Merge

Documentation/DMA-API.txt

@@ -444,4 +444,83 @@ dma_alloc_noncoherent(), starting at virtual address vaddr and
 continuing on for size.  Again, you *must* observe the cache line
 boundaries when doing this.
 
+int
+dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
+			    dma_addr_t device_addr, size_t size, int
+			    flags)
+
+
+Declare region of memory to be handed out by dma_alloc_coherent when
+it's asked for coherent memory for this device.
+
+bus_addr is the physical address to which the memory is currently
+assigned in the bus responding region (this will be used by the
+platform to perform the mapping)
+
+device_addr is the physical address the device needs to be programmed
+with actually to address this memory (this will be handed out as the
+dma_addr_t in dma_alloc_coherent())
+
+size is the size of the area (must be multiples of PAGE_SIZE).
+
+flags can be or'd together and are
+
+DMA_MEMORY_MAP - request that the memory returned from
+dma_alloc_coherent() be directly writeable.
+
+DMA_MEMORY_IO - request that the memory returned from
+dma_alloc_coherent() be addressable using read/write/memcpy_toio etc.
+
+One or both of these flags must be present
+
+DMA_MEMORY_INCLUDES_CHILDREN - make the declared memory be allocated by
+dma_alloc_coherent of any child devices of this one (for memory residing
+on a bridge).
+
+DMA_MEMORY_EXCLUSIVE - only allocate memory from the declared regions. 
+Do not allow dma_alloc_coherent() to fall back to system memory when
+it's out of memory in the declared region.
+
+The return value will be either DMA_MEMORY_MAP or DMA_MEMORY_IO and
+must correspond to a passed in flag (i.e. no returning DMA_MEMORY_IO
+if only DMA_MEMORY_MAP were passed in) for success or zero for
+failure.
+
+Note, for DMA_MEMORY_IO returns, all subsequent memory returned by
+dma_alloc_coherent() may no longer be accessed directly, but instead
+must be accessed using the correct bus functions.  If your driver
+isn't prepared to handle this contingency, it should not specify
+DMA_MEMORY_IO in the input flags.
+
+As a simplification for the platforms, only *one* such region of
+memory may be declared per device.
+
+For reasons of efficiency, most platforms choose to track the declared
+region only at the granularity of a page.  For smaller allocations,
+you should use the dma_pool() API.
+
+void
+dma_release_declared_memory(struct device *dev)
+
+Remove the memory region previously declared from the system.  This
+API performs *no* in-use checking for this region and will return
+unconditionally having removed all the required structures.  It is the
+drivers job to ensure that no parts of this memory region are
+currently in use.
+
+void *
+dma_mark_declared_memory_occupied(struct device *dev,
+				  dma_addr_t device_addr, size_t size)
+
+This is used to occupy specific regions of the declared space
+(dma_alloc_coherent() will hand out the first free region it finds).
+
+device_addr is the *device* address of the region requested
+
+size is the size (and should be a page sized multiple).
+
+The return value will be either a pointer to the processor virtual
+address of the memory, or an error (via PTR_ERR()) if any part of the
+region is occupied.
+
 

arch/i386/kernel/pci-dma.c

@@ -13,17 +13,40 @@
 #include <linux/pci.h>
 #include <asm/io.h>
 
+struct dma_coherent_mem {
+	void		*virt_base;
+	u32		device_base;
+	int		size;
+	int		flags;
+	unsigned long	*bitmap;
+};
+
 void *dma_alloc_coherent(struct device *dev, size_t size,
 			   dma_addr_t *dma_handle, int gfp)
 {
 	void *ret;
+	struct dma_coherent_mem *mem = dev ? dev->dma_mem : NULL;
+	int order = get_order(size);
 	/* ignore region specifiers */
 	gfp &= ~(__GFP_DMA | __GFP_HIGHMEM);
 
+	if (mem) {
+		int page = bitmap_find_free_region(mem->bitmap, mem->size,
+						     order);
+		if (page >= 0) {
+			*dma_handle = mem->device_base + (page << PAGE_SHIFT);
+			ret = mem->virt_base + (page << PAGE_SHIFT);
+			memset(ret, 0, size);
+			return ret;
+		}
+		if (mem->flags & DMA_MEMORY_EXCLUSIVE)
+			return NULL;
+	}
+
 	if (dev == NULL || (dev->coherent_dma_mask < 0xffffffff))
 		gfp |= GFP_DMA;
 
