[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Xen-devel] [PATCH 5/7] xen/iommu: smmu-v3: Add Xen specific code to enable the ported driver





On 10 March 2018 at 23:23, Manish Jaggi <mjaggi@xxxxxxxxxxxxxxxxxx> wrote:
> Hi Sameer,
>
>
>
> On 02/09/2018 08:40 AM, Sameer Goel wrote:
>>
>> This driver follows an approach similar to smmu driver. The intent here
>> is to reuse as much Linux code as possible.
>> - Glue code has been introduced to bridge the API calls.
>> - Called Linux functions from the Xen IOMMU function calls.
>> - Xen modifications are preceded by /*Xen: comment */
>> - xen/linux_compat: Add a Linux compat header
>>    For porting files directly from Linux it is useful to have a function
>> mapping
>>    definitions from Linux to Xen. This file adds common API functions and
>>    other defines that are needed for porting arm SMMU drivers.
>>
>> Signed-off-by: Sameer Goel <sameer.goel@xxxxxxxxxx>
>> ---
>>   xen/arch/arm/p2m.c                    |   1 +
>>   xen/drivers/Kconfig                   |   2 +
>>   xen/drivers/passthrough/arm/Kconfig   |   8 +
>>   xen/drivers/passthrough/arm/Makefile  |   1 +
>>   xen/drivers/passthrough/arm/smmu-v3.c | 892
>> ++++++++++++++++++++++++++++++++--
>>   xen/include/xen/linux_compat.h        |  84 ++++
>>   6 files changed, 959 insertions(+), 29 deletions(-)
>>   create mode 100644 xen/drivers/passthrough/arm/Kconfig
>>   create mode 100644 xen/include/xen/linux_compat.h
>>
>> diff --git a/xen/arch/arm/p2m.c b/xen/arch/arm/p2m.c
>> index 65e8b9c6ea..fef7605fd6 100644
>> --- a/xen/arch/arm/p2m.c
>> +++ b/xen/arch/arm/p2m.c
>> @@ -1460,6 +1460,7 @@ err:
>>   static void __init setup_virt_paging_one(void *data)
>>   {
>>       unsigned long val = (unsigned long)data;
>> +    /* SMMUv3 S2 cfg vtcr reuses the following value */
>>       WRITE_SYSREG32(val, VTCR_EL2);
>>       isb();
>>   }
>> diff --git a/xen/drivers/Kconfig b/xen/drivers/Kconfig
>> index bc3a54f0ea..612655386d 100644
>> --- a/xen/drivers/Kconfig
>> +++ b/xen/drivers/Kconfig
>> @@ -12,4 +12,6 @@ source "drivers/pci/Kconfig"
>>     source "drivers/video/Kconfig"
>>   +source "drivers/passthrough/arm/Kconfig"
>> +
>>   endmenu
>> diff --git a/xen/drivers/passthrough/arm/Kconfig
>> b/xen/drivers/passthrough/arm/Kconfig
>> new file mode 100644
>> index 0000000000..cda899f608
>> --- /dev/null
>> +++ b/xen/drivers/passthrough/arm/Kconfig
>> @@ -0,0 +1,8 @@
>> +
>> +config ARM_SMMU_v3
>> +       bool "ARM SMMUv3 Support"
>> +       depends on ARM_64
>> +       help
>> +        Support for implementations of the ARM System MMU architecture
>> +        version 3.
>> +
>> diff --git a/xen/drivers/passthrough/arm/Makefile
>> b/xen/drivers/passthrough/arm/Makefile
>> index f4cd26e15d..e14732b55c 100644
>> --- a/xen/drivers/passthrough/arm/Makefile
>> +++ b/xen/drivers/passthrough/arm/Makefile
>> @@ -1,2 +1,3 @@
>>   obj-y += iommu.o
>>   obj-y += smmu.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 e67ba6c40f..f43485fe6e 100644
>> --- a/xen/drivers/passthrough/arm/smmu-v3.c
>> +++ b/xen/drivers/passthrough/arm/smmu-v3.c
>> @@ -18,28 +18,414 @@
>>    * Author: Will Deacon <will.deacon@xxxxxxx>
>>    *
>>    * This driver is powered by bad coffee and bombay mix.
>> + *
>> + *
>> + * Based on Linux drivers/iommu/arm-smmu-v3.c
>> + * => commit 7aa8619a66aea52b145e04cbab4f8d6a4e5f3f3b
>> + *
>> + * Xen modifications:
>> + * Sameer Goel <sameer.goel@xxxxxxxxxx>
>> + * Copyright (C) 2017, The Linux Foundation, All rights reserved.
>> + *
>> + */
>> +
>> +#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/linux_compat.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 <acpi/acpi_iort.h>
>> +#include <asm/atomic.h>
>> +#include <asm/device.h>
>> +#include <asm/io.h>
>> +#include <asm/platform.h>
>> +
>> +/* 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
>> +
>> +/* Xen: Helpers to get device MMIO and IRQs */
>> +struct resource {
>> +       u64 addr;
>> +       u64 size;
>> +       unsigned int type;
>> +};
>> +
>> +#define resource_size(res) ((res)->size)
>> +
>> +#define platform_device device
>> +
>> +#define IORESOURCE_MEM 0
>> +#define IORESOURCE_IRQ 1
>> +
>> +static struct resource *platform_get_resource(struct platform_device
>> *pdev,
>> +                                             unsigned int type,
>> +                                             unsigned int num)
>> +{
>> +       /*
>> +        * The resource is only used between 2 calls of
>> platform_get_resource.
