Merge bk://kernel.bkbits.net/davem/sparc-2.5

into penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/linux

drivers/char/tty_io.c

@@ -1836,6 +1836,8 @@ static void __do_SAK(void *arg)
 #else
 	struct tty_struct *tty = arg;
 	struct task_struct *p;
+	struct list_head *l;
+	struct pid *pid;
 	int session;
 	int		i;
 	struct file	*filp;
@@ -1848,9 +1850,8 @@ static void __do_SAK(void *arg)
 	if (tty->driver.flush_buffer)
 		tty->driver.flush_buffer(tty);
 	read_lock(&tasklist_lock);
-	for_each_process(p) {
-		if ((p->tty == tty) ||
-		    ((session > 0) && (p->session == session))) {
+	for_each_task_pid(session, PIDTYPE_SID, p, l, pid) {
+		if (p->tty == tty || session > 0) {
 			printk(KERN_NOTICE "SAK: killed process %d"
 			    " (%s): p->session==tty->session\n",
 			    p->pid, p->comm);

fs/ext3/ialloc.c

@@ -417,7 +417,7 @@ struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode)
 	struct buffer_head *bitmap_bh = NULL;
 	struct buffer_head *bh2;
 	int group;
-	ino_t ino;
+	unsigned long ino;
 	struct inode * inode;
 	struct ext3_group_desc * gdp;
 	struct ext3_super_block * es;
@@ -619,9 +619,9 @@ struct inode *ext3_new_inode(handle_t *handle, struct inode * dir, int mode)
 }
 
 /* Verify that we are loading a valid orphan from disk */
-struct inode *ext3_orphan_get (struct super_block * sb, ino_t ino)
+struct inode *ext3_orphan_get(struct super_block *sb, unsigned long ino)
 {
-	ino_t max_ino = le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count);
+	unsigned long max_ino = le32_to_cpu(EXT3_SB(sb)->s_es->s_inodes_count);
 	unsigned long block_group;
 	int bit;
 	struct buffer_head *bitmap_bh = NULL;
@@ -630,8 +630,7 @@ struct inode *ext3_orphan_get (struct super_block * sb, ino_t ino)
 	/* Error cases - e2fsck has already cleaned up for us */
 	if (ino > max_ino) {
 		ext3_warning(sb, __FUNCTION__,
-			     "bad orphan ino %lu!  e2fsck was run?\n",
-			     (unsigned long) ino);
+			     "bad orphan ino %lu!  e2fsck was run?\n", ino);
 		goto out;
 	}
 
@@ -640,8 +639,7 @@ struct inode *ext3_orphan_get (struct super_block * sb, ino_t ino)
 	bitmap_bh = read_inode_bitmap(sb, block_group);
 	if (!bitmap_bh) {
 		ext3_warning(sb, __FUNCTION__,
-			     "inode bitmap error for orphan %lu\n",
-			     (unsigned long) ino);
+			     "inode bitmap error for orphan %lu\n", ino);
 		goto out;
 	}
 
@@ -653,19 +651,17 @@ struct inode *ext3_orphan_get (struct super_block * sb, ino_t ino)
 			!(inode = iget(sb, ino)) || is_bad_inode(inode) ||
 			NEXT_ORPHAN(inode) > max_ino) {
 		ext3_warning(sb, __FUNCTION__,
-			     "bad orphan inode %lu!  e2fsck was run?\n", (unsigned long)ino);
+			     "bad orphan inode %lu!  e2fsck was run?\n", ino);
 		printk(KERN_NOTICE "ext3_test_bit(bit=%d, block=%llu) = %d\n",
-		       bit, 
-			(unsigned long long)bitmap_bh->b_blocknr, 
+		       bit, (unsigned long long)bitmap_bh->b_blocknr,
 		       ext3_test_bit(bit, bitmap_bh->b_data));
 		printk(KERN_NOTICE "inode=%p\n", inode);
 		if (inode) {
 			printk(KERN_NOTICE "is_bad_inode(inode)=%d\n",
 			       is_bad_inode(inode));
-			printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%d\n",
+			printk(KERN_NOTICE "NEXT_ORPHAN(inode)=%u\n",
 			       NEXT_ORPHAN(inode));
-			printk(KERN_NOTICE "max_ino=%lu\n",
-			       (unsigned long) max_ino);
+			printk(KERN_NOTICE "max_ino=%lu\n", max_ino);
 		}
 		/* Avoid freeing blocks if we got a bad deleted inode */
 		if (inode && inode->i_nlink == 0)

