Just remarks

jsmaby at virgo.umeche.maine.edu jsmaby at virgo.umeche.maine.edu
Tue May 28 06:04:49 PDT 2002


> Just a thought: why do you need paper when you have linux? :-) I think
> that  if we are going to use these SBU's everywhere, maybe it wouldn't
> hurt  to  prepend  "time" to the static bash installation instructions

how about this:

begin=$(date +%s)
./configure --enable-static-link \
     --prefix=$LFS/static --with-curses &&
 make &&
 make install
end=$(date +%s)
export SBU=$(echo $[$end-$begin])
echo "Your SBU is $SBU seconds"

Now one wouldn't need to break out the calculator, just do
echo $[$SBU * 12]
to see how long gcc should take.

About rounding off, I suggest we keep two significant figures.
So that binutils would be 2.0 SBU instead of 2.  At least in
the scientific world, this would mean 2.0+/-0.1 instead of 2+/-1
which I think is a little closer to what we can expect for
acuracy.  A little extra precision never hurts.

Now when people calculate thier SBUs for all the packages, maybe
we should standardize the system build with as LFS-cvs, or rather
gcc-3.1, and no added optimizations.  Otherwise, the SBU will be
calculated with 2.95.3 (in my case at least), and compared with
what 3.1 gives in chroot.  One could do chapter 5 over after chapter
6 is done to get the numbers.  Or at least use one bash compile time
for chapter 5 (with host gcc), and compile a static bash after ch6
for its packages.

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



More information about the lfs-dev mailing list