lkml.org 
[lkml]   [2018]   [May]   [28]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    Patch in this message
    /
    From
    Subject[PATCH 4.4 017/268] do d_instantiate/unlock_new_inode combinations safely
    Date
    4.4-stable review patch.  If anyone has any objections, please let me know.

    ------------------

    From: Al Viro <viro@zeniv.linux.org.uk>

    commit 1e2e547a93a00ebc21582c06ca3c6cfea2a309ee upstream.

    For anything NFS-exported we do _not_ want to unlock new inode
    before it has grown an alias; original set of fixes got the
    ordering right, but missed the nasty complication in case of
    lockdep being enabled - unlock_new_inode() does
    lockdep_annotate_inode_mutex_key(inode)
    which can only be done before anyone gets a chance to touch
    ->i_mutex. Unfortunately, flipping the order and doing
    unlock_new_inode() before d_instantiate() opens a window when
    mkdir can race with open-by-fhandle on a guessed fhandle, leading
    to multiple aliases for a directory inode and all the breakage
    that follows from that.

    Correct solution: a new primitive (d_instantiate_new())
    combining these two in the right order - lockdep annotate, then
    d_instantiate(), then the rest of unlock_new_inode(). All
    combinations of d_instantiate() with unlock_new_inode() should
    be converted to that.

    Cc: stable@kernel.org # 2.6.29 and later
    Tested-by: Mike Marshall <hubcap@omnibond.com>
    Reviewed-by: Andreas Dilger <adilger@dilger.ca>
    Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
    Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

    ---
    fs/btrfs/inode.c | 16 ++++------------
    fs/dcache.c | 22 ++++++++++++++++++++++
    fs/ecryptfs/inode.c | 3 +--
    fs/ext2/namei.c | 6 ++----
    fs/ext4/namei.c | 6 ++----
    fs/f2fs/namei.c | 12 ++++--------
    fs/jffs2/dir.c | 12 ++++--------
    fs/jfs/namei.c | 12 ++++--------
    fs/nilfs2/namei.c | 6 ++----
    fs/reiserfs/namei.c | 12 ++++--------
    fs/udf/namei.c | 6 ++----
    fs/ufs/namei.c | 6 ++----
    include/linux/dcache.h | 1 +
    13 files changed, 54 insertions(+), 66 deletions(-)

    --- a/fs/btrfs/inode.c
    +++ b/fs/btrfs/inode.c
    @@ -6413,8 +6413,7 @@ static int btrfs_mknod(struct inode *dir
    goto out_unlock_inode;
    } else {
    btrfs_update_inode(trans, root, inode);
    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    }

    out_unlock:
    @@ -6489,8 +6488,7 @@ static int btrfs_create(struct inode *di
    goto out_unlock_inode;

    BTRFS_I(inode)->io_tree.ops = &btrfs_extent_io_ops;
    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);

    out_unlock:
    btrfs_end_transaction(trans, root);
    @@ -6633,12 +6631,7 @@ static int btrfs_mkdir(struct inode *dir
    if (err)
    goto out_fail_inode;

    - d_instantiate(dentry, inode);
    - /*
    - * mkdir is special. We're unlocking after we call d_instantiate
    - * to avoid a race with nfsd calling d_instantiate.
    - */
    - unlock_new_inode(inode);
    + d_instantiate_new(dentry, inode);
    drop_on_err = 0;

    out_fail:
    @@ -9789,8 +9782,7 @@ static int btrfs_symlink(struct inode *d
    goto out_unlock_inode;
    }

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);

    out_unlock:
    btrfs_end_transaction(trans, root);
    --- a/fs/dcache.c
    +++ b/fs/dcache.c
    @@ -1897,6 +1897,28 @@ struct dentry *d_instantiate_unique(stru

    EXPORT_SYMBOL(d_instantiate_unique);

