hash_utils_64.c

/*
 * PowerPC64 port by Mike Corrigan and Dave Engebretsen
 *   {mikejc|engebret}@us.ibm.com
 *
 *    Copyright (c) 2000 Mike Corrigan <mikejc@us.ibm.com>
 *
 * SMP scalability work:
 *    Copyright (C) 2001 Anton Blanchard <anton@au.ibm.com>, IBM
 * 
 *    Module name: htab.c
 *
 *    Description:
 *      PowerPC Hashed Page Table functions
 *
 * 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.
 */

#undef DEBUG
#undef DEBUG_LOW

#include <linux/spinlock.h>
#include <linux/errno.h>
#include <linux/sched.h>
#include <linux/proc_fs.h>
#include <linux/stat.h>
#include <linux/sysctl.h>
#include <linux/export.h>
#include <linux/ctype.h>
#include <linux/cache.h>
#include <linux/init.h>
#include <linux/signal.h>
#include <linux/memblock.h>
#include <linux/context_tracking.h>

#include <asm/processor.h>
#include <asm/pgtable.h>
#include <asm/mmu.h>
#include <asm/mmu_context.h>
#include <asm/page.h>
#include <asm/types.h>
#include <asm/uaccess.h>
#include <asm/machdep.h>
#include <asm/prom.h>
#include <asm/tlbflush.h>
#include <asm/io.h>
#include <asm/eeh.h>
#include <asm/tlb.h>
#include <asm/cacheflush.h>
#include <asm/cputable.h>
#include <asm/sections.h>
#include <asm/copro.h>
#include <asm/udbg.h>
#include <asm/code-patching.h>
#include <asm/fadump.h>
#include <asm/firmware.h>
#include <asm/tm.h>
#include <asm/trace.h>

#ifdef DEBUG
#define DBG(fmt...) udbg_printf(fmt)
#else
#define DBG(fmt...)
#endif

#ifdef DEBUG_LOW
#define DBG_LOW(fmt...) udbg_printf(fmt)
#else
#define DBG_LOW(fmt...)
#endif

#define KB (1024)
#define MB (1024*KB)
#define GB (1024L*MB)

/*
 * Note:  pte   --> Linux PTE
 *        HPTE  --> PowerPC Hashed Page Table Entry
 *
 * Execution context:
 *   htab_initialize is called with the MMU off (of course), but
 *   the kernel has been copied down to zero so it can directly
 *   reference global data.  At this point it is very difficult
 *   to print debug info.
 *
 */

#ifdef CONFIG_U3_DART
extern unsigned long dart_tablebase;
#endif /* CONFIG_U3_DART */

static unsigned long _SDR1;
struct mmu_psize_def mmu_psize_defs[MMU_PAGE_COUNT];
EXPORT_SYMBOL_GPL(mmu_psize_defs);

struct hash_pte *htab_address;
unsigned long htab_size_bytes;
unsigned long htab_hash_mask;
EXPORT_SYMBOL_GPL(htab_hash_mask);
int mmu_linear_psize = MMU_PAGE_4K;
EXPORT_SYMBOL_GPL(mmu_linear_psize);
int mmu_virtual_psize = MMU_PAGE_4K;
int mmu_vmalloc_psize = MMU_PAGE_4K;
#ifdef CONFIG_SPARSEMEM_VMEMMAP
int mmu_vmemmap_psize = MMU_PAGE_4K;
#endif
int mmu_io_psize = MMU_PAGE_4K;
int mmu_kernel_ssize = MMU_SEGSIZE_256M;
EXPORT_SYMBOL_GPL(mmu_kernel_ssize);
int mmu_highuser_ssize = MMU_SEGSIZE_256M;
u16 mmu_slb_size = 64;
EXPORT_SYMBOL_GPL(mmu_slb_size);
#ifdef CONFIG_PPC_64K_PAGES
int mmu_ci_restrictions;
#endif
#ifdef CONFIG_DEBUG_PAGEALLOC
static u8 *linear_map_hash_slots;
static unsigned long linear_map_hash_count;
static DEFINE_SPINLOCK(linear_map_hash_lock);
#endif /* CONFIG_DEBUG_PAGEALLOC */

/* There are definitions of page sizes arrays to be used when none
 * is provided by the firmware.
 */

/* Pre-POWER4 CPUs (4k pages only)
 */
static struct mmu_psize_def mmu_psize_defaults_old[] = {
	[MMU_PAGE_4K] = {
		.shift	= 12,
		.sllp	= 0,
		.penc   = {[MMU_PAGE_4K] = 0, [1 ... MMU_PAGE_COUNT - 1] = -1},
		.avpnm	= 0,
		.tlbiel = 0,
	},
};