fs/ext3/inode.c

@@ -1098,20 +1098,14 @@ static int ext3_prepare_write(struct file *file, struct page *page,
 	return ret;
 }
 
-static int journal_dirty_sync_data(handle_t *handle, struct buffer_head *bh)
-{
-	return ext3_journal_dirty_data(handle, bh, 0);
-}
-
-/*
- * For ext3_writepage().  We also brelse() the buffer to account for
- * the bget() which ext3_writepage() performs.
- */
-static int journal_dirty_async_data(handle_t *handle, struct buffer_head *bh)
+static int
+ext3_journal_dirty_data(handle_t *handle, struct buffer_head *bh)
 {
-	int ret = ext3_journal_dirty_data(handle, bh, 1);
-	__brelse(bh);
-	return ret;
+	int err = journal_dirty_data(handle, bh);
+	if (err)
+		ext3_journal_abort_handle(__FUNCTION__, __FUNCTION__,
+						bh, handle,err);
+	return err;
 }
 
 /* For commit_write() in data=journal mode */
@@ -1154,7 +1148,7 @@ static int ext3_commit_write(struct file *file, struct page *page,
 	} else {
 		if (ext3_should_order_data(inode)) {
 			ret = walk_page_buffers(handle, page_buffers(page),
-				from, to, NULL, journal_dirty_sync_data);
+				from, to, NULL, ext3_journal_dirty_data);
 		}
 		/* Be careful here if generic_commit_write becomes a
 		 * required invocation after block_prepare_write. */
@@ -1228,7 +1222,13 @@ static sector_t ext3_bmap(struct address_space *mapping, sector_t block)
 
 static int bget_one(handle_t *handle, struct buffer_head *bh)
 {
-	atomic_inc(&bh->b_count);
+	get_bh(bh);
+	return 0;
+}
+
+static int bput_one(handle_t *handle, struct buffer_head *bh)
+{
+	put_bh(bh);
 	return 0;
 }
 
@@ -1348,7 +1348,9 @@ static int ext3_writepage(struct page *page, struct writeback_control *wbc)
 	/* And attach them to the current transaction */
 	if (order_data) {
 		err = walk_page_buffers(handle, page_bufs,
-			0, PAGE_CACHE_SIZE, NULL, journal_dirty_async_data);
+			0, PAGE_CACHE_SIZE, NULL, ext3_journal_dirty_data);
+		walk_page_buffers(handle, page_bufs, 0,
+				PAGE_CACHE_SIZE, NULL, bput_one);
 		if (!ret)
 			ret = err;
 	}
@@ -1587,7 +1589,7 @@ static int ext3_block_truncate_page(handle_t *handle,
 		err = ext3_journal_dirty_metadata(handle, bh);
 	} else {
 		if (ext3_should_order_data(inode))
-			err = ext3_journal_dirty_data(handle, bh, 0);
+			err = ext3_journal_dirty_data(handle, bh);
 		mark_buffer_dirty(bh);
 	}
 

fs/ext3/namei.c

@@ -1871,7 +1871,7 @@ int ext3_orphan_del(handle_t *handle, struct inode *inode)
 	struct list_head *prev;
 	struct ext3_inode_info *ei = EXT3_I(inode);
 	struct ext3_sb_info *sbi;
-	ino_t ino_next; 
+	unsigned long ino_next;
 	struct ext3_iloc iloc;
 	int err = 0;
 
@@ -1885,7 +1885,7 @@ int ext3_orphan_del(handle_t *handle, struct inode *inode)
 	prev = ei->i_orphan.prev;
 	sbi = EXT3_SB(inode->i_sb);
 
