ext2: allocate ->s_blockgroup_lock separately
authorPekka J Enberg <penberg@cs.helsinki.fi>
Thu, 8 Jan 2009 02:07:19 +0000 (18:07 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 8 Jan 2009 16:31:00 +0000 (08:31 -0800)
commit18a82eb9f980b5e02cea651e4ecda26265d98933
tree7626c16d880eed47f0eaf323f8e8e9441c84d357
parent22d613d13445de9dea6edc3289c304237eb191f6
ext2: allocate ->s_blockgroup_lock separately

As spotted by kmemtrace, struct ext2_sb_info is 17024 bytes on 64-bit
which makes it a very bad fit for SLAB allocators.  The culprit of the
wasted memory is ->s_blockgroup_lock which can be as big as 16 KB when
NR_CPUS >= 32.

To fix that, allocate ->s_blockgroup_lock, which fits nicely in a order 2
page in the worst case, separately.  This shinks down struct ext2_sb_info
enough to fit a 1 KB slab cache so now we allocate 16 KB + 1 KB instead of
32 KB saving 15 KB of memory.

Acked-by: Andreas Dilger <adilger@sun.com>
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Cc: <linux-ext4@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
fs/ext2/super.c
include/linux/ext2_fs_sb.h