Cpuset aware writeback

Index: linux-2.6.20-rc4-mm1/include/linux/nodemask.h
===================================================================
@@ -352,7 +352,7 @@
 #define node_possible(node)	node_isset((node), node_possible_map)
 #define first_online_node	first_node(node_online_map)
 #define next_online_node(nid)	next_node((nid), node_online_map)
-int highest_possible_node_id(void);
+extern int nr_node_ids;
 #else
 #define num_online_nodes()	1
 #define num_possible_nodes()	1

@@ -360,7 +360,7 @@
 #define node_possible(node)	((node) == 0)
 #define first_online_node	0
 #define next_online_node(nid)	(MAX_NUMNODES)
-#define highest_possible_node_id()	0
+#define nr_node_ids		1
 #endif
 
 #define any_online_node(mask)			\

Index: linux-2.6.20-rc4-mm1/mm/page_alloc.c
===================================================================
@@ -679,6 +679,26 @@
 	return i;
 }
 
+#if MAX_NUMNODES > 1
+int nr_node_ids __read_mostly;
+EXPORT_SYMBOL(nr_node_ids);
+
+/*
+ * Figure out the number of possible node ids.
+ */
+static void __init setup_nr_node_ids(void)
+{
+	unsigned int node;
+	unsigned int highest = 0;
+
+	for_each_node_mask(node, node_possible_map)
+		highest = node;
+	nr_node_ids = highest + 1;
+}
+#else
+static void __init setup_nr_node_ids(void) {}
+#endif
+
 #ifdef CONFIG_NUMA
 /*
  * Called from the slab reaper to drain pagesets on a particular node that

@@ -3318,6 +3338,7 @@
 		min_free_kbytes = 65536;
 	setup_per_zone_pages_min();
 	setup_per_zone_lowmem_reserve();
+	setup_nr_node_ids();
 	return 0;
 }
 module_init(init_per_zone_pages_min)

@@ -3519,18 +3540,4 @@
 EXPORT_SYMBOL(page_to_pfn);
 #endif /* CONFIG_OUT_OF_LINE_PFN_TO_PAGE */
 
-#if MAX_NUMNODES > 1
-/*
- * Find the highest possible node id.
- */
-int highest_possible_node_id(void)
-{
-	unsigned int node;
-	unsigned int highest = 0;
 
-	for_each_node_mask(node, node_possible_map)
-		highest = node;
-	return highest;
-}
-EXPORT_SYMBOL(highest_possible_node_id);
-#endif

Index: linux-2.6.20-rc4-mm1/net/sunrpc/svc.c
===================================================================
@@ -116,7 +116,7 @@
 static int
 svc_pool_map_init_percpu(struct svc_pool_map *m)
 {
-	unsigned int maxpools = highest_possible_processor_id()+1;
+	unsigned int maxpools = nr_node_ids;
 	unsigned int pidx = 0;
 	unsigned int cpu;
 	int err;

@@ -144,7 +144,7 @@
 static int
 svc_pool_map_init_pernode(struct svc_pool_map *m)
 {
-	unsigned int maxpools = highest_possible_node_id()+1;
+	unsigned int maxpools = nr_node_ids;
 	unsigned int pidx = 0;
 	unsigned int node;
 	int err;

Index: linux-2.6.20-rc5/fs/fs-writeback.c
===================================================================
@@ -22,6 +22,7 @@
 #include <linux/blkdev.h>
 #include <linux/backing-dev.h>
 #include <linux/buffer_head.h>
+#include <linux/cpuset.h>
 #include "internal.h"
 
 /**

@@ -223,11 +224,13 @@ __sync_single_inode(struct inode *inode,
 			/*
 			 * The inode is clean, inuse
 			 */
+			cpuset_clear_dirty_nodes(inode);
 			list_move(&inode->i_list, &inode_in_use);
 		} else {
 			/*
 			 * The inode is clean, unused
 			 */
+			cpuset_clear_dirty_nodes(inode);
 			list_move(&inode->i_list, &inode_unused);
 		}
 	}

Index: linux-2.6.20-rc5/fs/inode.c
===================================================================
@@ -22,6 +22,7 @@
 #include <linux/bootmem.h>
 #include <linux/inotify.h>
 #include <linux/mount.h>
+#include <linux/cpuset.h>
 
 /*
  * This is needed for the following functions:

@@ -134,6 +135,7 @@ static struct inode *alloc_inode(struct 
 		inode->i_cdev = NULL;
 		inode->i_rdev = 0;
 		inode->dirtied_when = 0;
+		cpuset_clear_dirty_nodes(inode);
 		if (security_inode_alloc(inode)) {
 			if (inode->i_sb->s_op->destroy_inode)
 				inode->i_sb->s_op->destroy_inode(inode);

Index: linux-2.6.20-rc5/include/linux/fs.h
===================================================================
@@ -589,6 +589,9 @@ struct inode {
 	void			*i_security;
 #endif
 	void			*i_private; /* fs or device private pointer */
+#ifdef CONFIG_CPUSETS
+	nodemask_t		dirty_nodes;	/* Map of nodes with dirty pages */
+#endif
 };
 
 /*

Index: linux-2.6.20-rc5/mm/page-writeback.c
===================================================================
@@ -33,6 +33,7 @@
 #include <linux/syscalls.h>
 #include <linux/buffer_head.h>
 #include <linux/pagevec.h>
+#include <linux/cpuset.h>
 
 /*
  * The maximum number of pages to writeout in a single bdflush/kupdate

@@ -780,6 +781,7 @@ int __set_page_dirty_nobuffers(struct pa
 		if (mapping->host) {
 			/* !PageAnon && !swapper_space */
 			__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
+			cpuset_update_dirty_nodes(mapping->host, page);
 		}
 		return 1;
 	}

Index: linux-2.6.20-rc5/fs/buffer.c
===================================================================
@@ -42,6 +42,7 @@
 #include <linux/bitops.h>
 #include <linux/mpage.h>
 #include <linux/bit_spinlock.h>
+#include <linux/cpuset.h>
 
 static int fsync_buffers_list(spinlock_t *lock, struct list_head *list);
 static void invalidate_bh_lrus(void);

@@ -739,6 +740,7 @@ int __set_page_dirty_buffers(struct page
 	}
 	write_unlock_irq(&mapping->tree_lock);
 	__mark_inode_dirty(mapping->host, I_DIRTY_PAGES);
+	cpuset_update_dirty_nodes(mapping->host, page);
 	return 1;
 }
 EXPORT_SYMBOL(__set_page_dirty_buffers);

Index: linux-2.6.20-rc5/include/linux/cpuset.h
===================================================================
@@ -75,6 +75,19 @@ static inline int cpuset_do_slab_mem_spr
 
 extern void cpuset_track_online_nodes(void);
 
+/*
+ * We need macros since struct inode is not defined yet
+ */
+#define cpuset_update_dirty_nodes(__inode, __page) \
+	node_set(page_to_nid(__page), (__inode)->dirty_nodes)
+
+#define cpuset_clear_dirty_nodes(__inode) \
+		(__inode)->dirty_nodes = NODE_MASK_NONE
+
+#define cpuset_intersects_dirty_nodes(__inode, __nodemask_ptr) \
+		(!(__nodemask_ptr) || nodes_intersects((__inode)->dirty_nodes, \
+		*(__nodemask_ptr)))
+
 #else /* !CONFIG_CPUSETS */
 
 static inline int cpuset_init_early(void) { return 0; }

@@ -146,6 +159,17 @@ static inline int cpuset_do_slab_mem_spr
 
 static inline void cpuset_track_online_nodes(void) {}
 
+static inline void cpuset_update_dirty_nodes(struct inode *i,
+		struct page *page) {}
+
+static inline void cpuset_clear_dirty_nodes(struct inode *i) {}
+
+static inline int cpuset_intersects_dirty_nodes(struct inode *i,
+		nodemask_t *n)
+{
+	return 1;
+}
+
 #endif /* !CONFIG_CPUSETS */
 
 #endif /* _LINUX_CPUSET_H */

Index: linux-2.6.20-rc5/include/linux/writeback.h
===================================================================
@@ -81,7 +81,7 @@ static inline void wait_on_inode(struct 
 /*
  * mm/page-writeback.c
  */
-int wakeup_pdflush(long nr_pages);
+int wakeup_pdflush(long nr_pages, nodemask_t *nodes);
 void laptop_io_completion(void);
 void laptop_sync_completion(void);
 void throttle_vm_writeout(void);

@@ -109,7 +109,8 @@ balance_dirty_pages_ratelimited(struct a
 	balance_dirty_pages_ratelimited_nr(mapping, 1);
 }
 
