Merge branch 'for-linus' of git://git.monstr.eu/linux-2.6-microblaze

* 'for-linus' of git://git.monstr.eu/linux-2.6-microblaze: (27 commits)
  microblaze: entry.S use delay slot for return handlers
  microblaze: Save current task directly
  microblaze: Simplify entry.S - save/restore r3/r4 - ret_from_trap
  microblaze: PCI early support for noMMU system
  microblaze: Fix dma alloc and free coherent dma functions
  microblaze: Add consistent code
  microblaze: pgtable.h: move consistent functions
  microblaze: Remove ancient Kconfig option for consistent mapping
  microblaze: Remove VMALLOC_VMADDR
  microblaze: Add define for ASM_LOOP
  microblaze: Preliminary support for dma drivers
  microblaze: remove trailing space in messages
  microblaze: Use generic show_mem()
  microblaze: Change temp register for cmdline
  microblaze: Preliminary support for dma drivers
  microblaze: Move cache function to cache.c
  microblaze: Add support from PREEMPT
  microblaze: Add support for Xilinx PCI host bridge
  microblaze: Enable PCI, missing files
  microblaze: Add core PCI files
  ...

arch/microblaze/Kconfig

@@ -14,6 +14,8 @@ config MICROBLAZE
 	select USB_ARCH_HAS_EHCI
 	select ARCH_WANT_OPTIONAL_GPIOLIB
 	select HAVE_OPROFILE
+	select HAVE_DMA_ATTRS
+	select HAVE_DMA_API_DEBUG
 	select TRACING_SUPPORT
 
 config SWAP
@@ -76,9 +78,6 @@ config HAVE_LATENCYTOP_SUPPORT
 config PCI
 	def_bool n
 
-config NO_DMA
-	def_bool y
-
 config DTC
 	def_bool y
 
@@ -146,7 +145,6 @@ menu "Advanced setup"
 
 config ADVANCED_OPTIONS
 	bool "Prompt for advanced kernel configuration options"
-	depends on MMU
 	help
 	  This option will enable prompting for a variety of advanced kernel
 	  configuration options.  These options can cause the kernel to not
@@ -158,6 +156,15 @@ config ADVANCED_OPTIONS
 comment "Default settings for advanced configuration options are used"
 	depends on !ADVANCED_OPTIONS
 
+config XILINX_UNCACHED_SHADOW
+	bool "Are you using uncached shadow for RAM ?"
+	depends on ADVANCED_OPTIONS && !MMU
+	default n
+	help
+	  This is needed to be able to allocate uncachable memory regions.
+	  The feature requires the design to define the RAM memory controller
+	  window to be twice as large as the actual physical memory.
+
 config HIGHMEM_START_BOOL
 	bool "Set high memory pool address"
 	depends on ADVANCED_OPTIONS && HIGHMEM
@@ -175,7 +182,7 @@ config HIGHMEM_START
 
 config LOWMEM_SIZE_BOOL
 	bool "Set maximum low memory"
-	depends on ADVANCED_OPTIONS
+	depends on ADVANCED_OPTIONS && MMU
 	help
 	  This option allows you to set the maximum amount of memory which
 	  will be used as "low memory", that is, memory which the kernel can
@@ -187,7 +194,6 @@ config LOWMEM_SIZE_BOOL
 
 config LOWMEM_SIZE
 	hex "Maximum low memory size (in bytes)" if LOWMEM_SIZE_BOOL
-	depends on MMU
 	default "0x30000000"
 
 config KERNEL_START_BOOL
@@ -208,7 +214,7 @@ config KERNEL_START
 
 config TASK_SIZE_BOOL
 	bool "Set custom user task size"
-	depends on ADVANCED_OPTIONS
+	depends on ADVANCED_OPTIONS && MMU
 	help
 	  This option allows you to set the amount of virtual address space
 	  allocated to user tasks.  This can be useful in optimizing the
@@ -218,42 +224,34 @@ config TASK_SIZE_BOOL
 
 config TASK_SIZE
 	hex "Size of user task space" if TASK_SIZE_BOOL
-	depends on MMU
 	default "0x80000000"
 
-config CONSISTENT_START_BOOL
-	bool "Set custom consistent memory pool address"
-	depends on ADVANCED_OPTIONS && NOT_COHERENT_CACHE
-	help
-	  This option allows you to set the base virtual address
-	  of the the consistent memory pool.  This pool of virtual
-	  memory is used to make consistent memory allocations.
+endmenu
 
-config CONSISTENT_START
-	hex "Base virtual address of consistent memory pool" if CONSISTENT_START_BOOL
-	depends on MMU
-	default "0xff100000" if NOT_COHERENT_CACHE
+source "mm/Kconfig"
 
-config CONSISTENT_SIZE_BOOL
-	bool "Set custom consistent memory pool size"
-	depends on ADVANCED_OPTIONS && NOT_COHERENT_CACHE
-	help
-	  This option allows you to set the size of the the
-	  consistent memory pool.  This pool of virtual memory
-	  is used to make consistent memory allocations.
+menu "Exectuable file formats"
 
-config CONSISTENT_SIZE
-	hex "Size of consistent memory pool" if CONSISTENT_SIZE_BOOL
-	depends on MMU
-	default "0x00200000" if NOT_COHERENT_CACHE
+source "fs/Kconfig.binfmt"
 
 endmenu
 
-source "mm/Kconfig"
+menu "Bus Options"
 
-menu "Exectuable file formats"
+config PCI
+	bool "PCI support"
 
-source "fs/Kconfig.binfmt"
+config PCI_DOMAINS
+	def_bool PCI
+
+config PCI_SYSCALL
+	def_bool PCI
+
+config PCI_XILINX
+	bool "Xilinx PCI host bridge support"
+	depends on PCI
+
+source "drivers/pci/Kconfig"
 
 endmenu
 

arch/microblaze/Makefile

@@ -50,6 +50,7 @@ libs-y += $(LIBGCC)
 core-y += arch/microblaze/kernel/
 core-y += arch/microblaze/mm/
 core-y += arch/microblaze/platform/
+core-$(CONFIG_PCI) += arch/microblaze/pci/
 
 drivers-$(CONFIG_OPROFILE) += arch/microblaze/oprofile/
 

arch/microblaze/include/asm/device.h

@@ -14,6 +14,10 @@ struct device_node;
 struct dev_archdata {
 	/* Optional pointer to an OF device node */
 	struct device_node	*of_node;
+
+	/* DMA operations on that device */
+	struct dma_map_ops	*dma_ops;
+	void                    *dma_data;
 };
 
 struct pdev_archdata {

arch/microblaze/include/asm/dma-mapping.h

-#include <asm-generic/dma-mapping-broken.h>
+/*
+ * Implements the generic device dma API for microblaze and the pci
+ *
+ * Copyright (C) 2009-2010 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2009-2010 PetaLogix
+ *
+ * This file is subject to the terms and conditions of the GNU General
+ * Public License. See the file COPYING in the main directory of this
+ * archive for more details.
+ *
+ * This file is base on powerpc and x86 dma-mapping.h versions
+ * Copyright (C) 2004 IBM
+ */
+
+#ifndef _ASM_MICROBLAZE_DMA_MAPPING_H
+#define _ASM_MICROBLAZE_DMA_MAPPING_H
+
+/*
+ * See Documentation/PCI/PCI-DMA-mapping.txt and
+ * Documentation/DMA-API.txt for documentation.
+ */
+
+#include <linux/types.h>
+#include <linux/cache.h>
+#include <linux/mm.h>
+#include <linux/scatterlist.h>
+#include <linux/dma-debug.h>
+#include <linux/dma-attrs.h>
+#include <asm/io.h>
+#include <asm-generic/dma-coherent.h>
+
+#define DMA_ERROR_CODE		(~(dma_addr_t)0x0)
+
+#define __dma_alloc_coherent(dev, gfp, size, handle)	NULL
+#define __dma_free_coherent(size, addr)		((void)0)
+#define __dma_sync(addr, size, rw)		((void)0)
+
+static inline unsigned long device_to_mask(struct device *dev)
+{
+	if (dev->dma_mask && *dev->dma_mask)
+		return *dev->dma_mask;
+	/* Assume devices without mask can take 32 bit addresses */
+	return 0xfffffffful;
+}
+
+extern struct dma_map_ops *dma_ops;
+
+/*
+ * Available generic sets of operations
+ */
+extern struct dma_map_ops dma_direct_ops;
+
+static inline struct dma_map_ops *get_dma_ops(struct device *dev)
+{
+	/* We don't handle the NULL dev case for ISA for now. We could
+	 * do it via an out of line call but it is not needed for now. The
+	 * only ISA DMA device we support is the floppy and we have a hack
+	 * in the floppy driver directly to get a device for us.
+	 */
+	if (unlikely(!dev) || !dev->archdata.dma_ops)
+		return NULL;
+
+	return dev->archdata.dma_ops;
+}
+
+static inline void set_dma_ops(struct device *dev, struct dma_map_ops *ops)
+{
+	dev->archdata.dma_ops = ops;
+}
+
+static inline int dma_supported(struct device *dev, u64 mask)
+{
+	struct dma_map_ops *ops = get_dma_ops(dev);
+
+	if (unlikely(!ops))
+		return 0;
+	if (!ops->dma_supported)
+		return 1;
+	return ops->dma_supported(dev, mask);
+}
+
+#ifdef CONFIG_PCI
+/* We have our own implementation of pci_set_dma_mask() */
+#define HAVE_ARCH_PCI_SET_DMA_MASK
+
+#endif
+
+static inline int dma_set_mask(struct device *dev, u64 dma_mask)
+{
+	struct dma_map_ops *ops = get_dma_ops(dev);
+
+	if (unlikely(ops == NULL))
+		return -EIO;
+	if (ops->set_dma_mask)
+		return ops->set_dma_mask(dev, dma_mask);
+	if (!dev->dma_mask || !dma_supported(dev, dma_mask))
+		return -EIO;
+	*dev->dma_mask = dma_mask;
+	return 0;
+}
+
+#include <asm-generic/dma-mapping-common.h>
+
+static inline int dma_mapping_error(struct device *dev, dma_addr_t dma_addr)
+{
+	struct dma_map_ops *ops = get_dma_ops(dev);
+	if (ops->mapping_error)
+		return ops->mapping_error(dev, dma_addr);
+
+	return (dma_addr == DMA_ERROR_CODE);
+}
+
+#define dma_alloc_noncoherent(d, s, h, f) dma_alloc_coherent(d, s, h, f)
+#define dma_free_noncoherent(d, s, v, h) dma_free_coherent(d, s, v, h)
+#define dma_is_consistent(d, h)	(1)
+
+static inline void *dma_alloc_coherent(struct device *dev, size_t size,
+					dma_addr_t *dma_handle, gfp_t flag)
+{
+	struct dma_map_ops *ops = get_dma_ops(dev);
+	void *memory;
+
+	BUG_ON(!ops);
+
+	memory = ops->alloc_coherent(dev, size, dma_handle, flag);
+
+	debug_dma_alloc_coherent(dev, size, *dma_handle, memory);
+	return memory;
+}
+
+static inline void dma_free_coherent(struct device *dev, size_t size,
+				     void *cpu_addr, dma_addr_t dma_handle)
+{
+	struct dma_map_ops *ops = get_dma_ops(dev);
+
+	BUG_ON(!ops);
+	debug_dma_free_coherent(dev, size, cpu_addr, dma_handle);
+	ops->free_coherent(dev, size, cpu_addr, dma_handle);
+}
+
+static inline int dma_get_cache_alignment(void)
+{
+	return L1_CACHE_BYTES;
+}
+
+static inline void dma_cache_sync(struct device *dev, void *vaddr, size_t size,
+		enum dma_data_direction direction)
+{
+	BUG_ON(direction == DMA_NONE);
+	__dma_sync(vaddr, size, (int)direction);
+}
+
+#endif	/* _ASM_MICROBLAZE_DMA_MAPPING_H */

arch/microblaze/include/asm/io.h

@@ -15,7 +15,23 @@
 #include <asm/page.h>
 #include <linux/types.h>
 #include <linux/mm.h>          /* Get struct page {...} */
+#include <asm-generic/iomap.h>
 
