C++ pte_page函数代码示例

/*
 * This routine puts a long into any process space by following the page
 * tables. NOTE! You should check that the long isn't on a page boundary,
 * and that it is in the task area before calling this: this routine does
 * no checking.
 *
 * Now keeps R/W state of page so that a text page stays readonly
 * even if a debugger scribbles breakpoints into it.  -M.U-
 */
static void put_long(struct task_struct * tsk, struct vm_area_struct * vma, unsigned long addr,
	unsigned long data)
{
	pgd_t *pgdir;
	pmd_t *pgmiddle;
	pte_t *pgtable;
	unsigned long page;

repeat:
	pgdir = pgd_offset(vma->vm_mm, addr);
	if (!pgd_present(*pgdir)) {
		do_no_page(tsk, vma, addr, 1);
		goto repeat;
	}
	if (pgd_bad(*pgdir)) {
		printk("ptrace: bad page directory %08lx\n", pgd_val(*pgdir));
		pgd_clear(pgdir);
		return;
	}
	pgmiddle = pmd_offset(pgdir, addr);
	if (pmd_none(*pgmiddle)) {
		do_no_page(tsk, vma, addr, 1);
		goto repeat;
	}
	if (pmd_bad(*pgmiddle)) {
		printk("ptrace: bad page middle %08lx\n", pmd_val(*pgmiddle));
		pmd_clear(pgmiddle);
		return;
	}
	pgtable = pte_offset(pgmiddle, addr);
	if (!pte_present(*pgtable)) {
		do_no_page(tsk, vma, addr, 1);
		goto repeat;
	}
	page = pte_page(*pgtable);
	if (!pte_write(*pgtable)) {
		do_wp_page(tsk, vma, addr, 1);
		goto repeat;
	}
/* this is a hack for non-kernel-mapped video buffers and similar */
	if (page < high_memory)
		*(unsigned long *) (page + (addr & ~PAGE_MASK)) = data;
/* we're bypassing pagetables, so we have to set the dirty bit ourselves */
/* this should also re-instate whatever read-only mode there was before */
	set_pte(pgtable, pte_mkdirty(mk_pte(page, vma->vm_page_prot)));
	flush_tlb();
}

/**
 * flush_icache_page_range - Flush dcache and invalidate icache for part of a
 *				single page
 * @start: The starting virtual address of the page part.
 * @end: The ending virtual address of the page part.
 *
 * Invalidate the icache for part of a single page, as determined by the
 * virtual addresses given.  The page must be in the paged area.  The dcache is
 * not flushed as the cache must be in write-through mode to get here.
 */
static void flush_icache_page_range(unsigned long start, unsigned long end)
{
	unsigned long addr, size, off;
	struct page *page;
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *ppte, pte;

	/* work out how much of the page to flush */
	off = start & ~PAGE_MASK;
	size = end - start;

	/* get the physical address the page is mapped to from the page
	 * tables */
	pgd = pgd_offset(current->mm, start);
	if (!pgd || !pgd_val(*pgd))
		return;

	pud = pud_offset(pgd, start);
	if (!pud || !pud_val(*pud))
		return;

	pmd = pmd_offset(pud, start);
	if (!pmd || !pmd_val(*pmd))
		return;

	ppte = pte_offset_map(pmd, start);
	if (!ppte)
		return;
	pte = *ppte;
	pte_unmap(ppte);

	if (pte_none(pte))
		return;

	page = pte_page(pte);
	if (!page)
		return;

	addr = page_to_phys(page);

	/* invalidate the icache coverage on that region */
	mn10300_local_icache_inv_range2(addr + off, size);
	smp_cache_call(SMP_ICACHE_INV_FLUSH_RANGE, start, end);
}

/*
 * The performance critical leaf functions are made noinline otherwise gcc
 * inlines everything into a single function which results in too much
 * register pressure.
 */
static noinline int gup_pte_range(pmd_t pmd, unsigned long addr,
		unsigned long end, int write, struct page **pages, int *nr)
{
	unsigned long mask, result;
	pte_t *ptep;

	if (tlb_type == hypervisor) {
		result = _PAGE_PRESENT_4V|_PAGE_P_4V;
		if (write)
			result |= _PAGE_WRITE_4V;
	} else {
		result = _PAGE_PRESENT_4U|_PAGE_P_4U;
		if (write)
			result |= _PAGE_WRITE_4U;
	}
	mask = result | _PAGE_SPECIAL;

	ptep = pte_offset_kernel(&pmd, addr);
	do {
		struct page *page, *head;
		pte_t pte = *ptep;

		if ((pte_val(pte) & mask) != result)
			return 0;
		VM_BUG_ON(!pfn_valid(pte_pfn(pte)));