/* POWER4, GPUL, POWER5
 *
 * Support for 16Mb large pages
 */
static struct mmu_psize_def mmu_psize_defaults_gp[] = {
	[MMU_PAGE_4K] = {
		.shift	= 12,
		.sllp	= 0,
		.penc   = {[MMU_PAGE_4K] = 0, [1 ... MMU_PAGE_COUNT - 1] = -1},
		.avpnm	= 0,
		.tlbiel = 1,
	},
	[MMU_PAGE_16M] = {
		.shift	= 24,
		.sllp	= SLB_VSID_L,
		.penc   = {[0 ... MMU_PAGE_16M - 1] = -1, [MMU_PAGE_16M] = 0,
			    [MMU_PAGE_16M + 1 ... MMU_PAGE_COUNT - 1] = -1 },
		.avpnm	= 0x1UL,
		.tlbiel = 0,
	},
};

/*
 * 'R' and 'C' update notes:
 *  - Under pHyp or KVM, the updatepp path will not set C, thus it *will*
 *     create writeable HPTEs without C set, because the hcall H_PROTECT
 *     that we use in that case will not update C
 *  - The above is however not a problem, because we also don't do that
 *     fancy "no flush" variant of eviction and we use H_REMOVE which will
 *     do the right thing and thus we don't have the race I described earlier
 *
 *    - Under bare metal,  we do have the race, so we need R and C set
 *    - We make sure R is always set and never lost
 *    - C is _PAGE_DIRTY, and *should* always be set for a writeable mapping
 */
unsigned long htab_convert_pte_flags(unsigned long pteflags)
{
	unsigned long rflags = 0;

	/* _PAGE_EXEC -> NOEXEC */
	if ((pteflags & _PAGE_EXEC) == 0)
		rflags |= HPTE_R_N;
	/*
	 * PPP bits:
	 * Linux uses slb key 0 for kernel and 1 for user.
	 * kernel RW areas are mapped with PPP=0b000
	 * User area is mapped with PPP=0b010 for read/write
	 * or PPP=0b011 for read-only (including writeable but clean pages).
	 */
	if (pteflags & _PAGE_PRIVILEGED) {
		/*
		 * Kernel read only mapped with ppp bits 0b110
		 */
		if (!(pteflags & _PAGE_WRITE))
			rflags |= (HPTE_R_PP0 | 0x2);
	} else {
		if (pteflags & _PAGE_RWX)
			rflags |= 0x2;
		if (!((pteflags & _PAGE_WRITE) && (pteflags & _PAGE_DIRTY)))
			rflags |= 0x1;
	}
	/*
	 * We can't allow hardware to update hpte bits. Hence always
	 * set 'R' bit and set 'C' if it is a write fault
	 */
	rflags |=  HPTE_R_R;

	if (pteflags & _PAGE_DIRTY)
		rflags |= HPTE_R_C;
	/*
	 * Add in WIG bits
	 */

	if ((pteflags & _PAGE_CACHE_CTL) == _PAGE_TOLERANT)
		rflags |= HPTE_R_I;
	else if ((pteflags & _PAGE_CACHE_CTL) == _PAGE_NON_IDEMPOTENT)
		rflags |= (HPTE_R_I | HPTE_R_G);
	else if ((pteflags & _PAGE_CACHE_CTL) == _PAGE_SAO)
		rflags |= (HPTE_R_W | HPTE_R_I | HPTE_R_M);
	else
		/*
		 * Add memory coherence if cache inhibited is not set
		 */
		rflags |= HPTE_R_M;

	return rflags;
}

int htab_bolt_mapping(unsigned long vstart, unsigned long vend,
		      unsigned long pstart, unsigned long prot,
		      int psize, int ssize)
{
	unsigned long vaddr, paddr;
	unsigned int step, shift;
	int ret = 0;

	shift = mmu_psize_defs[psize].shift;
	step = 1 << shift;

	prot = htab_convert_pte_flags(prot);

	DBG("htab_bolt_mapping(%lx..%lx -> %lx (%lx,%d,%d)\n",
	    vstart, vend, pstart, prot, psize, ssize);

	for (vaddr = vstart, paddr = pstart; vaddr < vend;
	     vaddr += step, paddr += step) {
		unsigned long hash, hpteg;
		unsigned long vsid = get_kernel_vsid(vaddr, ssize);
		unsigned long vpn  = hpt_vpn(vaddr, vsid, ssize);
		unsigned long tprot = prot;