-	ret = (void *)__get_free_pages(gfp, get_order(size));
+	ret = (void *)__get_free_pages(gfp, order);
 
 	if (ret != NULL) {
 		memset(ret, 0, size);
@@ -35,5 +58,89 @@ void *dma_alloc_coherent(struct device *dev, size_t size,
 void dma_free_coherent(struct device *dev, size_t size,
 			 void *vaddr, dma_addr_t dma_handle)
 {
-	free_pages((unsigned long)vaddr, get_order(size));
+	struct dma_coherent_mem *mem = dev->dma_mem;
+	int order = get_order(size);
+	
+	if (mem && vaddr >= mem->virt_base && vaddr < (mem->virt_base + (mem->size << PAGE_SHIFT))) {
+		int page = (vaddr - mem->virt_base) >> PAGE_SHIFT;
+
+		bitmap_release_region(mem->bitmap, page, order);
+	} else
+		free_pages((unsigned long)vaddr, order);
+}
+
+int dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
+				dma_addr_t device_addr, size_t size, int flags)
+{
+	void *mem_base;
+	int pages = size >> PAGE_SHIFT;
+	int bitmap_size = (pages + 31)/32;
+
+	if ((flags & (DMA_MEMORY_MAP | DMA_MEMORY_IO)) == 0)
+		goto out;
+	if (!size)
+		goto out;
+	if (dev->dma_mem)
+		goto out;
+
+	/* FIXME: this routine just ignores DMA_MEMORY_INCLUDES_CHILDREN */
+
+	mem_base = ioremap(bus_addr, size);
+	if (!mem_base)
+		goto out;
+
+	dev->dma_mem = kmalloc(GFP_KERNEL, sizeof(struct dma_coherent_mem));
+	if (!dev->dma_mem)
+		goto out;
+	memset(dev->dma_mem, 0, sizeof(struct dma_coherent_mem));
+	dev->dma_mem->bitmap = kmalloc(GFP_KERNEL, bitmap_size);
+	if (!dev->dma_mem->bitmap)
+		goto free1_out;
+	memset(dev->dma_mem->bitmap, 0, bitmap_size);
+
+	dev->dma_mem->virt_base = mem_base;
+	dev->dma_mem->device_base = device_addr;
+	dev->dma_mem->size = pages;
+	dev->dma_mem->flags = flags;
+
+	if (flags & DMA_MEMORY_MAP)
+		return DMA_MEMORY_MAP;
+
+	return DMA_MEMORY_IO;
+
+ free1_out:
+	kfree(dev->dma_mem->bitmap);
+ out:
+	return 0;
+}
+EXPORT_SYMBOL(dma_declare_coherent_memory);
+
+void dma_release_declared_memory(struct device *dev)
+{
+	struct dma_coherent_mem *mem = dev->dma_mem;
+	
+	if(!mem)
+		return;
+	dev->dma_mem = NULL;
+	kfree(mem->bitmap);
+	kfree(mem);
+}
+EXPORT_SYMBOL(dma_release_declared_memory);
+
+void *dma_mark_declared_memory_occupied(struct device *dev,
+					dma_addr_t device_addr, size_t size)
+{
+	struct dma_coherent_mem *mem = dev->dma_mem;
+	int pages = (size + (device_addr & ~PAGE_MASK) + PAGE_SIZE - 1) >> PAGE_SHIFT;
+	int pos, err;
+
+	if (!mem)
+		return ERR_PTR(-EINVAL);
+
+	pos = (device_addr - mem->device_base) >> PAGE_SHIFT;
+	err = bitmap_allocate_region(mem->bitmap, pos, get_order(pages));
+	if (err != 0)
+		return ERR_PTR(err);
+	return mem->virt_base + (pos << PAGE_SHIFT);
 }
+EXPORT_SYMBOL(dma_mark_declared_memory_occupied);

drivers/scsi/NCR_Q720.c

@@ -216,7 +216,21 @@ NCR_Q720_probe(struct device *dev)
 		goto out_free;
 	}
 	
-	mem_base = (__u32)ioremap(base_addr, mem_size);
+	if (dma_declare_coherent_memory(dev, base_addr, base_addr,
+					mem_size, DMA_MEMORY_MAP)
+	    != DMA_MEMORY_MAP) {
+		printk(KERN_ERR "NCR_Q720: DMA declare memory failed\n");
+		goto out_release_region;
+	}
+
+	/* The first 1k of the memory buffer is a memory map of the registers
+	 */
+	mem_base = (__u32)dma_mark_declared_memory_occupied(dev, base_addr,
+							    1024);
+	if (IS_ERR((void *)mem_base)) {
+		printk("NCR_Q720 failed to reserve memory mapped region\n");
+		goto out_release;
+	}
 
