[PATCH] staging: Intel Restricted Access Region Handler

[Ossama Othman]

This is a resubmission of the Intel RAR Handler (memrar) driver that
add a TODO file, and also converts the Doxygen style comments to
kernel-doc style.

The Intel Restricted Access Region Handler provides a buffer allocation
mechanism to RAR users.  Since the intended usage model is to lock out
CPU access to RAR (the CPU will not be able to access RAR memory), this
driver does not access RAR memory, and merely keeps track of what areas
of RAR memory are in use.  It has it's own simple allocator that does
not rely on existing kernel allocators (SLAB, etc) since those
allocators are too tightly coupled with the paging mechanism, which isn't
needed for the intended RAR use cases.

An mmap() implementation is provided for debugging purposes to simplify
RAR memory access from the user space.  However, it will effectively be
a no-op when RAR access control is enabled since the CPU will not be
able to access RAR.

This driver should not be confused with the rar_register driver.  That
driver exposes an interface to access RAR registers on the Moorestown
platform.  The RAR handler driver relies on the rar_register driver for
low level RAR register reads and writes.

This patch was generated and built against the latest linux-2.6 master
branch.

Signed-off-by: Ossama Othman <ossama.othman@xxxxxxxxx>
---
 drivers/staging/Kconfig                   |    2 +
 drivers/staging/Makefile                  |    1 +
 drivers/staging/memrar/Kconfig            |   15 +
 drivers/staging/memrar/Makefile           |    2 +
 drivers/staging/memrar/TODO               |   43 ++
 drivers/staging/memrar/memrar.h           |  155 +++++
 drivers/staging/memrar/memrar_allocator.c |  432 +++++++++++++
 drivers/staging/memrar/memrar_allocator.h |  149 +++++
 drivers/staging/memrar/memrar_handler.c   |  937 +++++++++++++++++++++++++++++
 9 files changed, 1736 insertions(+), 0 deletions(-)
 create mode 100644 drivers/staging/memrar/Kconfig
 create mode 100644 drivers/staging/memrar/Makefile
 create mode 100644 drivers/staging/memrar/TODO
 create mode 100644 drivers/staging/memrar/memrar.h
 create mode 100644 drivers/staging/memrar/memrar_allocator.c
 create mode 100644 drivers/staging/memrar/memrar_allocator.h
 create mode 100644 drivers/staging/memrar/memrar_handler.c

diff --git a/drivers/staging/Kconfig b/drivers/staging/Kconfig
index 7696a66..a4b402e 100644
--- a/drivers/staging/Kconfig
+++ b/drivers/staging/Kconfig
@@ -111,6 +111,8 @@ source "drivers/staging/vme/Kconfig"
 
 source "drivers/staging/rar_register/Kconfig"
 
+source "drivers/staging/memrar/Kconfig"
+
 source "drivers/staging/sep/Kconfig"
 
 source "drivers/staging/iio/Kconfig"
diff --git a/drivers/staging/Makefile b/drivers/staging/Makefile
index ea2e70e..a517236 100644
--- a/drivers/staging/Makefile
+++ b/drivers/staging/Makefile
@@ -36,6 +36,7 @@ obj-$(CONFIG_FB_UDL)		+= udlfb/
 obj-$(CONFIG_HYPERV)		+= hv/
 obj-$(CONFIG_VME_BUS)		+= vme/
 obj-$(CONFIG_RAR_REGISTER)	+= rar_register/
+obj-$(CONFIG_MRST_RAR_HANDLER)	+= memrar/
 obj-$(CONFIG_DX_SEP)		+= sep/
 obj-$(CONFIG_IIO)		+= iio/
 obj-$(CONFIG_RAMZSWAP)		+= ramzswap/
diff --git a/drivers/staging/memrar/Kconfig b/drivers/staging/memrar/Kconfig
new file mode 100644
index 0000000..a5598a8
--- /dev/null
+++ b/drivers/staging/memrar/Kconfig
@@ -0,0 +1,15 @@
+config MRST_RAR_HANDLER
+	tristate "RAR handler driver for Intel Moorestown platform"
+	select RAR_REGISTER
+	---help---
+	  This driver provides a memory management interface to
+	  restricted access regions (RAR) available on the Intel
+	  Moorestown platform.
+
+	  Once locked down, restricted access regions are only
+	  accessible by specific hardware on the platform.  The x86
+	  CPU is typically not one of those platforms.  As such this
+	  driver does not access RAR, and only provides a buffer
+	  allocation/bookkeeping mechanism.
+
+	  If unsure, say N.
diff --git a/drivers/staging/memrar/Makefile b/drivers/staging/memrar/Makefile
new file mode 100644
index 0000000..a3336c0
--- /dev/null
+++ b/drivers/staging/memrar/Makefile
@@ -0,0 +1,2 @@
+obj-$(CONFIG_MRST_RAR_HANDLER)	+= memrar.o
+memrar-y			:= memrar_allocator.o memrar_handler.o
diff --git a/drivers/staging/memrar/TODO b/drivers/staging/memrar/TODO
new file mode 100644
index 0000000..0087447
--- /dev/null
+++ b/drivers/staging/memrar/TODO
@@ -0,0 +1,43 @@
+RAR Handler (memrar) Driver TODO Items
+======================================
+
+Maintainer: Ossama Othman <ossama.othman@xxxxxxxxx>
+
+memrar.h
+--------
+1. This header exposes the driver's user space and kernel space
+   interfaces.  It should be moved to <linux/rar/memrar.h>, or
+   something along those lines, when this memrar driver is moved out
+   of `staging'.
+     a. It would be ideal if staging/rar_register/rar_register.h was
+        moved to the same directory.
+
+memrar_allocator.[ch]
+---------------------
+1. Address potential fragmentation issues with the memrar_allocator.
+
+2. Hide struct memrar_allocator details/fields.  They need not be
+   exposed to the user.
+     a. Forward declare struct memrar_allocator.
+     b. Move all three struct definitions to `memrar_allocator.c'
+        source file.
+     c. Add a memrar_allocator_largest_free_area() function, or
+        something like that to get access to the value of the struct
+        memrar_allocator "largest_free_area" field.  This allows the
+        struct memrar_allocator fields to be completely hidden from
+        the user.  The memrar_handler code really only needs this for
+        statistic gathering on-demand.
+     d. Do the same for the "capacity" field as the
+        "largest_free_area" field.
+
+3. Move memrar_allocator.* to kernel `lib' directory since it is HW
+   neutral.
+     a. Alternatively, use lib/genalloc.c instead.
+     b. A kernel port of Doug Lea's malloc() implementation may also
+        be an option.
+
+memrar_handler.c
+----------------
+1. Split user space interface (ioctl code) from core/kernel code,
+   e.g.:
+     memrar_handler.c -> memrar_core.c, memrar_user.c
diff --git a/drivers/staging/memrar/memrar.h b/drivers/staging/memrar/memrar.h
new file mode 100644
index 0000000..0b735b8
--- /dev/null
+++ b/drivers/staging/memrar/memrar.h
@@ -0,0 +1,155 @@
+/*
+ *      RAR Handler (/dev/memrar) internal driver API.
+ *      Copyright (C) 2010 Intel Corporation. All rights reserved.
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of version 2 of the GNU General
+ *      Public License as published by the Free Software Foundation.
+ *
+ *      This program is distributed in the hope that it will be
+ *      useful, but WITHOUT ANY WARRANTY; without even the implied
+ *      warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ *      PURPOSE.  See the GNU General Public License for more details.