+#ifndef CONFIG_PCI
+#define _IO_BASE	0
+#define _ISA_MEM_BASE	0
+#define PCI_DRAM_OFFSET	0
+#else
+#define _IO_BASE	isa_io_base
+#define _ISA_MEM_BASE	isa_mem_base
+#define PCI_DRAM_OFFSET	pci_dram_offset
+#endif
+
+extern unsigned long isa_io_base;
+extern unsigned long pci_io_base;
+extern unsigned long pci_dram_offset;
+
+extern resource_size_t isa_mem_base;
 
 #define IO_SPACE_LIMIT (0xFFFFFFFF)
 
@@ -124,9 +140,6 @@ static inline void writel(unsigned int v, volatile void __iomem *addr)
 #define virt_to_phys(addr)	((unsigned long)__virt_to_phys(addr))
 #define virt_to_bus(addr)	((unsigned long)__virt_to_phys(addr))
 
-#define __page_address(page) \
-		(PAGE_OFFSET + (((page) - mem_map) << PAGE_SHIFT))
-#define page_to_phys(page)	virt_to_phys((void *)__page_address(page))
 #define page_to_bus(page)	(page_to_phys(page))
 #define bus_to_virt(addr)	(phys_to_virt(addr))
 
@@ -227,15 +240,7 @@ static inline void __iomem *__ioremap(phys_addr_t address, unsigned long size,
 #define out_8(a, v) __raw_writeb((v), (a))
 #define in_8(a) __raw_readb(a)
 
-/* FIXME */
-static inline void __iomem *ioport_map(unsigned long port, unsigned int len)
-{
-	return (void __iomem *) (port);
-}
-
-static inline void ioport_unmap(void __iomem *addr)
-{
-	/* Nothing to do */
-}
+#define ioport_map(port, nr)	((void __iomem *)(port))
+#define ioport_unmap(addr)
 
 #endif /* _ASM_MICROBLAZE_IO_H */

arch/microblaze/include/asm/irq.h

@@ -14,6 +14,12 @@
 
 #include <linux/interrupt.h>
 
+/* This type is the placeholder for a hardware interrupt number. It has to
+ * be big enough to enclose whatever representation is used by a given
+ * platform.
+ */
+typedef unsigned long irq_hw_number_t;
+
 extern unsigned int nr_irq;
 
 #define NO_IRQ (-1)
@@ -21,7 +27,8 @@ extern unsigned int nr_irq;
 struct pt_regs;
 extern void do_IRQ(struct pt_regs *regs);
 
-/* irq_of_parse_and_map - Parse and Map an interrupt into linux virq space
+/**
+ * irq_of_parse_and_map - Parse and Map an interrupt into linux virq space
  * @device: Device node of the device whose interrupt is to be mapped
  * @index: Index of the interrupt to map
  *
@@ -40,4 +47,32 @@ static inline void irq_dispose_mapping(unsigned int virq)
 	return;
 }
 
+struct irq_host;
+
+/**
+ * irq_create_mapping - Map a hardware interrupt into linux virq space
+ * @host: host owning this hardware interrupt or NULL for default host
+ * @hwirq: hardware irq number in that host space
+ *
+ * Only one mapping per hardware interrupt is permitted. Returns a linux
+ * virq number.
+ * If the sense/trigger is to be specified, set_irq_type() should be called
+ * on the number returned from that call.
+ */
+extern unsigned int irq_create_mapping(struct irq_host *host,
+					irq_hw_number_t hwirq);
+
+/**
+ * irq_create_of_mapping - Map a hardware interrupt into linux virq space
+ * @controller: Device node of the interrupt controller
+ * @inspec: Interrupt specifier from the device-tree
+ * @intsize: Size of the interrupt specifier from the device-tree
+ *
+ * This function is identical to irq_create_mapping except that it takes
+ * as input informations straight from the device-tree (typically the results
+ * of the of_irq_map_*() functions.
+ */
+extern unsigned int irq_create_of_mapping(struct device_node *controller,
+					u32 *intspec, unsigned int intsize);
+
 #endif /* _ASM_MICROBLAZE_IRQ_H */

arch/microblaze/include/asm/page.h

@@ -61,12 +61,6 @@ extern unsigned int __page_offset;
  */
 #define PAGE_OFFSET	CONFIG_KERNEL_START
 
-/*
- * MAP_NR -- given an address, calculate the index of the page struct which
- * points to the address's page.
- */
-#define MAP_NR(addr) (((unsigned long)(addr) - PAGE_OFFSET) >> PAGE_SHIFT)
-
 /*
  * The basic type of a PTE - 32 bit physical addressing.
  */
@@ -154,7 +148,11 @@ extern int page_is_ram(unsigned long pfn);
 # define pfn_to_virt(pfn)	__va(pfn_to_phys((pfn)))
 
 #  ifdef CONFIG_MMU
-#  define virt_to_page(kaddr) 	(mem_map +  MAP_NR(kaddr))
+
+#  define virt_to_page(kaddr)	(pfn_to_page(__pa(kaddr) >> PAGE_SHIFT))
+#  define page_to_virt(page)   __va(page_to_pfn(page) << PAGE_SHIFT)
+#  define page_to_phys(page)     (page_to_pfn(page) << PAGE_SHIFT)
+
 #  else /* CONFIG_MMU */
 #  define virt_to_page(vaddr)	(pfn_to_page(virt_to_pfn(vaddr)))
 #  define page_to_virt(page)	(pfn_to_virt(page_to_pfn(page)))

arch/microblaze/include/asm/pci-bridge.h

+#ifndef _ASM_MICROBLAZE_PCI_BRIDGE_H
+#define _ASM_MICROBLAZE_PCI_BRIDGE_H
+#ifdef __KERNEL__
+/*
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
 #include <linux/pci.h>
+#include <linux/list.h>
+#include <linux/ioport.h>
+
+struct device_node;
+
+enum {
+	/* Force re-assigning all resources (ignore firmware
+	 * setup completely)
+	 */
+	PCI_REASSIGN_ALL_RSRC	= 0x00000001,
+
+	/* Re-assign all bus numbers */
+	PCI_REASSIGN_ALL_BUS	= 0x00000002,
+
+	/* Do not try to assign, just use existing setup */
+	PCI_PROBE_ONLY		= 0x00000004,
+
+	/* Don't bother with ISA alignment unless the bridge has
+	 * ISA forwarding enabled
+	 */
+	PCI_CAN_SKIP_ISA_ALIGN	= 0x00000008,
+
+	/* Enable domain numbers in /proc */
+	PCI_ENABLE_PROC_DOMAINS	= 0x00000010,
+	/* ... except for domain 0 */
+	PCI_COMPAT_DOMAIN_0		= 0x00000020,
+};
+
+/*
+ * Structure of a PCI controller (host bridge)
+ */
+struct pci_controller {
+	struct pci_bus *bus;
+	char is_dynamic;
+	struct device_node *dn;
+	struct list_head list_node;
+	struct device *parent;
+
+	int first_busno;
+	int last_busno;
+
+	int self_busno;
+
+	void __iomem *io_base_virt;
+	resource_size_t io_base_phys;
+
+	resource_size_t pci_io_size;
+
+	/* Some machines (PReP) have a non 1:1 mapping of
+	 * the PCI memory space in the CPU bus space
+	 */
+	resource_size_t pci_mem_offset;
+
+	/* Some machines have a special region to forward the ISA
+	 * "memory" cycles such as VGA memory regions. Left to 0
+	 * if unsupported
+	 */
+	resource_size_t isa_mem_phys;
+	resource_size_t isa_mem_size;
+
+	struct pci_ops *ops;
+	unsigned int __iomem *cfg_addr;
+	void __iomem *cfg_data;
+
+	/*
+	 * Used for variants of PCI indirect handling and possible quirks:
+	 *  SET_CFG_TYPE - used on 4xx or any PHB that does explicit type0/1
+	 *  EXT_REG - provides access to PCI-e extended registers
+	 *  SURPRESS_PRIMARY_BUS - we surpress the setting of PCI_PRIMARY_BUS
+	 *   on Freescale PCI-e controllers since they used the PCI_PRIMARY_BUS
+	 *   to determine which bus number to match on when generating type0
+	 *   config cycles
+	 *  NO_PCIE_LINK - the Freescale PCI-e controllers have issues with
+	 *   hanging if we don't have link and try to do config cycles to
+	 *   anything but the PHB.  Only allow talking to the PHB if this is
+	 *   set.
+	 *  BIG_ENDIAN - cfg_addr is a big endian register
+	 *  BROKEN_MRM - the 440EPx/GRx chips have an errata that causes hangs
+	 *   on the PLB4.  Effectively disable MRM commands by setting this.
+	 */
+#define INDIRECT_TYPE_SET_CFG_TYPE		0x00000001
+#define INDIRECT_TYPE_EXT_REG		0x00000002
+#define INDIRECT_TYPE_SURPRESS_PRIMARY_BUS	0x00000004
+#define INDIRECT_TYPE_NO_PCIE_LINK		0x00000008
+#define INDIRECT_TYPE_BIG_ENDIAN		0x00000010
+#define INDIRECT_TYPE_BROKEN_MRM		0x00000020
+	u32 indirect_type;
+
+	/* Currently, we limit ourselves to 1 IO range and 3 mem
+	 * ranges since the common pci_bus structure can't handle more
+	 */
+	struct resource io_resource;
+	struct resource mem_resources[3];
+	int global_number;	/* PCI domain number */
+};
+
+static inline struct pci_controller *pci_bus_to_host(const struct pci_bus *bus)
+{
+	return bus->sysdata;
+}
+
+static inline int isa_vaddr_is_ioport(void __iomem *address)
+{
+	/* No specific ISA handling on ppc32 at this stage, it
+	 * all goes through PCI
+	 */
+	return 0;
+}
+
+/* These are used for config access before all the PCI probing
+   has been done. */
+extern int early_read_config_byte(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u8 *val);
+extern int early_read_config_word(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u16 *val);
+extern int early_read_config_dword(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u32 *val);
+extern int early_write_config_byte(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u8 val);
+extern int early_write_config_word(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u16 val);
+extern int early_write_config_dword(struct pci_controller *hose, int bus,
+			int dev_fn, int where, u32 val);
+
+extern int early_find_capability(struct pci_controller *hose, int bus,
+				 int dev_fn, int cap);
+
+extern void setup_indirect_pci(struct pci_controller *hose,
+			       resource_size_t cfg_addr,
+			       resource_size_t cfg_data, u32 flags);
+
+/* Get the PCI host controller for an OF device */
+extern struct pci_controller *pci_find_hose_for_OF_device(
+			struct device_node *node);
+
+/* Fill up host controller resources from the OF node */
+extern void pci_process_bridge_OF_ranges(struct pci_controller *hose,
+			struct device_node *dev, int primary);
+
+/* Allocate & free a PCI host bridge structure */
+extern struct pci_controller *pcibios_alloc_controller(struct device_node *dev);
+extern void pcibios_free_controller(struct pci_controller *phb);
+extern void pcibios_setup_phb_resources(struct pci_controller *hose);
+
+#ifdef CONFIG_PCI
+extern unsigned int pci_flags;
+
+static inline void pci_set_flags(int flags)
+{
+	pci_flags = flags;
+}
+
+static inline void pci_add_flags(int flags)
+{
+	pci_flags |= flags;
+}
+
+static inline int pci_has_flag(int flag)
+{
+	return pci_flags & flag;
+}
+
+extern struct list_head hose_list;
+
+extern unsigned long pci_address_to_pio(phys_addr_t address);
+extern int pcibios_vaddr_is_ioport(void __iomem *address);
+#else
+static inline unsigned long pci_address_to_pio(phys_addr_t address)
+{
+	return (unsigned long)-1;
+}
+static inline int pcibios_vaddr_is_ioport(void __iomem *address)
+{
+	return 0;
+}
+
+static inline void pci_set_flags(int flags) { }
+static inline void pci_add_flags(int flags) { }
+static inline int pci_has_flag(int flag)
+{
+	return 0;
+}
+#endif	/* CONFIG_PCI */
+
+#endif	/* __KERNEL__ */
+#endif	/* _ASM_MICROBLAZE_PCI_BRIDGE_H */