		/* The hugepage case is simplified on sparc64 because
		 * we encode the sub-page pfn offsets into the
		 * hugepage PTEs.  We could optimize this in the future
		 * use page_cache_add_speculative() for the hugepage case.
		 */
		page = pte_page(pte);
		head = compound_head(page);
		if (!page_cache_get_speculative(head))
			return 0;
		if (unlikely(pte_val(pte) != pte_val(*ptep))) {
			put_page(head);
			return 0;
		}

		pages[*nr] = page;
		(*nr)++;
	} while (ptep++, addr += PAGE_SIZE, addr != end);

	return 1;
}

static inline void forget_pte(pte_t page)
{
	if (pte_none(page))
		return;
	if (pte_present(page)) {
		struct page *ptpage = pte_page(page);
		if ((!VALID_PAGE(ptpage)) || PageReserved(ptpage))
			return;
		/* 
		 * free_page() used to be able to clear swap cache
		 * entries.  We may now have to do it manually.  
		 */
		free_page_and_swap_cache(ptpage);
		return;
	}
	swap_free(pte_to_swp_entry(page));
}

static int get_gate_page(struct mm_struct *mm, unsigned long address,
		unsigned int gup_flags, struct vm_area_struct **vma,
		struct page **page)
{
	pgd_t *pgd;
	p4d_t *p4d;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *pte;
	int ret = -EFAULT;

	/* user gate pages are read-only */
	if (gup_flags & FOLL_WRITE)
		return -EFAULT;
	if (address > TASK_SIZE)
		pgd = pgd_offset_k(address);
	else
		pgd = pgd_offset_gate(mm, address);
	BUG_ON(pgd_none(*pgd));
	p4d = p4d_offset(pgd, address);
	BUG_ON(p4d_none(*p4d));
	pud = pud_offset(p4d, address);
	BUG_ON(pud_none(*pud));
	pmd = pmd_offset(pud, address);
	if (pmd_none(*pmd))
		return -EFAULT;
	VM_BUG_ON(pmd_trans_huge(*pmd));
	pte = pte_offset_map(pmd, address);
	if (pte_none(*pte))
		goto unmap;
	*vma = get_gate_vma(mm);
	if (!page)
		goto out;
	*page = vm_normal_page(*vma, address, *pte);
	if (!*page) {
		if ((gup_flags & FOLL_DUMP) || !is_zero_pfn(pte_pfn(*pte)))
			goto unmap;
		*page = pte_page(*pte);
	}
	get_page(*page);
out:
	ret = 0;
unmap:
	pte_unmap(pte);
	return ret;
}

/*
 * Return indicates whether a page was freed so caller can adjust rss
 */
static inline int free_pte(pte_t pte)
{
	if (pte_present(pte)) {
		struct page *page = pte_page(pte);
		if ((!VALID_PAGE(page)) || PageReserved(page))
			return 0;
		/* 
		 * free_page() used to be able to clear swap cache
		 * entries.  We may now have to do it manually.  
		 */
		if (pte_dirty(pte) && page->mapping)
			set_page_dirty(page);
		free_page_and_swap_cache(page);
		return 1;
	}
	swap_free(pte_to_swp_entry(pte));
	return 0;
}

/*
 * The performance critical leaf functions are made noinline otherwise gcc
 * inlines everything into a single function which results in too much
 * register pressure.
 */
static noinline int gup_pte_range(pmd_t pmd, unsigned long addr,
		unsigned long end, int write, struct page **pages, int *nr)
{
	u64 mask, result;
	pte_t *ptep;

#ifdef CONFIG_X2TLB
	result = _PAGE_PRESENT | _PAGE_EXT(_PAGE_EXT_KERN_READ | _PAGE_EXT_USER_READ);
	if (write)
		result |= _PAGE_EXT(_PAGE_EXT_KERN_WRITE | _PAGE_EXT_USER_WRITE);
#elif defined(CONFIG_SUPERH64)
	result = _PAGE_PRESENT | _PAGE_USER | _PAGE_READ;
	if (write)
		result |= _PAGE_WRITE;
#else
	result = _PAGE_PRESENT | _PAGE_USER;
	if (write)
		result |= _PAGE_RW;
#endif

	mask = result | _PAGE_SPECIAL;

	ptep = pte_offset_map(&pmd, addr);
	do {
		pte_t pte = gup_get_pte(ptep);
		struct page *page;

		if ((pte_val(pte) & mask) != result) {
			pte_unmap(ptep);
			return 0;
		}
		VM_BUG_ON(!pfn_valid(pte_pfn(pte)));
		page = pte_page(pte);
		get_page(page);
		__flush_anon_page(page, addr);
		flush_dcache_page(page);
		pages[*nr] = page;
		(*nr)++;

