We need to get to a CONFIG_IA64_GENERIC kernel working, and the first
step to doing that seems to be fixing our usage of the i386 swiotlb and
pci-dma-xen. The patch below paravirtualizes lib/swiotlb.c and rips out
a lot of the code we had around for supporting the i386 versions. It
should be a little more straight-forward to switch to the generic kernel
flavor after this. This hasn't had much testing, so please don't run
this on a machine with data you care about. I'm running it with
swiotlb=force to try to exercise it, and it's holding up well. Comments
welcome. Thanks,
Alex
Signed-off-by: Alex Williamson <alex.williamson@xxxxxx>
---
diff -r 9313d0ce09f8 linux-2.6-xen-sparse/arch/ia64/xen/Makefile
--- a/linux-2.6-xen-sparse/arch/ia64/xen/Makefile Tue Apr 24 09:26:32
2007 -0600
+++ b/linux-2.6-xen-sparse/arch/ia64/xen/Makefile Fri Apr 27 10:05:04
2007 -0600
@@ -3,7 +3,5 @@
#
obj-y := hypercall.o xenivt.o xenentry.o xensetup.o xenpal.o xenhpski.o \
- hypervisor.o pci-dma-xen.o util.o xencomm.o xcom_hcall.o \
- xcom_mini.o xcom_privcmd.o mem.o
-
-pci-dma-xen-y := ../../i386/kernel/pci-dma-xen.o
+ hypervisor.o util.o xencomm.o xcom_hcall.o xcom_mini.o \
+ xcom_privcmd.o mem.o
diff -r 9313d0ce09f8 linux-2.6-xen-sparse/include/asm-ia64/dma-mapping.h
--- a/linux-2.6-xen-sparse/include/asm-ia64/dma-mapping.h Tue Apr 24
09:26:32 2007 -0600
+++ b/linux-2.6-xen-sparse/include/asm-ia64/dma-mapping.h Fri Apr 27
10:13:57 2007 -0600
@@ -13,7 +13,6 @@
#include <asm/swiotlb.h>
#endif
-#ifndef CONFIG_XEN
#define dma_alloc_coherent platform_dma_alloc_coherent
#define dma_alloc_noncoherent platform_dma_alloc_coherent /* coherent
mem. is cheap */
#define dma_free_coherent platform_dma_free_coherent
@@ -27,47 +26,6 @@
#define dma_sync_single_for_device platform_dma_sync_single_for_device
#define dma_sync_sg_for_device platform_dma_sync_sg_for_device
#define dma_mapping_error platform_dma_mapping_error
-#else
-int dma_map_sg(struct device *hwdev, struct scatterlist *sg, int nents,
- enum dma_data_direction direction);
-void dma_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nents,
- enum dma_data_direction direction);
-int dma_supported(struct device *dev, u64 mask);
-void *dma_alloc_coherent(struct device *dev, size_t size,
- dma_addr_t *dma_handle, gfp_t gfp);
-void dma_free_coherent(struct device *dev, size_t size, void *vaddr,
- dma_addr_t dma_handle);
-dma_addr_t dma_map_single(struct device *dev, void *ptr, size_t size,
- enum dma_data_direction direction);
-void dma_unmap_single(struct device *dev, dma_addr_t dma_addr, size_t size,
- enum dma_data_direction direction);
-void dma_sync_single_for_cpu(struct device *dev, dma_addr_t dma_handle,
- size_t size, enum dma_data_direction direction);
-void dma_sync_single_for_device(struct device *dev, dma_addr_t dma_handle,
- size_t size,
- enum dma_data_direction direction);
-int dma_mapping_error(dma_addr_t dma_addr);
-
-#define flush_write_buffers() do { } while (0)
-static inline void
-dma_sync_sg_for_cpu(struct device *dev, struct scatterlist *sg, int nelems,
- enum dma_data_direction direction)
-{
- if (swiotlb)
- swiotlb_sync_sg_for_cpu(dev,sg,nelems,direction);
- flush_write_buffers();
-}
-
-static inline void
-dma_sync_sg_for_device(struct device *dev, struct scatterlist *sg, int nelems,
- enum dma_data_direction direction)
-{
- if (swiotlb)
- swiotlb_sync_sg_for_device(dev,sg,nelems,direction);
- flush_write_buffers();
-}
-#endif
-
#define dma_map_page(dev, pg, off, size, dir) \
dma_map_single(dev, page_address(pg) + (off), (size), (dir))
#define dma_unmap_page(dev, dma_addr, size, dir) \
@@ -83,9 +41,7 @@ dma_sync_sg_for_device(struct device *de
#define dma_sync_single_range_for_device(dev, dma_handle, offset, size, dir)
\
dma_sync_single_for_device(dev, dma_handle, size, dir)
-#ifndef CONFIG_XEN
#define dma_supported platform_dma_supported
-#endif
static inline int
dma_set_mask (struct device *dev, u64 mask)