arch/microblaze/include/asm/pci.h

-#include <asm-generic/pci.h>
+/*
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ *
+ * Based on powerpc version
+ */
+
+#ifndef __ASM_MICROBLAZE_PCI_H
+#define __ASM_MICROBLAZE_PCI_H
+#ifdef __KERNEL__
+
+#include <linux/types.h>
+#include <linux/slab.h>
+#include <linux/string.h>
+#include <linux/dma-mapping.h>
+#include <linux/pci.h>
+
+#include <asm/scatterlist.h>
+#include <asm/io.h>
+#include <asm/prom.h>
+#include <asm/pci-bridge.h>
+
+#define PCIBIOS_MIN_IO		0x1000
+#define PCIBIOS_MIN_MEM		0x10000000
+
+struct pci_dev;
+
+/* Values for the `which' argument to sys_pciconfig_iobase syscall.  */
+#define IOBASE_BRIDGE_NUMBER	0
+#define IOBASE_MEMORY		1
+#define IOBASE_IO		2
+#define IOBASE_ISA_IO		3
+#define IOBASE_ISA_MEM		4
+
+#define pcibios_scan_all_fns(a, b)	0
+
+/*
+ * Set this to 1 if you want the kernel to re-assign all PCI
+ * bus numbers (don't do that on ppc64 yet !)
+ */
+#define pcibios_assign_all_busses() \
+	(pci_has_flag(PCI_REASSIGN_ALL_BUS))
+
+static inline void pcibios_set_master(struct pci_dev *dev)
+{
+	/* No special bus mastering setup handling */
+}
+
+static inline void pcibios_penalize_isa_irq(int irq, int active)
+{
+	/* We don't do dynamic PCI IRQ allocation */
+}
+
+#ifdef CONFIG_PCI
+extern void set_pci_dma_ops(struct dma_map_ops *dma_ops);
+extern struct dma_map_ops *get_pci_dma_ops(void);
+#else	/* CONFIG_PCI */
+#define set_pci_dma_ops(d)
+#define get_pci_dma_ops()	NULL
+#endif
+
+#ifdef CONFIG_PCI
+static inline void pci_dma_burst_advice(struct pci_dev *pdev,
+					enum pci_dma_burst_strategy *strat,
+					unsigned long *strategy_parameter)
+{
+	*strat = PCI_DMA_BURST_INFINITY;
+	*strategy_parameter = ~0UL;
+}
+#endif
+
+extern int pci_domain_nr(struct pci_bus *bus);
+
+/* Decide whether to display the domain number in /proc */
+extern int pci_proc_domain(struct pci_bus *bus);
+
+struct vm_area_struct;
+/* Map a range of PCI memory or I/O space for a device into user space */
+int pci_mmap_page_range(struct pci_dev *pdev, struct vm_area_struct *vma,
+			enum pci_mmap_state mmap_state, int write_combine);
+
+/* Tell drivers/pci/proc.c that we have pci_mmap_page_range() */
+#define HAVE_PCI_MMAP	1
+
+extern int pci_legacy_read(struct pci_bus *bus, loff_t port, u32 *val,
+			   size_t count);
+extern int pci_legacy_write(struct pci_bus *bus, loff_t port, u32 val,
+			   size_t count);
+extern int pci_mmap_legacy_page_range(struct pci_bus *bus,
+				      struct vm_area_struct *vma,
+				      enum pci_mmap_state mmap_state);
+
+#define HAVE_PCI_LEGACY	1
+
+/* pci_unmap_{page,single} is a nop so... */
+#define DECLARE_PCI_UNMAP_ADDR(ADDR_NAME)
+#define DECLARE_PCI_UNMAP_LEN(LEN_NAME)
+#define pci_unmap_addr(PTR, ADDR_NAME)		(0)
+#define pci_unmap_addr_set(PTR, ADDR_NAME, VAL)	do { } while (0)
+#define pci_unmap_len(PTR, LEN_NAME)		(0)
+#define pci_unmap_len_set(PTR, LEN_NAME, VAL)	do { } while (0)
+
+/* The PCI address space does equal the physical memory
+ * address space (no IOMMU).  The IDE and SCSI device layers use
+ * this boolean for bounce buffer decisions.
+ */
+#define PCI_DMA_BUS_IS_PHYS     (1)
+
+extern void pcibios_resource_to_bus(struct pci_dev *dev,
+			struct pci_bus_region *region,
+			struct resource *res);
+
+extern void pcibios_bus_to_resource(struct pci_dev *dev,
+			struct resource *res,
+			struct pci_bus_region *region);
+
+static inline struct resource *pcibios_select_root(struct pci_dev *pdev,
+			struct resource *res)
+{
+	struct resource *root = NULL;
+
+	if (res->flags & IORESOURCE_IO)
+		root = &ioport_resource;
+	if (res->flags & IORESOURCE_MEM)
+		root = &iomem_resource;
+
+	return root;
+}
+
+extern void pcibios_claim_one_bus(struct pci_bus *b);
+
+extern void pcibios_finish_adding_to_bus(struct pci_bus *bus);
+
+extern void pcibios_resource_survey(void);
+
+extern struct pci_controller *init_phb_dynamic(struct device_node *dn);
+extern int remove_phb_dynamic(struct pci_controller *phb);
+
+extern struct pci_dev *of_create_pci_dev(struct device_node *node,
+					struct pci_bus *bus, int devfn);
+
+extern void of_scan_pci_bridge(struct device_node *node,
+				struct pci_dev *dev);
+
+extern void of_scan_bus(struct device_node *node, struct pci_bus *bus);
+extern void of_rescan_bus(struct device_node *node, struct pci_bus *bus);
+
+extern int pci_read_irq_line(struct pci_dev *dev);
+
+extern int pci_bus_find_capability(struct pci_bus *bus,
+						unsigned int devfn, int cap);
+
+struct file;
+extern pgprot_t	pci_phys_mem_access_prot(struct file *file,
+					 unsigned long pfn,
+					 unsigned long size,
+					 pgprot_t prot);
+
+#define HAVE_ARCH_PCI_RESOURCE_TO_USER
+extern void pci_resource_to_user(const struct pci_dev *dev, int bar,
+				 const struct resource *rsrc,
+				 resource_size_t *start, resource_size_t *end);
+
+extern void pcibios_setup_bus_devices(struct pci_bus *bus);
+extern void pcibios_setup_bus_self(struct pci_bus *bus);
+
+/* This part of code was originaly in xilinx-pci.h */
+#ifdef CONFIG_PCI_XILINX
+extern void __init xilinx_pci_init(void);
+#else
+static inline void __init xilinx_pci_init(void) { return; }
+#endif
+
+#endif	/* __KERNEL__ */
+#endif /* __ASM_MICROBLAZE_PCI_H */

arch/microblaze/include/asm/pgalloc.h

@@ -19,6 +19,7 @@
 #include <asm/io.h>
 #include <asm/page.h>
 #include <asm/cache.h>
+#include <asm/pgtable.h>
 
 #define PGDIR_ORDER	0
 
@@ -111,7 +112,6 @@ static inline pte_t *pte_alloc_one_kernel(struct mm_struct *mm,
 		unsigned long address)
 {
 	pte_t *pte;
-	extern int mem_init_done;
 	extern void *early_get_page(void);
 	if (mem_init_done) {
 		pte = (pte_t *)__get_free_page(GFP_KERNEL |

arch/microblaze/include/asm/pgtable.h

@@ -16,6 +16,10 @@
 #define io_remap_pfn_range(vma, vaddr, pfn, size, prot)		\
 		remap_pfn_range(vma, vaddr, pfn, size, prot)
 
