[linux-2.6-block.git] / arch / powerpc / mm / pgtable_64.c

/*
 *  This file contains ioremap and related functions for 64-bit machines.
 *
 *  Derived from arch/ppc64/mm/init.c
 *    Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
 *
 *  Modifications by Paul Mackerras (PowerMac) (paulus@samba.org)
 *  and Cort Dougan (PReP) (cort@cs.nmt.edu)
 *    Copyright (C) 1996 Paul Mackerras
 *  Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk).
 *
 *  Derived from "arch/i386/mm/init.c"
 *    Copyright (C) 1991, 1992, 1993, 1994  Linus Torvalds
 *
 *  Dave Engebretsen <engebret@us.ibm.com>
 *      Rework for PPC64 port.
 *
 *  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/signal.h>
#include <linux/sched.h>
#include <linux/kernel.h>
#include <linux/errno.h>
#include <linux/string.h>
#include <linux/types.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/idr.h>
#include <linux/nodemask.h>
#include <linux/module.h>

#include <asm/pgalloc.h>
#include <asm/page.h>
#include <asm/prom.h>
#include <asm/lmb.h>
#include <asm/rtas.h>
#include <asm/io.h>
#include <asm/mmu_context.h>
#include <asm/pgtable.h>
#include <asm/mmu.h>
#include <asm/uaccess.h>
#include <asm/smp.h>
#include <asm/machdep.h>
#include <asm/tlb.h>
#include <asm/eeh.h>
#include <asm/processor.h>
#include <asm/mmzone.h>
#include <asm/cputable.h>
#include <asm/sections.h>
#include <asm/system.h>
#include <asm/iommu.h>
#include <asm/abs_addr.h>
#include <asm/vdso.h>
#include <asm/firmware.h>

#include "mmu_decl.h"

unsigned long ioremap_bot = IMALLOC_BASE;
static unsigned long phbs_io_bot = PHBS_IO_BASE;

/*
 * map_io_page currently only called by __ioremap
 * map_io_page adds an entry to the ioremap page table
 * and adds an entry to the HPT, possibly bolting it
 */
static int map_io_page(unsigned long ea, unsigned long pa, int flags)
{
	pgd_t *pgdp;
	pud_t *pudp;
	pmd_t *pmdp;
	pte_t *ptep;

	if (mem_init_done) {
		pgdp = pgd_offset_k(ea);
		pudp = pud_alloc(&init_mm, pgdp, ea);
		if (!pudp)
			return -ENOMEM;
		pmdp = pmd_alloc(&init_mm, pudp, ea);
		if (!pmdp)
			return -ENOMEM;
		ptep = pte_alloc_kernel(pmdp, ea);
		if (!ptep)
			return -ENOMEM;
		set_pte_at(&init_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT,
							  __pgprot(flags)));
	} else {
		/*
		 * If the mm subsystem is not fully up, we cannot create a
		 * linux page table entry for this mapping.  Simply bolt an
		 * entry in the hardware page table.
		 *
		 */
		if (htab_bolt_mapping(ea, ea + PAGE_SIZE, pa, flags,
				      mmu_io_psize)) {
			printk(KERN_ERR "Failed to do bolted mapping IO "
			       "memory at %016lx !\n", pa);
			return -ENOMEM;
		}
	}
	return 0;
}


static void __iomem * __ioremap_com(phys_addr_t addr, unsigned long pa,
			    unsigned long ea, unsigned long size,
			    unsigned long flags)
{
	unsigned long i;

	if ((flags & _PAGE_PRESENT) == 0)
		flags |= pgprot_val(PAGE_KERNEL);

	for (i = 0; i < size; i += PAGE_SIZE)
		if (map_io_page(ea+i, pa+i, flags))
			return NULL;

	return (void __iomem *) (ea + (addr & ~PAGE_MASK));
}

void __iomem * __ioremap(phys_addr_t addr, unsigned long size,
			 unsigned long flags)
{
	unsigned long pa, ea;
	void __iomem *ret;

