Re: [PATCH] register_blkdev

Joel Becker (Joel.Becker@oracle.com)
Fri, 7 Mar 2003 18:42:50 -0800


On Fri, Mar 07, 2003 at 06:18:19PM -0800, Linus Torvalds wrote:
> And realize that these things are often limited by on-disk / wire
> representations. Some of which are easier to fix than others (ie, think
> about NFS servers running old versions of Linux).

Yeah, I was chatting with Peter about that last night (he was
advocating 64 bits for dev_t). For some things (NFSv2) we could merely
truncate the space. It only really matters if /dev is on NFS, and we
can perhaps say "You need NFSv3 or better if you want the larger dev
space". That's just one possible way to approach it. A system big
enough to have 5000 disks attached likely isn't getting /dev from an
NFSv2 server.
I'm so-so on the 64bit vs 32bit dev_t argument, but I'd bet we
only want to change it once for the internal representation. How we
handle external representations (on disk, over the wire) is a different
matter.

Joel

-- 

"The one important thing i have learned over the years is the difference between taking one's work seriously and taking one's self seriously. The first is imperative and the second is disastrous." -Margot Fonteyn

Joel Becker Senior Member of Technical Staff Oracle Corporation E-mail: joel.becker@oracle.com Phone: (650) 506-8127 - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/