+#ifndef __ASSEMBLY__
+extern int mem_init_done;
+#endif
+
 #ifndef CONFIG_MMU
 
 #define pgd_present(pgd)	(1) /* pages are always present on non MMU */
@@ -51,6 +55,8 @@ static inline int pte_file(pte_t pte) { return 0; }
 
 #define arch_enter_lazy_cpu_mode()	do {} while (0)
 
+#define pgprot_noncached_wc(prot)	prot
+
 #else /* CONFIG_MMU */
 
 #include <asm-generic/4level-fixup.h>
@@ -68,7 +74,6 @@ static inline int pte_file(pte_t pte) { return 0; }
 
 extern unsigned long va_to_phys(unsigned long address);
 extern pte_t *va_to_pte(unsigned long address);
-extern unsigned long ioremap_bot, ioremap_base;
 
 /*
  * The following only work if pte_present() is true.
@@ -85,10 +90,24 @@ static inline pte_t pte_mkspecial(pte_t pte)	{ return pte; }
 #define VMALLOC_START	(CONFIG_KERNEL_START + \
 				max(32 * 1024 * 1024UL, memory_size))
 #define VMALLOC_END	ioremap_bot
-#define VMALLOC_VMADDR(x) ((unsigned long)(x))
 
 #endif /* __ASSEMBLY__ */
 
+/*
+ * Macro to mark a page protection value as "uncacheable".
+ */
+
+#define _PAGE_CACHE_CTL	(_PAGE_GUARDED | _PAGE_NO_CACHE | \
+							_PAGE_WRITETHRU)
+
+#define pgprot_noncached(prot) \
+			(__pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) | \
+					_PAGE_NO_CACHE | _PAGE_GUARDED))
+
+#define pgprot_noncached_wc(prot) \
+			 (__pgprot((pgprot_val(prot) & ~_PAGE_CACHE_CTL) | \
+							_PAGE_NO_CACHE))
+
 /*
  * The MicroBlaze MMU is identical to the PPC-40x MMU, and uses a hash
  * table containing PTEs, together with a set of 16 segment registers, to
@@ -397,7 +416,7 @@ static inline unsigned long pte_update(pte_t *p, unsigned long clr,
 	mts     rmsr, %2\n\
 	nop"
 	: "=&r" (old), "=&r" (tmp), "=&r" (msr), "=m" (*p)
-	: "r" ((unsigned long)(p+1) - 4), "r" (clr), "r" (set), "m" (*p)
+	: "r" ((unsigned long)(p + 1) - 4), "r" (clr), "r" (set), "m" (*p)
 	: "cc");
 
 	return old;
@@ -566,18 +585,11 @@ void mapin_ram(void);
 int map_page(unsigned long va, phys_addr_t pa, int flags);
 
 extern int mem_init_done;
-extern unsigned long ioremap_base;
-extern unsigned long ioremap_bot;
 
 asmlinkage void __init mmu_init(void);
 
 void __init *early_get_page(void);
 
-void *consistent_alloc(int gfp, size_t size, dma_addr_t *dma_handle);
-void consistent_free(void *vaddr);
-void consistent_sync(void *vaddr, size_t size, int direction);
-void consistent_sync_page(struct page *page, unsigned long offset,
-	size_t size, int direction);
 #endif /* __ASSEMBLY__ */
 #endif /* __KERNEL__ */
 
@@ -586,6 +598,14 @@ void consistent_sync_page(struct page *page, unsigned long offset,
 #ifndef __ASSEMBLY__
 #include <asm-generic/pgtable.h>
 
+extern unsigned long ioremap_bot, ioremap_base;
+
+void *consistent_alloc(int gfp, size_t size, dma_addr_t *dma_handle);
+void consistent_free(void *vaddr);
+void consistent_sync(void *vaddr, size_t size, int direction);
+void consistent_sync_page(struct page *page, unsigned long offset,
+	size_t size, int direction);
+
 void setup_memory(void);
 #endif /* __ASSEMBLY__ */
 

arch/microblaze/include/asm/prom.h

@@ -31,6 +31,21 @@
 /* Other Prototypes */
 extern int early_uartlite_console(void);
 
+#ifdef CONFIG_PCI
+/*
+ * PCI <-> OF matching functions
+ * (XXX should these be here?)
+ */
+struct pci_bus;
+struct pci_dev;
+extern int pci_device_from_OF_node(struct device_node *node,
+					u8 *bus, u8 *devfn);
+extern struct device_node *pci_busdev_to_OF_node(struct pci_bus *bus,
+							int devfn);
+extern struct device_node *pci_device_to_OF_node(struct pci_dev *dev);
+extern void pci_create_OF_bus_map(void);
+#endif
+
 /*
  * OF address retreival & translation
  */

arch/microblaze/include/asm/system.h

@@ -87,6 +87,9 @@ void free_initmem(void);
 extern char *klimit;
 extern void ret_from_fork(void);
 
+extern void *alloc_maybe_bootmem(size_t size, gfp_t mask);
+extern void *zalloc_maybe_bootmem(size_t size, gfp_t mask);
+
 #ifdef CONFIG_DEBUG_FS
 extern struct dentry *of_debugfs_root;
 #endif

arch/microblaze/include/asm/tlbflush.h

@@ -23,7 +23,7 @@
 extern void _tlbie(unsigned long address);
 extern void _tlbia(void);
 
-#define __tlbia()	_tlbia()
+#define __tlbia()	{ preempt_disable(); _tlbia(); preempt_enable(); }
 
 static inline void local_flush_tlb_all(void)
 	{ __tlbia(); }

arch/microblaze/kernel/Makefile

@@ -14,7 +14,7 @@ endif
 
 extra-y := head.o vmlinux.lds
 
-obj-y += exceptions.o \
+obj-y += dma.o exceptions.o \
 	hw_exception_handler.o init_task.o intc.o irq.o of_device.o \
 	of_platform.o process.o prom.o prom_parse.o ptrace.o \
 	setup.o signal.o sys_microblaze.o timer.o traps.o reset.o

arch/microblaze/kernel/asm-offsets.c

@@ -90,6 +90,7 @@ int main(int argc, char *argv[])
 	DEFINE(TI_FLAGS, offsetof(struct thread_info, flags));
 	DEFINE(TI_ADDR_LIMIT, offsetof(struct thread_info, addr_limit));
 	DEFINE(TI_CPU_CONTEXT, offsetof(struct thread_info, cpu_context));
+	DEFINE(TI_PREEMPT_COUNT, offsetof(struct thread_info, preempt_count));
 	BLANK();
 
 	/* struct cpu_context */

arch/microblaze/kernel/cpu/cache.c

@@ -15,25 +15,6 @@
 #include <asm/cpuinfo.h>
 #include <asm/pvr.h>
 
-static inline void __invalidate_flush_icache(unsigned int addr)
-{
-	__asm__ __volatile__ ("wic	%0, r0;"	\
-					: : "r" (addr));
-}
-
-static inline void __flush_dcache(unsigned int addr)
-{
-	__asm__ __volatile__ ("wdc.flush	%0, r0;"	\
-					: : "r" (addr));
-}
-
-static inline void __invalidate_dcache(unsigned int baseaddr,
-						unsigned int offset)
-{
-	__asm__ __volatile__ ("wdc.clear	%0, %1;"	\
-					: : "r" (baseaddr), "r" (offset));
-}
-
 static inline void __enable_icache_msr(void)
 {
 	__asm__ __volatile__ ("	msrset	r0, %0;		\
@@ -183,10 +164,14 @@ do {									\
 					"r" (line_length) : "memory");	\
 } while (0);
 
+#define ASM_LOOP
+
 static void __flush_icache_range_msr_irq(unsigned long start, unsigned long end)
 {
 	unsigned long flags;
-
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
 				(unsigned int)start, (unsigned int) end);
 
@@ -196,8 +181,13 @@ static void __flush_icache_range_msr_irq(unsigned long start, unsigned long end)
 	local_irq_save(flags);
 	__disable_icache_msr();
 
+#ifdef ASM_LOOP
 	CACHE_RANGE_LOOP_1(start, end, cpuinfo.icache_line_length, wic);
-
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wic	%0, r0;"	\
+				: : "r" (i));
+#endif
 	__enable_icache_msr();
 	local_irq_restore(flags);
 }
@@ -206,7 +196,9 @@ static void __flush_icache_range_nomsr_irq(unsigned long start,
 				unsigned long end)
 {
 	unsigned long flags;
-
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
 				(unsigned int)start, (unsigned int) end);
 
@@ -216,7 +208,13 @@ static void __flush_icache_range_nomsr_irq(unsigned long start,
 	local_irq_save(flags);
 	__disable_icache_nomsr();
 
+#ifdef ASM_LOOP
 	CACHE_RANGE_LOOP_1(start, end, cpuinfo.icache_line_length, wic);
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wic	%0, r0;"	\
+				: : "r" (i));
+#endif
 
 	__enable_icache_nomsr();
 	local_irq_restore(flags);
@@ -225,25 +223,41 @@ static void __flush_icache_range_nomsr_irq(unsigned long start,
 static void __flush_icache_range_noirq(unsigned long start,
 				unsigned long end)
 {
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
 				(unsigned int)start, (unsigned int) end);
 
