Index: linux-2.6.18-128.1.6/fs/ext4/ext4_sb.h
===================================================================
--- linux-2.6.18-128.1.6.orig/fs/ext4/ext4_sb.h	2009-05-28 17:16:51.000000000 +0530
+++ linux-2.6.18-128.1.6/fs/ext4/ext4_sb.h	2009-05-28 17:16:52.000000000 +0530
@@ -108,11 +108,14 @@
 
 	/* tunables */
 	unsigned long s_stripe;
-	unsigned long s_mb_stream_request;
+	unsigned long s_mb_small_req;
+	unsigned long s_mb_large_req;
 	unsigned long s_mb_max_to_scan;
 	unsigned long s_mb_min_to_scan;
 	unsigned long s_mb_stats;
 	unsigned long s_mb_order2_reqs;
+	unsigned long *s_mb_prealloc_table;
+	unsigned long s_mb_prealloc_table_size;
 	unsigned long s_mb_group_prealloc;
 	/* where last allocation was done - for stream allocation */
 	unsigned long s_mb_last_group;
Index: linux-2.6.18-128.1.6/fs/ext4/mballoc.c
===================================================================
--- linux-2.6.18-128.1.6.orig/fs/ext4/mballoc.c	2009-05-28 17:16:51.000000000 +0530
+++ linux-2.6.18-128.1.6/fs/ext4/mballoc.c	2009-05-28 17:19:57.000000000 +0530
@@ -1744,7 +1744,7 @@
 	if (size < isize)
 		size = isize;
 
-	if (size < sbi->s_mb_stream_request &&
+	if ((ac->ac_g_ex.fe_len < sbi->s_mb_large_req) &&
 			(ac->ac_flags & EXT4_MB_HINT_DATA)) {
 		/* TBD: may be hot point */
 		spin_lock(&sbi->s_md_lock);
@@ -2484,6 +2484,26 @@
 	return -ENOMEM;
 }
 
+static void ext4_mb_prealloc_table_add(struct ext4_sb_info *sbi, int value)
+{
+	int i;
+
+	if (value > (sbi->s_blocks_per_group - 1 - 1 - sbi->s_itb_per_group))
+		return;
+
+	for (i = 0; i < sbi->s_mb_prealloc_table_size; i++) {
+		if (sbi->s_mb_prealloc_table[i] == 0) {
+			sbi->s_mb_prealloc_table[i] = value;
+			return;
+		}
+
+		/* they should add values in order */
+		if (value <= sbi->s_mb_prealloc_table[i])
+			return;
+	}
+}
+
+
 int ext4_mb_init(struct super_block *sb, int needs_recovery)
 {
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
@@ -2542,15 +2562,59 @@
 	sbi->s_mb_max_to_scan = MB_DEFAULT_MAX_TO_SCAN;
 	sbi->s_mb_min_to_scan = MB_DEFAULT_MIN_TO_SCAN;
 	sbi->s_mb_stats = MB_DEFAULT_STATS;
-	sbi->s_mb_stream_request = MB_DEFAULT_STREAM_THRESHOLD;
 	sbi->s_mb_order2_reqs = MB_DEFAULT_ORDER2_REQS;
 	sbi->s_mb_history_filter = EXT4_MB_HISTORY_DEFAULT;
-	sbi->s_mb_group_prealloc = MB_DEFAULT_GROUP_PREALLOC;
+
+	if (sbi->s_stripe == 0) {
+		sbi->s_mb_prealloc_table_size = 8;
+		i = sbi->s_mb_prealloc_table_size * sizeof(unsigned long);
+		sbi->s_mb_prealloc_table = kmalloc(i, GFP_NOFS);
+		if (sbi->s_mb_prealloc_table == NULL) {
+				clear_opt(sbi->s_mount_opt, MBALLOC);
+				kfree(sbi->s_mb_offsets);
+				kfree(sbi->s_mb_maxs);
+				return -ENOMEM;
+		}
+		memset(sbi->s_mb_prealloc_table, 0, i);
+
+		ext4_mb_prealloc_table_add(sbi, 4);
+		ext4_mb_prealloc_table_add(sbi, 8);
+		ext4_mb_prealloc_table_add(sbi, 16);
+		ext4_mb_prealloc_table_add(sbi, 32);
+		ext4_mb_prealloc_table_add(sbi, 64);
+		ext4_mb_prealloc_table_add(sbi, 128);
+		ext4_mb_prealloc_table_add(sbi, 256);
+		ext4_mb_prealloc_table_add(sbi, 512);
+
+		sbi->s_mb_small_req = 256;
+		sbi->s_mb_large_req = 1024;
+		sbi->s_mb_group_prealloc = 512;
+	} else {
+		sbi->s_mb_prealloc_table_size = 3;
+		i = sbi->s_mb_prealloc_table_size * sizeof(unsigned long);
+		sbi->s_mb_prealloc_table = kmalloc(i, GFP_NOFS);
+		if (sbi->s_mb_prealloc_table == NULL) {
+			clear_opt(sbi->s_mount_opt, MBALLOC);
+			kfree(sbi->s_mb_offsets);
+			kfree(sbi->s_mb_maxs);
+			return -ENOMEM;
+		}
+		memset(sbi->s_mb_prealloc_table, 0, i);
+
+		ext4_mb_prealloc_table_add(sbi, sbi->s_stripe);
+		ext4_mb_prealloc_table_add(sbi, sbi->s_stripe * 2);
+		ext4_mb_prealloc_table_add(sbi, sbi->s_stripe * 4);
+
+		sbi->s_mb_small_req = sbi->s_stripe;
+		sbi->s_mb_large_req = sbi->s_stripe * 8;
+		sbi->s_mb_group_prealloc = sbi->s_stripe * 4;
+	}
 