>> +        * It's quite ugly but it's avoid to add too much code in the part
>> +        * imported from Linux
>> +        */
>> +       static struct resource res;
>> +       struct acpi_iort_node *iort_node;
>> +       struct acpi_iort_smmu_v3 *node_smmu_data;
>> +       int ret = 0;
>> +
>> +       res.type = type;
>> +
>> +       switch (type) {
>> +       case IORESOURCE_MEM:
>> +               if (pdev->type == DEV_ACPI) {
>> +                       ret = 1;
>> +                       iort_node = pdev->acpi_node;
>> +                       node_smmu_data =
>> +                               (struct acpi_iort_smmu_v3
>> *)iort_node->node_data;
>> +
>> +                       if (node_smmu_data != NULL) {
>> +                               res.addr = node_smmu_data->base_address;
>> +                               res.size = SZ_128K;
>> +                               ret = 0;
>> +                       }
>> +               } else {
>> +                       ret = dt_device_get_address(dev_to_dt(pdev), num,
>> +                                                   &res.addr, &res.size);
>> +               }
>> +
>> +               return ((ret) ? NULL : &res);
>> +
>> +       case IORESOURCE_IRQ:
>> +               /* ACPI case not implemented as there is no use case for
>> it */
>> +               ret = platform_get_irq(dev_to_dt(pdev), num);
>> +
>> +               if (ret < 0)
>> +                       return NULL;
>> +
>> +               res.addr = ret;
>> +               res.size = 1;
>> +
>> +               return &res;
>> +
>> +       default:
>> +               return NULL;
>> +       }
>> +}
>> +
>> +static int platform_get_irq_byname(struct platform_device *pdev, const
>> char *name)
>> +{
>> +       const struct dt_property *dtprop;
>> +       struct acpi_iort_node *iort_node;
>> +       struct acpi_iort_smmu_v3 *node_smmu_data;
>> +       int ret = 0;
>> +
>> +       if (pdev->type == DEV_ACPI) {
>> +               iort_node = pdev->acpi_node;
>> +               node_smmu_data = (struct acpi_iort_smmu_v3
>> *)iort_node->node_data;
>> +
>> +               if (node_smmu_data != NULL) {
>> +                       if (!strcmp(name, "eventq"))
>> +                               ret = node_smmu_data->event_gsiv;
>> +                       else if (!strcmp(name, "priq"))
>> +                               ret = node_smmu_data->pri_gsiv;
>> +                       else if (!strcmp(name, "cmdq-sync"))
>> +                               ret = node_smmu_data->sync_gsiv;
>> +                       else if (!strcmp(name, "gerror"))
>> +                               ret = node_smmu_data->gerr_gsiv;
>> +                       else
>> +                               ret = -EINVAL;
>> +               }
>> +       } else {
>> +               dtprop = dt_find_property(dev_to_dt(pdev),
>> "interrupt-names", NULL);
>> +               if (!dtprop)
>> +                       return -EINVAL;
>> +
>> +               if (!dtprop->value)
>> +                       return -ENODATA;
>> +       }
>> +
>> +       return ret;
>> +}
>> +
>> +/*
>> + * Xen: Helpers for DMA allocation. Just the function name is reused for
>> + * porting code these allocation are not managed allocations
>>    */
>>   -#include <linux/acpi.h>
>> -#include <linux/acpi_iort.h>
>> -#include <linux/delay.h>
>> -#include <linux/dma-iommu.h>
>> -#include <linux/err.h>
>> -#include <linux/interrupt.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/platform_device.h>
>> -
>> -#include <linux/amba/bus.h>
>> -
>> -#include "io-pgtable.h"
>> +static void *dmam_alloc_coherent(struct device *dev, size_t size,
>> +                          dma_addr_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)) {
>> +               dev_warn(dev, "Fixing alignment for the DMA buffer\n");
>> +               alignment = sizeof(void *);
>> +       }
>> +
>> +       vaddr = _xzalloc(size, alignment);
>> +       if (!vaddr) {
>> +               dev_err(dev, "DMA allocation failed\n");
>> +               return NULL;
>> +       }
>> +
>> +       *dma_handle = virt_to_maddr(vaddr);
>> +
>> +       return vaddr;
>> +}
>> +
>> +
>> +static void dmam_free_coherent(struct device *dev, size_t size, void
>> *vaddr,
>> +                       dma_addr_t dma_handle)
>> +{
>> +       xfree(vaddr);
>> +}
>> +
>> +/* Xen: Stub out DMA domain related functions */
>> +#define iommu_get_dma_cookie(dom) 0
>> +#define iommu_put_dma_cookie(dom)
>> +
>> +/* Xen: Stub out module param related function */
>> +#define module_param_named(a, b, c, d)
>> +#define MODULE_PARM_DESC(a, b)
>> +
>> +#define dma_set_mask_and_coherent(d, b) 0
>> +
>> +#define of_dma_is_coherent(n) 0
>> +
>> +#define MODULE_DEVICE_TABLE(type, name)
>> +
>> +static void __iomem *devm_ioremap_resource(struct device *dev,
>> +                                          struct resource *res)
>> +{
>> +       void __iomem *ptr;
>> +
>> +       if (!res || res->type != IORESOURCE_MEM) {
>> +               dev_err(dev, "Invalid resource\n");
>> +               return ERR_PTR(-EINVAL);
>> +       }
>> +
>> +       ptr = ioremap_nocache(res->addr, res->size);
>> +       if (!ptr) {
>> +               dev_err(dev,
>> +                       "ioremap failed (addr 0x%"PRIx64" size
>> 0x%"PRIx64")\n",
>> +                       res->addr, res->size);
>> +               return ERR_PTR(-ENOMEM);
>> +       }
>> +
>> +       return ptr;
>> +}
>> +
>> +/* Xen: Compatibility define for iommu_domain_geometry.*/
>> +struct iommu_domain_geometry {
>> +       dma_addr_t aperture_start; /* First address that can be mapped  
>> */
>> +       dma_addr_t aperture_end;   /* Last address that can be mapped    
>> */
>> +       bool force_aperture;       /* DMA only allowed in mappable range?
>> */
>> +};
>> +
>> +
>> +/* Xen: Type definitions for iommu_domain */
>> +#define IOMMU_DOMAIN_UNMANAGED 0
>> +#define IOMMU_DOMAIN_DMA 1
>> +#define IOMMU_DOMAIN_IDENTITY 2
>> +
>> +/* Xen: Dummy iommu_domain */
>> +struct iommu_domain {
>> +       /* Runtime SMMU configuration for this iommu_domain */
>> +       struct arm_smmu_domain          *priv;
>> +       unsigned int type;
>> +
>> +       /* Dummy compatibility defines */
>> +       unsigned long pgsize_bitmap;
>> +       struct iommu_domain_geometry geometry;
>> +
>> +       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;
>> +};
>> +
>> +/* Xen: 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;
>> +};
>> +
>> +/*
>> + * Xen: 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;
>> +};
>> +
>> +/*
>> + * Xen: io_pgtable compatibility defines.