 	/* now also enable accesses in asr 2 */
 	asr2 = inb(io_base + 0x0a);
@@ -296,7 +310,8 @@ NCR_Q720_probe(struct device *dev)
 	return 0;
 
  out_release:
-	iounmap((void *)mem_base);
+	dma_release_declared_memory(dev);
+ out_release_region:
 	release_mem_region(base_addr, mem_size);
  out_free:
 	kfree(p);
@@ -321,7 +336,7 @@ NCR_Q720_remove(struct device *dev)
 		if(p->hosts[i])
 			NCR_Q720_remove_one(p->hosts[i]);
 
-	iounmap((void *)p->mem_base);
+	dma_release_declared_memory(dev);
 	release_mem_region(p->phys_mem_base, p->mem_size);
 	free_irq(p->irq, p);
 	kfree(p);

include/asm-i386/dma-mapping.h

@@ -163,4 +163,16 @@ dma_cache_sync(void *vaddr, size_t size,
 	flush_write_buffers();
 }
 
+#define ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY
+extern int
+dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
+			    dma_addr_t device_addr, size_t size, int flags);
+
+extern void
+dma_release_declared_memory(struct device *dev);
+
+extern void *
+dma_mark_declared_memory_occupied(struct device *dev,
+				  dma_addr_t device_addr, size_t size);
+
 #endif

include/linux/bitmap.h

@@ -98,6 +98,9 @@ extern int bitmap_scnprintf(char *buf, unsigned int len,
 			const unsigned long *src, int nbits);
 extern int bitmap_parse(const char __user *ubuf, unsigned int ulen,
 			unsigned long *dst, int nbits);
+extern int bitmap_find_free_region(unsigned long *bitmap, int bits, int order);
+extern void bitmap_release_region(unsigned long *bitmap, int pos, int order);
+extern int bitmap_allocate_region(unsigned long *bitmap, int pos, int order);
 
 #define BITMAP_LAST_WORD_MASK(nbits)					\
 (									\

include/linux/device.h

@@ -285,6 +285,9 @@ struct device {
 
 	struct list_head	dma_pools;	/* dma pools (if dma'ble) */
 
+	struct dma_coherent_mem	*dma_mem; /* internal for coherent mem
+					     override */
+
 	void	(*release)(struct device * dev);
 };
 

include/linux/dma-mapping.h

 #ifndef _ASM_LINUX_DMA_MAPPING_H
 #define _ASM_LINUX_DMA_MAPPING_H
 
+#include <linux/err.h>
+
 /* These definitions mirror those in pci.h, so they can be used
  * interchangeably with their PCI_ counterparts */
 enum dma_data_direction {
@@ -21,6 +23,33 @@ enum dma_data_direction {
 
 extern u64 dma_get_required_mask(struct device *dev);
 
+/* flags for the coherent memory api */
+#define	DMA_MEMORY_MAP			0x01
+#define DMA_MEMORY_IO			0x02
+#define DMA_MEMORY_INCLUDES_CHILDREN	0x04
+#define DMA_MEMORY_EXCLUSIVE		0x08
+
+#ifndef ARCH_HAS_DMA_DECLARE_COHERENT_MEMORY
+static inline int
+dma_declare_coherent_memory(struct device *dev, dma_addr_t bus_addr,
+			    dma_addr_t device_addr, size_t size, int flags)
+{
+	return 0;
+}
+
+static inline void
+dma_release_declared_memory(struct device *dev)
+{
+}
+
+static inline void *
+dma_mark_declared_memory_occupied(struct device *dev,
+				  dma_addr_t device_addr, size_t size)
+{
+	return ERR_PTR(-EBUSY);
+}
+#endif
+
 #endif
 
 

