set_blocksize: Allow to set a larger block size than PAGE_SIZE

set_blocksize is changed to allow to specify a blocksize larger than a
page. If that occurs then we switch the device to use compound pages.

Signed-off-by: Christoph Lameter <clameter@sgi.com>

---
 fs/block_dev.c |   55 ++++++++++++++++++++++++++++++++++++++++++++++++-------
 fs/buffer.c    |   12 +++++-------
 fs/inode.c     |    7 +++++++
 3 files changed, 60 insertions(+), 14 deletions(-)

Index: linux-2.6.21-rc7-mm2/fs/block_dev.c
===================================================================
--- linux-2.6.21-rc7-mm2.orig/fs/block_dev.c	2007-04-27 15:27:37.000000000 -0700
+++ linux-2.6.21-rc7-mm2/fs/block_dev.c	2007-04-27 16:04:58.000000000 -0700
@@ -65,12 +65,41 @@ static void kill_bdev(struct block_devic
 		return;
 	invalidate_bh_lrus();
 	truncate_inode_pages(bdev->bd_inode->i_mapping, 0);
-}	
+}
+
+#ifdef CONFIG_LARGE_BLOCKSIZE
+
+static int page_cache_min_order = 0;
+static int page_cache_max_order = 4;	/* 64k */
+
+static int __init setup_page_cache_min_order(char *str)
+{
+	get_option (&str, &page_cache_min_order);
+	return 1;
+}
+
+__setup("page_cache_min_order=", setup_page_cache_min_order);
+
+static int __init setup_page_cache_max_order(char *str)
+{
+	get_option (&str, &page_cache_max_order);
+	return 1;
+}
+
+__setup("page_cache_max_order=", setup_page_cache_max_order);
+#else
+
+#define page_cache_min_order 0
+#define page_cache_max_order 0
+
+#endif
 
 int set_blocksize(struct block_device *bdev, int size)
 {
-	/* Size must be a power of two, and between 512 and PAGE_SIZE */
-	if (size > PAGE_SIZE || size < 512 || !is_power_of_2(size))
+	if (size < 512 || !is_power_of_2(size))
+		return -EINVAL;
+
+	if (size > (PAGE_SIZE << page_cache_max_order))
 		return -EINVAL;
 
 	/* Size cannot be smaller than the size supported by the device */
@@ -79,10 +108,20 @@ int set_blocksize(struct block_device *b
 
 	/* Don't change the size if it is same as current */
 	if (bdev->bd_block_size != size) {
+		int bits = blksize_bits(size);
+		int order;
+		struct address_space *mapping =
+			bdev->bd_inode->i_mapping;
+
 		sync_blockdev(bdev);
-		bdev->bd_block_size = size;
-		bdev->bd_inode->i_blkbits = blksize_bits(size);
 		kill_bdev(bdev);
+		bdev->bd_block_size = size;
+		bdev->bd_inode->i_blkbits = bits;
+		if (bits < PAGE_SHIFT + page_cache_min_order)
+			order = page_cache_min_order;
+		else
+			order = bits - PAGE_SHIFT;
+		set_mapping_order(mapping, order);
 	}
 	return 0;
 }
@@ -93,8 +132,10 @@ int sb_set_blocksize(struct super_block 
 {
 	if (set_blocksize(sb->s_bdev, size))
 		return 0;
-	/* If we get here, we know size is power of two
-	 * and it's value is between 512 and PAGE_SIZE */
+	/*
+	 * If we get here, we know size is power of two
+	 * and it's value is valid for the page cache
+	 */
 	sb->s_blocksize = size;
 	sb->s_blocksize_bits = blksize_bits(size);
 	return sb->s_blocksize;
Index: linux-2.6.21-rc7-mm2/fs/buffer.c
===================================================================
--- linux-2.6.21-rc7-mm2.orig/fs/buffer.c	2007-04-27 15:44:41.000000000 -0700
+++ linux-2.6.21-rc7-mm2/fs/buffer.c	2007-04-27 15:52:06.000000000 -0700
@@ -1086,7 +1086,7 @@ __getblk_slow(struct block_device *bdev,
 {
 	/* Size must be multiple of hard sectorsize */
 	if (unlikely(size & (bdev_hardsect_size(bdev)-1) ||
-			(size < 512 || size > PAGE_SIZE))) {
+		!page_cache_valid_blocksize(size))) {
 		printk(KERN_ERR "getblk(): invalid block size %d requested\n",
 					size);
 		printk(KERN_ERR "hardsect size: %d\n",
@@ -2101,20 +2101,18 @@ int cont_prepare_write(struct page *page
 			page_cache_release(new_page);
 			continue;
 		}
-		zerofrom = page_cache_offset(*bytes);
+		zerofrom = page_cache_offset(mapping, *bytes);
 		if (zerofrom & (blocksize-1)) {
 			*bytes |= (blocksize-1);
 			(*bytes)++;
 		}
 		status = __block_prepare_write(inode, new_page, zerofrom,
-					page_cache_size(mapping), get_block);
+					page_size, get_block);
 		if (status)
 			goto out_unmap;
 		zero_user_page(page, zerofrom,
-				page_cache_size(mapping) - zerofrom,
-				KM_USER0);
-		generic_commit_write(NULL, new_page, zerofrom,
-				page_cache_size(mapping));
+				page_size - zerofrom, KM_USER0);
+		generic_commit_write(NULL, new_page, zerofrom, page_size);
 		unlock_page(new_page);
 		page_cache_release(new_page);
 	}
Index: linux-2.6.21-rc7-mm2/fs/inode.c
===================================================================
--- linux-2.6.21-rc7-mm2.orig/fs/inode.c	2007-04-27 15:27:37.000000000 -0700
+++ linux-2.6.21-rc7-mm2/fs/inode.c	2007-04-27 15:52:06.000000000 -0700
@@ -155,6 +155,13 @@ static struct inode *alloc_inode(struct 
  		mapping->host = inode;
 		mapping->flags = 0;
 		mapping_set_gfp_mask(mapping, GFP_HIGH_MOVABLE);
+
+		if (inode->i_blkbits > PAGE_SHIFT)
+			set_mapping_order(mapping,
+				inode->i_blkbits - PAGE_SHIFT);
+		else
+			set_mapping_order(mapping, 0);
+
 		mapping->assoc_mapping = NULL;
 		mapping->backing_dev_info = &default_backing_dev_info;