	/*
	 * Choose an address to map it to.
	 * Once the imalloc system is running, we use it.
	 * Before that, we map using addresses going
	 * up from ioremap_bot.  imalloc will use
	 * the addresses from ioremap_bot through
	 * IMALLOC_END
	 * 
	 */
	pa = addr & PAGE_MASK;
	size = PAGE_ALIGN(addr + size) - pa;

	if ((size == 0) || (pa == 0))
		return NULL;

	if (mem_init_done) {
		struct vm_struct *area;
		area = im_get_free_area(size);
		if (area == NULL)
			return NULL;
		ea = (unsigned long)(area->addr);
		ret = __ioremap_com(addr, pa, ea, size, flags);
		if (!ret)
			im_free(area->addr);
	} else {
		ea = ioremap_bot;
		ret = __ioremap_com(addr, pa, ea, size, flags);
		if (ret)
			ioremap_bot += size;
	}
	return ret;
}


void __iomem * ioremap(phys_addr_t addr, unsigned long size)
{
	unsigned long flags = _PAGE_NO_CACHE | _PAGE_GUARDED;

	if (ppc_md.ioremap)
		return ppc_md.ioremap(addr, size, flags);
	return __ioremap(addr, size, flags);
}

void __iomem * ioremap_flags(phys_addr_t addr, unsigned long size,
			     unsigned long flags)
{
	if (ppc_md.ioremap)
		return ppc_md.ioremap(addr, size, flags);
	return __ioremap(addr, size, flags);
}


#define IS_PAGE_ALIGNED(_val) ((_val) == ((_val) & PAGE_MASK))

int __ioremap_explicit(phys_addr_t pa, unsigned long ea,
		       unsigned long size, unsigned long flags)
{
	struct vm_struct *area;
	void __iomem *ret;
	
	/* For now, require page-aligned values for pa, ea, and size */
	if (!IS_PAGE_ALIGNED(pa) || !IS_PAGE_ALIGNED(ea) ||
	    !IS_PAGE_ALIGNED(size)) {
		printk(KERN_ERR	"unaligned value in %s\n", __FUNCTION__);
		return 1;
	}
	
	if (!mem_init_done) {
		/* Two things to consider in this case:
		 * 1) No records will be kept (imalloc, etc) that the region
		 *    has been remapped
		 * 2) It won't be easy to iounmap() the region later (because
		 *    of 1)
		 */
		;
	} else {
		area = im_get_area(ea, size,
			IM_REGION_UNUSED|IM_REGION_SUBSET|IM_REGION_EXISTS);
		if (area == NULL) {
			/* Expected when PHB-dlpar is in play */
			return 1;
		}
		if (ea != (unsigned long) area->addr) {
			printk(KERN_ERR "unexpected addr return from "
			       "im_get_area\n");
			return 1;
		}
	}
	
	ret = __ioremap_com(pa, pa, ea, size, flags);
	if (ret == NULL) {
		printk(KERN_ERR "ioremap_explicit() allocation failure !\n");
		return 1;
	}
	if (ret != (void *) ea) {
		printk(KERN_ERR "__ioremap_com() returned unexpected addr\n");
		return 1;
	}

	return 0;
}

/*  
 * Unmap an IO region and remove it from imalloc'd list.
 * Access to IO memory should be serialized by driver.
 *
 * XXX	what about calls before mem_init_done (ie python_countermeasures())
 */
void __iounmap(volatile void __iomem *token)
{
	void *addr;

	if (!mem_init_done)
		return;
	
	addr = (void *) ((unsigned long __force) token & PAGE_MASK);

	im_free(addr);
}

void iounmap(volatile void __iomem *token)
{
	if (ppc_md.iounmap)
		ppc_md.iounmap(token);
	else
		__iounmap(token);
}

static int iounmap_subset_regions(unsigned long addr, unsigned long size)
{
	struct vm_struct *area;

