Well, not for some odd devices. See the next section for information
about that.
- DAC Addressing for Address Space Hungry Devices
-
-There exists a class of devices which do not mesh well with the PCI
-DMA mapping API. By definition these "mappings" are a finite
-resource. The number of total available mappings per bus is platform
-specific, but there will always be a reasonable amount.
-
-What is "reasonable"? Reasonable means that networking and block I/O
-devices need not worry about using too many mappings.
-
-As an example of a problematic device, consider compute cluster cards.
-They can potentially need to access gigabytes of memory at once via
-DMA. Dynamic mappings are unsuitable for this kind of access pattern.
-
-To this end we've provided a small API by which a device driver
-may use DAC cycles to directly address all of physical memory.
-Not all platforms support this, but most do. It is easy to determine
-whether the platform will work properly at probe time.
-
-First, understand that there may be a SEVERE performance penalty for
-using these interfaces on some platforms. Therefore, you MUST only
-use these interfaces if it is absolutely required. %99 of devices can
-use the normal APIs without any problems.
-
-Note that for streaming type mappings you must either use these
-interfaces, or the dynamic mapping interfaces above. You may not mix
-usage of both for the same device. Such an act is illegal and is
-guaranteed to put a banana in your tailpipe.
-
-However, consistent mappings may in fact be used in conjunction with
-these interfaces. Remember that, as defined, consistent mappings are
-always going to be SAC addressable.
-
-The first thing your driver needs to do is query the PCI platform
-layer if it is capable of handling your devices DAC addressing
-capabilities:
-
- int pci_dac_dma_supported(struct pci_dev *hwdev, u64 mask);
-
-You may not use the following interfaces if this routine fails.
-
-Next, DMA addresses using this API are kept track of using the
-dma64_addr_t type. It is guaranteed to be big enough to hold any
-DAC address the platform layer will give to you from the following
-routines. If you have consistent mappings as well, you still
-use plain dma_addr_t to keep track of those.
-
-All mappings obtained here will be direct. The mappings are not
-translated, and this is the purpose of this dialect of the DMA API.
-
-All routines work with page/offset pairs. This is the _ONLY_ way to
-portably refer to any piece of memory. If you have a cpu pointer
-(which may be validly DMA'd too) you may easily obtain the page
-and offset using something like this:
-
- struct page *page = virt_to_page(ptr);
- unsigned long offset = offset_in_page(ptr);
-
-Here are the interfaces:
-
- dma64_addr_t pci_dac_page_to_dma(struct pci_dev *pdev,
- struct page *page,
- unsigned long offset,
- int direction);
-
-The DAC address for the tuple PAGE/OFFSET are returned. The direction
-argument is the same as for pci_{map,unmap}_single(). The same rules
-for cpu/device access apply here as for the streaming mapping
-interfaces. To reiterate:
-
- The cpu may touch the buffer before pci_dac_page_to_dma.
- The device may touch the buffer after pci_dac_page_to_dma
- is made, but the cpu may NOT.
-
-When the DMA transfer is complete, invoke:
-
- void pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev,
- dma64_addr_t dma_addr,
- size_t len, int direction);
-
-This must be done before the CPU looks at the buffer again.
-This interface behaves identically to pci_dma_sync_{single,sg}_for_cpu().
-
-And likewise, if you wish to let the device get back at the buffer after
-the cpu has read/written it, invoke:
-
- void pci_dac_dma_sync_single_for_device(struct pci_dev *pdev,
- dma64_addr_t dma_addr,
- size_t len, int direction);
-
-before letting the device access the DMA area again.
-
-If you need to get back to the PAGE/OFFSET tuple from a dma64_addr_t
-the following interfaces are provided:
-
- struct page *pci_dac_dma_to_page(struct pci_dev *pdev,
- dma64_addr_t dma_addr);
- unsigned long pci_dac_dma_to_offset(struct pci_dev *pdev,
- dma64_addr_t dma_addr);
-
-This is possible with the DAC interfaces purely because they are
-not translated in any way.