-	jbd_debug(4, "remove inode %ld from orphan list\n", inode->i_ino);
+	jbd_debug(4, "remove inode %lu from orphan list\n", inode->i_ino);
 
 	list_del_init(&ei->i_orphan);
 
@@ -1901,7 +1901,7 @@ int ext3_orphan_del(handle_t *handle, struct inode *inode)
 		goto out_err;
 
 	if (prev == &sbi->s_orphan) {
-		jbd_debug(4, "superblock will point to %ld\n", ino_next);
+		jbd_debug(4, "superblock will point to %lu\n", ino_next);
 		BUFFER_TRACE(sbi->s_sbh, "get_write_access");
 		err = ext3_journal_get_write_access(handle, sbi->s_sbh);
 		if (err)
@@ -1913,7 +1913,7 @@ int ext3_orphan_del(handle_t *handle, struct inode *inode)
 		struct inode *i_prev =
 			&list_entry(prev, struct ext3_inode_info, i_orphan)->vfs_inode;
 
-		jbd_debug(4, "orphan inode %ld will point to %ld\n",
+		jbd_debug(4, "orphan inode %lu will point to %lu\n",
 			  i_prev->i_ino, ino_next);
 		err = ext3_reserve_inode_write(handle, i_prev, &iloc2);
 		if (err)

fs/hugetlbfs/inode.c

@@ -457,11 +457,8 @@ static struct address_space_operations hugetlbfs_aops = {
 };
 
 struct file_operations hugetlbfs_file_operations = {
-	.read		= generic_file_read,
-	.write		= generic_file_write,
 	.mmap		= hugetlbfs_file_mmap,
 	.fsync		= simple_sync_file,
-	.sendfile	= generic_file_sendfile,
 };
 
 static struct inode_operations hugetlbfs_dir_inode_operations = {

fs/jbd/checkpoint.c

@@ -585,7 +585,6 @@ void __journal_drop_transaction(journal_t *journal, transaction_t *transaction)
 	J_ASSERT (transaction->t_ilist == NULL);
 	J_ASSERT (transaction->t_buffers == NULL);
 	J_ASSERT (transaction->t_sync_datalist == NULL);
-	J_ASSERT (transaction->t_async_datalist == NULL);
 	J_ASSERT (transaction->t_forget == NULL);
 	J_ASSERT (transaction->t_iobuf_list == NULL);
 	J_ASSERT (transaction->t_shadow_list == NULL);

fs/jbd/commit.c

@@ -264,37 +264,6 @@ void journal_commit_transaction(journal_t *journal)
 	goto write_out_data_locked;
 
 sync_datalist_empty:
-	/*
-	 * Wait for all the async writepage data.  As they become unlocked
-	 * in end_buffer_async_write(), the only place where they can be
-	 * reaped is in try_to_free_buffers(), and we're locked against
-	 * that.
-	 */
-	while ((jh = commit_transaction->t_async_datalist)) {
-		struct buffer_head *bh = jh2bh(jh);
-		if (buffer_locked(bh)) {
-			spin_unlock(&journal_datalist_lock);
-			unlock_journal(journal);
-			wait_on_buffer(bh);
-			lock_journal(journal);
-			spin_lock(&journal_datalist_lock);
-			continue;	/* List may have changed */
-		}
-		if (jh->b_next_transaction) {
-			/*
-			 * For writepage() buffers in journalled data mode: a
-			 * later transaction may want the buffer for "metadata"
-			 */
-			__journal_refile_buffer(jh);
-		} else {
-			BUFFER_TRACE(bh, "finished async writeout: unfile");
-			__journal_unfile_buffer(jh);
-			jh->b_transaction = NULL;
-			__journal_remove_journal_head(bh);
-			BUFFER_TRACE(bh, "finished async writeout: refile");
-			__brelse(bh);
-		}
-	}
 	spin_unlock(&journal_datalist_lock);
 