+ *      You should have received a copy of the GNU General Public
+ *      License along with this program; if not, write to the Free
+ *      Software Foundation, Inc., 59 Temple Place - Suite 330,
+ *      Boston, MA  02111-1307, USA.
+ *      The full GNU General Public License is included in this
+ *      distribution in the file called COPYING.
+ */
+
+
+#ifndef _MEMRAR_H
+#define _MEMRAR_H
+
+#include <linux/ioctl.h>
+#include <linux/types.h>
+
+
+/**
+ * struct RAR_stat - RAR statistics structure
+ * @type:		Type of RAR memory (e.g., audio vs. video)
+ * @capacity:		Total size of RAR memory region.
+ * @largest_block_size:	Size of the largest reservable block.
+ *
+ * This structure is used for RAR_HANDLER_STAT ioctl and for the
+ * RAR_get_stat() user space wrapper function.
+ */
+struct RAR_stat {
+	__u32 type;
+	__u32 capacity;
+	__u32 largest_block_size;
+};
+
+
+/**
+ * struct RAR_block_info - user space struct that describes RAR buffer
+ * @type:	Type of RAR memory (e.g., audio vs. video)
+ * @size:	Requested size of a block to be reserved in RAR.
+ * @handle:	Handle that can be used to refer to reserved block.
+ *
+ * This is the basic structure exposed to the user space that
+ * describes a given RAR buffer.  The buffer's underlying bus address
+ * is not exposed to the user.  User space code refers to the buffer
+ * entirely by "handle".
+ */
+struct RAR_block_info {
+	__u32 type;
+	__u32 size;
+	__u32 handle;
+};
+
+
+#define RAR_IOCTL_BASE 0xE0
+
+/* Reserve RAR block. */
+#define RAR_HANDLER_RESERVE _IOWR(RAR_IOCTL_BASE, 0x00, struct RAR_block_info)
+
+/* Release previously reserved RAR block. */
+#define RAR_HANDLER_RELEASE _IOW(RAR_IOCTL_BASE, 0x01, __u32)
+
+/* Get RAR stats. */
+#define RAR_HANDLER_STAT    _IOWR(RAR_IOCTL_BASE, 0x02, struct RAR_stat)
+
+
+#ifdef __KERNEL__
+
+/* -------------------------------------------------------------- */
+/*               Kernel Side RAR Handler Interface                */
+/* -------------------------------------------------------------- */
+
+/**
+ * struct RAR_buffer - kernel space struct that describes RAR buffer
+ * @info:		structure containing base RAR buffer information
+ * @bus_address:	buffer bus address
+ *
+ * Structure that contains all information related to a given block of
+ * memory in RAR.  It is generally only used when retrieving RAR
+ * related bus addresses.
+ *
+ * Note: This structure is used only by RAR-enabled drivers, and is
+ *       not intended to be exposed to the user space.
+ */
+struct RAR_buffer {
+	struct RAR_block_info info;
+	dma_addr_t bus_address;
+};
+
+/**
+ * rar_reserve() - reserve RAR buffers
+ * @buffers:	array of RAR_buffers where type and size of buffers to
+ *		reserve are passed in, handle and bus address are
+ *		passed out
+ * @count:	number of RAR_buffers in the "buffers" array
+ *
+ * This function will reserve buffers in the restricted access regions
+ * of given types.
+ *
+ * It returns the number of successfully reserved buffers.  Successful
+ * buffer reservations will have the corresponding bus_address field
+ * set to a non-zero value in the given buffers vector.
+ */
+extern size_t rar_reserve(struct RAR_buffer *buffers,
+			  size_t count);
+
+/**
+ * rar_release() - release RAR buffers
+ * @buffers:	array of RAR_buffers where handles to buffers to be
+ *		released are passed in
+ * @count:	number of RAR_buffers in the "buffers" array
+ *
+ * This function will release RAR buffers that were retrieved through
+ * a call to rar_reserve() or rar_handle_to_bus() by decrementing the
+ * reference count.  The RAR buffer will be reclaimed when the
+ * reference count drops to zero.
+ *
+ * It returns the number of successfully released buffers.  Successful
+ * releases will have their handle field set to zero in the given
+ * buffers vector.
+ */
+extern size_t rar_release(struct RAR_buffer *buffers,
+			  size_t count);
+
+/**
+ * rar_handle_to_bus() - convert a vector of RAR handles to bus addresses
+ * @buffers:	array of RAR_buffers containing handles to be
+ *		converted to bus_addresses
+ * @count:	number of RAR_buffers in the "buffers" array
+
+ * This function will retrieve the RAR buffer bus addresses, type and
+ * size corresponding to the RAR handles provided in the buffers
+ * vector.
+ *
+ * It returns the number of successfully converted buffers.  The bus
+ * address will be set to 0 for unrecognized handles.
+ *
+ * The reference count for each corresponding buffer in RAR will be
+ * incremented.  Call rar_release() when done with the buffers.
+ */
+extern size_t rar_handle_to_bus(struct RAR_buffer *buffers,
+				size_t count);
+
+
+#endif  /* __KERNEL__ */
+
+#endif  /* _MEMRAR_H */
diff --git a/drivers/staging/memrar/memrar_allocator.c b/drivers/staging/memrar/memrar_allocator.c
new file mode 100644
index 0000000..a4f8c58
--- /dev/null
+++ b/drivers/staging/memrar/memrar_allocator.c
@@ -0,0 +1,432 @@
+/*
+ *      memrar_allocator 1.0:  An allocator for Intel RAR.
+ *
+ *      Copyright (C) 2010 Intel Corporation. All rights reserved.
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of version 2 of the GNU General
+ *      Public License as published by the Free Software Foundation.
+ *
+ *      This program is distributed in the hope that it will be
+ *      useful, but WITHOUT ANY WARRANTY; without even the implied
+ *      warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ *      PURPOSE.  See the GNU General Public License for more details.
+ *      You should have received a copy of the GNU General Public
+ *      License along with this program; if not, write to the Free
+ *      Software Foundation, Inc., 59 Temple Place - Suite 330,
+ *      Boston, MA  02111-1307, USA.
+ *      The full GNU General Public License is included in this
+ *      distribution in the file called COPYING.
+ *
+ *
+ *  ------------------------------------------------------------------
+ *
+ *      This simple allocator implementation provides a
+ *      malloc()/free()-like interface for reserving space within a
+ *      previously reserved block of memory.  It is not specific to
+ *      any hardware, nor is it coupled with the lower level paging
+ *      mechanism.
+ *
+ *      The primary goal of this implementation is to provide a means
+ *      to partition an arbitrary block of memory without actually
+ *      accessing the memory or incurring any hardware side-effects
+ *      (e.g. paging).  It is, in effect, a bookkeeping mechanism for
+ *      buffers.
+ */
+
+
+#include "memrar_allocator.h"
+#include <linux/slab.h>
+#include <linux/bug.h>
+#include <linux/kernel.h>
+
+
+struct memrar_allocator *memrar_create_allocator(unsigned long base,
+						 size_t capacity,
+						 size_t block_size)
+{
+	struct memrar_allocator *allocator  = NULL;
+	struct memrar_address_ranges *first_node = NULL;
+
+	/*
+	 * Make sure the base address is aligned on a block_size
+	 * boundary.
+	 *
+	 * @todo Is this necessary?
+	 */
+	/* base = ALIGN(base, block_size); */
+
+	/* Validate parameters.