>> + * Most of these are to port in the S1 translation code as is.
>> + */
>> +struct io_pgtable_ops {
>> +};
>> +
>> +struct iommu_gather_ops {
>> +       void (*tlb_flush_all)(void *cookie);
>> +       void (*tlb_add_flush)(unsigned long iova, size_t size, size_t
>> granule,
>> +                             bool leaf, void *cookie);
>> +       void (*tlb_sync)(void *cookie);
>> +};
>> +
>> +struct io_pgtable_cfg {
>> +       /*
>> +        * IO_PGTABLE_QUIRK_ARM_NS: (ARM formats) Set NS and NSTABLE bits
>> in
>> +        *      stage 1 PTEs, for hardware which insists on validating
>> them
>> +        *      even in non-secure state where they should normally be
>> ignored.
>> +        *
>> +        * IO_PGTABLE_QUIRK_NO_PERMS: Ignore the IOMMU_READ, IOMMU_WRITE
>> and
>> +        *      IOMMU_NOEXEC flags and map everything with full access,
>> for
>> +        *      hardware which does not implement the permissions of a
>> given
>> +        *      format, and/or requires some format-specific default
>> value.
>> +        *
>> +        * IO_PGTABLE_QUIRK_TLBI_ON_MAP: If the format forbids caching
>> invalid
>> +        *      (unmapped) entries but the hardware might do so anyway,
>> perform
>> +        *      TLB maintenance when mapping as well as when unmapping.
>> +        *
>> +        * IO_PGTABLE_QUIRK_ARM_MTK_4GB: (ARM v7s format) Set bit 9 in all
>> +        *      PTEs, for Mediatek IOMMUs which treat it as a 33rd address
>> bit
>> +        *      when the SoC is in "4GB mode" and they can only access the
>> high
>> +        *      remap of DRAM (0x1_00000000 to 0x1_ffffffff).
>> +        *
>> +        * IO_PGTABLE_QUIRK_NO_DMA: Guarantees that the tables will only
>> ever
>> +        *      be accessed by a fully cache-coherent IOMMU or CPU (e.g.
>> for a
>> +        *      software-emulated IOMMU), such that pagetable updates need
>> not
>> +        *      be treated as explicit DMA data.
>> +        */
>> +       #define IO_PGTABLE_QUIRK_ARM_NS         BIT(0)
>> +       #define IO_PGTABLE_QUIRK_NO_PERMS       BIT(1)
>> +       #define IO_PGTABLE_QUIRK_TLBI_ON_MAP    BIT(2)
>> +       #define IO_PGTABLE_QUIRK_ARM_MTK_4GB    BIT(3)
>> +       #define IO_PGTABLE_QUIRK_NO_DMA         BIT(4)
>> +       unsigned long                   quirks;
>> +       unsigned long                   pgsize_bitmap;
>> +       unsigned int                    ias;
>> +       unsigned int                    oas;
>> +       const struct iommu_gather_ops   *tlb;
>> +       struct device                   *iommu_dev;
>> +
>> +       /* Low-level data specific to the table format */
>> +       union {
>> +               struct {
>> +                       u64     ttbr[2];
>> +                       u64     tcr;
>> +                       u64     mair[2];
>> +               } arm_lpae_s1_cfg;
>> +
>> +               struct {
>> +                       u64     vttbr;
>> +                       u64     vtcr;
>> +               } arm_lpae_s2_cfg;
>> +
>> +               struct {
>> +                       u32     ttbr[2];
>> +                       u32     tcr;
>> +                       u32     nmrr;
>> +                       u32     prrr;
>> +               } arm_v7s_cfg;
>> +       };
>> +};
>> +
>> +enum io_pgtable_fmt {
>> +       ARM_32_LPAE_S1,
>> +       ARM_32_LPAE_S2,
>> +       ARM_64_LPAE_S1,
>> +       ARM_64_LPAE_S2,
>> +       ARM_V7S,
>> +       IO_PGTABLE_NUM_FMTS,
>> +};
>> +
>> +/*
>> + * Xen: The pgtable_ops are used by the S1 translations, so return the
>> dummy
>> + * address.
>> + */
>> +#define alloc_io_pgtable_ops(f, c, o) ((struct io_pgtable_ops *)0x0)
>
> Is this correct? it will make the caller null check fail
In your earlier patch it was

#define alloc_io_pgtable_ops(f, c, o) ((struct io_pgtable_ops *)0xDEADBEEF)

>
>> +#define free_io_pgtable_ops(o)
>> +
>> +/* Xen: Define wrapper for requesting IRQs */
>> +#define IRQF_ONESHOT 0
>> +
>> +typedef void (*irq_handler_t)(int, void *, struct cpu_user_regs *);
>> +
>> +static inline int devm_request_irq(struct device *dev, unsigned int irq,
>> +                                  irq_handler_t handler, unsigned long
>> irqflags,
>> +                                  const char *devname, void *dev_id)
>> +{
>> +       /*
>> +        * SMMUv3 implementation can support wired interrupt outputs that
>> are
>> +        * edge-triggered. Set the irq type as per the spec.
>> +        */
>> +       irq_set_type(irq, IRQ_TYPE_EDGE_BOTH);
>> +       return request_irq(irq, irqflags, handler, devname, dev_id);
>> +}
>> +
>> +/*
>> + * Xen does not have a concept of threaded irq, but eventually we can use
>> + * tasklets to optimize the following function.
