Merge mandrakesoft.com:/home/jgarzik/repo/linus-2.5

into mandrakesoft.com:/home/jgarzik/repo/net-drivers-2.5

8 files changed, 71 insertions, 43 deletions

diff --git a/include/linux/fs.h b/include/linux/fs.h
index 804ea47301f5..56f2bab87d7f 100644
--- a/include/linux/fs.h
+++ b/include/linux/fs.h
@@ -279,6 +279,7 @@ struct iattr {
  */
 struct page;
 struct address_space;
+struct writeback_control;
 
 struct address_space_operations {
 	int (*writepage)(struct page *);
@@ -286,10 +287,10 @@ struct address_space_operations {
 	int (*sync_page)(struct page *);
 
 	/* Write back some dirty pages from this mapping. */
-	int (*writepages)(struct address_space *, int *nr_to_write);
+	int (*writepages)(struct address_space *, struct writeback_control *);
 
 	/* Perform a writeback as a memory-freeing operation. */
-	int (*vm_writeback)(struct page *, int *nr_to_write);
+	int (*vm_writeback)(struct page *, struct writeback_control *);
 
 	/* Set a page dirty */
 	int (*set_page_dirty)(struct page *page);
@@ -1259,7 +1260,8 @@ extern loff_t generic_file_llseek(struct file *file, loff_t offset, int origin);
 extern loff_t remote_llseek(struct file *file, loff_t offset, int origin);
 extern int generic_file_open(struct inode * inode, struct file * filp);
 
-extern int generic_vm_writeback(struct page *page, int *nr_to_write);
+extern int generic_vm_writeback(struct page *page,
+				struct writeback_control *wbc);
 
 extern struct file_operations generic_ro_fops;
 
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 10021357c093..437572e2240b 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -39,18 +39,25 @@
  * can allocate highmem pages, the *get*page*() variants return
  * virtual kernel addresses to the allocated page(s).
  */
-extern struct page * FASTCALL(_alloc_pages(unsigned int gfp_mask, unsigned int order));
 extern struct page * FASTCALL(__alloc_pages(unsigned int gfp_mask, unsigned int order, struct zonelist *zonelist));
 extern struct page * alloc_pages_node(int nid, unsigned int gfp_mask, unsigned int order);
 
+/*
+ * We get the zone list from the current node and the gfp_mask.
+ * This zone list contains a maximum of MAXNODES*MAX_NR_ZONES zones.
+ *
+ * For the normal case of non-DISCONTIGMEM systems the NODE_DATA() gets
+ * optimized to &contig_page_data at compile-time.
+ */
 static inline struct page * alloc_pages(unsigned int gfp_mask, unsigned int order)
 {
-	/*
-	 * Gets optimized away by the compiler.
-	 */
-	if (order >= MAX_ORDER)
+	pg_data_t *pgdat = NODE_DATA(numa_node_id());
+	unsigned int idx = (gfp_mask & GFP_ZONEMASK);
+
+	if (unlikely(order >= MAX_ORDER))
 		return NULL;
-	return _alloc_pages(gfp_mask, order);
+
+	return __alloc_pages(gfp_mask, order, pgdat->node_zonelists + idx);
 }
 
 #define alloc_page(gfp_mask) alloc_pages(gfp_mask, 0)
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 7483c39e28dd..c63e4947387f 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -15,7 +15,10 @@
 #include <linux/rbtree.h>
 #include <linux/fs.h>
 
+#ifndef CONFIG_DISCONTIGMEM          /* Don't use mapnrs, do it properly */
 extern unsigned long max_mapnr;
+#endif
+
 extern unsigned long num_physpages;
 extern void * high_memory;
 extern int page_cluster;
@@ -345,8 +348,10 @@ static inline int page_mapped(struct page *page)
 #define VM_FAULT_MINOR	1
 #define VM_FAULT_MAJOR	2
 
-/* The array of struct pages */
+#ifndef CONFIG_DISCONTIGMEM
+/* The array of struct pages - for discontigmem use pgdat->lmem_map */
 extern struct page *mem_map;
+#endif 
 
 extern void show_free_areas(void);
 
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 8ebf441bdb47..580c39c4dcc1 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -10,11 +10,14 @@
 #include <linux/wait.h>
 #include <linux/cache.h>
 #include <asm/atomic.h>
+#ifdef CONFIG_DISCONTIGMEM
+#include <asm/numnodes.h>
+#endif
+#ifndef MAX_NUMNODES
+#define MAX_NUMNODES 1
+#endif
 