@@ -111,19 +67,6 @@ dma_cache_sync (void *vaddr, size_t size
#define dma_is_consistent(dma_handle) (1) /* all we do is coherent
memory... */
#ifdef CONFIG_XEN
-/* arch/i386/kernel/swiotlb.o requires */
-void contiguous_bitmap_init(unsigned long end_pfn);
-
-static inline int
-address_needs_mapping(struct device *hwdev, dma_addr_t addr)
-{
- dma_addr_t mask = DMA_64BIT_MASK;
- /* If the device has a mask, use it, otherwise default to 64 bits */
- if (hwdev && hwdev->dma_mask)
- mask = *hwdev->dma_mask;
- return (addr & ~mask) != 0;
-}
-
static inline int
range_straddles_page_boundary(void *p, size_t size)
{
diff -r 9313d0ce09f8 linux-2.6-xen-sparse/include/asm-ia64/machvec_dig.h
--- a/linux-2.6-xen-sparse/include/asm-ia64/machvec_dig.h Tue Apr 24
09:26:32 2007 -0600
+++ b/linux-2.6-xen-sparse/include/asm-ia64/machvec_dig.h Fri Apr 27
10:04:20 2007 -0600
@@ -13,19 +13,4 @@ extern ia64_mv_setup_t dig_setup;
#define platform_name "dig"
#define platform_setup dig_setup
-#ifdef CONFIG_XEN
-# define platform_dma_map_sg dma_map_sg
-# define platform_dma_unmap_sg dma_unmap_sg
-# define platform_dma_mapping_error dma_mapping_error
-# define platform_dma_supported dma_supported
-# define platform_dma_alloc_coherent dma_alloc_coherent
-# define platform_dma_free_coherent dma_free_coherent
-# define platform_dma_map_single dma_map_single
-# define platform_dma_unmap_single dma_unmap_single
-# define platform_dma_sync_single_for_cpu \
- dma_sync_single_for_cpu
-# define platform_dma_sync_single_for_device \
- dma_sync_single_for_device
-#endif
-
#endif /* _ASM_IA64_MACHVEC_DIG_h */
diff -r 9313d0ce09f8 linux-2.6-xen-sparse/lib/Makefile
--- a/linux-2.6-xen-sparse/lib/Makefile Tue Apr 24 09:26:32 2007 -0600
+++ b/linux-2.6-xen-sparse/lib/Makefile Fri Apr 27 10:03:11 2007 -0600
@@ -52,7 +52,9 @@ obj-$(CONFIG_AUDIT_GENERIC) += audit.o
obj-$(CONFIG_AUDIT_GENERIC) += audit.o
obj-$(CONFIG_SWIOTLB) += swiotlb.o
+ifeq ($(CONFIG_X86),y)
swiotlb-$(CONFIG_XEN) := ../arch/i386/kernel/swiotlb.o
+endif
hostprogs-y := gen_crc32table
clean-files := crc32table.h
diff -r 9313d0ce09f8 linux-2.6-xen-sparse/lib/swiotlb.c
--- /dev/null Thu Jan 01 00:00:00 1970 +0000
+++ b/linux-2.6-xen-sparse/lib/swiotlb.c Fri Apr 27 17:06:43 2007 -0600
@@ -0,0 +1,860 @@
+/*
+ * Dynamic DMA mapping support.
+ *
+ * This implementation is for IA-64 and EM64T platforms that do not support
+ * I/O TLBs (aka DMA address translation hardware).
+ * Copyright (C) 2000 Asit Mallick <Asit.K.Mallick@xxxxxxxxx>
+ * Copyright (C) 2000 Goutham Rao <goutham.rao@xxxxxxxxx>
+ * Copyright (C) 2000, 2003 Hewlett-Packard Co
+ * David Mosberger-Tang <davidm@xxxxxxxxxx>
+ *
+ * 03/05/07 davidm Switch from PCI-DMA to generic device DMA API.
+ * 00/12/13 davidm Rename to swiotlb.c and add mark_clean() to avoid
+ * unnecessary i-cache flushing.
+ * 04/07/.. ak Better overflow handling. Assorted fixes.
+ * 05/09/10 linville Add support for syncing ranges, support syncing for
+ * DMA_BIDIRECTIONAL mappings, miscellaneous cleanup.
+ */
+
+#include <linux/cache.h>
+#include <linux/dma-mapping.h>
+#include <linux/mm.h>
+#include <linux/module.h>
+#include <linux/spinlock.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ctype.h>
+
+#include <asm/io.h>
+#include <asm/dma.h>
+#include <asm/scatterlist.h>
+
+#include <linux/init.h>
+#include <linux/bootmem.h>
+
+#ifdef CONFIG_XEN
+#define VIRT_TO_PHYS(x) virt_to_bus(x)
+#define PHYS_TO_VIRT(x) bus_to_virt(x)
+#else
+#define VIRT_TO_PHYS(x) virt_to_phys(x)
+#define PHYS_TO_VIRT(x) phys_to_virt(x)
+#define range_straddles_page_boundary(x, y) (0)
+#endif
+
+#define OFFSET(val,align) ((unsigned long) \
+ ( (val) & ( (align) - 1)))
+
+#define SG_ENT_VIRT_ADDRESS(sg) (page_address((sg)->page) +
(sg)->offset)
+#define SG_ENT_PHYS_ADDRESS(SG) VIRT_TO_PHYS(SG_ENT_VIRT_ADDRESS(SG))
+
+/*
+ * Maximum allowable number of contiguous slabs to map,
+ * must be a power of 2. What is the appropriate value ?
