[PATCH] ZVC: Support NR_SLAB_RECLAIMABLE / NR_SLAB_UNRECLAIMABLE

Remove the atomic counter for slab_reclaim_pages and replace the counter
and NR_SLAB with two ZVC counter that account for unreclaimable and
reclaimable slab pages: NR_SLAB_RECLAIMABLE and NR_SLAB_UNRECLAIMABLE.

Change the check in vmscan.c to refer to to NR_SLAB_RECLAIMABLE.  The
intend seems to be to check for slab pages that could be freed.
Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

arch/i386/mm/pgtable.c

@@ -60,7 +60,9 @@ void show_mem(void)
 	printk(KERN_INFO "%lu pages writeback\n",
 					global_page_state(NR_WRITEBACK));
 	printk(KERN_INFO "%lu pages mapped\n", global_page_state(NR_FILE_MAPPED));
-	printk(KERN_INFO "%lu pages slab\n", global_page_state(NR_SLAB));
+	printk(KERN_INFO "%lu pages slab\n",
+		global_page_state(NR_SLAB_RECLAIMABLE) +
+		global_page_state(NR_SLAB_UNRECLAIMABLE));
 	printk(KERN_INFO "%lu pages pagetables\n",
 					global_page_state(NR_PAGETABLE));
 }

drivers/base/node.c

@@ -68,7 +68,9 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf)
 		       "Node %d PageTables:   %8lu kB\n"
 		       "Node %d NFS_Unstable: %8lu kB\n"
 		       "Node %d Bounce:       %8lu kB\n"
-		       "Node %d Slab:         %8lu kB\n",
+		       "Node %d Slab:         %8lu kB\n"
+		       "Node %d SReclaimable: %8lu kB\n"
+		       "Node %d SUnreclaim:   %8lu kB\n",
 		       nid, K(i.totalram),
 		       nid, K(i.freeram),
 		       nid, K(i.totalram - i.freeram),
@@ -88,7 +90,10 @@ static ssize_t node_read_meminfo(struct sys_device * dev, char * buf)
 		       nid, K(node_page_state(nid, NR_PAGETABLE)),
 		       nid, K(node_page_state(nid, NR_UNSTABLE_NFS)),
 		       nid, K(node_page_state(nid, NR_BOUNCE)),
-		       nid, K(node_page_state(nid, NR_SLAB)));
+		       nid, K(node_page_state(nid, NR_SLAB_RECLAIMABLE) +
+				node_page_state(nid, NR_SLAB_UNRECLAIMABLE)),
+		       nid, K(node_page_state(nid, NR_SLAB_RECLAIMABLE)),
+		       nid, K(node_page_state(nid, NR_SLAB_UNRECLAIMABLE)));
 	n += hugetlb_report_node_meminfo(nid, buf + n);
 	return n;
 }

fs/proc/proc_misc.c

@@ -170,6 +170,8 @@ static int meminfo_read_proc(char *page, char **start, off_t off,
 		"AnonPages:    %8lu kB\n"
 		"Mapped:       %8lu kB\n"
 		"Slab:         %8lu kB\n"
+		"SReclaimable: %8lu kB\n"
+		"SUnreclaim:   %8lu kB\n"
 		"PageTables:   %8lu kB\n"
 		"NFS_Unstable: %8lu kB\n"
 		"Bounce:       %8lu kB\n"
@@ -197,7 +199,10 @@ static int meminfo_read_proc(char *page, char **start, off_t off,
 		K(global_page_state(NR_WRITEBACK)),
 		K(global_page_state(NR_ANON_PAGES)),
 		K(global_page_state(NR_FILE_MAPPED)),
-		K(global_page_state(NR_SLAB)),
+		K(global_page_state(NR_SLAB_RECLAIMABLE) +
+				global_page_state(NR_SLAB_UNRECLAIMABLE)),
+		K(global_page_state(NR_SLAB_RECLAIMABLE)),
+		K(global_page_state(NR_SLAB_UNRECLAIMABLE)),
 		K(global_page_state(NR_PAGETABLE)),
 		K(global_page_state(NR_UNSTABLE_NFS)),
 		K(global_page_state(NR_BOUNCE)),

include/linux/mmzone.h

@@ -51,7 +51,8 @@ enum zone_stat_item {
 	NR_FILE_MAPPED,	/* pagecache pages mapped into pagetables.
 			   only modified from process context */
 	NR_FILE_PAGES,
-	NR_SLAB,	/* Pages used by slab allocator */
+	NR_SLAB_RECLAIMABLE,
+	NR_SLAB_UNRECLAIMABLE,
 	NR_PAGETABLE,	/* used for pagetables */
 	NR_FILE_DIRTY,
 	NR_WRITEBACK,

include/linux/slab.h

@@ -284,8 +284,6 @@ extern kmem_cache_t	*fs_cachep;
 extern kmem_cache_t	*sighand_cachep;
 extern kmem_cache_t	*bio_cachep;
 