	} while (ptep++, addr += PAGE_SIZE, addr != end);
	pte_unmap(ptep - 1);

	return 1;
}

struct page *kmem_vm_nopage(struct vm_area_struct *vma, unsigned long address, int write)
{
	unsigned long offset = vma->vm_pgoff << PAGE_SHIFT;
	unsigned long kaddr;
	pgd_t *pgd;
	pmd_t *pmd;
	pte_t *ptep, pte;
	struct page *page = NULL;

	/* address is user VA; convert to kernel VA of desired page */
	kaddr = (address - vma->vm_start) + offset;
	kaddr = VMALLOC_VMADDR(kaddr);

	spin_lock(&init_mm.page_table_lock);

	/* Lookup page structure for kernel VA */
	pgd = pgd_offset(&init_mm, kaddr);
	if (pgd_none(*pgd) || pgd_bad(*pgd))
		goto out;
	pmd = pmd_offset(pgd, kaddr);
	if (pmd_none(*pmd) || pmd_bad(*pmd))
		goto out;
	ptep = pte_offset(pmd, kaddr);
	if (!ptep)
		goto out;
	pte = *ptep;
	if (!pte_present(pte))
		goto out;
	if (write && !pte_write(pte))
		goto out;
	page = pte_page(pte);
	if (!VALID_PAGE(page)) {
		page = NULL;
		goto out;
	}

	/* Increment reference count on page */
	get_page(page);

out:
	spin_unlock(&init_mm.page_table_lock);

	return page;
}

/**
 * get_struct_page - Gets a struct page for a particular address
 * @address - the address of the page we need
 *
 * Two versions of this function have to be provided for working
 * between the 2.4 and 2.5 kernels. Rather than littering the
 * function with #defines, there is just two separate copies.
 * Look at the one that is relevant to the kernel you're using
 */
struct page *get_struct_page(unsigned long addr)
{
	struct mm_struct *mm;
	pgd_t *pgd;
	pud_t *pud;
	pmd_t *pmd;
	pte_t *ptep, pte;
	unsigned long pfn;
	struct page *page=NULL;

	mm = current->mm;
	/* Is this possible? */
	if (!mm) return NULL;

	spin_lock(&mm->page_table_lock);

	pgd = pgd_offset(mm, addr);
	if (!pgd_none(*pgd) && !pgd_bad(*pgd)) {
		pud = pud_offset(pgd, addr);
		if (!pud_none(*pud) && !pud_bad(*pud)) {
			pmd = pmd_offset(pud, addr);
			if (!pmd_none(*pmd) && !pmd_bad(*pmd)) {
				/*
			 	* disable preemption because of potential kmap().
			 	* page_table_lock should already have disabled
			 	* preemtion.  But, be paranoid.
			 	*/
				preempt_disable();
				ptep = pte_offset_map(pmd, addr);
				pte = *ptep;
				pte_unmap(ptep);
				preempt_enable();
				if (pte_present(pte)) {
					pfn = pte_pfn(pte);
					if (pfn_valid(pfn))
						page = pte_page(pte);
				}
			}
		}
	}

	spin_unlock(&mm->page_table_lock);
	return page;
}

/*
 * 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_address(page));
			set_bit(PG_arch_1, &page->flags);
		} else
			pp |= HPTE_R_N;
	}
	return pp;
}

void update_mmu_cache(struct vm_area_struct * vma,
		      unsigned long address, pte_t pte)
{
	unsigned long flags;
	unsigned long pteval;
	unsigned long vpn;

	/* Ptrace may call this routine. */
	if (vma && current->active_mm != vma->vm_mm)
		return;

#if defined(CONFIG_SH7705_CACHE_32KB)
	{
		struct page *page = pte_page(pte);
		unsigned long pfn = pte_pfn(pte);

		if (pfn_valid(pfn) && !test_bit(PG_mapped, &page->flags)) {
			unsigned long phys = pte_val(pte) & PTE_PHYS_MASK;

			__flush_wback_region((void *)P1SEGADDR(phys),
					     PAGE_SIZE);
			__set_bit(PG_mapped, &page->flags);
		}
	}
#endif

	local_irq_save(flags);

	/* Set PTEH register */
	vpn = (address & MMU_VPN_MASK) | get_asid();
	ctrl_outl(vpn, MMU_PTEH);

	pteval = pte_val(pte);

	/* Set PTEL register */
	pteval &= _PAGE_FLAGS_HARDWARE_MASK; /* drop software flags */
	/* conveniently, we want all the software flags to be 0 anyway */
	ctrl_outl(pteval, MMU_PTEL);

