xref: /openbmc/linux/drivers/iommu/dma-iommu.c (revision a9f4d93d)

// SPDX-License-Identifier: GPL-2.0
/*
 * A fairly generic DMA-API to IOMMU-API glue layer.
 *
 * Copyright (C) 2014-2015 ARM Ltd.
 *
 * based in part on arch/arm/mm/dma-mapping.c:
 * Copyright (C) 2000-2004 Russell King
 */

#include <linux/acpi_iort.h>
#include <linux/device.h>
#include <linux/dma-contiguous.h>
#include <linux/dma-iommu.h>
#include <linux/dma-noncoherent.h>
#include <linux/gfp.h>
#include <linux/huge_mm.h>
#include <linux/iommu.h>
#include <linux/iova.h>
#include <linux/irq.h>
#include <linux/mm.h>
#include <linux/pci.h>
#include <linux/scatterlist.h>
#include <linux/vmalloc.h>

struct iommu_dma_msi_page {
	struct list_head	list;
	dma_addr_t		iova;
	phys_addr_t		phys;
};

enum iommu_dma_cookie_type {
	IOMMU_DMA_IOVA_COOKIE,
	IOMMU_DMA_MSI_COOKIE,
};

struct iommu_dma_cookie {
	enum iommu_dma_cookie_type	type;
	union {
		/* Full allocator for IOMMU_DMA_IOVA_COOKIE */
		struct iova_domain	iovad;
		/* Trivial linear page allocator for IOMMU_DMA_MSI_COOKIE */
		dma_addr_t		msi_iova;
	};
	struct list_head		msi_page_list;
	spinlock_t			msi_lock;

	/* Domain for flush queue callback; NULL if flush queue not in use */
	struct iommu_domain		*fq_domain;
};

static inline size_t cookie_msi_granule(struct iommu_dma_cookie *cookie)
{
	if (cookie->type == IOMMU_DMA_IOVA_COOKIE)
		return cookie->iovad.granule;
	return PAGE_SIZE;
}

static struct iommu_dma_cookie *cookie_alloc(enum iommu_dma_cookie_type type)
{
	struct iommu_dma_cookie *cookie;

	cookie = kzalloc(sizeof(*cookie), GFP_KERNEL);
	if (cookie) {
		spin_lock_init(&cookie->msi_lock);
		INIT_LIST_HEAD(&cookie->msi_page_list);
		cookie->type = type;
	}
	return cookie;
}

/**
 * iommu_get_dma_cookie - Acquire DMA-API resources for a domain
 * @domain: IOMMU domain to prepare for DMA-API usage
 *
 * IOMMU drivers should normally call this from their domain_alloc
 * callback when domain->type == IOMMU_DOMAIN_DMA.
 */
int iommu_get_dma_cookie(struct iommu_domain *domain)
{
	if (domain->iova_cookie)
		return -EEXIST;

	domain->iova_cookie = cookie_alloc(IOMMU_DMA_IOVA_COOKIE);
	if (!domain->iova_cookie)
		return -ENOMEM;

	return 0;
}
EXPORT_SYMBOL(iommu_get_dma_cookie);

/**
 * iommu_get_msi_cookie - Acquire just MSI remapping resources
 * @domain: IOMMU domain to prepare
 * @base: Start address of IOVA region for MSI mappings
 *
 * Users who manage their own IOVA allocation and do not want DMA API support,
 * but would still like to take advantage of automatic MSI remapping, can use
 * this to initialise their own domain appropriately. Users should reserve a
 * contiguous IOVA region, starting at @base, large enough to accommodate the
 * number of PAGE_SIZE mappings necessary to cover every MSI doorbell address
 * used by the devices attached to @domain.
 */
int iommu_get_msi_cookie(struct iommu_domain *domain, dma_addr_t base)
{
	struct iommu_dma_cookie *cookie;

	if (domain->type != IOMMU_DOMAIN_UNMANAGED)
		return -EINVAL;

	if (domain->iova_cookie)
		return -EEXIST;

	cookie = cookie_alloc(IOMMU_DMA_MSI_COOKIE);
	if (!cookie)
		return -ENOMEM;

	cookie->msi_iova = base;
	domain->iova_cookie = cookie;
	return 0;
}
EXPORT_SYMBOL(iommu_get_msi_cookie);

/**
 * iommu_put_dma_cookie - Release a domain's DMA mapping resources
 * @domain: IOMMU domain previously prepared by iommu_get_dma_cookie() or
 *          iommu_get_msi_cookie()
 *
 * IOMMU drivers should normally call this from their domain_free callback.
 */
void iommu_put_dma_cookie(struct iommu_domain *domain)
{
	struct iommu_dma_cookie *cookie = domain->iova_cookie;
	struct iommu_dma_msi_page *msi, *tmp;

	if (!cookie)
		return;

	if (cookie->type == IOMMU_DMA_IOVA_COOKIE && cookie->iovad.granule)
		put_iova_domain(&cookie->iovad);

	list_for_each_entry_safe(msi, tmp, &cookie->msi_page_list, list) {
		list_del(&msi->list);
		kfree(msi);
	}
	kfree(cookie);
	domain->iova_cookie = NULL;
}
EXPORT_SYMBOL(iommu_put_dma_cookie);

/**
 * iommu_dma_get_resv_regions - Reserved region driver helper
 * @dev: Device from iommu_get_resv_regions()
 * @list: Reserved region list from iommu_get_resv_regions()
 *
 * IOMMU drivers can use this to implement their .get_resv_regions callback
 * for general non-IOMMU-specific reservations. Currently, this covers GICv3
 * ITS region reservation on ACPI based ARM platforms that may require HW MSI
 * reservation.
 */
void iommu_dma_get_resv_regions(struct device *dev, struct list_head *list)
{

	if (!is_of_node(dev_iommu_fwspec_get(dev)->iommu_fwnode))
		iort_iommu_msi_get_resv_regions(dev, list);