+	 *
+	 * Make sure we can allocate the entire memory space.  Zero
+	 * capacity or block size are obviously invalid.
+	 */
+	if (base == 0
+	    || capacity == 0
+	    || block_size == 0
+	    || ULONG_MAX - capacity < base
+	    || capacity < block_size)
+		return allocator;
+
+	/*
+	 * There isn't much point in creating a memory allocator that
+	 * is only capable of holding one block but we'll allow it,
+	 * and issue a diagnostic.
+	 */
+	WARN(capacity < block_size * 2,
+	     "memrar: Only one block available to allocator.\n");
+
+	allocator = kmalloc(sizeof(*allocator), GFP_KERNEL);
+
+	if (allocator == NULL)
+		return allocator;
+
+	mutex_init(&allocator->lock);
+	allocator->base = base;
+
+	/* Round the capacity down to a multiple of block_size. */
+	allocator->capacity = (capacity / block_size) * block_size;
+
+	allocator->block_size = block_size;
+
+	allocator->largest_free_area = allocator->capacity;
+
+	/* Initialize the handle and free lists. */
+	INIT_LIST_HEAD(&allocator->allocated_list.list);
+	INIT_LIST_HEAD(&allocator->free_list.list);
+
+	first_node = kmalloc(sizeof(*first_node), GFP_KERNEL);
+	if (first_node == NULL)	{
+		kfree(allocator);
+		allocator = NULL;
+	} else {
+		/* Full range of blocks is available. */
+		first_node->range.begin = base;
+		first_node->range.end   = base + allocator->capacity;
+		list_add(&first_node->list,
+			 &allocator->free_list.list);
+	}
+
+	return allocator;
+}
+
+void memrar_destroy_allocator(struct memrar_allocator *allocator)
+{
+	/*
+	 * Assume that the memory allocator lock isn't held at this
+	 * point in time.  Caller must ensure that.
+	 */
+
+	struct memrar_address_ranges *pos = NULL;
+	struct memrar_address_ranges *n   = NULL;
+
+	if (allocator == NULL)
+		return;
+
+	mutex_lock(&allocator->lock);
+
+	/* Reclaim free list resources. */
+	list_for_each_entry_safe(pos,
+				 n,
+				 &allocator->free_list.list,
+				 list) {
+		list_del(&pos->list);
+		kfree(pos);
+	}
+
+	mutex_unlock(&allocator->lock);
+
+	kfree(allocator);
+}
+
+unsigned long memrar_allocator_alloc(struct memrar_allocator *allocator,
+				     size_t size)
+{
+	struct memrar_address_ranges *pos = NULL;
+
+	size_t num_blocks;
+	unsigned long reserved_bytes;
+
+	/*
+	 * Address of allocated buffer.  We assume that zero is not a
+	 * valid address.
+	 */
+	unsigned long addr = 0;
+
+	if (allocator == NULL || size == 0)
+		return addr;
+
+	/* Reserve enough blocks to hold the amount of bytes requested. */
+	num_blocks = DIV_ROUND_UP(size, allocator->block_size);
+
+	reserved_bytes = num_blocks * allocator->block_size;
+
+	mutex_lock(&allocator->lock);
+
+	if (reserved_bytes > allocator->largest_free_area) {
+		mutex_unlock(&allocator->lock);
+		return addr;
+	}
+
+	/*
+	 * Iterate through the free list to find a suitably sized
+	 * range of free contiguous memory blocks.
+	 *
+	 * We also take the opportunity to reset the size of the
+	 * largest free area size statistic.
+	 */
+	list_for_each_entry(pos, &allocator->free_list.list, list) {
+		struct memrar_address_range * const fr = &pos->range;
+		size_t const curr_size = fr->end - fr->begin;
+
+		if (curr_size >= reserved_bytes && addr == 0) {
+			struct memrar_address_range *range = NULL;
+			struct memrar_address_ranges * const new_node =
+				kmalloc(sizeof(*new_node), GFP_KERNEL);
+
+			if (new_node == NULL)
+				break;
+
+			list_add(&new_node->list,
+				 &allocator->allocated_list.list);
+
+			/*
+			 * Carve out area of memory from end of free
+			 * range.
+			 */
+			range        = &new_node->range;
+			range->end   = fr->end;
+			fr->end     -= reserved_bytes;
+			range->begin = fr->end;
+			addr         = range->begin;
+
+			/*
+			 * Check if largest area has decreased in
+			 * size.  We'll need to continue scanning for
+			 * the next largest area if it has.
+			 */
+			if (curr_size == allocator->largest_free_area)
+				allocator->largest_free_area -=
+					reserved_bytes;
+			else
+				break;
+		}
+
+		/*
+		 * Reset largest free area size statistic as needed,
+		 * but only if we've actually allocated memory.
+		 */
+		if (addr != 0
+		    && curr_size > allocator->largest_free_area) {
+			allocator->largest_free_area = curr_size;
+			break;
+		}
+	}
+
+	mutex_unlock(&allocator->lock);
+
+	return addr;
+}
+
+long memrar_allocator_free(struct memrar_allocator *allocator,
+			   unsigned long addr)
+{
+	struct list_head *pos = NULL;
+	struct list_head *tmp = NULL;
+	struct list_head *dst = NULL;
+
+	struct memrar_address_ranges      *allocated = NULL;
+	struct memrar_address_range const *handle    = NULL;
+
+	unsigned long old_end        = 0;
+	unsigned long new_chunk_size = 0;
+
+	if (allocator == NULL)
+		return -EINVAL;
+
+	if (addr == 0)
+		return 0;  /* Ignore "free(0)". */
+
+	mutex_lock(&allocator->lock);
+
+	/* Find the corresponding handle. */
+	list_for_each_entry(allocated,
+			    &allocator->allocated_list.list,
+			    list) {
+		if (allocated->range.begin == addr) {
+			handle = &allocated->range;
+			break;
+		}
+	}
+
+	/* No such buffer created by this allocator. */
+	if (handle == NULL) {
+		mutex_unlock(&allocator->lock);
+		return -EFAULT;
+	}
+
+	/*
+	 * Coalesce adjacent chunks of memory if possible.
+	 *
+	 * @note This isn't full blown coalescing since we're only
+	 *       coalescing at most three chunks of memory.
+	 */
+	list_for_each_safe(pos, tmp, &allocator->free_list.list) {
+		/* @todo O(n) performance.  Optimize. */
+
+		struct memrar_address_range * const chunk =
+			&list_entry(pos,
+				    struct memrar_address_ranges,
+				    list)->range;
+
+		/* Extend size of existing free adjacent chunk. */
+		if (chunk->end == handle->begin) {
+			/*
+			 * Chunk "less than" than the one we're
+			 * freeing is adjacent.
+			 *
+			 * Before:
+			 *
+			 *   +-----+------+
+			 *   |chunk|handle|
+			 *   +-----+------+
+			 *
+			 * After:
+			 *
+			 *   +------------+
+			 *   |   chunk    |
+			 *   +------------+
+			 */
+
+			struct memrar_address_ranges const * const next =
+				list_entry(pos->next,
+					   struct memrar_address_ranges,
+					   list);
+
+			chunk->end = handle->end;
+
+			/*
+			 * Now check if next free chunk is adjacent to
+			 * the current extended free chunk.
