On Mon, Dec 23, 2002 at 11:20:25AM -0800, Tom Ball wrote:
> I also found 8.0 much slower for my smaller builds. One thing that's
> different in 8.0 is the ext3 filesystem, which I converted to when
> upgrading. By default it runs in ordered data mode, which causes writes
> to disk to happen much more frequently than on ext2 filesystems.
Yes, particularly on /tmp. Traditionally, ext2 was so fast that the
performance of /tmp was not an issue, but journalling changes that.
I've been using tmpfs on my systems without problems. There are a few
corner cases that have been cleaned up recently (see lkml), but in
practice it has not been a problem. tmpfs need never write to swap if
there is no memory pressure; it is a useful complement to the use of
journalling filesystems for truly persistent data. Remember that use
of meta-data only journalling poses a potential information leakage
(i.e., security) problem on a multiuser system, especially when used
on /tmp ...
To use tmpfs, one can just add a line like the following to /etc/fstab:
/tmp /tmp tmpfs defaults 0 0
See mount(8) in the man pages for details of the available options:
Mount options for tmpfs
The following parameters accept a suffix k, m or g for Ki, Mi, Gi
(binary kilo, mega and giga) and can be changed on remount.
size=nbytes
Override default size of the filesystem. The size is given in
bytes, and rounded down to entire pages. The default is half of
the memory.
nr_blocks=
Set number of blocks.
nr_inodes=
Set number of inodes.
mode= Set initial permissions of the root directory.
Regards,
Bill Rugolsky
--
Psyche-list mailing list
Psyche-list@redhat.com
https://listman.redhat.com/mailman/listinfo/psyche-list
