mm/vmalloc.c: halve the number of comparisons performed in pcpu_get_vm_areas()
authorWei Yang <richard.weiyang@gmail.com>
Wed, 6 Sep 2017 23:24:09 +0000 (16:24 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 7 Sep 2017 00:27:29 +0000 (17:27 -0700)
In pcpu_get_vm_areas(), it checks each range is not overlapped.  To make
sure it is, only (N^2)/2 comparison is necessary, while current code
does N^2 times.  By starting from the next range, it achieves the goal
and the continue could be removed.

Also,

 - the overlap check of two ranges could be done with one clause

 - one typo in comment is fixed.

Link: http://lkml.kernel.org/r/20170803063822.48702-1-richard.weiyang@gmail.com
Signed-off-by: Wei Yang <richard.weiyang@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Cc: Michal Hocko <mhocko@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/vmalloc.c

index a47e3894c775646cd636c3f48bcb75fd4b8771d6..fa409c9092be0aac5a11d2436cb9b8e1d272f656 100644 (file)
@@ -2482,7 +2482,7 @@ static unsigned long pvm_determine_end(struct vmap_area **pnext,
  * matching slot.  While scanning, if any of the areas overlaps with
  * existing vmap_area, the base address is pulled down to fit the
  * area.  Scanning is repeated till all the areas fit and then all
  * matching slot.  While scanning, if any of the areas overlaps with
  * existing vmap_area, the base address is pulled down to fit the
  * area.  Scanning is repeated till all the areas fit and then all
- * necessary data structres are inserted and the result is returned.
+ * necessary data structures are inserted and the result is returned.
  */
 struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
                                     const size_t *sizes, int nr_vms,
  */
 struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
                                     const size_t *sizes, int nr_vms,
@@ -2510,15 +2510,11 @@ struct vm_struct **pcpu_get_vm_areas(const unsigned long *offsets,
                if (start > offsets[last_area])
                        last_area = area;
 
                if (start > offsets[last_area])
                        last_area = area;
 
-               for (area2 = 0; area2 < nr_vms; area2++) {
+               for (area2 = area + 1; area2 < nr_vms; area2++) {
                        unsigned long start2 = offsets[area2];
                        unsigned long end2 = start2 + sizes[area2];
 
                        unsigned long start2 = offsets[area2];
                        unsigned long end2 = start2 + sizes[area2];
 
-                       if (area2 == area)
-                               continue;
-
-                       BUG_ON(start2 >= start && start2 < end);
-                       BUG_ON(end2 <= end && end2 > start);
+                       BUG_ON(start2 < end && start < end2);
                }
        }
        last_end = offsets[last_area] + sizes[last_area];
                }
        }
        last_end = offsets[last_area] + sizes[last_area];