	/* Check whether subsets of this region exist */
	area = im_get_area(addr, size, IM_REGION_SUPERSET);
	if (area == NULL)
		return 1;

	while (area) {
		iounmap((void __iomem *) area->addr);
		area = im_get_area(addr, size,
				IM_REGION_SUPERSET);
	}

	return 0;
}

int __iounmap_explicit(volatile void __iomem *start, unsigned long size)
{
	struct vm_struct *area;
	unsigned long addr;
	int rc;
	
	addr = (unsigned long __force) start & PAGE_MASK;

	/* Verify that the region either exists or is a subset of an existing
	 * region.  In the latter case, split the parent region to create 
	 * the exact region 
	 */
	area = im_get_area(addr, size, 
			    IM_REGION_EXISTS | IM_REGION_SUBSET);
	if (area == NULL) {
		/* Determine whether subset regions exist.  If so, unmap */
		rc = iounmap_subset_regions(addr, size);
		if (rc) {
			printk(KERN_ERR
			       "%s() cannot unmap nonexistent range 0x%lx\n",
 				__FUNCTION__, addr);
			return 1;
		}
	} else {
		iounmap((void __iomem *) area->addr);
	}
	/*
	 * FIXME! This can't be right:
	iounmap(area->addr);
	 * Maybe it should be "iounmap(area);"
	 */
	return 0;
}

EXPORT_SYMBOL(ioremap);
EXPORT_SYMBOL(ioremap_flags);
EXPORT_SYMBOL(__ioremap);
EXPORT_SYMBOL(iounmap);
EXPORT_SYMBOL(__iounmap);

static DEFINE_SPINLOCK(phb_io_lock);

void __iomem * reserve_phb_iospace(unsigned long size)
{
	void __iomem *virt_addr;
		
	if (phbs_io_bot >= IMALLOC_BASE) 
		panic("reserve_phb_iospace(): phb io space overflow\n");
			
	spin_lock(&phb_io_lock);
	virt_addr = (void __iomem *) phbs_io_bot;
	phbs_io_bot += size;
	spin_unlock(&phb_io_lock);

