LFS on older Sparcs

Bill Maltby - LFS Related lfsbill at wlmcs.com
Tue Nov 5 11:16:16 PST 2002

On Tue, 5 Nov 2002, Yann E. MORIN wrote:
>On Tuesday 05 November 2002 14:29, you wrote:

> > Configure:	28 minutes
> > Compile:	24 minutes
> > Install:	<1 minute
> >                 --
> > Total:		53 minutes
> > So 1 SBU is just about an hour.
>That's what I feared... :-/ I think configure is longer than compiling

>I'm heading right now (well, rather next week-end) to a x-compile from
>my PII-350. That shall be faster (I hope!)

It will be. I have a IBM6x86-II (read that as Cryix) PR-300 (equivalent to
a P2-300 supposedly). With older IDE/ATAPI drives, LFS 3.3 cvs 20020521
took about 6.5 hours for a complete install (well, the target drive was a
10K rpm IBM).

LFS 4.0 cvs 20021006 took 8.5 hours, approximately. This is due to the
changes for gcc-3.2, and related, and maybe some due to the addition of



You may have some improvement by fiddling with the block size on the
target HD. Since the default block is 1024 and almost *all* the files
through chapter8 are *much* larger than that, try using blocksize of 2048
or 4096. Reduce the number of i-nodes accordingly. Set m=0 when doing the
mke2fs. Make sure your swap is large enough (192MB - 256MB). If you do df
-i on the current LFS HD, it will show your free i-nodes so you can
estimate how much to cut. *Do not* cut to closely - you need to have some
for work files and room for growth.

Bill Maltby
billm at wlmcs.com

Unsubscribe: send email to listar at linuxfromscratch.org
and put 'unsubscribe lfs-support' in the subject header of the message

More information about the lfs-support mailing list