}
EXPORT_SYMBOL(iommu_dma_get_resv_regions);

static int cookie_init_hw_msi_region(struct iommu_dma_cookie *cookie,
		phys_addr_t start, phys_addr_t end)
{
	struct iova_domain *iovad = &cookie->iovad;
	struct iommu_dma_msi_page *msi_page;
	int i, num_pages;

	start -= iova_offset(iovad, start);
	num_pages = iova_align(iovad, end - start) >> iova_shift(iovad);

	msi_page = kcalloc(num_pages, sizeof(*msi_page), GFP_KERNEL);
	if (!msi_page)
		return -ENOMEM;

	for (i = 0; i < num_pages; i++) {
		msi_page[i].phys = start;
		msi_page[i].iova = start;
		INIT_LIST_HEAD(&msi_page[i].list);
		list_add(&msi_page[i].list, &cookie->msi_page_list);
		start += iovad->granule;
	}

	return 0;
}

static int iova_reserve_pci_windows(struct pci_dev *dev,
		struct iova_domain *iovad)
{
	struct pci_host_bridge *bridge = pci_find_host_bridge(dev->bus);
	struct resource_entry *window;
	unsigned long lo, hi;
	phys_addr_t start = 0, end;

	resource_list_for_each_entry(window, &bridge->windows) {
		if (resource_type(window->res) != IORESOURCE_MEM)
			continue;

		lo = iova_pfn(iovad, window->res->start - window->offset);
		hi = iova_pfn(iovad, window->res->end - window->offset);
		reserve_iova(iovad, lo, hi);
	}

	/* Get reserved DMA windows from host bridge */
	resource_list_for_each_entry(window, &bridge->dma_ranges) {
		end = window->res->start - window->offset;
resv_iova:
		if (end > start) {
			lo = iova_pfn(iovad, start);
			hi = iova_pfn(iovad, end);
			reserve_iova(iovad, lo, hi);
		} else {
			/* dma_ranges list should be sorted */
			dev_err(&dev->dev, "Failed to reserve IOVA\n");
			return -EINVAL;
		}

		start = window->res->end - window->offset + 1;
		/* If window is last entry */
		if (window->node.next == &bridge->dma_ranges &&
		    end != ~(dma_addr_t)0) {
			end = ~(dma_addr_t)0;
			goto resv_iova;
		}
	}

	return 0;
}

static int iova_reserve_iommu_regions(struct device *dev,
		struct iommu_domain *domain)
{
	struct iommu_dma_cookie *cookie = domain->iova_cookie;
	struct iova_domain *iovad = &cookie->iovad;
	struct iommu_resv_region *region;
	LIST_HEAD(resv_regions);
	int ret = 0;

	if (dev_is_pci(dev)) {
		ret = iova_reserve_pci_windows(to_pci_dev(dev), iovad);
		if (ret)
			return ret;
	}

	iommu_get_resv_regions(dev, &resv_regions);
	list_for_each_entry(region, &resv_regions, list) {
		unsigned long lo, hi;

		/* We ARE the software that manages these! */
		if (region->type == IOMMU_RESV_SW_MSI)
			continue;

		lo = iova_pfn(iovad, region->start);
		hi = iova_pfn(iovad, region->start + region->length - 1);
		reserve_iova(iovad, lo, hi);

		if (region->type == IOMMU_RESV_MSI)
			ret = cookie_init_hw_msi_region(cookie, region->start,
					region->start + region->length);
		if (ret)
			break;
	}
	iommu_put_resv_regions(dev, &resv_regions);

	return ret;
}

static void iommu_dma_flush_iotlb_all(struct iova_domain *iovad)
{
	struct iommu_dma_cookie *cookie;
	struct iommu_domain *domain;

	cookie = container_of(iovad, struct iommu_dma_cookie, iovad);
	domain = cookie->fq_domain;
	/*
	 * The IOMMU driver supporting DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE
	 * implies that ops->flush_iotlb_all must be non-NULL.
	 */
	domain->ops->flush_iotlb_all(domain);
}

/**
 * iommu_dma_init_domain - Initialise a DMA mapping domain
 * @domain: IOMMU domain previously prepared by iommu_get_dma_cookie()
 * @base: IOVA at which the mappable address space starts
 * @size: Size of IOVA space
 * @dev: Device the domain is being initialised for
 *
 * @base and @size should be exact multiples of IOMMU page granularity to
 * avoid rounding surprises. If necessary, we reserve the page at address 0
 * to ensure it is an invalid IOVA. It is safe to reinitialise a domain, but
 * any change which could make prior IOVAs invalid will fail.
 */
static int iommu_dma_init_domain(struct iommu_domain *domain, dma_addr_t base,
		u64 size, struct device *dev)
{
	struct iommu_dma_cookie *cookie = domain->iova_cookie;
	struct iova_domain *iovad = &cookie->iovad;
	unsigned long order, base_pfn;
	int attr;

	if (!cookie || cookie->type != IOMMU_DMA_IOVA_COOKIE)
		return -EINVAL;

	/* Use the smallest supported page size for IOVA granularity */
	order = __ffs(domain->pgsize_bitmap);
	base_pfn = max_t(unsigned long, 1, base >> order);

