>  		ias = smmu->ias;
>>  		oas = smmu->oas;
>> -		fmt = ARM_64_LPAE_S2;
>>  		finalise_stage_fn = arm_smmu_domain_finalise_s2;
>>  		break;
>>  	default:
>>  		return -EINVAL;
>>  	}
>>  -	pgtbl_cfg = (struct io_pgtable_cfg) {
>> -		.pgsize_bitmap	= smmu->pgsize_bitmap,
>> -		.ias		= ias,
>> -		.oas		= oas,
>> -		.coherent_walk	= smmu->features & ARM_SMMU_FEAT_COHERENCY,
>> -		.tlb		= &arm_smmu_flush_ops,
>> -		.iommu_dev	= smmu->dev,
>> -	};
>> -
>> -	if (smmu_domain->non_strict)
>> -		pgtbl_cfg.quirks |= IO_PGTABLE_QUIRK_NON_STRICT;
>> -
>> -	pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
>> -	if (!pgtbl_ops)
>> -		return -ENOMEM;
>> -
>> -	domain->pgsize_bitmap = pgtbl_cfg.pgsize_bitmap;
>> -	domain->geometry.aperture_end = (1UL << pgtbl_cfg.ias) - 1;
>> -	domain->geometry.force_aperture = true;
>> -
>>  	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
>>  	if (ret < 0) {
>> -		free_io_pgtable_ops(pgtbl_ops);
>>  		return ret;
>>  	}
>>  -	smmu_domain->pgtbl_ops = pgtbl_ops;
>>  	return 0;
>>  }
>>  @@ -1626,71 +1517,6 @@ out_unlock:
>>  	return ret;
>>  }
>>  -static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
>> -			phys_addr_t paddr, size_t size, int prot, gfp_t gfp)
>> -{
>> -	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
>> -
>> -	if (!ops)
>> -		return -ENODEV;
>> -
>> -	return ops->map(ops, iova, paddr, size, prot, gfp);
>> -}
>> -
>> -static size_t arm_smmu_unmap(struct iommu_domain *domain, unsigned long iova,
>> -			     size_t size, struct iommu_iotlb_gather *gather)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> -	struct io_pgtable_ops *ops = smmu_domain->pgtbl_ops;
>> -
>> -	if (!ops)
>> -		return 0;
>> -
>> -	return ops->unmap(ops, iova, size, gather);
>> -}
>> -
>> -static void arm_smmu_flush_iotlb_all(struct iommu_domain *domain)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> -
>> -	if (smmu_domain->smmu)
>> -		arm_smmu_tlb_inv_context(smmu_domain);
>> -}
>> -
>> -static void arm_smmu_iotlb_sync(struct iommu_domain *domain,
>> -				struct iommu_iotlb_gather *gather)
>> -{
>> -	struct arm_smmu_device *smmu = to_smmu_domain(domain)->smmu;
>> -
>> -	if (smmu)
>> -		arm_smmu_cmdq_issue_sync(smmu);
>> -}
>> -
>> -static phys_addr_t
>> -arm_smmu_iova_to_phys(struct iommu_domain *domain, dma_addr_t iova)
>> -{
>> -	struct io_pgtable_ops *ops = to_smmu_domain(domain)->pgtbl_ops;
>> -
>> -	if (domain->type == IOMMU_DOMAIN_IDENTITY)
>> -		return iova;
>> -
>> -	if (!ops)
>> -		return 0;
>> -
>> -	return ops->iova_to_phys(ops, iova);
>> -}
>> -
>> -static struct platform_driver arm_smmu_driver;
>> -
>> -static
>> -struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle *fwnode)
>> -{
>> -	struct device *dev = driver_find_device_by_fwnode(&arm_smmu_driver.driver,
>> -							  fwnode);
>> -	put_device(dev);
>> -	return dev ? dev_get_drvdata(dev) : NULL;
>> -}
>> -
>>  static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
>>  {
>>  	unsigned long limit = smmu->strtab_cfg.num_l1_ents;
>> @@ -1701,206 +1527,6 @@ static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32 sid)
>>  	return sid < limit;
>>  }
>>  -static struct iommu_ops arm_smmu_ops;
>> -
>> -static struct iommu_device *arm_smmu_probe_device(struct device *dev)
>> -{
> 
> Most of the code here looks useful to Xen. I think you want to keep the code and re-use it afterwards.

Ok. I removed the code here and added the XEN compatible code to add devices in next patch.
I will keep it in this patch and will modifying the code to make XEN compatible.

> 
>> -	int i, ret;
>> -	struct arm_smmu_device *smmu;
>> -	struct arm_smmu_master *master;
>> -	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> -
>> -	if (!fwspec || fwspec->ops != &arm_smmu_ops)
>> -		return ERR_PTR(-ENODEV);
>> -
>> -	if (WARN_ON_ONCE(dev_iommu_priv_get(dev)))
>> -		return ERR_PTR(-EBUSY);
>> -
>> -	smmu = arm_smmu_get_by_fwnode(fwspec->iommu_fwnode);
>> -	if (!smmu)
>> -		return ERR_PTR(-ENODEV);
>> -
>> -	master = kzalloc(sizeof(*master), GFP_KERNEL);
>> -	if (!master)
>> -		return ERR_PTR(-ENOMEM);
>> -
>> -	master->dev = dev;
>> -	master->smmu = smmu;
>> -	master->sids = fwspec->ids;
>> -	master->num_sids = fwspec->num_ids;
>> -	dev_iommu_priv_set(dev, master);
>> -
>> -	/* Check the SIDs are in range of the SMMU and our stream table */
>> -	for (i = 0; i < master->num_sids; i++) {
>> -		u32 sid = master->sids[i];
>> -
>> -		if (!arm_smmu_sid_in_range(smmu, sid)) {
>> -			ret = -ERANGE;
>> -			goto err_free_master;
>> -		}
>> -
>> -		/* Ensure l2 strtab is initialised */
>> -		if (smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB) {
>> -			ret = arm_smmu_init_l2_strtab(smmu, sid);
>> -			if (ret)
>> -				goto err_free_master;
>> -		}
>> -	}
>> -
>> -	return &smmu->iommu;
>> -
>> -err_free_master:
>> -	kfree(master);
>> -	dev_iommu_priv_set(dev, NULL);
>> -	return ERR_PTR(ret);
>> -}
>> -
>> -static void arm_smmu_release_device(struct device *dev)
>> -{
>> -	struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> -	struct arm_smmu_master *master;
>> -
>> -	if (!fwspec || fwspec->ops != &arm_smmu_ops)
>> -		return;
>> -
>> -	master = dev_iommu_priv_get(dev);
>> -	arm_smmu_detach_dev(master);
>> -	kfree(master);
>> -	iommu_fwspec_free(dev);
>> -}
>> -
>> -static struct iommu_group *arm_smmu_device_group(struct device *dev)
>> -{
>> -	struct iommu_group *group;
>> -
>> -	/*
>> -	 * We don't support devices sharing stream IDs other than PCI RID
>> -	 * aliases, since the necessary ID-to-device lookup becomes rather
>> -	 * impractical given a potential sparse 32-bit stream ID space.
>> -	 */
>> -	if (dev_is_pci(dev))
>> -		group = pci_device_group(dev);
>> -	else
>> -		group = generic_device_group(dev);
>> -
>> -	return group;
>> -}
>> -
>> -static int arm_smmu_domain_get_attr(struct iommu_domain *domain,
>> -				    enum iommu_attr attr, void *data)
>> -{
>> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> -
>> -	switch (domain->type) {
>> -	case IOMMU_DOMAIN_UNMANAGED:
>> -		switch (attr) {
>> -		case DOMAIN_ATTR_NESTING:
>> -			*(int *)data = (smmu_domain->stage == ARM_SMMU_DOMAIN_NESTED);
>> -			return 0;
>> -		default:
>> -			return -ENODEV;
>> -		}
>> -		break;
>> -	case IOMMU_DOMAIN_DMA:
>> -		switch (attr) {
>> -		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
>> -			*(int *)data = smmu_domain->non_strict;
>> -			return 0;
>> -		default:
>> -			return -ENODEV;
>> -		}
>> -		break;
>> -	default:
>> -		return -EINVAL;
>> -	}
>> -}
>> -
>> -static int arm_smmu_domain_set_attr(struct iommu_domain *domain,
>> -				    enum iommu_attr attr, void *data)
>> -{
>> -	int ret = 0;
>> -	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> -
>> -	mutex_lock(&smmu_domain->init_mutex);
>> -
>> -	switch (domain->type) {
>> -	case IOMMU_DOMAIN_UNMANAGED:
>> -		switch (attr) {
>> -		case DOMAIN_ATTR_NESTING:
>> -			if (smmu_domain->smmu) {
>> -				ret = -EPERM;
>> -				goto out_unlock;
>> -			}
>> -
>> -			if (*(int *)data)
>> -				smmu_domain->stage = ARM_SMMU_DOMAIN_NESTED;
>> -			else
>> -				smmu_domain->stage = ARM_SMMU_DOMAIN_S1;
>> -			break;
>> -		default:
>> -			ret = -ENODEV;
>> -		}
>> -		break;
>> -	case IOMMU_DOMAIN_DMA:
>> -		switch(attr) {
>> -		case DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE:
>> -			smmu_domain->non_strict = *(int *)data;
>> -			break;
>> -		default:
>> -			ret = -ENODEV;
>> -		}
>> -		break;
>> -	default:
>> -		ret = -EINVAL;
>> -	}
>> -
>> -out_unlock:
>> -	mutex_unlock(&smmu_domain->init_mutex);
>> -	return ret;
>> -}
>> -
>> -static int arm_smmu_of_xlate(struct device *dev, struct of_phandle_args *args)
>> -{
>> -	return iommu_fwspec_add_ids(dev, args->args, 1);
>> -}
>> -
>> -static void arm_smmu_get_resv_regions(struct device *dev,
>> -				      struct list_head *head)
>> -{
>> -	struct iommu_resv_region *region;
>> -	int prot = IOMMU_WRITE | IOMMU_NOEXEC | IOMMU_MMIO;
>> -
>> -	region = iommu_alloc_resv_region(MSI_IOVA_BASE, MSI_IOVA_LENGTH,
>> -					 prot, IOMMU_RESV_SW_MSI);
>> -	if (!region)
>> -		return;
>> -
>> -	list_add_tail(&region->list, head);
>> -
>> -	iommu_dma_get_resv_regions(dev, head);
>> -}
>> -
>> -static struct iommu_ops arm_smmu_ops = {
>> -	.capable		= arm_smmu_capable,
>> -	.domain_alloc		= arm_smmu_domain_alloc,
>> -	.domain_free		= arm_smmu_domain_free,
>> -	.attach_dev		= arm_smmu_attach_dev,
>> -	.map			= arm_smmu_map,
>> -	.unmap			= arm_smmu_unmap,
>> -	.flush_iotlb_all	= arm_smmu_flush_iotlb_all,
>> -	.iotlb_sync		= arm_smmu_iotlb_sync,
>> -	.iova_to_phys		= arm_smmu_iova_to_phys,
>> -	.probe_device		= arm_smmu_probe_device,
>> -	.release_device		= arm_smmu_release_device,
>> -	.device_group		= arm_smmu_device_group,
>> -	.domain_get_attr	= arm_smmu_domain_get_attr,
>> -	.domain_set_attr	= arm_smmu_domain_set_attr,
>> -	.of_xlate		= arm_smmu_of_xlate,
>> -	.get_resv_regions	= arm_smmu_get_resv_regions,
>> -	.put_resv_regions	= generic_iommu_put_resv_regions,
>> -	.pgsize_bitmap		= -1UL, /* Restricted during device attach */
>> -};
>> -
>>  /* Probing and initialisation functions */
>>  static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>>  				   struct arm_smmu_queue *q,
>> @@ -2406,7 +2032,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>>  	switch (FIELD_GET(IDR0_STALL_MODEL, reg)) {
>>  	case IDR0_STALL_MODEL_FORCE:
>>  		smmu->features |= ARM_SMMU_FEAT_STALL_FORCE;
>> -		fallthrough;
> 
> We should keep all the fallthrough documented. So I think we want to introduce the fallthrough in Xen as well.

Ok I will keep fallthrough documented in this patch. 

fallthrough implementation in XEN should be another patch. I am not sure when we can implement but we will try to implement.

> 
>>  	case IDR0_STALL_MODEL_STALL:
>>  		smmu->features |= ARM_SMMU_FEAT_STALLS;
>>  	}
>> @@ -2426,7 +2051,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>>  	switch (FIELD_GET(IDR0_TTF, reg)) {
>>  	case IDR0_TTF_AARCH32_64:
>>  		smmu->ias = 40;
>> -		fallthrough;
>>  	case IDR0_TTF_AARCH64:
>>  		break;
>>  	default:
>> @@ -2515,21 +2139,10 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>>  	default:
>>  		dev_info(smmu->dev,
>>  			"unknown output address size. Truncating to 48-bit\n");
>> -		fallthrough;
>>  	case IDR5_OAS_48_BIT:
>>  		smmu->oas = 48;
>>  	}
>>  -	if (arm_smmu_ops.pgsize_bitmap == -1UL)
>> -		arm_smmu_ops.pgsize_bitmap = smmu->pgsize_bitmap;
>> -	else
>> -		arm_smmu_ops.pgsize_bitmap |= smmu->pgsize_bitmap;
>> -
>> -	/* Set the DMA mask for our table walker */
>> -	if (dma_set_mask_and_coherent(smmu->dev, DMA_BIT_MASK(smmu->oas)))
>> -		dev_warn(smmu->dev,
>> -			 "failed to set DMA mask for table walker\n");
>> -
>>  	smmu->ias = max(smmu->ias, smmu->oas);
>>    	dev_info(smmu->dev, "ias %lu-bit, oas %lu-bit (features 0x%08x)\n",
>> @@ -2595,9 +2208,6 @@ static int arm_smmu_device_dt_probe(struct platform_device *pdev,
>>    	parse_driver_options(smmu);
>>  -	if (of_dma_is_coherent(dev->of_node))
>> -		smmu->features |= ARM_SMMU_FEAT_COHERENCY;
>> -
>>  	return ret;
>>  }
>>  @@ -2609,55 +2219,6 @@ 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)
> 
> This seems a bit odd that you are removing the function but not the callers. Shouldn't you keep this function until next patch?
> 
Ok I will keep this in this patch and will remove in next patch.

