Announcement

Collapse
No announcement yet.

XFS Adds Shared Data Extents For Linux 4.9

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

  • liam
    replied
    Originally posted by pal666 View Post
    so, your post should be interpreted like "now it is just eight years behind btrfs, nothing excited" ?
    how does that compare to you quote "am looking forward, instead, to upcoming xfs changes (snapshots! data scrubbing! COW!)" ?
    ???
    OK, I am genuinely confused as to why what I said is a problem.
    Did you think I was being sarcastic?
    To be quite clear: my post should be interpreted as saying that xfs has taken a very large step forward, and I'm looking forward to the future work which required this to land first.

    Leave a comment:


  • pal666
    replied
    Originally posted by liam View Post
    I didn't say "Hey everyone! Start using all the new xfs features!!!!"
    so, your post should be interpreted like "now it is just eight years behind btrfs, nothing excited" ?
    how does that compare to you quote "am looking forward, instead, to upcoming xfs changes (snapshots! data scrubbing! COW!)" ?

    Leave a comment:


  • liam
    replied
    Originally posted by pal666 View Post
    "It warns at mount time as an experimental feature and that it may eat data"

    i guess some people tired of btrfs to the point of switching off their brains completely
    Are you THAT interested in trolling?
    I didn't say "Hey everyone! Start using all the new xfs features!!!!"
    Chinner, as I've said before, is quite conservative about these things and usually places potential data eating changes behind such flags.

    Leave a comment:


  • pal666
    replied
    Originally posted by liam View Post
    This didn't take long:-)
    Now, waiting for the checksum work to land and we've got ourselves a pretty damn good fs.
    "It warns at mount time as an experimental feature and that it may eat data"

    i guess some people tired of btrfs to the point of switching off their brains completely

    Leave a comment:


  • liam
    replied
    This didn't take long:-)
    Now, waiting for the checksum work to land and we've got ourselves a pretty damn good fs.

    Edit: wanted to be clear that I'm referring to METADATA checksum
    Last edited by liam; 13 October 2016, 06:11 AM.

    Leave a comment:


  • starshipeleven
    replied
    Originally posted by Imroy View Post
    Will this allow deduplication, or does XFS already support that?
    afaik xfs does not have deduplication, this should go the way of enabling it.

    It seems like the tool duperemove (tool used to deduplicate data in btrfs) added support for xfs deduplication, in case someone wants to test this. https://github.com/markfasheh/duperemove

    Leave a comment:


  • Imroy
    replied
    Will this allow deduplication, or does XFS already support that?

    Leave a comment:


  • Tobu
    replied
    This is incredibly cool. There'll be a reflink ioctl building on this, and it paves the way for deduplication support.

    Leave a comment:


  • RahulSundaram
    replied
    Originally posted by pheldens View Post
    So whats the practical advantage in basic language?
    Essentially these are the underlying features that one can use to build user facing functionality including snapshots. XFS will become more competitive in features to Btrfs/ZFS.

    Leave a comment:


  • pheldens
    replied
    So whats the practical advantage in basic language?

    Leave a comment:

Working...
X