fs/splice.c | 22 ++--
fs/unionfs/Makefile | 2 +-
fs/unionfs/commonfops.c | 248 +++++++++++++++++++-----------
fs/unionfs/copyup.c | 2 +
fs/unionfs/dentry.c | 4 +-
fs/unionfs/file.c | 214 ++++++++++++++++++++++++--
fs/unionfs/inode.c | 43 +++---
fs/unionfs/mmap.c | 386 +++++++++--------------------------------------
fs/unionfs/union.h | 6 +-
include/linux/fs.h | 5 +
10 files changed, 480 insertions(+), 452 deletions(-)
New commits:
commit 34c50b75c8f191671d7164fa21a82d13d6f2073a
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Wed Apr 23 19:41:56 2008 -0400
Unionfs: release 2.3.3
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit afa6c3389b3a8253a7ed425b4c111843e0fd493d
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Wed Apr 23 19:13:41 2008 -0400
Unionfs: lock parent dentry branch config in write
Ensure that branch configuration is available to file_revalidate should a
copyup be required.
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 56317c869cc6b6d91bfe5f500e7b065e1551b071
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Wed Apr 23 19:13:32 2008 -0400
Unionfs: set lower file to NULL in file_release
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 9e5e123cf670109b85926c39b461840c4a02681b
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Wed Apr 23 19:13:25 2008 -0400
Unionfs: maintain one-open-file invariant for non-directories
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 3b3f89b4978e47522bcd15991db4c45ea2263a54
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Wed Apr 23 19:13:05 2008 -0400
Unionfs: reorganize file_revalidate for un/locking callers
Also clean up deep nesting/indentation.
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit c01b130d2d2665b3b92a120335c3acca7a4b7fd2
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 13:26:15 2008 -0400
Unionfs: copyup only if file is being written to
Before, we used to copyup on an open(2) call which used flags implying
writing (O_RDWR, O_WRONLY, O_APPEND). This meant that a file being opened
for writing, then immediately closed (without actually writing to the file),
will still have been copied up. Now, we don't copyup such files in ->open,
but defer the copyup till later. [Bug #591].
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 7b39c8dfba173646e36fa5d1e8563b07105cd08b
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 13:26:15 2008 -0400
Unionfs: set append offset correctly for copied-up files
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit b93df78bff136f5712759717a6f7c719a4f4e17d
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 13:26:15 2008 -0400
Unionfs: don't dereference dentry without lower branches in d_release
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 6569fe7c23a4fb05adbcf439850e6412fa51fea1
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 13:26:15 2008 -0400
Unionfs: stop as soon as first writeable branch is found
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 5c8eb9bc29a8a859ead59e8f3fcb8e6def065705
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 13:26:15 2008 -0400
Unionfs: lock inode around calls to notify_change()
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 53c9ea86e9b2eb5232da9528cf3e272acd60129b
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 13:26:15 2008 -0400
Unionfs: copy lower times in fsync/fasync only when needed
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 687b48be4cf8950a8241b74341e0f22cf002b41c
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 13:27:15 2008 -0400
Unionfs: minor code cleanups
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit f28b9858d6246c9597a3283cfb3ccde4a173bc82
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 15:17:57 2008 -0400
Unionfs: implement vm_operations->fault
As per recommendations made at LSF'08, a stackable file system which does
not change the data across layers, should try to use vm_operations instead
of address_space_operations. This means we now have to implement out own
->read and ->write methods because we cannot rely on VFS helpers which
require us to have a ->readpage method. Either way there are two caveats:
(1) It's not possible currently to set inode->i_mapping->a_ops to NULL,
because too many code paths expect i_mapping to be non-NULL.
(2) a small/dummy ->readpage is still needed because generic_file_mmap,
which we used in unionfs_mmap, still check for the existence of the
->readpage method. These code paths may have to be changed to remove the
need for readpage().
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 4c2b114f3e33650a24f9fb6a2c79fda340278e6f
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 16:20:58 2008 -0400
Unionfs: implement sendfile directly
Must implement sendfile, because we can no longer rely on
generic_file_sendfile: it needs the ->readpage address_space_operation
implemented, which we no longer have.
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 0c9ceaa3ee6cac81d52e11b08f3fdf5cd35bbb75
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 16:20:17 2008 -0400
Unionfs: implement splice_read/write methods directly
Must implement splice_read/write directly, using VFS helpers, because we can
no longer rely on generic_file_splice_read/write: they need
address_space_operations implemented, which we no longer have.
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
commit 3dc68d6ba74a315c7e8ff707bd1568c7a22d95b9
Author: Erez Zadok <[EMAIL PROTECTED]>
Date: Tue Apr 15 14:30:10 2008 -0400
VFS: rename do_splice_to/from to vfs_splice_* and export symbols
A stackable file system which uses vm_ops->fault, and does not implement
address_space_operations, cannot use generic_file_splice_read/write, but has
to implement ->splice_read/write itself. These two helper functions are
very useful to such a module.
Signed-off-by: Erez Zadok <[EMAIL PROTECTED]>
_______________________________________________
unionfs-cvs mailing list: http://unionfs.filesystems.org/
[email protected]
http://www.fsl.cs.sunysb.edu/mailman/listinfo/unionfs-cvs
