Greetings,
With the current way jHALFS creates the Makefile the following
functions when creating the Makefile are clobbering the -64bit, -n32
at the end of the filenames for some reason.
#----------------------------------#
LUSER_wrt_LogNewFiles() { #
#----------------------------------#
local name=`echo ${1} | sed 's/[0-9]-//'`
(
cat << EOF
@\$(call log_new_files_LUSER,$name)
EOF
) >> $MKFILE.tmp
}
#----------------------------------#
CHROOT_wrt_LogNewFiles() { #
#----------------------------------#
local name=`echo ${1} | sed 's/[0-9]-//'`
(
cat << EOF
@\$(call log_new_files,$name)
EOF
) >> $MKFILE.tmp
}
both eglibc and eglibc-64bit become eglibc and so I only have the log
of the 64bit installs.
It has been like this for a while, I just noticed it today and never
gave a thought about it back when building the install-logs for CLFS
2.0 and 2.1.
I added this to the script to see what was being used:
echo ${1}
echo ${1} | sed 's/[0-9]-//'
Output was this:
perl
perl
eglibc
eglibc
eglibc
eglibc
So clearly -64bit and -n32 is not being used at all.
It should have been in th eoutput from the first echo and second echo.
Sincerely,
William Harrington
--
http://linuxfromscratch.org/mailman/listinfo/alfs-discuss
FAQ: http://www.linuxfromscratch.org/faq/
Unsubscribe: See the above information page
