Re: [PATCH 5/6 v2] ext4: Introduce FALLOC_FL_ZERO_RANGE flag for fallocate

From: "Lukáš Czerner" <lczerner@redhat.com>
To: jon ernst <jonernst07@gmail.com>
Cc: linux-fsdevel@vger.kernel.org,
	"linux-ext4@vger.kernel.org List" <linux-ext4@vger.kernel.org>,
	Theodore Ts'o <tytso@mit.edu>,
	xfs@oss.sgi.com
Subject: Re: [PATCH 5/6 v2] ext4: Introduce FALLOC_FL_ZERO_RANGE flag for fallocate
Date: Thu, 27 Feb 2014 12:56:54 +0100 (CET)	[thread overview]
Message-ID: <alpine.LFD.2.00.1402271249570.2247@localhost.localdomain> (raw)
In-Reply-To: <CAGW2f1FcoMJpFv=7xqPB4+R3-yOqmn89CnAZAJKif_3a3ETcPg@mail.gmail.com>

On Wed, 26 Feb 2014, jon ernst wrote:

> Date: Wed, 26 Feb 2014 23:41:15 -0500
> From: jon ernst <jonernst07@gmail.com>
> To: Lukas Czerner <lczerner@redhat.com>
> Cc: "linux-ext4@vger.kernel.org List" <linux-ext4@vger.kernel.org>,
>     Theodore Ts'o <tytso@mit.edu>, linux-fsdevel@vger.kernel.org,
>     xfs@oss.sgi.com
> Subject: Re: [PATCH 5/6 v2] ext4: Introduce FALLOC_FL_ZERO_RANGE flag for
>     fallocate
> 
> On Wed, Feb 26, 2014 at 1:00 AM, jon ernst <jonernst07@gmail.com> wrote:
> > On Tue, Feb 25, 2014 at 2:14 PM, Lukas Czerner <lczerner@redhat.com> wrote:
> >> Introduce new FALLOC_FL_ZERO_RANGE flag for fallocate. This has the same
> >> functionality as xfs ioctl XFS_IOC_ZERO_RANGE.
> >>
> >> It can be used to convert a range of file to zeros preferably without
> >> issuing data IO. Blocks should be preallocated for the regions that span
> >> holes in the file, and the entire range is preferable converted to
> >> unwritten extents
> >>
> >> This can be also used to preallocate blocks past EOF in the same way as
> >> with fallocate. Flag FALLOC_FL_KEEP_SIZE which should cause the inode
> >> size to remain the same.
> >>
> >> Also add appropriate tracepoints.
> >>
> >> Signed-off-by: Lukas Czerner <lczerner@redhat.com>
> >> ---
> >>  fs/ext4/ext4.h              |   2 +
> >>  fs/ext4/extents.c           | 270 +++++++++++++++++++++++++++++++++++++++++---
> >>  fs/ext4/inode.c             |  17 ++-
> >>  include/trace/events/ext4.h |  64 +++++------
> 
> >>  static int
> >> +ext4_ext_convert_initialized_extent(handle_t *handle, struct inode *inode,
> >> +                       struct ext4_map_blocks *map,
> >> +                       struct ext4_ext_path *path, int flags,
> >> +                       unsigned int allocated, ext4_fsblk_t newblock)
> >> +{
> >> +       int ret = 0;
> >> +       int err = 0;
> >> +
> >> +       /*
> >> +        * Make sure that the extent is no bigger than we support with
> >> +        * uninitialized extent
> >> +        */
> >> +       if (map->m_len > EXT_UNINIT_MAX_LEN)
> >> +               map->m_len = EXT_UNINIT_MAX_LEN / 2;
> >
>  Pardon my possible dumb question. Why do you use
> "EXT_UNINIT_MAX_LEN/ 2;" here instead of "EXT_UNINIT_MAX_LEN"
>  I don't see the reason why we can't use EXT_UNINIT_MAX_LEN here.
> 
> 
> (resend, Ping on this question, thank you!)

Wow, that's an early ping :) I am sorry to disappoint you, my answer
is not going to be that exciting :)

Yes, we can just use EXT_UNINIT_MAX_LEN here. But
EXT_UNINIT_MAX_LEN/2 would make it much more evenly spread out.

I do not think there is any real world advantage to this and the
behaviour should be the same in both cases.

Thanks!
-Lukas

> 
> Thanks!
>  Jon
> 

_______________________________________________
xfs mailing list
xfs@oss.sgi.com
http://oss.sgi.com/mailman/listinfo/xfs