Re: /tmp in swap space

Aaron M. Ucko (amu@mit.edu)
27 May 1998 11:13:40 -0500


Benny Amorsen <amorsen@sscnet.com> writes:

> You are making a file containing just zeroes, and ext2fs is making a
> sparse file for you. A sparse file can be written at close to main
> memory speed, I would guess. There should not be much disk activity.
> Try /dev/urandom next time...

No, ext2fs only creates holes when you seek() over the relevant
blocks, which dd doesn't do; dd-ing from /dev/zero is a fast, reliable
way to create a file taking up as many real blocks as you wish. At
any rate, /dev/urandom has its own overhead, which would complicate
the issue.

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu