mm/slab: remove the checks for slab implementation bug
authorJoonsoo Kim <iamjoonsoo.kim@lge.com>
Tue, 15 Mar 2016 21:54:12 +0000 (14:54 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Tue, 15 Mar 2016 23:55:16 +0000 (16:55 -0700)
Some of "#if DEBUG" are for reporting slab implementation bug rather
than user usecase bug.  It's not really needed because slab is stable
for a quite long time and it makes code too dirty.  This patch remove
it.

Signed-off-by: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Acked-by: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Cc: Jesper Dangaard Brouer <brouer@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/slab.c

index 3634dc1..14c3f9c 100644 (file)
--- a/mm/slab.c
+++ b/mm/slab.c
@@ -2110,8 +2110,6 @@ __kmem_cache_create (struct kmem_cache *cachep, unsigned long flags)
        if (!(flags & SLAB_DESTROY_BY_RCU))
                flags |= SLAB_POISON;
 #endif
-       if (flags & SLAB_DESTROY_BY_RCU)
-               BUG_ON(flags & SLAB_POISON);
 #endif
 
        /*
@@ -2368,9 +2366,6 @@ static int drain_freelist(struct kmem_cache *cache,
                }
 
                page = list_entry(p, struct page, lru);
-#if DEBUG
-               BUG_ON(page->active);
-#endif
                list_del(&page->lru);
                /*
                 * Safe to drop the lock. The slab is no longer linked
@@ -2528,30 +2523,23 @@ static void kmem_flagcheck(struct kmem_cache *cachep, gfp_t flags)
        }
 }
 
-static void *slab_get_obj(struct kmem_cache *cachep, struct page *page,
-                               int nodeid)
+static void *slab_get_obj(struct kmem_cache *cachep, struct page *page)
 {
        void *objp;
 
        objp = index_to_obj(cachep, page, get_free_obj(page, page->active));
        page->active++;
-#if DEBUG
-       WARN_ON(page_to_nid(virt_to_page(objp)) != nodeid);
-#endif
 
        return objp;
 }
 
-static void slab_put_obj(struct kmem_cache *cachep, struct page *page,
-                               void *objp, int nodeid)
+static void slab_put_obj(struct kmem_cache *cachep,
+                       struct page *page, void *objp)
 {
        unsigned int objnr = obj_to_index(cachep, page, objp);
 #if DEBUG
        unsigned int i;
 
-       /* Verify that the slab belongs to the intended node */
-       WARN_ON(page_to_nid(virt_to_page(objp)) != nodeid);
-
        /* Verify double free bug */
        for (i = page->active; i < cachep->num; i++) {
                if (get_free_obj(page, i) == objnr) {
@@ -2817,8 +2805,7 @@ retry:
                        STATS_INC_ACTIVE(cachep);
                        STATS_SET_HIGH(cachep);
 
-                       ac_put_obj(cachep, ac, slab_get_obj(cachep, page,
-                                                                       node));
+                       ac_put_obj(cachep, ac, slab_get_obj(cachep, page));
                }
 
                /* move slabp to correct slabp list: */
@@ -3101,7 +3088,7 @@ retry:
 
        BUG_ON(page->active == cachep->num);
 
-       obj = slab_get_obj(cachep, page, nodeid);
+       obj = slab_get_obj(cachep, page);
        n->free_objects--;
        /* move slabp to correct slabp list: */
        list_del(&page->lru);
@@ -3252,7 +3239,7 @@ static void free_block(struct kmem_cache *cachep, void **objpp,
                page = virt_to_head_page(objp);
                list_del(&page->lru);
                check_spinlock_acquired_node(cachep, node);
-               slab_put_obj(cachep, page, objp, node);
+               slab_put_obj(cachep, page, objp);
                STATS_DEC_ACTIVE(cachep);
                n->free_objects++;
 
@@ -3282,9 +3269,7 @@ static void cache_flusharray(struct kmem_cache *cachep, struct array_cache *ac)
        LIST_HEAD(list);
 
        batchcount = ac->batchcount;
-#if DEBUG
-       BUG_ON(!batchcount || batchcount > ac->avail);
-#endif
+
        check_irq_off();
        n = get_node(cachep, node);
        spin_lock(&n->list_lock);