		/*
		 * If we hit a bad address return error.
		 */
		if (!vsid)
			return -1;
		/* Make kernel text executable */
		if (overlaps_kernel_text(vaddr, vaddr + step))
			tprot &= ~HPTE_R_N;

		/* Make kvm guest trampolines executable */
		if (overlaps_kvm_tmp(vaddr, vaddr + step))
			tprot &= ~HPTE_R_N;

		/*
		 * If relocatable, check if it overlaps interrupt vectors that
		 * are copied down to real 0. For relocatable kernel
		 * (e.g. kdump case) we copy interrupt vectors down to real
		 * address 0. Mark that region as executable. This is
		 * because on p8 system with relocation on exception feature
		 * enabled, exceptions are raised with MMU (IR=DR=1) ON. Hence
		 * in order to execute the interrupt handlers in virtual
		 * mode the vector region need to be marked as executable.
		 */
		if ((PHYSICAL_START > MEMORY_START) &&
			overlaps_interrupt_vector_text(vaddr, vaddr + step))
				tprot &= ~HPTE_R_N;

		hash = hpt_hash(vpn, shift, ssize);
		hpteg = ((hash & htab_hash_mask) * HPTES_PER_GROUP);

		BUG_ON(!ppc_md.hpte_insert);
		ret = ppc_md.hpte_insert(hpteg, vpn, paddr, tprot,
					 HPTE_V_BOLTED, psize, psize, ssize);

		if (ret < 0)
			break;

#ifdef CONFIG_DEBUG_PAGEALLOC
		if (debug_pagealloc_enabled() &&
			(paddr >> PAGE_SHIFT) < linear_map_hash_count)
			linear_map_hash_slots[paddr >> PAGE_SHIFT] = ret | 0x80;
#endif /* CONFIG_DEBUG_PAGEALLOC */
	}
	return ret < 0 ? ret : 0;
}

int htab_remove_mapping(unsigned long vstart, unsigned long vend,
		      int psize, int ssize)
{
	unsigned long vaddr;
	unsigned int step, shift;
	int rc;
	int ret = 0;

	shift = mmu_psize_defs[psize].shift;
	step = 1 << shift;

	if (!ppc_md.hpte_removebolted)
		return -ENODEV;

	for (vaddr = vstart; vaddr < vend; vaddr += step) {
		rc = ppc_md.hpte_removebolted(vaddr, psize, ssize);
		if (rc == -ENOENT) {
			ret = -ENOENT;
			continue;
		}
		if (rc < 0)
			return rc;
	}

	return ret;
}

static int __init htab_dt_scan_seg_sizes(unsigned long node,
					 const char *uname, int depth,
					 void *data)
{
	const char *type = of_get_flat_dt_prop(node, "device_type", NULL);
	const __be32 *prop;
	int size = 0;

	/* We are scanning "cpu" nodes only */
	if (type == NULL || strcmp(type, "cpu") != 0)
		return 0;

	prop = of_get_flat_dt_prop(node, "ibm,processor-segment-sizes", &size);
	if (prop == NULL)
		return 0;
	for (; size >= 4; size -= 4, ++prop) {
		if (be32_to_cpu(prop[0]) == 40) {
			DBG("1T segment support detected\n");
			cur_cpu_spec->mmu_features |= MMU_FTR_1T_SEGMENT;
			return 1;
		}
	}
	cur_cpu_spec->mmu_features &= ~MMU_FTR_NO_SLBIE_B;
	return 0;
}

static void __init htab_init_seg_sizes(void)
{
	of_scan_flat_dt(htab_dt_scan_seg_sizes, NULL);
}

static int __init get_idx_from_shift(unsigned int shift)
{
	int idx = -1;

	switch (shift) {
	case 0xc:
		idx = MMU_PAGE_4K;
		break;
	case 0x10:
		idx = MMU_PAGE_64K;
		break;
	case 0x14:
		idx = MMU_PAGE_1M;
		break;
	case 0x18:
		idx = MMU_PAGE_16M;
		break;
	case 0x22:
		idx = MMU_PAGE_16G;
		break;
	}
	return idx;
}

static int __init htab_dt_scan_page_sizes(unsigned long node,
					  const char *uname, int depth,
					  void *data)
{
	const char *type = of_get_flat_dt_prop(node, "device_type", NULL);
	const __be32 *prop;
	int size = 0;