+			 *
+			 * Before:
+			 *
+			 *   +------------+----+
+			 *   |   chunk    |next|
+			 *   +------------+----+
+			 *
+			 * After:
+			 *
+			 *   +-----------------+
+			 *   |      chunk      |
+			 *   +-----------------+
+			 */
+			if (!list_is_singular(pos)
+			    && chunk->end == next->range.begin) {
+				chunk->end = next->range.end;
+				list_del(pos->next);
+				kfree(next);
+			}
+
+			list_del(&allocated->list);
+
+			new_chunk_size = chunk->end - chunk->begin;
+
+			goto exit_memrar_free;
+
+		} else if (handle->end == chunk->begin) {
+			/*
+			 * Chunk "greater than" than the one we're
+			 * freeing is adjacent.
+			 *
+			 *   +------+-----+
+			 *   |handle|chunk|
+			 *   +------+-----+
+			 *
+			 * After:
+			 *
+			 *   +------------+
+			 *   |   chunk    |
+			 *   +------------+
+			 */
+
+			struct memrar_address_ranges const * const prev =
+				list_entry(pos->prev,
+					   struct memrar_address_ranges,
+					   list);
+
+			chunk->begin = handle->begin;
+
+			/*
+			 * Now check if previous free chunk is
+			 * adjacent to the current extended free
+			 * chunk.
+			 *
+			 *
+			 * Before:
+			 *
+			 *   +----+------------+
+			 *   |prev|   chunk    |
+			 *   +----+------------+
+			 *
+			 * After:
+			 *
+			 *   +-----------------+
+			 *   |      chunk      |
+			 *   +-----------------+
+			 */
+			if (!list_is_singular(pos)
+			    && prev->range.end == chunk->begin) {
+				chunk->begin = prev->range.begin;
+				list_del(pos->prev);
+				kfree(prev);
+			}
+
+			list_del(&allocated->list);
+
+			new_chunk_size = chunk->end - chunk->begin;
+
+			goto exit_memrar_free;
+
+		} else if (chunk->end < handle->begin
+			   && chunk->end > old_end) {
+			/* Keep track of where the entry could be
+			 * potentially moved from the "allocated" list
+			 * to the "free" list if coalescing doesn't
+			 * occur, making sure the "free" list remains
+			 * sorted.
+			 */
+			old_end = chunk->end;
+			dst = pos;
+		}
+	}
+
+	/*
+	 * Nothing to coalesce.
+	 *
+	 * Move the entry from the "allocated" list to the "free"
+	 * list.
+	 */
+	list_move(&allocated->list, dst);
+	new_chunk_size = handle->end - handle->begin;
+	allocated = NULL;
+
+exit_memrar_free:
+
+	if (new_chunk_size > allocator->largest_free_area)
+		allocator->largest_free_area = new_chunk_size;
+
+	mutex_unlock(&allocator->lock);
+
+	kfree(allocated);
+
+	return 0;
+}
+
+
+
+/*
+  Local Variables:
+    c-file-style: "linux"
+  End:
+*/
diff --git a/drivers/staging/memrar/memrar_allocator.h b/drivers/staging/memrar/memrar_allocator.h
new file mode 100644
index 0000000..0b80dea
--- /dev/null
+++ b/drivers/staging/memrar/memrar_allocator.h
@@ -0,0 +1,149 @@
+/*
+ *      Copyright (C) 2010 Intel Corporation. All rights reserved.
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of version 2 of the GNU General
+ *      Public License as published by the Free Software Foundation.
+ *
+ *      This program is distributed in the hope that it will be
+ *      useful, but WITHOUT ANY WARRANTY; without even the implied
+ *      warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ *      PURPOSE.  See the GNU General Public License for more details.
+ *      You should have received a copy of the GNU General Public
+ *      License along with this program; if not, write to the Free
+ *      Software Foundation, Inc., 59 Temple Place - Suite 330,
+ *      Boston, MA  02111-1307, USA.
+ *      The full GNU General Public License is included in this
+ *      distribution in the file called COPYING.
+ */
+
+#ifndef MEMRAR_ALLOCATOR_H
+#define MEMRAR_ALLOCATOR_H
+
+
+#include <linux/mutex.h>
+#include <linux/list.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+
+
+/**
+ * struct memrar_address_range - struct that describes a memory range
+ * @begin:	Beginning of available address range.
+ * @end:	End of available address range, one past the end,
+ *		i.e. [begin, end).
+ */
+struct memrar_address_range {
+/* private: internal use only */
+	unsigned long begin;
+	unsigned long end;
+};
+
+/**
+ * struct memrar_address_ranges - list of areas of memory.
+ * @list:	Linked list of address ranges.
+ * @range:	Memory address range corresponding to given list node.
+ */
+struct memrar_address_ranges {
+/* private: internal use only */
+	struct list_head list;
+	struct memrar_address_range range;
+};
+
+/**
+ * struct memrar_allocator - encapsulation of the memory allocator state
+ * @lock:		Lock used to synchronize access to the memory
+ *			allocator state.
+ * @base:		Base (start) address of the allocator memory
+ *			space.
+ * @capacity:		Size of the allocator memory space in bytes.
+ * @block_size:		The size in bytes of individual blocks within
+ *			the allocator memory space.
+ * @largest_free_area:	Largest free area of memory in the allocator
+ *			in bytes.
+ * @allocated_list:	List of allocated memory block address
+ *			ranges.
+ * @free_list:		List of free address ranges.
+ *
+ * This structure contains all memory allocator state, including the
+ * base address, capacity, free list, lock, etc.
+ */
+struct memrar_allocator {
+/* private: internal use only */
+	struct mutex lock;
+	unsigned long base;
+	size_t capacity;
+	size_t block_size;
+	size_t largest_free_area;
+	struct memrar_address_ranges allocated_list;
+	struct memrar_address_ranges free_list;
+};
+
+/**
+ * memrar_create_allocator() - create a memory allocator
+ * @base:	Address at which the memory allocator begins.
+ * @capacity:	Desired size of the memory allocator.  This value must
+ *		be larger than the block_size, ideally more than twice
+ *		as large since there wouldn't be much point in using a
+ *		memory allocator otherwise.
+ * @block_size:	The size of individual blocks within the memory
+ *		allocator.  This value must smaller than the
+ *		capacity.
+ *
+ * Create a memory allocator with the given capacity and block size.
+ * The capacity will be reduced to be a multiple of the block size, if
+ * necessary.
+ *
+ * Returns an instance of the memory allocator, if creation succeeds,
+ * otherwise zero if creation fails.  Failure may occur if not enough
+ * kernel memory exists to create the memrar_allocator instance
+ * itself, or if the capacity and block_size arguments are not
+ * compatible or make sense.
+ */
+struct memrar_allocator *memrar_create_allocator(unsigned long base,
+						 size_t capacity,
+						 size_t block_size);
+
+/**
+ * memrar_destroy_allocator() - destroy allocator
+ * @allocator:	The allocator being destroyed.
+ *
+ * Reclaim resources held by the memory allocator.  The caller must
+ * explicitly free all memory reserved by memrar_allocator_alloc()
+ * prior to calling this function.  Otherwise leaks will occur.
+ */
+void memrar_destroy_allocator(struct memrar_allocator *allocator);
+
+/**
+ * memrar_allocator_alloc() - reserve an area of memory of given size
+ * @allocator:	The allocator instance being used to reserve buffer.
+ * @size:	The size in bytes of the buffer to allocate.
+ *
+ * This functions reserves an area of memory managed by the given
+ * allocator.  It returns zero if allocation was not possible.
+ * Failure may occur if the allocator no longer has space available.