	/* Check the domain allows at least some access to the device... */
	if (domain->geometry.force_aperture) {
		if (base > domain->geometry.aperture_end ||
		    base + size <= domain->geometry.aperture_start) {
			pr_warn("specified DMA range outside IOMMU capability\n");
			return -EFAULT;
		}
		/* ...then finally give it a kicking to make sure it fits */
		base_pfn = max_t(unsigned long, base_pfn,
				domain->geometry.aperture_start >> order);
	}

	/* start_pfn is always nonzero for an already-initialised domain */
	if (iovad->start_pfn) {
		if (1UL << order != iovad->granule ||
		    base_pfn != iovad->start_pfn) {
			pr_warn("Incompatible range for DMA domain\n");
			return -EFAULT;
		}

		return 0;
	}

	init_iova_domain(iovad, 1UL << order, base_pfn);

	if (!cookie->fq_domain && !iommu_domain_get_attr(domain,
			DOMAIN_ATTR_DMA_USE_FLUSH_QUEUE, &attr) && attr) {
		cookie->fq_domain = domain;
		init_iova_flush_queue(iovad, iommu_dma_flush_iotlb_all, NULL);
	}

	if (!dev)
		return 0;

	return iova_reserve_iommu_regions(dev, domain);
}

/**
 * dma_info_to_prot - Translate DMA API directions and attributes to IOMMU API
 *                    page flags.
 * @dir: Direction of DMA transfer
 * @coherent: Is the DMA master cache-coherent?
 * @attrs: DMA attributes for the mapping
 *
 * Return: corresponding IOMMU API page protection flags
 */
static int dma_info_to_prot(enum dma_data_direction dir, bool coherent,
		     unsigned long attrs)
{
	int prot = coherent ? IOMMU_CACHE : 0;

	if (attrs & DMA_ATTR_PRIVILEGED)
		prot |= IOMMU_PRIV;

	switch (dir) {
	case DMA_BIDIRECTIONAL:
		return prot | IOMMU_READ | IOMMU_WRITE;
	case DMA_TO_DEVICE:
		return prot | IOMMU_READ;
	case DMA_FROM_DEVICE:
		return prot | IOMMU_WRITE;
	default:
		return 0;
	}
}

static dma_addr_t iommu_dma_alloc_iova(struct iommu_domain *domain,
		size_t size, dma_addr_t dma_limit, struct device *dev)
{
	struct iommu_dma_cookie *cookie = domain->iova_cookie;
	struct iova_domain *iovad = &cookie->iovad;
	unsigned long shift, iova_len, iova = 0;

	if (cookie->type == IOMMU_DMA_MSI_COOKIE) {
		cookie->msi_iova += size;
		return cookie->msi_iova - size;
	}

	shift = iova_shift(iovad);
	iova_len = size >> shift;
	/*
	 * Freeing non-power-of-two-sized allocations back into the IOVA caches
	 * will come back to bite us badly, so we have to waste a bit of space
	 * rounding up anything cacheable to make sure that can't happen. The
	 * order of the unadjusted size will still match upon freeing.
	 */
	if (iova_len < (1 << (IOVA_RANGE_CACHE_MAX_SIZE - 1)))
		iova_len = roundup_pow_of_two(iova_len);

	if (dev->bus_dma_mask)
		dma_limit &= dev->bus_dma_mask;

	if (domain->geometry.force_aperture)
		dma_limit = min(dma_limit, domain->geometry.aperture_end);

	/* Try to get PCI devices a SAC address */
	if (dma_limit > DMA_BIT_MASK(32) && dev_is_pci(dev))
		iova = alloc_iova_fast(iovad, iova_len,
				       DMA_BIT_MASK(32) >> shift, false);

	if (!iova)
		iova = alloc_iova_fast(iovad, iova_len, dma_limit >> shift,
				       true);

	return (dma_addr_t)iova << shift;
}

static void iommu_dma_free_iova(struct iommu_dma_cookie *cookie,
		dma_addr_t iova, size_t size)
{
	struct iova_domain *iovad = &cookie->iovad;

	/* The MSI case is only ever cleaning up its most recent allocation */
	if (cookie->type == IOMMU_DMA_MSI_COOKIE)
		cookie->msi_iova -= size;
	else if (cookie->fq_domain)	/* non-strict mode */
		queue_iova(iovad, iova_pfn(iovad, iova),
				size >> iova_shift(iovad), 0);
	else
		free_iova_fast(iovad, iova_pfn(iovad, iova),
				size >> iova_shift(iovad));
}

static void __iommu_dma_unmap(struct device *dev, dma_addr_t dma_addr,
		size_t size)
{
	struct iommu_domain *domain = iommu_get_dma_domain(dev);
	struct iommu_dma_cookie *cookie = domain->iova_cookie;
	struct iova_domain *iovad = &cookie->iovad;
	size_t iova_off = iova_offset(iovad, dma_addr);

	dma_addr -= iova_off;
	size = iova_align(iovad, size + iova_off);