	/* We are scanning "cpu" nodes only */
	if (type == NULL || strcmp(type, "cpu") != 0)
		return 0;

	prop = of_get_flat_dt_prop(node, "ibm,segment-page-sizes", &size);
	if (!prop)
		return 0;

	pr_info("Page sizes from device-tree:\n");
	size /= 4;
	cur_cpu_spec->mmu_features &= ~(MMU_FTR_16M_PAGE);
	while(size > 0) {
		unsigned int base_shift = be32_to_cpu(prop[0]);
		unsigned int slbenc = be32_to_cpu(prop[1]);
		unsigned int lpnum = be32_to_cpu(prop[2]);
		struct mmu_psize_def *def;
		int idx, base_idx;

		size -= 3; prop += 3;
		base_idx = get_idx_from_shift(base_shift);
		if (base_idx < 0) {
			/* skip the pte encoding also */
			prop += lpnum * 2; size -= lpnum * 2;
			continue;
		}
		def = &mmu_psize_defs[base_idx];
		if (base_idx == MMU_PAGE_16M)
			cur_cpu_spec->mmu_features |= MMU_FTR_16M_PAGE;

		def->shift = base_shift;
		if (base_shift <= 23)
			def->avpnm = 0;
		else
			def->avpnm = (1 << (base_shift - 23)) - 1;
		def->sllp = slbenc;
		/*
		 * We don't know for sure what's up with tlbiel, so
		 * for now we only set it for 4K and 64K pages
		 */
		if (base_idx == MMU_PAGE_4K || base_idx == MMU_PAGE_64K)
			def->tlbiel = 1;
		else
			def->tlbiel = 0;

		while (size > 0 && lpnum) {
			unsigned int shift = be32_to_cpu(prop[0]);
			int penc  = be32_to_cpu(prop[1]);

			prop += 2; size -= 2;
			lpnum--;

			idx = get_idx_from_shift(shift);
			if (idx < 0)
				continue;

			if (penc == -1)
				pr_err("Invalid penc for base_shift=%d "
				       "shift=%d\n", base_shift, shift);

			def->penc[idx] = penc;
			pr_info("base_shift=%d: shift=%d, sllp=0x%04lx,"
				" avpnm=0x%08lx, tlbiel=%d, penc=%d\n",
				base_shift, shift, def->sllp,
				def->avpnm, def->tlbiel, def->penc[idx]);
		}
	}

	return 1;
}

#ifdef CONFIG_HUGETLB_PAGE
/* Scan for 16G memory blocks that have been set aside for huge pages
 * and reserve those blocks for 16G huge pages.
 */
static int __init htab_dt_scan_hugepage_blocks(unsigned long node,
					const char *uname, int depth,
					void *data) {
	const char *type = of_get_flat_dt_prop(node, "device_type", NULL);
	const __be64 *addr_prop;
	const __be32 *page_count_prop;
	unsigned int expected_pages;
	long unsigned int phys_addr;
	long unsigned int block_size;

	/* We are scanning "memory" nodes only */
	if (type == NULL || strcmp(type, "memory") != 0)
		return 0;

	/* This property is the log base 2 of the number of virtual pages that
	 * will represent this memory block. */
	page_count_prop = of_get_flat_dt_prop(node, "ibm,expected#pages", NULL);
	if (page_count_prop == NULL)
		return 0;
	expected_pages = (1 << be32_to_cpu(page_count_prop[0]));
	addr_prop = of_get_flat_dt_prop(node, "reg", NULL);
	if (addr_prop == NULL)
		return 0;
	phys_addr = be64_to_cpu(addr_prop[0]);
	block_size = be64_to_cpu(addr_prop[1]);
	if (block_size != (16 * GB))
		return 0;
	printk(KERN_INFO "Huge page(16GB) memory: "
			"addr = 0x%lX size = 0x%lX pages = %d\n",
			phys_addr, block_size, expected_pages);
	if (phys_addr + (16 * GB) <= memblock_end_of_DRAM()) {
		memblock_reserve(phys_addr, block_size * expected_pages);
		add_gpage(phys_addr, block_size, expected_pages);
	}
	return 0;
}
#endif /* CONFIG_HUGETLB_PAGE */

static void mmu_psize_set_default_penc(void)
{
	int bpsize, apsize;
	for (bpsize = 0; bpsize < MMU_PAGE_COUNT; bpsize++)
		for (apsize = 0; apsize < MMU_PAGE_COUNT; apsize++)
			mmu_psize_defs[bpsize].penc[apsize] = -1;
}