-int pdflush_operation(void (*fn)(unsigned long), unsigned long arg0);
+int pdflush_operation(void (*fn)(unsigned long, nodemask_t *nodes),
+		unsigned long arg0, nodemask_t *nodes);
 extern int generic_writepages(struct address_space *mapping,
 			      struct writeback_control *wbc);
 int do_writepages(struct address_space *mapping, struct writeback_control *wbc);

Index: linux-2.6.20-rc5/mm/page-writeback.c
===================================================================
@@ -101,7 +101,7 @@ EXPORT_SYMBOL(laptop_mode);
 /* End of sysctl-exported parameters */
 
 
-static void background_writeout(unsigned long _min_pages);
+static void background_writeout(unsigned long _min_pages, nodemask_t *nodes);
 
 /*
  * Work out the current dirty-memory clamping and background writeout

@@ -244,7 +244,7 @@ static void balance_dirty_pages(struct a
 	 */
 	if ((laptop_mode && pages_written) ||
 	     (!laptop_mode && (nr_reclaimable > background_thresh)))
-		pdflush_operation(background_writeout, 0);
+		pdflush_operation(background_writeout, 0, NULL);
 }
 
 void set_page_dirty_balance(struct page *page)

@@ -325,7 +325,7 @@ void throttle_vm_writeout(void)
  * writeback at least _min_pages, and keep writing until the amount of dirty
  * memory is less than the background threshold, or until we're all clean.
  */
-static void background_writeout(unsigned long _min_pages)
+static void background_writeout(unsigned long _min_pages, nodemask_t *unused)
 {
 	long min_pages = _min_pages;
 	struct writeback_control wbc = {

@@ -365,12 +365,12 @@ static void background_writeout(unsigned
  * the whole world.  Returns 0 if a pdflush thread was dispatched.  Returns
  * -1 if all pdflush threads were busy.
  */
-int wakeup_pdflush(long nr_pages)
+int wakeup_pdflush(long nr_pages, nodemask_t *nodes)
 {
 	if (nr_pages == 0)
 		nr_pages = global_page_state(NR_FILE_DIRTY) +
 				global_page_state(NR_UNSTABLE_NFS);
-	return pdflush_operation(background_writeout, nr_pages);
+	return pdflush_operation(background_writeout, nr_pages, nodes);
 }
 
 static void wb_timer_fn(unsigned long unused);

@@ -394,7 +394,7 @@ static DEFINE_TIMER(laptop_mode_wb_timer
  * older_than_this takes precedence over nr_to_write.  So we'll only write back
  * all dirty pages if they are all attached to "old" mappings.
  */
-static void wb_kupdate(unsigned long arg)
+static void wb_kupdate(unsigned long arg, nodemask_t *unused)
 {
 	unsigned long oldest_jif;
 	unsigned long start_jif;

@@ -454,18 +454,18 @@ int dirty_writeback_centisecs_handler(ct
 
 static void wb_timer_fn(unsigned long unused)
 {
-	if (pdflush_operation(wb_kupdate, 0) < 0)
+	if (pdflush_operation(wb_kupdate, 0, NULL) < 0)
 		mod_timer(&wb_timer, jiffies + HZ); /* delay 1 second */
 }
 
-static void laptop_flush(unsigned long unused)
+static void laptop_flush(unsigned long unused, nodemask_t *unused2)
 {
 	sys_sync();
 }
 
 static void laptop_timer_fn(unsigned long unused)
 {
-	pdflush_operation(laptop_flush, 0);
+	pdflush_operation(laptop_flush, 0, NULL);
 }
 
 /*

Index: linux-2.6.20-rc5/mm/pdflush.c
===================================================================
@@ -83,10 +83,12 @@ static unsigned long last_empty_jifs;
  */
 struct pdflush_work {
 	struct task_struct *who;	/* The thread */
-	void (*fn)(unsigned long);	/* A callback function */
+	void (*fn)(unsigned long, nodemask_t *); /* A callback function */
 	unsigned long arg0;		/* An argument to the callback */
 	struct list_head list;		/* On pdflush_list, when idle */
 	unsigned long when_i_went_to_sleep;
+	int have_nodes;			/* Nodes were specified */
+	nodemask_t nodes;		/* Nodes of interest */
 };
 
 static int __pdflush(struct pdflush_work *my_work)

@@ -123,7 +125,8 @@ static int __pdflush(struct pdflush_work
 		}
 		spin_unlock_irq(&pdflush_lock);
 
-		(*my_work->fn)(my_work->arg0);
+		(*my_work->fn)(my_work->arg0,
+			my_work->have_nodes ? &my_work->nodes : NULL);
 
 		/*
 		 * Thread creation: For how long have there been zero

@@ -197,7 +200,8 @@ static int pdflush(void *dummy)
  * Returns zero if it indeed managed to find a worker thread, and passed your
  * payload to it.
  */
-int pdflush_operation(void (*fn)(unsigned long), unsigned long arg0)
+int pdflush_operation(void (*fn)(unsigned long, nodemask_t *),
+			unsigned long arg0, nodemask_t *nodes)
 {
 	unsigned long flags;
 	int ret = 0;

@@ -217,6 +221,11 @@ int pdflush_operation(void (*fn)(unsigne
 			last_empty_jifs = jiffies;
 		pdf->fn = fn;
 		pdf->arg0 = arg0;
+		if (nodes) {
+			pdf->nodes = *nodes;
+			pdf->have_nodes = 1;
+		} else
+			pdf->have_nodes = 0;
 		wake_up_process(pdf->who);
 		spin_unlock_irqrestore(&pdflush_lock, flags);
 	}

Index: linux-2.6.20-rc5/mm/vmscan.c
===================================================================
@@ -1065,7 +1065,7 @@ unsigned long try_to_free_pages(struct z
 		 */
 		if (total_scanned > sc.swap_cluster_max +
 					sc.swap_cluster_max / 2) {
-			wakeup_pdflush(laptop_mode ? 0 : total_scanned);
+			wakeup_pdflush(laptop_mode ? 0 : total_scanned, NULL);
 			sc.may_writepage = 1;
 		}
 

Index: linux-2.6.20-rc5/fs/buffer.c
===================================================================
@@ -357,7 +357,7 @@ static void free_more_memory(void)
 	struct zone **zones;
 	pg_data_t *pgdat;
 
-	wakeup_pdflush(1024);
+	wakeup_pdflush(1024, NULL);
 	yield();
 
 	for_each_online_pgdat(pgdat) {

Index: linux-2.6.20-rc5/fs/super.c
===================================================================
@@ -618,7 +618,7 @@ int do_remount_sb(struct super_block *sb
 	return 0;
 }
 
-static void do_emergency_remount(unsigned long foo)
+static void do_emergency_remount(unsigned long foo, nodemask_t *bar)
 {
 	struct super_block *sb;
 

@@ -646,7 +646,7 @@ static void do_emergency_remount(unsigne
 
 void emergency_remount(void)
 {
-	pdflush_operation(do_emergency_remount, 0);
+	pdflush_operation(do_emergency_remount, 0, NULL);
 }
 
 /*

Index: linux-2.6.20-rc5/fs/sync.c
===================================================================
@@ -21,9 +21,9 @@
  * sync everything.  Start out by waking pdflush, because that writes back
  * all queues in parallel.
  */
-static void do_sync(unsigned long wait)
+static void do_sync(unsigned long wait, nodemask_t *unused)
 {
-	wakeup_pdflush(0);
+	wakeup_pdflush(0, NULL);
 	sync_inodes(0);		/* All mappings, inodes and their blockdevs */
 	DQUOT_SYNC(NULL);
 	sync_supers();		/* Write the superblocks */

@@ -38,13 +38,13 @@ static void do_sync(unsigned long wait)
 
 asmlinkage long sys_sync(void)
 {
-	do_sync(1);
+	do_sync(1, NULL);
 	return 0;
 }
 
 void emergency_sync(void)
 {
-	pdflush_operation(do_sync, 0);
+	pdflush_operation(do_sync, 0, NULL);
 }
 
 /*

Index: linux-2.6.20-rc5/include/linux/writeback.h
===================================================================
@@ -59,11 +59,12 @@ struct writeback_control {
 	unsigned for_reclaim:1;		/* Invoked from the page allocator */
 	unsigned for_writepages:1;	/* This is a writepages() call */
 	unsigned range_cyclic:1;	/* range_start is cyclic */
+	nodemask_t *nodes;		/* Set of nodes of interest */
 };
 
 /*
  * fs/fs-writeback.c
- */	
+ */
 void writeback_inodes(struct writeback_control *wbc);
 void wake_up_inode(struct inode *inode);
 int inode_wait(void *);