	WARN_ON(iommu_unmap_fast(domain, dma_addr, size) != size);
	if (!cookie->fq_domain)
		iommu_tlb_sync(domain);
	iommu_dma_free_iova(cookie, dma_addr, size);
}

static dma_addr_t __iommu_dma_map(struct device *dev, phys_addr_t phys,
		size_t size, int prot)
{
	struct iommu_domain *domain = iommu_get_dma_domain(dev);
	struct iommu_dma_cookie *cookie = domain->iova_cookie;
	size_t iova_off = 0;
	dma_addr_t iova;

	if (cookie->type == IOMMU_DMA_IOVA_COOKIE) {
		iova_off = iova_offset(&cookie->iovad, phys);
		size = iova_align(&cookie->iovad, size + iova_off);
	}

	iova = iommu_dma_alloc_iova(domain, size, dma_get_mask(dev), dev);
	if (!iova)
		return DMA_MAPPING_ERROR;

	if (iommu_map(domain, iova, phys - iova_off, size, prot)) {
		iommu_dma_free_iova(cookie, iova, size);
		return DMA_MAPPING_ERROR;
	}
	return iova + iova_off;
}

static void __iommu_dma_free_pages(struct page **pages, int count)
{
	while (count--)
		__free_page(pages[count]);
	kvfree(pages);
}

static struct page **__iommu_dma_alloc_pages(struct device *dev,
		unsigned int count, unsigned long order_mask, gfp_t gfp)
{
	struct page **pages;
	unsigned int i = 0, nid = dev_to_node(dev);

	order_mask &= (2U << MAX_ORDER) - 1;
	if (!order_mask)
		return NULL;

	pages = kvzalloc(count * sizeof(*pages), GFP_KERNEL);
	if (!pages)
		return NULL;

	/* IOMMU can map any pages, so himem can also be used here */
	gfp |= __GFP_NOWARN | __GFP_HIGHMEM;

	while (count) {
		struct page *page = NULL;
		unsigned int order_size;

		/*
		 * Higher-order allocations are a convenience rather
		 * than a necessity, hence using __GFP_NORETRY until
		 * falling back to minimum-order allocations.
		 */
		for (order_mask &= (2U << __fls(count)) - 1;
		     order_mask; order_mask &= ~order_size) {
			unsigned int order = __fls(order_mask);
			gfp_t alloc_flags = gfp;

			order_size = 1U << order;
			if (order_mask > order_size)
				alloc_flags |= __GFP_NORETRY;
			page = alloc_pages_node(nid, alloc_flags, order);
			if (!page)
				continue;
			if (!order)
				break;
			if (!PageCompound(page)) {
				split_page(page, order);
				break;
			} else if (!split_huge_page(page)) {
				break;
			}
			__free_pages(page, order);
		}
		if (!page) {
			__iommu_dma_free_pages(pages, i);
			return NULL;
		}
		count -= order_size;
		while (order_size--)
			pages[i++] = page++;
	}
	return pages;
}

static struct page **__iommu_dma_get_pages(void *cpu_addr)
{
	struct vm_struct *area = find_vm_area(cpu_addr);

	if (!area || !area->pages)
		return NULL;
	return area->pages;
}

/**
 * iommu_dma_alloc_remap - Allocate and map a buffer contiguous in IOVA space
 * @dev: Device to allocate memory for. Must be a real device
 *	 attached to an iommu_dma_domain
 * @size: Size of buffer in bytes
 * @dma_handle: Out argument for allocated DMA handle
 * @gfp: Allocation flags
 * @attrs: DMA attributes for this allocation
 *
 * If @size is less than PAGE_SIZE, then a full CPU page will be allocated,
 * but an IOMMU which supports smaller pages might not map the whole thing.
 *
 * Return: Mapped virtual address, or NULL on failure.
 */
static void *iommu_dma_alloc_remap(struct device *dev, size_t size,
		dma_addr_t *dma_handle, gfp_t gfp, unsigned long attrs)
{
	struct iommu_domain *domain = iommu_get_dma_domain(dev);
	struct iommu_dma_cookie *cookie = domain->iova_cookie;
	struct iova_domain *iovad = &cookie->iovad;
	bool coherent = dev_is_dma_coherent(dev);
	int ioprot = dma_info_to_prot(DMA_BIDIRECTIONAL, coherent, attrs);
	pgprot_t prot = arch_dma_mmap_pgprot(dev, PAGE_KERNEL, attrs);
	unsigned int count, min_size, alloc_sizes = domain->pgsize_bitmap;
	struct page **pages;
	struct sg_table sgt;
	dma_addr_t iova;
	void *vaddr;

	*dma_handle = DMA_MAPPING_ERROR;

	min_size = alloc_sizes & -alloc_sizes;
	if (min_size < PAGE_SIZE) {
		min_size = PAGE_SIZE;
		alloc_sizes |= PAGE_SIZE;
	} else {
		size = ALIGN(size, min_size);
	}
	if (attrs & DMA_ATTR_ALLOC_SINGLE_PAGES)
		alloc_sizes = min_size;

	count = PAGE_ALIGN(size) >> PAGE_SHIFT;
	pages = __iommu_dma_alloc_pages(dev, count, alloc_sizes >> PAGE_SHIFT,
					gfp);
	if (!pages)
		return NULL;

	size = iova_align(iovad, size);
	iova = iommu_dma_alloc_iova(domain, size, dev->coherent_dma_mask, dev);
	if (!iova)
		goto out_free_pages;

	if (sg_alloc_table_from_pages(&sgt, pages, count, 0, size, GFP_KERNEL))
		goto out_free_iova;

	if (!(ioprot & IOMMU_CACHE)) {
		struct scatterlist *sg;
		int i;

		for_each_sg(sgt.sgl, sg, sgt.orig_nents, i)
			arch_dma_prep_coherent(sg_page(sg), sg->length);
	}

	if (iommu_map_sg(domain, iova, sgt.sgl, sgt.orig_nents, ioprot)
			< size)
		goto out_free_sg;

	vaddr = dma_common_pages_remap(pages, size, VM_USERMAP, prot,
			__builtin_return_address(0));
	if (!vaddr)
		goto out_unmap;