#ifdef CONFIG_PPC_64K_PAGES

static bool might_have_hea(void)
{
	/*
	 * The HEA ethernet adapter requires awareness of the
	 * GX bus. Without that awareness we can easily assume
	 * we will never see an HEA ethernet device.
	 */
#ifdef CONFIG_IBMEBUS
	return !cpu_has_feature(CPU_FTR_ARCH_207S);
#else
	return false;
#endif
}

#endif /* #ifdef CONFIG_PPC_64K_PAGES */

static void __init htab_init_page_sizes(void)
{
	int rc;

	/* se the invalid penc to -1 */
	mmu_psize_set_default_penc();

	/* Default to 4K pages only */
	memcpy(mmu_psize_defs, mmu_psize_defaults_old,
	       sizeof(mmu_psize_defaults_old));

	/*
	 * Try to find the available page sizes in the device-tree
	 */
	rc = of_scan_flat_dt(htab_dt_scan_page_sizes, NULL);
	if (rc != 0)  /* Found */
		goto found;

	/*
	 * Not in the device-tree, let's fallback on known size
	 * list for 16M capable GP & GR
	 */
	if (mmu_has_feature(MMU_FTR_16M_PAGE))
		memcpy(mmu_psize_defs, mmu_psize_defaults_gp,
		       sizeof(mmu_psize_defaults_gp));
found:
	if (!debug_pagealloc_enabled()) {
		/*
		 * Pick a size for the linear mapping. Currently, we only
		 * support 16M, 1M and 4K which is the default
		 */
		if (mmu_psize_defs[MMU_PAGE_16M].shift)
			mmu_linear_psize = MMU_PAGE_16M;
		else if (mmu_psize_defs[MMU_PAGE_1M].shift)
			mmu_linear_psize = MMU_PAGE_1M;
	}

#ifdef CONFIG_PPC_64K_PAGES
	/*
	 * Pick a size for the ordinary pages. Default is 4K, we support
	 * 64K for user mappings and vmalloc if supported by the processor.
	 * We only use 64k for ioremap if the processor
	 * (and firmware) support cache-inhibited large pages.
	 * If not, we use 4k and set mmu_ci_restrictions so that
	 * hash_page knows to switch processes that use cache-inhibited
	 * mappings to 4k pages.
	 */
	if (mmu_psize_defs[MMU_PAGE_64K].shift) {
		mmu_virtual_psize = MMU_PAGE_64K;
		mmu_vmalloc_psize = MMU_PAGE_64K;
		if (mmu_linear_psize == MMU_PAGE_4K)
			mmu_linear_psize = MMU_PAGE_64K;
		if (mmu_has_feature(MMU_FTR_CI_LARGE_PAGE)) {
			/*
			 * When running on pSeries using 64k pages for ioremap
			 * would stop us accessing the HEA ethernet. So if we
			 * have the chance of ever seeing one, stay at 4k.
			 */
			if (!might_have_hea() || !machine_is(pseries))
				mmu_io_psize = MMU_PAGE_64K;
		} else
			mmu_ci_restrictions = 1;
	}
#endif /* CONFIG_PPC_64K_PAGES */

#ifdef CONFIG_SPARSEMEM_VMEMMAP
	/* We try to use 16M pages for vmemmap if that is supported
	 * and we have at least 1G of RAM at boot
	 */
	if (mmu_psize_defs[MMU_PAGE_16M].shift &&
	    memblock_phys_mem_size() >= 0x40000000)
		mmu_vmemmap_psize = MMU_PAGE_16M;
	else if (mmu_psize_defs[MMU_PAGE_64K].shift)
		mmu_vmemmap_psize = MMU_PAGE_64K;
	else
		mmu_vmemmap_psize = MMU_PAGE_4K;
#endif /* CONFIG_SPARSEMEM_VMEMMAP */

	printk(KERN_DEBUG "Page orders: linear mapping = %d, "
	       "virtual = %d, io = %d"
#ifdef CONFIG_SPARSEMEM_VMEMMAP
	       ", vmemmap = %d"
#endif
	       "\n",
	       mmu_psize_defs[mmu_linear_psize].shift,
	       mmu_psize_defs[mmu_virtual_psize].shift,
	       mmu_psize_defs[mmu_io_psize].shift
#ifdef CONFIG_SPARSEMEM_VMEMMAP
	       ,mmu_psize_defs[mmu_vmemmap_psize].shift
#endif
	       );

#ifdef CONFIG_HUGETLB_PAGE
	/* Reserve 16G huge page memory sections for huge pages */
	of_scan_flat_dt(htab_dt_scan_hugepage_blocks, NULL);
#endif /* CONFIG_HUGETLB_PAGE */
}

static int __init htab_dt_scan_pftsize(unsigned long node,
				       const char *uname, int depth,
				       void *data)
{
	const char *type = of_get_flat_dt_prop(node, "device_type", NULL);
	const __be32 *prop;