-
Optimizing Unmap State Space Consumption
On many platforms, pci_unmap_{single,page}() is simply a nop.
p[i] = 0;
}
\f
+/* True if the machine supports DAC addressing, and DEV can
+ make use of it given MASK. */
+static int pci_dac_dma_supported(struct pci_dev *hwdev, u64 mask);
+
/* Map a single buffer of the indicated size for PCI DMA in streaming
mode. The 32-bit PCI bus mastering address to use is returned.
Once the device is given the dma address, the device owns this memory
/* True if the machine supports DAC addressing, and DEV can
make use of it given MASK. */
-int
+static int
pci_dac_dma_supported(struct pci_dev *dev, u64 mask)
{
dma64_addr_t dac_offset = alpha_mv.pci_dac_offset;
return ok;
}
-EXPORT_SYMBOL(pci_dac_dma_supported);
-
-dma64_addr_t
-pci_dac_page_to_dma(struct pci_dev *pdev, struct page *page,
- unsigned long offset, int direction)
-{
- return (alpha_mv.pci_dac_offset
- + __pa(page_address(page))
- + (dma64_addr_t) offset);
-}
-EXPORT_SYMBOL(pci_dac_page_to_dma);
-
-struct page *
-pci_dac_dma_to_page(struct pci_dev *pdev, dma64_addr_t dma_addr)
-{
- unsigned long paddr = (dma_addr & PAGE_MASK) - alpha_mv.pci_dac_offset;
- return virt_to_page(__va(paddr));
-}
-EXPORT_SYMBOL(pci_dac_dma_to_page);
-
-unsigned long
-pci_dac_dma_to_offset(struct pci_dev *pdev, dma64_addr_t dma_addr)
-{
- return (dma_addr & ~PAGE_MASK);
-}
-EXPORT_SYMBOL(pci_dac_dma_to_offset);
/* Helper for generic DMA-mapping functions. */
# Makefile for the PCI specific kernel interface routines under Linux.
#
-obj-y += pci.o pci-dac.o
+obj-y += pci.o
#
# PCI bus host bridge specific code
+++ /dev/null
-/*
- * This file is subject to the terms and conditions of the GNU General Public
- * License. See the file "COPYING" in the main directory of this archive
- * for more details.
- *
- * Copyright (C) 2000 Ani Joshi <ajoshi@unixbox.com>
- * Copyright (C) 2000, 2001, 06 Ralf Baechle <ralf@linux-mips.org>
- * swiped from i386, and cloned for MIPS by Geert, polished by Ralf.
- */
-
-#include <linux/types.h>
-#include <linux/dma-mapping.h>
-#include <linux/mm.h>
-#include <linux/module.h>
-#include <linux/string.h>
-
-#include <asm/cache.h>
-#include <asm/io.h>
-
-#include <dma-coherence.h>
-
-#include <linux/pci.h>
-
-dma64_addr_t pci_dac_page_to_dma(struct pci_dev *pdev,
- struct page *page, unsigned long offset, int direction)
-{
- struct device *dev = &pdev->dev;
-
- BUG_ON(direction == DMA_NONE);
-
- if (!plat_device_is_coherent(dev)) {
- unsigned long addr;
-
- addr = (unsigned long) page_address(page) + offset;