-/*
- * Free memory management - zoned buddy allocator.
- */
-
+/* Free memory management - zoned buddy allocator.  */
 #ifndef CONFIG_FORCE_MAX_ZONEORDER
 #define MAX_ORDER 11
 #else
@@ -112,7 +115,6 @@ struct zone {
 	struct page		*zone_mem_map;
 	/* zone_start_pfn == zone_start_paddr >> PAGE_SHIFT */
 	unsigned long		zone_start_pfn;
-	unsigned long		zone_start_mapnr;
 
 	/*
 	 * rarely used fields:
@@ -138,7 +140,7 @@ struct zone {
  * footprint of this construct is very small.
  */
 struct zonelist {
-	struct zone *zones[MAX_NR_ZONES+1]; // NULL delimited
+	struct zone *zones[MAX_NUMNODES * MAX_NR_ZONES + 1]; // NULL delimited
 };
 
 #define GFP_ZONEMASK	0x0f
@@ -163,7 +165,6 @@ typedef struct pglist_data {
 	unsigned long *valid_addr_bitmap;
 	struct bootmem_data *bdata;
 	unsigned long node_start_pfn;
-	unsigned long node_start_mapnr;
 	unsigned long node_size;
 	int node_id;
 	struct pglist_data *pgdat_next;
@@ -187,10 +188,12 @@ memclass(struct zone *pgzone, struct zone *classzone)
  * prototypes for the discontig memory code.
  */
 struct page;
-void free_area_init_core(int nid, pg_data_t *pgdat, struct page **gmap,
-  unsigned long *zones_size, unsigned long paddr, unsigned long *zholes_size,
-  struct page *pmap);
+extern void calculate_totalpages (pg_data_t *pgdat, unsigned long *zones_size,
+		unsigned long *zholes_size);
+extern void free_area_init_core(pg_data_t *pgdat, unsigned long *zones_size,
+		unsigned long *zholes_size);
 void get_zone_counts(unsigned long *active, unsigned long *inactive);
+extern void build_all_zonelists(void);
 
 extern pg_data_t contig_page_data;
 
diff --git a/include/linux/mpage.h b/include/linux/mpage.h
index 52253d90f55d..86aa7b676274 100644
--- a/include/linux/mpage.h
+++ b/include/linux/mpage.h
@@ -10,14 +10,16 @@
  * nested includes.  Get it right in the .c file).
  */
 
+struct writeback_control;
+
 int mpage_readpages(struct address_space *mapping, struct list_head *pages,
 				unsigned nr_pages, get_block_t get_block);
 int mpage_readpage(struct page *page, get_block_t get_block);
 int mpage_writepages(struct address_space *mapping,
-		int *nr_to_write, get_block_t get_block);
+		struct writeback_control *wbc, get_block_t get_block);
 
 static inline int
-generic_writepages(struct address_space *mapping, int *nr_to_write)
+generic_writepages(struct address_space *mapping, struct writeback_control *wbc)
 {
-	return mpage_writepages(mapping, nr_to_write, NULL);
+	return mpage_writepages(mapping, wbc, NULL);
 }
diff --git a/include/linux/sysctl.h b/include/linux/sysctl.h
index 3127165e7c13..9fd7d5c05605 100644
--- a/include/linux/sysctl.h
+++ b/include/linux/sysctl.h
@@ -128,7 +128,6 @@ enum
 	KERN_TAINTED=53,	/* int: various kernel tainted flags */
 	KERN_CADPID=54,		/* int: PID of the process to notify on CAD */
 	KERN_PIDMAX=55,		/* int: PID # limit */
-	KERN_HUGETLB_PAGE_NUM=56, /* int: Number of available Huge Pages */
 };
 