>> + */
>> +int devm_request_threaded_irq(struct device *dev, unsigned int irq,
>> irq_handler_t handler,
>> +                             irq_handler_t thread_fn, unsigned long
>> irqflags,
>> +                             const char *devname, void *dev_id)
>> +{
>> +       return devm_request_irq(dev, irq, thread_fn, irqflags, devname,
>> dev_id);
>> +}
>> +
>> +/* Xen: The mutex is used only during initialization so the typecast is
>> safe */
>> +#define mutex spinlock
>> +#define mutex_init spin_lock_init
>> +#define mutex_lock spin_lock
>> +#define mutex_unlock spin_unlock
>> +
>> +#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 VA_BITS 0 /* Only needed for S1 translations */
>>     /* MMIO registers */
>>   #define ARM_SMMU_IDR0                 0x0
>> @@ -433,6 +819,7 @@ enum pri_resp {
>>         PRI_RESP_SUCC,
>>   };
>>   +#if 0 /* Xen: No MSI support in this iteration */
>>   enum arm_smmu_msi_index {
>>         EVTQ_MSI_INDEX,
>>         GERROR_MSI_INDEX,
>> @@ -457,6 +844,7 @@ static phys_addr_t
>> arm_smmu_msi_cfg[ARM_SMMU_MAX_MSIS][3] = {
>>                 ARM_SMMU_PRIQ_IRQ_CFG2,
>>         },
>>   };
>> +#endif
>>     struct arm_smmu_cmdq_ent {
>>         /* Common fields */
>> @@ -561,6 +949,8 @@ struct arm_smmu_s2_cfg {
>>         u16                             vmid;
>>         u64                             vttbr;
>>         u64                             vtcr;
>> +       /* Xen: Domain associated to this configuration */
>> +       struct domain                   *domain;
>>   };
>>     struct arm_smmu_strtab_ent {
>> @@ -635,9 +1025,21 @@ struct arm_smmu_device {
>>         struct arm_smmu_strtab_cfg      strtab_cfg;
>>         /* IOMMU core code handle */
>> +#if 0 /*Xen: Generic iommu_device ref not needed here */
>>         struct iommu_device             iommu;
>> +#endif
>> +       /* Xen: Need to keep a list of SMMU devices */
>> +       struct list_head                devices;
>>   };
>>   +/* Xen: Keep a list of devices associated with this driver */
>> +static DEFINE_SPINLOCK(arm_smmu_devices_lock);
>> +static LIST_HEAD(arm_smmu_devices);
>> +/* Xen: Helper for finding a device using fwnode */
>> +static
>> +struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle
>> *fwnode);
>> +
>> +
>>   /* SMMU private data for each master */
>>   struct arm_smmu_master_data {
>>         struct arm_smmu_device          *smmu;
>> @@ -1232,7 +1634,7 @@ static void arm_smmu_handle_ppr(struct
>> arm_smmu_device *smmu, u64 *evt)
>>         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",
>> +                "\tsid 0x%08x.0x%05x: [%u%s] %sprivileged %s%s%s access
>> at iova %#" PRIx64 "\n",
>>                  sid, ssid, grpid, last ? "L" : "",
>>                  evt[0] & PRIQ_0_PERM_PRIV ? "" : "un",
>>                  evt[0] & PRIQ_0_PERM_READ ? "R" : "",
>> @@ -1346,6 +1748,8 @@ static irqreturn_t arm_smmu_combined_irq_handler(int
>> irq, void *dev)
>>   {
>>         arm_smmu_gerror_handler(irq, dev);
>>         arm_smmu_cmdq_sync_handler(irq, dev);
>> +       /*Xen: No threaded irq. So call the required function from here */
>> +       arm_smmu_combined_irq_thread(irq, dev);
>>         return IRQ_WAKE_THREAD;
>>   }
>>   @@ -1358,6 +1762,46 @@ static void __arm_smmu_tlb_sync(struct
>> arm_smmu_device *smmu)
>>         arm_smmu_cmdq_issue_cmd(smmu, &cmd);
>>   }
>>   +/*
>> + * Xen: Define the IRQ handlers for xen. The linux functions would be
>> + * modified to use the functions defined in the following code.
>> + */
>> +static void arm_smmu_evtq_thread_xen(int irq, void *dev,
>> +                                      struct cpu_user_regs *regs)
>> +{
>> +       arm_smmu_evtq_thread(irq, dev);
>> +}
>> +
>> +static void arm_smmu_priq_thread_xen(int irq, void *dev,
>> +                                      struct cpu_user_regs *regs)
>> +{
>> +       arm_smmu_priq_thread(irq, dev);
>> +}
>> +
>> +static void arm_smmu_cmdq_sync_handler_xen(int irq, void *dev,
>> +                                      struct cpu_user_regs *regs)
>> +{
>> +       arm_smmu_cmdq_sync_handler(irq, dev);
>> +}
>> +
>> +static void arm_smmu_gerror_handler_xen(int irq, void *dev,
>> +                                      struct cpu_user_regs *regs)
>> +{
>> +       arm_smmu_gerror_handler(irq, dev);
>> +}
>> +
>> +static void arm_smmu_combined_irq_handler_xen(int irq, void *dev,
>> +                                      struct cpu_user_regs *regs)
>> +{
>> +       arm_smmu_combined_irq_handler(irq, dev);
>> +}
>> +
>> +#define arm_smmu_evtq_thread arm_smmu_evtq_thread_xen
>> +#define arm_smmu_priq_thread arm_smmu_priq_thread_xen
>> +#define arm_smmu_cmdq_sync_handler arm_smmu_cmdq_sync_handler_xen
>> +#define arm_smmu_gerror_handler arm_smmu_gerror_handler_xen
>> +#define arm_smmu_combined_irq_handler arm_smmu_combined_irq_handler_xen
>> +
>>   static void arm_smmu_tlb_sync(void *cookie)
>>   {
>>         struct arm_smmu_domain *smmu_domain = cookie;
>> @@ -1415,6 +1859,7 @@ static const struct iommu_gather_ops
>> arm_smmu_gather_ops = {
>>         .tlb_sync       = arm_smmu_tlb_sync,
>>   };
>>   +#if 0 /*Xen: Unused functionality */
>>   /* IOMMU API */
>>   static bool arm_smmu_capable(enum iommu_cap cap)
>>   {
>> @@ -1427,6 +1872,7 @@ static bool arm_smmu_capable(enum iommu_cap cap)
>>                 return false;
>>         }
>>   }
>> +#endif
>>     static struct iommu_domain *arm_smmu_domain_alloc(unsigned type)
>>   {
>> @@ -1546,9 +1992,16 @@ static int arm_smmu_domain_finalise_s2(struct
>> arm_smmu_domain *smmu_domain,
>>         if (vmid < 0)
>>                 return vmid;
>>   -     cfg->vmid       = (u16)vmid;
>> -       cfg->vttbr      = pgtbl_cfg->arm_lpae_s2_cfg.vttbr;
>> -       cfg->vtcr       = pgtbl_cfg->arm_lpae_s2_cfg.vtcr;
>> +       /*
>> +        * Xen: Get the ttbr and vtcr values
>> +        * vttbr: This is a shared value with the domain page table
>> +        * vtcr: The TCR settings are the same as CPU since the page
>> +        * tables are shared
>> +        */
>> +
>> +       cfg->vmid       = vmid;
>> +       cfg->vttbr      = page_to_maddr(cfg->domain->arch.p2m.root);
>> +       cfg->vtcr       = READ_SYSREG32(VTCR_EL2) &
>> STRTAB_STE_2_VTCR_MASK;
>>         return 0;
>>   }
>>   @@ -1604,6 +2057,7 @@ static int arm_smmu_domain_finalise(struct
>> iommu_domain *domain)
>>         if (smmu->features & ARM_SMMU_FEAT_COHERENCY)
>>                 pgtbl_cfg.quirks = IO_PGTABLE_QUIRK_NO_DMA;
>>   +     /* Xen: pgtbl_ops gets an invalid address */
>>         pgtbl_ops = alloc_io_pgtable_ops(fmt, &pgtbl_cfg, smmu_domain);
>>         if (!pgtbl_ops)
>>                 return -ENOMEM;
>
> Will this not fail ?
