| To: | David Chinner <dgc@xxxxxxx> |
|---|---|
| Subject: | Re: preallocation |
| From: | Ying-Hung Chen <ying@xxxxxxxxxxxxxx> |
| Date: | Thu, 29 Sep 2005 17:22:43 +0800 |
| Cc: | linux-xfs@xxxxxxxxxxx, Ying-Hung Chen <ying@xxxxxxxxxxxxxx> |
| In-reply-to: | <20050929065949.GC8727140@melbourne.sgi.com> |
| References: | <4337ADB7.3060700@yingternet.com> <43381317.8040007@sgi.com> <433B8B8F.9010300@yingternet.com> <20050929065949.GC8727140@melbourne.sgi.com> |
| Sender: | linux-xfs-bounce@xxxxxxxxxxx |
| User-agent: | Mozilla Thunderbird 1.0.6 (Windows/20050716) |
> It truncates it. IOWS, it throws away the prealloc you did externally
> to this program. You need to do the preallocation after the truncation
> but before you start writing.
>
> Cheers,
>
> Dave.
Thanks Dave, now it seems to work!! just out of curiosity, when i run
xfs_bmap on each file, it shows
00001.ivf:
0: [0..106495]: 96..106591
which is cool,
but when i do xfs_db -r /dev/hdb4
and check the fragementation status, it shows
[root@localhost tmp]# xfs_db -r /dev/hdb4
xfs_db> frag
actual 1763, ideal 902, fragmentation factor 48.84%
where are all the fragments coming from? do I need to worry about it?
thanks,
-Ying
|
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: preallocation, David Chinner |
|---|---|
| Next by Date: | get allocated filesize after preallocate, Ying-Hung Chen |
| Previous by Thread: | Re: preallocation, David Chinner |
| Next by Thread: | Re: preallocation, Eric Sandeen |
| Indexes: | [Date] [Thread] [Top] [All Lists] |