+ */
+unsigned long memrar_allocator_alloc(struct memrar_allocator *allocator,
+				     size_t size);
+
+/**
+ * memrar_allocator_free() - release buffer starting at given address
+ * @allocator:	The allocator instance being used to release the buffer.
+ * @address:	The address of the buffer being released.
+ *
+ * Release an area of memory starting at the given address.  Failure
+ * could occur if the given address is not in the address space
+ * managed by the allocator.  Returns zero on success or an errno
+ * (negative value) on failure.
+ */
+long memrar_allocator_free(struct memrar_allocator *allocator,
+			   unsigned long address);
+
+#endif  /* MEMRAR_ALLOCATOR_H */
+
+
+/*
+  Local Variables:
+    c-file-style: "linux"
+  End:
+*/
diff --git a/drivers/staging/memrar/memrar_handler.c b/drivers/staging/memrar/memrar_handler.c
new file mode 100644
index 0000000..4bbf66f
--- /dev/null
+++ b/drivers/staging/memrar/memrar_handler.c
@@ -0,0 +1,937 @@
+/*
+ *      memrar_handler 1.0:  An Intel restricted access region handler device
+ *
+ *      Copyright (C) 2010 Intel Corporation. All rights reserved.
+ *
+ *      This program is free software; you can redistribute it and/or
+ *      modify it under the terms of version 2 of the GNU General
+ *      Public License as published by the Free Software Foundation.
+ *
+ *      This program is distributed in the hope that it will be
+ *      useful, but WITHOUT ANY WARRANTY; without even the implied
+ *      warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR
+ *      PURPOSE.  See the GNU General Public License for more details.
+ *      You should have received a copy of the GNU General Public
+ *      License along with this program; if not, write to the Free
+ *      Software Foundation, Inc., 59 Temple Place - Suite 330,
+ *      Boston, MA  02111-1307, USA.
+ *      The full GNU General Public License is included in this
+ *      distribution in the file called COPYING.
+ *
+ * -------------------------------------------------------------------
+ *
+ *      Moorestown restricted access regions (RAR) provide isolated
+ *      areas of main memory that are only acceessible by authorized
+ *      devices.
+ *
+ *      The Intel Moorestown RAR handler module exposes a kernel space
+ *      RAR memory management mechanism.  It is essentially a
+ *      RAR-specific allocator.
+ *
+ *      Besides providing RAR buffer management, the RAR handler also
+ *      behaves in many ways like an OS virtual memory manager.  For
+ *      example, the RAR "handles" created by the RAR handler are
+ *      analogous to user space virtual addresses.
+ *
+ *      RAR memory itself is never accessed directly by the RAR
+ *      handler.
+ */
+
+#include <linux/miscdevice.h>
+#include <linux/fs.h>
+#include <linux/slab.h>
+#include <linux/kref.h>
+#include <linux/mutex.h>
+#include <linux/kernel.h>
+#include <linux/uaccess.h>
+#include <linux/mm.h>
+#include <linux/ioport.h>
+#include <linux/io.h>
+
+#include "../rar_register/rar_register.h"
+
+#include "memrar.h"
+#include "memrar_allocator.h"
+
+
+#define MEMRAR_VER "1.0"
+
+/*
+ * Moorestown supports three restricted access regions.
+ *
+ * We only care about the first two, video and audio.  The third,
+ * reserved for Chaabi and the P-unit, will be handled by their
+ * respective drivers.
+ */
+#define MRST_NUM_RAR 2
+
+/* ---------------- -------------------- ------------------- */
+
+/**
+ * struct memrar_buffer_info - struct that keeps track of all RAR buffers
+ * @list:	Linked list of memrar_buffer_info objects.
+ * @buffer:	Core RAR buffer information.
+ * @refcount:	Reference count.
+ * @owner:	File handle corresponding to process that reserved the
+ *		block of memory in RAR.  This will be zero for buffers
+ *		allocated by other drivers instead of by a user space
+ *		process.
+ *
+ * This structure encapsulates a link list of RAR buffers, as well as
+ * other characteristics specific to a given list node, such as the
+ * reference count on the corresponding RAR buffer.
+ */
+struct memrar_buffer_info {
+	struct list_head list;
+	struct RAR_buffer buffer;
+	struct kref refcount;
+	struct file *owner;
+};
+
+/**
+ * struct memrar_rar_info - characteristics of a given RAR
+ * @base:	Base bus address of the RAR.
+ * @length:	Length of the RAR.
+ * @iobase:	Virtual address of RAR mapped into kernel.
+ * @allocator:	Allocator associated with the RAR.  Note the allocator
+ *		"capacity" may be smaller than the RAR length if the
+ *		length is not a multiple of the configured allocator
+ *		block size.
+ * @buffers:	Table that keeps track of all reserved RAR buffers.
+ * @lock:	Lock used to synchronize access to RAR-specific data
+ *		structures.
+ *
+ * Each RAR has an associated memrar_rar_info structure that describes
+ * where in memory the RAR is located, how large it is, and a list of
+ * reserved RAR buffers inside that RAR.  Each RAR also has a mutex
+ * associated with it to reduce lock contention when operations on
+ * multiple RARs are performed in parallel.
+ */
+struct memrar_rar_info {
+	dma_addr_t base;
+	unsigned long length;
+	void __iomem *iobase;
+	struct memrar_allocator *allocator;
+	struct memrar_buffer_info buffers;
+	struct mutex lock;
+};
+
+/*
+ * Array of RAR characteristics.
+ */
+static struct memrar_rar_info memrars[MRST_NUM_RAR];
+
+/* ---------------- -------------------- ------------------- */
+
+/* Validate RAR type. */
+static inline int memrar_is_valid_rar_type(u32 type)
+{
+	return type == RAR_TYPE_VIDEO || type == RAR_TYPE_AUDIO;
+}
+
+/* Check if an address/handle falls with the given RAR memory range. */
+static inline int memrar_handle_in_range(struct memrar_rar_info *rar,
+					 u32 vaddr)
+{
+	unsigned long const iobase = (unsigned long) (rar->iobase);
+	return (vaddr >= iobase && vaddr < iobase + rar->length);
+}
+
+/* Retrieve RAR information associated with the given handle. */
+static struct memrar_rar_info *memrar_get_rar_info(u32 vaddr)
+{
+	int i;
+	for (i = 0; i < MRST_NUM_RAR; ++i) {
+		struct memrar_rar_info * const rar = &memrars[i];
+		if (memrar_handle_in_range(rar, vaddr))
+			return rar;
+	}
+
+	return NULL;
+}
+
+/*
+ * Retrieve bus address from given handle.
+ *
+ * Returns address corresponding to given handle.  Zero if handle is
+ * invalid.
+ */
+static dma_addr_t memrar_get_bus_address(
+	struct memrar_rar_info *rar,
+	u32 vaddr)
+{
+	unsigned long const iobase = (unsigned long) (rar->iobase);
+
+	if (!memrar_handle_in_range(rar, vaddr))
+		return 0;
+
+	/*
+	 * An assumption is made that the virtual address offset is
+	 * the same as the bus address offset, at least based on the
+	 * way this driver is implemented.  For example, vaddr + 2 ==
+	 * baddr + 2.
+	 *
+	 * @todo Is that a valid assumption?
+	 */
+	return rar->base + (vaddr - iobase);
+}
+
+/*
+ * Retrieve physical address from given handle.
+ *
+ * Returns address corresponding to given handle.  Zero if handle is
+ * invalid.