- dma_cache_wback_inv(addr, PAGE_SIZE);
- }
-
- return plat_map_dma_mem_page(dev, page) + offset;
-}
-
-EXPORT_SYMBOL(pci_dac_page_to_dma);
-
-struct page *pci_dac_dma_to_page(struct pci_dev *pdev,
- dma64_addr_t dma_addr)
-{
- return pfn_to_page(plat_dma_addr_to_phys(dma_addr) >> PAGE_SHIFT);
-}
-
-EXPORT_SYMBOL(pci_dac_dma_to_page);
-
-unsigned long pci_dac_dma_to_offset(struct pci_dev *pdev,
- dma64_addr_t dma_addr)
-{
- return dma_addr & ~PAGE_MASK;
-}
-
-EXPORT_SYMBOL(pci_dac_dma_to_offset);
-
-void pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev,
- dma64_addr_t dma_addr, size_t len, int direction)
-{
- BUG_ON(direction == PCI_DMA_NONE);
-
- if (!plat_device_is_coherent(&pdev->dev))
- dma_cache_wback_inv(dma_addr + PAGE_OFFSET, len);
-}
-
-EXPORT_SYMBOL(pci_dac_dma_sync_single_for_cpu);
-
-void pci_dac_dma_sync_single_for_device(struct pci_dev *pdev,
- dma64_addr_t dma_addr, size_t len, int direction)
-{
- BUG_ON(direction == PCI_DMA_NONE);
-
- if (!plat_device_is_coherent(&pdev->dev))
- dma_cache_wback_inv(dma_addr + PAGE_OFFSET, len);
-}
-
-EXPORT_SYMBOL(pci_dac_dma_sync_single_for_device);
int iommu_bio_merge __read_mostly = 0;
EXPORT_SYMBOL(iommu_bio_merge);
-int iommu_sac_force __read_mostly = 0;
-EXPORT_SYMBOL(iommu_sac_force);
+static int iommu_sac_force __read_mostly = 0;
int no_iommu __read_mostly;
#ifdef CONFIG_IOMMU_DEBUG
extern int pci_dma_supported(struct pci_dev *hwdev, u64 mask);
-/* True if the machine supports DAC addressing, and DEV can
- make use of it given MASK. */
-extern int pci_dac_dma_supported(struct pci_dev *hwdev, u64 mask);
-
-/* Convert to/from DAC dma address and struct page. */
-extern dma64_addr_t pci_dac_page_to_dma(struct pci_dev *, struct page *,
- unsigned long, int);
-extern struct page *pci_dac_dma_to_page(struct pci_dev *, dma64_addr_t);
-extern unsigned long pci_dac_dma_to_offset(struct pci_dev *, dma64_addr_t);
-
-static inline void
-pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, dma64_addr_t dma_addr,
- size_t len, int direction)
-{
- /* Nothing to do. */
-}
-
-static inline void
-pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr,
- size_t len, int direction)
-{
- /* Nothing to do. */
-}
-
#ifdef CONFIG_PCI
static inline void pci_dma_burst_advice(struct pci_dev *pdev,
enum pci_dma_burst_strategy *strat,
*/
#define PCI_DMA_BUS_IS_PHYS (0)
-/*
- * We don't support DAC DMA cycles.
- */
-#define pci_dac_dma_supported(pci_dev, mask) (0)
-
/*
* Whether pci_unmap_{single,page} is a nop depends upon the
* configuration.