>> -{
>> -	struct resource res = {
>> -		.flags = IORESOURCE_MEM,
>> -		.start = start,
>> -		.end = start + size - 1,
>> -	};
>> -
>> -	return devm_ioremap_resource(dev, &res);
>> -}
>> -
>>  static int arm_smmu_device_probe(struct platform_device *pdev)
>>  {
>>  	int irq, ret;
>> @@ -2785,21 +2346,3 @@ static const struct of_device_id arm_smmu_of_match[] = {
>>  	{ .compatible = "arm,smmu-v3", },
>>  	{ },
>>  };
>> -MODULE_DEVICE_TABLE(of, arm_smmu_of_match);
>> -
>> -static struct platform_driver arm_smmu_driver = {
>> -	.driver	= {
>> -		.name			= "arm-smmu-v3",
>> -		.of_match_table		= arm_smmu_of_match,
>> -		.suppress_bind_attrs	= true,
>> -	},
>> -	.probe	= arm_smmu_device_probe,
>> -	.remove	= arm_smmu_device_remove,
>> -	.shutdown = arm_smmu_device_shutdown,
>> -};
>> -module_platform_driver(arm_smmu_driver);
>> -
>> -MODULE_DESCRIPTION("IOMMU API for ARM architected SMMUv3 implementations");
>> -MODULE_AUTHOR("Will Deacon <will@kernel.org>");
>> -MODULE_ALIAS("platform:arm-smmu-v3");
>> -MODULE_LICENSE("GPL v2");
> 
> -- 
> Julien Grall



^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-03  4:13       ` Stefano Stabellini
@ 2020-12-03 14:40         ` Rahul Singh
  2020-12-03 18:47           ` Stefano Stabellini
  0 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-12-03 14:40 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: Julien Grall, xen-devel, Bertrand Marquis, Andrew Cooper,
	George Dunlap, Ian Jackson, Jan Beulich, Wei Liu, Paul Durrant,
	Volodymyr Babchuk

Hello Stefano,

> On 3 Dec 2020, at 4:13 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
> 
> On Wed, 2 Dec 2020, Julien Grall wrote:
>> On 02/12/2020 02:51, Stefano Stabellini wrote:
>>> On Thu, 26 Nov 2020, Rahul Singh wrote:
>>>> +/* Alias to Xen device tree helpers */
>>>> +#define device_node dt_device_node
>>>> +#define of_phandle_args dt_phandle_args
>>>> +#define of_device_id dt_device_match
>>>> +#define of_match_node dt_match_node
>>>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np,
>>>> pname, out))
>>>> +#define of_property_read_bool dt_property_read_bool
>>>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
>>> 
>>> Given all the changes to the file by the previous patches we are
>>> basically fully (or almost fully) adapting this code to Xen.
>>> 
>>> So at that point I wonder if we should just as well make these changes
>>> (e.g. s/of_phandle_args/dt_phandle_args/g) to the code too.
>> 
>> I have already accepted the fact that keeping Linux code as-is is nearly
>> impossible without much workaround :). The benefits tends to also limited as
>> we noticed for the SMMU driver.
>> 
>> I would like to point out that this may make quite difficult (if not
>> impossible) to revert the previous patches which remove support for some
>> features (e.g. atomic, MSI, ATS).
>> 
>> If we are going to adapt the code to Xen (I'd like to keep Linux code style
>> though), then I think we should consider to keep code that may be useful in
>> the near future (at least MSI, ATS).
> 
> (I am fine with keeping the Linux code style.)
> 
> We could try to keep the code as similar to Linux as possible. This
> didn't work out in the past.
> 
> Otherwise, we could fully adapt the driver to Xen. If we fully adapt the
> driver to Xen (code style aside) it is better to be consistent and also
> do substitutions like s/of_phandle_args/dt_phandle_args/g. Then the
> policy becomes clear: the code comes from Linux but it is 100% adapted
> to Xen.
> 
> 
> Now the question about what to do about the MSI and ATS code is a good
> one. We know that we are going to want that code at some point in the
> next 2 years. Like you wrote, if we fully adapt the code to Xen and
> remove MSI and ATS code, then it is going to be harder to add it back.
> 
> So maybe keeping the MSI and ATS code for now, even if it cannot work,
> would be better. I think this strategy works well if the MSI and ATS
> code can be disabled easily, i.e. with a couple of lines of code in the
> init function rather than #ifdef everywhere. It doesn't work well if we
> have to add #ifdef everywhere.
> 
> It looks like MSI could be disabled adding a couple of lines to
> arm_smmu_setup_msis.
> 
> Similarly ATS seems to be easy to disable by forcing ats_enabled to
> false.
> 
> So yes, this looks like a good way forward. Rahul, what do you think?


I am ok to have the PCI ATS and MSI functionality in the code. 
As per the discussion next version of the patch will include below modification:Please let me know if there are any suggestion overall that should be added in next version.

1. Keep the PCI ATS and MSI functionality code.
2. Make the code with XEN compatible ( remove linux compatibility functions)
3. Keep the Linux coding style for code imported from Linux.
4. Fix all comments.

I have one query what will be coding style for new code to make driver work  for XEN ? 


> 
> 
> 
>>>> +#define FIELD_GET(_mask, _reg)          \
>>>> +    (typeof(_mask))(((_reg) & (_mask)) >> (__builtin_ffsll(_mask) - 1))
>>>> +
>>>> +#define WRITE_ONCE(x, val)                  \
>>>> +do {                                        \
>>>> +    *(volatile typeof(x) *)&(x) = (val);    \
>>>> +} while (0)
>>> 
>>> maybe we should define this in xen/include/xen/lib.h
>> 
>> I have attempted such discussion in the past and this resulted to more
>> bikeshed than it is worth it. So I would suggest to re-implement WRITE_ONCE()
>> as write_atomic() for now.
> 
> Good suggestion, less discussions more code :-)

I will use the write_atomic.

Regards,
Rahul




^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-03 14:40         ` Rahul Singh
@ 2020-12-03 18:47           ` Stefano Stabellini
  2020-12-07  8:33             ` Rahul Singh
  0 siblings, 1 reply; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-03 18:47 UTC (permalink / raw)
  To: Rahul Singh
  Cc: Stefano Stabellini, Julien Grall, xen-devel, Bertrand Marquis,
	Andrew Cooper, George Dunlap, Ian Jackson, Jan Beulich, Wei Liu,
	Paul Durrant, Volodymyr Babchuk

On Thu, 3 Dec 2020, Rahul Singh wrote:
> > On 3 Dec 2020, at 4:13 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
> > On Wed, 2 Dec 2020, Julien Grall wrote:
> >> On 02/12/2020 02:51, Stefano Stabellini wrote:
> >>> On Thu, 26 Nov 2020, Rahul Singh wrote:
> >>>> +/* Alias to Xen device tree helpers */
> >>>> +#define device_node dt_device_node
> >>>> +#define of_phandle_args dt_phandle_args
> >>>> +#define of_device_id dt_device_match
> >>>> +#define of_match_node dt_match_node
> >>>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np,
> >>>> pname, out))
> >>>> +#define of_property_read_bool dt_property_read_bool
> >>>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
> >>> 
> >>> Given all the changes to the file by the previous patches we are
> >>> basically fully (or almost fully) adapting this code to Xen.
> >>> 
> >>> So at that point I wonder if we should just as well make these changes
> >>> (e.g. s/of_phandle_args/dt_phandle_args/g) to the code too.
> >> 
> >> I have already accepted the fact that keeping Linux code as-is is nearly
> >> impossible without much workaround :). The benefits tends to also limited as
> >> we noticed for the SMMU driver.
> >> 
> >> I would like to point out that this may make quite difficult (if not
> >> impossible) to revert the previous patches which remove support for some
> >> features (e.g. atomic, MSI, ATS).
> >> 
> >> If we are going to adapt the code to Xen (I'd like to keep Linux code style
> >> though), then I think we should consider to keep code that may be useful in
> >> the near future (at least MSI, ATS).
> > 
> > (I am fine with keeping the Linux code style.)
> > 
> > We could try to keep the code as similar to Linux as possible. This
> > didn't work out in the past.
> > 
> > Otherwise, we could fully adapt the driver to Xen. If we fully adapt the
> > driver to Xen (code style aside) it is better to be consistent and also
> > do substitutions like s/of_phandle_args/dt_phandle_args/g. Then the
> > policy becomes clear: the code comes from Linux but it is 100% adapted
> > to Xen.
> > 
> > 
> > Now the question about what to do about the MSI and ATS code is a good
> > one. We know that we are going to want that code at some point in the
> > next 2 years. Like you wrote, if we fully adapt the code to Xen and
> > remove MSI and ATS code, then it is going to be harder to add it back.
> > 
> > So maybe keeping the MSI and ATS code for now, even if it cannot work,
> > would be better. I think this strategy works well if the MSI and ATS
> > code can be disabled easily, i.e. with a couple of lines of code in the
> > init function rather than #ifdef everywhere. It doesn't work well if we
> > have to add #ifdef everywhere.
> > 
> > It looks like MSI could be disabled adding a couple of lines to
> > arm_smmu_setup_msis.
> > 
> > Similarly ATS seems to be easy to disable by forcing ats_enabled to
> > false.
> > 
> > So yes, this looks like a good way forward. Rahul, what do you think?
> 
> 
> I am ok to have the PCI ATS and MSI functionality in the code. 
> As per the discussion next version of the patch will include below modification:Please let me know if there are any suggestion overall that should be added in next version.
> 
> 1. Keep the PCI ATS and MSI functionality code.

I'll repeat one point I wrote above that I think it is important: please
try to disable ATS and MSI without invasive changes, ideally just a
couple of lines to force-disable each feature.


> 2. Make the code with XEN compatible ( remove linux compatibility functions)
> 3. Keep the Linux coding style for code imported from Linux.
> 4. Fix all comments.

Sounds good.


> I have one query what will be coding style for new code to make driver work  for XEN ? 

We try to keep the same code style for the entirety of a source file. In
this case, the whole driver should use Linux code style (both imported
code and new code).


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 3/8] xen/arm: revert patch related to XArray
  2020-12-03 12:57     ` Rahul Singh
@ 2020-12-04  8:52       ` Julien Grall
  0 siblings, 0 replies; 53+ messages in thread
From: Julien Grall @ 2020-12-04  8:52 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, Bertrand Marquis, Stefano Stabellini, Volodymyr Babchuk



On 03/12/2020 12:57, Rahul Singh wrote:
> Hello Julien,

Hi,

>> On 2 Dec 2020, at 1:46 pm, Julien Grall <julien@xen.org> wrote:
>>
>> Hi Rahul,
>>
>> On 26/11/2020 17:02, Rahul Singh wrote:
>>> XArray is not implemented in XEN revert the patch that introduce the
>>> XArray code in SMMUv3 driver.
>>
>> Similar to the atomic revert, you are explaining why the revert but not the consequences. I think this is quite important to have them outline in the commit message as it looks like it means the SMMU driver would not scale.
> 
> Ok I will add.
>>
>>> Once XArray is implemented in XEN this patch can be added in XEN.
>>
>> What's the plan for that?
> 
> As of now there is no plan for Xarray from our side.
> Do we have requirement for Xarray implementation in XEN ?

That's going to depend on the consequence of reverting this patch.

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN
  2020-12-03 14:33     ` Rahul Singh
@ 2020-12-04  9:05       ` Julien Grall
  2020-12-07 10:36         ` Rahul Singh
  0 siblings, 1 reply; 53+ messages in thread
From: Julien Grall @ 2020-12-04  9:05 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, Bertrand Marquis, Stefano Stabellini, Volodymyr Babchuk

Hi Rahul,

On 03/12/2020 14:33, Rahul Singh wrote:
>> On 2 Dec 2020, at 2:45 pm, Julien Grall <julien@xen.org> wrote:
>>> -
>>> -static struct iommu_device *arm_smmu_probe_device(struct device *dev)
>>> -{
>>
>> Most of the code here looks useful to Xen. I think you want to keep the code and re-use it afterwards.
> 
> Ok. I removed the code here and added the XEN compatible code to add devices in next patch.
> I will keep it in this patch and will modifying the code to make XEN compatible.

In general, it is prefer if the code the code rather than dropping in 
patch A and then add it again differently patch B. This makes easier to 
check that the code outcome of the function is mostly the same.

>>> -static struct iommu_ops arm_smmu_ops = {
>>> -	.capable		= arm_smmu_capable,
>>> -	.domain_alloc		= arm_smmu_domain_alloc,
>>> -	.domain_free		= arm_smmu_domain_free,
>>> -	.attach_dev		= arm_smmu_attach_dev,
>>> -	.map			= arm_smmu_map,
>>> -	.unmap			= arm_smmu_unmap,
>>> -	.flush_iotlb_all	= arm_smmu_flush_iotlb_all,
>>> -	.iotlb_sync		= arm_smmu_iotlb_sync,
>>> -	.iova_to_phys		= arm_smmu_iova_to_phys,
>>> -	.probe_device		= arm_smmu_probe_device,
>>> -	.release_device		= arm_smmu_release_device,
>>> -	.device_group		= arm_smmu_device_group,
>>> -	.domain_get_attr	= arm_smmu_domain_get_attr,
>>> -	.domain_set_attr	= arm_smmu_domain_set_attr,
>>> -	.of_xlate		= arm_smmu_of_xlate,
>>> -	.get_resv_regions	= arm_smmu_get_resv_regions,
>>> -	.put_resv_regions	= generic_iommu_put_resv_regions,
>>> -	.pgsize_bitmap		= -1UL, /* Restricted during device attach */
>>> -};
>>> -
>>>   /* Probing and initialisation functions */
>>>   static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>>>   				   struct arm_smmu_queue *q,
>>> @@ -2406,7 +2032,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>>>   	switch (FIELD_GET(IDR0_STALL_MODEL, reg)) {
>>>   	case IDR0_STALL_MODEL_FORCE:
>>>   		smmu->features |= ARM_SMMU_FEAT_STALL_FORCE;
>>> -		fallthrough;
>>
>> We should keep all the fallthrough documented. So I think we want to introduce the fallthrough in Xen as well.
> 
> Ok I will keep fallthrough documented in this patch.
> 
> fallthrough implementation in XEN should be another patch. I am not sure when we can implement but we will try to implement.

Yes, I didn't ask to implement "fallthrough" in this patch, but instead 
as a pre-requirement patch.

I would implement it in include/xen/compiler.h.

