kbuild: full dependency check on asm-offsets.h
[linux-2.6.git] / arch / i386 / mm / hugetlbpage.c
index a8c4514..d524127 100644 (file)
 #include <asm/tlb.h>
 #include <asm/tlbflush.h>
 
-static pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_alloc(struct mm_struct *mm, unsigned long addr)
 {
        pgd_t *pgd;
        pud_t *pud;
-       pmd_t *pmd = NULL;
+       pte_t *pte = NULL;
 
        pgd = pgd_offset(mm, addr);
        pud = pud_alloc(mm, pgd, addr);
-       pmd = pmd_alloc(mm, pud, addr);
-       return (pte_t *) pmd;
+       if (pud)
+               pte = (pte_t *) pmd_alloc(mm, pud, addr);
+       BUG_ON(pte && !pte_none(*pte) && !pte_huge(*pte));
+
+       return pte;
 }
 
-static pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
+pte_t *huge_pte_offset(struct mm_struct *mm, unsigned long addr)
 {
        pgd_t *pgd;
        pud_t *pud;
        pmd_t *pmd = NULL;
 
        pgd = pgd_offset(mm, addr);
-       pud = pud_offset(pgd, addr);
-       pmd = pmd_offset(pud, addr);
+       if (pgd_present(*pgd)) {
+               pud = pud_offset(pgd, addr);
+               if (pud_present(*pud))
+                       pmd = pmd_offset(pud, addr);
+       }
        return (pte_t *) pmd;
 }
 
-static void set_huge_pte(struct mm_struct *mm, struct vm_area_struct *vma, struct page *page, pte_t * page_table, int write_access)
-{
-       pte_t entry;
-
-       add_mm_counter(mm, rss, HPAGE_SIZE / PAGE_SIZE);
-       if (write_access) {
-               entry =
-                   pte_mkwrite(pte_mkdirty(mk_pte(page, vma->vm_page_prot)));
-       } else
-               entry = pte_wrprotect(mk_pte(page, vma->vm_page_prot));
-       entry = pte_mkyoung(entry);
-       mk_pte_huge(entry);
-       set_pte(page_table, entry);
-}
-
 /*
  * This function checks for proper alignment of input addr and len parameters.
  */
@@ -69,77 +60,6 @@ int is_aligned_hugepage_range(unsigned long addr, unsigned long len)
        return 0;
 }
 
-int copy_hugetlb_page_range(struct mm_struct *dst, struct mm_struct *src,
-                       struct vm_area_struct *vma)
-{
-       pte_t *src_pte, *dst_pte, entry;
-       struct page *ptepage;
-       unsigned long addr = vma->vm_start;
-       unsigned long end = vma->vm_end;
-
-       while (addr < end) {
-               dst_pte = huge_pte_alloc(dst, addr);
-               if (!dst_pte)
-                       goto nomem;
-               src_pte = huge_pte_offset(src, addr);
-               entry = *src_pte;
-               ptepage = pte_page(entry);
-               get_page(ptepage);
-               set_pte(dst_pte, entry);
-               add_mm_counter(dst, rss, HPAGE_SIZE / PAGE_SIZE);
-               addr += HPAGE_SIZE;
-       }
-       return 0;
-
-nomem:
-       return -ENOMEM;
-}
-
-int
-follow_hugetlb_page(struct mm_struct *mm, struct vm_area_struct *vma,
-                   struct page **pages, struct vm_area_struct **vmas,
-                   unsigned long *position, int *length, int i)
-{
-       unsigned long vpfn, vaddr = *position;
-       int remainder = *length;
-
-       WARN_ON(!is_vm_hugetlb_page(vma));
-
-       vpfn = vaddr/PAGE_SIZE;
-       while (vaddr < vma->vm_end && remainder) {
-
-               if (pages) {
-                       pte_t *pte;
-                       struct page *page;
-
-                       pte = huge_pte_offset(mm, vaddr);
-
-                       /* hugetlb should be locked, and hence, prefaulted */
-                       WARN_ON(!pte || pte_none(*pte));
-
-                       page = &pte_page(*pte)[vpfn % (HPAGE_SIZE/PAGE_SIZE)];
-
-                       WARN_ON(!PageCompound(page));
-
-                       get_page(page);
-                       pages[i] = page;
-               }
-
-               if (vmas)
-                       vmas[i] = vma;
-
-               vaddr += PAGE_SIZE;
-               ++vpfn;
-               --remainder;
-               ++i;
-       }
-
-       *length = remainder;
-       *position = vaddr;
-
-       return i;
-}
-
 #if 0  /* This is just for testing */
 struct page *
 follow_huge_addr(struct mm_struct *mm, unsigned long address, int write)
@@ -204,92 +124,6 @@ follow_huge_pmd(struct mm_struct *mm, unsigned long address,
 }
 #endif
 