 	CACHE_LOOP_LIMITS(start, end,
 			cpuinfo.icache_line_length, cpuinfo.icache_size);
+#ifdef ASM_LOOP
 	CACHE_RANGE_LOOP_1(start, end, cpuinfo.icache_line_length, wic);
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wic	%0, r0;"	\
+				: : "r" (i));
+#endif
 }
 
 static void __flush_icache_all_msr_irq(void)
 {
 	unsigned long flags;
-
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s\n", __func__);
 
 	local_irq_save(flags);
 	__disable_icache_msr();
-
+#ifdef ASM_LOOP
 	CACHE_ALL_LOOP(cpuinfo.icache_size, cpuinfo.icache_line_length, wic);
-
+#else
+	for (i = 0; i < cpuinfo.icache_size;
+		 i += cpuinfo.icache_line_length)
+			__asm__ __volatile__ ("wic	%0, r0;" \
+					: : "r" (i));
+#endif
 	__enable_icache_msr();
 	local_irq_restore(flags);
 }
@@ -251,35 +265,59 @@ static void __flush_icache_all_msr_irq(void)
 static void __flush_icache_all_nomsr_irq(void)
 {
 	unsigned long flags;
-
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s\n", __func__);
 
 	local_irq_save(flags);
 	__disable_icache_nomsr();
-
+#ifdef ASM_LOOP
 	CACHE_ALL_LOOP(cpuinfo.icache_size, cpuinfo.icache_line_length, wic);
-
+#else
+	for (i = 0; i < cpuinfo.icache_size;
+		 i += cpuinfo.icache_line_length)
+			__asm__ __volatile__ ("wic	%0, r0;" \
+					: : "r" (i));
+#endif
 	__enable_icache_nomsr();
 	local_irq_restore(flags);
 }
 
 static void __flush_icache_all_noirq(void)
 {
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s\n", __func__);
+#ifdef ASM_LOOP
 	CACHE_ALL_LOOP(cpuinfo.icache_size, cpuinfo.icache_line_length, wic);
+#else
+	for (i = 0; i < cpuinfo.icache_size;
+		 i += cpuinfo.icache_line_length)
+			__asm__ __volatile__ ("wic	%0, r0;" \
+					: : "r" (i));
+#endif
 }
 
 static void __invalidate_dcache_all_msr_irq(void)
 {
 	unsigned long flags;
-
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s\n", __func__);
 
 	local_irq_save(flags);
 	__disable_dcache_msr();
-
+#ifdef ASM_LOOP
 	CACHE_ALL_LOOP(cpuinfo.dcache_size, cpuinfo.dcache_line_length, wdc);
-
+#else
+	for (i = 0; i < cpuinfo.dcache_size;
+		 i += cpuinfo.dcache_line_length)
+			__asm__ __volatile__ ("wdc	%0, r0;" \
+					: : "r" (i));
+#endif
 	__enable_dcache_msr();
 	local_irq_restore(flags);
 }
@@ -287,60 +325,107 @@ static void __invalidate_dcache_all_msr_irq(void)
 static void __invalidate_dcache_all_nomsr_irq(void)
 {
 	unsigned long flags;
-
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s\n", __func__);
 
 	local_irq_save(flags);
 	__disable_dcache_nomsr();
-
+#ifdef ASM_LOOP
 	CACHE_ALL_LOOP(cpuinfo.dcache_size, cpuinfo.dcache_line_length, wdc);
-
+#else
+	for (i = 0; i < cpuinfo.dcache_size;
+		 i += cpuinfo.dcache_line_length)
+			__asm__ __volatile__ ("wdc	%0, r0;" \
+					: : "r" (i));
+#endif
 	__enable_dcache_nomsr();
 	local_irq_restore(flags);
 }
 
 static void __invalidate_dcache_all_noirq_wt(void)
 {
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s\n", __func__);
+#ifdef ASM_LOOP
 	CACHE_ALL_LOOP(cpuinfo.dcache_size, cpuinfo.dcache_line_length, wdc)
+#else
+	for (i = 0; i < cpuinfo.dcache_size;
+		 i += cpuinfo.dcache_line_length)
+			__asm__ __volatile__ ("wdc	%0, r0;" \
+					: : "r" (i));
+#endif
 }
 
 /* FIXME this is weird - should be only wdc but not work
  * MS: I am getting bus errors and other weird things */
 static void __invalidate_dcache_all_wb(void)
 {
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s\n", __func__);
+#ifdef ASM_LOOP
 	CACHE_ALL_LOOP2(cpuinfo.dcache_size, cpuinfo.dcache_line_length,
 					wdc.clear)
+#else
+	for (i = 0; i < cpuinfo.dcache_size;
+		 i += cpuinfo.dcache_line_length)
+			__asm__ __volatile__ ("wdc.clear	%0, r0;" \
+					: : "r" (i));
+#endif
 }
 
 static void __invalidate_dcache_range_wb(unsigned long start,
 						unsigned long end)
 {
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
 				(unsigned int)start, (unsigned int) end);
 
 	CACHE_LOOP_LIMITS(start, end,
 			cpuinfo.dcache_line_length, cpuinfo.dcache_size);
+#ifdef ASM_LOOP
 	CACHE_RANGE_LOOP_2(start, end, cpuinfo.dcache_line_length, wdc.clear);
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wdc.clear	%0, r0;"	\
+				: : "r" (i));
+#endif
 }
 
 static void __invalidate_dcache_range_nomsr_wt(unsigned long start,
 							unsigned long end)
 {
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
 				(unsigned int)start, (unsigned int) end);
 	CACHE_LOOP_LIMITS(start, end,
 			cpuinfo.dcache_line_length, cpuinfo.dcache_size);
 
+#ifdef ASM_LOOP
 	CACHE_RANGE_LOOP_1(start, end, cpuinfo.dcache_line_length, wdc);
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wdc	%0, r0;"	\
+				: : "r" (i));
+#endif
 }
 
 static void __invalidate_dcache_range_msr_irq_wt(unsigned long start,
 							unsigned long end)
 {
 	unsigned long flags;
-
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
 				(unsigned int)start, (unsigned int) end);
 	CACHE_LOOP_LIMITS(start, end,
@@ -349,7 +434,13 @@ static void __invalidate_dcache_range_msr_irq_wt(unsigned long start,
 	local_irq_save(flags);
 	__disable_dcache_msr();
 
+#ifdef ASM_LOOP
 	CACHE_RANGE_LOOP_1(start, end, cpuinfo.dcache_line_length, wdc);
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wdc	%0, r0;"	\
+				: : "r" (i));
+#endif
 
 	__enable_dcache_msr();
 	local_irq_restore(flags);
@@ -359,7 +450,9 @@ static void __invalidate_dcache_range_nomsr_irq(unsigned long start,
 							unsigned long end)
 {
 	unsigned long flags;
-
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
 				(unsigned int)start, (unsigned int) end);
 
@@ -369,7 +462,13 @@ static void __invalidate_dcache_range_nomsr_irq(unsigned long start,
 	local_irq_save(flags);
 	__disable_dcache_nomsr();
 
+#ifdef ASM_LOOP
 	CACHE_RANGE_LOOP_1(start, end, cpuinfo.dcache_line_length, wdc);
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wdc	%0, r0;"	\
+				: : "r" (i));
+#endif
 
 	__enable_dcache_nomsr();
 	local_irq_restore(flags);
@@ -377,19 +476,38 @@ static void __invalidate_dcache_range_nomsr_irq(unsigned long start,
 
 static void __flush_dcache_all_wb(void)
 {
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s\n", __func__);
+#ifdef ASM_LOOP
 	CACHE_ALL_LOOP(cpuinfo.dcache_size, cpuinfo.dcache_line_length,
 				wdc.flush);
+#else
+	for (i = 0; i < cpuinfo.dcache_size;
+		 i += cpuinfo.dcache_line_length)
+			__asm__ __volatile__ ("wdc.flush	%0, r0;" \
+					: : "r" (i));
+#endif
 }
 
 static void __flush_dcache_range_wb(unsigned long start, unsigned long end)
 {
+#ifndef ASM_LOOP
+	int i;
+#endif
 	pr_debug("%s: start 0x%x, end 0x%x\n", __func__,
 				(unsigned int)start, (unsigned int) end);
 
 	CACHE_LOOP_LIMITS(start, end,
 			cpuinfo.dcache_line_length, cpuinfo.dcache_size);
+#ifdef ASM_LOOP
 	CACHE_RANGE_LOOP_2(start, end, cpuinfo.dcache_line_length, wdc.flush);
+#else
+	for (i = start; i < end; i += cpuinfo.icache_line_length)
+		__asm__ __volatile__ ("wdc.flush	%0, r0;"	\
+				: : "r" (i));
+#endif
 }
 
 /* struct for wb caches and for wt caches */
@@ -493,7 +611,7 @@ const struct scache wt_nomsr_noirq = {
 #define CPUVER_7_20_A	0x0c
 #define CPUVER_7_20_D	0x0f
 
-#define INFO(s)	printk(KERN_INFO "cache: " s " \n");
+#define INFO(s)	printk(KERN_INFO "cache: " s "\n");
 
 void microblaze_cache_init(void)
 {
@@ -532,4 +650,9 @@ void microblaze_cache_init(void)
 			}
 		}
 	}
+	invalidate_dcache();
+	enable_dcache();
+
+	invalidate_icache();
+	enable_icache();
 }

arch/microblaze/kernel/dma.c

+/*
+ * Copyright (C) 2009-2010 PetaLogix
+ * Copyright (C) 2006 Benjamin Herrenschmidt, IBM Corporation
+ *
+ * Provide default implementations of the DMA mapping callbacks for
+ * directly mapped busses.
+ */
+
+#include <linux/device.h>
+#include <linux/dma-mapping.h>
+#include <linux/dma-debug.h>
+#include <asm/bug.h>
+#include <asm/cacheflush.h>
+
+/*
+ * Generic direct DMA implementation
+ *
+ * This implementation supports a per-device offset that can be applied if
+ * the address at which memory is visible to devices is not 0. Platform code
+ * can set archdata.dma_data to an unsigned long holding the offset. By
+ * default the offset is PCI_DRAM_OFFSET.
+ */
+static inline void __dma_sync_page(unsigned long paddr, unsigned long offset,
+				size_t size, enum dma_data_direction direction)
+{
+	switch (direction) {
+	case DMA_TO_DEVICE:
+		flush_dcache_range(paddr + offset, paddr + offset + size);
+		break;
+	case DMA_FROM_DEVICE:
+		invalidate_dcache_range(paddr + offset, paddr + offset + size);
+		break;
+	default:
+		BUG();
+	}
+}
+
+static unsigned long get_dma_direct_offset(struct device *dev)
+{
+	if (dev)
+		return (unsigned long)dev->archdata.dma_data;
+
+	return PCI_DRAM_OFFSET; /* FIXME Not sure if is correct */
+}
+
+#define NOT_COHERENT_CACHE
+
+static void *dma_direct_alloc_coherent(struct device *dev, size_t size,
+				dma_addr_t *dma_handle, gfp_t flag)
+{
+#ifdef NOT_COHERENT_CACHE
+	return consistent_alloc(flag, size, dma_handle);
+#else
+	void *ret;
+	struct page *page;
+	int node = dev_to_node(dev);
+
+	/* ignore region specifiers */
+	flag  &= ~(__GFP_HIGHMEM);
+
+	page = alloc_pages_node(node, flag, get_order(size));
+	if (page == NULL)
+		return NULL;
+	ret = page_address(page);
+	memset(ret, 0, size);
+	*dma_handle = virt_to_phys(ret) + get_dma_direct_offset(dev);
+
+	return ret;
+#endif
+}
+
+static void dma_direct_free_coherent(struct device *dev, size_t size,
+			      void *vaddr, dma_addr_t dma_handle)
+{
+#ifdef NOT_COHERENT_CACHE
+	consistent_free(vaddr);
+#else
+	free_pages((unsigned long)vaddr, get_order(size));
+#endif
+}
+
+static int dma_direct_map_sg(struct device *dev, struct scatterlist *sgl,
+			     int nents, enum dma_data_direction direction,
+			     struct dma_attrs *attrs)
+{
+	struct scatterlist *sg;
+	int i;
+
+	/* FIXME this part of code is untested */
+	for_each_sg(sgl, sg, nents, i) {
+		sg->dma_address = sg_phys(sg) + get_dma_direct_offset(dev);
+		sg->dma_length = sg->length;
+		__dma_sync_page(page_to_phys(sg_page(sg)), sg->offset,
+							sg->length, direction);
+	}
+
+	return nents;
+}
+
+static void dma_direct_unmap_sg(struct device *dev, struct scatterlist *sg,
+				int nents, enum dma_data_direction direction,
+				struct dma_attrs *attrs)
+{
+}
+
+static int dma_direct_dma_supported(struct device *dev, u64 mask)
+{
+	return 1;
+}
+
+static inline dma_addr_t dma_direct_map_page(struct device *dev,
+					     struct page *page,
+					     unsigned long offset,
+					     size_t size,
+					     enum dma_data_direction direction,
+					     struct dma_attrs *attrs)
+{
+	__dma_sync_page(page_to_phys(page), offset, size, direction);
+	return page_to_phys(page) + offset + get_dma_direct_offset(dev);
+}
+
+static inline void dma_direct_unmap_page(struct device *dev,
+					 dma_addr_t dma_address,
+					 size_t size,
+					 enum dma_data_direction direction,
+					 struct dma_attrs *attrs)
+{
+/* There is not necessary to do cache cleanup
+ *
+ * phys_to_virt is here because in __dma_sync_page is __virt_to_phys and
+ * dma_address is physical address
+ */
+	__dma_sync_page(dma_address, 0 , size, direction);
+}
+
+struct dma_map_ops dma_direct_ops = {
+	.alloc_coherent	= dma_direct_alloc_coherent,
+	.free_coherent	= dma_direct_free_coherent,
+	.map_sg		= dma_direct_map_sg,
+	.unmap_sg	= dma_direct_unmap_sg,
+	.dma_supported	= dma_direct_dma_supported,
+	.map_page	= dma_direct_map_page,
+	.unmap_page	= dma_direct_unmap_page,
+};
+EXPORT_SYMBOL(dma_direct_ops);
+
+/* Number of entries preallocated for DMA-API debugging */
+#define PREALLOC_DMA_DEBUG_ENTRIES (1 << 16)
+
+static int __init dma_init(void)
+{
+       dma_debug_init(PREALLOC_DMA_DEBUG_ENTRIES);
+
+       return 0;
+}
+fs_initcall(dma_init);