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-03 18:47           ` Stefano Stabellini
@ 2020-12-07  8:33             ` Rahul Singh
  0 siblings, 0 replies; 53+ messages in thread
From: Rahul Singh @ 2020-12-07  8:33 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: Julien Grall, xen-devel, Bertrand Marquis, Andrew Cooper,
	George Dunlap, Ian Jackson, Jan Beulich, Wei Liu, Paul Durrant,
	Volodymyr Babchuk

Hello Stefano,

> On 3 Dec 2020, at 6:47 pm, Stefano Stabellini <sstabellini@kernel.org> wrote:
> 
> On Thu, 3 Dec 2020, Rahul Singh wrote:
>>> On 3 Dec 2020, at 4:13 am, Stefano Stabellini <sstabellini@kernel.org> wrote:
>>> On Wed, 2 Dec 2020, Julien Grall wrote:
>>>> On 02/12/2020 02:51, Stefano Stabellini wrote:
>>>>> On Thu, 26 Nov 2020, Rahul Singh wrote:
>>>>>> +/* Alias to Xen device tree helpers */
>>>>>> +#define device_node dt_device_node
>>>>>> +#define of_phandle_args dt_phandle_args
>>>>>> +#define of_device_id dt_device_match
>>>>>> +#define of_match_node dt_match_node
>>>>>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np,
>>>>>> pname, out))
>>>>>> +#define of_property_read_bool dt_property_read_bool
>>>>>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
>>>>> 
>>>>> Given all the changes to the file by the previous patches we are
>>>>> basically fully (or almost fully) adapting this code to Xen.
>>>>> 
>>>>> So at that point I wonder if we should just as well make these changes
>>>>> (e.g. s/of_phandle_args/dt_phandle_args/g) to the code too.
>>>> 
>>>> I have already accepted the fact that keeping Linux code as-is is nearly
>>>> impossible without much workaround :). The benefits tends to also limited as
>>>> we noticed for the SMMU driver.
>>>> 
>>>> I would like to point out that this may make quite difficult (if not
>>>> impossible) to revert the previous patches which remove support for some
>>>> features (e.g. atomic, MSI, ATS).
>>>> 
>>>> If we are going to adapt the code to Xen (I'd like to keep Linux code style
>>>> though), then I think we should consider to keep code that may be useful in
>>>> the near future (at least MSI, ATS).
>>> 
>>> (I am fine with keeping the Linux code style.)
>>> 
>>> We could try to keep the code as similar to Linux as possible. This
>>> didn't work out in the past.
>>> 
>>> Otherwise, we could fully adapt the driver to Xen. If we fully adapt the
>>> driver to Xen (code style aside) it is better to be consistent and also
>>> do substitutions like s/of_phandle_args/dt_phandle_args/g. Then the
>>> policy becomes clear: the code comes from Linux but it is 100% adapted
>>> to Xen.
>>> 
>>> 
>>> Now the question about what to do about the MSI and ATS code is a good
>>> one. We know that we are going to want that code at some point in the
>>> next 2 years. Like you wrote, if we fully adapt the code to Xen and
>>> remove MSI and ATS code, then it is going to be harder to add it back.
>>> 
>>> So maybe keeping the MSI and ATS code for now, even if it cannot work,
>>> would be better. I think this strategy works well if the MSI and ATS
>>> code can be disabled easily, i.e. with a couple of lines of code in the
>>> init function rather than #ifdef everywhere. It doesn't work well if we
>>> have to add #ifdef everywhere.
>>> 
>>> It looks like MSI could be disabled adding a couple of lines to
>>> arm_smmu_setup_msis.
>>> 
>>> Similarly ATS seems to be easy to disable by forcing ats_enabled to
>>> false.
>>> 
>>> So yes, this looks like a good way forward. Rahul, what do you think?
>> 
>> 
>> I am ok to have the PCI ATS and MSI functionality in the code. 
>> As per the discussion next version of the patch will include below modification:Please let me know if there are any suggestion overall that should be added in next version.
>> 
>> 1. Keep the PCI ATS and MSI functionality code.
> 
> I'll repeat one point I wrote above that I think it is important: please
> try to disable ATS and MSI without invasive changes, ideally just a
> couple of lines to force-disable each feature

Yes I will disable the feature.
> 
> 
>> 2. Make the code with XEN compatible ( remove linux compatibility functions)
>> 3. Keep the Linux coding style for code imported from Linux.
>> 4. Fix all comments.
> 
> Sounds good.
> 
> 
>> I have one query what will be coding style for new code to make driver work  for XEN ? 
> 
> We try to keep the same code style for the entirety of a source file. In
> this case, the whole driver should use Linux code style (both imported
> code and new code).

Ok.

Regards,
Rahul	

^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN
  2020-12-04  9:05       ` Julien Grall
@ 2020-12-07 10:36         ` Rahul Singh
  2020-12-07 10:55           ` Julien Grall
  0 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-12-07 10:36 UTC (permalink / raw)
  To: Julien Grall
  Cc: xen-devel, Bertrand Marquis, Stefano Stabellini, Volodymyr Babchuk

Hello Julien,	

> On 4 Dec 2020, at 9:05 am, Julien Grall <julien@xen.org> wrote:
> 
> Hi Rahul,
> 
> On 03/12/2020 14:33, Rahul Singh wrote:
>>> On 2 Dec 2020, at 2:45 pm, Julien Grall <julien@xen.org> wrote:
>>>> -
>>>> -static struct iommu_device *arm_smmu_probe_device(struct device *dev)
>>>> -{
>>> 
>>> Most of the code here looks useful to Xen. I think you want to keep the code and re-use it afterwards.
>> Ok. I removed the code here and added the XEN compatible code to add devices in next patch.
>> I will keep it in this patch and will modifying the code to make XEN compatible.
> 
> In general, it is prefer if the code the code rather than dropping in patch A and then add it again differently patch B. This makes easier to check that the code outcome of the function is mostly the same.
> 

Ok.

>>>> -static struct iommu_ops arm_smmu_ops = {
>>>> -	.capable		= arm_smmu_capable,
>>>> -	.domain_alloc		= arm_smmu_domain_alloc,
>>>> -	.domain_free		= arm_smmu_domain_free,
>>>> -	.attach_dev		= arm_smmu_attach_dev,
>>>> -	.map			= arm_smmu_map,
>>>> -	.unmap			= arm_smmu_unmap,
>>>> -	.flush_iotlb_all	= arm_smmu_flush_iotlb_all,
>>>> -	.iotlb_sync		= arm_smmu_iotlb_sync,
>>>> -	.iova_to_phys		= arm_smmu_iova_to_phys,
>>>> -	.probe_device		= arm_smmu_probe_device,
>>>> -	.release_device		= arm_smmu_release_device,
>>>> -	.device_group		= arm_smmu_device_group,
>>>> -	.domain_get_attr	= arm_smmu_domain_get_attr,
>>>> -	.domain_set_attr	= arm_smmu_domain_set_attr,
>>>> -	.of_xlate		= arm_smmu_of_xlate,
>>>> -	.get_resv_regions	= arm_smmu_get_resv_regions,
>>>> -	.put_resv_regions	= generic_iommu_put_resv_regions,
>>>> -	.pgsize_bitmap		= -1UL, /* Restricted during device attach */
>>>> -};
>>>> -
>>>>  /* Probing and initialisation functions */
>>>>  static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>>>>  				   struct arm_smmu_queue *q,
>>>> @@ -2406,7 +2032,6 @@ static int arm_smmu_device_hw_probe(struct arm_smmu_device *smmu)
>>>>  	switch (FIELD_GET(IDR0_STALL_MODEL, reg)) {
>>>>  	case IDR0_STALL_MODEL_FORCE:
>>>>  		smmu->features |= ARM_SMMU_FEAT_STALL_FORCE;
>>>> -		fallthrough;
>>> 
>>> We should keep all the fallthrough documented. So I think we want to introduce the fallthrough in Xen as well.
>> Ok I will keep fallthrough documented in this patch.
>> fallthrough implementation in XEN should be another patch. I am not sure when we can implement but we will try to implement.
> 
> Yes, I didn't ask to implement "fallthrough" in this patch, but instead as a pre-requirement patch.
> 
> I would implement it in include/xen/compiler.h.

Ok. I will implement and will share the patch for “__attribute__((__fallthrough__)) ” but for SMMUv3 is that ok if we can proceed with “ /* fallthrough */  ". 

As “fallthorugh” implementation is common for all architecture it required review for all stakeholder as per my understanding. I don’t want to block SMMUv3 because of this.

Regards,
Rahul
 
> 
> Cheers,
> 
> -- 
> Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN
  2020-12-07 10:36         ` Rahul Singh
@ 2020-12-07 10:55           ` Julien Grall
  0 siblings, 0 replies; 53+ messages in thread
From: Julien Grall @ 2020-12-07 10:55 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, Bertrand Marquis, Stefano Stabellini, Volodymyr Babchuk



On 07/12/2020 10:36, Rahul Singh wrote:
>> I would implement it in include/xen/compiler.h.
> 
> Ok. I will implement and will share the patch for “__attribute__((__fallthrough__)) ” but for SMMUv3 is that ok if we can proceed with “ /* fallthrough */  ".
The first approach should always be to work with the community to 
introduce such functionality tree-wide. Note that I am not suggesting to 
replace existing "/* fallthrough */" with "fallthrough".

If it is taking too long or the task is significant, then we can discuss 
about way to temporally work around it.

I don't believe we are at that stage yet here.

> As “fallthorugh” implementation is common for all architecture it required review for all stakeholder as per my understanding. I don’t want to block SMMUv3 because of this.

It only requires review from "THE REST" maintainers.

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-02 16:22   ` Julien Grall
@ 2020-12-07 12:12     ` Rahul Singh
  2020-12-07 17:39       ` Julien Grall
  0 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-12-07 12:12 UTC (permalink / raw)
  To: Julien Grall
  Cc: xen-devel, Bertrand Marquis, Andrew Cooper, George Dunlap,
	Ian Jackson, Jan Beulich, Stefano Stabellini, Wei Liu,
	Paul Durrant, Volodymyr Babchuk

Hello Julien,

> On 2 Dec 2020, at 4:22 pm, Julien Grall <julien@xen.org> wrote:
> 
> Hi Rahul,
> 
> On 26/11/2020 17:02, Rahul Singh wrote:
>> Add support for ARM architected SMMUv3 implementation. It is based on
>> the Linux SMMUv3 driver.
>> Major differences with regard to Linux driver are as follows:
>> 1. Only Stage-2 translation is supported as compared to the Linux driver
>>    that supports both Stage-1 and Stage-2 translations.
>> 2. Use P2M  page table instead of creating one as SMMUv3 has the
>>    capability to share the page tables with the CPU.
>> 3. Tasklets are used in place of threaded IRQ's in Linux for event queue
>>    and priority queue IRQ handling.
> 
> On the previous version, we discussed that using tasklets is not a suitable replacement for threaded IRQs. What's the plan to address it?

As suggested by you in earlier discussions I will try to use the timer to schedule the work and will share the findings. 
> 
>> 4. Latest version of the Linux SMMUv3 code implements the commands queue
>>    access functions based on atomic operations implemented in Linux.
>>    Atomic functions used by the commands queue access functions are not
>>    implemented in XEN therefore we decided to port the earlier version
>>    of the code. Once the proper atomic operations will be available in
>>    XEN the driver can be updated.
>> 5. Driver is currently supported as Tech Preview.
>> Signed-off-by: Rahul Singh <rahul.singh@arm.com>
>> ---
>>  MAINTAINERS                           |   6 +
>>  SUPPORT.md                            |   1 +
>>  xen/drivers/passthrough/Kconfig       |  10 +
>>  xen/drivers/passthrough/arm/Makefile  |   1 +
>>  xen/drivers/passthrough/arm/smmu-v3.c | 986 +++++++++++++++++++++-----
>>  5 files changed, 814 insertions(+), 190 deletions(-)
>> diff --git a/MAINTAINERS b/MAINTAINERS
>> index dab38a6a14..1d63489eec 100644
>> --- a/MAINTAINERS
>> +++ b/MAINTAINERS
>> @@ -249,6 +249,12 @@ F:	xen/include/asm-arm/
>>  F:	xen/include/public/arch-arm/
>>  F:	xen/include/public/arch-arm.h
>>  +ARM SMMUv3
>> +M:	Bertrand Marquis <bertrand.marquis@arm.com>
>> +M:	Rahul Singh <rahul.singh@arm.com>
>> +S:	Supported
>> +F:	xen/drivers/passthrough/arm/smmu-v3.c
>> +
>>  Change Log
>>  M:	Paul Durrant <paul@xen.org>
>>  R:	Community Manager <community.manager@xenproject.org>
>> diff --git a/SUPPORT.md b/SUPPORT.md
>> index ab02aca5f4..e402c7202d 100644
>> --- a/SUPPORT.md
>> +++ b/SUPPORT.md
>> @@ -68,6 +68,7 @@ For the Cortex A57 r0p0 - r1p1, see Errata 832075.
>>      Status, ARM SMMUv1: Supported, not security supported
>>      Status, ARM SMMUv2: Supported, not security supported
>>      Status, Renesas IPMMU-VMSA: Supported, not security supported
>> +    Status, ARM SMMUv3: Tech Preview
> 
> Please move this right after "ARM SMMUv2”.

Ok.

> 
>>    ### ARM/GICv3 ITS
>>  diff --git a/xen/drivers/passthrough/Kconfig b/xen/drivers/passthrough/Kconfig
>> index 0036007ec4..5b71c59f47 100644
>> --- a/xen/drivers/passthrough/Kconfig
>> +++ b/xen/drivers/passthrough/Kconfig
>> @@ -13,6 +13,16 @@ config ARM_SMMU
>>  	  Say Y here if your SoC includes an IOMMU device implementing the
>>  	  ARM SMMU architecture.
>>  +config ARM_SMMU_V3
>> +	bool "ARM Ltd. System MMU Version 3 (SMMUv3) Support" if EXPERT
>> +	depends on ARM_64
>> +	---help---
>> +	 Support for implementations of the ARM System MMU architecture
>> +	 version 3.
>> +
>> +	 Say Y here if your system includes an IOMMU device implementing
>> +	 the ARM SMMUv3 architecture.
>> +
>>  config IPMMU_VMSA
>>  	bool "Renesas IPMMU-VMSA found in R-Car Gen3 SoCs"
>>  	depends on ARM_64
>> diff --git a/xen/drivers/passthrough/arm/Makefile b/xen/drivers/passthrough/arm/Makefile
>> index fcd918ea3e..c5fb3b58a5 100644
>> --- a/xen/drivers/passthrough/arm/Makefile
>> +++ b/xen/drivers/passthrough/arm/Makefile
>> @@ -1,3 +1,4 @@
>>  obj-y += iommu.o iommu_helpers.o iommu_fwspec.o
>>  obj-$(CONFIG_ARM_SMMU) += smmu.o
>>  obj-$(CONFIG_IPMMU_VMSA) += ipmmu-vmsa.o
>> +obj-$(CONFIG_ARM_SMMU_V3) += smmu-v3.o
>> diff --git a/xen/drivers/passthrough/arm/smmu-v3.c b/xen/drivers/passthrough/arm/smmu-v3.c
>> index 55d1cba194..8f2337e7f2 100644
>> --- a/xen/drivers/passthrough/arm/smmu-v3.c
>> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
>> @@ -2,36 +2,280 @@
>>  /*
>>   * IOMMU API for ARM architected SMMUv3 implementations.
>>   *
>> - * Copyright (C) 2015 ARM Limited
>> + * Based on Linux's SMMUv3 driver:
>> + *    drivers/iommu/arm/arm-smmu-v3/arm-smmu-v3.c
>> + *    commit: 951cbbc386ff01b50da4f46387e994e81d9ab431
>> + * and Xen's SMMU driver:
>> + *    xen/drivers/passthrough/arm/smmu.c
> 
> I would suggest to list the major differences here as well.