-void unmap_hugepage_range(struct vm_area_struct *vma,
-               unsigned long start, unsigned long end)
-{
-       struct mm_struct *mm = vma->vm_mm;
-       unsigned long address;
-       pte_t pte, *ptep;
-       struct page *page;
-
-       BUG_ON(start & (HPAGE_SIZE - 1));
-       BUG_ON(end & (HPAGE_SIZE - 1));
-
-       for (address = start; address < end; address += HPAGE_SIZE) {
-               ptep = huge_pte_offset(mm, address);
-               if (!ptep)
-                       continue;
-               pte = ptep_get_and_clear(mm, address, ptep);
-               if (pte_none(pte))
-                       continue;
-               page = pte_page(pte);
-               put_page(page);
-       }
-       add_mm_counter(mm ,rss, -((end - start) >> PAGE_SHIFT));
-       flush_tlb_range(vma, start, end);
-}
-
-int hugetlb_prefault(struct address_space *mapping, struct vm_area_struct *vma)
-{
-       struct mm_struct *mm = current->mm;
-       unsigned long addr;
-       int ret = 0;
-
-       BUG_ON(vma->vm_start & ~HPAGE_MASK);
-       BUG_ON(vma->vm_end & ~HPAGE_MASK);
-
-       spin_lock(&mm->page_table_lock);
-       for (addr = vma->vm_start; addr < vma->vm_end; addr += HPAGE_SIZE) {
-               unsigned long idx;
-               pte_t *pte = huge_pte_alloc(mm, addr);
-               struct page *page;
-
-               if (!pte) {
-                       ret = -ENOMEM;
-                       goto out;
-               }
-
-               if (!pte_none(*pte)) {
-                       pmd_t *pmd = (pmd_t *) pte;
-
-                       page = pmd_page(*pmd);
-                       pmd_clear(pmd);
-                       mm->nr_ptes--;
-                       dec_page_state(nr_page_table_pages);
-                       page_cache_release(page);
-               }
-
-               idx = ((addr - vma->vm_start) >> HPAGE_SHIFT)
-                       + (vma->vm_pgoff >> (HPAGE_SHIFT - PAGE_SHIFT));
-               page = find_get_page(mapping, idx);
-               if (!page) {
-                       /* charge the fs quota first */
-                       if (hugetlb_get_quota(mapping)) {
-                               ret = -ENOMEM;
-                               goto out;
-                       }
-                       page = alloc_huge_page();
-                       if (!page) {
-                               hugetlb_put_quota(mapping);
-                               ret = -ENOMEM;
-                               goto out;
-                       }
-                       ret = add_to_page_cache(page, mapping, idx, GFP_ATOMIC);
-                       if (! ret) {
-                               unlock_page(page);
-                       } else {
-                               hugetlb_put_quota(mapping);
-                               free_huge_page(page);
-                               goto out;
-                       }
-               }
-               set_huge_pte(mm, vma, page, pte, vma->vm_flags & VM_WRITE);
-       }
-out:
-       spin_unlock(&mm->page_table_lock);
-       return ret;
-}
-
 /* x86_64 also uses this file */
 
