Re: [PATCH 00/13] overlay filesystem: request for inclusion (v15)

[Jan Engelhardt]

On Thursday 2012-09-20 22:48, Miklos Szeredi wrote:
>
>> Miklos, how do you think about this?
>> 
>> Do you think UnionMount is totally gone?
>
>Unionmount provides almost the same functionality as overlayfs.  The big
>difference between the two is that unionmounts resides 100% in the VFS
>while 95% of overlayfs is plain filesystem code.  I think that's the
>biggest advantage: filesystem code is easier to maintain, has less
>impact on core complexity, etc.

The big advantage is actually that the unioned view is in a separate
namespace (vfsmount).

>Aufs provides much better filesystem semantics than either unionmounts
>or overlayfs.  But that does come at a price:
>
>aufs:   98 files changed, 29893 insertions(+), 7 deletions(-)
>overlayfs:  22 files changed, 2981 insertions(+), 10 deletions(-)

In two years time when sufficient user requests have come in,
overlayfs is likely to have wrong as much.
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH 00/13] overlay filesystem: request for inclusion (v15)

[J. R. Okajima]

Miklos Szeredi:
> Aufs provides much better filesystem semantics than either unionmounts
> or overlayfs.  But that does come at a price:
>
> aufs:   98 files changed, 29893 insertions(+), 7 deletions(-)
> overlayfs:  22 files changed, 2981 insertions(+), 10 deletions(-)

Yes, I have to admit that aufs grew up very big.
The fundamental (by design) difference between them is considering how
important the filesystem semantics is. You may dislike a big module,
but I am interested in how you (or someone else) will implement the
missing overlayfs features to keep the semantics. One approach is
implemented in aufs. I guess you will try another one. That is what I am
interesed.


J. R. Okajima
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH 00/13] overlay filesystem: request for inclusion (v15)

[Miklos Szeredi]

"J. R. Okajima"  writes:

> Miklos, how do you think about this?
> 
> Do you think UnionMount is totally gone?

Unionmount provides almost the same functionality as overlayfs.  The big
difference between the two is that unionmounts resides 100% in the VFS
while 95% of overlayfs is plain filesystem code.  I think that's the
biggest advantage: filesystem code is easier to maintain, has less
impact on core complexity, etc.

Aufs provides much better filesystem semantics than either unionmounts
or overlayfs.  But that does come at a price:

aufs:   98 files changed, 29893 insertions(+), 7 deletions(-)
overlayfs:  22 files changed, 2981 insertions(+), 10 deletions(-)

Thanks,
Miklos
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Re: [PATCH 00/13] overlay filesystem: request for inclusion (v15)

[J. R. Okajima]

Thank you for CC-ing me.

Miklos Szeredi:
> Please consider overlayfs for inclusion into 3.7.
>
> It's used by Ubuntu and by various other projects.  I regularly get emails
> asking when it will be included in mainline.

AUFS is in very same situation for years.
Miklos, how do you think about this?

Do you think UnionMount is totally gone?


> Git tree is here:
>
>   git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs.git overlayfs.v15

Years ago, I also switched the AUFS development base (and releases) to
GIT on SourceForge due to the request from Ubuntu people.


J. R. Okajima
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majord...@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

[PATCH 00/13] overlay filesystem: request for inclusion (v15)

[Miklos Szeredi]

Al and Linus,

Please consider overlayfs for inclusion into 3.7.

It's used by Ubuntu and by various other projects.  I regularly get emails
asking when it will be included in mainline.

Git tree is here:

  git://git.kernel.org/pub/scm/linux/kernel/git/mszeredi/vfs.git overlayfs.v15

Thanks,
Miklos

---
Andy Whitcroft (3):
  overlayfs: add statfs support
  ovl: switch to __inode_permission()
  overlayfs: copy up i_uid/i_gid from the underlying inode

Erez Zadok (1):
  overlayfs: implement show_options

Miklos Szeredi (6):
  vfs: add i_op->dentry_open()
  vfs: export do_splice_direct() to modules
  vfs: introduce clone_private_mount()
  overlay filesystem
  fs: limit filesystem stacking depth
  vfs: export __inode_permission() to modules

Neil Brown (1):
  overlay: overlay filesystem documentation

Robin Dong (2):
  overlayfs: fix possible leak in ovl_new_inode
  overlayfs: create new inode in ovl_link

---
 Documentation/filesystems/Locking   |2 +
 Documentation/filesystems/overlayfs.txt |  199 +
 Documentation/filesystems/vfs.txt   |7 +
 MAINTAINERS |7 +
 fs/Kconfig  |1 +
 fs/Makefile |1 +
 fs/ecryptfs/main.c  |7 +
 fs/internal.h   |5 -
 fs/namei.c  |   10 +-
 fs/namespace.c  |   18 +
 fs/open.c   |   23 +-
 fs/overlayfs/Kconfig|4 +
 fs/overlayfs/Makefile   |7 +
 fs/overlayfs/copy_up.c  |  385 +
 fs/overlayfs/dir.c  |  604 +++
 fs/overlayfs/inode.c|  372 +
 fs/overlayfs/overlayfs.h|   70 
 fs/overlayfs/readdir.c  |  566 +
 fs/overlayfs/super.c|  685 +++
 fs/splice.c |1 +
 include/linux/fs.h  |   14 +
 include/linux/mount.h   |3 +
 22 files changed, 2981 insertions(+), 10 deletions(-)
 create mode 100644 Documentation/filesystems/overlayfs.txt
 create mode 100644 fs/overlayfs/Kconfig
 create mode 100644 fs/overlayfs/Makefile
 create mode 100644 fs/overlayfs/copy_up.c
 create mode 100644 fs/overlayfs/dir.c
 create mode 100644 fs/overlayfs/inode.c
 create mode 100644 fs/overlayfs/overlayfs.h
 create mode 100644 fs/overlayfs/readdir.c
 create mode 100644 fs/overlayfs/super.c

--
Changes from v14 to v15

- fix compile error with usernamespace

- statfs: calculate correct f_namelen, fill f_type with own magic

- documentation fixes

--
Changes from v13 to v14

- update to 3.6

- copy i_uid/i_gid from the underlying inode (patch by Andy Whitcroft)

--
Changes from v12 to v13

- create new inode in ovl_link (patch by Robin Dong)

- switch to __inode_permission() (patch by Andy Whitcroft)

--
Changes from v11 to v12

- update to for-next of vfs tree

- split __dentry_open argument cleanup patch from vfs-add-i_op-open.patch

- change i_op->open and vfs_open so that they take "struct file *"

--
Changes from v10 to v11

- fix overlayfs over overlayfs

- improve stack use of lookup and readdir

- add limitations to documentation

- make lower mount read-only

- update permission and fsync to new API

--
Changes from v9 to v10

- prevent d_delete() from turning upperdentry negative (reported by
  Erez Zadok)

- show mount options in /proc/mounts and friends (patch by Erez Zadok)

- fix off-by-one error in readdir (reported by Jordi Pujol)

--
Changes from v8 to v9

- support xattr on tmpfs

- fix build after split-up

- fix remove after rename (reported by Jordi Pujol)

- fix rename failure case

--
Changes from v7 to v8:

- split overlayfs.c into smaller files

- fix locking for copy up (reported by Al Viro)

- locking analysis of copy up vs. directory rename added as a comment

- tested with lockdep, fixed one lock annotation

- other bug fixes

--
Changes from v6 to v7

- added patches from Felix Fietkau to fix deadlocks on jffs2

- optimized directory removal

- properly clean up after copy-up and other failures

-