Ok. 
> 
>>   *
>> - * Author: Will Deacon <will.deacon@arm.com>
>> + * Copyright (C) 2015 ARM Limited Will Deacon <will.deacon@arm.com>
> 
> Why did you merge the Author and copyright line?

I  Will fix this in next version.

>>   *
>> - * This driver is powered by bad coffee and bombay mix.
>> + * Copyright (C) 2020 Arm Ltd.
>> + *
>> + * 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, see <http://www.gnu.org/licenses/>.
>> + *
>> + */
>> +
>> +#include <xen/acpi.h>
>> +#include <xen/config.h>
>> +#include <xen/delay.h>
>> +#include <xen/errno.h>
>> +#include <xen/err.h>
>> +#include <xen/irq.h>
>> +#include <xen/lib.h>
>> +#include <xen/list.h>
>> +#include <xen/mm.h>
>> +#include <xen/rbtree.h>
>> +#include <xen/sched.h>
>> +#include <xen/sizes.h>
>> +#include <xen/vmap.h>
>> +#include <asm/atomic.h>
>> +#include <asm/device.h>
>> +#include <asm/io.h>
>> +#include <asm/platform.h>
>> +#include <asm/iommu_fwspec.h>
> 
> All the headers seem to be alphabetically ordered but this one.

Ack.
> 
>> +
>> +/* Linux compatibility functions. */
> 
> Some of the helpers here seem to be similar to the SMMU driver. Can we have an header that can be shared between the two?

As we agreed that I will make the driver XEN compatible and I will remove the compatibility functions from here. If there is any helper that is common I will create the common header file.
> 
>> +typedef paddr_t dma_addr_t;
>> +typedef unsigned int gfp_t;
>> +
>> +#define platform_device device
>> +
>> +#define GFP_KERNEL 0
>> +
>> +/* Alias to Xen device tree helpers */
>> +#define device_node dt_device_node
>> +#define of_phandle_args dt_phandle_args
>> +#define of_device_id dt_device_match
>> +#define of_match_node dt_match_node
>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, pname, out))
>> +#define of_property_read_bool dt_property_read_bool
>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
>> +
>> +/* Alias to Xen lock functions */
>> +#define mutex spinlock
>> +#define mutex_init spin_lock_init
>> +#define mutex_lock spin_lock
>> +#define mutex_unlock spin_unlock
> 
> Hmm... mutex are not spinlock. Can you explain why this is fine to switch to spinlock?

Yes mutex are not spinlock. As mutex is not implemented in XEN I thought of using spinlock in place of mutex as this is the only locking mechanism available in XEN. 
Let me know if there is another blocking lock available in XEN. I will check if we can use that.

> 
>> +
>> +/* Alias to Xen time functions */
>> +#define ktime_t s_time_t
>> +#define ktime_get()             (NOW())
>> +#define ktime_add_us(t,i)       (t + MICROSECS(i))
>> +#define ktime_compare(t,i)      (t > (i))
>> +
>> +/* Alias to Xen allocation helpers */
>> +#define kzalloc(size, flags)    _xzalloc(size, sizeof(void *))
>> +#define kfree xfree
>> +#define devm_kzalloc(dev, size, flags)  _xzalloc(size, sizeof(void *))
>> +
>> +/* Device logger functions */
>> +#define dev_name(dev) dt_node_full_name(dev->of_node)
>> +#define dev_dbg(dev, fmt, ...)      \
>> +    printk(XENLOG_DEBUG "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +#define dev_notice(dev, fmt, ...)   \
>> +    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +#define dev_warn(dev, fmt, ...)     \
>> +    printk(XENLOG_WARNING "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +#define dev_err(dev, fmt, ...)      \
>> +    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +#define dev_info(dev, fmt, ...)     \
>> +    printk(XENLOG_INFO "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +#define dev_err_ratelimited(dev, fmt, ...)      \
>> +    printk(XENLOG_ERR "SMMUv3: %s: " fmt, dev_name(dev), ## __VA_ARGS__)
>> +
>> +/*
>> + * Periodically poll an address and wait between reads in us until a
>> + * condition is met or a timeout occurs.
>> + */
>> +#define readx_poll_timeout(op, addr, val, cond, sleep_us, timeout_us) \
>> +({ \
>> +     s_time_t deadline = NOW() + MICROSECS(timeout_us); \
>> +     for (;;) { \
>> +        (val) = op(addr); \
>> +        if (cond) \
>> +            break; \
>> +        if (NOW() > deadline) { \
>> +            (val) = op(addr); \
>> +            break; \
>> +        } \
>> +        udelay(sleep_us); \
>> +     } \
>> +     (cond) ? 0 : -ETIMEDOUT; \
>> +})
>> +
>> +#define readl_relaxed_poll_timeout(addr, val, cond, delay_us, timeout_us) \
>> +    readx_poll_timeout(readl_relaxed, addr, val, cond, delay_us, timeout_us)
>> +
>> +#define FIELD_PREP(_mask, _val)         \
>> +    (((typeof(_mask))(_val) << (__builtin_ffsll(_mask) - 1)) & (_mask))
>> +
>> +#define FIELD_GET(_mask, _reg)          \
>> +    (typeof(_mask))(((_reg) & (_mask)) >> (__builtin_ffsll(_mask) - 1))
>> +
>> +#define WRITE_ONCE(x, val)                  \
>> +do {                                        \
>> +    *(volatile typeof(x) *)&(x) = (val);    \
>> +} while (0)
> 
> Please implement it with write_atomic() or ACCESS_ONCE().

OK. 
> 
>> +
>> +/* Xen: Stub out DMA domain related functions */
>> +#define iommu_get_dma_cookie(dom) 0
>> +#define iommu_put_dma_cookie(dom)
>> +
>> +/*
>> + * Helpers for DMA allocation. Just the function name is reused for
>> + * porting code, these allocation are not managed allocations
>>   */
>> +static void *dmam_alloc_coherent(struct device *dev, size_t size,
>> +                                 paddr_t *dma_handle, gfp_t gfp)
>> +{
>> +    void *vaddr;
>> +    unsigned long alignment = size;
>> +
>> +    /*
>> +     * _xzalloc requires that the (align & (align -1)) = 0. Most of the
>> +     * allocations in SMMU code should send the right value for size. In
>> +     * case this is not true print a warning and align to the size of a
>> +     * (void *)
>> +     */
>> +    if ( size & (size - 1) )
> 
> We should use the same coding style within the file. As the file is imported from Linux, new code should follow Linux coding style.

Ok.
> 
>> +    {
>> +        printk(XENLOG_WARNING "SMMUv3: Fixing alignment for the DMA buffer\n");
>> +        alignment = sizeof(void *);
>> +    }
>> +
>> +    vaddr = _xzalloc(size, alignment);
>> +    if ( !vaddr )
>> +    {
>> +        printk(XENLOG_ERR "SMMUv3: DMA allocation failed\n");
>> +        return NULL;
>> +    }
>> +
>> +    *dma_handle = virt_to_maddr(vaddr);
>> +
>> +    return vaddr;
>> +}
>> +
>> +/* Xen: Type definitions for iommu_domain */
>> +#define IOMMU_DOMAIN_UNMANAGED 0
>> +#define IOMMU_DOMAIN_DMA 1
>> +#define IOMMU_DOMAIN_IDENTITY 2
>> +
>> +/* Xen specific code. */
>> +struct iommu_domain {
>> +    /* Runtime SMMU configuration for this iommu_domain */
>> +    atomic_t ref;
>> +    /*
>> +     * Used to link iommu_domain contexts for a same domain.
>> +     * There is at least one per-SMMU to used by the domain.
>> +     */
>> +    struct list_head    list;
>> +};
>> +
>> +/* Describes information required for a Xen domain */
>> +struct arm_smmu_xen_domain {
>> +    spinlock_t      lock;
>> +
>> +    /* List of iommu domains associated to this domain */
>> +    struct list_head    contexts;
>> +};
>> +
>> +/*
>> + * Information about each device stored in dev->archdata.iommu
>> + * The dev->archdata.iommu stores the iommu_domain (runtime configuration of
>> + * the SMMU).
>> + */
>> +struct arm_smmu_xen_device {
>> +    struct iommu_domain *domain;
>> +};
>> +
>> +/* Keep a list of devices associated with this driver */
>> +static DEFINE_SPINLOCK(arm_smmu_devices_lock);
>> +static LIST_HEAD(arm_smmu_devices);
>> +
>> +
>> +static inline void *dev_iommu_priv_get(struct device *dev)
>> +{
>> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> +
>> +    return fwspec && fwspec->iommu_priv ? fwspec->iommu_priv : NULL;
>> +}
>> +
>> +static inline void dev_iommu_priv_set(struct device *dev, void *priv)
>> +{
>> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> +
>> +    fwspec->iommu_priv = priv;
>> +}
>> +
>> +int dt_property_match_string(const struct dt_device_node *np,
>> +                             const char *propname, const char *string)
> 
> I think this should be implemented in device_tree.c

Ok I will move this to device_tree.c file
> 
>> +{
>> +    const struct dt_property *dtprop = dt_find_property(np, propname, NULL);
>> +    size_t l;
>> +    int i;
>> +    const char *p, *end;
>> +
>> +    if ( !dtprop )
>> +        return -EINVAL;
>> +
>> +    if ( !dtprop->value )
>> +        return -ENODATA;
>> +
>> +    p = dtprop->value;
>> +    end = p + dtprop->length;
>> +
>> +    for ( i = 0; p < end; i++, p += l )
>> +    {
>> +        l = strnlen(p, end - p) + 1;
>> +
>> +        if ( p + l > end )
>> +            return -EILSEQ;
>> +
>> +        if ( strcmp(string, p) == 0 )
>> +            return i; /* Found it; return index */
>> +    }
>> +
>> +    return -ENODATA;
>> +}
>> +
>> +static int platform_get_irq_byname_optional(struct device *dev,
>> +                                            const char *name)
>> +{
>> +    int index, ret;
>> +    struct dt_device_node *np  = dev_to_dt(dev);
>> +
>> +    if ( unlikely(!name) )
>> +        return -EINVAL;
>> +
>> +    index = dt_property_match_string(np, "interrupt-names", name);
>> +    if ( index < 0 )
>> +    {
>> +        dev_info(dev, "IRQ %s not found\n", name);
>> +        return index;
>> +    }
>>  -#include <linux/acpi.h>
>> -#include <linux/acpi_iort.h>
>> -#include <linux/bitfield.h>
>> -#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>
>> -#include <linux/iommu.h>
>> -#include <linux/iopoll.h>
>> -#include <linux/module.h>
>> -#include <linux/msi.h>
>> -#include <linux/of.h>
>> -#include <linux/of_address.h>
>> -#include <linux/of_iommu.h>
>> -#include <linux/of_platform.h>
>> -#include <linux/pci.h>
>> -#include <linux/pci-ats.h>
>> -#include <linux/platform_device.h>
>> -
>> -#include <linux/amba/bus.h>
>> +    ret = platform_get_irq(np, index);
>> +    if ( ret < 0 )
>> +    {
>> +        dev_err(dev, "failed to get irq index %d\n", index);
>> +        return -ENODEV;
>> +    }
>> +
>> +    return ret;
>> +}
>> +
>> +/* Start of Linux SMMUv3 code */
>>    /* MMIO registers */
>>  #define ARM_SMMU_IDR0			0x0
>> @@ -507,6 +751,7 @@ struct arm_smmu_s2_cfg {
>>  	u16				vmid;
>>  	u64				vttbr;
>>  	u64				vtcr;
>> +	struct domain		*domain;
>>  };
>>    struct arm_smmu_strtab_cfg {
>> @@ -567,8 +812,13 @@ struct arm_smmu_device {
>>    	struct arm_smmu_strtab_cfg	strtab_cfg;
>>  -	/* IOMMU core code handle */
>> -	struct iommu_device		iommu;
>> +	/* Need to keep a list of SMMU devices */
>> +	struct list_head		devices;
>> +
>> +	/* Tasklets for handling evts/faults and pci page request IRQs*/
>> +	struct tasklet		evtq_irq_tasklet;
>> +	struct tasklet		priq_irq_tasklet;
>> +	struct tasklet		combined_irq_tasklet;
>>  };
>>    /* SMMU private data for each master */
>> @@ -1110,7 +1360,7 @@ static int arm_smmu_init_l2_strtab(struct arm_smmu_device *smmu, u32 sid)
>>  }
>>    /* IRQ and event handlers */
>> -static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
>> +static void arm_smmu_evtq_thread(void *dev)
>>  {
>>  	int i;
>>  	struct arm_smmu_device *smmu = dev;
>> @@ -1140,7 +1390,6 @@ static irqreturn_t arm_smmu_evtq_thread(int irq, void *dev)
>>  	/* Sync our overflow flag, as we believe we're up to speed */
>>  	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
>>  		    Q_IDX(llq, llq->cons);
>> -	return IRQ_HANDLED;
>>  }
>>    static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
>> @@ -1181,7 +1430,7 @@ static void arm_smmu_handle_ppr(struct arm_smmu_device *smmu, u64 *evt)
>>  	}
>>  }
>>  -static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
>> +static void arm_smmu_priq_thread(void *dev)
>>  {
>>  	struct arm_smmu_device *smmu = dev;
>>  	struct arm_smmu_queue *q = &smmu->priq.q;
>> @@ -1200,12 +1449,12 @@ static irqreturn_t arm_smmu_priq_thread(int irq, void *dev)
>>  	llq->cons = Q_OVF(llq->prod) | Q_WRP(llq, llq->cons) |
>>  		      Q_IDX(llq, llq->cons);
>>  	queue_sync_cons_out(q);
>> -	return IRQ_HANDLED;
>>  }
>>    static int arm_smmu_device_disable(struct arm_smmu_device *smmu);
>>  -static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>> +static void arm_smmu_gerror_handler(int irq, void *dev,
>> +				struct cpu_user_regs *regs)
>>  {
>>  	u32 gerror, gerrorn, active;
>>  	struct arm_smmu_device *smmu = dev;
>> @@ -1215,7 +1464,7 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>>    	active = gerror ^ gerrorn;
>>  	if (!(active & GERROR_ERR_MASK))
>> -		return IRQ_NONE; /* No errors pending */
>> +		return; /* No errors pending */
>>    	dev_warn(smmu->dev,
>>  		 "unexpected global error reported (0x%08x), this could be serious\n",
>> @@ -1248,26 +1497,42 @@ static irqreturn_t arm_smmu_gerror_handler(int irq, void *dev)
>>  		arm_smmu_cmdq_skip_err(smmu);
>>    	writel(gerror, smmu->base + ARM_SMMU_GERRORN);
>> -	return IRQ_HANDLED;
>>  }
>>  -static irqreturn_t arm_smmu_combined_irq_thread(int irq, void *dev)
>> +static void arm_smmu_combined_irq_handler(int irq, void *dev,
>> +				struct cpu_user_regs *regs)
>> +{
>> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
> 
> The cast is not necessary.