+ * The complexity of {map,unmap}_single is linearly dependent on this value.
+ */
+#define IO_TLB_SEGSIZE 128
+
+/*
+ * log of the size of each IO TLB slab. The number of slabs is command line
+ * controllable.
+ */
+#define IO_TLB_SHIFT 11
+
+#define SLABS_PER_PAGE (1 << (PAGE_SHIFT - IO_TLB_SHIFT))
+
+/*
+ * Minimum IO TLB size to bother booting with. Systems with mainly
+ * 64bit capable cards will only lightly use the swiotlb. If we can't
+ * allocate a contiguous 1MB, we're probably in trouble anyway.
+ */
+#define IO_TLB_MIN_SLABS ((1<<20) >> IO_TLB_SHIFT)
+
+/*
+ * Enumeration for sync targets
+ */
+enum dma_sync_target {
+ SYNC_FOR_CPU = 0,
+ SYNC_FOR_DEVICE = 1,
+};
+
+int swiotlb_force;
+
+/*
+ * Used to do a quick range check in swiotlb_unmap_single and
+ * swiotlb_sync_single_*, to see if the memory was in fact allocated by this
+ * API.
+ */
+static char *io_tlb_start, *io_tlb_end;
+
+/*
+ * The number of IO TLB blocks (in groups of 64) betweeen io_tlb_start and
+ * io_tlb_end. This is command line adjustable via setup_io_tlb_npages.
+ */
+static unsigned long io_tlb_nslabs;
+
+/*
+ * When the IOMMU overflows we return a fallback buffer. This sets the size.
+ */
+static unsigned long io_tlb_overflow = 32*1024;
+
+void *io_tlb_overflow_buffer;
+
+/*
+ * This is a free list describing the number of free entries available from
+ * each index
+ */
+static unsigned int *io_tlb_list;
+static unsigned int io_tlb_index;
+
+/*
+ * We need to save away the original address corresponding to a mapped entry
+ * for the sync operations.
+ */
+static unsigned char **io_tlb_orig_addr;
+
+/*
+ * Protect the above data structures in the map and unmap calls
+ */
+static DEFINE_SPINLOCK(io_tlb_lock);
+
+static int __init
+setup_io_tlb_npages(char *str)
+{
+ if (isdigit(*str)) {
+ io_tlb_nslabs = simple_strtoul(str, &str, 0);
+ /* avoid tail segment of size < IO_TLB_SEGSIZE */
+ io_tlb_nslabs = ALIGN(io_tlb_nslabs, IO_TLB_SEGSIZE);
+ }
+ if (*str == ',')
+ ++str;
+ if (!strcmp(str, "force"))
+ swiotlb_force = 1;
+ return 1;
+}
+__setup("swiotlb=", setup_io_tlb_npages);
+/* make io_tlb_overflow tunable too? */
+
+/*
+ * Statically reserve bounce buffer space and initialize bounce buffer data
+ * structures for the software IO TLB used to implement the DMA API.
+ */
+void
+swiotlb_init_with_default_size (size_t default_size)
+{
+ unsigned long i;
+
+ if (!io_tlb_nslabs) {
+ io_tlb_nslabs = (default_size >> IO_TLB_SHIFT);
+ io_tlb_nslabs = ALIGN(io_tlb_nslabs, IO_TLB_SEGSIZE);
+ }
+
+#ifdef CONFIG_XEN
+ if (is_running_on_xen()) {
+ /* Round up to a power of two */
+ while (io_tlb_nslabs & (io_tlb_nslabs - 1))
+ io_tlb_nslabs += io_tlb_nslabs & ~(io_tlb_nslabs - 1);
+ }
+#endif
+ /*
+ * Get IO TLB memory from the low pages
+ */
+ io_tlb_start = alloc_bootmem_low_pages(io_tlb_nslabs * (1 <<
IO_TLB_SHIFT));
+ if (!io_tlb_start)
+ panic("Cannot allocate SWIOTLB buffer");
+ io_tlb_end = io_tlb_start + io_tlb_nslabs * (1 << IO_TLB_SHIFT);
+
+#ifdef CONFIG_XEN
+ /*
+ * Map these into 32bit space
+ * TODO: add logic for trying to get even lower pages
+ */
+ for (i = 0 ; i < io_tlb_nslabs ; i += IO_TLB_SEGSIZE) {
+ if (xen_create_contiguous_region((unsigned long)io_tlb_start +
(i << IO_TLB_SHIFT), get_order(IO_TLB_SEGSIZE << IO_TLB_SHIFT), 30))
+ panic("Failed to setup Xen contiguous region");
+ }
+#endif
+
+ /*
+ * Allocate and initialize the free list array. This array is used
+ * to find contiguous free memory regions of size up to IO_TLB_SEGSIZE
+ * between io_tlb_start and io_tlb_end.