#define pci_unmap_len(PTR, LEN_NAME) (0)
#define pci_unmap_len_set(PTR, LEN_NAME, VAL) do { } while (0)
-/* This is always fine. */
-#define pci_dac_dma_supported(pci_dev, mask) (1)
-
-static inline dma64_addr_t
-pci_dac_page_to_dma(struct pci_dev *pdev, struct page *page, unsigned long offset, int direction)
-{
- return ((dma64_addr_t) page_to_phys(page) +
- (dma64_addr_t) offset);
-}
-
-static inline struct page *
-pci_dac_dma_to_page(struct pci_dev *pdev, dma64_addr_t dma_addr)
-{
- return pfn_to_page(dma_addr >> PAGE_SHIFT);
-}
-
-static inline unsigned long
-pci_dac_dma_to_offset(struct pci_dev *pdev, dma64_addr_t dma_addr)
-{
- return (dma_addr & ~PAGE_MASK);
-}
-
-static inline void
-pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
-{
-}
-
-static inline void
-pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
-{
-}
-
#define HAVE_PCI_MMAP
extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
enum pci_mmap_state mmap_state, int write_combine);
extern void pci_free_consistent(struct pci_dev *hwdev, size_t size,
void *vaddr, dma_addr_t dma_handle);
-/* This is always fine. */
-#define pci_dac_dma_supported(pci_dev, mask) (1)
-
/* Return the index of the PCI controller for device PDEV. */
#define pci_controller_num(PDEV) (0)
#define pci_unmap_len(PTR, LEN_NAME) (0)
#define pci_unmap_len_set(PTR, LEN_NAME, VAL) do { } while (0)
-/* This is always fine. */
-#define pci_dac_dma_supported(pci_dev, mask) (1)
-
-static inline dma64_addr_t
-pci_dac_page_to_dma(struct pci_dev *pdev, struct page *page, unsigned long offset, int direction)
-{
- return ((dma64_addr_t) page_to_phys(page) +
- (dma64_addr_t) offset);
-}
-
-static inline struct page *
-pci_dac_dma_to_page(struct pci_dev *pdev, dma64_addr_t dma_addr)
-{
- return pfn_to_page(dma_addr >> PAGE_SHIFT);
-}
-
-static inline unsigned long
-pci_dac_dma_to_offset(struct pci_dev *pdev, dma64_addr_t dma_addr)
-{
- return (dma_addr & ~PAGE_MASK);
-}
-
-static inline void
-pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
-{
-}
-
-static inline void
-pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
-{
- flush_write_buffers();
-}
-
#define HAVE_PCI_MMAP
extern int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
enum pci_mmap_state mmap_state, int write_combine);
#define pci_unmap_len_set(PTR, LEN_NAME, VAL) \
(((PTR)->LEN_NAME) = (VAL))
-/* The ia64 platform always supports 64-bit addressing. */
-#define pci_dac_dma_supported(pci_dev, mask) (1)
-#define pci_dac_page_to_dma(dev,pg,off,dir) ((dma_addr_t) page_to_bus(pg) + (off))
-#define pci_dac_dma_to_page(dev,dma_addr) (virt_to_page(bus_to_virt(dma_addr)))
-#define pci_dac_dma_to_offset(dev,dma_addr) offset_in_page(dma_addr)
-#define pci_dac_dma_sync_single_for_cpu(dev,dma_addr,len,dir) do { } while (0)
-#define pci_dac_dma_sync_single_for_device(dev,dma_addr,len,dir) do { mb(); } while (0)
-
#ifdef CONFIG_PCI
static inline void pci_dma_burst_advice(struct pci_dev *pdev,
enum pci_dma_burst_strategy *strat,
return 1;
}
-/*
- * Not supporting more than 32-bit PCI bus addresses now, but
- * must satisfy references to this function. Change if needed.
- */
-#define pci_dac_dma_supported(pci_dev, mask) (0)
-
#endif /* CONFIG_COMEMPCI */
#endif /* M68KNOMMU_PCI_H */
#endif /* CONFIG_DMA_NEED_PCI_MAP_STATE */
-/* This is always fine. */
-#define pci_dac_dma_supported(pci_dev, mask) (1)
-
-extern dma64_addr_t pci_dac_page_to_dma(struct pci_dev *pdev,
- struct page *page, unsigned long offset, int direction);
-extern struct page *pci_dac_dma_to_page(struct pci_dev *pdev,
- dma64_addr_t dma_addr);
-extern unsigned long pci_dac_dma_to_offset(struct pci_dev *pdev,
- dma64_addr_t dma_addr);
-extern void pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev,
- dma64_addr_t dma_addr, size_t len, int direction);
-extern void pci_dac_dma_sync_single_for_device(struct pci_dev *pdev,
- dma64_addr_t dma_addr, size_t len, int direction);
-
#ifdef CONFIG_PCI
static inline void pci_dma_burst_advice(struct pci_dev *pdev,
enum pci_dma_burst_strategy *strat,
#define PCIBIOS_MIN_IO 0x10
#define PCIBIOS_MIN_MEM 0x1000 /* NBPG - but pci/setup-res.c dies */
-/* Don't support DAC yet. */
-#define pci_dac_dma_supported(pci_dev, mask) (0)
-
/* export the pci_ DMA API in terms of the dma_ one */
#include <asm-generic/pci-dma-compat.h>
void (*unmap_sg)(struct device *dev, struct scatterlist *sg,
int nents, enum dma_data_direction direction);
int (*dma_supported)(struct device *dev, u64 mask);
- int (*dac_dma_supported)(struct device *dev, u64 mask);
int (*set_dma_mask)(struct device *dev, u64 dma_mask);
};
extern void set_pci_dma_ops(struct dma_mapping_ops *dma_ops);
extern struct dma_mapping_ops *get_pci_dma_ops(void);
-/* For DAC DMA, we currently don't support it by default, but
- * we let 64-bit platforms override this.