    +/*
    + * This should be equivalent to d_instantiate() + unlock_new_inode(),
    + * with lockdep-related part of unlock_new_inode() done before
    + * anything else. Use that instead of open-coding d_instantiate()/
    + * unlock_new_inode() combinations.
    + */
    +void d_instantiate_new(struct dentry *entry, struct inode *inode)
    +{
    + BUG_ON(!hlist_unhashed(&entry->d_u.d_alias));
    + BUG_ON(!inode);
    + lockdep_annotate_inode_mutex_key(inode);
    + security_d_instantiate(entry, inode);
    + spin_lock(&inode->i_lock);
    + __d_instantiate(entry, inode);
    + WARN_ON(!(inode->i_state & I_NEW));
    + inode->i_state &= ~I_NEW;
    + smp_mb();
    + wake_up_bit(&inode->i_state, __I_NEW);
    + spin_unlock(&inode->i_lock);
    +}
    +EXPORT_SYMBOL(d_instantiate_new);
    +
    /**
    * d_instantiate_no_diralias - instantiate a non-aliased dentry
    * @entry: dentry to complete
    --- a/fs/ecryptfs/inode.c
    +++ b/fs/ecryptfs/inode.c
    @@ -287,8 +287,7 @@ ecryptfs_create(struct inode *directory_
    iput(ecryptfs_inode);
    goto out;
    }
    - unlock_new_inode(ecryptfs_inode);
    - d_instantiate(ecryptfs_dentry, ecryptfs_inode);
    + d_instantiate_new(ecryptfs_dentry, ecryptfs_inode);
    out:
    return rc;
    }
    --- a/fs/ext2/namei.c
    +++ b/fs/ext2/namei.c
    @@ -40,8 +40,7 @@ static inline int ext2_add_nondir(struct
    {
    int err = ext2_add_link(dentry, inode);
    if (!err) {
    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    return 0;
    }
    inode_dec_link_count(inode);
    @@ -267,8 +266,7 @@ static int ext2_mkdir(struct inode * dir
    if (err)
    goto out_fail;

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    out:
    return err;

    --- a/fs/ext4/namei.c
    +++ b/fs/ext4/namei.c
    @@ -2429,8 +2429,7 @@ static int ext4_add_nondir(handle_t *han
    int err = ext4_add_entry(handle, dentry, inode);
    if (!err) {
    ext4_mark_inode_dirty(handle, inode);
    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    return 0;
    }
    drop_nlink(inode);
    @@ -2669,8 +2668,7 @@ out_clear_inode:
    err = ext4_mark_inode_dirty(handle, dir);
    if (err)
    goto out_clear_inode;
    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    if (IS_DIRSYNC(dir))
    ext4_handle_sync(handle);

    --- a/fs/f2fs/namei.c
    +++ b/fs/f2fs/namei.c
    @@ -150,8 +150,7 @@ static int f2fs_create(struct inode *dir

    alloc_nid_done(sbi, ino);

    - d_instantiate(dentry, inode);
    - unlock_new_inode(inode);
    + d_instantiate_new(dentry, inode);

    if (IS_DIRSYNC(dir))
    f2fs_sync_fs(sbi->sb, 1);
    @@ -399,8 +398,7 @@ static int f2fs_symlink(struct inode *di
    err = page_symlink(inode, p_str, p_len);

    err_out:
    - d_instantiate(dentry, inode);
    - unlock_new_inode(inode);
    + d_instantiate_new(dentry, inode);

    /*
    * Let's flush symlink data in order to avoid broken symlink as much as
    @@ -454,8 +452,7 @@ static int f2fs_mkdir(struct inode *dir,

    alloc_nid_done(sbi, inode->i_ino);

    - d_instantiate(dentry, inode);
    - unlock_new_inode(inode);
    + d_instantiate_new(dentry, inode);

    if (IS_DIRSYNC(dir))
    f2fs_sync_fs(sbi->sb, 1);
    @@ -499,8 +496,7 @@ static int f2fs_mknod(struct inode *dir,

    alloc_nid_done(sbi, inode->i_ino);

    - d_instantiate(dentry, inode);
    - unlock_new_inode(inode);
    + d_instantiate_new(dentry, inode);

    if (IS_DIRSYNC(dir))
    f2fs_sync_fs(sbi->sb, 1);
    --- a/fs/jffs2/dir.c
    +++ b/fs/jffs2/dir.c
    @@ -207,8 +207,7 @@ static int jffs2_create(struct inode *di
    __func__, inode->i_ino, inode->i_mode, inode->i_nlink,
    f->inocache->pino_nlink, inode->i_mapping->nrpages);

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    return 0;

    fail:
    @@ -428,8 +427,7 @@ static int jffs2_symlink (struct inode *
    mutex_unlock(&dir_f->sem);
    jffs2_complete_reservation(c);

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    return 0;

    fail:
    @@ -573,8 +571,7 @@ static int jffs2_mkdir (struct inode *di
    mutex_unlock(&dir_f->sem);
    jffs2_complete_reservation(c);

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    return 0;

    fail:
    @@ -745,8 +742,7 @@ static int jffs2_mknod (struct inode *di
    mutex_unlock(&dir_f->sem);
    jffs2_complete_reservation(c);

