linux-toradex.git/fs/nilfs2/super.c, branch v3.0.29

nilfs2: always set back pointer to host inode in mapping->host

2011-05-10T13:21:57+00:00

In the current nilfs, page cache for btree nodes and meta data files
do not set a valid back pointer to the host inode in mapping->host.

This will change it so that every address space in nilfs uses
mapping->host to hold its host inode.

Signed-off-by: Ryusuke Konishi

nilfs2: implement resize ioctl

2011-05-10T13:21:46+00:00

This adds resize ioctl which makes online resize possible.

Signed-off-by: Ryusuke Konishi

nilfs2: add routine to move secondary super block

2011-05-10T13:21:45+00:00

After resizing the filesystem, the secondary super block must be moved
to a new location.  This adds a helper function for this.

Signed-off-by: Ryusuke Konishi

nilfs2: get rid of nilfs_sb_info structure

2011-03-09T02:54:26+00:00

This directly uses sb->s_fs_info to keep a nilfs filesystem object and
fully removes the intermediate nilfs_sb_info structure.  With this
change, the hierarchy of on-memory structures of nilfs will be
simplified as follows:

Before:
  super_block
       -> nilfs_sb_info
             -> the_nilfs
                   -> cptree --+-> nilfs_root (current file system)
                               +-> nilfs_root (snapshot A)
                               +-> nilfs_root (snapshot B)
                               :
             -> nilfs_sc_info (log writer structure)
After:
  super_block
       -> the_nilfs
             -> cptree --+-> nilfs_root (current file system)
                         +-> nilfs_root (snapshot A)
                         +-> nilfs_root (snapshot B)
                         :
             -> nilfs_sc_info (log writer structure)

The reason why we didn't design so from the beginning is because the
initial shape also differed from the above.  The early hierachy was
composed of "per-mount-point" super_block -> nilfs_sb_info pairs and a
shared nilfs object.  On the kernel 2.6.37, it was changed to the
current shape in order to unify super block instances into one per
device, and this cleanup became applicable as the result.

Signed-off-by: Ryusuke Konishi

nilfs2: use sb instance instead of nilfs_sb_info struct

2011-03-09T02:05:08+00:00

This replaces sbi uses with direct reference to sb instance.

Signed-off-by: Ryusuke Konishi

nilfs2: move next generation counter into nilfs object

2011-03-09T02:05:08+00:00

Moves s_next_generation counter and a spinlock protecting it to nilfs
object from nilfs_sb_info structure.

Signed-off-by: Ryusuke Konishi

nilfs2: move s_inode_lock and s_dirty_files into nilfs object

2011-03-09T02:05:07+00:00

Moves s_inode_lock spinlock and s_dirty_files list to nilfs object
from nilfs_sb_info structure.

Signed-off-by: Ryusuke Konishi

nilfs2: move parameters on nilfs_sb_info into nilfs object

2011-03-09T02:05:07+00:00

This moves four parameter variables on nilfs_sb_info s_resuid,
s_resgid, s_interval and s_watermark to the nilfs object.

Signed-off-by: Ryusuke Konishi

nilfs2: move mount options to nilfs object

2011-03-09T02:05:07+00:00

This moves mount_opt local variable to nilfs object from nilfs_sb_info
struct.

Signed-off-by: Ryusuke Konishi

mm: prevent concurrent unmap_mapping_range() on the same inode

2011-02-24T03:52:52+00:00

Michael Leun reported that running parallel opens on a fuse filesystem
can trigger a "kernel BUG at mm/truncate.c:475"

Gurudas Pai reported the same bug on NFS.

The reason is, unmap_mapping_range() is not prepared for more than
one concurrent invocation per inode.  For example:

  thread1: going through a big range, stops in the middle of a vma and
     stores the restart address in vm_truncate_count.

  thread2: comes in with a small (e.g. single page) unmap request on
     the same vma, somewhere before restart_address, finds that the
     vma was already unmapped up to the restart address and happily
     returns without doing anything.

Another scenario would be two big unmap requests, both having to
restart the unmapping and each one setting vm_truncate_count to its
own value.  This could go on forever without any of them being able to
finish.

Truncate and hole punching already serialize with i_mutex.  Other
callers of unmap_mapping_range() do not, and it's difficult to get
i_mutex protection for all callers.  In particular ->d_revalidate(),
which calls invalidate_inode_pages2_range() in fuse, may be called
with or without i_mutex.

This patch adds a new mutex to 'struct address_space' to prevent
running multiple concurrent unmap_mapping_range() on the same mapping.

[ We'll hopefully get rid of all this with the upcoming mm
  preemptibility series by Peter Zijlstra, the "mm: Remove i_mmap_mutex
  lockbreak" patch in particular.  But that is for 2.6.39 ]

Signed-off-by: Miklos Szeredi 
Reported-by: Michael Leun 
Reported-by: Gurudas Pai 
Tested-by: Gurudas Pai 
Acked-by: Hugh Dickins 
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds