On Mon, 22 Jun 1998 18:26:30 -0600 (MDT), Colin Plumb <colin@nyx.net>
said:
> Building a fake device out of bits of real devices is not that complicated.
> The RAID code does this and the file system doesn't even need to know about
> it.
> The tricky part comes when you want to add or remove real devices, because
> then your fake device changes size, and the file system needs to know
> about *that*.
Correct. There is in fact so much filesystem interaction required that
I'm not at all convinced that a block-device LVM is needed or even
useful. Virtual disks for redundancy or performance are just fine, but
when it comes to filesystem sizing, the fs has to be actively involved
in any change. Given that, we can actually implement the whole thing in
the filesystem.
Miguel's prototype LVM stuff works by letting you mke2fs a new partition
and then daisy-chain that new device on to the end of the existing
filesystem, at run time, while it is all mounted. Removing such a
partition from the middle of a logical volume set is harder but
certainly feasible in theory. Is there really any overwhelming
justification for needing extra device-level support for this
functionality? Given that we _need_ filesystem support, my own reaction
is that splitting the support between fs and block device just
complicates the matter; it's better just to do it in one place.
--Stephen
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu