The patch titled ext2: allocate ->s_blockgroup_lock separately has been added to the -mm tree. Its filename is ext2-allocate-s_blockgroup_lock-separately.patch Before you just go and hit "reply", please: a) Consider who else should be cc'ed b) Prefer to cc a suitable mailing list as well c) Ideally: find the original patch on the mailing list and do a reply-to-all to that, adding suitable additional cc's *** Remember to use Documentation/SubmitChecklist when testing your code *** See http://userweb.kernel.org/~akpm/stuff/added-to-mm.txt to find out what to do about this The current -mm tree may be found at http://userweb.kernel.org/~akpm/mmotm/ ------------------------------------------------------ Subject: ext2: allocate ->s_blockgroup_lock separately From: Pekka J Enberg <penberg@xxxxxxxxxxxxxx> 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. Cc: Andreas Dilger <adilger@xxxxxxx> Signed-off-by: Pekka Enberg <penberg@xxxxxxxxxxxxxx> Cc: <linux-ext4@xxxxxxxxxxxxxxx> Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx> --- fs/ext2/super.c | 10 +++++++++- include/linux/ext2_fs_sb.h | 4 ++-- 2 files changed, 11 insertions(+), 3 deletions(-) diff -puN fs/ext2/super.c~ext2-allocate-s_blockgroup_lock-separately fs/ext2/super.c --- a/fs/ext2/super.c~ext2-allocate-s_blockgroup_lock-separately +++ a/fs/ext2/super.c @@ -132,6 +132,7 @@ static void ext2_put_super (struct super percpu_counter_destroy(&sbi->s_dirs_counter); brelse (sbi->s_sbh); sb->s_fs_info = NULL; + kfree(sbi->s_blockgroup_lock); kfree(sbi); return; @@ -765,6 +766,13 @@ static int ext2_fill_super(struct super_ sbi = kzalloc(sizeof(*sbi), GFP_KERNEL); if (!sbi) return -ENOMEM; + + sbi->s_blockgroup_lock = + kzalloc(sizeof(struct blockgroup_lock), GFP_KERNEL); + if (!sbi->s_blockgroup_lock) { + kfree(sbi); + return -ENOMEM; + } sb->s_fs_info = sbi; sbi->s_sb_block = sb_block; @@ -992,7 +1000,7 @@ static int ext2_fill_super(struct super_ printk ("EXT2-fs: not enough memory\n"); goto failed_mount; } - bgl_lock_init(&sbi->s_blockgroup_lock); + bgl_lock_init(sbi->s_blockgroup_lock); sbi->s_debts = kcalloc(sbi->s_groups_count, sizeof(*sbi->s_debts), GFP_KERNEL); if (!sbi->s_debts) { printk ("EXT2-fs: not enough memory\n"); diff -puN include/linux/ext2_fs_sb.h~ext2-allocate-s_blockgroup_lock-separately include/linux/ext2_fs_sb.h --- a/include/linux/ext2_fs_sb.h~ext2-allocate-s_blockgroup_lock-separately +++ a/include/linux/ext2_fs_sb.h @@ -101,7 +101,7 @@ struct ext2_sb_info { struct percpu_counter s_freeblocks_counter; struct percpu_counter s_freeinodes_counter; struct percpu_counter s_dirs_counter; - struct blockgroup_lock s_blockgroup_lock; + struct blockgroup_lock *s_blockgroup_lock; /* root of the per fs reservation window tree */ spinlock_t s_rsv_window_lock; struct rb_root s_rsv_window_root; @@ -111,7 +111,7 @@ struct ext2_sb_info { static inline spinlock_t * sb_bgl_lock(struct ext2_sb_info *sbi, unsigned int block_group) { - return bgl_lock_ptr(&sbi->s_blockgroup_lock, block_group); + return bgl_lock_ptr(sbi->s_blockgroup_lock, block_group); } #endif /* _LINUX_EXT2_FS_SB */ _ Patches currently in -mm which might be from penberg@xxxxxxxxxxxxxx are fs-introduce-bgl_lock_ptr.patch repeatable-slab-corruption-with-ltp-msgctl08.patch linux-next.patch ext4-allocate-s_blockgroup_lock-separately.patch ext2-allocate-s_blockgroup_lock-separately.patch ext3-allocate-s_blockgroup_lock-separately.patch slab-leaks3-default-y.patch -- To unsubscribe from this list: send the line "unsubscribe mm-commits" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html