Re: Block I/O Enchancements, 2.5.1-pre2

[Linus Torvalds <torvalds@transmeta.com>]

On Tue, 27 Nov 2001, Jeff Garzik wrote:
>
> Oh yeah, I meant to ask: do we get 64-bit inode numbers and 64-bit block
> numbers on x86 sometime in 2.5?

Well, the 64-bit sector number skeleton is already there in pre2..

We will probably _not_ get 64-bit page index numbers, though. I don't want
to make the page structure bigger/slower for very little gain. So the page
cache is probably going to be limited to about 44 bits (45+ if people
start doing large pages, which is probably worth it). So there would still
be partition/file limits on the order of 16-64 TB in the next few years.

(In a longer timeframe, assuming RAM keeps getting cheaper and cheaper,
and 64-bit computing starts hppening on PC's, a few years down the line we
can re-visit this - that particular transition is not going to be too
painful).

And yes, I realize that you can already build big arrays and use LVM etc
to make them be more than 16TB. I just do not think it's a problem yet,
and I'd rather cater to "normal" people than to peopel who can't bother
to partition their data at all.

		Linus