+ */
+ io_tlb_list = alloc_bootmem(io_tlb_nslabs * sizeof(int));
+ for (i = 0; i < io_tlb_nslabs; i++)
+ io_tlb_list[i] = IO_TLB_SEGSIZE - OFFSET(i, IO_TLB_SEGSIZE);
+ io_tlb_index = 0;
+ io_tlb_orig_addr = alloc_bootmem(io_tlb_nslabs * sizeof(char *));
+
+ /*
+ * Get the overflow emergency buffer
+ */
+ io_tlb_overflow_buffer = alloc_bootmem_low(io_tlb_overflow);
+#ifdef CONFIG_XEN
+ if (xen_create_contiguous_region((unsigned long)io_tlb_overflow_buffer,
+ get_order(io_tlb_overflow), 32))
+ panic("Failed to setup Xen contiguous region for overflow");
+#endif
+ printk(KERN_INFO "Placing software IO TLB between 0x%lx - 0x%lx\n",
+ virt_to_phys(io_tlb_start), virt_to_phys(io_tlb_end));
+}
+
+void
+swiotlb_init (void)
+{
+ swiotlb_init_with_default_size(64 * (1<<20)); /* default to 64MB */
+}
+
+/*
+ * Systems with larger DMA zones (those that don't support ISA) can
+ * initialize the swiotlb later using the slab allocator if needed.
+ * This should be just like above, but with some error catching.
+ */
+int
+swiotlb_late_init_with_default_size (size_t default_size)
+{
+ unsigned long i, req_nslabs = io_tlb_nslabs;
+ unsigned int order;
+
+#ifdef CONFIG_XEN
+ if (is_running_on_xen())
+ panic("TODO\n");
+#endif
+ if (!io_tlb_nslabs) {
+ io_tlb_nslabs = (default_size >> IO_TLB_SHIFT);
+ io_tlb_nslabs = ALIGN(io_tlb_nslabs, IO_TLB_SEGSIZE);
+ }
+
+ /*
+ * Get IO TLB memory from the low pages
+ */
+ order = get_order(io_tlb_nslabs * (1 << IO_TLB_SHIFT));
+ io_tlb_nslabs = SLABS_PER_PAGE << order;
+
+ while ((SLABS_PER_PAGE << order) > IO_TLB_MIN_SLABS) {
+ io_tlb_start = (char *)__get_free_pages(GFP_DMA | __GFP_NOWARN,
+ order);
+ if (io_tlb_start)
+ break;
+ order--;
+ }
+
+ if (!io_tlb_start)
+ goto cleanup1;
+
+ if (order != get_order(io_tlb_nslabs * (1 << IO_TLB_SHIFT))) {
+ printk(KERN_WARNING "Warning: only able to allocate %ld MB "
+ "for software IO TLB\n", (PAGE_SIZE << order) >> 20);
+ io_tlb_nslabs = SLABS_PER_PAGE << order;
+ }
+ io_tlb_end = io_tlb_start + io_tlb_nslabs * (1 << IO_TLB_SHIFT);
+ memset(io_tlb_start, 0, io_tlb_nslabs * (1 << IO_TLB_SHIFT));
+
+ /*
+ * Allocate and initialize the free list array. This array is used
+ * to find contiguous free memory regions of size up to IO_TLB_SEGSIZE
+ * between io_tlb_start and io_tlb_end.
+ */
+ io_tlb_list = (unsigned int *)__get_free_pages(GFP_KERNEL,
+ get_order(io_tlb_nslabs * sizeof(int)));
+ if (!io_tlb_list)
+ goto cleanup2;
+
+ for (i = 0; i < io_tlb_nslabs; i++)
+ io_tlb_list[i] = IO_TLB_SEGSIZE - OFFSET(i, IO_TLB_SEGSIZE);
+ io_tlb_index = 0;
+
+ io_tlb_orig_addr = (unsigned char **)__get_free_pages(GFP_KERNEL,
+ get_order(io_tlb_nslabs * sizeof(char *)));
+ if (!io_tlb_orig_addr)
+ goto cleanup3;
+
+ memset(io_tlb_orig_addr, 0, io_tlb_nslabs * sizeof(char *));
+
+ /*
+ * Get the overflow emergency buffer
+ */
+ io_tlb_overflow_buffer = (void *)__get_free_pages(GFP_DMA,
+ get_order(io_tlb_overflow));
+ if (!io_tlb_overflow_buffer)
+ goto cleanup4;
+
+ printk(KERN_INFO "Placing %ldMB software IO TLB between 0x%lx - "
+ "0x%lx\n", (io_tlb_nslabs * (1 << IO_TLB_SHIFT)) >> 20,
+ virt_to_phys(io_tlb_start), virt_to_phys(io_tlb_end));
+
+ return 0;
+
+cleanup4:
+ free_pages((unsigned long)io_tlb_orig_addr, get_order(io_tlb_nslabs *
+ sizeof(char *)));
+ io_tlb_orig_addr = NULL;
+cleanup3:
+ free_pages((unsigned long)io_tlb_list, get_order(io_tlb_nslabs *
+ sizeof(int)));
+ io_tlb_list = NULL;
+ io_tlb_end = NULL;
+cleanup2:
+ free_pages((unsigned long)io_tlb_start, order);
+ io_tlb_start = NULL;
+cleanup1:
+ io_tlb_nslabs = req_nslabs;
+ return -ENOMEM;
+}
+
+static inline int
+address_needs_mapping(struct device *hwdev, dma_addr_t addr)
+{
+ dma_addr_t mask = 0xffffffff;
+ /* If the device has a mask, use it, otherwise default to 32 bits */
+ if (hwdev && hwdev->dma_mask)
+ mask = *hwdev->dma_mask;
+ return (addr & ~mask) != 0;
+}
+
+/*
+ * Allocates bounce buffer and returns its kernel virtual address.
