ksm: five little cleanups
authorHugh Dickins <hugh.dickins@tiscali.co.uk>
Tue, 22 Sep 2009 00:02:15 +0000 (17:02 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 22 Sep 2009 14:17:32 +0000 (07:17 -0700)
1. We don't use __break_cow entry point now: merge it into break_cow.
2. remove_all_slot_rmap_items is just a special case of
   remove_trailing_rmap_items: use the latter instead.
3. Extend comment on unmerge_ksm_pages and rmap_items.
4. try_to_merge_two_pages should use try_to_merge_with_ksm_page
   instead of duplicating its code; and so swap them around.
5. Comment on cmp_and_merge_page described last year's: update it.

Signed-off-by: Hugh Dickins <hugh.dickins@tiscali.co.uk>
Acked-by: Izik Eidus <ieidus@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/ksm.c

index 2849422..c49bb71 100644 (file)
--- a/mm/ksm.c
+++ b/mm/ksm.c
@@ -315,22 +315,18 @@ static void break_ksm(struct vm_area_struct *vma, unsigned long addr)
        /* Which leaves us looping there if VM_FAULT_OOM: hmmm... */
 }
 
-static void __break_cow(struct mm_struct *mm, unsigned long addr)
+static void break_cow(struct mm_struct *mm, unsigned long addr)
 {
        struct vm_area_struct *vma;
 
+       down_read(&mm->mmap_sem);
        vma = find_vma(mm, addr);
        if (!vma || vma->vm_start > addr)
-               return;
+               goto out;
        if (!(vma->vm_flags & VM_MERGEABLE) || !vma->anon_vma)
-               return;
+               goto out;
        break_ksm(vma, addr);
-}
-
-static void break_cow(struct mm_struct *mm, unsigned long addr)
-{
-       down_read(&mm->mmap_sem);
-       __break_cow(mm, addr);
+out:
        up_read(&mm->mmap_sem);
 }
 
@@ -439,17 +435,6 @@ static void remove_rmap_item_from_tree(struct rmap_item *rmap_item)
        cond_resched();         /* we're called from many long loops */
 }
 
-static void remove_all_slot_rmap_items(struct mm_slot *mm_slot)
-{
-       struct rmap_item *rmap_item, *node;
-
-       list_for_each_entry_safe(rmap_item, node, &mm_slot->rmap_list, link) {
-               remove_rmap_item_from_tree(rmap_item);
-               list_del(&rmap_item->link);
-               free_rmap_item(rmap_item);
-       }
-}
-
 static void remove_trailing_rmap_items(struct mm_slot *mm_slot,
                                       struct list_head *cur)
 {
@@ -471,6 +456,11 @@ static void remove_trailing_rmap_items(struct mm_slot *mm_slot,
  * page and upping mmap_sem.  Nor does it fit with the way we skip dup'ing
  * rmap_items from parent to child at fork time (so as not to waste time
  * if exit comes before the next scan reaches it).
+ *
+ * Similarly, although we'd like to remove rmap_items (so updating counts
+ * and freeing memory) when unmerging an area, it's easier to leave that
+ * to the next pass of ksmd - consider, for example, how ksmd might be
+ * in cmp_and_merge_page on one of the rmap_items we would be removing.
  */
 static void unmerge_ksm_pages(struct vm_area_struct *vma,
                              unsigned long start, unsigned long end)
@@ -495,7 +485,7 @@ static void unmerge_and_remove_all_rmap_items(void)
                                continue;
                        unmerge_ksm_pages(vma, vma->vm_start, vma->vm_end);
                }
-               remove_all_slot_rmap_items(mm_slot);
+               remove_trailing_rmap_items(mm_slot, mm_slot->rmap_list.next);
                up_read(&mm->mmap_sem);
        }
 
@@ -533,7 +523,7 @@ static void remove_mm_from_lists(struct mm_struct *mm)
        list_del(&mm_slot->mm_list);
        spin_unlock(&ksm_mmlist_lock);
 
-       remove_all_slot_rmap_items(mm_slot);
+       remove_trailing_rmap_items(mm_slot, mm_slot->rmap_list.next);
        free_mm_slot(mm_slot);
        clear_bit(MMF_VM_MERGEABLE, &mm->flags);
 }
@@ -740,6 +730,29 @@ out:
 }
 