221 lines not on this page · raw message

Index: linux-2.6.20-rc5/fs/fs-writeback.c
===================================================================
@@ -352,6 +352,12 @@ sync_sb_inodes(struct super_block *sb, s
 			continue;		/* blockdev has wrong queue */
 		}
 
+		if (!cpuset_intersects_dirty_nodes(inode, wbc->nodes)) {
+			/* No pages on the nodes under writeback */
+			list_move(&inode->i_list, &sb->s_dirty);
+			continue;
+		}
+
 		/* Was this inode dirtied after sync_sb_inodes was called? */
 		if (time_after(inode->dirtied_when, start))
 			break;

Index: linux-2.6.20-rc5/mm/vmscan.c
===================================================================
@@ -1065,7 +1065,8 @@ unsigned long try_to_free_pages(struct z
 		 */
 		if (total_scanned > sc.swap_cluster_max +
 					sc.swap_cluster_max / 2) {
-			wakeup_pdflush(laptop_mode ? 0 : total_scanned, NULL);
+			wakeup_pdflush(laptop_mode ? 0 : total_scanned,
+				&cpuset_current_mems_allowed);
 			sc.may_writepage = 1;
 		}
 

Index: linux-2.6.20-rc5/include/linux/writeback.h
===================================================================
@@ -85,7 +85,7 @@ static inline void wait_on_inode(struct 
 int wakeup_pdflush(long nr_pages, nodemask_t *nodes);
 void laptop_io_completion(void);
 void laptop_sync_completion(void);
-void throttle_vm_writeout(void);
+void throttle_vm_writeout(nodemask_t *);
 
 /* These are exported to sysctl. */
 extern int dirty_background_ratio;

Index: linux-2.6.20-rc5/mm/page-writeback.c
===================================================================
@@ -349,12 +349,12 @@ void balance_dirty_pages_ratelimited_nr(
 }
 EXPORT_SYMBOL(balance_dirty_pages_ratelimited_nr);
 
-void throttle_vm_writeout(void)
+void throttle_vm_writeout(nodemask_t *nodes)
 {
 	struct dirty_limits dl;
 
         for ( ; ; ) {
-		get_dirty_limits(&dl, NULL, &node_online_map);
+		get_dirty_limits(&dl, NULL, nodes);
 
                 /*
                  * Boost the allowable dirty threshold a bit for page

Index: linux-2.6.20-rc5/mm/vmscan.c
===================================================================
@@ -949,7 +949,7 @@ static unsigned long shrink_zone(int pri
 		}
 	}
 
-	throttle_vm_writeout();
+	throttle_vm_writeout(&cpuset_current_mems_allowed);
 
 	atomic_dec(&zone->reclaim_in_progress);
 	return nr_reclaimed;

Index: linux-2.6.20-rc5/include/linux/mmzone.h
===================================================================
@@ -53,6 +53,7 @@ enum zone_stat_item {
 	NR_FILE_PAGES,
 	NR_SLAB_RECLAIMABLE,
 	NR_SLAB_UNRECLAIMABLE,
+	NR_UNRECLAIMABLE,
 	NR_PAGETABLE,	/* used for pagetables */
 	NR_FILE_DIRTY,
 	NR_WRITEBACK,

Index: linux-2.6.20-rc5/fs/proc/proc_misc.c
===================================================================
@@ -174,6 +174,7 @@ static int meminfo_read_proc(char *page,
 		"Slab:         %8lu kB\n"
 		"SReclaimable: %8lu kB\n"
 		"SUnreclaim:   %8lu kB\n"
+		"Unreclaimabl: %8lu kB\n"
 		"PageTables:   %8lu kB\n"
 		"NFS_Unstable: %8lu kB\n"
 		"Bounce:       %8lu kB\n"

@@ -205,6 +206,7 @@ static int meminfo_read_proc(char *page,
 				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_UNRECLAIMABLE)),
 		K(global_page_state(NR_PAGETABLE)),
 		K(global_page_state(NR_UNSTABLE_NFS)),
 		K(global_page_state(NR_BOUNCE)),

Index: linux-2.6.20-rc5/mm/hugetlb.c
===================================================================
@@ -115,6 +115,8 @@ static int alloc_fresh_huge_page(void)
 		nr_huge_pages_node[page_to_nid(page)]++;
 		spin_unlock(&hugetlb_lock);
 		put_page(page); /* free it into the hugepage allocator */
+		mod_zone_page_state(page_zone(page), NR_UNRECLAIMABLE,
+					HPAGE_SIZE / PAGE_SIZE);
 		return 1;
 	}
 	return 0;

@@ -183,6 +185,8 @@ static void update_and_free_page(struct 
 				1 << PG_dirty | 1 << PG_active | 1 << PG_reserved |
 				1 << PG_private | 1<< PG_writeback);
 	}
+	mod_zone_page_state(page_zone(page), NR_UNRECLAIMABLE,
+					- (HPAGE_SIZE / PAGE_SIZE));
 	page[1].lru.next = NULL;
 	set_page_refcounted(page);
 	__free_pages(page, HUGETLB_PAGE_ORDER);

Index: linux-2.6.20-rc5/mm/vmstat.c
===================================================================
@@ -459,6 +459,7 @@ static const char * const vmstat_text[] 
 	"nr_file_pages",
 	"nr_slab_reclaimable",
 	"nr_slab_unreclaimable",
+	"nr_unreclaimable",
 	"nr_page_table_pages",
 	"nr_dirty",
 	"nr_writeback",

Index: linux-2.6.20-rc5/mm/page-writeback.c
===================================================================
@@ -165,7 +165,9 @@ get_dirty_limits(struct dirty_limits *dl
 			dl->nr_writeback +=
 				node_page_state(node, NR_WRITEBACK);
 			available_memory +=
-				NODE_DATA(node)->node_present_pages;
+				NODE_DATA(node)->node_present_pages
+				- node_page_state(node, NR_UNRECLAIMABLE)
+				- node_page_state(node, NR_SLAB_UNRECLAIMABLE);
 #ifdef CONFIG_HIGHMEM
 			high_memory += NODE_DATA(node)
 				->node_zones[ZONE_HIGHMEM]->present_pages;

@@ -180,7 +182,9 @@ get_dirty_limits(struct dirty_limits *dl
 		dl->nr_dirty = global_page_state(NR_FILE_DIRTY);
 		dl->nr_unstable = global_page_state(NR_UNSTABLE_NFS);
 		dl->nr_writeback = global_page_state(NR_WRITEBACK);
-		available_memory = vm_total_pages;
+		available_memory = vm_total_pages
+				- global_page_state(NR_UNRECLAIMABLE)
+				- global_page_state(NR_SLAB_UNRECLAIMABLE);
 		high_memory = totalhigh_pages;
 		nr_mapped = global_page_state(NR_FILE_MAPPED) +
 				global_page_state(NR_ANON_PAGES);

Index: linux-2.6.20-rc5/drivers/base/node.c
===================================================================
@@ -70,7 +70,8 @@ static ssize_t node_read_meminfo(struct 
 		       "Node %d Bounce:       %8lu kB\n"
 		       "Node %d Slab:         %8lu kB\n"
 		       "Node %d SReclaimable: %8lu kB\n"
-		       "Node %d SUnreclaim:   %8lu kB\n",
+		       "Node %d SUnreclaim:   %8lu kB\n"
+		       "Node %d Unreclaimabl: %8lu kB\n",
 		       nid, K(i.totalram),
 		       nid, K(i.freeram),
 		       nid, K(i.totalram - i.freeram),

@@ -93,7 +94,8 @@ static ssize_t node_read_meminfo(struct 
 		       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)));
+		       nid, K(node_page_state(nid, NR_SLAB_UNRECLAIMABLE)),
+		       nid, K(node_page_state(nid, NR_UNRECLAIMABLE)));
 	n += hugetlb_report_node_meminfo(nid, buf + n);
 	return n;
 }

Index: git-linux/mm/page-writeback.c
===================================================================
@@ -101,6 +101,18 @@ EXPORT_SYMBOL(laptop_mode);
 
 static void background_writeout(unsigned long _min_pages);
 
+/* Maximal number of pages that can be consumed by pageable caches. */
+static unsigned long total_pageable_pages(void)
+{
+	unsigned long active;
+	unsigned long inactive;
+	unsigned long free;
+
+	get_zone_counts(&active, &inactive, &free);
+	/* +1 to never return 0. */
+	return active + inactive + free + 1;
+}
+
 /*
  * Work out the current dirty-memory clamping and background writeout
  * thresholds.

