About SBUs and disk usage.

Ken Moffat ken at linuxfromscratch.org
Mon Nov 7 13:33:34 PST 2005

On Mon, 7 Nov 2005, Jim Gifford wrote:

> The only chapters that can get SBU's is the final-system, due to our build 
> method. So I don't really think SBU's is going to be a good measurement 
> system here in cross-lfs.
> I'm all for dropping them altogether. Anyone have any suggestions.

  If you are on a _slow_ machine, an estimation of how long something 
will take is often welcome.  But we don't have anything useful to use 
in the native build.  I think Manuel suggested (some months ago) the 
time to build and test tcl : my experience is that with gcc-4 this is 
sufficiently different from the old SBU that it won't usefully carry 
through to b(c)lfs.

  For space, I like knowing how big things are going to be, but again it 
doesn't seem possible to do it for the cross-build, only once we are 
running native.

  But then, I'm a traditionalist in lfs terms - I like to run the 
testsuites, and I get worried about failures.

  das eine Mal als Tragödie, das andere Mal als Farce

More information about the cross-lfs mailing list