 #ifdef HAVE_ARCH_HUGETLB_UNMAPPED_AREA
@@ -301,7 +135,12 @@ static unsigned long hugetlb_get_unmapped_area_bottomup(struct file *file,
        struct vm_area_struct *vma;
        unsigned long start_addr;
 
-       start_addr = mm->free_area_cache;
+       if (len > mm->cached_hole_size) {
+               start_addr = mm->free_area_cache;
+       } else {
+               start_addr = TASK_UNMAPPED_BASE;
+               mm->cached_hole_size = 0;
+       }
 
 full_search:
        addr = ALIGN(start_addr, HPAGE_SIZE);
@@ -315,6 +154,7 @@ full_search:
                         */
                        if (start_addr != TASK_UNMAPPED_BASE) {
                                start_addr = TASK_UNMAPPED_BASE;
+                               mm->cached_hole_size = 0;
                                goto full_search;
                        }
                        return -ENOMEM;
@@ -323,6 +163,8 @@ full_search:
                        mm->free_area_cache = addr + len;
                        return addr;
                }
+               if (addr + mm->cached_hole_size < vma->vm_start)
+                       mm->cached_hole_size = vma->vm_start - addr;
                addr = ALIGN(vma->vm_end, HPAGE_SIZE);
        }
 }
@@ -334,12 +176,17 @@ static unsigned long hugetlb_get_unmapped_area_topdown(struct file *file,
        struct mm_struct *mm = current->mm;
        struct vm_area_struct *vma, *prev_vma;
        unsigned long base = mm->mmap_base, addr = addr0;
+       unsigned long largest_hole = mm->cached_hole_size;
        int first_time = 1;
 
        /* don't allow allocations above current base */
        if (mm->free_area_cache > base)
                mm->free_area_cache = base;
 
+       if (len <= largest_hole) {
+               largest_hole = 0;
+               mm->free_area_cache  = base;
+       }
 try_again:
        /* make sure it can fit in the remaining address space */
        if (mm->free_area_cache < len)
@@ -360,13 +207,21 @@ try_again:
                 * vma->vm_start, use it:
                 */
                if (addr + len <= vma->vm_start &&
-                               (!prev_vma || (addr >= prev_vma->vm_end)))
+                           (!prev_vma || (addr >= prev_vma->vm_end))) {
                        /* remember the address as a hint for next time */
-                       return (mm->free_area_cache = addr);
-               else
+                       mm->cached_hole_size = largest_hole;
+                       return (mm->free_area_cache = addr);
+               } else {
                        /* pull free_area_cache down to the first hole */
-                       if (mm->free_area_cache == vma->vm_end)
+                       if (mm->free_area_cache == vma->vm_end) {
                                mm->free_area_cache = vma->vm_start;
+                               mm->cached_hole_size = largest_hole;
+                       }
+               }
+
+               /* remember the largest hole we saw so far */
+               if (addr + largest_hole < vma->vm_start)
+                       largest_hole = vma->vm_start - addr;
 
                /* try just below the current vma->vm_start */
                addr = (vma->vm_start - len) & HPAGE_MASK;
@@ -379,6 +234,7 @@ fail:
         */
        if (first_time) {
                mm->free_area_cache = base;
+               largest_hole = 0;
                first_time = 0;
                goto try_again;
        }
@@ -389,6 +245,7 @@ fail:
         * allocations.
         */
        mm->free_area_cache = TASK_UNMAPPED_BASE;
+       mm->cached_hole_size = ~0UL;
        addr = hugetlb_get_unmapped_area_bottomup(file, addr0,
                        len, pgoff, flags);
 
@@ -396,6 +253,7 @@ fail:
         * Restore the topdown base:
         */
        mm->free_area_cache = base;
+       mm->cached_hole_size = ~0UL;
 
        return addr;
 }