@@ -127,22 +139,31 @@ get_dirty_limits(long *pbackground, long
 	int unmapped_ratio;
 	long background;
 	long dirty;
-	unsigned long available_memory = vm_total_pages;
+	unsigned long total_pages;
+	unsigned long available_memory;
 	struct task_struct *tsk;
 
+	available_memory = total_pages = total_pageable_pages();
+
 #ifdef CONFIG_HIGHMEM
 	/*
 	 * If this mapping can only allocate from low memory,
 	 * we exclude high memory from our count.
 	 */
-	if (mapping && !(mapping_gfp_mask(mapping) & __GFP_HIGHMEM))
+	if (mapping && !(mapping_gfp_mask(mapping) & __GFP_HIGHMEM)) {
+		if (available_memory > totalhigh_pages)
 		available_memory -= totalhigh_pages;
+		else
+			available_memory = 1;
+	}
 #endif
 
 
 	unmapped_ratio = 100 - ((global_page_state(NR_FILE_MAPPED) +
 				global_page_state(NR_ANON_PAGES)) * 100) /
-					vm_total_pages;
+					total_pages;
+	if (unmapped_ratio < 0)
+		unmapped_ratio = 0;
 
 	dirty_ratio = vm_dirty_ratio;
 	if (dirty_ratio > unmapped_ratio / 2)

@@ -513,7 +534,7 @@ void laptop_sync_completion(void)
 
 void writeback_set_ratelimit(void)
 {
-	ratelimit_pages = vm_total_pages / (num_online_cpus() * 32);
+	ratelimit_pages = total_pageable_pages() / (num_online_cpus() * 32);
 	if (ratelimit_pages < 16)
 		ratelimit_pages = 16;
 	if (ratelimit_pages * PAGE_CACHE_SIZE > 4096 * 1024)

@@ -542,7 +563,7 @@ void __init page_writeback_init(void)
 	long buffer_pages = nr_free_buffer_pages();
 	long correction;
 
-	correction = (100 * 4 * buffer_pages) / vm_total_pages;
+	correction = (100 * 4 * buffer_pages) / total_pageable_pages();
 
 	if (correction < 100) {
 		dirty_background_ratio *= correction;

Index: linux-2.6.20-rc5/fs/xfs/linux-2.6/xfs_super.c
===================================================================
@@ -370,7 +370,9 @@ xfs_fs_inode_init_once(
 STATIC int
 xfs_init_zones(void)
 {
-	xfs_vnode_zone = kmem_zone_init_flags(sizeof(bhv_vnode_t), "xfs_vnode",
+	xfs_vnode_zone = kmem_zone_init_flags(sizeof(bhv_vnode_t)
+						- unused_numa_nodemask_bytes,
+					"xfs_vnode",
 					KM_ZONE_HWALIGN | KM_ZONE_RECLAIM |
 					KM_ZONE_SPREAD,
 					xfs_fs_inode_init_once);

Index: linux-2.6.20-rc5/include/linux/fs.h
===================================================================
@@ -591,6 +591,14 @@ struct inode {
 	void			*i_private; /* fs or device private pointer */
 #ifdef CONFIG_CPUSETS
 	nodemask_t		dirty_nodes;	/* Map of nodes with dirty pages */
+	/*
+	 * Note that we may only use a portion of the bitmap in dirty_nodes
+	 * if we have a large MAX_NUMNODES but the number of possible nodes
+	 * is small in order to reduce the size of the inode.
+	 *
+	 * Bits after nr_node_ids (one node beyond the last possible
+	 * node_id) may not be accessed.
+	 */
 #endif
 };
 

Index: linux-2.6.20-rc5/fs/ext3/super.c
===================================================================
@@ -480,7 +480,8 @@ static void init_once(void * foo, struct
 static int init_inodecache(void)
 {
 	ext3_inode_cachep = kmem_cache_create("ext3_inode_cache",
-					     sizeof(struct ext3_inode_info),
+					     sizeof(struct ext3_inode_info)
+						- unused_numa_nodemask_bytes,
 					     0, (SLAB_RECLAIM_ACCOUNT|
 						SLAB_MEM_SPREAD),
 					     init_once, NULL);

Index: linux-2.6.20-rc5/fs/inode.c
===================================================================
@@ -1399,7 +1399,8 @@ void __init inode_init(unsigned long mem
 
 	/* inode slab cache */
 	inode_cachep = kmem_cache_create("inode_cache",
-					 sizeof(struct inode),
+					 sizeof(struct inode)
+						- unused_numa_nodemask_bytes,
 					 0,
 					 (SLAB_RECLAIM_ACCOUNT|SLAB_PANIC|
 					 SLAB_MEM_SPREAD),

Index: linux-2.6.20-rc5/fs/reiserfs/super.c
===================================================================
@@ -525,11 +525,11 @@ static void init_once(void *foo, struct 
 static int init_inodecache(void)
 {
 	reiserfs_inode_cachep = kmem_cache_create("reiser_inode_cache",
-						  sizeof(struct
-							 reiserfs_inode_info),
-						  0, (SLAB_RECLAIM_ACCOUNT|
-							SLAB_MEM_SPREAD),
-						  init_once, NULL);
+					  sizeof(struct reiserfs_inode_info)
+						 - unused_numa_nodemask_bytes,
+					  0, (SLAB_RECLAIM_ACCOUNT|
+						SLAB_MEM_SPREAD),
+					  init_once, NULL);
 	if (reiserfs_inode_cachep == NULL)
 		return -ENOMEM;
 	return 0;

Index: linux-2.6.20-rc5/include/linux/cpuset.h
===================================================================
@@ -82,11 +82,13 @@ extern void cpuset_track_online_nodes(vo
 	node_set(page_to_nid(__page), (__inode)->dirty_nodes)
 
 #define cpuset_clear_dirty_nodes(__inode) \
-		(__inode)->dirty_nodes = NODE_MASK_NONE
+	bitmap_zero(nodes_addr((__inode)->dirty_nodes), nr_node_ids)
 
 #define cpuset_intersects_dirty_nodes(__inode, __nodemask_ptr) \
-		(!(__nodemask_ptr) || nodes_intersects((__inode)->dirty_nodes, \
-		*(__nodemask_ptr)))
+		(!(__nodemask_ptr) || bitmap_intersects( \
+			nodes_addr((__inode)->dirty_nodes), \
+			nodes_addr(*(__nodemask_ptr)), \
+			nr_node_ids))
 
 #else /* !CONFIG_CPUSETS */
 

Index: linux-2.6.20-rc5/include/linux/nodemask.h
===================================================================
@@ -353,6 +353,13 @@ extern nodemask_t node_possible_map;
 #define first_online_node	first_node(node_online_map)
 #define next_online_node(nid)	next_node((nid), node_online_map)
 extern int nr_node_ids;
+
+#if MAX_NUMNODES > BITS_PER_LONG
+extern int unused_numa_nodemask_bytes;
+#else
+#define unused_numa_nodemask_bytes 0
+#endif
+
 #else
 #define num_online_nodes()	1
 #define num_possible_nodes()	1

@@ -361,6 +368,7 @@ extern int nr_node_ids;
 #define first_online_node	0
 #define next_online_node(nid)	(MAX_NUMNODES)
 #define nr_node_ids		1
+#define unused_numa_nodemask_bytes	0
 #endif
 
 #define any_online_node(mask)			\

Index: linux-2.6.20-rc5/mm/page_alloc.c
===================================================================
@@ -663,6 +663,10 @@ static int rmqueue_bulk(struct zone *zon
 #if MAX_NUMNODES > 1
 int nr_node_ids __read_mostly;
 EXPORT_SYMBOL(nr_node_ids);
+#if MAX_NUMNODES > BITS_PER_LONG
+int unused_numa_nodemask_bytes __read_mostly;
+EXPORT_SYMBOL(unused_numa_nodemask_bytes);
+#endif
 
 /*
  * Figure out the number of possible node ids.

