[PATCH] ext4: do not normalize block requests from fallocate.

From: Vivek Haldar
Date: Fri May 13 2011 - 17:19:55 EST


Currently, an fallocate request of size slightly larger than a power of
2 is turned into two block requests, each a power of 2, with the extra
blocks pre-allocated for future use. When an application calls
fallocate, it already has an idea about how large the file may grow so
there is usually little benefit to reserve extra blocks on the
preallocation list. This reduces disk fragmentation.

Tested: fsstress. Also verified manually that fallocat'ed files are
contiguously laid out with this change (whereas without it they begin at
power-of-2 boundaries, leaving blocks in between). CPU usage of
fallocate is not appreciably higher. In a tight fallocate loop, CPU
usage hovers between 5%-8% with this change, and 5%-7% without it.

Signed-off-by: Vivek Haldar <haldar@xxxxxxxxxx>
---
fs/ext4/ext4.h | 2 ++
fs/ext4/extents.c | 5 ++++-
2 files changed, 6 insertions(+), 1 deletions(-)

diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
index 076c5d2..e606d34 100644
--- a/fs/ext4/ext4.h
+++ b/fs/ext4/ext4.h
@@ -512,6 +512,8 @@ struct ext4_new_group_data {
/* Convert extent to initialized after IO complete */
#define EXT4_GET_BLOCKS_IO_CONVERT_EXT (EXT4_GET_BLOCKS_CONVERT|\
EXT4_GET_BLOCKS_CREATE_UNINIT_EXT)
+ /* Don't normalize when fallocat'ing */
+#define EXT4_GET_BLOCKS_NO_NORMALIZE 0x0020

/*
* Flags used by ext4_free_blocks
diff --git a/fs/ext4/extents.c b/fs/ext4/extents.c
index e363f21..f02dd52 100644
--- a/fs/ext4/extents.c
+++ b/fs/ext4/extents.c
@@ -3304,6 +3304,8 @@ int ext4_ext_map_blocks(handle_t *handle, struct inode *inode,
else
/* disable in-core preallocation for non-regular files */
ar.flags = 0;
+ if (flags & EXT4_GET_BLOCKS_NO_NORMALIZE)
+ ar.flags |= EXT4_MB_HINT_NOPREALLOC;
newblock = ext4_mb_new_blocks(handle, &ar, &err);
if (!newblock)
goto out2;
@@ -3549,7 +3551,8 @@ retry:
break;
}
ret = ext4_map_blocks(handle, inode, &map,
- EXT4_GET_BLOCKS_CREATE_UNINIT_EXT);
+ EXT4_GET_BLOCKS_CREATE_UNINIT_EXT |
+ EXT4_GET_BLOCKS_NO_NORMALIZE);
if (ret <= 0) {
#ifdef EXT4FS_DEBUG
WARN_ON(ret <= 0);
--
1.7.3.1

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/