	/* We are scanning "cpu" nodes only */
	if (type == NULL || strcmp(type, "cpu") != 0)
		return 0;

	prop = of_get_flat_dt_prop(node, "ibm,pft-size", NULL);
	if (prop != NULL) {
		/* pft_size[0] is the NUMA CEC cookie */
		ppc64_pft_size = be32_to_cpu(prop[1]);
		return 1;
	}
	return 0;
}

unsigned htab_shift_for_mem_size(unsigned long mem_size)
{
	unsigned memshift = __ilog2(mem_size);
	unsigned pshift = mmu_psize_defs[mmu_virtual_psize].shift;
	unsigned pteg_shift;

	/* round mem_size up to next power of 2 */
	if ((1UL << memshift) < mem_size)
		memshift += 1;

	/* aim for 2 pages / pteg */
	pteg_shift = memshift - (pshift + 1);

	/*
	 * 2^11 PTEGS of 128 bytes each, ie. 2^18 bytes is the minimum htab
	 * size permitted by the architecture.
	 */
	return max(pteg_shift + 7, 18U);
}

static unsigned long __init htab_get_table_size(void)
{
	/* If hash size isn't already provided by the platform, we try to
	 * retrieve it from the device-tree. If it's not there neither, we
	 * calculate it now based on the total RAM size
	 */
	if (ppc64_pft_size == 0)
		of_scan_flat_dt(htab_dt_scan_pftsize, NULL);
	if (ppc64_pft_size)
		return 1UL << ppc64_pft_size;

	return 1UL << htab_shift_for_mem_size(memblock_phys_mem_size());
}

#ifdef CONFIG_MEMORY_HOTPLUG
int create_section_mapping(unsigned long start, unsigned long end)
{
	int rc = htab_bolt_mapping(start, end, __pa(start),
				   pgprot_val(PAGE_KERNEL), mmu_linear_psize,
				   mmu_kernel_ssize);

	if (rc < 0) {
		int rc2 = htab_remove_mapping(start, end, mmu_linear_psize,
					      mmu_kernel_ssize);
		BUG_ON(rc2 && (rc2 != -ENOENT));
	}
	return rc;
}

int remove_section_mapping(unsigned long start, unsigned long end)
{
	int rc = htab_remove_mapping(start, end, mmu_linear_psize,
				     mmu_kernel_ssize);
	WARN_ON(rc < 0);
	return rc;
}
#endif /* CONFIG_MEMORY_HOTPLUG */

static void __init hash_init_partition_table(phys_addr_t hash_table,
					     unsigned long pteg_count)
{
	unsigned long ps_field;
	unsigned long htab_size;
	unsigned long patb_size = 1UL << PATB_SIZE_SHIFT;

	/*
	 * slb llp encoding for the page size used in VPM real mode.
	 * We can ignore that for lpid 0
	 */
	ps_field = 0;
	htab_size =  __ilog2(pteg_count) - 11;

	BUILD_BUG_ON_MSG((PATB_SIZE_SHIFT > 24), "Partition table size too large.");
	partition_tb = __va(memblock_alloc_base(patb_size, patb_size,
						MEMBLOCK_ALLOC_ANYWHERE));

	/* Initialize the Partition Table with no entries */
	memset((void *)partition_tb, 0, patb_size);
	partition_tb->patb0 = cpu_to_be64(ps_field | hash_table | htab_size);
	/*
	 * FIXME!! This should be done via update_partition table
	 * For now UPRT is 0 for us.
	 */
	partition_tb->patb1 = 0;
	DBG("Partition table %p\n", partition_tb);
	/*
	 * update partition table control register,
	 * 64 K size.
	 */
	mtspr(SPRN_PTCR, __pa(partition_tb) | (PATB_SIZE_SHIFT - 12));

}

static void __init htab_initialize(void)
{
	unsigned long table;
	unsigned long pteg_count;
	unsigned long prot;
	unsigned long base = 0, size = 0, limit;
	struct memblock_region *reg;

	DBG(" -> htab_initialize()\n");

	/* Initialize segment sizes */
	htab_init_seg_sizes();

	/* Initialize page sizes */
	htab_init_page_sizes();

	if (mmu_has_feature(MMU_FTR_1T_SEGMENT)) {
		mmu_kernel_ssize = MMU_SEGSIZE_1T;
		mmu_highuser_ssize = MMU_SEGSIZE_1T;
		printk(KERN_INFO "Using 1TB segments\n");
	}