Ok.
> 
>> +
>> +	arm_smmu_gerror_handler(irq, dev, regs);
>> +
>> +	tasklet_schedule(&(smmu->combined_irq_tasklet));
>> +}
>> +
>> +static void arm_smmu_combined_irq_thread(void *dev)
>>  {
>>  	struct arm_smmu_device *smmu = dev;
>>  -	arm_smmu_evtq_thread(irq, dev);
>> +	arm_smmu_evtq_thread(dev);
>>  	if (smmu->features & ARM_SMMU_FEAT_PRI)
>> -		arm_smmu_priq_thread(irq, dev);
>> -
>> -	return IRQ_HANDLED;
>> +		arm_smmu_priq_thread(dev);
>>  }
>>  -static irqreturn_t arm_smmu_combined_irq_handler(int irq, void *dev)
>> +static void arm_smmu_evtq_irq_tasklet(int irq, void *dev,
>> +				struct cpu_user_regs *regs)
>>  {
>> -	arm_smmu_gerror_handler(irq, dev);
>> -	return IRQ_WAKE_THREAD;
>> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
> 
> Ditto.
> 
>> +
>> +	tasklet_schedule(&(smmu->evtq_irq_tasklet));
>>  }
>>  +static void arm_smmu_priq_irq_tasklet(int irq, void *dev,
>> +				struct cpu_user_regs *regs)
>> +{
>> +	struct arm_smmu_device *smmu = (struct arm_smmu_device *)dev;
> 
> Ditto.
> 
>> +
>> +	tasklet_schedule(&(smmu->priq_irq_tasklet));
>> +}
>>    /* IO_PGTABLE API */
>>  static void arm_smmu_tlb_inv_context(void *cookie)
>> @@ -1354,27 +1619,69 @@ static void arm_smmu_domain_free(struct iommu_domain *domain)
>>  }
>>    static int arm_smmu_domain_finalise_s2(struct arm_smmu_domain *smmu_domain,
>> -				       struct arm_smmu_master *master,
>> -				       struct io_pgtable_cfg *pgtbl_cfg)
>> +				       struct arm_smmu_master *master)
>>  {
>>  	int vmid;
>> +	u64 reg;
> 
> I think uint32_t is sufficient here.

Ok. 
> 
>>  	struct arm_smmu_device *smmu = smmu_domain->smmu;
>>  	struct arm_smmu_s2_cfg *cfg = &smmu_domain->s2_cfg;
>>  +	/* VTCR */
>> +	reg = VTCR_RES1 | VTCR_SH0_IS | VTCR_IRGN0_WBWA | VTCR_ORGN0_WBWA;
> 
> VTCR_RES1 will set bit 31 to 1. However, from the spec it looks like the equivalent bit in the entry (see 5.2 in ARM IHI 0070C.a) will be RES0.

Yes I will fix this.

> 
>> +
>> +	switch (PAGE_SIZE) {
>> +	case SZ_4K:
>> +		reg |= VTCR_TG0_4K;
>> +		break;
>> +	case SZ_16K:
>> +		reg |= VTCR_TG0_16K;
>> +		break;
>> +	case SZ_64K:
>> +		reg |= VTCR_TG0_4K;
>> +		break;
>> +	}
> 
> I would just handle 4K here and add a BUILD_BUG_ON(PAGE_SIZE != SZ_4K).

Ok.
> 
>> + > +	switch (smmu->oas) {
> 
> AFAICT smmu->oas and ...
> 
>> +	case 32:
>> +		reg |= VTCR_PS(_AC(0x0,ULL));
>> +		break;
>> +	case 36:
>> +		reg |= VTCR_PS(_AC(0x1,ULL));
>> +		break;
>> +	case 40:
>> +		reg |= VTCR_PS(_AC(0x2,ULL));
>> +		break;
>> +	case 42:
>> +		reg |= VTCR_PS(_AC(0x3,ULL));
>> +		break;
>> +	case 44:
>> +		reg |= VTCR_PS(_AC(0x4,ULL));
>> +		break;
>> +		case 48:
>> +		reg |= VTCR_PS(_AC(0x5,ULL));
>> +		break;
>> +	case 52:
>> +		reg |= VTCR_PS(_AC(0x6,ULL));
>> +		break;
>> +	}
>> +
>> +	reg |= VTCR_T0SZ(64ULL - smmu->ias);
> 
> ... are directly taken from the SMMU configuration. However, as we share the P2M, we need to make sure the value match what the CPU is using.

Ok let me double check again on this. As I have tested SMMUv3 driver on N1SDP board it is observed as same.
> 
> For the IAS, you will want to use p2m_ipa_bits and for the output, we will want to cap to PADDR_BITS.
> 
>> +	reg |= VTCR_SL0(0x2);
> 
> Similar to above, the starting level will depend on how the P2M was configured.
> 
>> +	reg |= VTCR_VS;
> 
> AFAICT, the bit 179 (bit 19 in the word) is indicating whether AArch32 or AArch64 translation table is used. However, bit 19 in VTCR_EL2 indicates whether we are using 8-bit or 16-bit VMID.

Ok let me double check all the configuration for Stream Table Entry.

> 
>> +
>> +	cfg->vtcr   = reg;
> 
> It would be better to initialize vtcr exactly at the same place as Linux does. This would make easier to match the code.

Ok. 
>> +
>>  	vmid = arm_smmu_bitmap_alloc(smmu->vmid_map, smmu->vmid_bits);
>>  	if (vmid < 0)
>>  		return vmid;
>> +	cfg->vmid  = (u16)vmid;
>> +
>> +	cfg->vttbr  = page_to_maddr(cfg->domain->arch.p2m.root);
>> +
>> +	printk(XENLOG_DEBUG
>> +		   "SMMUv3: d%u: vmid 0x%x vtcr 0x%"PRIpaddr" p2maddr 0x%"PRIpaddr"\n",
>> +		   cfg->domain->domain_id, cfg->vmid, cfg->vtcr, cfg->vttbr);
>>  -	vtcr = &pgtbl_cfg->arm_lpae_s2_cfg.vtcr;
>> -	cfg->vmid	= (u16)vmid;
>> -	cfg->vttbr	= pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
>> -	cfg->vtcr	= FIELD_PREP(STRTAB_STE_2_VTCR_S2T0SZ, vtcr->tsz) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SL0, vtcr->sl) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2IR0, vtcr->irgn) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2OR0, vtcr->orgn) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2SH0, vtcr->sh) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2TG, vtcr->tg) |
>> -			  FIELD_PREP(STRTAB_STE_2_VTCR_S2PS, vtcr->ps);
>>  	return 0;
>>  }
>>  @@ -1382,28 +1689,12 @@ static int arm_smmu_domain_finalise(struct iommu_domain *domain,
>>  				    struct arm_smmu_master *master)
>>  {
>>  	int ret;
>> -	unsigned long ias, oas;
>> -	int (*finalise_stage_fn)(struct arm_smmu_domain *,
>> -				 struct arm_smmu_master *,
>> -				 struct io_pgtable_cfg *);
>>  	struct arm_smmu_domain *smmu_domain = to_smmu_domain(domain);
>> -	struct arm_smmu_device *smmu = smmu_domain->smmu;
>>    	/* Restrict the stage to what we can actually support */
>>  	smmu_domain->stage = ARM_SMMU_DOMAIN_S2;
>>  -	switch (smmu_domain->stage) {
>> -	case ARM_SMMU_DOMAIN_NESTED:
>> -	case ARM_SMMU_DOMAIN_S2:
>> -		ias = smmu->ias;
>> -		oas = smmu->oas;
>> -		finalise_stage_fn = arm_smmu_domain_finalise_s2;
>> -		break;
>> -	default:
>> -		return -EINVAL;
>> -	}
>> -
>> -	ret = finalise_stage_fn(smmu_domain, master, &pgtbl_cfg);
> 
> It is not entirely clear why this code is removed here and not the previous patch?

I will removed this code in previous patch. Somehow I missed.
> 
>> +	ret = arm_smmu_domain_finalise_s2(smmu_domain, master);
>>  	if (ret < 0) {
>>  		return ret;
>>  	}
>> @@ -1553,7 +1844,8 @@ static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>>  		return -ENOMEM;
>>  	}
>>  -	if (!WARN_ON(q->base_dma & (qsz - 1))) {
>> +	WARN_ON(q->base_dma & (qsz - 1));
> 
> This is a call to rework how our WARN_ON() in Xen.

Yes.
> 
>> +	if (unlikely(q->base_dma & (qsz - 1))) {
>>  		dev_info(smmu->dev, "allocated %u entries for %s\n",
>>  			 1 << q->llq.max_n_shift, name);
>>  	}
>> @@ -1758,9 +2050,7 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>>  	/* Request interrupt lines */
>>  	irq = smmu->evtq.q.irq;
>>  	if (irq) {
>> -		ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
>> -						arm_smmu_evtq_thread,
>> -						IRQF_ONESHOT,
>> +		ret = request_irq(irq, 0, arm_smmu_evtq_irq_tasklet,
>>  						"arm-smmu-v3-evtq", smmu);
>>  		if (ret < 0)
>>  			dev_warn(smmu->dev, "failed to enable evtq irq\n");
>> @@ -1770,8 +2060,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>>    	irq = smmu->gerr_irq;
>>  	if (irq) {
>> -		ret = devm_request_irq(smmu->dev, irq, arm_smmu_gerror_handler,
>> -				       0, "arm-smmu-v3-gerror", smmu);
>> +		ret = request_irq(irq, 0, arm_smmu_gerror_handler,
>> +						"arm-smmu-v3-gerror", smmu);
>>  		if (ret < 0)
>>  			dev_warn(smmu->dev, "failed to enable gerror irq\n");
>>  	} else {
>> @@ -1781,11 +2071,8 @@ static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>>  	if (smmu->features & ARM_SMMU_FEAT_PRI) {
>>  		irq = smmu->priq.q.irq;
>>  		if (irq) {
>> -			ret = devm_request_threaded_irq(smmu->dev, irq, NULL,
>> -							arm_smmu_priq_thread,
>> -							IRQF_ONESHOT,
>> -							"arm-smmu-v3-priq",
>> -							smmu);
>> +			ret = request_irq(irq, 0, arm_smmu_priq_irq_tasklet,
>> +							"arm-smmu-v3-priq", smmu);
>>  			if (ret < 0)
>>  				dev_warn(smmu->dev,
>>  					 "failed to enable priq irq\n");
>> @@ -1814,11 +2101,8 @@ static int arm_smmu_setup_irqs(struct arm_smmu_device *smmu)
>>  		 * Cavium ThunderX2 implementation doesn't support unique irq
>>  		 * lines. Use a single irq line for all the SMMUv3 interrupts.
>>  		 */
>> -		ret = devm_request_threaded_irq(smmu->dev, irq,
>> -					arm_smmu_combined_irq_handler,
>> -					arm_smmu_combined_irq_thread,
>> -					IRQF_ONESHOT,
>> -					"arm-smmu-v3-combined-irq", smmu);
>> +		ret = request_irq(irq, 0, arm_smmu_combined_irq_handler,
>> +						"arm-smmu-v3-combined-irq", smmu);
>>  		if (ret < 0)
>>  			dev_warn(smmu->dev, "failed to enable combined irq\n");
>>  	} else
>> @@ -1857,7 +2141,7 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>>  	reg = readl_relaxed(smmu->base + ARM_SMMU_CR0);
>>  	if (reg & CR0_SMMUEN) {
>>  		dev_warn(smmu->dev, "SMMU currently enabled! Resetting...\n");
>> -		WARN_ON(is_kdump_kernel() && !disable_bypass);
>> +		WARN_ON(!disable_bypass);
>>  		arm_smmu_update_gbpa(smmu, GBPA_ABORT, 0);
>>  	}
>>  @@ -1952,8 +2236,11 @@ static int arm_smmu_device_reset(struct arm_smmu_device *smmu, bool bypass)
>>  		return ret;
>>  	}
>>  -	if (is_kdump_kernel())
>> -		enables &= ~(CR0_EVTQEN | CR0_PRIQEN);
>> +	/* Initialize tasklets for threaded IRQs*/
>> +	tasklet_init(&smmu->evtq_irq_tasklet, arm_smmu_evtq_thread, smmu);
>> +	tasklet_init(&smmu->priq_irq_tasklet, arm_smmu_priq_thread, smmu);
>> +	tasklet_init(&smmu->combined_irq_tasklet, arm_smmu_combined_irq_thread,
>> +				 smmu);
>>    	/* Enable the SMMU interface, or ensure bypass */
>>  	if (!bypass || disable_bypass) {
>> @@ -2195,7 +2482,7 @@ static inline int arm_smmu_device_acpi_probe(struct platform_device *pdev,
>>  static int arm_smmu_device_dt_probe(struct platform_device *pdev,
>>  				    struct arm_smmu_device *smmu)
>>  {
>> -	struct device *dev = &pdev->dev;
>> +	struct device *dev = pdev;
>>  	u32 cells;
>>  	int ret = -EINVAL;
>>  @@ -2219,130 +2506,449 @@ static unsigned long arm_smmu_resource_size(struct arm_smmu_device *smmu)
>>  		return SZ_128K;
>>  }
>>  +/* Start of Xen specific code. */
>>  static int arm_smmu_device_probe(struct platform_device *pdev)
>>  {
>> -	int irq, ret;
>> -	struct resource *res;
>> -	resource_size_t ioaddr;
>> -	struct arm_smmu_device *smmu;
>> -	struct device *dev = &pdev->dev;
>> -	bool bypass;
>> +    int irq, ret;
>> +    paddr_t ioaddr, iosize;
>> +    struct arm_smmu_device *smmu;
>> +    bool bypass;
>> +
>> +    smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
>> +    if ( !smmu )
>> +    {
>> +        dev_err(pdev, "failed to allocate arm_smmu_device\n");
>> +        return -ENOMEM;
>> +    }
>> +    smmu->dev = pdev;
>> +
>> +    if ( pdev->of_node )
>> +    {
>> +        ret = arm_smmu_device_dt_probe(pdev, smmu);
>> +    } else
>> +    {
>> +        ret = arm_smmu_device_acpi_probe(pdev, smmu);
>> +        if ( ret == -ENODEV )
>> +            return ret;
>> +    }
>> +
>> +    /* Set bypass mode according to firmware probing result */
>> +    bypass = !!ret;
> 
> AFAICT, bypass would be set if the device-tree is buggy. For Xen, I think it would be saner to not continue as this would break isolation.

