[Top] [All Lists]

Re: DMAPI: DM attribute/region space allocation

To: James A Goodwin <jagoodwi@xxxxxxxxxx>, Dean Roehrich <roehrich@xxxxxxx>
Subject: Re: DMAPI: DM attribute/region space allocation
From: Nathan Scott <nathans@xxxxxxx>
Date: Tue, 10 Feb 2004 09:50:47 +1100
Cc: linux-xfs@xxxxxxxxxxx
In-reply-to: <200402091835.i19IZ29i6889939@xxxxxxxxxxxxxxxxxxxxxx>
References: <200402091835.i19IZ29i6889939@xxxxxxxxxxxxxxxxxxxxxx>
Sender: linux-xfs-bounce@xxxxxxxxxxx
User-agent: Mutt/1.5.3i
On Mon, Feb 09, 2004 at 12:35:02PM -0600, Dean Roehrich wrote:
> >From:  Nathan Scott <nathans@xxxxxxx>
> >
> >I believe SGIs HSM uses the XFS RESBLKS interface to reserve a set
> >amount of filesystem space, which is only available when allocating
> >extended attribute space for the root namespace (where the DMAPI
> >attributes reside).
> Yes.  Mentioned briefly in the xfs(5) manpage.
> Dunno if this works, but here's how it might be done...
>         xfs_fsop_resblks_t counts;
>         fd = open(device, ...);

(not the block device, the ioctl needs to get into XFS.)

>         counts.resblks = 100; /* experiment with it */
>         ioctl(fd, XFS_IOC_SET_RESBLKS, &counts);
> I believe you have to do that everytime you mount the filesystem.

Yes; there is also an xfs_io(8) interface for this - just point
it at the mount point or a file in the filesystem, and you can
get & set this value (and see available reserved space too).  I
played with making this a more generic interface once, to allow
root to have access to an extra chunk of space for allocations,
like other fs's do.  But I didn't get around to tidying it up,
testing it properly, etc (& the patch is long since lost before
someone asks for it ;) - but it wasn't too difficult iirc).



<Prev in Thread] Current Thread [Next in Thread>