@@ -675,6 +679,10 @@ static void __init setup_nr_node_ids(voi
 	for_each_node_mask(node, node_possible_map)
 		highest = node;
 	nr_node_ids = highest + 1;
+#if MAX_NUMNODES > BITS_PER_LONG
+	unused_numa_nodemask_bytes = BITS_TO_LONGS(MAX_NUMNODES)
+				- BITS_TO_LONGS(nr_node_ids);
+#endif
 }
 #else
 static void __init setup_nr_node_ids(void) {}

Index: linux-2.6.20-rc5/fs/nfs/inode.c
===================================================================
@@ -1136,7 +1136,8 @@ static void init_once(void * foo, struct
 static int __init nfs_init_inodecache(void)
 {
 	nfs_inode_cachep = kmem_cache_create("nfs_inode_cache",
-					     sizeof(struct nfs_inode),
+					     sizeof(struct nfs_inode)
+					     - unused_numa_nodemask_bytes,
 					     0, (SLAB_RECLAIM_ACCOUNT|
 						SLAB_MEM_SPREAD),
 					     init_once, NULL);

Index: linux-2.6-git/fs/nfs/write.c
===================================================================
@@ -89,8 +89,6 @@ static struct kmem_cache *nfs_wdata_cach
 static mempool_t *nfs_wdata_mempool;
 static mempool_t *nfs_commit_mempool;
 
-static DECLARE_WAIT_QUEUE_HEAD(nfs_write_congestion);
-
 struct nfs_write_data *nfs_commit_alloc(void)
 {
 	struct nfs_write_data *p = mempool_alloc(nfs_commit_mempool, GFP_NOFS);

@@ -245,6 +243,91 @@ static int wb_priority(struct writeback_
 }
 
 /*
+ * NFS congestion control
+ *
+ * Congestion is composed of two parts, writeback and commit pages.
+ * Writeback pages are active, that is, all that is required to get out of
+ * writeback state is sit around and wait. Commit pages on the other hand
+ * are not and they need a little nudge to go away.
+ *
+ * Normally we want to maximise the number of commit pages for it allows the
+ * server to make best use of unstable storage. However when we are running
+ * low on memory we do need to reduce the commit pages.
+ *
+ * Hence writeback pages are managed in the conventional way, but for commit
+ * pages we poll on every write.
+ *
+ * The threshold is picked so that it allows for 16M of in flight data
+ * given current transfer speeds that seems reasonable.
+ */
+
+#define NFS_CONGESTION_SIZE		16
+#define NFS_CONGESTION_ON_THRESH 	(NFS_CONGESTION_SIZE << (20 - PAGE_SHIFT))
+#define NFS_CONGESTION_OFF_THRESH	\
+	(NFS_CONGESTION_ON_THRESH - (NFS_CONGESTION_ON_THRESH >> 2))
+
+/*
+ * Include the commit pages into the congestion logic when there is either
+ * pressure from the total dirty limit or when we're in direct reclaim.
+ */
+static inline int commit_pressure(struct inode *inode)
+{
+	return dirty_pages_exceeded(inode->i_mapping) ||
+		(current->flags & PF_MEMALLOC);
+}
+
+static void nfs_congestion_on(struct inode *inode)
+{
+	struct nfs_server *nfss = NFS_SERVER(inode);
+
+	if (atomic_read(&nfss->writeback) > NFS_CONGESTION_ON_THRESH)
+		set_bdi_congested(&nfss->backing_dev_info, WRITE);
+}
+
+static void nfs_congestion_off(struct inode *inode)
+{
+	struct nfs_server *nfss = NFS_SERVER(inode);
+
+	if (atomic_read(&nfss->writeback) < NFS_CONGESTION_OFF_THRESH) {
+		clear_bdi_congested(&nfss->backing_dev_info, WRITE);
+		congestion_end(WRITE);
+	}
+}
+
+static inline void nfs_set_page_writeback(struct page *page)
+{
+	if (!test_set_page_writeback(page)) {
+		struct inode *inode = page->mapping->host;
+		atomic_inc(&NFS_SERVER(inode)->writeback);
+		nfs_congestion_on(inode);
+	}
+}
+
+static inline void nfs_end_page_writeback(struct page *page)
+{
+	struct inode *inode = page->mapping->host;
+	end_page_writeback(page);
+	atomic_dec(&NFS_SERVER(inode)->writeback);
+	nfs_congestion_off(inode);
+}
+
+static inline void nfs_set_page_commit(struct page *page)
+{
+	struct inode *inode = page->mapping->host;
+	inc_zone_page_state(page, NR_UNSTABLE_NFS);
+	atomic_inc(&NFS_SERVER(inode)->commit);
+	nfs_congestion_on(inode);
+}
+
+static inline void nfs_end_page_commit(struct page *page)
+{
+	struct inode *inode = page->mapping->host;
+	dec_zone_page_state(page, NR_UNSTABLE_NFS);
+	atomic_dec(&NFS_SERVER(inode)->commit);
+	nfs_congestion_off(inode);
+}
+
+/*
  * Find an associated nfs write request, and prepare to flush it out
  * Returns 1 if there was no write request, or if the request was
  * already tagged by nfs_set_page_dirty.Returns 0 if the request

@@ -281,7 +364,7 @@ static int nfs_page_mark_flush(struct pa
 	spin_unlock(req_lock);
 	if (test_and_set_bit(PG_FLUSHING, &req->wb_flags) == 0) {
 		nfs_mark_request_dirty(req);
-		set_page_writeback(page);
+		nfs_set_page_writeback(page);
 	}
 	ret = test_bit(PG_NEED_FLUSH, &req->wb_flags);
 	nfs_unlock_request(req);

@@ -336,13 +419,8 @@ int nfs_writepage(struct page *page, str
 	return err; 
 }
 
-/*
- * Note: causes nfs_update_request() to block on the assumption
- * 	 that the writeback is generated due to memory pressure.
- */
 int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
 {
-	struct backing_dev_info *bdi = mapping->backing_dev_info;
 	struct inode *inode = mapping->host;
 	int err;
 

@@ -351,11 +429,6 @@ int nfs_writepages(struct address_space 
 	err = generic_writepages(mapping, wbc);
 	if (err)
 		return err;
-	while (test_and_set_bit(BDI_write_congested, &bdi->state) != 0) {
-		if (wbc->nonblocking)
-			return 0;
-		nfs_wait_on_write_congestion(mapping, 0);
-	}
 	err = nfs_flush_mapping(mapping, wbc, wb_priority(wbc));
 	if (err < 0)
 		goto out;

@@ -369,9 +442,6 @@ int nfs_writepages(struct address_space 
 	if (err > 0)
 		err = 0;
 out:
-	clear_bit(BDI_write_congested, &bdi->state);
-	wake_up_all(&nfs_write_congestion);
-	congestion_end(WRITE);
 	return err;
 }
 

@@ -401,7 +471,7 @@ static int nfs_inode_add_request(struct 
 }
 
 /*
- * Insert a write request into an inode
+ * Remove a write request from an inode
  */
 static void nfs_inode_remove_request(struct nfs_page *req)
 {

@@ -473,7 +543,7 @@ nfs_mark_request_commit(struct nfs_page 
 	nfs_list_add_request(req, &nfsi->commit);
 	nfsi->ncommit++;
 	spin_unlock(&nfsi->req_lock);
-	inc_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
+	nfs_set_page_commit(req->wb_page);
 	__mark_inode_dirty(inode, I_DIRTY_DATASYNC);
 }
 #endif

@@ -544,7 +614,7 @@ static void nfs_cancel_commit_list(struc
 
 	while(!list_empty(head)) {
 		req = nfs_list_entry(head->next);
-		dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
+		nfs_end_page_commit(req->wb_page);
 		nfs_list_remove_request(req);
 		nfs_inode_remove_request(req);
 		nfs_unlock_request(req);

@@ -563,13 +633,14 @@ static void nfs_cancel_commit_list(struc
  * The requests are *not* checked to ensure that they form a contiguous set.
  */
 static int
-nfs_scan_commit(struct inode *inode, struct list_head *dst, unsigned long idx_start, unsigned int npages)
+nfs_scan_commit(struct inode *inode, struct list_head *dst,
+		unsigned long idx_start, unsigned int npages, unsigned int maxpages)
 {
 	struct nfs_inode *nfsi = NFS_I(inode);
 	int res = 0;
 
 	if (nfsi->ncommit != 0) {
-		res = nfs_scan_list(nfsi, &nfsi->commit, dst, idx_start, npages);
+		res = nfs_scan_list(nfsi, &nfsi->commit, dst, idx_start, npages, maxpages);
 		nfsi->ncommit -= res;
 		if ((nfsi->ncommit == 0) != list_empty(&nfsi->commit))
 			printk(KERN_ERR "NFS: desynchronized value of nfs_i.ncommit.\n");

@@ -577,48 +648,93 @@ nfs_scan_commit(struct inode *inode, str
 	return res;
 }
 #else
-static inline int nfs_scan_commit(struct inode *inode, struct list_head *dst, unsigned long idx_start, unsigned int npages)
+static inline int nfs_scan_commit(struct inode *inode, struct list_head *dst,
+		unsigned long idx_start, unsigned int npages, unsigned int maxpages)
 {
 	return 0;
 }
 #endif
 