+ */
+static dma_addr_t memrar_get_physical_address(
+	struct memrar_rar_info *rar,
+	u32 vaddr)
+{
+	/*
+	 * @todo This assumes that the bus address and physical
+	 *       address are the same.  That is true for Moorestown
+	 *       but not necessarily on other platforms.  This
+	 *       deficiency should be addressed at some point.
+	 */
+	return memrar_get_bus_address(rar, vaddr);
+}
+
+/*
+ * Core block release code.
+ *
+ * Note: This code removes the node from a list.  Make sure any list
+ *       iteration is performed using list_for_each_safe().
+ */
+static void memrar_release_block_i(struct kref *ref)
+{
+	/*
+	 * Last reference is being released.  Remove from the table,
+	 * and reclaim resources.
+	 */
+
+	struct memrar_buffer_info * const node =
+		container_of(ref, struct memrar_buffer_info, refcount);
+
+	struct RAR_block_info * const user_info =
+		&node->buffer.info;
+
+	struct memrar_allocator * const allocator =
+		memrars[user_info->type].allocator;
+
+	list_del(&node->list);
+
+	memrar_allocator_free(allocator, user_info->handle);
+
+	kfree(node);
+}
+
+/*
+ * Initialize RAR parameters, such as bus addresses, etc.
+ */
+static int memrar_init_rar_resources(char const *devname)
+{
+	/* ---- Sanity Checks ----
+	 * 1. RAR bus addresses in both Lincroft and Langwell RAR
+	 *    registers should be the same.
+	 *    a. There's no way we can do this through IA.
+	 *
+	 * 2. Secure device ID in Langwell RAR registers should be set
+	 *    appropriately, e.g. only LPE DMA for the audio RAR, and
+	 *    security for the other Langwell based RAR registers.
+	 *    a. There's no way we can do this through IA.
+	 *
+	 * 3. Audio and video RAR registers and RAR access should be
+	 *    locked down.  If not, enable RAR access control.  Except
+	 *    for debugging purposes, there is no reason for them to
+	 *    be unlocked.
+	 *    a.  We can only do this for the Lincroft (IA) side.
+	 *
+	 * @todo Should the RAR handler driver even be aware of audio
+	 *       and video RAR settings?
+	 */
+
+	/*
+	 * RAR buffer block size.
+	 *
+	 * We choose it to be the size of a page to simplify the
+	 * /dev/memrar mmap() implementation and usage.  Otherwise
+	 * paging is not involved once an RAR is locked down.
+	 */
+	static size_t const RAR_BLOCK_SIZE = PAGE_SIZE;
+
+	int z;
+	int found_rar = 0;
+
+	BUG_ON(MRST_NUM_RAR != ARRAY_SIZE(memrars));
+
+	for (z = 0; z != MRST_NUM_RAR; ++z) {
+		dma_addr_t low, high;
+		struct memrar_rar_info * const rar = &memrars[z];
+
+		BUG_ON(!memrar_is_valid_rar_type(z));
+
+		mutex_init(&rar->lock);
+
+		/*
+		 * Initialize the process table before we reach any
+		 * code that exit on failure since the finalization
+		 * code requires an initialized list.
+		 */
+		INIT_LIST_HEAD(&rar->buffers.list);
+
+		if (rar_get_address(z, &low, &high) != 0) {
+			/* No RAR is available. */
+			break;
+		} else if (low == 0 || high == 0) {
+			/*
+			 * We don't immediately break out of the loop
+			 * since the next type of RAR may be enabled.
+			 */
+			rar->base      = 0;
+			rar->length    = 0;
+			rar->iobase    = NULL;
+			rar->allocator = NULL;
+			continue;
+		}
+
+		/*
+		 * @todo Verify that LNC and LNW RAR register contents
+		 *       addresses, security, etc are compatible and
+		 *       consistent).
+		 */
+
+		rar->length = high - low + 1;
+
+		/* Claim RAR memory as our own. */
+		if (request_mem_region(low, rar->length, devname) == NULL) {
+			rar->length = 0;
+
+			pr_err("%s: Unable to claim RAR[%d] memory.\n",
+			       devname,
+			       z);
+			pr_err("%s: RAR[%d] disabled.\n", devname, z);
+
+			/*
+			 * Rather than break out of the loop by
+			 * returning -EBUSY, for example, we may be
+			 * able to claim memory of the next RAR region
+			 * as our own.
+			 */
+			continue;
+		}
+
+		rar->base = low;
+
+		/*
+		 * Now map it into the kernel address space.
+		 *
+		 * Note that the RAR memory may only be accessed by IA
+		 * when debugging.  Otherwise attempts to access the
+		 * RAR memory when it is locked down will result in
+		 * behavior similar to writing to /dev/null and
+		 * reading from /dev/zero.  This behavior is enforced
+		 * by the hardware.  Even if we don't access the
+		 * memory, mapping it into the kernel provides us with
+		 * a convenient RAR handle to bus address mapping.
+		 */
+		rar->iobase = ioremap_nocache(rar->base, rar->length);
+		if (rar->iobase == NULL) {
+			pr_err("%s: Unable to map RAR memory.\n",
+			       devname);
+			return -ENOMEM;
+		}
+
+		/* Initialize corresponding memory allocator. */
+		rar->allocator = memrar_create_allocator(
+			(unsigned long) rar->iobase,
+			rar->length,
+			RAR_BLOCK_SIZE);
+		if (rar->allocator == NULL)
+			return -1;
+
+		/*
+		 * -------------------------------------------------
+		 * Make sure all RARs handled by us are locked down.
+		 * -------------------------------------------------
+		 */
+
+		/* Enable RAR protection on the Lincroft side. */
+		if (0) {
+			/*
+			 * This is mostly a sanity check since the
+			 * vendor should have locked down RAR in the
+			 * SMIP header RAR configuration.
+			 */
+			rar_lock(z);
+		} else {
+			pr_warning("%s: LNC RAR[%d] no lock sanity check.\n",
+				   devname,
+				   z);
+		}
+
+		/* ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ */
+		/* |||||||||||||||||||||||||||||||||||||||||||||||||| */
+
+		/*
+		 * It would be nice if we could verify that RAR
+		 * protection on the Langwell side is enabled, but
+		 * there is no way to do that from here.  The
+		 * necessary Langwell RAR registers are not accessible
+		 * from the Lincroft (IA) side.
+		 *
+		 * Hopefully the ODM did the right thing and enabled
+		 * Langwell side RAR protection in the integrated
+		 * firmware SMIP header.
+		 */
+
+		pr_info("%s: BRAR[%d] bus address range = "
+			"[0x%lx, 0x%lx]\n",
+			devname,
+			z,
+			(unsigned long) low,
+			(unsigned long) high);
+
+		pr_info("%s: BRAR[%d] size = %u KiB\n",
+			devname,
+			z,
+			rar->allocator->capacity / 1024);
+
+		found_rar = 1;
+	}
+
+	if (!found_rar)	{
+		/*
+		 * No RAR support.  Don't bother continuing.
+		 *
+		 * Note that this is not a failure.
+		 */
+		pr_info("%s: No Moorestown RAR support available.\n",
+			devname);
+		return -ENODEV;
+	}
+
+	return 0;
+}
+
+/*
+ * Finalize RAR resources.
+ */
+static void memrar_fini_rar_resources(void)
+{
+	int z;
+	struct memrar_buffer_info *pos;
+	struct memrar_buffer_info *tmp;
+
+	/*
+	 * @todo Do we need to hold a lock at this point in time?