 	/*
@@ -304,7 +273,6 @@ void journal_commit_transaction(journal_t *journal)
 	 * clean by now, so check that it is in fact empty.
 	 */
 	J_ASSERT (commit_transaction->t_sync_datalist == NULL);
-	J_ASSERT (commit_transaction->t_async_datalist == NULL);
 
 	jbd_debug (3, "JBD: commit phase 3\n");
 
@@ -629,7 +597,6 @@ void journal_commit_transaction(journal_t *journal)
 	jbd_debug(3, "JBD: commit phase 7\n");
 
 	J_ASSERT(commit_transaction->t_sync_datalist == NULL);
-	J_ASSERT(commit_transaction->t_async_datalist == NULL);
 	J_ASSERT(commit_transaction->t_buffers == NULL);
 	J_ASSERT(commit_transaction->t_checkpoint_list == NULL);
 	J_ASSERT(commit_transaction->t_iobuf_list == NULL);

fs/jbd/transaction.c

@@ -578,9 +578,6 @@ static void jbd_unexpected_dirty_buffer(struct journal_head *jh)
  * part of the transaction, that is).
  *
  * Returns an error code or 0 on success.
- *
- * In full data journalling mode the buffer may be of type BJ_AsyncData,
- * because we're write()ing a buffer which is also part of a shared mapping.
  */
 
 static int
@@ -949,26 +946,16 @@ int journal_get_undo_access (handle_t *handle, struct buffer_head *bh)
  * The buffer is placed on the transaction's data list and is marked as
  * belonging to the transaction.
  *
- * If `async' is set then the writebabk will be initiated by the caller
- * using submit_bh -> end_buffer_async_write.  We put the buffer onto
- * t_async_datalist.
- * 
  * Returns error number or 0 on success.  
  *
  * journal_dirty_data() can be called via page_launder->ext3_writepage
  * by kswapd.  So it cannot block.  Happily, there's nothing here
  * which needs lock_journal if `async' is set.
- *
- * When the buffer is on the current transaction we freely move it
- * between BJ_AsyncData and BJ_SyncData according to who tried to
- * change its state last.
  */
-
-int journal_dirty_data (handle_t *handle, struct buffer_head *bh, int async)
+int journal_dirty_data (handle_t *handle, struct buffer_head *bh)
 {
 	journal_t *journal = handle->h_transaction->t_journal;
 	int need_brelse = 0;
-	int wanted_jlist = async ? BJ_AsyncData : BJ_SyncData;
 	struct journal_head *jh;
 
