1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
|
#!/bin/sh
#
# Copyright (c) 1994-1999 The University of Utah and the Flux Group.
#
# This file is part of the OSKit Linux Boot Loader, which is free software,
# also known as "open source;" you can redistribute it and/or modify it under
# the terms of the GNU General Public License (GPL), version 2, as published
# by the Free Software Foundation (FSF).
#
# The OSKit is distributed in the hope that it will be useful, but WITHOUT ANY
# WARRANTY; without even the implied warranty of MERCHANTABILITY or FITNESS
# FOR A PARTICULAR PURPOSE. See the GPL for more details. You should have
# received a copy of the GPL along with the OSKit; see the file COPYING. If
# not, write to the FSF, 59 Temple Place #330, Boston, MA 02111-1307, USA.
#
# This little script uses GNU ld to build a Linux 16-bit compressed boot image
# from a set of boot modules.
#
# Usage: mklinuximage [options] [files ...]
#
# Options:
# -o filename
# Specify the output <filename> of the resulting Linux image. The
# default is "zImage" in the current directory.
# -stdin
# Take file specifications from stdin, in addition to those on the
# command line. This is useful when using another script to build
# a list of file specifications which can then be piped into this
# script.
# -save-temps
# Debugging option to save temporary files instead of deleting them
# after the image is built.
# [files ...]
# A list of file specifications in the format pathname1:pathname2,
# where pathname1 is the name of a local module which is placed in the
# boot image. If :pathname2 is provided, it specifies the name to give
# the module in the image. If :pathname2 is omitted, it defaults to
# pathname1.
#
# The first file specification is typically the name of the Oskit kernel.
#
bootdir=${BOOTDIR-@prefix@/lib/oskit}
cc=${CC-@CC@}
ld=${LD-@LD@}
modules=
outfile=zImage
savetemps=
# Parse the command-line options.
until [ $# -eq 0 ]
do
case "$1" in
-o ) outfile="$2"; shift; shift;;
-stdin) fromstdin="yes"; shift;;
-save-temps) savetemps="$1"; shift;;
* ) modules="$modules $1"; shift;;
esac
done
#
# Read the rest of the modules from stdin.
#
if [ "X${fromstdin}" != "X" ]; then
read mod
while [ "X${mod}" != "X" ];
do
modules="$modules $mod"
read mod
done
fi
# Wrap each of the input files in a .o file.
# At the same time, build an assembly language module
# containing a table describing the boot modules.
echo >$outfile.mods.S ".text; .long 0xf00baabb"
files=
for module in $modules; do
# Split out the associated string, if any.
file=`echo $module | sed -e 's,:.*$,,'`
string=`echo $module | sed -e 's,^[^:]*:,,'`
if test -z "$string"; then string=$file; fi
# Add this file to the list of files to be included,
# but only if it hasn't already been seen before
# (e.g., with a different attached string).
alreadythere=
for afile in $files; do
if [ $file = $afile ]; then alreadythere=yes; fi
done
if [ -z $alreadythere ]; then files="$files $file"; fi
# Convert all non-alphanumeric chars to underscores.
# The BFD binary input format will do the same thing
# to produce the symbol names that it "wraps around" the input files.
sym_name=`echo $module | sed -e 's,[^a-zA-Z0-9],_,g'`
echo >>$outfile.mods.S ".long _binary_$sym_name""_start"
echo >>$outfile.mods.S ".long _binary_$sym_name""_end"
echo >>$outfile.mods.S ".long cmdline_$sym_name"
echo >>$outfile.mods.S ".data"
echo >>$outfile.mods.S "cmdline_$sym_name: .ascii \"$string\\0\""
echo >>$outfile.mods.S ".text"
done
echo >>$outfile.mods.S ".long 0; .data; .align 4"
# Assemble the module vector file.
$cc -c -o $outfile.mods.o $outfile.mods.S || exit 1
# Link the module vector file with the boot module files.
# Use the binary bfd backend for both the input bmod files and the output file.
# There is no meaning to "entry point address" in the binary backend,
# but the linker doesn't know that, so it looks for _start and warns
# when it doesn't find it (and falls back to zero). To stifle the warning
# we can define the `_start' symbol on the command line; I think this
# works with all GNU ld versions.
$ld -Ttext 0 -defsym _start=0 -oformat binary -o $outfile.tmp \
$outfile.mods.o -format binary $files -format default \
|| exit 1
# Compress the whole output file as one big glob.
gzip -9 <$outfile.tmp >$outfile.tmp.gz
# Create the final boot image by tacking that onto the end of 'linuxboot.bin'.
cat $bootdir/linuxboot.bin $outfile.tmp.gz >$outfile
if test -z "$savetemps"; then
rm -f $outfile.mods.S $outfile.mods.o $outfile.tmp $outfile.tmp.gz
fi
exit 0
|