>
>> @@ -1721,6 +2175,7 @@ out_unlock:
>>         return ret;
>>   }
>>   +#if 0 /* Xen: Unused functionality */
>>   static int arm_smmu_map(struct iommu_domain *domain, unsigned long iova,
>>                         phys_addr_t paddr, size_t size, int prot)
>>   {
>> @@ -1772,6 +2227,7 @@ struct arm_smmu_device
>> *arm_smmu_get_by_fwnode(struct fwnode_handle *fwnode)
>>         put_device(dev);
>>         return dev ? dev_get_drvdata(dev) : NULL;
>>   }
>> +#endif
>>     static bool arm_smmu_sid_in_range(struct arm_smmu_device *smmu, u32
>> sid)
>>   {
>> @@ -1783,7 +2239,14 @@ static bool arm_smmu_sid_in_range(struct
>> arm_smmu_device *smmu, u32 sid)
>>         return sid < limit;
>>   }
>>   +/* Xen: Unused */
>> +#if 0
>>   static struct iommu_ops arm_smmu_ops;
>> +#endif
>> +
>> +/* Xen: Redefine arm_smmu_ops to what fwspec should evaluate */
>> +static const struct iommu_ops arm_smmu_iommu_ops;
>> +#define arm_smmu_ops arm_smmu_iommu_ops
>>     static int arm_smmu_add_device(struct device *dev)
>>   {
>> @@ -1791,8 +2254,11 @@ static int arm_smmu_add_device(struct device *dev)
>>         struct arm_smmu_device *smmu;
>>         struct arm_smmu_master_data *master;
>>         struct iommu_fwspec *fwspec = dev->iommu_fwspec;
>> +#if 0 /*Xen: iommu_group is not needed */
>>         struct iommu_group *group;
>> +#endif
>>   +     /* Xen: fwspec->ops are not needed */
>>         if (!fwspec || fwspec->ops != &arm_smmu_ops)
>>                 return -ENODEV;
>>         /*
>> @@ -1830,6 +2296,11 @@ static int arm_smmu_add_device(struct device *dev)
>>                 }
>>         }
>>   +/*
>> + * Xen: Do not need an iommu group as the stream data is carried by the
>> SMMU
>> + * master device object
>> + */
>> +#if 0
>>         group = iommu_group_get_for_dev(dev);
>>         if (!IS_ERR(group)) {
>>                 iommu_group_put(group);
>> @@ -1837,8 +2308,16 @@ static int arm_smmu_add_device(struct device *dev)
>>         }
>>         return PTR_ERR_OR_ZERO(group);
>> +#endif
>> +       return 0;
>>   }
>>   +/*
>> + * Xen: We can potentially support this function and destroy a device.
>> This
>> + * will be relevant for PCI hotplug. So, will be implemented as needed
>> after
>> + * passthrough support is available.
>> + */
>> +#if 0
>>   static void arm_smmu_remove_device(struct device *dev)
>>   {
>>         struct iommu_fwspec *fwspec = dev->iommu_fwspec;
>> @@ -1974,6 +2453,7 @@ static struct iommu_ops arm_smmu_ops = {
>>         .put_resv_regions       = arm_smmu_put_resv_regions,
>>         .pgsize_bitmap          = -1UL, /* Restricted during device attach
>> */
>>   };
>> +#endif
>>     /* Probing and initialisation functions */
>>   static int arm_smmu_init_one_queue(struct arm_smmu_device *smmu,
>> @@ -2182,6 +2662,7 @@ static int arm_smmu_update_gbpa(struct
>> arm_smmu_device *smmu, u32 set, u32 clr)
>>                                           1, ARM_SMMU_POLL_TIMEOUT_US);
>>   }
>>   +#if 0 /* Xen: There is no MSI support as yet */
>>   static void arm_smmu_free_msis(void *data)
>>   {
>>         struct device *dev = data;
>> @@ -2247,12 +2728,15 @@ static void arm_smmu_setup_msis(struct
>> arm_smmu_device *smmu)
>>         /* Add callback to free MSIs on teardown */
>>         devm_add_action(dev, arm_smmu_free_msis, dev);
>>   }
>> +#endif
>>     static void arm_smmu_setup_unique_irqs(struct arm_smmu_device *smmu)
>>   {
>>         int irq, ret;
>>   +#if 0 /*Xen: Cannot setup msis for now */
>>         arm_smmu_setup_msis(smmu);
>> +#endif
>>         /* Request interrupt lines */
>>         irq = smmu->evtq.q.irq;
>> @@ -2316,9 +2800,13 @@ static int arm_smmu_setup_irqs(struct
>> arm_smmu_device *smmu)
>>                  * Cavium ThunderX2 implementation doesn't not support
>> unique
>>                  * irq lines. Use single irq line for all the SMMUv3
>> interrupts.
>>                  */
>> -               ret = devm_request_threaded_irq(smmu->dev, irq,
>> +               /*
>> +                * Xen: Does not support threaded irqs, so serialise the
>> setup.