+ */
+static void *
+map_single(struct device *hwdev, char *buffer, size_t size, int dir)
+{
+ unsigned long flags;
+ char *dma_addr;
+ unsigned int nslots, stride, index, wrap;
+ int i;
+
+ /*
+ * For mappings greater than a page, we limit the stride (and
+ * hence alignment) to a page size.
+ */
+ nslots = ALIGN(size, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT;
+ if (size > PAGE_SIZE)
+ stride = (1 << (PAGE_SHIFT - IO_TLB_SHIFT));
+ else
+ stride = 1;
+
+ BUG_ON(!nslots);
+
+ /*
+ * Find suitable number of IO TLB entries size that will fit this
+ * request and allocate a buffer from that IO TLB pool.
+ */
+ spin_lock_irqsave(&io_tlb_lock, flags);
+ {
+ wrap = index = ALIGN(io_tlb_index, stride);
+
+ if (index >= io_tlb_nslabs)
+ wrap = index = 0;
+
+ do {
+ /*
+ * If we find a slot that indicates we have 'nslots'
+ * number of contiguous buffers, we allocate the
+ * buffers from that slot and mark the entries as '0'
+ * indicating unavailable.
+ */
+ if (io_tlb_list[index] >= nslots) {
+ int count = 0;
+
+ for (i = index; i < (int) (index + nslots); i++)
+ io_tlb_list[i] = 0;
+ for (i = index - 1; (OFFSET(i, IO_TLB_SEGSIZE)
!= IO_TLB_SEGSIZE -1) && io_tlb_list[i]; i--)
+ io_tlb_list[i] = ++count;
+ dma_addr = io_tlb_start + (index <<
IO_TLB_SHIFT);
+
+ /*
+ * Update the indices to avoid searching in
+ * the next round.
+ */
+ io_tlb_index = ((index + nslots) < io_tlb_nslabs
+ ? (index + nslots) : 0);
+
+ goto found;
+ }
+ index += stride;
+ if (index >= io_tlb_nslabs)
+ index = 0;
+ } while (index != wrap);
+
+ spin_unlock_irqrestore(&io_tlb_lock, flags);
+ return NULL;
+ }
+ found:
+ spin_unlock_irqrestore(&io_tlb_lock, flags);
+
+ /*
+ * Save away the mapping from the original address to the DMA address.
+ * This is needed when we sync the memory. Then we sync the buffer if
+ * needed.
+ */
+ io_tlb_orig_addr[index] = buffer;
+ if (dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL)
+ memcpy(dma_addr, buffer, size);
+
+ return dma_addr;
+}
+
+/*
+ * dma_addr is the kernel virtual address of the bounce buffer to unmap.
+ */
+static void
+unmap_single(struct device *hwdev, char *dma_addr, size_t size, int dir)
+{
+ unsigned long flags;
+ int i, count, nslots = ALIGN(size, 1 << IO_TLB_SHIFT) >> IO_TLB_SHIFT;
+ int index = (dma_addr - io_tlb_start) >> IO_TLB_SHIFT;
+ char *buffer = io_tlb_orig_addr[index];
+
+ /*
+ * First, sync the memory before unmapping the entry
+ */
+ if (buffer && ((dir == DMA_FROM_DEVICE) || (dir == DMA_BIDIRECTIONAL)))
+ /*
+ * bounce... copy the data back into the original buffer * and
+ * delete the bounce buffer.
+ */
+ memcpy(buffer, dma_addr, size);
+
+ /*
+ * Return the buffer to the free list by setting the corresponding
+ * entries to indicate the number of contigous entries available.
+ * While returning the entries to the free list, we merge the entries
+ * with slots below and above the pool being returned.
+ */
+ spin_lock_irqsave(&io_tlb_lock, flags);
+ {
+ count = ((index + nslots) < ALIGN(index + 1, IO_TLB_SEGSIZE) ?
+ io_tlb_list[index + nslots] : 0);
+ /*
+ * Step 1: return the slots to the free list, merging the
+ * slots with superceeding slots
+ */
+ for (i = index + nslots - 1; i >= index; i--)
+ io_tlb_list[i] = ++count;
+ /*
+ * Step 2: merge the returned slots with the preceding slots,
+ * if available (non zero)
+ */
+ for (i = index - 1; (OFFSET(i, IO_TLB_SEGSIZE) !=
IO_TLB_SEGSIZE -1) && io_tlb_list[i]; i--)
+ io_tlb_list[i] = ++count;
+ }
+ spin_unlock_irqrestore(&io_tlb_lock, flags);
+}
+
+static void
+sync_single(struct device *hwdev, char *dma_addr, size_t size,
+ int dir, int target)
+{
+ int index = (dma_addr - io_tlb_start) >> IO_TLB_SHIFT;
+ char *buffer = io_tlb_orig_addr[index];
+
+ switch (target) {
+ case SYNC_FOR_CPU:
+ if (likely(dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL))
+ memcpy(buffer, dma_addr, size);
+ else
+ BUG_ON(dir != DMA_TO_DEVICE);
+ break;
+ case SYNC_FOR_DEVICE:
+ if (likely(dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL))
+ memcpy(dma_addr, buffer, size);
+ else
+ BUG_ON(dir != DMA_FROM_DEVICE);
+ break;
+ default:
+ BUG();
+ }
+}
+
+void *
+swiotlb_alloc_coherent(struct device *hwdev, size_t size,
+ dma_addr_t *dma_handle, gfp_t flags)
+{
+ unsigned long dev_addr;
+ void *ret;
+ int order = get_order(size);
+
+ /*
+ * XXX fix me: the DMA API should pass us an explicit DMA mask
+ * instead, or use ZONE_DMA32 (ia64 overloads ZONE_DMA to be a ~32
+ * bit range instead of a 16MB one).
