/*
* linux/mm/mincore.c
*
- * Copyright (C) 1994-1999 Linus Torvalds
+ * Copyright (C) 1994-2006 Linus Torvalds
*/
/*
#include <linux/mm.h>
#include <linux/mman.h>
#include <linux/syscalls.h>
+#include <linux/swap.h>
+#include <linux/swapops.h>
#include <asm/uaccess.h>
#include <asm/pgtable.h>
* and is up to date; i.e. that no page-in operation would be required
* at this time if an application were to map and access this page.
*/
-static unsigned char mincore_page(struct vm_area_struct * vma,
- unsigned long pgoff)
+static unsigned char mincore_page(struct address_space *mapping, pgoff_t pgoff)
{
unsigned char present = 0;
- struct address_space * as = vma->vm_file->f_mapping;
- struct page * page;
+ struct page *page;
- page = find_get_page(as, pgoff);
+ /*
+ * When tmpfs swaps out a page from a file, any process mapping that
+ * file will not get a swp_entry_t in its pte, but rather it is like
+ * any other file mapping (ie. marked !present and faulted in with
+ * tmpfs's .fault). So swapped out tmpfs mappings are tested here.
+ *
+ * However when tmpfs moves the page from pagecache and into swapcache,
+ * it is still in core, but the find_get_page below won't find it.
+ * No big deal, but make a note of it.
+ */
+ page = find_get_page(mapping, pgoff);
if (page) {
present = PageUptodate(page);
page_cache_release(page);
return present;
}
-static long mincore_vma(struct vm_area_struct * vma,
- unsigned long start, unsigned long end, unsigned char __user * vec)
+/*
+ * Do a chunk of "sys_mincore()". We've already checked
+ * all the arguments, we hold the mmap semaphore: we should
+ * just return the amount of info we're asked for.
+ */
+static long do_mincore(unsigned long addr, unsigned char *vec, unsigned long pages)
{
- long error, i, remaining;
- unsigned char * tmp;
-
- error = -ENOMEM;
- if (!vma->vm_file)
- return error;
+ pgd_t *pgd;
+ pud_t *pud;
+ pmd_t *pmd;
+ pte_t *ptep;
+ spinlock_t *ptl;
+ unsigned long nr;
+ int i;
+ pgoff_t pgoff;
+ struct vm_area_struct *vma = find_vma(current->mm, addr);
- start = ((start - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff;
- if (end > vma->vm_end)
- end = vma->vm_end;
- end = ((end - vma->vm_start) >> PAGE_SHIFT) + vma->vm_pgoff;
-
- error = -EAGAIN;
- tmp = (unsigned char *) __get_free_page(GFP_KERNEL);
- if (!tmp)
- return error;
-
- /* (end - start) is # of pages, and also # of bytes in "vec */
- remaining = (end - start),
+ /*
+ * find_vma() didn't find anything above us, or we're
+ * in an unmapped hole in the address space: ENOMEM.
+ */
+ if (!vma || addr < vma->vm_start)
+ return -ENOMEM;
- error = 0;
- for (i = 0; remaining > 0; remaining -= PAGE_SIZE, i++) {
- int j = 0;
- long thispiece = (remaining < PAGE_SIZE) ?
- remaining : PAGE_SIZE;
+ /*
+ * Calculate how many pages there are left in the last level of the
+ * PTE array for our address.
+ */
+ nr = PTRS_PER_PTE - ((addr >> PAGE_SHIFT) & (PTRS_PER_PTE-1));
- while (j < thispiece)
- tmp[j++] = mincore_page(vma, start++);
+ /*
+ * Don't overrun this vma
+ */
+ nr = min(nr, (vma->vm_end - addr) >> PAGE_SHIFT);
- if (copy_to_user(vec + PAGE_SIZE * i, tmp, thispiece)) {
- error = -EFAULT;
- break;
+ /*
+ * Don't return more than the caller asked for
+ */
+ nr = min(nr, pages);
+
+ pgd = pgd_offset(vma->vm_mm, addr);
+ if (pgd_none_or_clear_bad(pgd))
+ goto none_mapped;
+ pud = pud_offset(pgd, addr);
+ if (pud_none_or_clear_bad(pud))
+ goto none_mapped;
+ pmd = pmd_offset(pud, addr);
+ if (pmd_none_or_clear_bad(pmd))
+ goto none_mapped;
+
+ ptep = pte_offset_map_lock(vma->vm_mm, pmd, addr, &ptl);
+ for (i = 0; i < nr; i++, ptep++, addr += PAGE_SIZE) {
+ unsigned char present;
+ pte_t pte = *ptep;
+
+ if (pte_present(pte)) {
+ present = 1;
+
+ } else if (pte_none(pte)) {