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    return 0;

    fail:
    --- a/fs/jfs/namei.c
    +++ b/fs/jfs/namei.c
    @@ -178,8 +178,7 @@ static int jfs_create(struct inode *dip,
    unlock_new_inode(ip);
    iput(ip);
    } else {
    - unlock_new_inode(ip);
    - d_instantiate(dentry, ip);
    + d_instantiate_new(dentry, ip);
    }

    out2:
    @@ -313,8 +312,7 @@ static int jfs_mkdir(struct inode *dip,
    unlock_new_inode(ip);
    iput(ip);
    } else {
    - unlock_new_inode(ip);
    - d_instantiate(dentry, ip);
    + d_instantiate_new(dentry, ip);
    }

    out2:
    @@ -1058,8 +1056,7 @@ static int jfs_symlink(struct inode *dip
    unlock_new_inode(ip);
    iput(ip);
    } else {
    - unlock_new_inode(ip);
    - d_instantiate(dentry, ip);
    + d_instantiate_new(dentry, ip);
    }

    out2:
    @@ -1443,8 +1440,7 @@ static int jfs_mknod(struct inode *dir,
    unlock_new_inode(ip);
    iput(ip);
    } else {
    - unlock_new_inode(ip);
    - d_instantiate(dentry, ip);
    + d_instantiate_new(dentry, ip);
    }

    out1:
    --- a/fs/nilfs2/namei.c
    +++ b/fs/nilfs2/namei.c
    @@ -50,8 +50,7 @@ static inline int nilfs_add_nondir(struc
    {
    int err = nilfs_add_link(dentry, inode);
    if (!err) {
    - d_instantiate(dentry, inode);
    - unlock_new_inode(inode);
    + d_instantiate_new(dentry, inode);
    return 0;
    }
    inode_dec_link_count(inode);
    @@ -246,8 +245,7 @@ static int nilfs_mkdir(struct inode *dir
    goto out_fail;

    nilfs_mark_inode_dirty(inode);
    - d_instantiate(dentry, inode);
    - unlock_new_inode(inode);
    + d_instantiate_new(dentry, inode);
    out:
    if (!err)
    err = nilfs_transaction_commit(dir->i_sb);
    --- a/fs/reiserfs/namei.c
    +++ b/fs/reiserfs/namei.c
    @@ -687,8 +687,7 @@ static int reiserfs_create(struct inode
    reiserfs_update_inode_transaction(inode);
    reiserfs_update_inode_transaction(dir);

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    retval = journal_end(&th);

    out_failed:
    @@ -771,8 +770,7 @@ static int reiserfs_mknod(struct inode *
    goto out_failed;
    }

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    retval = journal_end(&th);

    out_failed:
    @@ -871,8 +869,7 @@ static int reiserfs_mkdir(struct inode *
    /* the above add_entry did not update dir's stat data */
    reiserfs_update_sd(&th, dir);

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    retval = journal_end(&th);
    out_failed:
    reiserfs_write_unlock(dir->i_sb);
    @@ -1186,8 +1183,7 @@ static int reiserfs_symlink(struct inode
    goto out_failed;
    }

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    retval = journal_end(&th);
    out_failed:
    reiserfs_write_unlock(parent_dir->i_sb);
    --- a/fs/udf/namei.c
    +++ b/fs/udf/namei.c
    @@ -611,8 +611,7 @@ static int udf_add_nondir(struct dentry
    if (fibh.sbh != fibh.ebh)
    brelse(fibh.ebh);
    brelse(fibh.sbh);
    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);

    return 0;
    }
    @@ -722,8 +721,7 @@ static int udf_mkdir(struct inode *dir,
    inc_nlink(dir);
    dir->i_ctime = dir->i_mtime = current_fs_time(dir->i_sb);
    mark_inode_dirty(dir);
    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    if (fibh.sbh != fibh.ebh)
    brelse(fibh.ebh);
    brelse(fibh.sbh);
    --- a/fs/ufs/namei.c
    +++ b/fs/ufs/namei.c
    @@ -38,8 +38,7 @@ static inline int ufs_add_nondir(struct
    {
    int err = ufs_add_link(dentry, inode);
    if (!err) {
    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    return 0;
    }
    inode_dec_link_count(inode);
    @@ -191,8 +190,7 @@ static int ufs_mkdir(struct inode * dir,
    if (err)
    goto out_fail;

    - unlock_new_inode(inode);
    - d_instantiate(dentry, inode);
    + d_instantiate_new(dentry, inode);
    return 0;

    out_fail:
    --- a/include/linux/dcache.h
    +++ b/include/linux/dcache.h
    @@ -236,6 +236,7 @@ extern seqlock_t rename_lock;
    * These are the low-level FS interfaces to the dcache..
    */
    extern void d_instantiate(struct dentry *, struct inode *);
    +extern void d_instantiate_new(struct dentry *, struct inode *);
    extern struct dentry * d_instantiate_unique(struct dentry *, struct inode *);
    extern int d_instantiate_no_diralias(struct dentry *, struct inode *);
    extern void __d_drop(struct dentry *dentry);

    \
     
     \ /
      Last update: 2018-05-28 13:24    [W:6.120 / U:0.304 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site