Ok . I will not configure the SMMU in bypass mode if device-tree probe is failed. I will report error to the user.

> 
>> +
>> +    /* Base address */
>> +    ret = dt_device_get_address(dev_to_dt(pdev), 0, &ioaddr, &iosize);
>> +    if( ret )
>> +        return -ENODEV;
>> +
>> +    if ( iosize < arm_smmu_resource_size(smmu) )
>> +    {
>> +        dev_err(pdev, "MMIO region too small (%lx)\n", iosize);
>> +        return -EINVAL;
>> +    }
>> +
>> +    /*
>> +     * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
>> +     * the PMCG registers which are reserved by the PMU driver.
>> +     */
> 
> This comment doesn't seem to match the code. Did you intend to...

Yes I will fix the code and will use ARM_SMMU_REG_SZ as size.
> 
> 
> 
>> +    smmu->base = ioremap_nocache(ioaddr, iosize);
> 
> ... use ARM_SMMU_REG_SZ which would only map the necessary region?

Yes I will use ARM_SMMU_REG_SZ .
> 
> 
>> +    if ( IS_ERR(smmu->base) )
>> +        return PTR_ERR(smmu->base);
>> +
>> +    if ( iosize > SZ_64K )
>> +    {
>> +        smmu->page1 = ioremap_nocache(ioaddr + SZ_64K, ARM_SMMU_REG_SZ);
>> +        if ( IS_ERR(smmu->page1) )
>> +            return PTR_ERR(smmu->page1);
>> +    }
>> +    else
>> +    {
>> +        smmu->page1 = smmu->base;
>> +    }
>> +
>> +    /* Interrupt lines */
>> +
>> +    irq = platform_get_irq_byname_optional(pdev, "combined");
>> +    if ( irq > 0 )
>> +        smmu->combined_irq = irq;
>> +    else
>> +    {
>> +        irq = platform_get_irq_byname_optional(pdev, "eventq");
>> +        if ( irq > 0 )
>> +            smmu->evtq.q.irq = irq;
>> +
>> +        irq = platform_get_irq_byname_optional(pdev, "priq");
>> +        if ( irq > 0 )
>> +            smmu->priq.q.irq = irq;
>> +
>> +        irq = platform_get_irq_byname_optional(pdev, "gerror");
>> +        if ( irq > 0 )
>> +            smmu->gerr_irq = irq;
>> +    }
>> +    /* Probe the h/w */
>> +    ret = arm_smmu_device_hw_probe(smmu);
>> +    if ( ret )
>> +        return ret;
>> +
>> +    /* Initialise in-memory data structures */
>> +    ret = arm_smmu_init_structures(smmu);
>> +    if ( ret )
>> +        return ret;
>> +
>> +    /* Reset the device */
>> +    ret = arm_smmu_device_reset(smmu, bypass);
>> +    if ( ret )
>> +        return ret;
>> +
>> +    /*
>> +     * Keep a list of all probed devices. This will be used to query
>> +     * the smmu devices based on the fwnode.
>> +     */
>> +    INIT_LIST_HEAD(&smmu->devices);
>> +
>> +    spin_lock(&arm_smmu_devices_lock);
>> +    list_add(&smmu->devices, &arm_smmu_devices);
>> +    spin_unlock(&arm_smmu_devices_lock);
>> +
>> +    return 0;
>> +}
>> +
>> +static int __must_check arm_smmu_iotlb_flush_all(struct domain *d)
>> +{
>> +    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
>> +    struct iommu_domain *io_domain;
>> +
>> +    spin_lock(&xen_domain->lock);
>> +
>> +    list_for_each_entry( io_domain, &xen_domain->contexts, list )
>> +    {
>> +        /*
>> +         * Only invalidate the context when SMMU is present.
>> +         * This is because the context initialization is delayed
>> +         * until a master has been added.
>> +         */
>> +        if ( unlikely(!ACCESS_ONCE(to_smmu_domain(io_domain)->smmu)) )
>> +            continue;
>> +
>> +        arm_smmu_tlb_inv_context(to_smmu_domain(io_domain));
>> +    }
>>  -	smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
>> -	if (!smmu) {
>> -		dev_err(dev, "failed to allocate arm_smmu_device\n");
>> -		return -ENOMEM;
>> -	}
>> -	smmu->dev = dev;
>> +    spin_unlock(&xen_domain->lock);
>> +
>> +    return 0;
>> +}
>> +
>> +static int __must_check arm_smmu_iotlb_flush(struct domain *d, dfn_t dfn,
>> +                                             unsigned long page_count,
>> +                                             unsigned int flush_flags)
>> +{
>> +    return arm_smmu_iotlb_flush_all(d);
>> +}
>>  -	if (dev->of_node) {
>> -		ret = arm_smmu_device_dt_probe(pdev, smmu);
>> -	} else {
>> -		ret = arm_smmu_device_acpi_probe(pdev, smmu);
>> -		if (ret == -ENODEV)
>> -			return ret;
>> -	}
>> +static struct arm_smmu_device *arm_smmu_get_by_dev(struct device *dev)
>> +{
>> +    struct arm_smmu_device *smmu = NULL;
>>  -	/* Set bypass mode according to firmware probing result */
>> -	bypass = !!ret;
>> +    spin_lock(&arm_smmu_devices_lock);
>> +    list_for_each_entry( smmu, &arm_smmu_devices, devices )
>> +    {
>> +        if ( smmu->dev  == dev )
>> +        {
>> +            spin_unlock(&arm_smmu_devices_lock);
>> +            return smmu;
>> +        }
>> +    }
>> +    spin_unlock(&arm_smmu_devices_lock);
>>  -	/* Base address */
>> -	res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
>> -	if (resource_size(res) < arm_smmu_resource_size(smmu)) {
>> -		dev_err(dev, "MMIO region too small (%pr)\n", res);
>> -		return -EINVAL;
>> -	}
>> -	ioaddr = res->start;
> 
> The code removed is basically the same as the one you added except the coding style change. This patch is already quite long to review, so can we please keep the change to the strict minimum?
> 
> If you want to do to clean-up then they should be done before/after.

Ok.
> 
>> +    return NULL;
>> +}
>>  -	/*
>> -	 * Don't map the IMPLEMENTATION DEFINED regions, since they may contain
>> -	 * the PMCG registers which are reserved by the PMU driver.
>> -	 */
>> -	smmu->base = arm_smmu_ioremap(dev, ioaddr, ARM_SMMU_REG_SZ);
>> -	if (IS_ERR(smmu->base))
>> -		return PTR_ERR(smmu->base);
>> -
>> -	if (arm_smmu_resource_size(smmu) > SZ_64K) {
>> -		smmu->page1 = arm_smmu_ioremap(dev, ioaddr + SZ_64K,
>> -					       ARM_SMMU_REG_SZ);
>> -		if (IS_ERR(smmu->page1))
>> -			return PTR_ERR(smmu->page1);
>> -	} else {
>> -		smmu->page1 = smmu->base;
>> -	}
>> +/* Probing and initialisation functions */
>> +static struct iommu_domain *arm_smmu_get_domain(struct domain *d,
>> +                                                struct device *dev)
>> +{
>> +    struct iommu_domain *io_domain;
>> +    struct arm_smmu_domain *smmu_domain;
>> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> +    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
>> +    struct arm_smmu_device *smmu = arm_smmu_get_by_dev(fwspec->iommu_dev);
>> +
>> +    if ( !smmu )
>> +        return NULL;
>>  -	/* Interrupt lines */
>> +    /*
>> +     * Loop through the &xen_domain->contexts to locate a context
>> +     * assigned to this SMMU
>> +     */
>> +    list_for_each_entry( io_domain, &xen_domain->contexts, list )
>> +    {
>> +        smmu_domain = to_smmu_domain(io_domain);
>> +        if ( smmu_domain->smmu == smmu )
>> +            return io_domain;
>> +    }
>>  -	irq = platform_get_irq_byname_optional(pdev, "combined");
>> -	if (irq > 0)
>> -		smmu->combined_irq = irq;
>> -	else {
>> -		irq = platform_get_irq_byname_optional(pdev, "eventq");
>> -		if (irq > 0)
>> -			smmu->evtq.q.irq = irq;
>> +    return NULL;
>> +}
>>  -		irq = platform_get_irq_byname_optional(pdev, "priq");
>> -		if (irq > 0)
>> -			smmu->priq.q.irq = irq;
>> +static void arm_smmu_destroy_iommu_domain(struct iommu_domain *io_domain)
>> +{
>> +    list_del(&io_domain->list);
>> +    arm_smmu_domain_free(io_domain);
>> +}
>>  -		irq = platform_get_irq_byname_optional(pdev, "gerror");
>> -		if (irq > 0)
>> -			smmu->gerr_irq = irq;
>> -	}
>> -	/* Probe the h/w */
>> -	ret = arm_smmu_device_hw_probe(smmu);
>> -	if (ret)
>> -		return ret;
>> +static int arm_smmu_assign_dev(struct domain *d, u8 devfn,
>> +                               struct device *dev, u32 flag)
>> +{
>> +    int ret = 0;
>> +    struct iommu_domain *io_domain;
>> +    struct arm_smmu_domain *smmu_domain;
>> +    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
>>  -	/* Initialise in-memory data structures */
>> -	ret = arm_smmu_init_structures(smmu);
>> -	if (ret)
>> -		return ret;
>> +    if ( !dev->archdata.iommu )
>> +    {
>> +        dev->archdata.iommu = xzalloc(struct arm_smmu_xen_device);
>> +        if ( !dev->archdata.iommu )
>> +            return -ENOMEM;
>> +    }
>>  -	/* Record our private device structure */
>> -	platform_set_drvdata(pdev, smmu);
>> +    spin_lock(&xen_domain->lock);
>>  -	/* Reset the device */
>> -	ret = arm_smmu_device_reset(smmu, bypass);
>> -	if (ret)
>> -		return ret;
>> +    /*
>> +     * Check to see if an iommu_domain already exists for this xen domain
>> +     * under the same SMMU
>> +     */
>> +    io_domain = arm_smmu_get_domain(d, dev);
>> +    if ( !io_domain )
>> +    {
>> +        io_domain = arm_smmu_domain_alloc(IOMMU_DOMAIN_DMA);
>> +        if ( !io_domain )
>> +        {
>> +            ret = -ENOMEM;
>> +            goto out;
>> +        }
>>  -	/* And we're up. Go go go! */
>> -	ret = iommu_device_sysfs_add(&smmu->iommu, dev, NULL,
>> -				     "smmu3.%pa", &ioaddr);
>> -	if (ret)
>> -		return ret;
>> +        smmu_domain = to_smmu_domain(io_domain);
>> +        smmu_domain->s2_cfg.domain = d;
>>  -	iommu_device_set_ops(&smmu->iommu, &arm_smmu_ops);
>> -	iommu_device_set_fwnode(&smmu->iommu, dev->fwnode);
>> +        /* Chain the new context to the domain */
>> +        list_add(&io_domain->list, &xen_domain->contexts);
>>  -	ret = iommu_device_register(&smmu->iommu);
>> -	if (ret) {
>> -		dev_err(dev, "Failed to register iommu\n");
>> -		return ret;
>> -	}
>> +    }
>> +
>> +    ret = arm_smmu_attach_dev(io_domain, dev);
>> +    if ( ret )
>> +    {
>> +        if ( io_domain->ref.counter == 0 )
>> +            arm_smmu_destroy_iommu_domain(io_domain);
>> +    }
>> +    else
>> +    {
>> +        atomic_inc(&io_domain->ref);
>> +    }
>>  -	return arm_smmu_set_bus_ops(&arm_smmu_ops);
>> +out:
>> +    spin_unlock(&xen_domain->lock);
>> +    return ret;
>>  }
>>  -static int arm_smmu_device_remove(struct platform_device *pdev)
>> +static int arm_smmu_deassign_dev(struct domain *d, struct device *dev)
>>  {
>> -	struct arm_smmu_device *smmu = platform_get_drvdata(pdev);
>> +    struct iommu_domain *io_domain = arm_smmu_get_domain(d, dev);
>> +    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
>> +    struct arm_smmu_domain *arm_smmu = to_smmu_domain(io_domain);
>> +    struct arm_smmu_master *master = dev_iommu_priv_get(dev);
>>  -	arm_smmu_set_bus_ops(NULL);
>> -	iommu_device_unregister(&smmu->iommu);
>> -	iommu_device_sysfs_remove(&smmu->iommu);
>> -	arm_smmu_device_disable(smmu);
>> +    if ( !arm_smmu || arm_smmu->s2_cfg.domain != d )
>> +    {
>> +        dev_err(dev, " not attached to domain %d\n", d->domain_id);
>> +        return -ESRCH;
>> +    }
>>  -	return 0;
>> +    spin_lock(&xen_domain->lock);
>> +
>> +    arm_smmu_detach_dev(master);
>> +    atomic_dec(&io_domain->ref);
>> +
>> +    if ( io_domain->ref.counter == 0 )
>> +        arm_smmu_destroy_iommu_domain(io_domain);
>> +
>> +    spin_unlock(&xen_domain->lock);
>> +
>> +    return 0;
>> +}
>> +
>> +static int arm_smmu_reassign_dev(struct domain *s, struct domain *t,
>> +                                 u8 devfn,  struct device *dev)
>> +{
>> +    int ret = 0;
>> +
>> +    /* Don't allow remapping on other domain than hwdom */
>> +    if ( t && t != hardware_domain )
>> +        return -EPERM;
>> +
>> +    if ( t == s )
>> +        return 0;
>> +
>> +    ret = arm_smmu_deassign_dev(s, dev);
>> +    if ( ret )
>> +        return ret;
>> +
>> +    if ( t )
>> +    {
>> +        /* No flags are defined for ARM. */
>> +        ret = arm_smmu_assign_dev(t, devfn, dev, 0);
>> +        if ( ret )
>> +            return ret;
>> +    }
>> +
>> +    return 0;
>> +}
>> +
>> +static int arm_smmu_iommu_xen_domain_init(struct domain *d)
>> +{
>> +    struct arm_smmu_xen_domain *xen_domain;
>> +
>> +    xen_domain = xzalloc(struct arm_smmu_xen_domain);
>> +    if ( !xen_domain )
>> +        return -ENOMEM;
>> +
>> +    spin_lock_init(&xen_domain->lock);
>> +    INIT_LIST_HEAD(&xen_domain->contexts);
>> +
>> +    dom_iommu(d)->arch.priv = xen_domain;
>> +
>> +    return 0;
>> +}
>> +
>> +static void __hwdom_init arm_smmu_iommu_hwdom_init(struct domain *d)
>> +{
>> +}
>> +
>> +static void arm_smmu_iommu_xen_domain_teardown(struct domain *d)
>> +{
>> +    struct arm_smmu_xen_domain *xen_domain = dom_iommu(d)->arch.priv;
>> +
>> +    ASSERT(list_empty(&xen_domain->contexts));
>> +    xfree(xen_domain);
>> +}
>> +
>> +static int arm_smmu_dt_xlate(struct device *dev,
>> +                             const struct dt_phandle_args *args)
>> +{
>> +    int ret;
>> +    struct iommu_fwspec *fwspec = dev_iommu_fwspec_get(dev);
>> +
>> +    ret = iommu_fwspec_add_ids(dev, args->args, 1);
>> +    if ( ret )
>> +        return ret;
>> +
>> +    if ( dt_device_is_protected(dev_to_dt(dev)) )
>> +    {
>> +        dev_err(dev, "Already added to SMMUv3\n");
>> +        return -EEXIST;
>> +    }
>> +
>> +    /* Let Xen know that the master device is protected by an IOMMU. */
>> +    dt_device_set_protected(dev_to_dt(dev));
>> +
>> +    dev_info(dev, "Added master device (SMMUv3 %s StreamIds %u)\n",
>> +            dev_name(fwspec->iommu_dev), fwspec->num_ids);
>> +
>> +    return 0;
>>  }
>>  -static void arm_smmu_device_shutdown(struct platform_device *pdev)
> 
> I think this function should have been dropped in the previous patch.