+ if (vma->vm_file) {
+ pgoff = linear_page_index(vma, addr);
+ present = mincore_page(vma->vm_file->f_mapping,
+ pgoff);
+ } else
+ present = 0;
+
+ } else if (pte_file(pte)) {
+ pgoff = pte_to_pgoff(pte);
+ present = mincore_page(vma->vm_file->f_mapping, pgoff);
+
+ } else { /* pte is a swap entry */
+ swp_entry_t entry = pte_to_swp_entry(pte);
+ if (is_migration_entry(entry)) {
+ /* migration entries are always uptodate */
+ present = 1;
+ } else {
+#ifdef CONFIG_SWAP
+ pgoff = entry.val;
+ present = mincore_page(&swapper_space, pgoff);
+#else
+ WARN_ON(1);
+ present = 1;
+#endif
+ }
}
+
+ vec[i] = present;
+ }
+ pte_unmap_unlock(ptep-1, ptl);
+
+ return nr;
+
+none_mapped:
+ if (vma->vm_file) {
+ pgoff = linear_page_index(vma, addr);
+ for (i = 0; i < nr; i++, pgoff++)
+ vec[i] = mincore_page(vma->vm_file->f_mapping, pgoff);
+ } else {
+ for (i = 0; i < nr; i++)
+ vec[i] = 0;
}
- free_page((unsigned long) tmp);
- return error;
+ return nr;
}
/*
asmlinkage long sys_mincore(unsigned long start, size_t len,
unsigned char __user * vec)
{
- int index = 0;
- unsigned long end, limit;
- struct vm_area_struct * vma;
- size_t max;
- int unmapped_error = 0;
- long error;
-
- /* check the arguments */
- if (start & ~PAGE_CACHE_MASK)
- goto einval;
-
- if (start < FIRST_USER_PGD_NR * PGDIR_SIZE)
- goto enomem;
-
- limit = TASK_SIZE;
- if (start >= limit)
- goto enomem;
-
- if (!len)
- return 0;
+ long retval;
+ unsigned long pages;
+ unsigned char *tmp;
- max = limit - start;
- len = PAGE_CACHE_ALIGN(len);
- if (len > max || !len)
- goto enomem;
-
- end = start + len;
+ /* Check the start address: needs to be page-aligned.. */
+ if (start & ~PAGE_CACHE_MASK)
+ return -EINVAL;
- /* check the output buffer whilst holding the lock */
- error = -EFAULT;
- down_read(¤t->mm->mmap_sem);
+ /* ..and we need to be passed a valid user-space range */
+ if (!access_ok(VERIFY_READ, (void __user *) start, len))
+ return -ENOMEM;
- if (!access_ok(VERIFY_WRITE, vec, len >> PAGE_SHIFT))
- goto out;
+ /* This also avoids any overflows on PAGE_CACHE_ALIGN */
+ pages = len >> PAGE_SHIFT;
+ pages += (len & ~PAGE_MASK) != 0;
- /*
- * If the interval [start,end) covers some unmapped address
- * ranges, just ignore them, but return -ENOMEM at the end.
- */
- error = 0;
-
- vma = find_vma(current->mm, start);
- while (vma) {
- /* Here start < vma->vm_end. */
- if (start < vma->vm_start) {
- unmapped_error = -ENOMEM;
- start = vma->vm_start;
- }
+ if (!access_ok(VERIFY_WRITE, vec, pages))
+ return -EFAULT;
- /* Here vma->vm_start <= start < vma->vm_end. */
- if (end <= vma->vm_end) {
- if (start < end) {
- error = mincore_vma(vma, start, end,
- &vec[index]);
- if (error)
- goto out;
- }
- error = unmapped_error;
- goto out;
+ tmp = (void *) __get_free_page(GFP_USER);
+ if (!tmp)
+ return -EAGAIN;
+
+ retval = 0;
+ while (pages) {
+ /*
+ * Do at most PAGE_SIZE entries per iteration, due to
+ * the temporary buffer size.
+ */
+ down_read(¤t->mm->mmap_sem);
+ retval = do_mincore(start, tmp, min(pages, PAGE_SIZE));
+ up_read(¤t->mm->mmap_sem);
+
+ if (retval <= 0)
+ break;
+ if (copy_to_user(vec, tmp, retval)) {
+ retval = -EFAULT;
+ break;
}
-
- /* Here vma->vm_start <= start < vma->vm_end < end. */
- error = mincore_vma(vma, start, vma->vm_end, &vec[index]);
- if (error)
- goto out;
- index += (vma->vm_end - start) >> PAGE_CACHE_SHIFT;
- start = vma->vm_end;
- vma = vma->vm_next;
+ pages -= retval;
+ vec += retval;
+ start += retval << PAGE_SHIFT;
+ retval = 0;
}
-
- /* we found a hole in the area queried if we arrive here */
- error = -ENOMEM;
-
-out:
- up_read(¤t->mm->mmap_sem);
- return error;
-
-einval:
- return -EINVAL;
-enomem:
- return -ENOMEM;
+ free_page((unsigned long) tmp);
+ return retval;
}