 	if (is_handle_aborted(handle))
@@ -1046,8 +1033,7 @@ int journal_dirty_data (handle_t *handle, struct buffer_head *bh, int async)
 			 * the write() data.
 			 */
 			if (jh->b_jlist != BJ_None &&
-					jh->b_jlist != BJ_SyncData &&
-					jh->b_jlist != BJ_AsyncData) {
+					jh->b_jlist != BJ_SyncData) {
 				JBUFFER_TRACE(jh, "Not stealing");
 				goto no_journal;
 			}
@@ -1058,7 +1044,7 @@ int journal_dirty_data (handle_t *handle, struct buffer_head *bh, int async)
 			 * again because that can cause the write-out loop in
 			 * commit to never terminate.
 			 */
-			if (!async && buffer_dirty(bh)) {
+			if (buffer_dirty(bh)) {
 				atomic_inc(&bh->b_count);
 				spin_unlock(&journal_datalist_lock);
 				need_brelse = 1;
@@ -1084,18 +1070,18 @@ int journal_dirty_data (handle_t *handle, struct buffer_head *bh, int async)
 		 * committing transaction, so might still be left on that
 		 * transaction's metadata lists.
 		 */
-		if (jh->b_jlist != wanted_jlist) {
+		if (jh->b_jlist != BJ_SyncData) {
 			JBUFFER_TRACE(jh, "not on correct data list: unfile");
 			J_ASSERT_JH(jh, jh->b_jlist != BJ_Shadow);
 			__journal_unfile_buffer(jh);
 			jh->b_transaction = NULL;
 			JBUFFER_TRACE(jh, "file as data");
 			__journal_file_buffer(jh, handle->h_transaction,
-						wanted_jlist);
+						BJ_SyncData);
 		}
 	} else {
 		JBUFFER_TRACE(jh, "not on a transaction");
-		__journal_file_buffer(jh, handle->h_transaction, wanted_jlist);
+		__journal_file_buffer(jh, handle->h_transaction, BJ_SyncData);
 	}
 no_journal:
 	spin_unlock(&journal_datalist_lock);
@@ -1559,12 +1545,12 @@ __blist_del_buffer(struct journal_head **list, struct journal_head *jh)
  * Remove a buffer from the appropriate transaction list.
  *
  * Note that this function can *change* the value of
- * bh->b_transaction->t_sync_datalist, t_async_datalist, t_buffers, t_forget,
+ * bh->b_transaction->t_sync_datalist, t_buffers, t_forget,
  * t_iobuf_list, t_shadow_list, t_log_list or t_reserved_list.  If the caller
  * is holding onto a copy of one of thee pointers, it could go bad.
  * Generally the caller needs to re-read the pointer from the transaction_t.
  *
- * If bh->b_jlist is BJ_SyncData or BJ_AsyncData then we may have been called
+ * If bh->b_jlist is BJ_SyncData then we may have been called
  * via journal_try_to_free_buffer() or journal_clean_data_list().  In that
  * case, journal_datalist_lock will be held, and the journal may not be locked.
  */
@@ -1590,9 +1576,6 @@ void __journal_unfile_buffer(struct journal_head *jh)
 	case BJ_SyncData:
 		list = &transaction->t_sync_datalist;
 		break;
-	case BJ_AsyncData:
-		list = &transaction->t_async_datalist;
-		break;
 	case BJ_Metadata:
 		transaction->t_nr_buffers--;
 		J_ASSERT_JH(jh, transaction->t_nr_buffers >= 0);
@@ -1658,7 +1641,7 @@ static inline int __journal_try_to_free_buffer(struct buffer_head *bh)
 		goto out;
 
 	if (jh->b_transaction != 0 && jh->b_cp_transaction == 0) {
-		if (jh->b_jlist == BJ_SyncData || jh->b_jlist==BJ_AsyncData) {
+		if (jh->b_jlist == BJ_SyncData) {
 			/* A written-back ordered data buffer */
 			JBUFFER_TRACE(jh, "release data");
 			__journal_unfile_buffer(jh);
@@ -1994,9 +1977,6 @@ void __journal_file_buffer(struct journal_head *jh,
 	case BJ_SyncData:
 		list = &transaction->t_sync_datalist;
 		break;
-	case BJ_AsyncData:
-		list = &transaction->t_async_datalist;
-		break;
 	case BJ_Metadata:
 		transaction->t_nr_buffers++;
 		list = &transaction->t_buffers;

fs/proc/inode.c

@@ -221,7 +221,6 @@ printk("proc_iget: using deleted entry %s, count=%d\n", de->name, atomic_read(&d
 int proc_fill_super(struct super_block *s, void *data, int silent)
 {
 	struct inode * root_inode;
-	struct task_struct *p;
 
 	s->s_blocksize = 1024;
 	s->s_blocksize_bits = 10;
@@ -234,11 +233,7 @@ int proc_fill_super(struct super_block *s, void *data, int silent)
 	/*
 	 * Fixup the root inode's nlink value
 	 */
-	read_lock(&tasklist_lock);
-	for_each_process(p)
-		if (p->pid)
-			root_inode->i_nlink++;
-	read_unlock(&tasklist_lock);
+	root_inode->i_nlink += nr_processes();
 	s->s_root = d_alloc_root(root_inode);
 	if (!s->s_root)
 		goto out_no_root;

fs/proc/proc_misc.c

@@ -140,9 +140,10 @@ static int meminfo_read_proc(char *page, char **start, off_t off,
 	struct page_state ps;
 	unsigned long inactive;
 	unsigned long active;
+	unsigned long free;
 
 	get_page_state(&ps);
-	get_zone_counts(&active, &inactive);
+	get_zone_counts(&active, &inactive, &free);
 