 
@@ -147,12 +146,12 @@ enum
 	VM_PAGE_CLUSTER=10,	/* int: set number of pages to swap together */
 	VM_DIRTY_BACKGROUND=11,	/* dirty_background_ratio */
 	VM_DIRTY_ASYNC=12,	/* dirty_async_ratio */
-	VM_DIRTY_SYNC=13,	/* dirty_sync_ratio */
-	VM_DIRTY_WB_CS=14,	/* dirty_writeback_centisecs */
-	VM_DIRTY_EXPIRE_CS=15,	/* dirty_expire_centisecs */
-	VM_NR_PDFLUSH_THREADS=16, /* nr_pdflush_threads */
-	VM_OVERCOMMIT_RATIO=17, /* percent of RAM to allow overcommit in */
-	VM_PAGEBUF=18		/* struct: Control pagebuf parameters */
+	VM_DIRTY_WB_CS=13,	/* dirty_writeback_centisecs */
+	VM_DIRTY_EXPIRE_CS=14,	/* dirty_expire_centisecs */
+	VM_NR_PDFLUSH_THREADS=15, /* nr_pdflush_threads */
+	VM_OVERCOMMIT_RATIO=16, /* percent of RAM to allow overcommit in */
+	VM_PAGEBUF=17,		/* struct: Control pagebuf parameters */
+	VM_HUGETLB_PAGES=18,	/* int: Number of available Huge Pages */
 };
 
 
diff --git a/include/linux/uio.h b/include/linux/uio.h
index ec098c8e6793..85b2f0ec9d3f 100644
--- a/include/linux/uio.h
+++ b/include/linux/uio.h
@@ -35,7 +35,11 @@ struct iovec
 #endif
 
 /*
- * Total number of bytes covered by an iovec
+ * Total number of bytes covered by an iovec.
+ *
+ * NOTE that it is not safe to use this function until all the iovec's
+ * segment lengths have been validated.  Because the individual lengths can
+ * overflow a size_t when added together.
  */
 static inline size_t iov_length(const struct iovec *iov, unsigned long nr_segs)
 {
diff --git a/include/linux/writeback.h b/include/linux/writeback.h
index 5de884cd6a7c..c35b96eb6a90 100644
--- a/include/linux/writeback.h
+++ b/include/linux/writeback.h
@@ -27,22 +27,29 @@ static inline int current_is_pdflush(void)
  * fs/fs-writeback.c
  */
 enum writeback_sync_modes {
-	WB_SYNC_NONE =  0,	/* Don't wait on anything */
-	WB_SYNC_LAST =  1,	/* Wait on the last-written mapping */
-	WB_SYNC_ALL =   2,	/* Wait on every mapping */
-	WB_SYNC_HOLD =  3,	/* Hold the inode on sb_dirty for sys_sync() */
+	WB_SYNC_NONE,	/* Don't wait on anything */
+	WB_SYNC_ALL,	/* Wait on every mapping */
+	WB_SYNC_HOLD,	/* Hold the inode on sb_dirty for sys_sync() */
 };
 
-void writeback_unlocked_inodes(int *nr_to_write,
-			       enum writeback_sync_modes sync_mode,
-			       unsigned long *older_than_this);
+/*
+ * A control structure which tells the writeback code what to do
+ */
+struct writeback_control {
+	struct backing_dev_info *bdi;	/* If !NULL, only write back this
+					   queue */
+	enum writeback_sync_modes sync_mode;
+	unsigned long *older_than_this;	/* If !NULL, only write back inodes
+					   older than this */
+	long nr_to_write;		/* Write this many pages, and decrement
+					   this for each page written */
+};
+	
+void writeback_inodes(struct writeback_control *wbc);
 void wake_up_inode(struct inode *inode);
 void __wait_on_inode(struct inode * inode);
 void sync_inodes_sb(struct super_block *, int wait);
 void sync_inodes(int wait);
-void writeback_backing_dev(struct backing_dev_info *bdi, int *nr_to_write,
-			enum writeback_sync_modes sync_mode,
-			unsigned long *older_than_this);
 
 /* writeback.h requires fs.h; it, too, is not included from here. */
 static inline void wait_on_inode(struct inode *inode)
@@ -57,7 +64,6 @@ static inline void wait_on_inode(struct inode *inode)
 /* These 5 are exported to sysctl. */
 extern int dirty_background_ratio;
 extern int dirty_async_ratio;
-extern int dirty_sync_ratio;
 extern int dirty_writeback_centisecs;
 extern int dirty_expire_centisecs;
 
@@ -65,7 +71,7 @@ extern int dirty_expire_centisecs;
 void balance_dirty_pages(struct address_space *mapping);
 void balance_dirty_pages_ratelimited(struct address_space *mapping);
 int pdflush_operation(void (*fn)(unsigned long), unsigned long arg0);
-int do_writepages(struct address_space *mapping, int *nr_to_write);
+int do_writepages(struct address_space *mapping, struct writeback_control *wbc);
 
 /* pdflush.c */
 extern int nr_pdflush_threads;	/* Global so it can be exported to sysctl