+ */
+ flags |= GFP_DMA;
+
+ ret = (void *)__get_free_pages(flags, order);
+#ifdef CONFIG_XEN
+ if (ret && is_running_on_xen()) {
+ u64 mask = hwdev->coherent_dma_mask ? hwdev->coherent_dma_mask :
+ DMA_32BIT_MASK;
+ if (xen_create_contiguous_region((unsigned long)ret, order,
+ fls64(mask))) {
+ free_pages((unsigned long)ret, order);
+ ret = NULL;
+ }
+ }
+#endif
+ if (ret && address_needs_mapping(hwdev, VIRT_TO_PHYS(ret))) {
+ /*
+ * The allocated memory isn't reachable by the device.
+ * Fall back on swiotlb_map_single().
+ */
+ free_pages((unsigned long) ret, order);
+ ret = NULL;
+ }
+ if (!ret) {
+ /*
+ * We are either out of memory or the device can't DMA
+ * to GFP_DMA memory; fall back on
+ * swiotlb_map_single(), which will grab memory from
+ * the lowest available address range.
+ */
+ dma_addr_t handle;
+ handle = swiotlb_map_single(NULL, NULL, size, DMA_FROM_DEVICE);
+ if (swiotlb_dma_mapping_error(handle))
+ return NULL;
+
+ ret = PHYS_TO_VIRT(handle);
+ }
+
+ memset(ret, 0, size);
+ dev_addr = VIRT_TO_PHYS(ret);
+
+ /* Confirm address can be DMA'd by device */
+ if (address_needs_mapping(hwdev, dev_addr)) {
+ printk("hwdev DMA mask = 0x%016Lx, dev_addr = 0x%016lx\n",
+ (unsigned long long)*hwdev->dma_mask, dev_addr);
+ panic("swiotlb_alloc_coherent: allocated memory is out of "
+ "range for device");
+ }
+ *dma_handle = dev_addr;
+ return ret;
+}
+
+void
+swiotlb_free_coherent(struct device *hwdev, size_t size, void *vaddr,
+ dma_addr_t dma_handle)
+{
+ if (!(vaddr >= (void *)io_tlb_start
+ && vaddr < (void *)io_tlb_end)) {
+#ifdef CONFIG_XEN
+ if (is_running_on_xen())
+ xen_destroy_contiguous_region((unsigned long)vaddr,
+ get_order(size));
+#endif
+ free_pages((unsigned long) vaddr, get_order(size));
+ } else
+ /* DMA_TO_DEVICE to avoid memcpy in unmap_single */
+ swiotlb_unmap_single (hwdev, dma_handle, size, DMA_TO_DEVICE);
+}
+
+static void
+swiotlb_full(struct device *dev, size_t size, int dir, int do_panic)
+{
+ /*
+ * Ran out of IOMMU space for this operation. This is very bad.
+ * Unfortunately the drivers cannot handle this operation properly.
+ * unless they check for dma_mapping_error (most don't)
+ * When the mapping is small enough return a static buffer to limit
+ * the damage, or panic when the transfer is too big.
+ */
+ printk(KERN_ERR "DMA: Out of SW-IOMMU space for %lu bytes at "
+ "device %s\n", size, dev ? dev->bus_id : "?");
+
+ if (size > io_tlb_overflow && do_panic) {
+ if (dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL)
+ panic("DMA: Memory would be corrupted\n");
+ if (dir == DMA_TO_DEVICE || dir == DMA_BIDIRECTIONAL)
+ panic("DMA: Random memory would be DMAed\n");
+ }
+}
+
+/*
+ * Map a single buffer of the indicated size for DMA in streaming mode. The
+ * physical address to use is returned.
+ *
+ * Once the device is given the dma address, the device owns this memory until
+ * either swiotlb_unmap_single or swiotlb_dma_sync_single is performed.
+ */
+dma_addr_t
+swiotlb_map_single(struct device *hwdev, void *ptr, size_t size, int dir)
+{
+ unsigned long dev_addr = VIRT_TO_PHYS(ptr);
+ void *map;
+
+ BUG_ON(dir == DMA_NONE);
+ /*
+ * If the pointer passed in happens to be in the device's DMA window,
+ * we can safely return the device addr and not worry about bounce
+ * buffering it.
+ */
+ if (!range_straddles_page_boundary(ptr, size) &&
+ !address_needs_mapping(hwdev, dev_addr) && !swiotlb_force)
+ return dev_addr;
+
+ /*
+ * Oh well, have to allocate and map a bounce buffer.