+	 *       (module initialization failure or exit?)
+	 */
+
+	for (z = MRST_NUM_RAR; z-- != 0; ) {
+		struct memrar_rar_info * const rar = &memrars[z];
+
+		/* Clean up remaining resources. */
+
+		list_for_each_entry_safe(pos,
+					 tmp,
+					 &rar->buffers.list,
+					 list) {
+			kref_put(&pos->refcount, memrar_release_block_i);
+		}
+
+		memrar_destroy_allocator(rar->allocator);
+		rar->allocator = NULL;
+
+		iounmap(rar->iobase);
+		rar->iobase = NULL;
+
+		release_mem_region(rar->base, rar->length);
+		rar->base = 0;
+
+		rar->length = 0;
+	}
+}
+
+static long memrar_reserve_block(struct RAR_buffer *request,
+				 struct file *filp)
+{
+	struct RAR_block_info * const rinfo = &request->info;
+	struct RAR_buffer *buffer;
+	struct memrar_buffer_info *buffer_info;
+	u32 handle;
+	struct memrar_rar_info *rar = NULL;
+
+	/* Prevent array overflow. */
+	if (!memrar_is_valid_rar_type(rinfo->type))
+		return -EINVAL;
+
+	rar = &memrars[rinfo->type];
+
+	/* Reserve memory in RAR. */
+	handle = memrar_allocator_alloc(rar->allocator, rinfo->size);
+	if (handle == 0)
+		return -ENOMEM;
+
+	buffer_info = kmalloc(sizeof(*buffer_info), GFP_KERNEL);
+
+	if (buffer_info == NULL) {
+		memrar_allocator_free(rar->allocator, handle);
+		return -ENOMEM;
+	}
+
+	buffer = &buffer_info->buffer;
+	buffer->info.type = rinfo->type;
+	buffer->info.size = rinfo->size;
+
+	/* Memory handle corresponding to the bus address. */
+	buffer->info.handle = handle;
+	buffer->bus_address = memrar_get_bus_address(rar, handle);
+
+	/*
+	 * Keep track of owner so that we can later cleanup if
+	 * necessary.
+	 */
+	buffer_info->owner = filp;
+
+	kref_init(&buffer_info->refcount);
+
+	mutex_lock(&rar->lock);
+	list_add(&buffer_info->list, &rar->buffers.list);
+	mutex_unlock(&rar->lock);
+
+	rinfo->handle = buffer->info.handle;
+	request->bus_address = buffer->bus_address;
+
+	return 0;
+}
+
+static long memrar_release_block(u32 addr)
+{
+	struct memrar_buffer_info *pos;
+	struct memrar_buffer_info *tmp;
+	struct memrar_rar_info * const rar = memrar_get_rar_info(addr);
+	long result = -EINVAL;
+
+	if (rar == NULL)
+		return -EFAULT;
+
+	mutex_lock(&rar->lock);
+
+	/*
+	 * Iterate through the buffer list to find the corresponding
+	 * buffer to be released.
+	 */
+	list_for_each_entry_safe(pos,
+				 tmp,
+				 &rar->buffers.list,
+				 list) {
+		struct RAR_block_info * const info =
+			&pos->buffer.info;
+
+		/*
+		 * Take into account handle offsets that may have been
+		 * added to the base handle, such as in the following
+		 * scenario:
+		 *
+		 *     u32 handle = base + offset;
+		 *     rar_handle_to_bus(handle);
+		 *     rar_release(handle);
+		 */
+		if (addr >= info->handle
+		    && addr < (info->handle + info->size)
+		    && memrar_is_valid_rar_type(info->type)) {
+			kref_put(&pos->refcount, memrar_release_block_i);
+			result = 0;
+			break;
+		}
+	}
+
+	mutex_unlock(&rar->lock);
+
+	return result;
+}
+
+static long memrar_get_stat(struct RAR_stat *r)
+{
+	long result = -EINVAL;
+
+	if (likely(r != NULL) && memrar_is_valid_rar_type(r->type)) {
+		struct memrar_allocator * const allocator =
+			memrars[r->type].allocator;
+
+		BUG_ON(allocator == NULL);
+
+		/*
+		 * Allocator capacity doesn't change over time.  No
+		 * need to synchronize.
+		 */
+		r->capacity = allocator->capacity;
+
+		mutex_lock(&allocator->lock);
+
+		r->largest_block_size = allocator->largest_free_area;
+
+		mutex_unlock(&allocator->lock);
+
+		result = 0;
+	}
+
+	return result;
+}
+
+static long memrar_ioctl(struct file *filp,
+			 unsigned int cmd,
+			 unsigned long arg)
+{
+	void __user *argp = (void __user *)arg;
+	long result = 0;
+
+	struct RAR_buffer buffer;
+	struct RAR_block_info * const request = &buffer.info;
+	struct RAR_stat rar_info;
+	u32 rar_handle;
+
+	switch (cmd) {
+	case RAR_HANDLER_RESERVE:
+		if (copy_from_user(request,
+				   argp,
+				   sizeof(*request)))
+			return -EFAULT;
+
+		result = memrar_reserve_block(&buffer, filp);
+		if (result != 0)
+			return result;
+
+		return copy_to_user(argp, request, sizeof(*request));
+
+	case RAR_HANDLER_RELEASE:
+		if (copy_from_user(&rar_handle,
+				   argp,
+				   sizeof(rar_handle)))
+			return -EFAULT;
+
+		return memrar_release_block(rar_handle);
+
+	case RAR_HANDLER_STAT:
+		if (copy_from_user(&rar_info,
+				   argp,
+				   sizeof(rar_info)))
+			return -EFAULT;
+
+		/*
+		 * Populate the RAR_stat structure based on the RAR
+		 * type given by the user
+		 */
+		if (memrar_get_stat(&rar_info) != 0)
+			return -EINVAL;
+
+		/*
+		 * @todo Do we need to verify destination pointer
+		 *       "argp" is non-zero?  Is that already done by
+		 *       copy_to_user()?
+		 */
+		return copy_to_user(argp,
+				    &rar_info,
+				    sizeof(rar_info)) ? -EFAULT : 0;
+
+	default:
+		return -ENOTTY;
+	}
+
+	return 0;
+}
+
+static int memrar_mmap(struct file *filp, struct vm_area_struct *vma)
+{
+	/*
+	 * This mmap() implementation is predominantly useful for
+	 * debugging since the CPU will be prevented from accessing
+	 * RAR memory by the hardware when RAR is properly locked
+	 * down.
+	 *
+	 * In order for this implementation to be useful RAR memory
+	 * must be not be locked down.  However, we only want to do
+	 * that when debugging.  DO NOT leave RAR memory unlocked in a
+	 * deployed device that utilizes RAR.
+	 */
+
+	size_t const size = vma->vm_end - vma->vm_start;
+
+	/* Users pass the RAR handle as the mmap() offset parameter. */
+	unsigned long const handle = vma->vm_pgoff << PAGE_SHIFT;
+
+	struct memrar_rar_info * const rar = memrar_get_rar_info(handle);
+
+	unsigned long pfn;
+
+	/* Invalid RAR handle or size passed to mmap(). */
+	if (rar == NULL
+	    || handle == 0
+	    || size > (handle - (unsigned long) rar->iobase))
+		return -EINVAL;
+
+	/*
+	 * Retrieve physical address corresponding to the RAR handle,
+	 * and convert it to a page frame.