 /*
  * display in kilobytes.

include/linux/ext3_fs.h

@@ -707,7 +707,7 @@ extern int ext3fs_dirhash(const char *name, int len, struct
 /* ialloc.c */
 extern struct inode * ext3_new_inode (handle_t *, struct inode *, int);
 extern void ext3_free_inode (handle_t *, struct inode *);
-extern struct inode * ext3_orphan_get (struct super_block *, ino_t);
+extern struct inode * ext3_orphan_get (struct super_block *, unsigned long);
 extern unsigned long ext3_count_free_inodes (struct super_block *);
 extern unsigned long ext3_count_dirs (struct super_block *);
 extern void ext3_check_inodes_bitmap (struct super_block *);

include/linux/ext3_jbd.h

@@ -132,16 +132,6 @@ __ext3_journal_get_write_access(const char *where,
 	return err;
 }
 
-static inline int
-__ext3_journal_dirty_data(const char *where,
-			  handle_t *handle, struct buffer_head *bh, int async)
-{
-	int err = journal_dirty_data(handle, bh, async);
-	if (err)
-		ext3_journal_abort_handle(where, __FUNCTION__, bh, handle,err);
-	return err;
-}
-
 static inline void
 ext3_journal_forget(handle_t *handle, struct buffer_head *bh)
 {
@@ -183,8 +173,6 @@ __ext3_journal_dirty_metadata(const char *where,
 	__ext3_journal_get_undo_access(__FUNCTION__, (handle), (bh))
 #define ext3_journal_get_write_access(handle, bh) \
 	__ext3_journal_get_write_access(__FUNCTION__, (handle), (bh))
-#define ext3_journal_dirty_data(handle, bh, async) \
-	__ext3_journal_dirty_data(__FUNCTION__, (handle), (bh), (async))
 #define ext3_journal_revoke(handle, blocknr, bh) \
 	__ext3_journal_revoke(__FUNCTION__, (handle), (blocknr), (bh))
 #define ext3_journal_get_create_access(handle, bh) \

include/linux/jbd.h

@@ -360,13 +360,6 @@ struct transaction_s
 	 */
 	struct journal_head *	t_sync_datalist;
 	
-	/*
-	 * Doubly-linked circular list of all writepage data buffers
-	 * still to be written before this transaction can be committed.
-	 * Protected by journal_datalist_lock.
-	 */
-	struct journal_head *	t_async_datalist;
-	
 	/* Doubly-linked circular list of all forget buffers (superseded
            buffers which we can un-checkpoint once this transaction
            commits) */
@@ -654,8 +647,7 @@ extern int	 journal_extend (handle_t *, int nblocks);
 extern int	 journal_get_write_access (handle_t *, struct buffer_head *);
 extern int	 journal_get_create_access (handle_t *, struct buffer_head *);
 extern int	 journal_get_undo_access (handle_t *, struct buffer_head *);
-extern int	 journal_dirty_data (handle_t *,
-				struct buffer_head *, int async);
+extern int	 journal_dirty_data (handle_t *, struct buffer_head *);
 extern int	 journal_dirty_metadata (handle_t *, struct buffer_head *);
 extern void	 journal_release_buffer (handle_t *, struct buffer_head *);
 extern void	 journal_forget (handle_t *, struct buffer_head *);
@@ -806,14 +798,13 @@ extern int journal_blocks_per_page(struct inode *inode);
 /* journaling buffer types */
 #define BJ_None		0	/* Not journaled */
 #define BJ_SyncData	1	/* Normal data: flush before commit */
-#define BJ_AsyncData	2	/* writepage data: wait on it before commit */
-#define BJ_Metadata	3	/* Normal journaled metadata */
-#define BJ_Forget	4	/* Buffer superseded by this transaction */
-#define BJ_IO		5	/* Buffer is for temporary IO use */
-#define BJ_Shadow	6	/* Buffer contents being shadowed to the log */
-#define BJ_LogCtl	7	/* Buffer contains log descriptors */
-#define BJ_Reserved	8	/* Buffer is reserved for access by journal */
-#define BJ_Types	9
+#define BJ_Metadata	2	/* Normal journaled metadata */
+#define BJ_Forget	3	/* Buffer superseded by this transaction */
+#define BJ_IO		4	/* Buffer is for temporary IO use */
+#define BJ_Shadow	5	/* Buffer contents being shadowed to the log */
+#define BJ_LogCtl	6	/* Buffer contains log descriptors */
+#define BJ_Reserved	7	/* Buffer is reserved for access by journal */
+#define BJ_Types	8
  
