On 4/19/12 2:20 PM, Theodore Ts'o wrote:
> Wire up the use of the O_HOT and O_COLD open flags so that when an
> inode is being created, it can influence which part of the disk gets
> used on rotational storage devices.
I'm curious to know how this will work for example on a linear device
make up of rotational devices (possibly a concat of raids, etc).
At least for dm, it will be still marked as rotational,
but the relative speed of regions of the linear device can't be inferred
from the offset within the device.
Do we really have enough information about the storage under us to
know what parts are "fast" and what parts are "slow?"
-Eric
> Signed-off-by: "Theodore Ts'o" <tytso@xxxxxxx>
> ---
> fs/ext4/ext4.h | 8 +++++++-
> fs/ext4/ialloc.c | 33 +++++++++++++++++++++++++++------
> fs/ext4/migrate.c | 2 +-
> fs/ext4/namei.c | 15 +++++++++++----
> 4 files changed, 46 insertions(+), 12 deletions(-)
>
> diff --git a/fs/ext4/ext4.h b/fs/ext4/ext4.h
> index 0e01e90..6539c9a 100644
> --- a/fs/ext4/ext4.h
> +++ b/fs/ext4/ext4.h
> @@ -1553,6 +1553,12 @@ struct ext4_dir_entry_2 {
> #define EXT4_MAX_REC_LEN ((1<<16)-1)
>
> /*
> + * Flags for ext4_new_inode()
> + */
> +#define EXT4_NEWI_HOT 0x0001
> +#define EXT4_NEWI_COLD 0x0002
> +
> +/*
> * If we ever get support for fs block sizes > page_size, we'll need
> * to remove the #if statements in the next two functions...
> */
> @@ -1850,7 +1856,7 @@ extern int ext4fs_dirhash(const char *name, int len, struct
> /* ialloc.c */
> extern struct inode *ext4_new_inode(handle_t *, struct inode *, umode_t,
> const struct qstr *qstr, __u32 goal,
> - uid_t *owner);
> + uid_t *owner, int flags);
> extern void ext4_free_inode(handle_t *, struct inode *);
> extern struct inode * ext4_orphan_get(struct super_block *, unsigned long);
> extern unsigned long ext4_count_free_inodes(struct super_block *);
> diff --git a/fs/ext4/ialloc.c b/fs/ext4/ialloc.c
> index 409c2ee..3dcc8c8 100644
> --- a/fs/ext4/ialloc.c
> +++ b/fs/ext4/ialloc.c
> @@ -363,7 +363,7 @@ static void get_orlov_stats(struct super_block *sb, ext4_group_t g,
>
> static int find_group_orlov(struct super_block *sb, struct inode *parent,
> ext4_group_t *group, umode_t mode,
> - const struct qstr *qstr)
> + const struct qstr *qstr, int flags)
> {
> ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
> struct ext4_sb_info *sbi = EXT4_SB(sb);
> @@ -508,13 +508,20 @@ fallback_retry:
> }
>
> static int find_group_other(struct super_block *sb, struct inode *parent,
> - ext4_group_t *group, umode_t mode)
> + ext4_group_t *group, umode_t mode, int flags)
> {
> ext4_group_t parent_group = EXT4_I(parent)->i_block_group;
> ext4_group_t i, last, ngroups = ext4_get_groups_count(sb);
> struct ext4_group_desc *desc;
> int flex_size = ext4_flex_bg_size(EXT4_SB(sb));
>
> + if ((flags & EXT4_NEWI_HOT) && (ngroups > 3) &&
> + (parent_group > ngroups / 3))
> + parent_group = 0;
> + if ((flags & EXT4_NEWI_COLD) && (ngroups > 3) &&
> + (parent_group < (2 * (ngroups / 3))))
> + parent_group = 2 * (ngroups / 3);
> +
> /*
> * Try to place the inode is the same flex group as its
> * parent. If we can't find space, use the Orlov algorithm to
> @@ -550,7 +557,7 @@ static int find_group_other(struct super_block *sb, struct inode *parent,
> *group = parent_group + flex_size;
> if (*group > ngroups)
> *group = 0;
> - return find_group_orlov(sb, parent, group, mode, NULL);
> + return find_group_orlov(sb, parent, group, mode, NULL, flags);
> }
>
> /*
> @@ -614,7 +621,8 @@ static int find_group_other(struct super_block *sb, struct inode *parent,
> * group to find a free inode.