arch/microblaze/kernel/head.S

@@ -99,8 +99,8 @@ no_fdt_arg:
 	tophys(r4,r4)			/* convert to phys address */
 	ori	r3, r0, COMMAND_LINE_SIZE - 1 /* number of loops */
 _copy_command_line:
-	lbu	r7, r5, r6 /* r7=r5+r6 - r5 contain pointer to command line */
-	sb	r7, r4, r6		/* addr[r4+r6]= r7*/
+	lbu	r2, r5, r6 /* r7=r5+r6 - r5 contain pointer to command line */
+	sb	r2, r4, r6		/* addr[r4+r6]= r7*/
 	addik	r6, r6, 1		/* increment counting */
 	bgtid	r3, _copy_command_line	/* loop for all entries       */
 	addik	r3, r3, -1		/* descrement loop */
@@ -136,6 +136,11 @@ _invalidate:
 	addik	r3, r3, -1
 	/* sync */
 
+	/* Setup the kernel PID */
+	mts	rpid,r0			/* Load the kernel PID */
+	nop
+	bri	4
+
 	/*
 	 * We should still be executing code at physical address area
 	 * RAM_BASEADDR at this point. However, kernel code is at
@@ -146,10 +151,6 @@ _invalidate:
 	addik	r3,r0, CONFIG_KERNEL_START /* Load the kernel virtual address */
 	tophys(r4,r3)			/* Load the kernel physical address */
 
-	mts	rpid,r0			/* Load the kernel PID */
-	nop
-	bri	4
-
 	/*
 	 * Configure and load two entries into TLB slots 0 and 1.
 	 * In case we are pinning TLBs, these are reserved in by the

arch/microblaze/kernel/irq.c

@@ -93,3 +93,18 @@ int show_interrupts(struct seq_file *p, void *v)
 	}
 	return 0;
 }
+
+/* MS: There is no any advance mapping mechanism. We are using simple 32bit
+  intc without any cascades or any connection that's why mapping is 1:1 */
+unsigned int irq_create_mapping(struct irq_host *host, irq_hw_number_t hwirq)
+{
+	return hwirq;
+}
+EXPORT_SYMBOL_GPL(irq_create_mapping);
+
+unsigned int irq_create_of_mapping(struct device_node *controller,
+					u32 *intspec, unsigned int intsize)
+{
+	return intspec[0];
+}
+EXPORT_SYMBOL_GPL(irq_create_of_mapping);

arch/microblaze/kernel/setup.c

@@ -22,7 +22,10 @@
 #include <linux/io.h>
 #include <linux/bug.h>
 #include <linux/param.h>
+#include <linux/pci.h>
 #include <linux/cache.h>
+#include <linux/of_platform.h>
+#include <linux/dma-mapping.h>
 #include <asm/cacheflush.h>
 #include <asm/entry.h>
 #include <asm/cpuinfo.h>
@@ -54,14 +57,10 @@ void __init setup_arch(char **cmdline_p)
 
 	microblaze_cache_init();
 
-	invalidate_dcache();
-	enable_dcache();
-
-	invalidate_icache();
-	enable_icache();
-
 	setup_memory();
 
+	xilinx_pci_init();
+
 #if defined(CONFIG_SELFMOD_INTC) || defined(CONFIG_SELFMOD_TIMER)
 	printk(KERN_NOTICE "Self modified code enable\n");
 #endif
@@ -188,3 +187,37 @@ static int microblaze_debugfs_init(void)
 }
 arch_initcall(microblaze_debugfs_init);
 #endif
+
+static int dflt_bus_notify(struct notifier_block *nb,
+				unsigned long action, void *data)
+{
+	struct device *dev = data;
+
+	/* We are only intereted in device addition */
+	if (action != BUS_NOTIFY_ADD_DEVICE)
+		return 0;
+
+	set_dma_ops(dev, &dma_direct_ops);
+
+	return NOTIFY_DONE;
+}
+
+static struct notifier_block dflt_plat_bus_notifier = {
+	.notifier_call = dflt_bus_notify,
+	.priority = INT_MAX,
+};
+
+static struct notifier_block dflt_of_bus_notifier = {
+	.notifier_call = dflt_bus_notify,
+	.priority = INT_MAX,
+};
+
+static int __init setup_bus_notifier(void)
+{
+	bus_register_notifier(&platform_bus_type, &dflt_plat_bus_notifier);
+	bus_register_notifier(&of_platform_bus_type, &dflt_of_bus_notifier);
+
+	return 0;
+}
+
+arch_initcall(setup_bus_notifier);