OK. 
> 
>> +static int arm_smmu_add_device(u8 devfn, struct device *dev)
>>  {
>> -	arm_smmu_device_remove(pdev);
>> +    int i, ret;
>> +    struct arm_smmu_device *smmu;
>> +    struct arm_smmu_master *master;
>> +    struct iommu_fwspec *fwspec;
>> +
>> +    fwspec = dev_iommu_fwspec_get(dev);
>> +    if ( !fwspec )
>> +        return -ENODEV;
>> +
>> +    smmu = arm_smmu_get_by_dev(fwspec->iommu_dev);
>> +    if ( !smmu )
>> +        return -ENODEV;
>> +
>> +    master = xzalloc(struct arm_smmu_master);
>> +    if ( !master )
>> +        return -ENOMEM;
>> +
>> +    master->dev = dev;
>> +    master->smmu = smmu;
>> +    master->sids = fwspec->ids;
>> +    master->num_sids = fwspec->num_ids;
>> +
>> +    dev_iommu_priv_set(dev, master);
>> +
>> +    /* Check the SIDs are in range of the SMMU and our stream table */
>> +    for ( i = 0; i < master->num_sids; i++ )
>> +    {
>> +        u32 sid = master->sids[i];
>> +
>> +        if ( !arm_smmu_sid_in_range(smmu, sid) )
>> +        {
>> +            ret = -ERANGE;
>> +            goto err_free_master;
>> +        }
>> +
>> +        /* Ensure l2 strtab is initialised */
>> +        if ( smmu->features & ARM_SMMU_FEAT_2_LVL_STRTAB )
>> +        {
>> +            ret = arm_smmu_init_l2_strtab(smmu, sid);
>> +            if ( ret )
>> +                goto err_free_master;
>> +        }
>> +    }
>> +
>> +    return 0;
>> +
>> +err_free_master:
>> +    xfree(master);
>> +    dev_iommu_priv_set(dev, NULL);
>> +    return ret;
>>  }
>>  -static const struct of_device_id arm_smmu_of_match[] = {
>> -	{ .compatible = "arm,smmu-v3", },
>> -	{ },
>> +static const struct iommu_ops arm_smmu_iommu_ops = {
>> +    .init = arm_smmu_iommu_xen_domain_init,
>> +    .hwdom_init = arm_smmu_iommu_hwdom_init,
>> +    .teardown = arm_smmu_iommu_xen_domain_teardown,
>> +    .iotlb_flush = arm_smmu_iotlb_flush,
>> +    .iotlb_flush_all = arm_smmu_iotlb_flush_all,
>> +    .assign_device = arm_smmu_assign_dev,
>> +    .reassign_device = arm_smmu_reassign_dev,
>> +    .map_page = arm_iommu_map_page,
>> +    .unmap_page = arm_iommu_unmap_page,
>> +    .dt_xlate = arm_smmu_dt_xlate,
>> +    .add_device = arm_smmu_add_device,
>> +};
>> +
>> +static const struct dt_device_match arm_smmu_of_match[] = {
>> +    { .compatible = "arm,smmu-v3", },
>> +    { },
>>  };
>> +
>> +static __init int arm_smmu_dt_init(struct dt_device_node *dev,
>> +                                   const void *data)
>> +{
>> +    int rc;
>> +
>> +    /*
>> +     * Even if the device can't be initialized, we don't want to
>> +     * give the SMMU device to dom0.
>> +     */
>> +    dt_device_set_used_by(dev, DOMID_XEN);
>> +
>> +    rc = arm_smmu_device_probe(dt_to_dev(dev));
>> +    if ( rc )
>> +        return rc;
>> +
>> +    iommu_set_ops(&arm_smmu_iommu_ops);
>> +    return 0;
>> +}
>> +
>> +DT_DEVICE_START(smmuv3, "ARM SMMU V3", DEVICE_IOMMU)
>> +    .dt_match = arm_smmu_of_match,
>> +    .init = arm_smmu_dt_init,
>> +DT_DEVICE_END
> 
> Cheers,
> 
> -- 
> Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-07 12:12     ` Rahul Singh
@ 2020-12-07 17:39       ` Julien Grall
  2020-12-07 18:42         ` Rahul Singh
  0 siblings, 1 reply; 53+ messages in thread
From: Julien Grall @ 2020-12-07 17:39 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, Bertrand Marquis, Andrew Cooper, George Dunlap,
	Ian Jackson, Jan Beulich, Stefano Stabellini, Wei Liu,
	Paul Durrant, Volodymyr Babchuk



On 07/12/2020 12:12, Rahul Singh wrote:
>>> +typedef paddr_t dma_addr_t;
>>> +typedef unsigned int gfp_t;
>>> +
>>> +#define platform_device device
>>> +
>>> +#define GFP_KERNEL 0
>>> +
>>> +/* Alias to Xen device tree helpers */
>>> +#define device_node dt_device_node
>>> +#define of_phandle_args dt_phandle_args
>>> +#define of_device_id dt_device_match
>>> +#define of_match_node dt_match_node
>>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, pname, out))
>>> +#define of_property_read_bool dt_property_read_bool
>>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
>>> +
>>> +/* Alias to Xen lock functions */
>>> +#define mutex spinlock
>>> +#define mutex_init spin_lock_init
>>> +#define mutex_lock spin_lock
>>> +#define mutex_unlock spin_unlock
>>
>> Hmm... mutex are not spinlock. Can you explain why this is fine to switch to spinlock?
> 
> Yes mutex are not spinlock. As mutex is not implemented in XEN I thought of using spinlock in place of mutex as this is the only locking mechanism available in XEN.
> Let me know if there is another blocking lock available in XEN. I will check if we can use that.

There are no blocking lock available in Xen so far. However, if Linux 
were using mutex instead of spinlock, then it likely means they 
operations in the critical section can be long running.

How did you came to the conclusion that using spinlock in the SMMU 
driver would be fine?

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-07 17:39       ` Julien Grall
@ 2020-12-07 18:42         ` Rahul Singh
  2020-12-08 19:05           ` Julien Grall
  0 siblings, 1 reply; 53+ messages in thread
From: Rahul Singh @ 2020-12-07 18:42 UTC (permalink / raw)
  To: Julien Grall
  Cc: xen-devel, Bertrand Marquis, Andrew Cooper, George Dunlap,
	Ian Jackson, Jan Beulich, Stefano Stabellini, Wei Liu,
	Paul Durrant, Volodymyr Babchuk

Hello Julien,

> On 7 Dec 2020, at 5:39 pm, Julien Grall <julien@xen.org> wrote:
> 
> 
> 
> On 07/12/2020 12:12, Rahul Singh wrote:
>>>> +typedef paddr_t dma_addr_t;
>>>> +typedef unsigned int gfp_t;
>>>> +
>>>> +#define platform_device device
>>>> +
>>>> +#define GFP_KERNEL 0
>>>> +
>>>> +/* Alias to Xen device tree helpers */
>>>> +#define device_node dt_device_node
>>>> +#define of_phandle_args dt_phandle_args
>>>> +#define of_device_id dt_device_match
>>>> +#define of_match_node dt_match_node
>>>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, pname, out))
>>>> +#define of_property_read_bool dt_property_read_bool
>>>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
>>>> +
>>>> +/* Alias to Xen lock functions */
>>>> +#define mutex spinlock
>>>> +#define mutex_init spin_lock_init
>>>> +#define mutex_lock spin_lock
>>>> +#define mutex_unlock spin_unlock
>>> 
>>> Hmm... mutex are not spinlock. Can you explain why this is fine to switch to spinlock?
>> Yes mutex are not spinlock. As mutex is not implemented in XEN I thought of using spinlock in place of mutex as this is the only locking mechanism available in XEN.
>> Let me know if there is another blocking lock available in XEN. I will check if we can use that.
> 
> There are no blocking lock available in Xen so far. However, if Linux were using mutex instead of spinlock, then it likely means they operations in the critical section can be long running.

Yes you are right Linux is using mutex when attaching a device to the SMMU as this operation might take longer time.
> 
> How did you came to the conclusion that using spinlock in the SMMU driver would be fine?

Mutex is replaced by spinlock  in the SMMU driver when there is a request to assign a device to the guest. As we are in user context at that time its ok to use spinlock.
As per my understanding there is one scenario when CPU will spin when there is a request from the user at the same time to assign another device to the SMMU and I think that is very rare.

Please suggest how we can proceed on this.

Regards
Rahul
> 
> Cheers,
> 
> -- 
> Julien Grall



^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-07 18:42         ` Rahul Singh
@ 2020-12-08 19:05           ` Julien Grall
  2020-12-09  1:19             ` Stefano Stabellini
  0 siblings, 1 reply; 53+ messages in thread
From: Julien Grall @ 2020-12-08 19:05 UTC (permalink / raw)
  To: Rahul Singh
  Cc: xen-devel, Bertrand Marquis, Andrew Cooper, George Dunlap,
	Ian Jackson, Jan Beulich, Stefano Stabellini, Wei Liu,
	Paul Durrant, Volodymyr Babchuk



On 07/12/2020 18:42, Rahul Singh wrote:
> Hello Julien,

Hi,

> 
>> On 7 Dec 2020, at 5:39 pm, Julien Grall <julien@xen.org> wrote:
>>
>>
>>
>> On 07/12/2020 12:12, Rahul Singh wrote:
>>>>> +typedef paddr_t dma_addr_t;
>>>>> +typedef unsigned int gfp_t;
>>>>> +
>>>>> +#define platform_device device
>>>>> +
>>>>> +#define GFP_KERNEL 0
>>>>> +
>>>>> +/* Alias to Xen device tree helpers */
>>>>> +#define device_node dt_device_node
>>>>> +#define of_phandle_args dt_phandle_args
>>>>> +#define of_device_id dt_device_match
>>>>> +#define of_match_node dt_match_node
>>>>> +#define of_property_read_u32(np, pname, out) (!dt_property_read_u32(np, pname, out))
>>>>> +#define of_property_read_bool dt_property_read_bool
>>>>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
>>>>> +
>>>>> +/* Alias to Xen lock functions */
>>>>> +#define mutex spinlock
>>>>> +#define mutex_init spin_lock_init
>>>>> +#define mutex_lock spin_lock
>>>>> +#define mutex_unlock spin_unlock
>>>>
>>>> Hmm... mutex are not spinlock. Can you explain why this is fine to switch to spinlock?
>>> Yes mutex are not spinlock. As mutex is not implemented in XEN I thought of using spinlock in place of mutex as this is the only locking mechanism available in XEN.
>>> Let me know if there is another blocking lock available in XEN. I will check if we can use that.
>>
>> There are no blocking lock available in Xen so far. However, if Linux were using mutex instead of spinlock, then it likely means they operations in the critical section can be long running.
> 
> Yes you are right Linux is using mutex when attaching a device to the SMMU as this operation might take longer time.
>>
>> How did you came to the conclusion that using spinlock in the SMMU driver would be fine?
> 
> Mutex is replaced by spinlock  in the SMMU driver when there is a request to assign a device to the guest. As we are in user context at that time its ok to use spinlock.

I am not sure to understand what you mean by "user context" here. Can 
you clarify it?

> As per my understanding there is one scenario when CPU will spin when there is a request from the user at the same time to assign another device to the SMMU and I think that is very rare.

What "user" are you referring to?

> 
> Please suggest how we can proceed on this.

I am guessing that what you are saying is the request to 
assign/de-assign device will be issued by the toolstack and therefore 
they should be trusted.

My concern here is not about someone waiting on the lock to be released. 
It is more the fact that using a mutex() is an insight that the 
operation protected can be long. Depending on the length, this may 
result to unwanted side effect (e.g. other vCPU not scheduled, RCU stall 
in dom0, watchdog hit...).

I recall a discussion from a couple of years ago mentioning that STE 
programming operations can take quite a long time. So I would like to 
understand how long the operation is meant to last.

For a tech preview, this is probably OK to replace the mutex with an 
spinlock. But I would not want this to go past the tech preview stage 
without a proper analysis.

Stefano, what do you think?

