Re: FreeBSD-11.0-BETA1-amd64-disc1.iso is too big for my 700MB CD-r

From: Glen Barber <gjb_at_FreeBSD.org>
Date: Wed, 13 Jul 2016 14:06:03 +0000
On Wed, Jul 13, 2016 at 01:52:13PM +0000, Glen Barber wrote:
> On Mon, Jul 11, 2016 at 05:23:32PM -0700, Maxim Sobolev wrote:
> > P.S. Just in case if somebody wants to integrate this method into FreeBSD
> > liveCD build, we do a bit of trick there by making normal ISO9660 file
> > system with compressed kernel and relevant boot pieces and then also
> > sticking in BSD label on the same disk image. It turns out ISO9660 and BSD
> > disklabel structures do not overlap, so it works nicely since 2005 or
> > so. Then we append UFS image compressed with mkuzip at the end of it.
> > Resulting image can be used just as any ISO would. We also cook up UFS with
> > unique label and then use GEOM_LABEL to easily find relevant file system on
> > boot regardless of the physical device name.
> > 
> >   mkuzip -dL -S -s 65536 -o ${UZPFILE} ${UFSFILE}
> >   mkisofs -b boot/${CDBOOT} -no-emul-boot -r -o ${ISOFILE} ${CDIR}
> >   eval $(stat -s ${UZPFILE})
> >   UZPSIZE=$((st_size + 2048 - (st_size % 2048)))
> >   truncate -s ${UZPSIZE} ${UZPFILE}
> >   eval $(stat -s ${ISOFILE})
> >   ISOSIZE=${st_size}
> >   echo "bytes/sector:  2048"                                            >
> > ${TDIR}/label.txt
> >   echo "sectors/unit:  $(((UZPSIZE + ISOSIZE) / 2048))"                >>
> > ${TDIR}/label.txt
> >   echo "a:             $((UZPSIZE / 2048)) $((ISOSIZE / 2048)) unused" >>
> > ${TDIR}/label.txt
> >   echo "c: $(((UZPSIZE + ISOSIZE) / 2048))                   0 unused" >>
> > ${TDIR}/label.txt
> >   truncate -s $((ISOSIZE + UZPSIZE)) ${ISOFILE}
> >   disklabel -A -R -f ${ISOFILE} ${TDIR}/label.txt
> >   truncate -s ${ISOSIZE} ${ISOFILE}
> >   cat ${UZPFILE} >> ${ISOFILE}
> > 
> 
> I'm very interested in implementing this, provided it solves the
> disc1.iso size issue.  Could you provide a bit explanation on what the
> variables above represent?  At the moment, I'm having trouble parsing
> the file paths and input/output paths.
> 

For now, I think ignore my reply.  I think I now see what you're doing.

Glen


Received on Wed Jul 13 2016 - 12:06:04 UTC

This archive was generated by hypermail 2.4.0 : Wed May 19 2021 - 11:41:06 UTC