 extern int jbd_blocks_per_page(struct inode *inode);
 
@@ -860,8 +851,7 @@ static inline int buffer_jdirty(struct buffer_head *bh)
 static inline int buffer_jbd_data(struct buffer_head *bh)
 {
 	return SPLICE_LOCK(buffer_jbd(bh),
-			bh2jh(bh)->b_jlist == BJ_SyncData ||
-			bh2jh(bh)->b_jlist == BJ_AsyncData);
+			bh2jh(bh)->b_jlist == BJ_SyncData);
 }
 
 #ifdef CONFIG_SMP

include/linux/mmzone.h

@@ -193,7 +193,8 @@ typedef struct pglist_data {
 extern int numnodes;
 extern struct pglist_data *pgdat_list;
 
-void get_zone_counts(unsigned long *active, unsigned long *inactive);
+void get_zone_counts(unsigned long *active, unsigned long *inactive,
+			unsigned long *free);
 void build_all_zonelists(void);
 void wakeup_kswapd(struct zone *zone);
 

include/linux/sched.h

@@ -27,6 +27,7 @@
 #include <linux/compiler.h>
 #include <linux/completion.h>
 #include <linux/pid.h>
+#include <linux/percpu.h>
 
 struct exec_domain;
 
@@ -87,6 +88,8 @@ extern unsigned long avenrun[];		/* Load averages */
 
 extern int nr_threads;
 extern int last_pid;
+DECLARE_PER_CPU(unsigned long, process_counts);
+extern int nr_processes(void);
 extern unsigned long nr_running(void);
 extern unsigned long nr_uninterruptible(void);
 extern unsigned long nr_iowait(void);

kernel/exit.c

@@ -41,6 +41,8 @@ static struct dentry * __unhash_process(struct task_struct *p)
 	if (thread_group_leader(p)) {
 		detach_pid(p, PIDTYPE_PGID);
 		detach_pid(p, PIDTYPE_SID);
+		if (p->pid)
+			per_cpu(process_counts, smp_processor_id())--;
 	}
 
 	REMOVE_LINKS(p);
@@ -191,7 +193,7 @@ int is_orphaned_pgrp(int pgrp)
 	return will_become_orphaned_pgrp(pgrp, 0);
 }
 
-static inline int __has_stopped_jobs(int pgrp)
+static inline int has_stopped_jobs(int pgrp)
 {
 	int retval = 0;
 	struct task_struct *p;
@@ -207,17 +209,6 @@ static inline int __has_stopped_jobs(int pgrp)
 	return retval;
 }
 
-static inline int has_stopped_jobs(int pgrp)
-{
-	int retval;
-
-	read_lock(&tasklist_lock);
-	retval = __has_stopped_jobs(pgrp);
-	read_unlock(&tasklist_lock);
-
-	return retval;
-}
-
 /**
  * reparent_to_init() - Reparent the calling kernel thread to the init task.
  *
@@ -504,7 +495,7 @@ static inline void reparent_thread(task_t *p, task_t *father, int traced)
 	    (p->session == father->session)) {
 		int pgrp = p->pgrp;
 