arch/microblaze/mm/Makefile

@@ -2,6 +2,6 @@
 # Makefile
 #
 
-obj-y := init.o
+obj-y := consistent.o init.o
 
 obj-$(CONFIG_MMU) += pgtable.o mmu_context.o fault.o

arch/microblaze/mm/consistent.c

+/*
+ * Microblaze support for cache consistent memory.
+ * Copyright (C) 2010 Michal Simek <monstr@monstr.eu>
+ * Copyright (C) 2010 PetaLogix
+ * Copyright (C) 2005 John Williams <jwilliams@itee.uq.edu.au>
+ *
+ * Based on PowerPC version derived from arch/arm/mm/consistent.c
+ * Copyright (C) 2001 Dan Malek (dmalek@jlc.net)
+ * Copyright (C) 2000 Russell King
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/module.h>
+#include <linux/signal.h>
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/ptrace.h>
+#include <linux/mman.h>
+#include <linux/mm.h>
+#include <linux/swap.h>
+#include <linux/stddef.h>
+#include <linux/vmalloc.h>
+#include <linux/init.h>
+#include <linux/delay.h>
+#include <linux/bootmem.h>
+#include <linux/highmem.h>
+#include <linux/pci.h>
+#include <linux/interrupt.h>
+
+#include <asm/pgalloc.h>
+#include <linux/io.h>
+#include <linux/hardirq.h>
+#include <asm/mmu_context.h>
+#include <asm/mmu.h>
+#include <linux/uaccess.h>
+#include <asm/pgtable.h>
+#include <asm/cpuinfo.h>
+
+#ifndef CONFIG_MMU
+
+/* I have to use dcache values because I can't relate on ram size */
+#define UNCACHED_SHADOW_MASK (cpuinfo.dcache_high - cpuinfo.dcache_base + 1)
+
+/*
+ * Consistent memory allocators. Used for DMA devices that want to
+ * share uncached memory with the processor core.
+ * My crufty no-MMU approach is simple. In the HW platform we can optionally
+ * mirror the DDR up above the processor cacheable region.  So, memory accessed
+ * in this mirror region will not be cached.  It's alloced from the same
+ * pool as normal memory, but the handle we return is shifted up into the
+ * uncached region.  This will no doubt cause big problems if memory allocated
+ * here is not also freed properly. -- JW
+ */
+void *consistent_alloc(int gfp, size_t size, dma_addr_t *dma_handle)
+{
+	struct page *page, *end, *free;
+	unsigned long order;
+	void *ret, *virt;
+
+	if (in_interrupt())
+		BUG();
+
+	size = PAGE_ALIGN(size);
+	order = get_order(size);
+
+	page = alloc_pages(gfp, order);
+	if (!page)
+		goto no_page;
+
+	/* We could do with a page_to_phys and page_to_bus here. */
+	virt = page_address(page);
+	ret = ioremap(virt_to_phys(virt), size);
+	if (!ret)
+		goto no_remap;
+
+	/*
+	 * Here's the magic!  Note if the uncached shadow is not implemented,
+	 * it's up to the calling code to also test that condition and make
+	 * other arranegments, such as manually flushing the cache and so on.
+	 */
+#ifdef CONFIG_XILINX_UNCACHED_SHADOW
+	ret = (void *)((unsigned) ret | UNCACHED_SHADOW_MASK);
+#endif
+	/* dma_handle is same as physical (shadowed) address */
+	*dma_handle = (dma_addr_t)ret;
+
+	/*
+	 * free wasted pages.  We skip the first page since we know
+	 * that it will have count = 1 and won't require freeing.
+	 * We also mark the pages in use as reserved so that
+	 * remap_page_range works.
+	 */
+	page = virt_to_page(virt);
+	free = page + (size >> PAGE_SHIFT);
+	end  = page + (1 << order);
+
+	for (; page < end; page++) {
+		init_page_count(page);
+		if (page >= free)
+			__free_page(page);
+		else
+			SetPageReserved(page);
+	}
+
+	return ret;
+no_remap:
+	__free_pages(page, order);
+no_page:
+	return NULL;
+}
+
+#else
+
+void *consistent_alloc(int gfp, size_t size, dma_addr_t *dma_handle)
+{
+	int order, err, i;
+	unsigned long page, va, flags;
+	phys_addr_t pa;
+	struct vm_struct *area;
+	void	 *ret;
+
+	if (in_interrupt())
+		BUG();
+
+	/* Only allocate page size areas. */
+	size = PAGE_ALIGN(size);
+	order = get_order(size);
+
+	page = __get_free_pages(gfp, order);
+	if (!page) {
+		BUG();
+		return NULL;
+	}
+
+	/*
+	 * we need to ensure that there are no cachelines in use,
+	 * or worse dirty in this area.
+	 */
+	flush_dcache_range(virt_to_phys(page), virt_to_phys(page) + size);
+
+	/* Allocate some common virtual space to map the new pages. */
+	area = get_vm_area(size, VM_ALLOC);
+	if (area == NULL) {
+		free_pages(page, order);
+		return NULL;
+	}
+	va = (unsigned long) area->addr;
+	ret = (void *)va;
+
+	/* This gives us the real physical address of the first page. */
+	*dma_handle = pa = virt_to_bus((void *)page);
+
+	/* MS: This is the whole magic - use cache inhibit pages */
+	flags = _PAGE_KERNEL | _PAGE_NO_CACHE;
+
+	/*
+	 * Set refcount=1 on all pages in an order>0
+	 * allocation so that vfree() will actually
+	 * free all pages that were allocated.
+	 */
+	if (order > 0) {
+		struct page *rpage = virt_to_page(page);
+		for (i = 1; i < (1 << order); i++)
+			init_page_count(rpage+i);
+	}
+
+	err = 0;
+	for (i = 0; i < size && err == 0; i += PAGE_SIZE)
+		err = map_page(va+i, pa+i, flags);
+
+	if (err) {
+		vfree((void *)va);
+		return NULL;
+	}
+
+	return ret;
+}
+#endif /* CONFIG_MMU */
+EXPORT_SYMBOL(consistent_alloc);
+
+/*
+ * free page(s) as defined by the above mapping.
+ */
+void consistent_free(void *vaddr)
+{
+	if (in_interrupt())
+		BUG();
+
+	/* Clear SHADOW_MASK bit in address, and free as per usual */
+#ifdef CONFIG_XILINX_UNCACHED_SHADOW
+	vaddr = (void *)((unsigned)vaddr & ~UNCACHED_SHADOW_MASK);
+#endif
+	vfree(vaddr);
+}
+EXPORT_SYMBOL(consistent_free);
+
+/*
+ * make an area consistent.
+ */
+void consistent_sync(void *vaddr, size_t size, int direction)
+{
+	unsigned long start;
+	unsigned long end;
+
+	start = (unsigned long)vaddr;
+
+	/* Convert start address back down to unshadowed memory region */
+#ifdef CONFIG_XILINX_UNCACHED_SHADOW
+	start &= ~UNCACHED_SHADOW_MASK;
+#endif
+	end = start + size;
+
+	switch (direction) {
+	case PCI_DMA_NONE:
+		BUG();
+	case PCI_DMA_FROMDEVICE:	/* invalidate only */
+		flush_dcache_range(start, end);
+		break;
+	case PCI_DMA_TODEVICE:		/* writeback only */
+		flush_dcache_range(start, end);
+		break;
+	case PCI_DMA_BIDIRECTIONAL:	/* writeback and invalidate */
+		flush_dcache_range(start, end);
+		break;
+	}
+}
+EXPORT_SYMBOL(consistent_sync);
+
+/*
+ * consistent_sync_page makes memory consistent. identical
+ * to consistent_sync, but takes a struct page instead of a
+ * virtual address
+ */
+void consistent_sync_page(struct page *page, unsigned long offset,
+	size_t size, int direction)
+{
+	unsigned long start = (unsigned long)page_address(page) + offset;
+	consistent_sync((void *)start, size, direction);
+}
+EXPORT_SYMBOL(consistent_sync_page);

arch/microblaze/mm/init.c

@@ -23,6 +23,9 @@
 #include <asm/sections.h>
 #include <asm/tlb.h>
 
+/* Use for MMU and noMMU because of PCI generic code */
+int mem_init_done;
+
 #ifndef CONFIG_MMU
 unsigned int __page_offset;
 EXPORT_SYMBOL(__page_offset);
@@ -30,7 +33,6 @@ EXPORT_SYMBOL(__page_offset);
 #else
 DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);
 
-int mem_init_done;
 static int init_bootmem_done;
 #endif /* CONFIG_MMU */
 
@@ -193,12 +195,6 @@ void free_initmem(void)
 			(unsigned long)(&__init_end));
 }
 
-/* FIXME from arch/powerpc/mm/mem.c*/
-void show_mem(void)
-{
-	printk(KERN_NOTICE "%s\n", __func__);
-}
-
 void __init mem_init(void)
 {
 	high_memory = (void *)__va(memory_end);
@@ -208,9 +204,7 @@ void __init mem_init(void)
 	printk(KERN_INFO "Memory: %luk/%luk available\n",
 	       nr_free_pages() << (PAGE_SHIFT-10),
 	       num_physpages << (PAGE_SHIFT-10));
-#ifdef CONFIG_MMU
 	mem_init_done = 1;
-#endif
 }
 
 #ifndef CONFIG_MMU
@@ -222,6 +216,10 @@ int ___range_ok(unsigned long addr, unsigned long size)
 }
 EXPORT_SYMBOL(___range_ok);
 
+int page_is_ram(unsigned long pfn)
+{
+	return __range_ok(pfn, 0);
+}
 #else
 int page_is_ram(unsigned long pfn)
 {
@@ -349,4 +347,27 @@ void __init *early_get_page(void)
 	}
 	return p;
 }
+
 #endif /* CONFIG_MMU */
+
+void * __init_refok alloc_maybe_bootmem(size_t size, gfp_t mask)
+{
+	if (mem_init_done)
+		return kmalloc(size, mask);
+	else
+		return alloc_bootmem(size);
+}
+
+void * __init_refok zalloc_maybe_bootmem(size_t size, gfp_t mask)
+{
+	void *p;
+
+	if (mem_init_done)
+		p = kzalloc(size, mask);
+	else {
+		p = alloc_bootmem(size);
+		if (p)
+			memset(p, 0, size);
+	}
+	return p;
+}

arch/microblaze/mm/pgtable.c

@@ -103,7 +103,7 @@ static void __iomem *__ioremap(phys_addr_t addr, unsigned long size,
 		area = get_vm_area(size, VM_IOREMAP);
 		if (area == NULL)
 			return NULL;
-		v = VMALLOC_VMADDR(area->addr);
+		v = (unsigned long) area->addr;
 	} else {
 		v = (ioremap_bot -= size);
 	}

arch/microblaze/pci/Makefile

+#
+# Makefile
+#
+
+obj-$(CONFIG_PCI)		+= pci_32.o pci-common.o indirect_pci.o iomap.o
+obj-$(CONFIG_PCI_XILINX)	+= xilinx_pci.o

arch/microblaze/pci/indirect_pci.c

+/*
+ * Support for indirect PCI bridges.
+ *
+ * Copyright (C) 1998 Gabriel Paubert.
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License
+ * as published by the Free Software Foundation; either version
+ * 2 of the License, or (at your option) any later version.
+ */
+
+#include <linux/kernel.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/string.h>
+#include <linux/init.h>
+
+#include <asm/io.h>
+#include <asm/prom.h>
+#include <asm/pci-bridge.h>
+
+static int
+indirect_read_config(struct pci_bus *bus, unsigned int devfn, int offset,
+		     int len, u32 *val)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	volatile void __iomem *cfg_data;
+	u8 cfg_type = 0;
+	u32 bus_no, reg;
+
+	if (hose->indirect_type & INDIRECT_TYPE_NO_PCIE_LINK) {
+		if (bus->number != hose->first_busno)
+			return PCIBIOS_DEVICE_NOT_FOUND;
+		if (devfn != 0)
+			return PCIBIOS_DEVICE_NOT_FOUND;
+	}
+
+	if (hose->indirect_type & INDIRECT_TYPE_SET_CFG_TYPE)
+		if (bus->number != hose->first_busno)
+			cfg_type = 1;
+
+	bus_no = (bus->number == hose->first_busno) ?
+			hose->self_busno : bus->number;
+
+	if (hose->indirect_type & INDIRECT_TYPE_EXT_REG)
+		reg = ((offset & 0xf00) << 16) | (offset & 0xfc);
+	else
+		reg = offset & 0xfc; /* Only 3 bits for function */
+
+	if (hose->indirect_type & INDIRECT_TYPE_BIG_ENDIAN)
+		out_be32(hose->cfg_addr, (0x80000000 | (bus_no << 16) |
+			 (devfn << 8) | reg | cfg_type));
+	else
+		out_le32(hose->cfg_addr, (0x80000000 | (bus_no << 16) |
+			 (devfn << 8) | reg | cfg_type));
+
+	/*
+	 * Note: the caller has already checked that offset is
+	 * suitably aligned and that len is 1, 2 or 4.
+	 */
+	cfg_data = hose->cfg_data + (offset & 3); /* Only 3 bits for function */
+	switch (len) {
+	case 1:
+		*val = in_8(cfg_data);
+		break;
+	case 2:
+		*val = in_le16(cfg_data);
+		break;
+	default:
+		*val = in_le32(cfg_data);
+		break;
+	}
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static int
+indirect_write_config(struct pci_bus *bus, unsigned int devfn, int offset,
+		      int len, u32 val)
+{
+	struct pci_controller *hose = pci_bus_to_host(bus);
+	volatile void __iomem *cfg_data;
+	u8 cfg_type = 0;
+	u32 bus_no, reg;
+
+	if (hose->indirect_type & INDIRECT_TYPE_NO_PCIE_LINK) {
+		if (bus->number != hose->first_busno)
+			return PCIBIOS_DEVICE_NOT_FOUND;
+		if (devfn != 0)
+			return PCIBIOS_DEVICE_NOT_FOUND;
+	}
+
+	if (hose->indirect_type & INDIRECT_TYPE_SET_CFG_TYPE)
+		if (bus->number != hose->first_busno)
+			cfg_type = 1;
+
+	bus_no = (bus->number == hose->first_busno) ?
+			hose->self_busno : bus->number;
+
+	if (hose->indirect_type & INDIRECT_TYPE_EXT_REG)
+		reg = ((offset & 0xf00) << 16) | (offset & 0xfc);
+	else
+		reg = offset & 0xfc;
+
+	if (hose->indirect_type & INDIRECT_TYPE_BIG_ENDIAN)
+		out_be32(hose->cfg_addr, (0x80000000 | (bus_no << 16) |
+			 (devfn << 8) | reg | cfg_type));
+	else
+		out_le32(hose->cfg_addr, (0x80000000 | (bus_no << 16) |
+			 (devfn << 8) | reg | cfg_type));
+
+	/* surpress setting of PCI_PRIMARY_BUS */
+	if (hose->indirect_type & INDIRECT_TYPE_SURPRESS_PRIMARY_BUS)
+		if ((offset == PCI_PRIMARY_BUS) &&
+			(bus->number == hose->first_busno))
+			val &= 0xffffff00;
+
+	/* Workaround for PCI_28 Errata in 440EPx/GRx */
+	if ((hose->indirect_type & INDIRECT_TYPE_BROKEN_MRM) &&
+			offset == PCI_CACHE_LINE_SIZE) {
+		val = 0;
+	}
+
+	/*
+	 * Note: the caller has already checked that offset is
+	 * suitably aligned and that len is 1, 2 or 4.
+	 */
+	cfg_data = hose->cfg_data + (offset & 3);
+	switch (len) {
+	case 1:
+		out_8(cfg_data, val);
+		break;
+	case 2:
+		out_le16(cfg_data, val);
+		break;
+	default:
+		out_le32(cfg_data, val);
+		break;
+	}
+
+	return PCIBIOS_SUCCESSFUL;
+}
+
+static struct pci_ops indirect_pci_ops = {
+	.read = indirect_read_config,
+	.write = indirect_write_config,
+};
+
+void __init
+setup_indirect_pci(struct pci_controller *hose,
+		   resource_size_t cfg_addr,
+		   resource_size_t cfg_data, u32 flags)
+{
+	resource_size_t base = cfg_addr & PAGE_MASK;
+	void __iomem *mbase;
+
+	mbase = ioremap(base, PAGE_SIZE);
+	hose->cfg_addr = mbase + (cfg_addr & ~PAGE_MASK);
+	if ((cfg_data & PAGE_MASK) != base)
+		mbase = ioremap(cfg_data & PAGE_MASK, PAGE_SIZE);
+	hose->cfg_data = mbase + (cfg_data & ~PAGE_MASK);
+	hose->ops = &indirect_pci_ops;
+	hose->indirect_type = flags;
+}