	*dma_handle = iova;
	sg_free_table(&sgt);
	return vaddr;

out_unmap:
	__iommu_dma_unmap(dev, iova, size);
out_free_sg:
	sg_free_table(&sgt);
out_free_iova:
	iommu_dma_free_iova(cookie, iova, size);
out_free_pages:
	__iommu_dma_free_pages(pages, count);
	return NULL;
}

/**
 * __iommu_dma_mmap - Map a buffer into provided user VMA
 * @pages: Array representing buffer from __iommu_dma_alloc()
 * @size: Size of buffer in bytes
 * @vma: VMA describing requested userspace mapping
 *
 * Maps the pages of the buffer in @pages into @vma. The caller is responsible
 * for verifying the correct size and protection of @vma beforehand.
 */
static int __iommu_dma_mmap(struct page **pages, size_t size,
		struct vm_area_struct *vma)
{
	return vm_map_pages(vma, pages, PAGE_ALIGN(size) >> PAGE_SHIFT);
}

static void iommu_dma_sync_single_for_cpu(struct device *dev,
		dma_addr_t dma_handle, size_t size, enum dma_data_direction dir)
{
	phys_addr_t phys;

	if (dev_is_dma_coherent(dev))
		return;

	phys = iommu_iova_to_phys(iommu_get_dma_domain(dev), dma_handle);
	arch_sync_dma_for_cpu(dev, phys, size, dir);
}

static void iommu_dma_sync_single_for_device(struct device *dev,
		dma_addr_t dma_handle, size_t size, enum dma_data_direction dir)
{
	phys_addr_t phys;

	if (dev_is_dma_coherent(dev))
		return;

	phys = iommu_iova_to_phys(iommu_get_dma_domain(dev), dma_handle);
	arch_sync_dma_for_device(dev, phys, size, dir);
}

static void iommu_dma_sync_sg_for_cpu(struct device *dev,
		struct scatterlist *sgl, int nelems,
		enum dma_data_direction dir)
{
	struct scatterlist *sg;
	int i;

	if (dev_is_dma_coherent(dev))
		return;

	for_each_sg(sgl, sg, nelems, i)
		arch_sync_dma_for_cpu(dev, sg_phys(sg), sg->length, dir);
}

static void iommu_dma_sync_sg_for_device(struct device *dev,
		struct scatterlist *sgl, int nelems,
		enum dma_data_direction dir)
{
	struct scatterlist *sg;
	int i;

	if (dev_is_dma_coherent(dev))
		return;

	for_each_sg(sgl, sg, nelems, i)
		arch_sync_dma_for_device(dev, sg_phys(sg), sg->length, dir);
}

static dma_addr_t iommu_dma_map_page(struct device *dev, struct page *page,
		unsigned long offset, size_t size, enum dma_data_direction dir,
		unsigned long attrs)
{
	phys_addr_t phys = page_to_phys(page) + offset;
	bool coherent = dev_is_dma_coherent(dev);
	int prot = dma_info_to_prot(dir, coherent, attrs);
	dma_addr_t dma_handle;

	dma_handle =__iommu_dma_map(dev, phys, size, prot);
	if (!coherent && !(attrs & DMA_ATTR_SKIP_CPU_SYNC) &&
	    dma_handle != DMA_MAPPING_ERROR)
		arch_sync_dma_for_device(dev, phys, size, dir);
	return dma_handle;
}

static void iommu_dma_unmap_page(struct device *dev, dma_addr_t dma_handle,
		size_t size, enum dma_data_direction dir, unsigned long attrs)
{
	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
		iommu_dma_sync_single_for_cpu(dev, dma_handle, size, dir);
	__iommu_dma_unmap(dev, dma_handle, size);
}

/*
 * Prepare a successfully-mapped scatterlist to give back to the caller.
 *
 * At this point the segments are already laid out by iommu_dma_map_sg() to
 * avoid individually crossing any boundaries, so we merely need to check a
 * segment's start address to avoid concatenating across one.
 */
static int __finalise_sg(struct device *dev, struct scatterlist *sg, int nents,
		dma_addr_t dma_addr)
{
	struct scatterlist *s, *cur = sg;
	unsigned long seg_mask = dma_get_seg_boundary(dev);
	unsigned int cur_len = 0, max_len = dma_get_max_seg_size(dev);
	int i, count = 0;

	for_each_sg(sg, s, nents, i) {
		/* Restore this segment's original unaligned fields first */
		unsigned int s_iova_off = sg_dma_address(s);
		unsigned int s_length = sg_dma_len(s);
		unsigned int s_iova_len = s->length;

		s->offset += s_iova_off;
		s->length = s_length;
		sg_dma_address(s) = DMA_MAPPING_ERROR;
		sg_dma_len(s) = 0;

		/*
		 * Now fill in the real DMA data. If...
		 * - there is a valid output segment to append to
		 * - and this segment starts on an IOVA page boundary
		 * - but doesn't fall at a segment boundary
		 * - and wouldn't make the resulting output segment too long
		 */
		if (cur_len && !s_iova_off && (dma_addr & seg_mask) &&
		    (cur_len + s_length <= max_len)) {
			/* ...then concatenate it with the previous one */
			cur_len += s_length;
		} else {
			/* Otherwise start the next output segment */
			if (i > 0)
				cur = sg_next(cur);
			cur_len = s_length;
			count++;

			sg_dma_address(cur) = dma_addr + s_iova_off;
		}

		sg_dma_len(cur) = cur_len;
		dma_addr += s_iova_len;

		if (s_length + s_iova_off < s_iova_len)
			cur_len = 0;
	}
	return count;
}

/*
 * If mapping failed, then just restore the original list,
 * but making sure the DMA fields are invalidated.
 */
static void __invalidate_sg(struct scatterlist *sg, int nents)
{
	struct scatterlist *s;
	int i;

	for_each_sg(sg, s, nents, i) {
		if (sg_dma_address(s) != DMA_MAPPING_ERROR)
			s->offset += sg_dma_address(s);
		if (sg_dma_len(s))
			s->length = sg_dma_len(s);
		sg_dma_address(s) = DMA_MAPPING_ERROR;
		sg_dma_len(s) = 0;
	}
}