+ */
+ map = map_single(hwdev, ptr, size, dir);
+ if (!map) {
+ swiotlb_full(hwdev, size, dir, 1);
+ map = io_tlb_overflow_buffer;
+ }
+
+ dev_addr = VIRT_TO_PHYS(map);
+
+ /*
+ * Ensure that the address returned is DMA'ble
+ */
+ if (address_needs_mapping(hwdev, dev_addr))
+ panic("map_single: bounce buffer is not DMA'ble");
+
+ return dev_addr;
+}
+
+/*
+ * Since DMA is i-cache coherent, any (complete) pages that were written via
+ * DMA can be marked as "clean" so that lazy_mmu_prot_update() doesn't have to
+ * flush them when they get mapped into an executable vm-area.
+ */
+static void
+mark_clean(void *addr, size_t size)
+{
+ unsigned long pg_addr, end;
+
+#ifndef XEN /* FIXME */
+ if (is_running_on_xen())
+ return;
+#endif
+ pg_addr = PAGE_ALIGN((unsigned long) addr);
+ end = (unsigned long) addr + size;
+ while (pg_addr + PAGE_SIZE <= end) {
+ struct page *page = virt_to_page(pg_addr);
+ set_bit(PG_arch_1, &page->flags);
+ pg_addr += PAGE_SIZE;
+ }
+}
+
+/*
+ * Unmap a single streaming mode DMA translation. The dma_addr and size must
+ * match what was provided for in a previous swiotlb_map_single call. All
+ * other usages are undefined.
+ *
+ * After this call, reads by the cpu to the buffer are guaranteed to see
+ * whatever the device wrote there.
+ */
+void
+swiotlb_unmap_single(struct device *hwdev, dma_addr_t dev_addr, size_t size,
+ int dir)
+{
+ char *dma_addr = PHYS_TO_VIRT(dev_addr);
+
+ BUG_ON(dir == DMA_NONE);
+ if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end)
+ unmap_single(hwdev, dma_addr, size, dir);
+ else if (dir == DMA_FROM_DEVICE)
+ mark_clean(dma_addr, size);
+}
+
+/*
+ * Make physical memory consistent for a single streaming mode DMA translation
+ * after a transfer.
+ *
+ * If you perform a swiotlb_map_single() but wish to interrogate the buffer
+ * using the cpu, yet do not wish to teardown the dma mapping, you must
+ * call this function before doing so. At the next point you give the dma
+ * address back to the card, you must first perform a
+ * swiotlb_dma_sync_for_device, and then the device again owns the buffer
+ */
+static inline void
+swiotlb_sync_single(struct device *hwdev, dma_addr_t dev_addr,
+ size_t size, int dir, int target)
+{
+ char *dma_addr = PHYS_TO_VIRT(dev_addr);
+
+ BUG_ON(dir == DMA_NONE);
+ if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end)
+ sync_single(hwdev, dma_addr, size, dir, target);
+ else if (dir == DMA_FROM_DEVICE)
+ mark_clean(dma_addr, size);
+}
+
+void
+swiotlb_sync_single_for_cpu(struct device *hwdev, dma_addr_t dev_addr,
+ size_t size, int dir)
+{
+ swiotlb_sync_single(hwdev, dev_addr, size, dir, SYNC_FOR_CPU);
+}
+
+void
+swiotlb_sync_single_for_device(struct device *hwdev, dma_addr_t dev_addr,
+ size_t size, int dir)
+{
+ swiotlb_sync_single(hwdev, dev_addr, size, dir, SYNC_FOR_DEVICE);
+}
+
+/*
+ * Same as above, but for a sub-range of the mapping.
+ */
+static inline void
+swiotlb_sync_single_range(struct device *hwdev, dma_addr_t dev_addr,
+ unsigned long offset, size_t size,
+ int dir, int target)
+{
+ char *dma_addr = PHYS_TO_VIRT(dev_addr) + offset;
+
+ BUG_ON(dir == DMA_NONE);
+ if (dma_addr >= io_tlb_start && dma_addr < io_tlb_end)
+ sync_single(hwdev, dma_addr, size, dir, target);
+ else if (dir == DMA_FROM_DEVICE)
+ mark_clean(dma_addr, size);
+}
+
+void
+swiotlb_sync_single_range_for_cpu(struct device *hwdev, dma_addr_t dev_addr,
+ unsigned long offset, size_t size, int dir)
+{
+ swiotlb_sync_single_range(hwdev, dev_addr, offset, size, dir,
+ SYNC_FOR_CPU);
+}
+
+void
+swiotlb_sync_single_range_for_device(struct device *hwdev, dma_addr_t dev_addr,
+ unsigned long offset, size_t size, int dir)
+{
+ swiotlb_sync_single_range(hwdev, dev_addr, offset, size, dir,
+ SYNC_FOR_DEVICE);
+}
+
+/*
+ * Map a set of buffers described by scatterlist in streaming mode for DMA.
+ * This is the scatter-gather version of the above swiotlb_map_single
+ * interface. Here the scatter gather list elements are each tagged with the
+ * appropriate dma address and length. They are obtained via
+ * sg_dma_{address,length}(SG).
+ *
+ * NOTE: An implementation may be able to use a smaller number of
+ * DMA address/length pairs than there are SG table elements.