	/*
	 * Calculate the required size of the htab.  We want the number of
	 * PTEGs to equal one half the number of real pages.
	 */ 
	htab_size_bytes = htab_get_table_size();
	pteg_count = htab_size_bytes >> 7;

	htab_hash_mask = pteg_count - 1;

	if (firmware_has_feature(FW_FEATURE_LPAR)) {
		/* Using a hypervisor which owns the htab */
		htab_address = NULL;
		_SDR1 = 0; 
#ifdef CONFIG_FA_DUMP
		/*
		 * If firmware assisted dump is active firmware preserves
		 * the contents of htab along with entire partition memory.
		 * Clear the htab if firmware assisted dump is active so
		 * that we dont end up using old mappings.
		 */
		if (is_fadump_active() && ppc_md.hpte_clear_all)
			ppc_md.hpte_clear_all();
#endif
	} else {
		/* Find storage for the HPT.  Must be contiguous in
		 * the absolute address space. On cell we want it to be
		 * in the first 2 Gig so we can use it for IOMMU hacks.
		 */
		if (machine_is(cell))
			limit = 0x80000000;
		else
			limit = MEMBLOCK_ALLOC_ANYWHERE;

		table = memblock_alloc_base(htab_size_bytes, htab_size_bytes, limit);

		DBG("Hash table allocated at %lx, size: %lx\n", table,
		    htab_size_bytes);

		htab_address = __va(table);

		/* htab absolute addr + encoded htabsize */
		_SDR1 = table + __ilog2(pteg_count) - 11;

		/* Initialize the HPT with no entries */
		memset((void *)table, 0, htab_size_bytes);

		if (!cpu_has_feature(CPU_FTR_ARCH_300))
			/* Set SDR1 */
			mtspr(SPRN_SDR1, _SDR1);
		else
			hash_init_partition_table(table, pteg_count);
	}

	prot = pgprot_val(PAGE_KERNEL);

#ifdef CONFIG_DEBUG_PAGEALLOC
	if (debug_pagealloc_enabled()) {
		linear_map_hash_count = memblock_end_of_DRAM() >> PAGE_SHIFT;
		linear_map_hash_slots = __va(memblock_alloc_base(
				linear_map_hash_count, 1, ppc64_rma_size));
		memset(linear_map_hash_slots, 0, linear_map_hash_count);
	}
#endif /* CONFIG_DEBUG_PAGEALLOC */

	/* On U3 based machines, we need to reserve the DART area and
	 * _NOT_ map it to avoid cache paradoxes as it's remapped non
	 * cacheable later on
	 */

	/* create bolted the linear mapping in the hash table */
	for_each_memblock(memory, reg) {
		base = (unsigned long)__va(reg->base);
		size = reg->size;

		DBG("creating mapping for region: %lx..%lx (prot: %lx)\n",
		    base, size, prot);

#ifdef CONFIG_U3_DART
		/* Do not map the DART space. Fortunately, it will be aligned
		 * in such a way that it will not cross two memblock regions and
		 * will fit within a single 16Mb page.
		 * The DART space is assumed to be a full 16Mb region even if
		 * we only use 2Mb of that space. We will use more of it later
		 * for AGP GART. We have to use a full 16Mb large page.
		 */
		DBG("DART base: %lx\n", dart_tablebase);

		if (dart_tablebase != 0 && dart_tablebase >= base
		    && dart_tablebase < (base + size)) {
			unsigned long dart_table_end = dart_tablebase + 16 * MB;
			if (base != dart_tablebase)
				BUG_ON(htab_bolt_mapping(base, dart_tablebase,
							__pa(base), prot,
							mmu_linear_psize,
							mmu_kernel_ssize));
			if ((base + size) > dart_table_end)
				BUG_ON(htab_bolt_mapping(dart_tablebase+16*MB,
							base + size,
							__pa(dart_table_end),
							 prot,
							 mmu_linear_psize,
							 mmu_kernel_ssize));
			continue;
		}
#endif /* CONFIG_U3_DART */
		BUG_ON(htab_bolt_mapping(base, base + size, __pa(base),
				prot, mmu_linear_psize, mmu_kernel_ssize));
	}
	memblock_set_current_limit(MEMBLOCK_ALLOC_ANYWHERE);