>> +                * This is the same for pris and event interrupt lines on
>> other
>> +                * systems
>> +                */
>> +               ret = devm_request_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)
>> @@ -2542,8 +3030,14 @@ static int arm_smmu_device_hw_probe(struct
>> arm_smmu_device *smmu)
>>                 smmu->features |= ARM_SMMU_FEAT_STALLS;
>>         }
>>   +/*
>> + * Xen: Block stage 1 translations. By doing this here we do not need to
>> set the
>> + * domain->stage explicitly.
>> + */
>> +#if 0
>>         if (reg & IDR0_S1P)
>>                 smmu->features |= ARM_SMMU_FEAT_TRANS_S1;
>> +#endif
>>         if (reg & IDR0_S2P)
>>                 smmu->features |= ARM_SMMU_FEAT_TRANS_S2;
>> @@ -2616,10 +3110,12 @@ static int arm_smmu_device_hw_probe(struct
>> arm_smmu_device *smmu)
>>         if (reg & IDR5_GRAN4K)
>>                 smmu->pgsize_bitmap |= SZ_4K | SZ_2M | SZ_1G;
>>   +#if 0 /* Xen: SMMU ops do not have a pgsize_bitmap member for Xen */
>>         if (arm_smmu_ops.pgsize_bitmap == -1UL)
>>                 arm_smmu_ops.pgsize_bitmap = smmu->pgsize_bitmap;
>>         else
>>                 arm_smmu_ops.pgsize_bitmap |= smmu->pgsize_bitmap;
>> +#endif
>>         /* Output address size */
>>         switch (reg & IDR5_OAS_MASK << IDR5_OAS_SHIFT) {
>> @@ -2680,7 +3176,8 @@ static int arm_smmu_device_acpi_probe(struct
>> platform_device *pdev,
>>         struct device *dev = smmu->dev;
>>         struct acpi_iort_node *node;
>>   -     node = *(struct acpi_iort_node **)dev_get_platdata(dev);
>> +       /* Xen: Modification to get iort_node */
>> +       node = (struct acpi_iort_node *)dev->acpi_node;
>>         /* Retrieve SMMUv3 specific data */
>>         iort_smmu = (struct acpi_iort_smmu_v3 *)node->node_data;
>> @@ -2703,7 +3200,7 @@ 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;
>>   @@ -2716,6 +3213,7 @@ static int arm_smmu_device_dt_probe(struct
>> platform_device *pdev,
>>         parse_driver_options(smmu);
>>   +     /* Xen: of_dma_is_coherent is a stub till dt support is introduced
>> */
>>         if (of_dma_is_coherent(dev->of_node))
>>                 smmu->features |= ARM_SMMU_FEAT_COHERENCY;
>>   @@ -2734,9 +3232,11 @@ static int arm_smmu_device_probe(struct
>> platform_device *pdev)
>>   {
>>         int irq, ret;
>>         struct resource *res;
>> +#if 0 /*Xen: Do not need to setup sysfs */
>>         resource_size_t ioaddr;
>> +#endif
>>         struct arm_smmu_device *smmu;
>> -       struct device *dev = &pdev->dev;
>> +       struct device *dev = pdev;/* Xen: dev is ignored */
>>         bool bypass;
>>         smmu = devm_kzalloc(dev, sizeof(*smmu), GFP_KERNEL);
>> @@ -2763,7 +3263,9 @@ static int arm_smmu_device_probe(struct
>> platform_device *pdev)
>>                 dev_err(dev, "MMIO region too small (%pr)\n", res);
>>                 return -EINVAL;
>>         }
>> +#if 0 /*Xen: Do not need to setup sysfs */
>>         ioaddr = res->start;
>> +#endif
>>         smmu->base = devm_ioremap_resource(dev, res);
>>         if (IS_ERR(smmu->base))
>> @@ -2802,13 +3304,18 @@ static int arm_smmu_device_probe(struct
>> platform_device *pdev)
>>                 return ret;
>>         /* Record our private device structure */
>> +       /* Xen: SMMU is not treated a a platform device*/
>> +#if 0
>>         platform_set_drvdata(pdev, smmu);
>> +#endif
>>         /* Reset the device */
>>         ret = arm_smmu_device_reset(smmu, bypass);
>>         if (ret)
>>                 return ret;
>>   +/* Xen: Not creating an IOMMU device list for Xen */
>> +#if 0
>>         /* And we're up. Go go go! */
>>         ret = iommu_device_sysfs_add(&smmu->iommu, dev, NULL,
>>                                      "smmu3.%pa", &ioaddr);
>> @@ -2844,9 +3351,20 @@ static int arm_smmu_device_probe(struct
>> platform_device *pdev)
>>                 if (ret)
>>                         return ret;
>>         }
>> +#endif
>> +       /*
>> +        * Xen: 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;
>>   }
>>   +/* Xen: Unused function */
>> +#if 0
>>   static int arm_smmu_device_remove(struct platform_device *pdev)
>>   {
>>         struct arm_smmu_device *smmu = platform_get_drvdata(pdev);
>> @@ -2860,6 +3378,8 @@ static void arm_smmu_device_shutdown(struct
>> platform_device *pdev)
>>   {
>>         arm_smmu_device_remove(pdev);
>>   }
>> +#endif
>> +
>>     static const struct of_device_id arm_smmu_of_match[] = {
>>         { .compatible = "arm,smmu-v3", },
>> @@ -2867,6 +3387,7 @@ static const struct of_device_id arm_smmu_of_match[]
>> = {
>>   };
>>   MODULE_DEVICE_TABLE(of, arm_smmu_of_match);
>>   +#if 0
>>   static struct platform_driver arm_smmu_driver = {
>>         .driver = {
>>                 .name           = "arm-smmu-v3",
>> @@ -2883,3 +3404,316 @@ IOMMU_OF_DECLARE(arm_smmuv3, "arm,smmu-v3", NULL);
>>   MODULE_DESCRIPTION("IOMMU API for ARM architected SMMUv3
>> implementations");
>>   MODULE_AUTHOR("Will Deacon <will.deacon@xxxxxxx>");
>>   MODULE_LICENSE("GPL v2");
>> +#endif
>> +
>> +/***** Start of Xen specific code *****/
>> +
>> +static int __must_check arm_smmu_iotlb_flush_all(struct domain *d)
>> +{
>> +       struct arm_smmu_xen_domain *smmu_domain = dom_iommu(d)->arch.priv;
>> +       struct iommu_domain *cfg;
>> +
>> +       spin_lock(&smmu_domain->lock);