	return virt_addr;
}
Commit	Line	Data
14cf11af PM	1	/*
	2	* This file contains ioremap and related functions for 64-bit machines.
	3	*
	4	* Derived from arch/ppc64/mm/init.c
	5	* Copyright (C) 1995-1996 Gary Thomas (gdt@linuxppc.org)
	6	*
	7	* Modifications by Paul Mackerras (PowerMac) (paulus@samba.org)
	8	* and Cort Dougan (PReP) (cort@cs.nmt.edu)
	9	* Copyright (C) 1996 Paul Mackerras
	10	* Amiga/APUS changes by Jesper Skov (jskov@cygnus.co.uk).
	11	*
	12	* Derived from "arch/i386/mm/init.c"
	13	* Copyright (C) 1991, 1992, 1993, 1994 Linus Torvalds
	14	*
	15	* Dave Engebretsen <engebret@us.ibm.com>
	16	* Rework for PPC64 port.
	17	*
	18	* This program is free software; you can redistribute it and/or
	19	* modify it under the terms of the GNU General Public License
	20	* as published by the Free Software Foundation; either version
	21	* 2 of the License, or (at your option) any later version.
	22	*
	23	*/
	24
14cf11af PM	25	#include <linux/signal.h>
	26	#include <linux/sched.h>
	27	#include <linux/kernel.h>
	28	#include <linux/errno.h>
	29	#include <linux/string.h>
	30	#include <linux/types.h>
	31	#include <linux/mman.h>
	32	#include <linux/mm.h>
	33	#include <linux/swap.h>
	34	#include <linux/stddef.h>
	35	#include <linux/vmalloc.h>
	36	#include <linux/init.h>
	37	#include <linux/delay.h>
	38	#include <linux/bootmem.h>
	39	#include <linux/highmem.h>
	40	#include <linux/idr.h>
	41	#include <linux/nodemask.h>
	42	#include <linux/module.h>
	43
	44	#include <asm/pgalloc.h>
	45	#include <asm/page.h>
	46	#include <asm/prom.h>
	47	#include <asm/lmb.h>
	48	#include <asm/rtas.h>
	49	#include <asm/io.h>
	50	#include <asm/mmu_context.h>
	51	#include <asm/pgtable.h>
	52	#include <asm/mmu.h>
	53	#include <asm/uaccess.h>
	54	#include <asm/smp.h>
	55	#include <asm/machdep.h>
	56	#include <asm/tlb.h>
	57	#include <asm/eeh.h>
	58	#include <asm/processor.h>
	59	#include <asm/mmzone.h>
	60	#include <asm/cputable.h>
14cf11af PM	61	#include <asm/sections.h>
	62	#include <asm/system.h>
	63	#include <asm/iommu.h>
	64	#include <asm/abs_addr.h>
	65	#include <asm/vdso.h>
5e203d68	66	#include <asm/firmware.h>
800fc3ee DG	67
800fc3ee DG	68	#include "mmu_decl.h"
14cf11af	69
14cf11af PM	70	unsigned long ioremap_bot = IMALLOC_BASE;
	71	static unsigned long phbs_io_bot = PHBS_IO_BASE;
	72
14cf11af PM	73	/*
	74	* map_io_page currently only called by __ioremap
	75	* map_io_page adds an entry to the ioremap page table
	76	* and adds an entry to the HPT, possibly bolting it
	77	*/
	78	static int map_io_page(unsigned long ea, unsigned long pa, int flags)
	79	{
	80	pgd_t *pgdp;
	81	pud_t *pudp;
	82	pmd_t *pmdp;
	83	pte_t *ptep;
14cf11af PM	84
14cf11af PM	85	if (mem_init_done) {
14cf11af PM	86	pgdp = pgd_offset_k(ea);
	87	pudp = pud_alloc(&init_mm, pgdp, ea);
	88	if (!pudp)
	89	return -ENOMEM;
	90	pmdp = pmd_alloc(&init_mm, pudp, ea);
	91	if (!pmdp)
	92	return -ENOMEM;
23fd0775	93	ptep = pte_alloc_kernel(pmdp, ea);
14cf11af PM	94	if (!ptep)
	95	return -ENOMEM;
	96	set_pte_at(&init_mm, ea, ptep, pfn_pte(pa >> PAGE_SHIFT,
	97	__pgprot(flags)));
14cf11af	98	} else {
14cf11af PM	99	/*
	100	* If the mm subsystem is not fully up, we cannot create a
	101	* linux page table entry for this mapping. Simply bolt an
	102	* entry in the hardware page table.
3c726f8d	103	*
14cf11af	104	*/