 	i = sizeof(struct ext4_locality_group) * num_possible_cpus();
 	sbi->s_locality_groups = kmalloc(i, GFP_KERNEL);
 	if (sbi->s_locality_groups == NULL) {
 		clear_opt(sbi->s_mount_opt, MBALLOC);
+		kfree(sbi->s_mb_prealloc_table);
 		kfree(sbi->s_mb_offsets);
 		kfree(sbi->s_mb_maxs);
 		return -ENOMEM;
@@ -2725,10 +2789,82 @@
 #define EXT4_MB_MAX_TO_SCAN_NAME	"max_to_scan"
 #define EXT4_MB_MIN_TO_SCAN_NAME	"min_to_scan"
 #define EXT4_MB_ORDER2_REQ		"order2_req"
-#define EXT4_MB_STREAM_REQ		"stream_req"
+#define EXT4_MB_SMALL_REQ          	"small_req"
+#define EXT4_MB_LARGE_REQ            	"large_req"
+#define EXT4_MB_PREALLOC_TABLE          "prealloc_table"
 #define EXT4_MB_GROUP_PREALLOC		"group_prealloc"
 
+static int ext4_mb_prealloc_table_proc_read(char *page, char **start, off_t off,
+					    int count, int *eof, void *data)
+{
+	struct ext4_sb_info *sbi = data;
+	int len = 0;
+	int i;
+
+	*eof = 1;
+	if (off != 0)
+		return 0;
+
+	for (i = 0; i < sbi->s_mb_prealloc_table_size; i++)
+		len += sprintf(page + len, "%ld ",
+			       sbi->s_mb_prealloc_table[i]);
+	len += sprintf(page + len, "\n");
+
+	*start = page;
+	return len;
+}
+
+static int ext4_mb_prealloc_table_proc_write(struct file *file,
+					     const char __user *buf,
+					     unsigned long cnt, void *data)
+{
+	struct ext4_sb_info *sbi = data;
+	unsigned long value;
+	unsigned long prev = 0;
+	char str[128];
+	char *cur;
+	char *end;
+	unsigned long *new_table;
+	int num = 0;
+	int i = 0;
+
+	if (cnt >= sizeof(str))
+		return -EINVAL;
+	if (copy_from_user(str, buf, cnt))
+		return -EFAULT;
+
+	num = 0;
+	cur = str;
+	end = str + cnt;
+	while (cur < end) {
+		while ((cur < end) && (*cur == ' ')) cur++;
+		value = simple_strtol(cur, &cur, 0);
+		if (value == 0)
+			break;
+		if (value <= prev)
+			return -EINVAL;
+		prev = value;
+		num++;
+	}
 
+	new_table = kmalloc(num * sizeof(*new_table), GFP_KERNEL);
+	if (new_table == NULL)
+		return -ENOMEM;
+	kfree(sbi->s_mb_prealloc_table);
+	memset(new_table, 0, num * sizeof(*new_table));
+	sbi->s_mb_prealloc_table = new_table;
+	sbi->s_mb_prealloc_table_size = num;
+	cur = str;
+	end = str + cnt;
+	while (cur < end && i < num) {
+	while ((cur < end) && (*cur == ' ')) cur++;
+		value = simple_strtol(cur, &cur, 0);
+		ext4_mb_prealloc_table_add(sbi, value);
+		i++;
+	}
+
+	return cnt;
+}
 