lib/bitmap.c

@@ -408,3 +408,85 @@ int bitmap_parse(const char __user *ubuf, unsigned int ubuflen,
 	return 0;
 }
 EXPORT_SYMBOL(bitmap_parse);
+
+/**
+ *	bitmap_find_free_region - find a contiguous aligned mem region
+ *	@bitmap: an array of unsigned longs corresponding to the bitmap
+ *	@bits: number of bits in the bitmap
+ *	@order: region size to find (size is actually 1<<order)
+ *
+ * This is used to allocate a memory region from a bitmap.  The idea is
+ * that the region has to be 1<<order sized and 1<<order aligned (this
+ * makes the search algorithm much faster).
+ *
+ * The region is marked as set bits in the bitmap if a free one is
+ * found.
+ *
+ * Returns either beginning of region or negative error
+ */
+int bitmap_find_free_region(unsigned long *bitmap, int bits, int order)
+{
+	unsigned long mask;
+	int pages = 1 << order;
+	int i;
+
+	if(pages > BITS_PER_LONG)
+		return -EINVAL;
+
+	/* make a mask of the order */
+	mask = (1ul << (pages - 1));
+	mask += mask - 1;
+
+	/* run up the bitmap pages bits at a time */
+	for (i = 0; i < bits; i += pages) {
+		int index = i/BITS_PER_LONG;
+		int offset = i - (index * BITS_PER_LONG);
+		if((bitmap[index] & (mask << offset)) == 0) {
+			/* set region in bimap */
+			bitmap[index] |= (mask << offset);
+			return i;
+		}
+	}
+	return -ENOMEM;
+}
+EXPORT_SYMBOL(bitmap_find_free_region);
+
+/**
+ *	bitmap_release_region - release allocated bitmap region
+ *	@bitmap: a pointer to the bitmap
+ *	@pos: the beginning of the region
+ *	@order: the order of the bits to release (number is 1<<order)
+ *
+ * This is the complement to __bitmap_find_free_region and releases
+ * the found region (by clearing it in the bitmap).
+ */
+void bitmap_release_region(unsigned long *bitmap, int pos, int order)
+{
+	int pages = 1 << order;
+	unsigned long mask = (1ul << (pages - 1));
+	int index = pos/BITS_PER_LONG;
+	int offset = pos - (index * BITS_PER_LONG);
+	mask += mask - 1;
+	bitmap[index] &= ~(mask << offset);
+}
+EXPORT_SYMBOL(bitmap_release_region);
+
+int bitmap_allocate_region(unsigned long *bitmap, int pos, int order)
+{
+	int pages = 1 << order;
+	unsigned long mask = (1ul << (pages - 1));
+	int index = pos/BITS_PER_LONG;
+	int offset = pos - (index * BITS_PER_LONG);
+
+	/* We don't do regions of pages > BITS_PER_LONG.  The
+	 * algorithm would be a simple look for multiple zeros in the
+	 * array, but there's no driver today that needs this.  If you
+	 * trip this BUG(), you get to code it... */
+	BUG_ON(pages > BITS_PER_LONG);
+	mask += mask - 1;
+	if (bitmap[index] & (mask << offset))
+		return -EBUSY;
+	bitmap[index] |= (mask << offset);
+	return 0;
+}
+EXPORT_SYMBOL(bitmap_allocate_region);

mm/vmalloc.c

@@ -179,11 +179,26 @@ int map_vm_area(struct vm_struct *area, pgprot_t prot, struct page ***pages)
 	return err;
 }
 
+#define IOREMAP_MAX_ORDER	(7 + PAGE_SHIFT)	/* 128 pages */
+
 struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags,
 				unsigned long start, unsigned long end)
 {
 	struct vm_struct **p, *tmp, *area;
-	unsigned long addr = start;
+	unsigned long align = 1;
+	unsigned long addr;
+
+	if (flags & VM_IOREMAP) {
+		int bit = fls(size);
+
+		if (bit > IOREMAP_MAX_ORDER)
+			bit = IOREMAP_MAX_ORDER;
+		else if (bit < PAGE_SHIFT)
+			bit = PAGE_SHIFT;
+
+		align = 1ul << bit;
+	}
+	addr = ALIGN(start, align);
 
 	area = kmalloc(sizeof(*area), GFP_KERNEL);
 	if (unlikely(!area))
@@ -200,13 +215,17 @@ struct vm_struct *__get_vm_area(unsigned long size, unsigned long flags,
 
 	write_lock(&vmlist_lock);
 	for (p = &vmlist; (tmp = *p) != NULL ;p = &tmp->next) {
-		if ((unsigned long)tmp->addr < addr)
+		if ((unsigned long)tmp->addr < addr) {
+			if((unsigned long)tmp->addr + tmp->size >= addr)
+				addr = ALIGN(tmp->size + 
+					     (unsigned long)tmp->addr, align);
 			continue;
+		}
 		if ((size + addr) < addr)
 			goto out;
 		if (size + addr <= (unsigned long)tmp->addr)
 			goto found;
-		addr = tmp->size + (unsigned long)tmp->addr;
+		addr = ALIGN(tmp->size + (unsigned long)tmp->addr, align);
 		if (addr > end - size)
 			goto out;
 	}