+static int nfs_do_commit(struct super_block *sb, int npages);
+
+static int nfs_commit_congestion(struct inode *inode)
+{
+	struct nfs_server *nfss = NFS_SERVER(inode);
+	int thresh, excess_commit;
+	int ret = 0;
+
+	if (!commit_pressure(inode))
+		goto out;
+
+	thresh = atomic_read(&nfss->writeback);
+	if (thresh < 128)
+		thresh = 128;
+	if (thresh > NFS_CONGESTION_OFF_THRESH)
+		thresh = NFS_CONGESTION_OFF_THRESH;
+
+	excess_commit = atomic_read(&nfss->commit) - thresh;
+
+	/*
+	 * limit the flush to stable storage - this should
+	 * avoid flushing most of the unstable pages when there
+	 * is not much real pressure but they just tripped the
+	 * dirty page boundary.
+	 */
+	thresh /= 2;
+	if (excess_commit > thresh)
+		excess_commit = thresh;
+
+	while (excess_commit > 0 && commit_pressure(inode)) {
+		int npages = min(16, excess_commit);
+		npages = nfs_do_commit(inode->i_sb, npages);
+		if (npages <= 0)
+			break;
+		excess_commit -= npages;
+		ret = 1;
+		cond_resched();
+	}
+out:
+	return ret;
+}
+
 static int nfs_wait_on_write_congestion(struct address_space *mapping, int intr)
 {
+	struct inode *inode = mapping->host;
 	struct backing_dev_info *bdi = mapping->backing_dev_info;
-	DEFINE_WAIT(wait);
 	int ret = 0;
 
 	might_sleep();
 
-	if (!bdi_write_congested(bdi))
+	if (!bdi_write_congested(bdi)) {
+		nfs_commit_congestion(inode);
 		return 0;
+	}
 
-	nfs_inc_stats(mapping->host, NFSIOS_CONGESTIONWAIT);
+	nfs_inc_stats(inode, NFSIOS_CONGESTIONWAIT);
 
-	if (intr) {
-		struct rpc_clnt *clnt = NFS_CLIENT(mapping->host);
-		sigset_t oldset;
-
-		rpc_clnt_sigmask(clnt, &oldset);
-		prepare_to_wait(&nfs_write_congestion, &wait, TASK_INTERRUPTIBLE);
-		if (bdi_write_congested(bdi)) {
-			if (signalled())
-				ret = -ERESTARTSYS;
-			else
-				schedule();
+	do {
+		if (nfs_commit_congestion(inode) &&
+				!bdi_write_congested(bdi))
+			break;
+
+		if (intr) {
+			struct rpc_clnt *clnt = NFS_CLIENT(inode);
+			sigset_t oldset;
+
+			rpc_clnt_sigmask(clnt, &oldset);
+			ret = congestion_wait_interruptible(WRITE, HZ/10);
+			rpc_clnt_sigunmask(clnt, &oldset);
+			if (ret == -ERESTARTSYS)
+				return ret;
+			ret = 0;
+		} else {
+			congestion_wait(WRITE, HZ/10);
 		}
-		rpc_clnt_sigunmask(clnt, &oldset);
-	} else {
-		prepare_to_wait(&nfs_write_congestion, &wait, TASK_UNINTERRUPTIBLE);
-		if (bdi_write_congested(bdi))
-			schedule();
-	}
-	finish_wait(&nfs_write_congestion, &wait);
+	} while (bdi_write_congested(bdi));
+
 	return ret;
 }
 
-
 /*
  * Try to update any existing write request, or create one if there is none.
  * In order to match, the request's credentials must match those of

@@ -779,7 +895,7 @@ int nfs_updatepage(struct file *file, st
 
 static void nfs_writepage_release(struct nfs_page *req)
 {
-	end_page_writeback(req->wb_page);
+	nfs_end_page_writeback(req->wb_page);
 
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
 	if (!PageError(req->wb_page)) {

@@ -1095,12 +1211,12 @@ static void nfs_writeback_done_full(stru
 			ClearPageUptodate(page);
 			SetPageError(page);
 			req->wb_context->error = task->tk_status;
-			end_page_writeback(page);
+			nfs_end_page_writeback(page);
 			nfs_inode_remove_request(req);
 			dprintk(", error = %d\n", task->tk_status);
 			goto next;
 		}
-		end_page_writeback(page);
+		nfs_end_page_writeback(page);
 
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
 		if (data->args.stable != NFS_UNSTABLE || data->verf.committed == NFS_FILE_SYNC) {

@@ -1277,7 +1393,7 @@ nfs_commit_list(struct inode *inode, str
 		req = nfs_list_entry(head->next);
 		nfs_list_remove_request(req);
 		nfs_mark_request_commit(req);
-		dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
+		nfs_end_page_commit(req->wb_page);
 		nfs_clear_page_writeback(req);
 	}
 	return -ENOMEM;

@@ -1301,7 +1417,7 @@ static void nfs_commit_done(struct rpc_t
 	while (!list_empty(&data->pages)) {
 		req = nfs_list_entry(data->pages.next);
 		nfs_list_remove_request(req);
-		dec_zone_page_state(req->wb_page, NR_UNSTABLE_NFS);
+		nfs_end_page_commit(req->wb_page);
 
 		dprintk("NFS: commit (%s/%Ld %d@%Ld)",
 			req->wb_context->dentry->d_inode->i_sb->s_id,

@@ -1367,7 +1483,7 @@ int nfs_commit_inode(struct inode *inode
 	int res;
 
 	spin_lock(&nfsi->req_lock);
-	res = nfs_scan_commit(inode, &head, 0, 0);
+	res = nfs_scan_commit(inode, &head, 0, 0, 0);
 	spin_unlock(&nfsi->req_lock);
 	if (res) {
 		int error = nfs_commit_list(inode, &head, how);

@@ -1378,6 +1494,43 @@ int nfs_commit_inode(struct inode *inode
 }
 #endif
 
+static int nfs_do_commit(struct super_block *sb, int npages)
+{
+	struct inode *inode;
+	int ret = 0;
+
+	down_read(&sb->s_umount);
+	spin_lock(&inode_lock);
+	list_for_each_entry(inode, &sb->s_inodes, i_sb_list) {
+		struct nfs_inode *nsfi = NFS_I(inode);
+		LIST_HEAD(head);
+		int res;
+
+		if (inode->i_state & (I_FREEING|I_WILL_FREE))
+			continue;
+
+		spin_lock(&nsfi->req_lock);
+		res = nfs_scan_commit(inode, &head, 0, 0, npages);
+		spin_unlock(&nsfi->req_lock);
+		ret += res;
+		npages -= res;
+		if (res) {
+			res = nfs_commit_list(inode, &head, FLUSH_HIGHPRI);
+			if (res < 0) {
+				ret = res;
+				break;
+			}
+		}
+
+		if (npages <= 0)
+			break;
+	}
+	spin_unlock(&inode_lock);
+	up_read(&sb->s_umount);
+
+	return ret;
+}
+
 long nfs_sync_mapping_wait(struct address_space *mapping, struct writeback_control *wbc, int how)
 {
 	struct inode *inode = mapping->host;

@@ -1424,7 +1577,7 @@ long nfs_sync_mapping_wait(struct addres
 			continue;
 		if (nocommit)
 			break;
-		pages = nfs_scan_commit(inode, &head, idx_start, npages);
+		pages = nfs_scan_commit(inode, &head, idx_start, npages, 0);
 		if (pages == 0) {
 			if (wbc->pages_skipped != 0)
 				continue;

@@ -1437,7 +1590,7 @@ long nfs_sync_mapping_wait(struct addres
 			spin_lock(&nfsi->req_lock);
 			continue;
 		}
-		pages += nfs_scan_commit(inode, &head, 0, 0);
+		pages += nfs_scan_commit(inode, &head, 0, 0, 0);
 		spin_unlock(&nfsi->req_lock);
 		ret = nfs_commit_list(inode, &head, how);
 		spin_lock(&nfsi->req_lock);

Index: linux-2.6-git/mm/backing-dev.c
===================================================================
@@ -55,6 +55,22 @@ long congestion_wait(int rw, long timeou
 }
 EXPORT_SYMBOL(congestion_wait);
 
+long congestion_wait_interruptible(int rw, long timeout)
+{
+	long ret;
+	DEFINE_WAIT(wait);
+	wait_queue_head_t *wqh = &congestion_wqh[rw];
+
+	prepare_to_wait(wqh, &wait, TASK_INTERRUPTIBLE);
+	if (signal_pending(current))
+		ret = -ERESTARTSYS;
+	else
+		ret = io_schedule_timeout(timeout);
+	finish_wait(wqh, &wait);
+	return ret;
+}
+EXPORT_SYMBOL(congestion_wait_interruptible);
+
 /**
  * congestion_end - wake up sleepers on a congested backing_dev_info
  * @rw: READ or WRITE

Index: linux-2.6-git/include/linux/nfs_fs_sb.h
===================================================================
@@ -82,6 +82,8 @@ struct nfs_server {
 	struct rpc_clnt *	client_acl;	/* ACL RPC client handle */
 	struct nfs_iostats *	io_stats;	/* I/O statistics */
 	struct backing_dev_info	backing_dev_info;
+	atomic_t		writeback;	/* number of writeback pages */
+	atomic_t		commit;		/* number of commit pages */
 	int			flags;		/* various flags */
 	unsigned int		caps;		/* server capabilities */
 	unsigned int		rsize;		/* read size */

