ARM: Add fixed PCI i/o mapping

This adds a fixed virtual mapping for PCI i/o addresses. The mapping is
located at the last 2MB of vmalloc region (0xfee00000-0xff000000). 2MB
is used to align with PMD size, but IO_SPACE_LIMIT is 1MB. The space
is reserved after .map_io and can be mapped at any time later with
pci_ioremap_io. Platforms which need early i/o mapping (e.g. for vga
console) can call pci_map_io_early in their .map_io function.

This has changed completely from the 1st implementation which only
supported creating the static mapping at .map_io.
Signed-off-by: Rob Herring <rob.herring@calxeda.com>
Cc: Russell King <linux@arm.linux.org.uk>
Acked-by: Nicolas Pitre <nico@linaro.org>

Documentation/arm/memory.txt

@@ -51,6 +51,9 @@ ffc00000	ffefffff	DMA memory mapping region.  Memory returned
 ff000000	ffbfffff	Reserved for future expansion of DMA
 				mapping region.
 
+fee00000	feffffff	Mapping of PCI I/O space. This is a static
+				mapping within the vmalloc space.
+
 VMALLOC_START	VMALLOC_END-1	vmalloc() / ioremap() space.
 				Memory returned by vmalloc/ioremap will
 				be dynamically placed in this region.

arch/arm/include/asm/io.h

@@ -113,11 +113,19 @@ static inline void __iomem *__typesafe_io(unsigned long addr)
 #define __iowmb()		do { } while (0)
 #endif
 
+/* PCI fixed i/o mapping */
+#define PCI_IO_VIRT_BASE	0xfee00000
+
+extern int pci_ioremap_io(unsigned int offset, phys_addr_t phys_addr);
+
 /*
  * Now, pick up the machine-defined IO definitions
  */
 #ifdef CONFIG_NEED_MACH_IO_H
 #include <mach/io.h>
+#elif defined(CONFIG_PCI)
+#define IO_SPACE_LIMIT	((resource_size_t)0xfffff)
+#define __io(a)		__typesafe_io(PCI_IO_VIRT_BASE + ((a) & IO_SPACE_LIMIT))
 #else
 #define __io(a)		__typesafe_io((a) & IO_SPACE_LIMIT)
 #endif

arch/arm/include/asm/mach/map.h

@@ -9,6 +9,9 @@
  *
  *  Page table mapping constructs and function prototypes
  */
+#ifndef __ASM_MACH_MAP_H
+#define __ASM_MACH_MAP_H
+
 #include <asm/io.h>
 
 struct map_desc {
@@ -34,6 +37,8 @@ struct map_desc {
 
 #ifdef CONFIG_MMU
 extern void iotable_init(struct map_desc *, int);
+extern void vm_reserve_area_early(unsigned long addr, unsigned long size,
+				  void *caller);
 
 struct mem_type;
 extern const struct mem_type *get_mem_type(unsigned int type);
@@ -44,4 +49,7 @@ extern int ioremap_page(unsigned long virt, unsigned long phys,
 			const struct mem_type *mtype);
 #else
 #define iotable_init(map,num)	do { } while (0)
+#define vm_reserve_area_early(a,s,c)	do { } while (0)
+#endif
+
 #endif

arch/arm/include/asm/mach/pci.h

@@ -11,6 +11,7 @@
 #ifndef __ASM_MACH_PCI_H
 #define __ASM_MACH_PCI_H
 
+
 struct pci_sys_data;
 struct pci_ops;
 struct pci_bus;
@@ -54,6 +55,15 @@ struct pci_sys_data {
  */
 void pci_common_init(struct hw_pci *);
 
+/*
+ * Setup early fixed I/O mapping.
+ */
+#if defined(CONFIG_PCI)
+extern void pci_map_io_early(unsigned long pfn);
+#else
+static inline void pci_map_io_early(unsigned long pfn) {}
+#endif
+
 /*
  * PCI controllers
  */

arch/arm/kernel/bios32.c

@@ -13,6 +13,7 @@
 #include <linux/io.h>
 
 #include <asm/mach-types.h>
+#include <asm/mach/map.h>
 #include <asm/mach/pci.h>
 
 static int debug_pci;
@@ -627,3 +628,15 @@ int pci_mmap_page_range(struct pci_dev *dev, struct vm_area_struct *vma,
 
 	return 0;
 }
+
+void __init pci_map_io_early(unsigned long pfn)
+{
+	struct map_desc pci_io_desc = {
+		.virtual	= PCI_IO_VIRT_BASE,
+		.type		= MT_DEVICE,
+		.length		= SZ_64K,
+	};
+
+	pci_io_desc.pfn = pfn;
+	iotable_init(&pci_io_desc, 1);
+}

arch/arm/mm/ioremap.c

@@ -36,6 +36,7 @@
 #include <asm/system_info.h>
 
 #include <asm/mach/map.h>
+#include <asm/mach/pci.h>
 #include "mm.h"
 