+	 */
+	pfn = memrar_get_physical_address(rar, handle) >> PAGE_SHIFT;
+
+
+	pr_debug("memrar: mapping RAR range [0x%lx, 0x%lx) into user space.\n",
+		 handle,
+		 handle + size);
+
+	/*
+	 * Map RAR memory into user space.  This is really only useful
+	 * for debugging purposes since the memory won't be
+	 * accessible, i.e. reads return zero and writes are ignored,
+	 * when RAR access control is enabled.
+	 */
+	if (remap_pfn_range(vma,
+			    vma->vm_start,
+			    pfn,
+			    size,
+			    vma->vm_page_prot))
+		return -EAGAIN;
+
+	/* vma->vm_ops = &memrar_mem_ops; */
+
+	return 0;
+}
+
+static int memrar_open(struct inode *inode, struct file *filp)
+{
+	/* Nothing to do yet. */
+
+	return 0;
+}
+
+static int memrar_release(struct inode *inode, struct file *filp)
+{
+	/* Free all regions associated with the given file handle. */
+
+	struct memrar_buffer_info *pos;
+	struct memrar_buffer_info *tmp;
+	int z;
+
+	for (z = 0; z != MRST_NUM_RAR; ++z) {
+		struct memrar_rar_info * const rar = &memrars[z];
+
+		mutex_lock(&rar->lock);
+
+		list_for_each_entry_safe(pos,
+					 tmp,
+					 &rar->buffers.list,
+					 list) {
+			if (filp == pos->owner)
+				kref_put(&pos->refcount,
+					 memrar_release_block_i);
+		}
+
+		mutex_unlock(&rar->lock);
+	}
+
+	return 0;
+}
+
+/*
+ * This function is part of the kernel space memrar driver API.
+ */
+size_t rar_reserve(struct RAR_buffer *buffers, size_t count)
+{
+	struct RAR_buffer * const end =
+		(buffers == NULL ? buffers : buffers + count);
+	struct RAR_buffer *i;
+
+	size_t reserve_count = 0;
+
+	for (i = buffers; i != end; ++i) {
+		if (memrar_reserve_block(i, NULL) == 0)
+			++reserve_count;
+		else
+			i->bus_address = 0;
+	}
+
+	return reserve_count;
+}
+EXPORT_SYMBOL(rar_reserve);
+
+/*
+ * This function is part of the kernel space memrar driver API.
+ */
+size_t rar_release(struct RAR_buffer *buffers, size_t count)
+{
+	struct RAR_buffer * const end =
+		(buffers == NULL ? buffers : buffers + count);
+	struct RAR_buffer *i;
+
+	size_t release_count = 0;
+
+	for (i = buffers; i != end; ++i) {
+		u32 * const handle = &i->info.handle;
+		if (memrar_release_block(*handle) == 0) {
+			/*
+			 * @todo We assume we should do this each time
+			 *       the ref count is decremented.  Should
+			 *       we instead only do this when the ref
+			 *       count has dropped to zero, and the
+			 *       buffer has been completely
+			 *       released/unmapped?
+			 */
+			*handle = 0;
+			++release_count;
+		}
+	}
+
+	return release_count;
+}
+EXPORT_SYMBOL(rar_release);
+
+/*
+ * This function is part of the kernel space driver API.
+ */
+size_t rar_handle_to_bus(struct RAR_buffer *buffers, size_t count)
+{
+	struct RAR_buffer * const end =
+		(buffers == NULL ? buffers : buffers + count);
+	struct RAR_buffer *i;
+	struct memrar_buffer_info *pos;
+
+	size_t conversion_count = 0;
+
+	/*
+	 * Find all bus addresses corresponding to the given handles.
+	 *
+	 * @todo Not liking this nested loop.  Optimize.
+	 */
+	for (i = buffers; i != end; ++i) {
+		struct memrar_rar_info * const rar =
+			memrar_get_rar_info(i->info.handle);
+
+		/*
+		 * Check if we have a bogus handle, and then continue
+		 * with remaining buffers.
+		 */
+		if (rar == NULL) {
+			i->bus_address = 0;
+			continue;
+		}
+
+		mutex_lock(&rar->lock);
+
+		list_for_each_entry(pos, &rar->buffers.list, list) {
+			struct RAR_block_info * const user_info =
+				&pos->buffer.info;
+
+			/*
+			 * Take into account handle offsets that may
+			 * have been added to the base handle, such as
+			 * in the following scenario:
+			 *
+			 *     u32 handle = base + offset;
+			 *     rar_handle_to_bus(handle);
+			 */
+
+			if (i->info.handle >= user_info->handle
+			    && i->info.handle < (user_info->handle
+						 + user_info->size)) {
+				u32 const offset =
+					i->info.handle - user_info->handle;
+
+				i->info.type = user_info->type;
+				i->info.size = user_info->size - offset;
+				i->bus_address =
+					pos->buffer.bus_address
+					+ offset;
+
+				/* Increment the reference count. */
+				kref_get(&pos->refcount);
+
+				++conversion_count;
+				break;
+			} else {
+				i->bus_address = 0;
+			}
+		}
+
+		mutex_unlock(&rar->lock);
+	}
+
+	return conversion_count;
+}
+EXPORT_SYMBOL(rar_handle_to_bus);
+
+static const struct file_operations memrar_fops = {
+	.owner = THIS_MODULE,
+	.unlocked_ioctl = memrar_ioctl,
+	.mmap           = memrar_mmap,
+	.open           = memrar_open,
+	.release        = memrar_release,
+};
+
+static struct miscdevice memrar_miscdev = {
+	.minor = MISC_DYNAMIC_MINOR,    /* dynamic allocation */
+	.name = "memrar",               /* /dev/memrar */
+	.fops = &memrar_fops
+};
+
+static char const banner[] __initdata =
+	KERN_INFO
+	"Intel RAR Handler: " MEMRAR_VER " initialized.\n";
+
+static int memrar_registration_callback(void *ctx)
+{
+	/*
+	 * We initialize the RAR parameters early on so that we can
+	 * discontinue memrar device initialization and registration
+	 * if suitably configured RARs are not available.
+	 */
+	int result = memrar_init_rar_resources(memrar_miscdev.name);
+
+	if (result != 0)
+		return result;
+
+	result = misc_register(&memrar_miscdev);
+
+	if (result != 0) {
+		pr_err("%s: misc_register() failed.\n",
+			memrar_miscdev.name);
+
+		/* Clean up resources previously reserved. */
+		memrar_fini_rar_resources();
+	}
+
+	return result;
+}
+
+static int __init memrar_init(void)
+{
+	printk(banner);
+
+	return register_rar(&memrar_registration_callback, 0);
+}
+
+static void __exit memrar_exit(void)
+{
+	memrar_fini_rar_resources();
+
+	misc_deregister(&memrar_miscdev);
+}
+
+
+module_init(memrar_init);
+module_exit(memrar_exit);
+
+
+MODULE_AUTHOR("Ossama Othman <ossama.othman@xxxxxxxxx>");
+MODULE_DESCRIPTION("Intel Restricted Access Region Handler");
+MODULE_LICENSE("GPL");
+MODULE_ALIAS_MISCDEV(MISC_DYNAMIC_MINOR);
+MODULE_VERSION(MEMRAR_VER);
+
+
+
+/*
+  Local Variables:
+    c-file-style: "linux"
+  End:
+*/
-- 
1.6.3.3

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/