Creating sparse file on XFS and EXT3 has different results

[Greg Freemyer]

Apologies for the top post.  I simply forgot to not do that.

On Aug 5, 2011 8:11 AM, "Greg Freemyer" <greg.freemyer at gmail.com> wrote:
>
> There are no "specs" as to how a sparse file is handled in response to
writes.
>
> Sparse is mostly beneficial when the holes are very large.
>
> If an app really wants to have minimal on disk space, you should
pre-allocate space with fallocate.
>
> You may even need to hole punch after the writes.  Both xfs and ext4
support both fallocate and hole punching. (I don't know the userspace call
to hole punch.  I think its a relatively new feature for ext4.)
>
> Greg
>
> On Aug 4, 2011 10:16 PM, "Ashish Sangwan" <ashishsangwan2 at gmail.com>
wrote:

Greg
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.kernelnewbies.org/pipermail/kernelnewbies/attachments/20110805/3f6fe291/attachment.html