 int ioremap_page(unsigned long virt, unsigned long phys,
@@ -383,3 +384,16 @@ void __arm_iounmap(volatile void __iomem *io_addr)
 	arch_iounmap(io_addr);
 }
 EXPORT_SYMBOL(__arm_iounmap);
+
+#ifdef CONFIG_PCI
+int pci_ioremap_io(unsigned int offset, phys_addr_t phys_addr)
+{
+	BUG_ON(offset + SZ_64K > IO_SPACE_LIMIT);
+
+	return ioremap_page_range(PCI_IO_VIRT_BASE + offset,
+				  PCI_IO_VIRT_BASE + offset + SZ_64K,
+				  phys_addr,
+				  __pgprot(get_mem_type(MT_DEVICE)->prot_pte));
+}
+EXPORT_SYMBOL_GPL(pci_ioremap_io);
+#endif

arch/arm/mm/mmu.c

@@ -31,6 +31,7 @@
 
 #include <asm/mach/arch.h>
 #include <asm/mach/map.h>
+#include <asm/mach/pci.h>
 
 #include "mm.h"
 
@@ -216,7 +217,7 @@ static struct mem_type mem_types[] = {
 		.prot_l1	= PMD_TYPE_TABLE,
 		.prot_sect	= PROT_SECT_DEVICE | PMD_SECT_WB,
 		.domain		= DOMAIN_IO,
-	},	
+	},
 	[MT_DEVICE_WC] = {	/* ioremap_wc */
 		.prot_pte	= PROT_PTE_DEVICE | L_PTE_MT_DEV_WC,
 		.prot_l1	= PMD_TYPE_TABLE,
@@ -783,14 +784,27 @@ void __init iotable_init(struct map_desc *io_desc, int nr)
 		create_mapping(md);
 		vm->addr = (void *)(md->virtual & PAGE_MASK);
 		vm->size = PAGE_ALIGN(md->length + (md->virtual & ~PAGE_MASK));
-		vm->phys_addr = __pfn_to_phys(md->pfn); 
-		vm->flags = VM_IOREMAP | VM_ARM_STATIC_MAPPING; 
+		vm->phys_addr = __pfn_to_phys(md->pfn);
+		vm->flags = VM_IOREMAP | VM_ARM_STATIC_MAPPING;
 		vm->flags |= VM_ARM_MTYPE(md->type);
 		vm->caller = iotable_init;
 		vm_area_add_early(vm++);
 	}
 }
 
+void __init vm_reserve_area_early(unsigned long addr, unsigned long size,
+				  void *caller)
+{
+	struct vm_struct *vm;
+
+	vm = early_alloc_aligned(sizeof(*vm), __alignof__(*vm));
+	vm->addr = (void *)addr;
+	vm->size = size;
+	vm->flags = VM_IOREMAP | VM_ARM_STATIC_MAPPING;
+	vm->caller = caller;
+	vm_area_add_early(vm);
+}
+
 #ifndef CONFIG_ARM_LPAE
 
 /*
@@ -808,14 +822,7 @@ void __init iotable_init(struct map_desc *io_desc, int nr)
 
 static void __init pmd_empty_section_gap(unsigned long addr)
 {
-	struct vm_struct *vm;
-
-	vm = early_alloc_aligned(sizeof(*vm), __alignof__(*vm));
-	vm->addr = (void *)addr;
-	vm->size = SECTION_SIZE;
-	vm->flags = VM_IOREMAP | VM_ARM_STATIC_MAPPING;
-	vm->caller = pmd_empty_section_gap;
-	vm_area_add_early(vm);
+	vm_reserve_area_early(addr, SECTION_SIZE, pmd_empty_section_gap);
 }
 
 static void __init fill_pmd_gaps(void)
@@ -864,6 +871,28 @@ static void __init fill_pmd_gaps(void)
 #define fill_pmd_gaps() do { } while (0)
 #endif
 
+#if defined(CONFIG_PCI) && !defined(CONFIG_NEED_MACH_IO_H)
+static void __init pci_reserve_io(void)
+{
+	struct vm_struct *vm;
+	unsigned long addr;
+
+	/* we're still single threaded hence no lock needed here */
+	for (vm = vmlist; vm; vm = vm->next) {
+		if (!(vm->flags & VM_ARM_STATIC_MAPPING))
+			continue;
+		addr = (unsigned long)vm->addr;
+		addr &= ~(SZ_2M - 1);
+		if (addr == PCI_IO_VIRT_BASE)
+			return;
+
+	}
+	vm_reserve_area_early(PCI_IO_VIRT_BASE, SZ_2M, pci_reserve_io);
+}
+#else
+#define pci_reserve_io() do { } while (0)
+#endif
+
 static void * __initdata vmalloc_min =
 	(void *)(VMALLOC_END - (240 << 20) - VMALLOC_OFFSET);
 
@@ -1147,6 +1176,9 @@ static void __init devicemaps_init(struct machine_desc *mdesc)
 		mdesc->map_io();
 	fill_pmd_gaps();
 
+	/* Reserve fixed i/o space in VMALLOC region */
+	pci_reserve_io();
+
 	/*
 	 * Finally flush the caches and tlb to ensure that we're in a
 	 * consistent state wrt the writebuffer.  This also ensures that