arch/microblaze/pci/iomap.c

+/*
+ * ppc64 "iomap" interface implementation.
+ *
+ * (C) Copyright 2004 Linus Torvalds
+ */
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/mm.h>
+#include <asm/io.h>
+#include <asm/pci-bridge.h>
+
+void __iomem *pci_iomap(struct pci_dev *dev, int bar, unsigned long max)
+{
+	resource_size_t start = pci_resource_start(dev, bar);
+	resource_size_t len = pci_resource_len(dev, bar);
+	unsigned long flags = pci_resource_flags(dev, bar);
+
+	if (!len)
+		return NULL;
+	if (max && len > max)
+		len = max;
+	if (flags & IORESOURCE_IO)
+		return ioport_map(start, len);
+	if (flags & IORESOURCE_MEM)
+		return ioremap(start, len);
+	/* What? */
+	return NULL;
+}
+EXPORT_SYMBOL(pci_iomap);
+
+void pci_iounmap(struct pci_dev *dev, void __iomem *addr)
+{
+	if (isa_vaddr_is_ioport(addr))
+		return;
+	if (pcibios_vaddr_is_ioport(addr))
+		return;
+	iounmap(addr);
+}
+EXPORT_SYMBOL(pci_iounmap);

arch/microblaze/pci/xilinx_pci.c

+/*
+ * PCI support for Xilinx plbv46_pci soft-core which can be used on
+ * Xilinx Virtex ML410 / ML510 boards.
+ *
+ * Copyright 2009 Roderick Colenbrander
+ * Copyright 2009 Secret Lab Technologies Ltd.
+ *
+ * The pci bridge fixup code was copied from ppc4xx_pci.c and was written
+ * by Benjamin Herrenschmidt.
+ * Copyright 2007 Ben. Herrenschmidt <benh@kernel.crashing.org>, IBM Corp.
+ *
+ * This file is licensed under the terms of the GNU General Public License
+ * version 2. This program is licensed "as is" without any warranty of any
+ * kind, whether express or implied.
+ */
+
+#include <linux/ioport.h>
+#include <linux/of.h>
+#include <linux/pci.h>
+#include <asm/io.h>
+
+#define XPLB_PCI_ADDR 0x10c
+#define XPLB_PCI_DATA 0x110
+#define XPLB_PCI_BUS  0x114
+
+#define PCI_HOST_ENABLE_CMD (PCI_COMMAND_SERR | PCI_COMMAND_PARITY | \
+				PCI_COMMAND_MASTER | PCI_COMMAND_MEMORY)
+
+static struct of_device_id xilinx_pci_match[] = {
+	{ .compatible = "xlnx,plbv46-pci-1.03.a", },
+	{}
+};
+
+/**
+ * xilinx_pci_fixup_bridge - Block Xilinx PHB configuration.
+ */
+static void xilinx_pci_fixup_bridge(struct pci_dev *dev)
+{
+	struct pci_controller *hose;
+	int i;
+
+	if (dev->devfn || dev->bus->self)
+		return;
+
+	hose = pci_bus_to_host(dev->bus);
+	if (!hose)
+		return;
+
+	if (!of_match_node(xilinx_pci_match, hose->dn))
+		return;
+
+	/* Hide the PCI host BARs from the kernel as their content doesn't
+	 * fit well in the resource management
+	 */
+	for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
+		dev->resource[i].start = 0;
+		dev->resource[i].end = 0;
+		dev->resource[i].flags = 0;
+	}
+
+	dev_info(&dev->dev, "Hiding Xilinx plb-pci host bridge resources %s\n",
+		 pci_name(dev));
+}
+DECLARE_PCI_FIXUP_HEADER(PCI_ANY_ID, PCI_ANY_ID, xilinx_pci_fixup_bridge);
+
+#ifdef DEBUG
+/**
+ * xilinx_pci_exclude_device - Don't do config access for non-root bus
+ *
+ * This is a hack.  Config access to any bus other than bus 0 does not
+ * currently work on the ML510 so we prevent it here.
+ */
+static int
+xilinx_pci_exclude_device(struct pci_controller *hose, u_char bus, u8 devfn)
+{
+	return (bus != 0);
+}
+
+/**
+ * xilinx_early_pci_scan - List pci config space for available devices
+ *
+ * List pci devices in very early phase.
+ */
+void __init xilinx_early_pci_scan(struct pci_controller *hose)
+{
+	u32 bus = 0;
+	u32 val, dev, func, offset;
+
+	/* Currently we have only 2 device connected - up-to 32 devices */
+	for (dev = 0; dev < 2; dev++) {
+		/* List only first function number - up-to 8 functions */
+		for (func = 0; func < 1; func++) {
+			printk(KERN_INFO "%02x:%02x:%02x", bus, dev, func);
+			/* read the first 64 standardized bytes */
+			/* Up-to 192 bytes can be list of capabilities */
+			for (offset = 0; offset < 64; offset += 4) {
+				early_read_config_dword(hose, bus,
+					PCI_DEVFN(dev, func), offset, &val);
+				if (offset == 0 && val == 0xFFFFFFFF) {
+					printk(KERN_CONT "\nABSENT");
+					break;
+				}
+				if (!(offset % 0x10))
+					printk(KERN_CONT "\n%04x:    ", offset);
+
+				printk(KERN_CONT "%08x  ", val);
+			}
+			printk(KERN_INFO "\n");
+		}
+	}
+}
+#else
+void __init xilinx_early_pci_scan(struct pci_controller *hose)
+{
+}
+#endif
+
+/**
+ * xilinx_pci_init - Find and register a Xilinx PCI host bridge
+ */
+void __init xilinx_pci_init(void)
+{
+	struct pci_controller *hose;
+	struct resource r;
+	void __iomem *pci_reg;
+	struct device_node *pci_node;
+
+	pci_node = of_find_matching_node(NULL, xilinx_pci_match);
+	if (!pci_node)
+		return;
+
+	if (of_address_to_resource(pci_node, 0, &r)) {
+		pr_err("xilinx-pci: cannot resolve base address\n");
+		return;
+	}
+
+	hose = pcibios_alloc_controller(pci_node);
+	if (!hose) {
+		pr_err("xilinx-pci: pcibios_alloc_controller() failed\n");
+		return;
+	}
+
+	/* Setup config space */
+	setup_indirect_pci(hose, r.start + XPLB_PCI_ADDR,
+			   r.start + XPLB_PCI_DATA,
+			   INDIRECT_TYPE_SET_CFG_TYPE);
+
+	/* According to the xilinx plbv46_pci documentation the soft-core starts
+	 * a self-init when the bus master enable bit is set. Without this bit
+	 * set the pci bus can't be scanned.
+	 */
+	early_write_config_word(hose, 0, 0, PCI_COMMAND, PCI_HOST_ENABLE_CMD);
+
+	/* Set the max latency timer to 255 */
+	early_write_config_byte(hose, 0, 0, PCI_LATENCY_TIMER, 0xff);
+
+	/* Set the max bus number to 255, and bus/subbus no's to 0 */
+	pci_reg = of_iomap(pci_node, 0);
+	out_be32(pci_reg + XPLB_PCI_BUS, 0x000000ff);
+	iounmap(pci_reg);
+
+	/* Register the host bridge with the linux kernel! */
+	pci_process_bridge_OF_ranges(hose, pci_node,
+					INDIRECT_TYPE_SET_CFG_TYPE);
+
+	pr_info("xilinx-pci: Registered PCI host bridge\n");
+	xilinx_early_pci_scan(hose);
+}

drivers/pci/Makefile

@@ -48,6 +48,7 @@ obj-$(CONFIG_PPC) += setup-bus.o
 obj-$(CONFIG_MIPS) += setup-bus.o setup-irq.o
 obj-$(CONFIG_X86_VISWS) += setup-irq.o
 obj-$(CONFIG_MN10300) += setup-bus.o
+obj-$(CONFIG_MICROBLAZE) += setup-bus.o
 
 #
 # ACPI Related PCI FW Functions