- */
-static inline int pci_dac_dma_supported(struct pci_dev *hwdev,u64 mask)
-{
- struct dma_mapping_ops *d = get_pci_dma_ops();
-
- if (d && d->dac_dma_supported)
- return d->dac_dma_supported(&hwdev->dev, mask);
- return 0;
-}
-
static inline void pci_dma_burst_advice(struct pci_dev *pdev,
enum pci_dma_burst_strategy *strat,
unsigned long *strategy_parameter)
}
#endif
-/*
- * At present there are very few 32-bit PPC machines that can have
- * memory above the 4GB point, and we don't support that.
- */
-#define pci_dac_dma_supported(pci_dev, mask) (0)
-
/* Return the index of the PCI controller for device PDEV. */
#define pci_domain_nr(bus) ((struct pci_controller *)(bus)->sysdata)->index
}
#endif
-/*
- * At present there are very few 32-bit PPC machines that can have
- * memory above the 4GB point, and we don't support that.
- */
-#define pci_dac_dma_supported(pci_dev, mask) (0)
-
/* Return the index of the PCI controller for device PDEV. */
#define pci_domain_nr(bus) ((struct pci_controller *)(bus)->sysdata)->index
#define pci_unmap_len_set(PTR, LEN_NAME, VAL) do { } while (0)
#endif
-/* Not supporting more than 32-bit PCI bus addresses now, but
- * must satisfy references to this function. Change if needed.
- */
-#define pci_dac_dma_supported(pci_dev, mask) (0)
-
#ifdef CONFIG_PCI
static inline void pci_dma_burst_advice(struct pci_dev *pdev,
enum pci_dma_burst_strategy *strat,
#define pci_unmap_len_set(PTR, LEN_NAME, VAL) do { } while (0)
#endif
-/* Not supporting more than 32-bit PCI bus addresses now, but
- * must satisfy references to this function. Change if needed.
- */
-#define pci_dac_dma_supported(pci_dev, mask) (0)
-
/* These macros should be used after a pci_map_sg call has been done
* to get bus addresses of each of the SG entries and their lengths.
* You should only work with the number of sg entries pci_map_sg
return 1;
}
-#define pci_dac_dma_supported(dev, mask) (0)
-
#ifdef CONFIG_PCI
static inline void pci_dma_burst_advice(struct pci_dev *pdev,
enum pci_dma_burst_strategy *strat,
#define PCI64_REQUIRED_MASK (~(dma64_addr_t)0)
#define PCI64_ADDR_BASE 0xfffc000000000000UL
-/* Usage of the pci_dac_foo interfaces is only valid if this
- * test passes.