Index: linux-2.6-git/include/linux/backing-dev.h
===================================================================
@@ -93,6 +93,7 @@ static inline int bdi_rw_congested(struc
 void clear_bdi_congested(struct backing_dev_info *bdi, int rw);
 void set_bdi_congested(struct backing_dev_info *bdi, int rw);
 long congestion_wait(int rw, long timeout);
+long congestion_wait_interruptible(int rw, long timeout);
 void congestion_end(int rw);
 
 #define bdi_cap_writeback_dirty(bdi) \

Index: linux-2.6-git/include/linux/writeback.h
===================================================================
@@ -100,6 +100,7 @@ int dirty_writeback_centisecs_handler(st
 				      void __user *, size_t *, loff_t *);
 
 void page_writeback_init(void);
+int dirty_pages_exceeded(struct address_space *mapping);
 void balance_dirty_pages_ratelimited_nr(struct address_space *mapping,
 					unsigned long nr_pages_dirtied);
 

Index: linux-2.6-git/mm/page-writeback.c
===================================================================
@@ -167,6 +167,12 @@ get_dirty_limits(long *pbackground, long
 	*pdirty = dirty;
 }
 
+int dirty_pages_exceeded(struct address_space *mapping)
+{
+	return dirty_exceeded;
+}
+EXPORT_SYMBOL_GPL(dirty_pages_exceeded);
+
 /*
  * balance_dirty_pages() must be called by processes which are generating dirty
  * data.  It looks at the number of dirty pages in the machine and will force

Index: linux-2.6-git/fs/nfs/pagelist.c
===================================================================
@@ -356,7 +356,7 @@ out:
  */
 int nfs_scan_list(struct nfs_inode *nfsi, struct list_head *head,
 		struct list_head *dst, unsigned long idx_start,
-		unsigned int npages)
+		unsigned int npages, unsigned int maxpages)
 {
 	struct nfs_page *pgvec[NFS_SCAN_MAXENTRIES];
 	struct nfs_page *req;

@@ -370,6 +370,9 @@ int nfs_scan_list(struct nfs_inode *nfsi
 	else
 		idx_end = idx_start + npages - 1;
 
+	if (maxpages == 0)
+		maxpages = ~0;
+
 	for (;;) {
 		found = radix_tree_gang_lookup(&nfsi->nfs_page_tree,
 				(void **)&pgvec[0], idx_start,

@@ -387,7 +390,10 @@ int nfs_scan_list(struct nfs_inode *nfsi
 				nfs_list_remove_request(req);
 				nfs_list_add_request(req, dst);
 				res++;
+				maxpages--;
 			}
+			if (!maxpages)
+				goto out;
 		}
 
 	}

Index: linux-2.6-git/include/linux/nfs_page.h
===================================================================
@@ -66,7 +66,7 @@ extern	long nfs_scan_dirty(struct addres
 				struct writeback_control *wbc,
 				struct list_head *dst);
 extern	int nfs_scan_list(struct nfs_inode *nfsi, struct list_head *head, struct list_head *dst,
-			  unsigned long idx_start, unsigned int npages);
+			  unsigned long idx_start, unsigned int npages, unsigned int maxpages);
 extern	int nfs_coalesce_requests(struct list_head *, struct list_head *,
 				  unsigned int);
 extern  int nfs_wait_on_request(struct nfs_page *);

Index: linux-2.6-git/fs/inode.c
===================================================================
@@ -81,6 +81,7 @@ static struct hlist_head *inode_hashtabl
  * the i_state of an inode while it is in use..
  */
 DEFINE_SPINLOCK(inode_lock);
+EXPORT_SYMBOL_GPL(inode_lock);
 
 /*
  * iprune_mutex provides exclusion between the kswapd or try_to_free_pages

Index: linux-2.6-git/fs/nfs/write.c
===================================================================
@@ -89,8 +89,6 @@ static struct kmem_cache *nfs_wdata_cach
 static mempool_t *nfs_wdata_mempool;
 static mempool_t *nfs_commit_mempool;
 
-static DECLARE_WAIT_QUEUE_HEAD(nfs_write_congestion);
-
 struct nfs_write_data *nfs_commit_alloc(void)
 {
 	struct nfs_write_data *p = mempool_alloc(nfs_commit_mempool, GFP_NOFS);

@@ -245,6 +243,39 @@ static int wb_priority(struct writeback_
 }
 
 /*
+ * NFS congestion control
+ */
+
+static unsigned long nfs_congestion_pages;
+
+#define NFS_CONGESTION_ON_THRESH 	nfs_congestion_pages
+#define NFS_CONGESTION_OFF_THRESH	\
+	(NFS_CONGESTION_ON_THRESH - (NFS_CONGESTION_ON_THRESH >> 2))
+
+static inline void nfs_set_page_writeback(struct page *page)
+{
+	if (!test_set_page_writeback(page)) {
+		struct inode *inode = page->mapping->host;
+		struct nfs_server *nfss = NFS_SERVER(inode);
+
+		if (atomic_inc_return(&nfss->writeback) > NFS_CONGESTION_ON_THRESH)
+			set_bdi_congested(&nfss->backing_dev_info, WRITE);
+	}
+}
+
+static inline void nfs_end_page_writeback(struct page *page)
+{
+	struct inode *inode = page->mapping->host;
+	struct nfs_server *nfss = NFS_SERVER(inode);
+
+	end_page_writeback(page);
+	if (atomic_dec_return(&nfss->writeback) < NFS_CONGESTION_OFF_THRESH) {
+		clear_bdi_congested(&nfss->backing_dev_info, WRITE);
+		congestion_end(WRITE);
+	}
+}
+
+/*
  * Find an associated nfs write request, and prepare to flush it out
  * Returns 1 if there was no write request, or if the request was
  * already tagged by nfs_set_page_dirty.Returns 0 if the request

@@ -281,7 +312,7 @@ static int nfs_page_mark_flush(struct pa
 	spin_unlock(req_lock);
 	if (test_and_set_bit(PG_FLUSHING, &req->wb_flags) == 0) {
 		nfs_mark_request_dirty(req);
-		set_page_writeback(page);
+		nfs_set_page_writeback(page);
 	}
 	ret = test_bit(PG_NEED_FLUSH, &req->wb_flags);
 	nfs_unlock_request(req);

@@ -336,13 +367,8 @@ int nfs_writepage(struct page *page, str
 	return err; 
 }
 
-/*
- * Note: causes nfs_update_request() to block on the assumption
- * 	 that the writeback is generated due to memory pressure.
- */
 int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
 {
-	struct backing_dev_info *bdi = mapping->backing_dev_info;
 	struct inode *inode = mapping->host;
 	int err;
 

@@ -351,11 +377,6 @@ int nfs_writepages(struct address_space 
 	err = generic_writepages(mapping, wbc);
 	if (err)
 		return err;
-	while (test_and_set_bit(BDI_write_congested, &bdi->state) != 0) {
-		if (wbc->nonblocking)
-			return 0;
-		nfs_wait_on_write_congestion(mapping, 0);
-	}
 	err = nfs_flush_mapping(mapping, wbc, wb_priority(wbc));
 	if (err < 0)
 		goto out;

@@ -369,9 +390,6 @@ int nfs_writepages(struct address_space 
 	if (err > 0)
 		err = 0;
 out:
-	clear_bit(BDI_write_congested, &bdi->state);
-	wake_up_all(&nfs_write_congestion);
-	congestion_end(WRITE);
 	return err;
 }
 

@@ -401,7 +419,7 @@ static int nfs_inode_add_request(struct 
 }
 
 /*
- * Insert a write request into an inode
+ * Remove a write request from an inode
  */
 static void nfs_inode_remove_request(struct nfs_page *req)
 {

@@ -585,8 +603,8 @@ static inline int nfs_scan_commit(struct
 
 static int nfs_wait_on_write_congestion(struct address_space *mapping, int intr)
 {
+	struct inode *inode = mapping->host;
 	struct backing_dev_info *bdi = mapping->backing_dev_info;
-	DEFINE_WAIT(wait);
 	int ret = 0;
 
 	might_sleep();

@@ -594,31 +612,27 @@ static int nfs_wait_on_write_congestion(
 	if (!bdi_write_congested(bdi))
 		return 0;
 
-	nfs_inc_stats(mapping->host, NFSIOS_CONGESTIONWAIT);
+	nfs_inc_stats(inode, NFSIOS_CONGESTIONWAIT);
 