/*
 * The DMA API client is passing in a scatterlist which could describe
 * any old buffer layout, but the IOMMU API requires everything to be
 * aligned to IOMMU pages. Hence the need for this complicated bit of
 * impedance-matching, to be able to hand off a suitably-aligned list,
 * but still preserve the original offsets and sizes for the caller.
 */
static int iommu_dma_map_sg(struct device *dev, struct scatterlist *sg,
		int nents, enum dma_data_direction dir, unsigned long attrs)
{
	struct iommu_domain *domain = iommu_get_dma_domain(dev);
	struct iommu_dma_cookie *cookie = domain->iova_cookie;
	struct iova_domain *iovad = &cookie->iovad;
	struct scatterlist *s, *prev = NULL;
	int prot = dma_info_to_prot(dir, dev_is_dma_coherent(dev), attrs);
	dma_addr_t iova;
	size_t iova_len = 0;
	unsigned long mask = dma_get_seg_boundary(dev);
	int i;

	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC))
		iommu_dma_sync_sg_for_device(dev, sg, nents, dir);

	/*
	 * Work out how much IOVA space we need, and align the segments to
	 * IOVA granules for the IOMMU driver to handle. With some clever
	 * trickery we can modify the list in-place, but reversibly, by
	 * stashing the unaligned parts in the as-yet-unused DMA fields.
	 */
	for_each_sg(sg, s, nents, i) {
		size_t s_iova_off = iova_offset(iovad, s->offset);
		size_t s_length = s->length;
		size_t pad_len = (mask - iova_len + 1) & mask;

		sg_dma_address(s) = s_iova_off;
		sg_dma_len(s) = s_length;
		s->offset -= s_iova_off;
		s_length = iova_align(iovad, s_length + s_iova_off);
		s->length = s_length;

		/*
		 * Due to the alignment of our single IOVA allocation, we can
		 * depend on these assumptions about the segment boundary mask:
		 * - If mask size >= IOVA size, then the IOVA range cannot
		 *   possibly fall across a boundary, so we don't care.
		 * - If mask size < IOVA size, then the IOVA range must start
		 *   exactly on a boundary, therefore we can lay things out
		 *   based purely on segment lengths without needing to know
		 *   the actual addresses beforehand.
		 * - The mask must be a power of 2, so pad_len == 0 if
		 *   iova_len == 0, thus we cannot dereference prev the first
		 *   time through here (i.e. before it has a meaningful value).
		 */
		if (pad_len && pad_len < s_length - 1) {
			prev->length += pad_len;
			iova_len += pad_len;
		}

		iova_len += s_length;
		prev = s;
	}

	iova = iommu_dma_alloc_iova(domain, iova_len, dma_get_mask(dev), dev);
	if (!iova)
		goto out_restore_sg;

	/*
	 * We'll leave any physical concatenation to the IOMMU driver's
	 * implementation - it knows better than we do.
	 */
	if (iommu_map_sg(domain, iova, sg, nents, prot) < iova_len)
		goto out_free_iova;

	return __finalise_sg(dev, sg, nents, iova);

out_free_iova:
	iommu_dma_free_iova(cookie, iova, iova_len);
out_restore_sg:
	__invalidate_sg(sg, nents);
	return 0;
}

static void iommu_dma_unmap_sg(struct device *dev, struct scatterlist *sg,
		int nents, enum dma_data_direction dir, unsigned long attrs)
{
	dma_addr_t start, end;
	struct scatterlist *tmp;
	int i;

	if (!(attrs & DMA_ATTR_SKIP_CPU_SYNC) == 0)
		iommu_dma_sync_sg_for_cpu(dev, sg, nents, dir);

	/*
	 * The scatterlist segments are mapped into a single
	 * contiguous IOVA allocation, so this is incredibly easy.
	 */
	start = sg_dma_address(sg);
	for_each_sg(sg_next(sg), tmp, nents - 1, i) {
		if (sg_dma_len(tmp) == 0)
			break;
		sg = tmp;
	}
	end = sg_dma_address(sg) + sg_dma_len(sg);
	__iommu_dma_unmap(dev, start, end - start);
}

static dma_addr_t iommu_dma_map_resource(struct device *dev, phys_addr_t phys,
		size_t size, enum dma_data_direction dir, unsigned long attrs)
{
	return __iommu_dma_map(dev, phys, size,
			dma_info_to_prot(dir, false, attrs) | IOMMU_MMIO);
}

static void iommu_dma_unmap_resource(struct device *dev, dma_addr_t handle,
		size_t size, enum dma_data_direction dir, unsigned long attrs)
{
	__iommu_dma_unmap(dev, handle, size);
}

static void __iommu_dma_free(struct device *dev, size_t size, void *cpu_addr)
{
	size_t alloc_size = PAGE_ALIGN(size);
	int count = alloc_size >> PAGE_SHIFT;
	struct page *page = NULL, **pages = NULL;

