init.c

linux-2.6.15.6

/*
 * Initialize MMU support.
 *
 * Copyright (C) 1998-2003 Hewlett-Packard Co
 *	David Mosberger-Tang <[e-mail]>
 */
#include <linux/config.h>
#include <linux/kernel.h>
#include <linux/init.h>

#include <linux/bootmem.h>
#include <linux/efi.h>
#include <linux/elf.h>
#include <linux/mm.h>
#include <linux/mmzone.h>
#include <linux/module.h>
#include <linux/personality.h>
#include <linux/reboot.h>
#include <linux/slab.h>
#include <linux/swap.h>
#include <linux/proc_fs.h>
#include <linux/bitops.h>

#include <asm/a.out.h>
#include <asm/dma.h>
#include <asm/ia32.h>
#include <asm/io.h>
#include <asm/machvec.h>
#include <asm/numa.h>
#include <asm/patch.h>
#include <asm/pgalloc.h>
#include <asm/sal.h>
#include <asm/sections.h>
#include <asm/system.h>
#include <asm/tlb.h>
#include <asm/uaccess.h>
#include <asm/unistd.h>
#include <asm/mca.h>

DEFINE_PER_CPU(struct mmu_gather, mmu_gathers);

DEFINE_PER_CPU(unsigned long *, __pgtable_quicklist);
DEFINE_PER_CPU(long, __pgtable_quicklist_size);

extern void ia64_tlb_init (void);

unsigned long MAX_DMA_ADDRESS = PAGE_OFFSET + 0x100000000UL;

#ifdef CONFIG_VIRTUAL_MEM_MAP
unsigned long vmalloc_end = VMALLOC_END_INIT;
EXPORT_SYMBOL(vmalloc_end);
struct page *vmem_map;
EXPORT_SYMBOL(vmem_map);
#endif

struct page *zero_page_memmap_ptr;	/* map entry for zero page */
EXPORT_SYMBOL(zero_page_memmap_ptr);

#define MIN_PGT_PAGES			25UL
#define MAX_PGT_FREES_PER_PASS		16L
#define PGT_FRACTION_OF_NODE_MEM	16

static inline long
max_pgt_pages(void)
{
	u64 node_free_pages, max_pgt_pages;

#ifndef	CONFIG_NUMA
	node_free_pages = nr_free_pages();
#else
	node_free_pages = nr_free_pages_pgdat(NODE_DATA(numa_node_id()));
#endif
	max_pgt_pages = node_free_pages / PGT_FRACTION_OF_NODE_MEM;
	max_pgt_pages = max(max_pgt_pages, MIN_PGT_PAGES);
	return max_pgt_pages;
}

static inline long
min_pages_to_free(void)
{
	long pages_to_free;

	pages_to_free = pgtable_quicklist_size - max_pgt_pages();
	pages_to_free = min(pages_to_free, MAX_PGT_FREES_PER_PASS);
	return pages_to_free;
}

void
check_pgt_cache(void)
{
	long pages_to_free;

	if (unlikely(pgtable_quicklist_size <= MIN_PGT_PAGES))
		return;

	preempt_disable();
	while (unlikely((pages_to_free = min_pages_to_free()) > 0)) {
		while (pages_to_free--) {
			free_page((unsigned long)pgtable_quicklist_alloc());
		}
		preempt_enable();
		preempt_disable();
	}
	preempt_enable();
}

void
lazy_mmu_prot_update (pte_t pte)
{
	unsigned long addr;
	struct page *page;

	if (!pte_exec(pte))
		return;				/* not an executable page... */

	page = pte_page(pte);
	addr = (unsigned long) page_address(page);

	if (test_bit(PG_arch_1, &page->flags))
		return;				/* i-cache is already coherent with d-cache */

	flush_icache_range(addr, addr + PAGE_SIZE);
	set_bit(PG_arch_1, &page->flags);	/* mark page as clean */
}

inline void
ia64_set_rbs_bot (void)
{
	unsigned long stack_size = current->signal->rlim[RLIMIT_STACK].rlim_max & -16;

	if (stack_size > MAX_USER_STACK_SIZE)
		stack_size = MAX_USER_STACK_SIZE;
	current->thread.rbs_bot = STACK_TOP - stack_size;
}

/*
 * This performs some platform-dependent address space initialization.
 * On IA-64, we want to setup the VM area for the register backing
 * store (which grows upwards) and install the gateway page which is
 * used for signal trampolines, etc.
 */
void
ia64_init_addr_space (void)
{
	struct vm_area_struct *vma;

	ia64_set_rbs_bot();

	/*
	 * If we're out of memory and kmem_cache_alloc() returns NULL, we simply ignore
	 * the problem.  When the process attempts to write to the register backing store
	 * for the first time, it will get a SEGFAULT in this case.
	 */
	vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL);
	if (vma) {
		memset(vma, 0, sizeof(*vma));
		vma->vm_mm = current->mm;
		vma->vm_start = current->thread.rbs_bot & PAGE_MASK;
		vma->vm_end = vma->vm_start + PAGE_SIZE;
		vma->vm_page_prot = protection_map[VM_DATA_DEFAULT_FLAGS & 0x7];
		vma->vm_flags = VM_DATA_DEFAULT_FLAGS|VM_GROWSUP|VM_ACCOUNT;
		down_write(&current->mm->mmap_sem);
		if (insert_vm_struct(current->mm, vma)) {
			up_write(&current->mm->mmap_sem);
			kmem_cache_free(vm_area_cachep, vma);
			return;
		}
		up_write(&current->mm->mmap_sem);
	}

	/* map NaT-page at address zero to speed up speculative dereferencing of NULL: */
	if (!(current->personality & MMAP_PAGE_ZERO)) {
		vma = kmem_cache_alloc(vm_area_cachep, SLAB_KERNEL);
		if (vma) {
			memset(vma, 0, sizeof(*vma));
			vma->vm_mm = current->mm;
			vma->vm_end = PAGE_SIZE;
			vma->vm_page_prot = __pgprot(pgprot_val(PAGE_READONLY) | _PAGE_MA_NAT);
			vma->vm_flags = VM_READ | VM_MAYREAD | VM_IO | VM_RESERVED;
			down_write(&current->mm->mmap_sem);
			if (insert_vm_struct(current->mm, vma)) {
				up_write(&current->mm->mmap_sem);
				kmem_cache_free(vm_area_cachep, vma);
				return;
			}
			up_write(&current->mm->mmap_sem);
		}
	}
}

void
free_initmem (void)
{
	unsigned long addr, eaddr;

	addr = (unsigned long) ia64_imva(__init_begin);
	eaddr = (unsigned long) ia64_imva(__init_end);
	while (addr < eaddr) {
		ClearPageReserved(virt_to_page(addr));
		set_page_count(virt_to_page(addr), 1);
		free_page(addr);
		++totalram_pages;
		addr += PAGE_SIZE;
	}
	printk(KERN_INFO "Freeing unused kernel memory: %ldkB freed\n",
	       (__init_end - __init_begin) >> 10);
}

void
free_initrd_mem (unsigned long start, unsigned long end)
{
	struct page *page;
	/*
	 * EFI uses 4KB pages while the kernel can use 4KB or bigger.
	 * Thus EFI and the kernel may have different page sizes. It is
	 * therefore possible to have the initrd share the same page as
	 * the end of the kernel (given current setup).
	 *
	 * To avoid freeing/using the wrong page (kernel sized) we:
	 *	- align up the beginning of initrd
	 *	- align down the end of initrd
	 *
	 *  |             |
	 *  |=============| a000
	 *  |             |
	 *  |             |
	 *  |             | 9000
	 *  |/////////////|
	 *  |/////////////|
	 *  |=============| 8000
	 *  |///INITRD////|
	 *  |/////////////|
	 *  |/////////////| 7000
	 *  |             |
	 *  |KKKKKKKKKKKKK|
	 *  |=============| 6000
	 *  |KKKKKKKKKKKKK|
	 *  |KKKKKKKKKKKKK|
	 *  K=kernel using 8KB pages
	 *
	 * In this example, we must free page 8000 ONLY. So we must align up
	 * initrd_start and keep initrd_end as is.
	 */
	start = PAGE_ALIGN(start);
	end = end & PAGE_MASK;

	if (start < end)
		printk(KERN_INFO "Freeing initrd memory: %ldkB freed\n", (end - start) >> 10);

	for (; start < end; start += PAGE_SIZE) {
		if (!virt_addr_valid(start))
			continue;
		page = virt_to_page(start);
		ClearPageReserved(page);
		set_page_count(page, 1);
		free_page(start);
		++totalram_pages;
	}
}

/*
 * This installs a clean page in the kernel's page table.
 */
struct page *
put_kernel_page (struct page *page, unsigned long address, pgprot_t pgprot)
{
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;

	if (!PageReserved(page))
		printk(KERN_ERR "put_kernel_page: page at 0x%p not in reserved memory\n",
		       page_address(page));

	pgd = pgd_offset_k(address);		/* note: this is NOT pgd_offset()! */

	{
		pud = pud_alloc(&init_mm, pgd, address);
		if (!pud)
			goto out;
		pmd = pmd_alloc(&init_mm, pud, address);
		if (!pmd)
			goto out;
		pte = pte_alloc_kernel(pmd, address);
		if (!pte)
			goto out;
		if (!pte_none(*pte))
			goto out;
		set_pte(pte, mk_pte(page, pgprot));
	}
  out:
	/* no need for flush_tlb */
	return page;
}

static void
setup_gate (void)
{
	struct page *page;

	/*
	 * Map the gate page twice: once read-only to export the ELF
	 * headers etc. and once execute-only page to enable
	 * privilege-promotion via "epc":
	 */
	page = virt_to_page(ia64_imva(__start_gate_section));
	put_kernel_page(page, GATE_ADDR, PAGE_READONLY);
#ifdef HAVE_BUGGY_SEGREL
	page = virt_to_page(ia64_imva(__start_gate_section + PAGE_SIZE));
	put_kernel_page(page, GATE_ADDR + PAGE_SIZE, PAGE_GATE);
#else
	put_kernel_page(page, GATE_ADDR + PERCPU_PAGE_SIZE, PAGE_GATE);
	/* Fill in the holes (if any) with read-only zero pages: */
	{
		unsigned long addr;

		for (addr = GATE_ADDR + PAGE_SIZE;
		     addr < GATE_ADDR + PERCPU_PAGE_SIZE;