>> +       list_for_each_entry(cfg, &smmu_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(cfg->priv->smmu)))
>> +                       continue;
>> +               arm_smmu_tlb_inv_context(cfg->priv);
>> +       }
>> +       spin_unlock(&smmu_domain->lock);
>> +       return 0;
>> +}
>> +
>> +static int __must_check arm_smmu_iotlb_flush(struct domain *d,
>> +                                            unsigned long gfn,
>> +                                            unsigned int page_count)
>> +{
>> +       return arm_smmu_iotlb_flush_all(d);
>> +}
>> +
>> +static struct iommu_domain *arm_smmu_get_domain(struct domain *d,
>> +                                               struct device *dev)
>> +{
>> +       struct iommu_domain *domain;
>> +       struct arm_smmu_xen_domain *xen_domain;
>> +       struct arm_smmu_device *smmu;
>> +       struct arm_smmu_domain *smmu_domain;
>> +
>> +       xen_domain = dom_iommu(d)->arch.priv;
>> +
>> +       smmu = arm_smmu_get_by_fwnode(dev->iommu_fwspec->iommu_fwnode);
>> +       if (!smmu)
>> +               return NULL;
>> +
>> +       /*
>> +        * Loop through the &xen_domain->contexts to locate a context
>> +        * assigned to this SMMU
>> +        */
>> +       list_for_each_entry(domain, &xen_domain->contexts, list) {
>> +               smmu_domain = to_smmu_domain(domain);
>> +               if (smmu_domain->smmu == smmu)
>> +                       return domain;
>> +       }
>> +
>> +       return NULL;
>> +}
>> +
>> +static void arm_smmu_destroy_iommu_domain(struct iommu_domain *domain)
>> +{
>> +       list_del(&domain->list);
>> +       arm_smmu_domain_free(domain);
>> +}
>> +
>> +static int arm_smmu_assign_dev(struct domain *d, u8 devfn,
>> +                              struct device *dev, u32 flag)
>> +{
>> +       int ret = 0;
>> +       struct iommu_domain *domain;
>> +       struct arm_smmu_xen_domain *xen_domain;
>> +       struct arm_smmu_domain *arm_smmu;
>> +
>> +       xen_domain = dom_iommu(d)->arch.priv;
>> +
>> +       if (!dev->archdata.iommu) {
>> +               dev->archdata.iommu = xzalloc(struct arm_smmu_xen_device);
>> +               if (!dev->archdata.iommu)
>> +                       return -ENOMEM;
>> +       }
>> +
>> +       ret = arm_smmu_add_device(dev);
>> +       if (ret)
>> +               return ret;
>> +
>> +       spin_lock(&xen_domain->lock);
>> +
>> +       /*
>> +        * Check to see if an iommu_domain already exists for this xen
>> domain
>> +        * under the same SMMU
>> +        */
>> +       domain = arm_smmu_get_domain(d, dev);
>> +       if (!domain) {
>> +
>> +               domain = arm_smmu_domain_alloc(IOMMU_DOMAIN_DMA);
>> +               if (!domain) {
>> +                       ret = -ENOMEM;
>> +                       goto out;
>> +               }
>> +
>> +               arm_smmu = to_smmu_domain(domain);
>> +               arm_smmu->s2_cfg.domain = d;
>> +
>> +               /* Chain the new context to the domain */
>> +               list_add(&domain->list, &xen_domain->contexts);
>> +
>> +       }
>> +
>> +       ret = arm_smmu_attach_dev(domain, dev);
>> +       if (ret) {
>> +               if (domain->ref.counter == 0)
>> +                       arm_smmu_destroy_iommu_domain(domain);
>> +       } else {
>> +               atomic_inc(&domain->ref);
>> +       }
>> +
>> +out:
>> +       spin_unlock(&xen_domain->lock);
>> +       return ret;
>> +}
>> +
>> +static int arm_smmu_deassign_dev(struct domain *d, struct device *dev)
>> +{
>> +       struct iommu_domain *domain = arm_smmu_get_domain(d, dev);
>> +       struct arm_smmu_xen_domain *xen_domain;
>> +       struct arm_smmu_domain *arm_smmu = to_smmu_domain(domain);
>> +
>> +       xen_domain = dom_iommu(d)->arch.priv;
>> +
>> +       if (!arm_smmu || arm_smmu->s2_cfg.domain != d) {
>> +               dev_err(dev, " not attached to domain %d\n",
>> d->domain_id);
>> +               return -ESRCH;
>> +       }
>> +
>> +       spin_lock(&xen_domain->lock);
>> +
>> +       arm_smmu_detach_dev(dev);
>> +       atomic_dec(&domain->ref);
>> +
>> +       if (domain->ref.counter == 0)
>> +               arm_smmu_destroy_iommu_domain(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_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_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 __must_check arm_smmu_map_page(struct domain *d, unsigned long
>> gfn,
>> +                       unsigned long mfn, unsigned int flags)
>> +{
>> +       p2m_type_t t;
>> +
>> +       /*
>> +        * Grant mappings can be used for DMA requests. The dev_bus_addr
>> +        * returned by the hypercall is the MFN (not the IPA). For device
>> +        * protected by an IOMMU, Xen needs to add a 1:1 mapping in the
>> domain
>> +        * p2m to allow DMA request to work.
>> +        * This is only valid when the domain is directed mapped. Hence
>> this
>> +        * function should only be used by gnttab code with gfn == mfn.
>> +        */
>> +       BUG_ON(!is_domain_direct_mapped(d));
>> +       BUG_ON(mfn != gfn);
>> +
>> +       /* We only support readable and writable flags */
>> +       if (!(flags & (IOMMUF_readable | IOMMUF_writable)))
>> +               return -EINVAL;
>> +
>> +       t = (flags & IOMMUF_writable) ? p2m_iommu_map_rw :
>> p2m_iommu_map_ro;
>> +
>> +       /*
>> +        * The function guest_physmap_add_entry replaces the current
>> mapping
>> +        * if there is already one...