	/* Non-coherent atomic allocation? Easy */
	if (IS_ENABLED(CONFIG_DMA_DIRECT_REMAP) &&
	    dma_free_from_pool(cpu_addr, alloc_size))
		return;

	if (IS_ENABLED(CONFIG_DMA_REMAP) && is_vmalloc_addr(cpu_addr)) {
		/*
		 * If it the address is remapped, then it's either non-coherent
		 * or highmem CMA, or an iommu_dma_alloc_remap() construction.
		 */
		pages = __iommu_dma_get_pages(cpu_addr);
		if (!pages)
			page = vmalloc_to_page(cpu_addr);
		dma_common_free_remap(cpu_addr, alloc_size, VM_USERMAP);
	} else {
		/* Lowmem means a coherent atomic or CMA allocation */
		page = virt_to_page(cpu_addr);
	}

	if (pages)
		__iommu_dma_free_pages(pages, count);
	if (page && !dma_release_from_contiguous(dev, page, count))
		__free_pages(page, get_order(alloc_size));
}

static void iommu_dma_free(struct device *dev, size_t size, void *cpu_addr,
		dma_addr_t handle, unsigned long attrs)
{
	__iommu_dma_unmap(dev, handle, size);
	__iommu_dma_free(dev, size, cpu_addr);
}

static void *iommu_dma_alloc_pages(struct device *dev, size_t size,
		struct page **pagep, gfp_t gfp, unsigned long attrs)
{
	bool coherent = dev_is_dma_coherent(dev);
	size_t alloc_size = PAGE_ALIGN(size);
	struct page *page = NULL;
	void *cpu_addr;

	if (gfpflags_allow_blocking(gfp))
		page = dma_alloc_from_contiguous(dev, alloc_size >> PAGE_SHIFT,
						 get_order(alloc_size),
						 gfp & __GFP_NOWARN);
	if (!page)
		page = alloc_pages(gfp, get_order(alloc_size));
	if (!page)
		return NULL;

	if (IS_ENABLED(CONFIG_DMA_REMAP) && (!coherent || PageHighMem(page))) {
		pgprot_t prot = arch_dma_mmap_pgprot(dev, PAGE_KERNEL, attrs);

		cpu_addr = dma_common_contiguous_remap(page, alloc_size,
				VM_USERMAP, prot, __builtin_return_address(0));
		if (!cpu_addr)
			goto out_free_pages;

		if (!coherent)
			arch_dma_prep_coherent(page, size);
	} else {
		cpu_addr = page_address(page);
	}

	*pagep = page;
	memset(cpu_addr, 0, alloc_size);
	return cpu_addr;
out_free_pages:
	if (!dma_release_from_contiguous(dev, page, alloc_size >> PAGE_SHIFT))
		__free_pages(page, get_order(alloc_size));
	return NULL;
}

static void *iommu_dma_alloc(struct device *dev, size_t size,
		dma_addr_t *handle, gfp_t gfp, unsigned long attrs)
{
	bool coherent = dev_is_dma_coherent(dev);
	int ioprot = dma_info_to_prot(DMA_BIDIRECTIONAL, coherent, attrs);
	struct page *page = NULL;
	void *cpu_addr;

	gfp |= __GFP_ZERO;

	if (IS_ENABLED(CONFIG_DMA_REMAP) && gfpflags_allow_blocking(gfp) &&
	    !(attrs & DMA_ATTR_FORCE_CONTIGUOUS))
		return iommu_dma_alloc_remap(dev, size, handle, gfp, attrs);

	if (IS_ENABLED(CONFIG_DMA_DIRECT_REMAP) &&
	    !gfpflags_allow_blocking(gfp) && !coherent)
		cpu_addr = dma_alloc_from_pool(PAGE_ALIGN(size), &page, gfp);
	else
		cpu_addr = iommu_dma_alloc_pages(dev, size, &page, gfp, attrs);
	if (!cpu_addr)
		return NULL;

	*handle = __iommu_dma_map(dev, page_to_phys(page), size, ioprot);
	if (*handle == DMA_MAPPING_ERROR) {
		__iommu_dma_free(dev, size, cpu_addr);
		return NULL;
	}

	return cpu_addr;
}

static int iommu_dma_mmap(struct device *dev, struct vm_area_struct *vma,
		void *cpu_addr, dma_addr_t dma_addr, size_t size,
		unsigned long attrs)
{
	unsigned long nr_pages = PAGE_ALIGN(size) >> PAGE_SHIFT;
	unsigned long pfn, off = vma->vm_pgoff;
	int ret;

	vma->vm_page_prot = arch_dma_mmap_pgprot(dev, vma->vm_page_prot, attrs);

	if (dma_mmap_from_dev_coherent(dev, vma, cpu_addr, size, &ret))
		return ret;

	if (off >= nr_pages || vma_pages(vma) > nr_pages - off)
		return -ENXIO;

	if (IS_ENABLED(CONFIG_DMA_REMAP) && is_vmalloc_addr(cpu_addr)) {
		struct page **pages = __iommu_dma_get_pages(cpu_addr);

		if (pages)
			return __iommu_dma_mmap(pages, size, vma);
		pfn = vmalloc_to_pfn(cpu_addr);
	} else {
		pfn = page_to_pfn(virt_to_page(cpu_addr));
	}

	return remap_pfn_range(vma, vma->vm_start, pfn + off,
			       vma->vm_end - vma->vm_start,
			       vma->vm_page_prot);
}

static int iommu_dma_get_sgtable(struct device *dev, struct sg_table *sgt,
		void *cpu_addr, dma_addr_t dma_addr, size_t size,
		unsigned long attrs)
{
	struct page *page;
	int ret;