 #define MB_PROC_FOPS(name)					\
 static int ext4_mb_##name##_proc_show(struct seq_file *m, void *v)	\
@@ -2774,7 +2910,8 @@
 MB_PROC_FOPS(max_to_scan);
 MB_PROC_FOPS(min_to_scan);
 MB_PROC_FOPS(order2_reqs);
-MB_PROC_FOPS(stream_request);
+MB_PROC_FOPS(small_req);
+MB_PROC_FOPS(large_req);
 MB_PROC_FOPS(group_prealloc);
 
 #define	MB_PROC_HANDLER(name, var)					\
@@ -2795,6 +2932,7 @@
 	mode_t mode = S_IFREG | S_IRUGO | S_IWUSR;
 	struct ext4_sb_info *sbi = EXT4_SB(sb);
 	struct proc_dir_entry *proc;
+	struct proc_dir_entry *proc_entry;
 	char devname[64];
 
 	if (proc_root_ext4 == NULL) {
@@ -2808,15 +2946,29 @@
 	MB_PROC_HANDLER(EXT4_MB_MAX_TO_SCAN_NAME, max_to_scan);
 	MB_PROC_HANDLER(EXT4_MB_MIN_TO_SCAN_NAME, min_to_scan);
 	MB_PROC_HANDLER(EXT4_MB_ORDER2_REQ, order2_reqs);
-	MB_PROC_HANDLER(EXT4_MB_STREAM_REQ, stream_request);
+	MB_PROC_HANDLER(EXT4_MB_SMALL_REQ, small_req);
+	MB_PROC_HANDLER(EXT4_MB_LARGE_REQ, large_req);
 	MB_PROC_HANDLER(EXT4_MB_GROUP_PREALLOC, group_prealloc);
 
+	proc_entry = create_proc_entry(EXT4_MB_PREALLOC_TABLE, S_IFREG |
+				       S_IRUGO | S_IWUSR, sbi->s_mb_proc);
+	if (proc_entry == NULL) {
+		printk(KERN_ERR "EXT4-fs: unable to create %s\n",
+		       EXT4_MB_PREALLOC_TABLE);
+		goto err_out;
+	}
+	proc_entry->data = sbi;
+	proc_entry->read_proc = ext4_mb_prealloc_table_proc_read;
+	proc_entry->write_proc = ext4_mb_prealloc_table_proc_write;
+
 	return 0;
 
 err_out:
 	printk(KERN_ERR "EXT4-fs: Unable to create %s\n", devname);
 	remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_PREALLOC_TABLE, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_LARGE_REQ, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_SMALL_REQ, sbi->s_mb_proc);
 	remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc);
 	remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc);
 	remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc);
@@ -2838,7 +2990,9 @@
 
 	bdevname(sb->s_bdev, devname);
 	remove_proc_entry(EXT4_MB_GROUP_PREALLOC, sbi->s_mb_proc);
-	remove_proc_entry(EXT4_MB_STREAM_REQ, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_PREALLOC_TABLE, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_LARGE_REQ, sbi->s_mb_proc);
+	remove_proc_entry(EXT4_MB_SMALL_REQ, sbi->s_mb_proc);
 	remove_proc_entry(EXT4_MB_ORDER2_REQ, sbi->s_mb_proc);
 	remove_proc_entry(EXT4_MB_MIN_TO_SCAN_NAME, sbi->s_mb_proc);
 	remove_proc_entry(EXT4_MB_MAX_TO_SCAN_NAME, sbi->s_mb_proc);
@@ -3032,11 +3186,12 @@
 ext4_mb_normalize_request(struct ext4_allocation_context *ac,
 				struct ext4_allocation_request *ar)
 {
-	int bsbits, max;
+	int bsbits, i, wind;
 	ext4_lblk_t end;
-	loff_t size, orig_size, start_off;
+	loff_t size, orig_size;
 	ext4_lblk_t start, orig_start;
 	struct ext4_inode_info *ei = EXT4_I(ac->ac_inode);
+	struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
 	struct ext4_prealloc_space *pa;
 
 	/* do normalize only data requests, metadata requests
@@ -3066,49 +3221,35 @@
 	size = size << bsbits;
 	if (size < i_size_read(ac->ac_inode))
 		size = i_size_read(ac->ac_inode);
+	size = (size + ac->ac_sb->s_blocksize - 1) >> bsbits;
 
-	/* max size of free chunks */
-	max = 2 << bsbits;
+	start = wind = 0;
 