 /*
+ * try_to_merge_with_ksm_page - like try_to_merge_two_pages,
+ * but no new kernel page is allocated: kpage must already be a ksm page.
+ */
+static int try_to_merge_with_ksm_page(struct mm_struct *mm1,
+                                     unsigned long addr1,
+                                     struct page *page1,
+                                     struct page *kpage)
+{
+       struct vm_area_struct *vma;
+       int err = -EFAULT;
+
+       down_read(&mm1->mmap_sem);
+       vma = find_vma(mm1, addr1);
+       if (!vma || vma->vm_start > addr1)
+               goto out;
+
+       err = try_to_merge_one_page(vma, page1, kpage);
+out:
+       up_read(&mm1->mmap_sem);
+       return err;
+}
+
+/*
  * try_to_merge_two_pages - take two identical pages and prepare them
  * to be merged into one page.
  *
@@ -772,9 +785,8 @@ static int try_to_merge_two_pages(struct mm_struct *mm1, unsigned long addr1,
        down_read(&mm1->mmap_sem);
        vma = find_vma(mm1, addr1);
        if (!vma || vma->vm_start > addr1) {
-               put_page(kpage);
                up_read(&mm1->mmap_sem);
-               return err;
+               goto out;
        }
 
        copy_user_highpage(kpage, page1, addr1, vma);
@@ -782,56 +794,20 @@ static int try_to_merge_two_pages(struct mm_struct *mm1, unsigned long addr1,
        up_read(&mm1->mmap_sem);
 
        if (!err) {
-               down_read(&mm2->mmap_sem);
-               vma = find_vma(mm2, addr2);
-               if (!vma || vma->vm_start > addr2) {
-                       put_page(kpage);
-                       up_read(&mm2->mmap_sem);
-                       break_cow(mm1, addr1);
-                       return -EFAULT;
-               }
-
-               err = try_to_merge_one_page(vma, page2, kpage);
-               up_read(&mm2->mmap_sem);
-
+               err = try_to_merge_with_ksm_page(mm2, addr2, page2, kpage);
                /*
-                * If the second try_to_merge_one_page failed, we have a
-                * ksm page with just one pte pointing to it, so break it.
+                * If that fails, we have a ksm page with only one pte
+                * pointing to it: so break it.
                 */
                if (err)
                        break_cow(mm1, addr1);
        }
-
+out:
        put_page(kpage);
        return err;
 }
 
 /*
- * try_to_merge_with_ksm_page - like try_to_merge_two_pages,
- * but no new kernel page is allocated: kpage must already be a ksm page.
- */
-static int try_to_merge_with_ksm_page(struct mm_struct *mm1,
-                                     unsigned long addr1,
-                                     struct page *page1,
-                                     struct page *kpage)
-{
-       struct vm_area_struct *vma;
-       int err = -EFAULT;
-
-       down_read(&mm1->mmap_sem);
-       vma = find_vma(mm1, addr1);
-       if (!vma || vma->vm_start > addr1) {
-               up_read(&mm1->mmap_sem);
-               return err;
-       }
-
-       err = try_to_merge_one_page(vma, page1, kpage);
-       up_read(&mm1->mmap_sem);
-
-       return err;
-}
-
-/*
  * stable_tree_search - search page inside the stable tree
  * @page: the page that we are searching identical pages to.
  * @page2: pointer into identical page that we are holding inside the stable
@@ -1033,10 +1009,10 @@ static void stable_tree_append(struct rmap_item *rmap_item,
 }
 
 /*
- * cmp_and_merge_page - take a page computes its hash value and check if there
- * is similar hash value to different page,
- * in case we find that there is similar hash to different page we call to
- * try_to_merge_two_pages().
+ * cmp_and_merge_page - first see if page can be merged into the stable tree;
+ * if not, compare checksum to previous and if it's the same, see if page can
+ * be inserted into the unstable tree, or merged with a page already there and
+ * both transferred to the stable tree.
  *
  * @page: the page that we are searching identical page to.
  * @rmap_item: the reverse mapping into the virtual address of this page