	/*
	 * If we have a memory_limit and we've allocated TCEs then we need to
	 * explicitly map the TCE area at the top of RAM. We also cope with the
	 * case that the TCEs start below memory_limit.
	 * tce_alloc_start/end are 16MB aligned so the mapping should work
	 * for either 4K or 16MB pages.
	 */
	if (tce_alloc_start) {
		tce_alloc_start = (unsigned long)__va(tce_alloc_start);
		tce_alloc_end = (unsigned long)__va(tce_alloc_end);

		if (base + size >= tce_alloc_start)
			tce_alloc_start = base + size + 1;

		BUG_ON(htab_bolt_mapping(tce_alloc_start, tce_alloc_end,
					 __pa(tce_alloc_start), prot,
					 mmu_linear_psize, mmu_kernel_ssize));
	}


	DBG(" <- htab_initialize()\n");
}
#undef KB
#undef MB

void __init hash__early_init_mmu(void)
{
	/*
	 * initialize page table size
	 */
	__pte_frag_nr = H_PTE_FRAG_NR;
	__pte_frag_size_shift = H_PTE_FRAG_SIZE_SHIFT;

	__pte_index_size = H_PTE_INDEX_SIZE;
	__pmd_index_size = H_PMD_INDEX_SIZE;
	__pud_index_size = H_PUD_INDEX_SIZE;
	__pgd_index_size = H_PGD_INDEX_SIZE;
	__pmd_cache_index = H_PMD_CACHE_INDEX;
	__pte_table_size = H_PTE_TABLE_SIZE;
	__pmd_table_size = H_PMD_TABLE_SIZE;
	__pud_table_size = H_PUD_TABLE_SIZE;
	__pgd_table_size = H_PGD_TABLE_SIZE;
	/*
	 * 4k use hugepd format, so for hash set then to
	 * zero
	 */
	__pmd_val_bits = 0;
	__pud_val_bits = 0;
	__pgd_val_bits = 0;

	__kernel_virt_start = H_KERN_VIRT_START;
	__kernel_virt_size = H_KERN_VIRT_SIZE;
	__vmalloc_start = H_VMALLOC_START;
	__vmalloc_end = H_VMALLOC_END;
	vmemmap = (struct page *)H_VMEMMAP_BASE;
	ioremap_bot = IOREMAP_BASE;

	/* Initialize the MMU Hash table and create the linear mapping
	 * of memory. Has to be done before SLB initialization as this is
	 * currently where the page size encoding is obtained.
	 */
	htab_initialize();

	/* Initialize SLB management */
	slb_initialize();
}

#ifdef CONFIG_SMP
void hash__early_init_mmu_secondary(void)
{
	/* Initialize hash table for that CPU */
	if (!firmware_has_feature(FW_FEATURE_LPAR)) {
		if (!cpu_has_feature(CPU_FTR_ARCH_300))
			mtspr(SPRN_SDR1, _SDR1);
		else
			mtspr(SPRN_PTCR,
			      __pa(partition_tb) | (PATB_SIZE_SHIFT - 12));
	}
	/* Initialize SLB */
	slb_initialize();
}
#endif /* CONFIG_SMP */

/*
 * Called by asm hashtable.S for doing lazy icache flush
 */
unsigned int hash_page_do_lazy_icache(unsigned int pp, pte_t pte, int trap)
{
	struct page *page;

	if (!pfn_valid(pte_pfn(pte)))
		return pp;

	page = pte_page(pte);

	/* page is dirty */
	if (!test_bit(PG_arch_1, &page->flags) && !PageReserved(page)) {
		if (trap == 0x400) {
			flush_dcache_icache_page(page);
			set_bit(PG_arch_1, &page->flags);
		} else
			pp |= HPTE_R_N;
	}
	return pp;
}

#ifdef CONFIG_PPC_MM_SLICES
static unsigned int get_paca_psize(unsigned long addr)
{
	u64 lpsizes;
	unsigned char *hpsizes;
	unsigned long index, mask_index;

	if (addr < SLICE_LOW_TOP) {
		lpsizes = get_paca()->mm_ctx_low_slices_psize;
		index = GET_LOW_SLICE_INDEX(addr);
		return (lpsizes >> (index * 4)) & 0xF;
	}
	hpsizes = get_paca()->mm_ctx_high_slices_psize;
	index = GET_HIGH_SLICE_INDEX(addr);
	mask_index = index & 0x1;
	return (hpsizes[index >> 1] >> (mask_index * 4)) & 0xF;
}

#else
unsigned int get_paca_psize(unsigned long addr)
{
	return get_paca()->mm_ctx_user_psize;
}
#endif

/*
 * Demote a segment to using 4k pages.