>> +        */
>> +       return guest_physmap_add_entry(d, _gfn(gfn), _mfn(mfn), 0, t);
>> +}
>> +
>> +static int __must_check arm_smmu_unmap_page(struct domain *d, unsigned
>> long gfn)
>> +{
>> +       /*
>> +        * This function should only be used by gnttab code when the
>> domain
>> +        * is direct mapped
>> +        */
>> +       if (!is_domain_direct_mapped(d))
>> +               return -EINVAL;
>> +
>> +       return guest_physmap_remove_page(d, _gfn(gfn), _mfn(gfn), 0);
>> +}
>> +
>> +static const struct iommu_ops arm_smmu_iommu_ops = {
>> +       .init = arm_smmu_iommu_domain_init,
>> +       .hwdom_init = arm_smmu_iommu_hwdom_init,
>> +       .teardown = arm_smmu_iommu_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_smmu_map_page,
>> +       .unmap_page = arm_smmu_unmap_page,
>> +};
>> +
>> +static
>> +struct arm_smmu_device *arm_smmu_get_by_fwnode(struct fwnode_handle
>> *fwnode)
>> +{
>> +       struct arm_smmu_device *smmu = NULL;
>> +
>> +       spin_lock(&arm_smmu_devices_lock);
>> +       list_for_each_entry(smmu, &arm_smmu_devices, devices) {
>> +               if (smmu->dev->fwnode == fwnode)
>> +                       break;
>> +       }
>> +       spin_unlock(&arm_smmu_devices_lock);
>> +
>> +       return smmu;
>> +}
>> +
>> +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
>> +
>> +#ifdef CONFIG_ACPI
>> +/* Set up the IOMMU */
>> +static int __init arm_smmu_acpi_init(const void *data)
>> +{
>> +       int rc;
>> +
>> +       rc = arm_smmu_device_probe((struct device *)data);
>> +       if (rc)
>> +               return rc;
>> +
>> +       iommu_set_ops(&arm_smmu_iommu_ops);
>> +       return 0;
>> +}
>> +
>> +ACPI_DEVICE_START(asmmuv3, "ARM SMMU V3", DEVICE_IOMMU)
>> +       .class_type = ACPI_IORT_NODE_SMMU_V3,
>> +       .init = arm_smmu_acpi_init,
>> +ACPI_DEVICE_END
>> +
>> +#endif
>> diff --git a/xen/include/xen/linux_compat.h
>> b/xen/include/xen/linux_compat.h
>> new file mode 100644
>> index 0000000000..8037be0a3e
>> --- /dev/null
>> +++ b/xen/include/xen/linux_compat.h
>> @@ -0,0 +1,84 @@
>>
>> +/******************************************************************************
>> + * include/xen/linux_compat.h
>> + *
>> + * Compatibility defines for porting code from Linux to Xen
>> + *
>> + * Copyright (c) 2017 Linaro Limited
>> + *
>> + * This program is free software; you can redistribute it and/or modify
>> + * it under the terms of the GNU General Public License as published by
>> + * the Free Software Foundation; either version 2 of the License, or
>> + * (at your option) any later version.
>> + *
>> + * 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/>.
>> + */
>> +
>> +#ifndef __XEN_LINUX_COMPAT_H__
>> +#define __XEN_LINUX_COMPAT_H__
>> +
>> +#include <asm/types.h>
>> +
>> +typedef paddr_t phys_addr_t;
>> +typedef paddr_t dma_addr_t;
>> +
>> +typedef unsigned int gfp_t;
>> +#define GFP_KERNEL 0
>> +#define __GFP_ZERO 0x01U
>> +
>> +/* Helpers for IRQ functions */
>> +#define free_irq release_irq
>> +
>> +enum irqreturn {
>> +    IRQ_NONE,
>> +    IRQ_HANDLED,
>> +    IRQ_WAKE_THREAD,
>> +};
>> +
>> +typedef enum irqreturn irqreturn_t;
>> +
>> +/* Device logger functions */
>> +#define dev_dbg(dev, fmt, ...) printk(XENLOG_DEBUG fmt, ## __VA_ARGS__)
>> +#define dev_notice(dev, fmt, ...) printk(XENLOG_INFO fmt, ## __VA_ARGS__)
>> +#define dev_warn(dev, fmt, ...) printk(XENLOG_WARNING fmt, ##
>> __VA_ARGS__)
>> +#define dev_err(dev, fmt, ...) printk(XENLOG_ERR fmt, ## __VA_ARGS__)
>> +#define dev_info(dev, fmt, ...) printk(XENLOG_INFO fmt, ## __VA_ARGS__)
>> +
>> +#define dev_err_ratelimited(dev, fmt, ...)                  \
>> +     printk(XENLOG_ERR fmt, ## __VA_ARGS__)
>> +
>> +#define dev_name(dev) dt_node_full_name(dev_to_dt(dev))
>> +
>> +/* Alias to Xen allocation helpers */
>> +#define kfree xfree
>> +#define kmalloc(size, flags) ({\
>> +       void *__ret_alloc = NULL; \
>> +       if (flags & __GFP_ZERO) \
>> +               __ret_alloc = _xzalloc(size, sizeof(void *)); \
>> +       else \
>> +               __ret_alloc = _xmalloc(size, sizeof(void *)); \
>> +       __ret_alloc; \
>> +})
>> +#define kzalloc(size, flags)        _xzalloc(size, sizeof(void *))
>> +#define devm_kzalloc(dev, size, flags)  _xzalloc(size, sizeof(void *))
>> +#define kmalloc_array(size, n, flags) ({\
>> +       void *__ret_alloc = NULL; \
>> +       if (flags & __GFP_ZERO) \
>> +               __ret_alloc = _xzalloc_array(size, sizeof(void *), n); \
>> +       else \
>> +               __ret_alloc = _xmalloc_array(size, sizeof(void *), n); \
>> +       __ret_alloc; \
>> +})
>> +
>> +/* 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))
>> +
>> +#endif /* __XEN_LINUX_COMPAT_H__ */
>
>
>
> _______________________________________________
> Xen-devel mailing list
> Xen-devel@xxxxxxxxxxxxxxxxxxxx
> https://lists.xenproject.org/mailman/listinfo/xen-devel
_______________________________________________
Xen-devel mailing list
Xen-devel@xxxxxxxxxxxxxxxxxxxx
https://lists.xenproject.org/mailman/listinfo/xen-devel

 


Rackspace

Lists.xenproject.org is hosted with RackSpace, monitoring our
servers 24x7x365 and backed by RackSpace's Fanatical Support®.