[rfc] use ext3 instead of ext2

Bryan Kadzban bryan at kadzban.is-a-geek.net
Fri Jul 16 04:01:52 PDT 2004

Sébastien Maerten wrote:
> Yesterday, I read this: 
> http://kerneltrap.org/node/view/3466?PHPSESSID=4d3cbf0a2a03eacb8ff88b6e57782483
> Which IMHO seems to imply that the last point preventing replacement
> of ext2 by ext3 is performance with low latency work.

Does low latency even matter with a preemptible kernel?

I was under the impression that it didn't, since the only things
adversely affecting preemption are held locks (in the sense that the
process that does the preempt can't get them) and preempt_disable calls,
but I could be wrong.

More information about the lfs-dev mailing list