3c726f8d	105	if (htab_bolt_mapping(ea, ea + PAGE_SIZE, pa, flags,
a3252544	106	mmu_io_psize)) {
77ac166f BH	107	printk(KERN_ERR "Failed to do bolted mapping IO "
	108	"memory at %016lx !\n", pa);
	109	return -ENOMEM;
	110	}
14cf11af PM	111	}
	112	return 0;
	113	}
	114
	115
68a64357	116	static void __iomem * __ioremap_com(phys_addr_t addr, unsigned long pa,
14cf11af PM	117	unsigned long ea, unsigned long size,
	118	unsigned long flags)
	119	{
	120	unsigned long i;
	121
	122	if ((flags & _PAGE_PRESENT) == 0)
	123	flags \|= pgprot_val(PAGE_KERNEL);
	124
	125	for (i = 0; i < size; i += PAGE_SIZE)
	126	if (map_io_page(ea+i, pa+i, flags))
	127	return NULL;
	128
	129	return (void __iomem *) (ea + (addr & ~PAGE_MASK));
	130	}
	131
68a64357	132	void __iomem * __ioremap(phys_addr_t addr, unsigned long size,
14cf11af PM	133	unsigned long flags)
	134	{
	135	unsigned long pa, ea;
	136	void __iomem *ret;
	137
	138	/*
	139	* Choose an address to map it to.
	140	* Once the imalloc system is running, we use it.
	141	* Before that, we map using addresses going
	142	* up from ioremap_bot. imalloc will use
	143	* the addresses from ioremap_bot through
	144	* IMALLOC_END
	145	*
	146	*/
	147	pa = addr & PAGE_MASK;
	148	size = PAGE_ALIGN(addr + size) - pa;
	149
d177c207	150	if ((size == 0) \|\| (pa == 0))
14cf11af PM	151	return NULL;
	152
	153	if (mem_init_done) {
	154	struct vm_struct *area;
	155	area = im_get_free_area(size);
	156	if (area == NULL)
	157	return NULL;
	158	ea = (unsigned long)(area->addr);
	159	ret = __ioremap_com(addr, pa, ea, size, flags);
	160	if (!ret)
	161	im_free(area->addr);
	162	} else {
	163	ea = ioremap_bot;
	164	ret = __ioremap_com(addr, pa, ea, size, flags);
	165	if (ret)
	166	ioremap_bot += size;
	167	}
	168	return ret;
	169	}
	170
4cb3cee0	171
68a64357	172	void __iomem * ioremap(phys_addr_t addr, unsigned long size)
4cb3cee0 BH	173	{
	174	unsigned long flags = _PAGE_NO_CACHE \| _PAGE_GUARDED;
	175
	176	if (ppc_md.ioremap)
	177	return ppc_md.ioremap(addr, size, flags);
	178	return __ioremap(addr, size, flags);
	179	}
	180
68a64357	181	void __iomem * ioremap_flags(phys_addr_t addr, unsigned long size,
4cb3cee0 BH	182	unsigned long flags)
	183	{
	184	if (ppc_md.ioremap)
	185	return ppc_md.ioremap(addr, size, flags);
	186	return __ioremap(addr, size, flags);
	187	}
	188
	189
14cf11af PM	190	#define IS_PAGE_ALIGNED(_val) ((_val) == ((_val) & PAGE_MASK))
14cf11af PM	191
68a64357	192	int __ioremap_explicit(phys_addr_t pa, unsigned long ea,
14cf11af PM	193	unsigned long size, unsigned long flags)
	194	{
	195	struct vm_struct *area;
	196	void __iomem *ret;
	197
	198	/* For now, require page-aligned values for pa, ea, and size */
	199	if (!IS_PAGE_ALIGNED(pa) \|\| !IS_PAGE_ALIGNED(ea) \|\|
	200	!IS_PAGE_ALIGNED(size)) {
	201	printk(KERN_ERR "unaligned value in %s\n", __FUNCTION__);
	202	return 1;
	203	}
	204
	205	if (!mem_init_done) {
	206	/* Two things to consider in this case:
	207	* 1) No records will be kept (imalloc, etc) that the region
	208	* has been remapped
	209	* 2) It won't be easy to iounmap() the region later (because
	210	* of 1)
	211	*/
	212	;
	213	} else {
	214	area = im_get_area(ea, size,
	215	IM_REGION_UNUSED\|IM_REGION_SUBSET\|IM_REGION_EXISTS);
	216	if (area == NULL) {
	217	/* Expected when PHB-dlpar is in play */