	if (IS_ENABLED(CONFIG_DMA_REMAP) && is_vmalloc_addr(cpu_addr)) {
		struct page **pages = __iommu_dma_get_pages(cpu_addr);

		if (pages) {
			return sg_alloc_table_from_pages(sgt, pages,
					PAGE_ALIGN(size) >> PAGE_SHIFT,
					0, size, GFP_KERNEL);
		}

		page = vmalloc_to_page(cpu_addr);
	} else {
		page = virt_to_page(cpu_addr);
	}

	ret = sg_alloc_table(sgt, 1, GFP_KERNEL);
	if (!ret)
		sg_set_page(sgt->sgl, page, PAGE_ALIGN(size), 0);
	return ret;
}

static const struct dma_map_ops iommu_dma_ops = {
	.alloc			= iommu_dma_alloc,
	.free			= iommu_dma_free,
	.mmap			= iommu_dma_mmap,
	.get_sgtable		= iommu_dma_get_sgtable,
	.map_page		= iommu_dma_map_page,
	.unmap_page		= iommu_dma_unmap_page,
	.map_sg			= iommu_dma_map_sg,
	.unmap_sg		= iommu_dma_unmap_sg,
	.sync_single_for_cpu	= iommu_dma_sync_single_for_cpu,
	.sync_single_for_device	= iommu_dma_sync_single_for_device,
	.sync_sg_for_cpu	= iommu_dma_sync_sg_for_cpu,
	.sync_sg_for_device	= iommu_dma_sync_sg_for_device,
	.map_resource		= iommu_dma_map_resource,
	.unmap_resource		= iommu_dma_unmap_resource,
};

/*
 * The IOMMU core code allocates the default DMA domain, which the underlying
 * IOMMU driver needs to support via the dma-iommu layer.
 */
void iommu_setup_dma_ops(struct device *dev, u64 dma_base, u64 size)
{
	struct iommu_domain *domain = iommu_get_domain_for_dev(dev);

	if (!domain)
		goto out_err;

	/*
	 * The IOMMU core code allocates the default DMA domain, which the
	 * underlying IOMMU driver needs to support via the dma-iommu layer.
	 */
	if (domain->type == IOMMU_DOMAIN_DMA) {
		if (iommu_dma_init_domain(domain, dma_base, size, dev))
			goto out_err;
		dev->dma_ops = &iommu_dma_ops;
	}

	return;
out_err:
	 pr_warn("Failed to set up IOMMU for device %s; retaining platform DMA ops\n",
		 dev_name(dev));
}

static struct iommu_dma_msi_page *iommu_dma_get_msi_page(struct device *dev,
		phys_addr_t msi_addr, struct iommu_domain *domain)
{
	struct iommu_dma_cookie *cookie = domain->iova_cookie;
	struct iommu_dma_msi_page *msi_page;
	dma_addr_t iova;
	int prot = IOMMU_WRITE | IOMMU_NOEXEC | IOMMU_MMIO;
	size_t size = cookie_msi_granule(cookie);

	msi_addr &= ~(phys_addr_t)(size - 1);
	list_for_each_entry(msi_page, &cookie->msi_page_list, list)
		if (msi_page->phys == msi_addr)
			return msi_page;

	msi_page = kzalloc(sizeof(*msi_page), GFP_ATOMIC);
	if (!msi_page)
		return NULL;

	iova = __iommu_dma_map(dev, msi_addr, size, prot);
	if (iova == DMA_MAPPING_ERROR)
		goto out_free_page;

	INIT_LIST_HEAD(&msi_page->list);
	msi_page->phys = msi_addr;
	msi_page->iova = iova;
	list_add(&msi_page->list, &cookie->msi_page_list);
	return msi_page;

out_free_page:
	kfree(msi_page);
	return NULL;
}

int iommu_dma_prepare_msi(struct msi_desc *desc, phys_addr_t msi_addr)
{
	struct device *dev = msi_desc_to_dev(desc);
	struct iommu_domain *domain = iommu_get_domain_for_dev(dev);
	struct iommu_dma_cookie *cookie;
	struct iommu_dma_msi_page *msi_page;
	unsigned long flags;

	if (!domain || !domain->iova_cookie) {
		desc->iommu_cookie = NULL;
		return 0;
	}

	cookie = domain->iova_cookie;

	/*
	 * We disable IRQs to rule out a possible inversion against
	 * irq_desc_lock if, say, someone tries to retarget the affinity
	 * of an MSI from within an IPI handler.
	 */
	spin_lock_irqsave(&cookie->msi_lock, flags);
	msi_page = iommu_dma_get_msi_page(dev, msi_addr, domain);
	spin_unlock_irqrestore(&cookie->msi_lock, flags);

	msi_desc_set_iommu_cookie(desc, msi_page);

	if (!msi_page)
		return -ENOMEM;
	return 0;
}

void iommu_dma_compose_msi_msg(struct msi_desc *desc,
			       struct msi_msg *msg)
{
	struct device *dev = msi_desc_to_dev(desc);
	const struct iommu_domain *domain = iommu_get_domain_for_dev(dev);
	const struct iommu_dma_msi_page *msi_page;

	msi_page = msi_desc_get_iommu_cookie(desc);

	if (!domain || !domain->iova_cookie || WARN_ON(!msi_page))
		return;

	msg->address_hi = upper_32_bits(msi_page->iova);
	msg->address_lo &= cookie_msi_granule(domain->iova_cookie) - 1;
	msg->address_lo += lower_32_bits(msi_page->iova);
}

static int iommu_dma_init(void)
{
	return iova_cache_get();
}
arch_initcall(iommu_dma_init);