-		if (__will_become_orphaned_pgrp(pgrp, 0) && __has_stopped_jobs(pgrp)) {
+		if (__will_become_orphaned_pgrp(pgrp, 0) && has_stopped_jobs(pgrp)) {
 			__kill_pg_info(SIGHUP, (void *)1, pgrp);
 			__kill_pg_info(SIGCONT, (void *)1, pgrp);
 		}
@@ -589,7 +580,7 @@ static void exit_notify(void)
 	if ((t->pgrp != current->pgrp) &&
 	    (t->session == current->session) &&
 	    __will_become_orphaned_pgrp(current->pgrp, current) &&
-	    __has_stopped_jobs(current->pgrp)) {
+	    has_stopped_jobs(current->pgrp)) {
 		__kill_pg_info(SIGHUP, (void *)1, current->pgrp);
 		__kill_pg_info(SIGCONT, (void *)1, current->pgrp);
 	}

kernel/fork.c

@@ -48,6 +48,8 @@ int nr_threads;
 int max_threads;
 unsigned long total_forks;	/* Handle normal Linux uptimes. */
 
+DEFINE_PER_CPU(unsigned long, process_counts) = 0;
+
 rwlock_t tasklist_lock __cacheline_aligned = RW_LOCK_UNLOCKED;  /* outer */
 
 /*
@@ -57,6 +59,18 @@ rwlock_t tasklist_lock __cacheline_aligned = RW_LOCK_UNLOCKED;  /* outer */
  */
 static task_t *task_cache[NR_CPUS] __cacheline_aligned;
 
+int nr_processes(void)
+{
+	int cpu;
+	int total = 0;
+
+	for (cpu = 0; cpu < NR_CPUS; cpu++) {
+		if (cpu_online(cpu))
+			total += per_cpu(process_counts, cpu);
+	}
+	return total;
+}
+
 void __put_task_struct(struct task_struct *tsk)
 {
 	if (tsk != current) {
@@ -931,6 +945,8 @@ static struct task_struct *copy_process(unsigned long clone_flags,
 		attach_pid(p, PIDTYPE_TGID, p->tgid);
 		attach_pid(p, PIDTYPE_PGID, p->pgrp);
 		attach_pid(p, PIDTYPE_SID, p->session);
+		if (p->pid)
+			per_cpu(process_counts, smp_processor_id())++;
 	} else
 		link_pid(p, p->pids + PIDTYPE_TGID, &p->group_leader->pids[PIDTYPE_TGID].pid);
 

mm/page_alloc.c

@@ -776,15 +776,18 @@ void get_full_page_state(struct page_state *ret)
 	__get_page_state(ret, sizeof(*ret) / sizeof(unsigned long));
 }
 
-void get_zone_counts(unsigned long *active, unsigned long *inactive)
+void get_zone_counts(unsigned long *active,
+		unsigned long *inactive, unsigned long *free)
 {
 	struct zone *zone;
 
 	*active = 0;
 	*inactive = 0;
+	*free = 0;
 	for_each_zone(zone) {
 		*active += zone->nr_active;
 		*inactive += zone->nr_inactive;
+		*free += zone->free_pages;
 	}
 }
 
@@ -838,6 +841,7 @@ void show_free_areas(void)
 	int cpu, temperature;
 	unsigned long active;
 	unsigned long inactive;
+	unsigned long free;
 	struct zone *zone;
 
 	for_each_zone(zone) {
@@ -863,7 +867,7 @@ void show_free_areas(void)
 	}
 
 	get_page_state(&ps);
-	get_zone_counts(&active, &inactive);
+	get_zone_counts(&active, &inactive, &free);
 
 	printk("\nFree pages: %11ukB (%ukB HighMem)\n",
 		K(nr_free_pages()),

mm/readahead.c

@@ -477,7 +477,8 @@ unsigned long max_sane_readahead(unsigned long nr)
 {
 	unsigned long active;
 	unsigned long inactive;
+	unsigned long free;
 
-	get_zone_counts(&active, &inactive);
-	return min(nr, inactive / 2);
+	get_zone_counts(&active, &inactive, &free);
+	return min(nr, (inactive + free) / 2);
 }