On Fri, Jun 14, 2002 at 10:27:59AM -0800, Ethan Benson wrote:
> On Fri, Jun 14, 2002 at 01:23:50PM -0400, Charles Shannon Hendrix wrote:
> >
> > For many systems, especially those which lack large tape backup systems,
> > tree level exclusion is a necessity.
>
> that does not mean it needs to be done via an extended attribute.
I said that tree-level exclusion is necessary. I said nothing about
extended attributes.
> > XFS is already huge kernel bloat, so a few more K isn't going to matter.
>
> thats a terribly foolish attitude, and will ensure XFS would never go
> into the mainline kernel.
The changes being talked about might not even amount to a few K.
They could be very small. You won't know until it's done.
But otherwise, I'm only stating the facts: XFS is huge, and a few K more
or less is not likely to affect its inclusion into the kernel.
The Linux kernel on my system is 28MB of code compressed. A build
directory on my system is 180MB. XFS itself is over 5MB. I have trimmed
my kernel to only necessary components. The kernel is ~3MB with 2.1MB
of loadable modules. The days of a lean kernel have long been gone.
--
UNIX/Perl/C/Pizza__________________________________shannon@xxxxxxxxxxxxx
|