> */
> struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, umode_t mode,
> - const struct qstr *qstr, __u32 goal, uid_t *owner)
> + const struct qstr *qstr, __u32 goal, uid_t *owner,
> + int flags)
> {
> struct super_block *sb;
> struct buffer_head *inode_bitmap_bh = NULL;
> @@ -643,6 +651,19 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, umode_t mode,
> ei = EXT4_I(inode);
> sbi = EXT4_SB(sb);
>
> + if (blk_queue_nonrot(bdev_get_queue(sb->s_bdev)))
> + flags &= ~(EXT4_NEWI_HOT | EXT4_NEWI_COLD);
> +
> + /*
> + * We will only allow the HOT flag if the user passes the
> + * reserved uid/gid check, or if she has CAP_SYS_RESOURCE
> + */
> + if ((flags & EXT4_NEWI_HOT) &&
> + !(sbi->s_resuid == current_fsuid() ||
> + ((sbi->s_resgid != 0) && in_group_p(sbi->s_resgid)) ||
> + capable(CAP_SYS_RESOURCE)))
> + flags &= ~EXT4_NEWI_HOT;
> +
> if (!goal)
> goal = sbi->s_inode_goal;
>
> @@ -654,9 +675,9 @@ struct inode *ext4_new_inode(handle_t *handle, struct inode *dir, umode_t mode,
> }
>
> if (S_ISDIR(mode))
> - ret2 = find_group_orlov(sb, dir, &group, mode, qstr);
> + ret2 = find_group_orlov(sb, dir, &group, mode, qstr, flags);
> else
> - ret2 = find_group_other(sb, dir, &group, mode);
> + ret2 = find_group_other(sb, dir, &group, mode, flags);
>
> got_group:
> EXT4_I(dir)->i_last_alloc_group = group;
> diff --git a/fs/ext4/migrate.c b/fs/ext4/migrate.c
> index f39f80f..2b3d65c 100644
> --- a/fs/ext4/migrate.c
> +++ b/fs/ext4/migrate.c
> @@ -469,7 +469,7 @@ int ext4_ext_migrate(struct inode *inode)
> owner[0] = inode->i_uid;
> owner[1] = inode->i_gid;
> tmp_inode = ext4_new_inode(handle, inode->i_sb->s_root->d_inode,
> - S_IFREG, NULL, goal, owner);
> + S_IFREG, NULL, goal, owner, 0);
> if (IS_ERR(tmp_inode)) {
> retval = PTR_ERR(tmp_inode);
> ext4_journal_stop(handle);
> diff --git a/fs/ext4/namei.c b/fs/ext4/namei.c
> index 6f48ff8..222a419 100644
> --- a/fs/ext4/namei.c
> +++ b/fs/ext4/namei.c
> @@ -1742,6 +1742,7 @@ static int ext4_create(struct inode *dir, struct dentry *dentry, umode_t mode,
> handle_t *handle;
> struct inode *inode;
> int err, retries = 0;
> + int flags = 0;
>
> dquot_initialize(dir);
>
> @@ -1755,7 +1756,13 @@ retry:
> if (IS_DIRSYNC(dir))
> ext4_handle_sync(handle);
>
> - inode = ext4_new_inode(handle, dir, mode, &dentry->d_name, 0, NULL);
> + if (op && op->open_flag & O_HOT)
> + flags |= EXT4_NEWI_HOT;
> + if (op && op->open_flag & O_COLD)
> + flags |= EXT4_NEWI_COLD;
> +
> + inode = ext4_new_inode(handle, dir, mode, &dentry->d_name, 0,
> + NULL, flags);
> err = PTR_ERR(inode);
> if (!IS_ERR(inode)) {
> inode->i_op = &ext4_file_inode_operations;
> @@ -1791,7 +1798,7 @@ retry:
> if (IS_DIRSYNC(dir))
> ext4_handle_sync(handle);
>
> - inode = ext4_new_inode(handle, dir, mode, &dentry->d_name, 0, NULL);
> + inode = ext4_new_inode(handle, dir, mode, &dentry->d_name, 0, NULL, 0);
> err = PTR_ERR(inode);
> if (!IS_ERR(inode)) {
> init_special_inode(inode, inode->i_mode, rdev);
> @@ -1831,7 +1838,7 @@ retry:
> ext4_handle_sync(handle);
>
> inode = ext4_new_inode(handle, dir, S_IFDIR | mode,
> - &dentry->d_name, 0, NULL);
> + &dentry->d_name, 0, NULL, 0);
> err = PTR_ERR(inode);
> if (IS_ERR(inode))
> goto out_stop;
> @@ -2278,7 +2285,7 @@ retry:
> ext4_handle_sync(handle);
>
> inode = ext4_new_inode(handle, dir, S_IFLNK|S_IRWXUGO,
> - &dentry->d_name, 0, NULL);
> + &dentry->d_name, 0, NULL, 0);
> err = PTR_ERR(inode);
> if (IS_ERR(inode))
> goto out_stop;
--
To unsubscribe from this list: send the line "unsubscribe linux-ext4" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