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-08 19:05           ` Julien Grall
@ 2020-12-09  1:19             ` Stefano Stabellini
  2020-12-09  7:55               ` Bertrand Marquis
  0 siblings, 1 reply; 53+ messages in thread
From: Stefano Stabellini @ 2020-12-09  1:19 UTC (permalink / raw)
  To: Julien Grall
  Cc: Rahul Singh, xen-devel, Bertrand Marquis, Andrew Cooper,
	George Dunlap, Ian Jackson, Jan Beulich, Stefano Stabellini,
	Wei Liu, Paul Durrant, Volodymyr Babchuk

On Tue, 8 Dec 2020, Julien Grall wrote:
> On 07/12/2020 18:42, Rahul Singh wrote:
> > > On 7 Dec 2020, at 5:39 pm, Julien Grall <julien@xen.org> wrote:
> > > On 07/12/2020 12:12, Rahul Singh wrote:
> > > > > > +typedef paddr_t dma_addr_t;
> > > > > > +typedef unsigned int gfp_t;
> > > > > > +
> > > > > > +#define platform_device device
> > > > > > +
> > > > > > +#define GFP_KERNEL 0
> > > > > > +
> > > > > > +/* Alias to Xen device tree helpers */
> > > > > > +#define device_node dt_device_node
> > > > > > +#define of_phandle_args dt_phandle_args
> > > > > > +#define of_device_id dt_device_match
> > > > > > +#define of_match_node dt_match_node
> > > > > > +#define of_property_read_u32(np, pname, out)
> > > > > > (!dt_property_read_u32(np, pname, out))
> > > > > > +#define of_property_read_bool dt_property_read_bool
> > > > > > +#define of_parse_phandle_with_args dt_parse_phandle_with_args
> > > > > > +
> > > > > > +/* Alias to Xen lock functions */
> > > > > > +#define mutex spinlock
> > > > > > +#define mutex_init spin_lock_init
> > > > > > +#define mutex_lock spin_lock
> > > > > > +#define mutex_unlock spin_unlock
> > > > > 
> > > > > Hmm... mutex are not spinlock. Can you explain why this is fine to
> > > > > switch to spinlock?
> > > > Yes mutex are not spinlock. As mutex is not implemented in XEN I thought
> > > > of using spinlock in place of mutex as this is the only locking
> > > > mechanism available in XEN.
> > > > Let me know if there is another blocking lock available in XEN. I will
> > > > check if we can use that.
> > > 
> > > There are no blocking lock available in Xen so far. However, if Linux were
> > > using mutex instead of spinlock, then it likely means they operations in
> > > the critical section can be long running.
> > 
> > Yes you are right Linux is using mutex when attaching a device to the SMMU
> > as this operation might take longer time.
> > > 
> > > How did you came to the conclusion that using spinlock in the SMMU driver
> > > would be fine?
> > 
> > Mutex is replaced by spinlock  in the SMMU driver when there is a request to
> > assign a device to the guest. As we are in user context at that time its ok
> > to use spinlock.
> 
> I am not sure to understand what you mean by "user context" here. Can you
> clarify it?
> 
> > As per my understanding there is one scenario when CPU will spin when there
> > is a request from the user at the same time to assign another device to the
> > SMMU and I think that is very rare.
> 
> What "user" are you referring to?
> 
> > 
> > Please suggest how we can proceed on this.
> 
> I am guessing that what you are saying is the request to assign/de-assign
> device will be issued by the toolstack and therefore they should be trusted.
> 
> My concern here is not about someone waiting on the lock to be released. It is
> more the fact that using a mutex() is an insight that the operation protected
> can be long. Depending on the length, this may result to unwanted side effect
> (e.g. other vCPU not scheduled, RCU stall in dom0, watchdog hit...).
> 
> I recall a discussion from a couple of years ago mentioning that STE
> programming operations can take quite a long time. So I would like to
> understand how long the operation is meant to last.
> 
> For a tech preview, this is probably OK to replace the mutex with an spinlock.
> But I would not want this to go past the tech preview stage without a proper
> analysis.
> 
> Stefano, what do you think?

In short, I agree.


We need to be very careful replacing mutexes with spinlocks. We need to
look closely at the ways the spinlocks could introduce unwanted
latencies. Concurrent assign_device operations are possible but rare
and, more importantly, they are user-driven so they could be mitigated.
I am more worried about other possible scenarios, e.g. STE or other
operations.

Rahul clearly put a lot of work into this series already and I think it
is better to take this incrementally, which will allow us to do better
testing and also move faster overall. So I am fine to take the series as
is now, pending an investigation on the spinlocks later.


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-09  1:19             ` Stefano Stabellini
@ 2020-12-09  7:55               ` Bertrand Marquis
  2020-12-09  9:18                 ` Julien Grall
  0 siblings, 1 reply; 53+ messages in thread
From: Bertrand Marquis @ 2020-12-09  7:55 UTC (permalink / raw)
  To: Stefano Stabellini
  Cc: Julien Grall, Rahul Singh, xen-devel, Andrew Cooper,
	George Dunlap, Ian Jackson, Jan Beulich, Wei Liu, Paul Durrant,
	Volodymyr Babchuk

Hi,

> On 9 Dec 2020, at 01:19, Stefano Stabellini <sstabellini@kernel.org> wrote:
> 
> On Tue, 8 Dec 2020, Julien Grall wrote:
>> On 07/12/2020 18:42, Rahul Singh wrote:
>>>> On 7 Dec 2020, at 5:39 pm, Julien Grall <julien@xen.org> wrote:
>>>> On 07/12/2020 12:12, Rahul Singh wrote:
>>>>>>> +typedef paddr_t dma_addr_t;
>>>>>>> +typedef unsigned int gfp_t;
>>>>>>> +
>>>>>>> +#define platform_device device
>>>>>>> +
>>>>>>> +#define GFP_KERNEL 0
>>>>>>> +
>>>>>>> +/* Alias to Xen device tree helpers */
>>>>>>> +#define device_node dt_device_node
>>>>>>> +#define of_phandle_args dt_phandle_args
>>>>>>> +#define of_device_id dt_device_match
>>>>>>> +#define of_match_node dt_match_node
>>>>>>> +#define of_property_read_u32(np, pname, out)
>>>>>>> (!dt_property_read_u32(np, pname, out))
>>>>>>> +#define of_property_read_bool dt_property_read_bool
>>>>>>> +#define of_parse_phandle_with_args dt_parse_phandle_with_args
>>>>>>> +
>>>>>>> +/* Alias to Xen lock functions */
>>>>>>> +#define mutex spinlock
>>>>>>> +#define mutex_init spin_lock_init
>>>>>>> +#define mutex_lock spin_lock
>>>>>>> +#define mutex_unlock spin_unlock
>>>>>> 
>>>>>> Hmm... mutex are not spinlock. Can you explain why this is fine to
>>>>>> switch to spinlock?
>>>>> Yes mutex are not spinlock. As mutex is not implemented in XEN I thought
>>>>> of using spinlock in place of mutex as this is the only locking
>>>>> mechanism available in XEN.
>>>>> Let me know if there is another blocking lock available in XEN. I will
>>>>> check if we can use that.
>>>> 
>>>> There are no blocking lock available in Xen so far. However, if Linux were
>>>> using mutex instead of spinlock, then it likely means they operations in
>>>> the critical section can be long running.
>>> 
>>> Yes you are right Linux is using mutex when attaching a device to the SMMU
>>> as this operation might take longer time.
>>>> 
>>>> How did you came to the conclusion that using spinlock in the SMMU driver
>>>> would be fine?
>>> 
>>> Mutex is replaced by spinlock  in the SMMU driver when there is a request to
>>> assign a device to the guest. As we are in user context at that time its ok
>>> to use spinlock.
>> 
>> I am not sure to understand what you mean by "user context" here. Can you
>> clarify it?
>> 
>>> As per my understanding there is one scenario when CPU will spin when there
>>> is a request from the user at the same time to assign another device to the
>>> SMMU and I think that is very rare.
>> 
>> What "user" are you referring to?
>> 
>>> 
>>> Please suggest how we can proceed on this.
>> 
>> I am guessing that what you are saying is the request to assign/de-assign
>> device will be issued by the toolstack and therefore they should be trusted.
>> 
>> My concern here is not about someone waiting on the lock to be released. It is
>> more the fact that using a mutex() is an insight that the operation protected
>> can be long. Depending on the length, this may result to unwanted side effect
>> (e.g. other vCPU not scheduled, RCU stall in dom0, watchdog hit...).
>> 
>> I recall a discussion from a couple of years ago mentioning that STE
>> programming operations can take quite a long time. So I would like to
>> understand how long the operation is meant to last.
>> 
>> For a tech preview, this is probably OK to replace the mutex with an spinlock.
>> But I would not want this to go past the tech preview stage without a proper
>> analysis.
>> 
>> Stefano, what do you think?
> 
> In short, I agree.
> 
> 
> We need to be very careful replacing mutexes with spinlocks. We need to
> look closely at the ways the spinlocks could introduce unwanted
> latencies. Concurrent assign_device operations are possible but rare
> and, more importantly, they are user-driven so they could be mitigated.
> I am more worried about other possible scenarios, e.g. STE or other
> operations.
> 
> Rahul clearly put a lot of work into this series already and I think it
> is better to take this incrementally, which will allow us to do better
> testing and also move faster overall. So I am fine to take the series as
> is now, pending an investigation on the spinlocks later.

I also agree with the issue on the spinlock but we have no equivalent of something
looking like a mutex for now in Xen so this would require some major redesign and
will take us far from the linux driver.

I would suggest to add a comment before this part of the code with a “TODO” so that
it is clear inside the code.
We could also add some comment in Kconfig to mention this possible “faulty” behaviour.

Would you agree on going forward like this ?

Regards
Bertrand



^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-09  7:55               ` Bertrand Marquis
@ 2020-12-09  9:18                 ` Julien Grall
  2020-12-09 18:37                   ` Rahul Singh
  0 siblings, 1 reply; 53+ messages in thread
From: Julien Grall @ 2020-12-09  9:18 UTC (permalink / raw)
  To: Bertrand Marquis, Stefano Stabellini
  Cc: Rahul Singh, xen-devel, Andrew Cooper, George Dunlap,
	Ian Jackson, Jan Beulich, Wei Liu, Paul Durrant,
	Volodymyr Babchuk



On 09/12/2020 07:55, Bertrand Marquis wrote:
> Hi,

Hi,

> I also agree with the issue on the spinlock but we have no equivalent of something
> looking like a mutex for now in Xen so this would require some major redesign and
> will take us far from the linux driver.

I agree that keeping the Xen implementation close to the Linux one is 
important. However, I view this has a the secondary goal. The primary 
goal is to have a safe and secure driver.

If it means introducing a new set of lock or diverging from Linux, then 
so it be.

> 
> I would suggest to add a comment before this part of the code with a “TODO” so that
> it is clear inside the code.
> We could also add some comment in Kconfig to mention this possible “faulty” behaviour.

I think it would be enough to write in the Kconfig that the driver is 
"experimental and should not be used in production".

Ideally, I would like a list of known issues for the driver (could be in 
the cover letter or/and at the top of the source file) so we can track 
what's missing to get it supported.

Cheers,

-- 
Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

* Re: [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver
  2020-12-09  9:18                 ` Julien Grall
@ 2020-12-09 18:37                   ` Rahul Singh
  0 siblings, 0 replies; 53+ messages in thread
From: Rahul Singh @ 2020-12-09 18:37 UTC (permalink / raw)
  To: Julien Grall
  Cc: Bertrand Marquis, Stefano Stabellini, xen-devel, Andrew Cooper,
	George Dunlap, Ian Jackson, Jan Beulich, Wei Liu, Paul Durrant,
	Volodymyr Babchuk

Hello Julien,

> On 9 Dec 2020, at 9:18 am, Julien Grall <julien@xen.org> wrote:
> 
> 
> 
> On 09/12/2020 07:55, Bertrand Marquis wrote:
>> Hi,
> 
> Hi,
> 
>> I also agree with the issue on the spinlock but we have no equivalent of something
>> looking like a mutex for now in Xen so this would require some major redesign and
>> will take us far from the linux driver.
> 
> I agree that keeping the Xen implementation close to the Linux one is important. However, I view this has a the secondary goal. The primary goal is to have a safe and secure driver.
> 
> If it means introducing a new set of lock or diverging from Linux, then so it be.
>> I would suggest to add a comment before this part of the code with a “TODO” so that
>> it is clear inside the code.
>> We could also add some comment in Kconfig to mention this possible “faulty” behaviour.
> 
> I think it would be enough to write in the Kconfig that the driver is "experimental and should not be used in production”.

Ok . I will add in Kconfig.
> 
> Ideally, I would like a list of known issues for the driver (could be in the cover letter or/and at the top of the source file) so we can track what's missing to get it supported.

I will list all known issues and short coming of the driver at the top of the source file.

Regards,
Rahul
> 
> Cheers,
> 
> -- 
> Julien Grall


^ permalink raw reply	[flat|nested] 53+ messages in thread

end of thread, other threads:[~2020-12-09 18:37 UTC | newest]

Thread overview: 53+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-11-26 17:01 [PATCH v2 0/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
2020-11-26 17:02 ` [PATCH v2 1/8] xen/arm: Import the SMMUv3 driver from Linux Rahul Singh
2020-12-01 22:01   ` Stefano Stabellini
2020-11-26 17:02 ` [PATCH v2 2/8] xen/arm: revert atomic operation related command-queue insertion patch Rahul Singh
2020-12-01 22:23   ` Stefano Stabellini
2020-12-02 13:05     ` Rahul Singh
2020-12-02 13:44   ` Julien Grall
2020-12-03 11:49     ` Rahul Singh
2020-11-26 17:02 ` [PATCH v2 3/8] xen/arm: revert patch related to XArray Rahul Singh
2020-12-02  0:20   ` Stefano Stabellini
2020-12-02 13:46   ` Julien Grall
2020-12-03 12:57     ` Rahul Singh
2020-12-04  8:52       ` Julien Grall
2020-11-26 17:02 ` [PATCH v2 4/8] xen/arm: Remove support for MSI on SMMUv3 Rahul Singh
2020-12-02  0:33   ` Stefano Stabellini
2020-12-02  0:40     ` Stefano Stabellini
2020-12-02 13:12       ` Rahul Singh
2020-12-02 14:11         ` Julien Grall
2020-12-03 12:59           ` Rahul Singh
2020-11-26 17:02 ` [PATCH v2 5/8] xen/arm: Remove support for PCI ATS " Rahul Singh
2020-12-02  0:39   ` Stefano Stabellini
2020-12-02 13:07     ` Rahul Singh
2020-12-02 13:57   ` Julien Grall
2020-11-26 17:02 ` [PATCH v2 6/8] xen/arm: Remove support for Stage-1 translation " Rahul Singh
2020-12-02  0:53   ` Stefano Stabellini
2020-12-02 13:13     ` Rahul Singh
2020-11-26 17:02 ` [PATCH v2 7/8] xen/arm: Remove Linux specific code that is not usable in XEN Rahul Singh
2020-12-02  1:48   ` Stefano Stabellini
2020-12-02 14:34     ` Rahul Singh
2020-12-02 14:39       ` Julien Grall
2020-12-02 14:45   ` Julien Grall
2020-12-03 14:33     ` Rahul Singh
2020-12-04  9:05       ` Julien Grall
2020-12-07 10:36         ` Rahul Singh
2020-12-07 10:55           ` Julien Grall
2020-11-26 17:02 ` [PATCH v2 8/8] xen/arm: Add support for SMMUv3 driver Rahul Singh
2020-12-02  2:51   ` Stefano Stabellini
2020-12-02 16:27     ` Rahul Singh
2020-12-02 19:26       ` Rahul Singh
2020-12-02 16:47     ` Julien Grall
2020-12-03  4:13       ` Stefano Stabellini
2020-12-03 14:40         ` Rahul Singh
2020-12-03 18:47           ` Stefano Stabellini
2020-12-07  8:33             ` Rahul Singh
2020-12-02 16:22   ` Julien Grall
2020-12-07 12:12     ` Rahul Singh
2020-12-07 17:39       ` Julien Grall
2020-12-07 18:42         ` Rahul Singh
2020-12-08 19:05           ` Julien Grall
2020-12-09  1:19             ` Stefano Stabellini
2020-12-09  7:55               ` Bertrand Marquis
2020-12-09  9:18                 ` Julien Grall
2020-12-09 18:37                   ` Rahul Singh

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).