-	if (intr) {
-		struct rpc_clnt *clnt = NFS_CLIENT(mapping->host);
-		sigset_t oldset;
-
-		rpc_clnt_sigmask(clnt, &oldset);
-		prepare_to_wait(&nfs_write_congestion, &wait, TASK_INTERRUPTIBLE);
-		if (bdi_write_congested(bdi)) {
-			if (signalled())
-				ret = -ERESTARTSYS;
-			else
-				schedule();
+	do {
+		if (intr) {
+			struct rpc_clnt *clnt = NFS_CLIENT(inode);
+			sigset_t oldset;
+
+			rpc_clnt_sigmask(clnt, &oldset);
+			ret = congestion_wait_interruptible(WRITE, HZ/10);
+			rpc_clnt_sigunmask(clnt, &oldset);
+			if (ret == -ERESTARTSYS)
+				return ret;
+			ret = 0;
+		} else {
+			congestion_wait(WRITE, HZ/10);
 		}
-		rpc_clnt_sigunmask(clnt, &oldset);
-	} else {
-		prepare_to_wait(&nfs_write_congestion, &wait, TASK_UNINTERRUPTIBLE);
-		if (bdi_write_congested(bdi))
-			schedule();
-	}
-	finish_wait(&nfs_write_congestion, &wait);
+	} while (bdi_write_congested(bdi));
+
 	return ret;
 }
 
-
 /*
  * Try to update any existing write request, or create one if there is none.
  * In order to match, the request's credentials must match those of

@@ -779,7 +793,7 @@ int nfs_updatepage(struct file *file, st
 
 static void nfs_writepage_release(struct nfs_page *req)
 {
-	end_page_writeback(req->wb_page);
+	nfs_end_page_writeback(req->wb_page);
 
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
 	if (!PageError(req->wb_page)) {

@@ -1095,12 +1109,12 @@ static void nfs_writeback_done_full(stru
 			ClearPageUptodate(page);
 			SetPageError(page);
 			req->wb_context->error = task->tk_status;
-			end_page_writeback(page);
+			nfs_end_page_writeback(page);
 			nfs_inode_remove_request(req);
 			dprintk(", error = %d\n", task->tk_status);
 			goto next;
 		}
-		end_page_writeback(page);
+		nfs_end_page_writeback(page);
 
 #if defined(CONFIG_NFS_V3) || defined(CONFIG_NFS_V4)
 		if (data->args.stable != NFS_UNSTABLE || data->verf.committed == NFS_FILE_SYNC) {

@@ -1565,6 +1579,23 @@ int __init nfs_init_writepagecache(void)
 	if (nfs_commit_mempool == NULL)
 		return -ENOMEM;
 
+	/*
+	 * NFS congestion size, scale with available memory.
+	 *
+	 *  64MB:    8192k
+	 * 128MB:   11585k
+	 * 256MB:   16384k
+	 * 512MB:   23170k
+	 *   1GB:   32768k
+	 *   2GB:   46340k
+	 *   4GB:   65536k
+	 *   8GB:   92681k
+	 *  16GB:  131072k
+	 *
+	 * This allows larger machines to have larger/more transfers.
+	 */
+	nfs_congestion_size = 32*int_sqrt(totalram_pages);
+
 	return 0;
 }
 

Index: linux-2.6-git/mm/backing-dev.c
===================================================================
@@ -55,6 +55,22 @@ long congestion_wait(int rw, long timeou
 }
 EXPORT_SYMBOL(congestion_wait);
 
+long congestion_wait_interruptible(int rw, long timeout)
+{
+	long ret;
+	DEFINE_WAIT(wait);
+	wait_queue_head_t *wqh = &congestion_wqh[rw];
+
+	prepare_to_wait(wqh, &wait, TASK_INTERRUPTIBLE);
+	if (signal_pending(current))
+		ret = -ERESTARTSYS;
+	else
+		ret = io_schedule_timeout(timeout);
+	finish_wait(wqh, &wait);
+	return ret;
+}
+EXPORT_SYMBOL(congestion_wait_interruptible);
+
 /**
  * congestion_end - wake up sleepers on a congested backing_dev_info
  * @rw: READ or WRITE

Index: linux-2.6-git/include/linux/nfs_fs_sb.h
===================================================================
@@ -82,6 +82,7 @@ struct nfs_server {
 	struct rpc_clnt *	client_acl;	/* ACL RPC client handle */
 	struct nfs_iostats *	io_stats;	/* I/O statistics */
 	struct backing_dev_info	backing_dev_info;
+	atomic_t		writeback;	/* number of writeback pages */
 	int			flags;		/* various flags */
 	unsigned int		caps;		/* server capabilities */
 	unsigned int		rsize;		/* read size */

Index: linux-2.6-git/include/linux/backing-dev.h
===================================================================
@@ -93,6 +93,7 @@ static inline int bdi_rw_congested(struc
 void clear_bdi_congested(struct backing_dev_info *bdi, int rw);
 void set_bdi_congested(struct backing_dev_info *bdi, int rw);
 long congestion_wait(int rw, long timeout);
+long congestion_wait_interruptible(int rw, long timeout);
 void congestion_end(int rw);
 
 #define bdi_cap_writeback_dirty(bdi) \

Index: linux-2.6-git/fs/nfs/write.c
===================================================================
@@ -89,8 +89,6 @@ static struct kmem_cache *nfs_wdata_cach
 static mempool_t *nfs_wdata_mempool;
 static mempool_t *nfs_commit_mempool;
 
-static DECLARE_WAIT_QUEUE_HEAD(nfs_write_congestion);
-
 struct nfs_write_data *nfs_commit_alloc(void)
 {
 	struct nfs_write_data *p = mempool_alloc(nfs_commit_mempool, GFP_NOFS);

@@ -245,6 +243,39 @@ static int wb_priority(struct writeback_
 }
 
 /*
+ * NFS congestion control
+ */
+
+int nfs_congestion_kb;
+
+#define NFS_CONGESTION_ON_THRESH 	(nfs_congestion_kb >> (PAGE_SHIFT-10))
+#define NFS_CONGESTION_OFF_THRESH	\
+	(NFS_CONGESTION_ON_THRESH - (NFS_CONGESTION_ON_THRESH >> 2))
+
+static inline void nfs_set_page_writeback(struct page *page)
+{
+	if (!test_set_page_writeback(page)) {
+		struct inode *inode = page->mapping->host;
+		struct nfs_server *nfss = NFS_SERVER(inode);
+
+		if (atomic_inc_return(&nfss->writeback) > NFS_CONGESTION_ON_THRESH)
+			set_bdi_congested(&nfss->backing_dev_info, WRITE);
+	}
+}
+
+static inline void nfs_end_page_writeback(struct page *page)
+{
+	struct inode *inode = page->mapping->host;
+	struct nfs_server *nfss = NFS_SERVER(inode);
+
+	end_page_writeback(page);
+	if (atomic_dec_return(&nfss->writeback) < NFS_CONGESTION_OFF_THRESH) {
+		clear_bdi_congested(&nfss->backing_dev_info, WRITE);
+		congestion_end(WRITE);
+	}
+}
+
+/*
  * Find an associated nfs write request, and prepare to flush it out
  * Returns 1 if there was no write request, or if the request was
  * already tagged by nfs_set_page_dirty.Returns 0 if the request

@@ -281,7 +312,7 @@ static int nfs_page_mark_flush(struct pa
 	spin_unlock(req_lock);
 	if (test_and_set_bit(PG_FLUSHING, &req->wb_flags) == 0) {
 		nfs_mark_request_dirty(req);
-		set_page_writeback(page);
+		nfs_set_page_writeback(page);
 	}
 	ret = test_bit(PG_NEED_FLUSH, &req->wb_flags);
 	nfs_unlock_request(req);

@@ -336,13 +367,8 @@ int nfs_writepage(struct page *page, str
 	return err; 
 }
 
-/*
- * Note: causes nfs_update_request() to block on the assumption
- * 	 that the writeback is generated due to memory pressure.
- */
 int nfs_writepages(struct address_space *mapping, struct writeback_control *wbc)
 {
-	struct backing_dev_info *bdi = mapping->backing_dev_info;
 	struct inode *inode = mapping->host;
 	int err;
 

@@ -351,11 +377,6 @@ int nfs_writepages(struct address_space 
 	err = generic_writepages(mapping, wbc);
 	if (err)
 		return err;
-	while (test_and_set_bit(BDI_write_congested, &bdi->state) != 0) {
-		if (wbc->nonblocking)
-			return 0;
-		nfs_wait_on_write_congestion(mapping, 0);
-	}
 	err = nfs_flush_mapping(mapping, wbc, wb_priority(wbc));
 	if (err < 0)
 		goto out;