	218	return 1;
	219	}
	220	if (ea != (unsigned long) area->addr) {
	221	printk(KERN_ERR "unexpected addr return from "
	222	"im_get_area\n");
	223	return 1;
	224	}
	225	}
	226
	227	ret = __ioremap_com(pa, pa, ea, size, flags);
	228	if (ret == NULL) {
	229	printk(KERN_ERR "ioremap_explicit() allocation failure !\n");
	230	return 1;
	231	}
	232	if (ret != (void *) ea) {
	233	printk(KERN_ERR "__ioremap_com() returned unexpected addr\n");
	234	return 1;
	235	}
	236
	237	return 0;
	238	}
	239
	240	/*
	241	* Unmap an IO region and remove it from imalloc'd list.
	242	* Access to IO memory should be serialized by driver.
14cf11af PM	243	*
	244	* XXX what about calls before mem_init_done (ie python_countermeasures())
	245	*/
68a64357	246	void __iounmap(volatile void __iomem *token)
14cf11af PM	247	{
	248	void *addr;
	249
	250	if (!mem_init_done)
	251	return;
	252
	253	addr = (void *) ((unsigned long __force) token & PAGE_MASK);
	254
	255	im_free(addr);
	256	}
	257
68a64357	258	void iounmap(volatile void __iomem *token)
4cb3cee0 BH	259	{
	260	if (ppc_md.iounmap)
	261	ppc_md.iounmap(token);
	262	else
	263	__iounmap(token);
	264	}
	265
14cf11af PM	266	static int iounmap_subset_regions(unsigned long addr, unsigned long size)
	267	{
	268	struct vm_struct *area;
	269
	270	/* Check whether subsets of this region exist */
	271	area = im_get_area(addr, size, IM_REGION_SUPERSET);
	272	if (area == NULL)
	273	return 1;
	274
	275	while (area) {
	276	iounmap((void __iomem *) area->addr);
	277	area = im_get_area(addr, size,
	278	IM_REGION_SUPERSET);
	279	}
	280
	281	return 0;
	282	}
	283
68a64357	284	int __iounmap_explicit(volatile void __iomem *start, unsigned long size)
14cf11af PM	285	{
	286	struct vm_struct *area;
	287	unsigned long addr;
	288	int rc;
	289
	290	addr = (unsigned long __force) start & PAGE_MASK;
	291
	292	/* Verify that the region either exists or is a subset of an existing
	293	* region. In the latter case, split the parent region to create
	294	* the exact region
	295	*/
	296	area = im_get_area(addr, size,
	297	IM_REGION_EXISTS \| IM_REGION_SUBSET);
	298	if (area == NULL) {
	299	/* Determine whether subset regions exist. If so, unmap */
	300	rc = iounmap_subset_regions(addr, size);
	301	if (rc) {
	302	printk(KERN_ERR
	303	"%s() cannot unmap nonexistent range 0x%lx\n",
	304	__FUNCTION__, addr);
	305	return 1;
	306	}
	307	} else {
	308	iounmap((void __iomem *) area->addr);
	309	}
	310	/*
	311	* FIXME! This can't be right:
	312	iounmap(area->addr);
	313	* Maybe it should be "iounmap(area);"
	314	*/
	315	return 0;
	316	}
	317
14cf11af	318	EXPORT_SYMBOL(ioremap);
4cb3cee0	319	EXPORT_SYMBOL(ioremap_flags);
14cf11af PM	320	EXPORT_SYMBOL(__ioremap);
14cf11af PM	321	EXPORT_SYMBOL(iounmap);
4cb3cee0	322	EXPORT_SYMBOL(__iounmap);
ab1f9dac	323
017e3c53 BH	324	static DEFINE_SPINLOCK(phb_io_lock);
017e3c53 BH	325
ab1f9dac PM	326	void __iomem * reserve_phb_iospace(unsigned long size)
	327	{
	328	void __iomem *virt_addr;
	329
	330	if (phbs_io_bot >= IMALLOC_BASE)
	331	panic("reserve_phb_iospace(): phb io space overflow\n");
	332
017e3c53	333	spin_lock(&phb_io_lock);
ab1f9dac PM	334	virt_addr = (void __iomem *) phbs_io_bot;
ab1f9dac PM	335	phbs_io_bot += size;
017e3c53	336	spin_unlock(&phb_io_lock);
ab1f9dac PM	337
	338	return virt_addr;
	339	}