	/* Load the TLB */
	asm volatile("ldtlb": /* no output */ : /* no input */ : "memory");
	local_irq_restore(flags);
}

static noinline int gup_pte_range(pmd_t pmd, unsigned long addr,
		unsigned long end, int write, struct page **pages, int *nr)
{
	unsigned long mask, result;
	pte_t *ptep;

	if (tlb_type == hypervisor) {
		result = _PAGE_PRESENT_4V|_PAGE_P_4V;
		if (write)
			result |= _PAGE_WRITE_4V;
	} else {
		result = _PAGE_PRESENT_4U|_PAGE_P_4U;
		if (write)
			result |= _PAGE_WRITE_4U;
	}
	mask = result | _PAGE_SPECIAL;

	ptep = pte_offset_kernel(&pmd, addr);
	do {
		struct page *page, *head;
		pte_t pte = *ptep;

		if ((pte_val(pte) & mask) != result)
			return 0;
		VM_BUG_ON(!pfn_valid(pte_pfn(pte)));

		page = pte_page(pte);
		head = compound_head(page);
		if (!page_cache_get_speculative(head))
			return 0;
		if (unlikely(pte_val(pte) != pte_val(*ptep))) {
			put_page(head);
			return 0;
		}
		if (head != page)
			get_huge_page_tail(page);

		pages[*nr] = page;
		(*nr)++;
	} while (ptep++, addr += PAGE_SIZE, addr != end);

	return 1;
}

/* Ensure all existing pages follow the policy. */
static int
verify_pages(struct mm_struct *mm,
	     unsigned long addr, unsigned long end, unsigned long *nodes)
{
	while (addr < end) {
		struct page *p;
		pte_t *pte;
		pmd_t *pmd;
		pud_t *pud;
		pgd_t *pgd;
		pgd = pgd_offset(mm, addr);
		if (pgd_none(*pgd)) {
			unsigned long next = (addr + PGDIR_SIZE) & PGDIR_MASK;
			if (next > addr)
				break;
			addr = next;
			continue;
		}
		pud = pud_offset(pgd, addr);
		if (pud_none(*pud)) {
			addr = (addr + PUD_SIZE) & PUD_MASK;
			continue;
		}
		pmd = pmd_offset(pud, addr);
		if (pmd_none(*pmd)) {
			addr = (addr + PMD_SIZE) & PMD_MASK;
			continue;
		}
		p = NULL;
		pte = pte_offset_map(pmd, addr);
		if (pte_present(*pte))
			p = pte_page(*pte);
		pte_unmap(pte);
		if (p) {
			unsigned nid = page_to_nid(p);
			if (!test_bit(nid, nodes))
				return -EIO;
		}
		addr += PAGE_SIZE;
	}
	return 0;
}

/*
 * This routine gets a long from any process space by following the page
 * tables. NOTE! You should check that the long isn't on a page boundary,
 * and that it is in the task area before calling this: this routine does
 * no checking.
 *
 */
static unsigned long get_long(struct task_struct * tsk, 
	struct vm_area_struct * vma, unsigned long addr)
{
	pgd_t * pgdir;
	pmd_t * pgmiddle;
	pte_t * pgtable;
	unsigned long page;

repeat:
	pgdir = pgd_offset(vma->vm_mm, addr);
	if (pgd_none(*pgdir)) {
		do_no_page(tsk, vma, addr, 0);
		goto repeat;
	}
	if (pgd_bad(*pgdir)) {
		printk("ptrace: bad page directory %08lx\n", pgd_val(*pgdir));
		pgd_clear(pgdir);
		return 0;
	}
	pgmiddle = pmd_offset(pgdir,addr);
	if (pmd_none(*pgmiddle)) {
		do_no_page(tsk, vma, addr, 0);
		goto repeat;
	}
	if (pmd_bad(*pgmiddle)) {
		printk("ptrace: bad page directory %08lx\n",
		       pmd_val(*pgmiddle));
		pmd_clear(pgmiddle);
		return 0;
	}
	pgtable = pte_offset(pgmiddle, addr);
	if (!pte_present(*pgtable)) {
		do_no_page(tsk, vma, addr, 0);
		goto repeat;
	}
	page = pte_page(*pgtable);
/* this is a hack for non-kernel-mapped video buffers and similar */
	if (page >= high_memory)
		return 0;
	page += addr & ~PAGE_MASK;
	return *(unsigned long *) page;
}

static inline unsigned long uvirt_to_kva(pgd_t *pgd, unsigned long adr)
{
  unsigned long ret = 0UL;
  pmd_t *pmd;
  pte_t *ptep, pte;
  
  if (!pgd_none(*pgd)) {
    pmd = pmd_offset(pgd, adr);
    if (!pmd_none(*pmd)) {
      ptep = pte_offset_kernel(pmd, adr);
      pte = *ptep;
      if(pte_present(pte)) {
	ret = (unsigned long) page_address(pte_page(pte));
	ret |= (adr & (PAGE_SIZE - 1));
      }
    }
  }
  //  printk(KERN_INFO "uv2kva(%lx-->%lx) \n", adr, ret);
  return ret;
}