-#define NRL_CHECK_SIZE(req, size, max, chunk_size)	\
-		(req <= (size) || max <= (chunk_size))
+	/* let's choose preallocation window depending on file size */
+	for (i = 0; i < sbi->s_mb_prealloc_table_size; i++) {
+		if (size <= sbi->s_mb_prealloc_table[i]) {
+			wind = sbi->s_mb_prealloc_table[i];
+			break;
+		}
+	}
+	size = wind;
 
-	/* first, try to predict filesize */
-	/* XXX: should this table be tunable? */
-	start_off = 0;
-	if (size <= 16 * 1024) {
-		size = 16 * 1024;
-	} else if (size <= 32 * 1024) {
-		size = 32 * 1024;
-	} else if (size <= 64 * 1024) {
-		size = 64 * 1024;
-	} else if (size <= 128 * 1024) {
-		size = 128 * 1024;
-	} else if (size <= 256 * 1024) {
-		size = 256 * 1024;
-	} else if (size <= 512 * 1024) {
-		size = 512 * 1024;
-	} else if (size <= 1024 * 1024) {
-		size = 1024 * 1024;
-	} else if (NRL_CHECK_SIZE(size, 4 * 1024 * 1024, max, 2 * 1024)) {
-		start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
-						(21 - bsbits)) << 21;
-		size = 2 * 1024 * 1024;
-	} else if (NRL_CHECK_SIZE(size, 8 * 1024 * 1024, max, 4 * 1024)) {
-		start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
-							(22 - bsbits)) << 22;
-		size = 4 * 1024 * 1024;
-	} else if (NRL_CHECK_SIZE(ac->ac_o_ex.fe_len,
-					(8<<20)>>bsbits, max, 8 * 1024)) {
-		start_off = ((loff_t)ac->ac_o_ex.fe_logical >>
-							(23 - bsbits)) << 23;
-		size = 8 * 1024 * 1024;
-	} else {
-		start_off = (loff_t)ac->ac_o_ex.fe_logical << bsbits;
-		size	  = ac->ac_o_ex.fe_len << bsbits;
+	if (wind == 0) {
+		__u64 tstart, tend;
+		/* file is quite large, we now preallocate with
+		 * the biggest configured window with regart to
+		 * logical offset */
+		wind = sbi->s_mb_prealloc_table[i - 1];
+		tstart = ac->ac_o_ex.fe_logical;
+		do_div(tstart, wind);
+		start = tstart * wind;
+		tend = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len - 1;
+		do_div(tend, wind);
+		tend = tend * wind + wind;
+		size = tend - start;
 	}
-	orig_size = size = size >> bsbits;
-	orig_start = start = start_off >> bsbits;
+	orig_size = size;
+	orig_start = start;
 
 	/* don't cover already allocated blocks in selected range */
 	if (ar->pleft && start <= ar->lleft) {
@@ -3185,7 +3326,6 @@
 	}
 	BUG_ON(start + size <= ac->ac_o_ex.fe_logical &&
 			start > ac->ac_o_ex.fe_logical);
-	BUG_ON(size <= 0 || size >= EXT4_BLOCKS_PER_GROUP(ac->ac_sb));
 
 	/* now prepare goal request */
 
@@ -4077,22 +4217,32 @@
 {
 	struct ext4_sb_info *sbi = EXT4_SB(ac->ac_sb);
 	int bsbits = ac->ac_sb->s_blocksize_bits;
-	loff_t size, isize;
+	loff_t size;
 
 	if (!(ac->ac_flags & EXT4_MB_HINT_DATA))
 		return;
 
-	size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len;
-	isize = i_size_read(ac->ac_inode) >> bsbits;
-	size = max(size, isize);
-
-	/* don't use group allocation for large files */
-	if (size >= sbi->s_mb_stream_request)
+	if (ac->ac_o_ex.fe_len >= sbi->s_mb_small_req)
 		return;
 
 	if (unlikely(ac->ac_flags & EXT4_MB_HINT_GOAL_ONLY))
 		return;
 
+	/* request is so large that we don't care about
+	 * streaming - it overweights any possible seek */
+	if (ac->ac_o_ex.fe_len >= sbi->s_mb_large_req)
+		return;
+
+	size = ac->ac_o_ex.fe_logical + ac->ac_o_ex.fe_len;
+	size = size << bsbits;
+	if (size < i_size_read(ac->ac_inode))
+		size = i_size_read(ac->ac_inode);
+	size = (size + ac->ac_sb->s_blocksize - 1) >> bsbits;
+
+	/* don't use group allocation for large files */
+	if (size >= sbi->s_mb_large_req)
+		return;
+
 	BUG_ON(ac->ac_lg != NULL);
 	/*
 	 * locality group prealloc space are per cpu. The reason for having