-extern atomic_t slab_reclaim_pages;
-
 #endif	/* __KERNEL__ */
 
 #endif	/* _LINUX_SLAB_H */

mm/mmap.c

@@ -116,7 +116,7 @@ int __vm_enough_memory(long pages, int cap_sys_admin)
 		 * which are reclaimable, under pressure.  The dentry
 		 * cache and most inode caches should fall into this
 		 */
-		free += atomic_read(&slab_reclaim_pages);
+		free += global_page_state(NR_SLAB_RECLAIMABLE);
 
 		/*
 		 * Leave the last 3% for root

mm/nommu.c

@@ -1133,7 +1133,7 @@ int __vm_enough_memory(long pages, int cap_sys_admin)
 		 * which are reclaimable, under pressure.  The dentry
 		 * cache and most inode caches should fall into this
 		 */
-		free += atomic_read(&slab_reclaim_pages);
+		free += global_page_state(NR_SLAB_RECLAIMABLE);
 
 		/*
 		 * Leave the last 3% for root

mm/page_alloc.c

@@ -1304,7 +1304,8 @@ void show_free_areas(void)
 		global_page_state(NR_WRITEBACK),
 		global_page_state(NR_UNSTABLE_NFS),
 		nr_free_pages(),
-		global_page_state(NR_SLAB),
+		global_page_state(NR_SLAB_RECLAIMABLE) +
+			global_page_state(NR_SLAB_UNRECLAIMABLE),
 		global_page_state(NR_FILE_MAPPED),
 		global_page_state(NR_PAGETABLE));
 

mm/slab.c

@@ -735,14 +735,6 @@ static inline void init_lock_keys(void)
 static DEFINE_MUTEX(cache_chain_mutex);
 static struct list_head cache_chain;
 
-/*
- * vm_enough_memory() looks at this to determine how many slab-allocated pages
- * are possibly freeable under pressure
- *
- * SLAB_RECLAIM_ACCOUNT turns this on per-slab
- */
-atomic_t slab_reclaim_pages;
-
 /*
  * chicken and egg problem: delay the per-cpu array allocation
  * until the general caches are up.
@@ -1580,8 +1572,11 @@ static void *kmem_getpages(struct kmem_cache *cachep, gfp_t flags, int nodeid)
 
 	nr_pages = (1 << cachep->gfporder);
 	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
-		atomic_add(nr_pages, &slab_reclaim_pages);
-	add_zone_page_state(page_zone(page), NR_SLAB, nr_pages);
+		add_zone_page_state(page_zone(page),
+			NR_SLAB_RECLAIMABLE, nr_pages);
+	else
+		add_zone_page_state(page_zone(page),
+			NR_SLAB_UNRECLAIMABLE, nr_pages);
 	for (i = 0; i < nr_pages; i++)
 		__SetPageSlab(page + i);
 	return page_address(page);
@@ -1596,7 +1591,12 @@ static void kmem_freepages(struct kmem_cache *cachep, void *addr)
 	struct page *page = virt_to_page(addr);
 	const unsigned long nr_freed = i;
 
-	sub_zone_page_state(page_zone(page), NR_SLAB, nr_freed);
+	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
+		sub_zone_page_state(page_zone(page),
+				NR_SLAB_RECLAIMABLE, nr_freed);
+	else
+		sub_zone_page_state(page_zone(page),
+				NR_SLAB_UNRECLAIMABLE, nr_freed);
 	while (i--) {
 		BUG_ON(!PageSlab(page));
 		__ClearPageSlab(page);
@@ -1605,8 +1605,6 @@ static void kmem_freepages(struct kmem_cache *cachep, void *addr)
 	if (current->reclaim_state)
 		current->reclaim_state->reclaimed_slab += nr_freed;
 	free_pages((unsigned long)addr, cachep->gfporder);
-	if (cachep->flags & SLAB_RECLAIM_ACCOUNT)
-		atomic_sub(1 << cachep->gfporder, &slab_reclaim_pages);
 }
 
 static void kmem_rcu_free(struct rcu_head *head)

mm/slob.c

@@ -339,7 +339,3 @@ void kmem_cache_init(void)
 
 	mod_timer(&slob_timer, jiffies + HZ);
 }
-
-atomic_t slab_reclaim_pages = ATOMIC_INIT(0);
-EXPORT_SYMBOL(slab_reclaim_pages);
-

mm/vmscan.c

@@ -1378,7 +1378,7 @@ unsigned long shrink_all_memory(unsigned long nr_pages)
 	for_each_zone(zone)
 		lru_pages += zone->nr_active + zone->nr_inactive;
 
-	nr_slab = global_page_state(NR_SLAB);
+	nr_slab = global_page_state(NR_SLAB_RECLAIMABLE);
 	/* If slab caches are huge, it's better to hit them first */
 	while (nr_slab >= lru_pages) {
 		reclaim_state.reclaimed_slab = 0;

mm/vmstat.c

@@ -458,7 +458,8 @@ static char *vmstat_text[] = {
 	"nr_anon_pages",
 	"nr_mapped",
 	"nr_file_pages",
-	"nr_slab",
+	"nr_slab_reclaimable",
+	"nr_slab_unreclaimable",
 	"nr_page_table_pages",
 	"nr_dirty",
 	"nr_writeback",