I dont think so ... i am working on AIX whole day so i know the
concept of an LVM. It was just an example as you may read - Sure
it implements many more features.
> not a mechanism by which filesystems are resized. Rather, it
> provides a way to GRACEFULLY combine multiple physical volumes
> into a virtual contiguous logical volume, and gives a STANDARD
> interface to applying alocation policies and even RAID.
Yep - And even this Ted thought of beeing integrated into the ext234
filesystem
> I only put some words in CAPS because I KNOW somebody is gonna
> yell in my ear telling about MD. Well, MD isn't as clean as LVM,
> and I never thought it was graceful and definitely NOT standard.
>
> > Id like to see the LVM and an resizeable filesystem etc but
> > this seems not to be the way it will go.
>
> Having support for LVM in the kernel is so minimal it isn't
> even funny. It is userland to the hilt. The patch to kernel
> (ftp://ftp.msede.com/pub/linux/lvm/lvm-2.1-patch) is a grand
> total of 5k, and the rest is userland utilities and a SUPER
> handy lvm library.
I know these LVM implementation and talked to heinz on the Linux
Kongress ... but ...
> I see no bloat, only value added. Changes NOTHING for those
> who don't want it.
>
> Oh well, I guess even core developers can be sticks in the mud.
> Who would have guessed...
I also heard the speak of Ted about resizeable ext2 end future
whith holes in the fs etc, letting ext2 handle the multi-PHYSICAL-volume
things etc.
This obsoletes the LVM abstraction layer which i think is not "the unix way"
i think of it like doing everything with small functional "LEGO bricks"
BTW: If you read into the kernel mailing list archive we had this discussion
before.
Flo
-- Florian.Lohoff@mediaWays.net +49-5241-80-7085 Good, Fast, Cheap: Pick any two (you can't have all three). (RFC 1925)- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.rutgers.edu Please read the FAQ at http://www.tux.org/lkml/