+ * (for example via virtual mapping capabilities)
+ * The routine returns the number of addr/length pairs actually
+ * used, at most nents.
+ *
+ * Device ownership issues as mentioned above for swiotlb_map_single are the
+ * same here.
+ */
+int
+swiotlb_map_sg(struct device *hwdev, struct scatterlist *sg, int nelems,
+ int dir)
+{
+ void *addr;
+ unsigned long dev_addr;
+ int i;
+
+ BUG_ON(dir == DMA_NONE);
+
+ for (i = 0; i < nelems; i++, sg++) {
+ addr = SG_ENT_VIRT_ADDRESS(sg);
+ dev_addr = VIRT_TO_PHYS(addr);
+ if (swiotlb_force || address_needs_mapping(hwdev, dev_addr)) {
+ void *map = map_single(hwdev, addr, sg->length, dir);
+ sg->dma_address = virt_to_bus(map);
+ if (!map) {
+ /* Don't panic here, we expect map_sg users
+ to do proper error handling. */
+ swiotlb_full(hwdev, sg->length, dir, 0);
+ swiotlb_unmap_sg(hwdev, sg - i, i, dir);
+ sg[0].dma_length = 0;
+ return 0;
+ }
+ } else
+ sg->dma_address = dev_addr;
+ sg->dma_length = sg->length;
+ }
+ return nelems;
+}
+
+/*
+ * Unmap a set of streaming mode DMA translations. Again, cpu read rules
+ * concerning calls here are the same as for swiotlb_unmap_single() above.
+ */
+void
+swiotlb_unmap_sg(struct device *hwdev, struct scatterlist *sg, int nelems,
+ int dir)
+{
+ int i;
+
+ BUG_ON(dir == DMA_NONE);
+
+ for (i = 0; i < nelems; i++, sg++)
+ if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg))
+ unmap_single(hwdev, (void *)
PHYS_TO_VIRT(sg->dma_address), sg->dma_length, dir);
+ else if (dir == DMA_FROM_DEVICE)
+ mark_clean(SG_ENT_VIRT_ADDRESS(sg), sg->dma_length);
+}
+
+/*
+ * Make physical memory consistent for a set of streaming mode DMA translations
+ * after a transfer.
+ *
+ * The same as swiotlb_sync_single_* but for a scatter-gather list, same rules
+ * and usage.
+ */
+static inline void
+swiotlb_sync_sg(struct device *hwdev, struct scatterlist *sg,
+ int nelems, int dir, int target)
+{
+ int i;
+
+ BUG_ON(dir == DMA_NONE);
+
+ for (i = 0; i < nelems; i++, sg++)
+ if (sg->dma_address != SG_ENT_PHYS_ADDRESS(sg))
+ sync_single(hwdev, (void *) sg->dma_address,
+ sg->dma_length, dir, target);
+}
+
+void
+swiotlb_sync_sg_for_cpu(struct device *hwdev, struct scatterlist *sg,
+ int nelems, int dir)
+{
+ swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_CPU);
+}
+
+void
+swiotlb_sync_sg_for_device(struct device *hwdev, struct scatterlist *sg,
+ int nelems, int dir)
+{
+ swiotlb_sync_sg(hwdev, sg, nelems, dir, SYNC_FOR_DEVICE);
+}
+
+int
+swiotlb_dma_mapping_error(dma_addr_t dma_addr)
+{
+ return (dma_addr == VIRT_TO_PHYS(io_tlb_overflow_buffer));
+}
+
+/*
+ * Return whether the given device DMA address mask can be supported
+ * properly. For example, if your device can only drive the low 24-bits
+ * during bus mastering, then you would pass 0x00ffffff as the mask to
+ * this function.
+ */
+int
+swiotlb_dma_supported (struct device *hwdev, u64 mask)
+{
+ return (VIRT_TO_PHYS(io_tlb_end) - 1) <= mask;
+}
+
+EXPORT_SYMBOL(swiotlb_init);
+EXPORT_SYMBOL(swiotlb_map_single);
+EXPORT_SYMBOL(swiotlb_unmap_single);
+EXPORT_SYMBOL(swiotlb_map_sg);
+EXPORT_SYMBOL(swiotlb_unmap_sg);
+EXPORT_SYMBOL(swiotlb_sync_single_for_cpu);
+EXPORT_SYMBOL(swiotlb_sync_single_for_device);
+EXPORT_SYMBOL_GPL(swiotlb_sync_single_range_for_cpu);
+EXPORT_SYMBOL_GPL(swiotlb_sync_single_range_for_device);
+EXPORT_SYMBOL(swiotlb_sync_sg_for_cpu);
+EXPORT_SYMBOL(swiotlb_sync_sg_for_device);
+EXPORT_SYMBOL(swiotlb_dma_mapping_error);
+EXPORT_SYMBOL(swiotlb_alloc_coherent);
+EXPORT_SYMBOL(swiotlb_free_coherent);
+EXPORT_SYMBOL(swiotlb_dma_supported);
_______________________________________________
Xen-ia64-devel mailing list
Xen-ia64-devel@xxxxxxxxxxxxxxxxxxx
http://lists.xensource.com/xen-ia64-devel
|
Home