Thread (283 messages) 283 messages, 37 authors, 2007-07-12

Re: [PATCH 4/7][TAKE5] support new modes in fallocate

From: David Chinner <hidden>
Date: 2007-06-26 23:32:53
Also in: linux-fsdevel, linux-xfs, lkml

On Tue, Jun 26, 2007 at 11:42:50AM -0400, Andreas Dilger wrote:
On Jun 26, 2007  16:15 +0530, Amit K. Arora wrote:
quoted
On Mon, Jun 25, 2007 at 03:52:39PM -0600, Andreas Dilger wrote:
quoted
In XFS one of the (many) ALLOC modes is to zero existing data on allocate.
For ext4 all this would mean is calling ext4_ext_mark_uninitialized() on
each extent.  For some workloads this would be much faster than truncate
and reallocate of all the blocks in a file.
In ext4, we already mark each extent having preallocated blocks as
uninitialized. This is done as part of following code (which is part of
patch 5/7) in ext4_ext_get_blocks() :  
What I meant is that with XFS_IOC_ALLOCSP the previously-written data
is ZEROED OUT, unlike with fallocate() which leaves previously-written
data alone and only allocates in holes.

So, if you had a sparse file with some data in it:

     AAAAA         BBBBBB

fallocate() would allocate the holes:

00000AAAAA000000000BBBBBB00000000

XFS_IOC_ALLOCSP would overwrite everything:

000000000000000000000000000000000
No, it wouldn't. XFS_IOC_ALLOCSP would give you:


      AAAAA         BBBBBB00000000

because it only allocates the space between the old EOF and the new
EOF. Graphic demonstration - write 4k @ 4k, 4k @ 16k, allocsp out to 32k:

budgie:~ # xfs_io -f \
-c "pwrite 4096 4096" \
-c "pwrite 16384 4096" \
-c "bmap -vvp" \
-c "allocsp 32768 0" \
-c "bmap -vvp" \
/mnt/test/alfred
wrote 4096/4096 bytes at offset 4096
4 KiB, 1 ops; 0.0000 sec (108.507 MiB/sec and 27777.7778 ops/sec)
wrote 4096/4096 bytes at offset 16384
4 KiB, 1 ops; 0.0000 sec (260.417 MiB/sec and 66666.6667 ops/sec)
/mnt/test/alfred:
 EXT: FILE-OFFSET      BLOCK-RANGE      AG AG-OFFSET          TOTAL
   0: [0..7]:          hole                                       8
   1: [8..15]:         5226864..5226871  4 (1022160..1022167)     8
   2: [16..31]:        hole                                      16
   3: [32..39]:        5226888..5226895  4 (1022184..1022191)     8
/mnt/test/alfred:
 EXT: FILE-OFFSET      BLOCK-RANGE      AG AG-OFFSET          TOTAL
   0: [0..7]:          hole                                       8
   1: [8..15]:         5226864..5226871  4 (1022160..1022167)     8
   2: [16..31]:        hole                                      16
   3: [32..63]:        5226888..5226919  4 (1022184..1022215)    32
budgie:~ #

Cheers,

Dave.
-- 
Dave Chinner
Principal Engineer
SGI Australian Software Group
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help