- */
-#define pci_dac_dma_supported(pci_dev, mask) \
- ((((mask) & PCI64_REQUIRED_MASK) == PCI64_REQUIRED_MASK) ? 1 : 0)
-
-static inline dma64_addr_t
-pci_dac_page_to_dma(struct pci_dev *pdev, struct page *page, unsigned long offset, int direction)
-{
- return (PCI64_ADDR_BASE +
- __pa(page_address(page)) + offset);
-}
-
-static inline struct page *
-pci_dac_dma_to_page(struct pci_dev *pdev, dma64_addr_t dma_addr)
-{
- unsigned long paddr = (dma_addr & PAGE_MASK) - PCI64_ADDR_BASE;
-
- return virt_to_page(__va(paddr));
-}
-
-static inline unsigned long
-pci_dac_dma_to_offset(struct pci_dev *pdev, dma64_addr_t dma_addr)
-{
- return (dma_addr & ~PAGE_MASK);
-}
-
-static inline void
-pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
-{
- /* DAC cycle addressing does not make use of the
- * PCI controller's streaming cache, so nothing to do.
- */
-}
-
-static inline void
-pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
-{
- /* DAC cycle addressing does not make use of the
- * PCI controller's streaming cache, so nothing to do.
- */
-}
-
#define PCI_DMA_ERROR_CODE (~(dma_addr_t)0x0)
static inline int pci_dma_mapping_error(dma_addr_t dma_addr)
/* As we don't really support PCI DMA to cpu memory, and use bounce-buffers
instead, perversely enough, this becomes always true! */
# define pci_dma_supported(dev, mask) 1
-# define pci_dac_dma_supported(dev, mask) 0
# define pcibios_assign_all_busses() 1
#endif /* CONFIG_RTE_MB_A_PCI */
#if defined(CONFIG_IOMMU) || defined(CONFIG_CALGARY_IOMMU)
-/*
- * x86-64 always supports DAC, but sometimes it is useful to force
- * devices through the IOMMU to get automatic sg list merging.
- * Optional right now.
- */
-extern int iommu_sac_force;
-#define pci_dac_dma_supported(pci_dev, mask) (!iommu_sac_force)
-
#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME) \
dma_addr_t ADDR_NAME;
#define DECLARE_PCI_UNMAP_LEN(LEN_NAME) \
#else
/* No IOMMU */
-#define pci_dac_dma_supported(pci_dev, mask) 1
-
#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)
#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)
#define pci_unmap_addr(PTR, ADDR_NAME) (0)
#include <asm-generic/pci-dma-compat.h>
-static inline dma64_addr_t
-pci_dac_page_to_dma(struct pci_dev *pdev, struct page *page, unsigned long offset, int direction)
-{
- return ((dma64_addr_t) page_to_phys(page) +
- (dma64_addr_t) offset);
-}
-
-static inline struct page *
-pci_dac_dma_to_page(struct pci_dev *pdev, dma64_addr_t dma_addr)
-{
- return virt_to_page(__va(dma_addr));
-}
-
-static inline unsigned long
-pci_dac_dma_to_offset(struct pci_dev *pdev, dma64_addr_t dma_addr)
-{
- return (dma_addr & ~PAGE_MASK);
-}
-
-static inline void
-pci_dac_dma_sync_single_for_cpu(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
-{
-}
-
-static inline void
-pci_dac_dma_sync_single_for_device(struct pci_dev *pdev, dma64_addr_t dma_addr, size_t len, int direction)
-{
- flush_write_buffers();
-}
-
#ifdef CONFIG_PCI
static inline void pci_dma_burst_advice(struct pci_dev *pdev,
enum pci_dma_burst_strategy *strat,
#define pci_ubnmap_len(PTR, LEN_NAME) (0)
#define pci_unmap_len_set(PTR, LEN_NAME, VAL) do { } while (0)
-/* We cannot access memory above 4GB */
-#define pci_dac_dma_supported(pci_dev, mask) (0)
-
/* Map a range of PCI memory or I/O space for a device into user space */
int pci_mmap_page_range(struct pci_dev *pdev, struct vm_area_struct *vma,
enum pci_mmap_state mmap_state, int write_combine);
projects / GitHub / moto-9609 / android_kernel_motorola_exynos9610.git / commitdiff