| blob | b1bac81558f52408d07247ef4a82dec59188aad6 |
1 Linux kernel release 2.2.xx
3 These are the release notes for Linux version 2.2. Read them carefully,
4 as they tell you what this is all about, explain how to install the
5 kernel, and what to do if something goes wrong.
7 However, please make sure you don't ask questions which are already answered
8 in various files in the Documentation directory. See DOCUMENTATION below.
10 WHAT IS LINUX?
12 Linux is a Unix clone written from scratch by Linus Torvalds with
13 assistance from a loosely-knit team of hackers across the Net.
14 It aims towards POSIX compliance.
16 It has all the features you would expect in a modern fully-fledged
17 Unix, including true multitasking, virtual memory, shared libraries,
18 demand loading, shared copy-on-write executables, proper memory
19 management and TCP/IP networking.
21 It is distributed under the GNU General Public License - see the
22 accompanying COPYING file for more details.
24 ON WHAT HARDWARE DOES IT RUN?
26 Linux was first developed for 386/486-based PCs. These days it also
27 runs on ARMs, DEC Alphas, SUN Sparcs, M68000 machines (like Atari and
28 Amiga), MIPS and PowerPC, and others.
30 DOCUMENTATION:
32 - There is a lot of documentation available both in electronic form on
33 the Internet and in books, both Linux-specific and pertaining to
34 general UNIX questions. I'd recommend looking into the documentation
35 subdirectories on any Linux ftp site for the LDP (Linux Documentation
36 Project) books. This README is not meant to be documentation on the
37 system: there are much better sources available.
39 - There are various readme's in the kernel Documentation/ subdirectory:
40 these typically contain kernel-specific installation notes for some
41 drivers for example. See ./Documentation/00-INDEX for a list of what
42 is contained in each file. Please read the Changes file, as it
43 contains information about the problems, which may result by upgrading
44 your kernel.
46 INSTALLING the kernel:
48 - If you install the full sources, do a
50 cd /usr/src
51 gzip -cd linux-2.2.XX.tar.gz | tar xfv -
53 to get it all put in place. Replace "XX" with the version number of the
54 latest kernel.
56 - You can also upgrade between 2.2.xx releases by patching. Patches are
57 distributed in the traditional gzip and the new bzip2 format. To
58 install by patching, get all the newer patch files and do
60 cd /usr/src
61 gzip -cd patchXX.gz | patch -p0
63 or
64 cd /usr/src
65 bzip2 -dc patchXX.bz2 | patch -p0
67 (repeat xx for all versions bigger than the version of your current
68 source tree, _in_order_) and you should be ok. You may want to remove
69 the backup files (xxx~ or xxx.orig), and make sure that there are no
70 failed patches (xxx# or xxx.rej). If there are, either you or me has
71 made a mistake.
73 Alternatively, the script patch-kernel can be used to automate this
74 process. It determines the current kernel version and applies any
75 patches found.
77 cd /usr/src
78 linux/scripts/patch-kernel
80 The default directory for the kernel source is /usr/src/linux, but
81 can be specified as the first argument. Patches are applied from
82 the current directory, but an alternative directory can be specified
83 as the second argument.
85 - Make sure you have no stale .o files and dependencies lying around:
87 cd /usr/src/linux
88 make mrproper
90 You should now have the sources correctly installed.
92 SOFTWARE REQUIREMENTS
94 Compiling and running the 2.2.x kernels requires up-to-date
95 versions of various software packages. Consult
96 ./Documentation/Changes for the minimum version numbers required
97 and how to get updates for these packages. Beware that using
98 excessively old versions of these packages can cause indirect
99 errors that are very difficult to track down, so don't assume that
100 you can just update packages when obvious problems arise during
101 build or operation.
103 CONFIGURING the kernel:
105 - Do a "make config" to configure the basic kernel. "make config" needs
106 bash to work: it will search for bash in $BASH, /bin/bash and /bin/sh
107 (in that order), so one of those must be correct for it to work.
109 Do not skip this step even if you are only upgrading one minor
110 version. New configuration options are added in each release, and
111 odd problems will turn up if the configuration files are not set up
112 as expected. If you want to carry your existing configuration to a
113 new version with minimal work, use "make oldconfig", which will
114 only ask you for the answers to new questions.
116 - Alternate configuration commands are:
117 "make menuconfig" Text based color menus, radiolists & dialogs.
118 "make xconfig" X windows based configuration tool.
119 "make oldconfig" Default all questions based on the contents of
120 your existing ./.config file.
122 NOTES on "make config":
123 - having unnecessary drivers will make the kernel bigger, and can
124 under some circumstances lead to problems: probing for a
125 nonexistent controller card may confuse your other controllers
126 - compiling the kernel with "Processor type" set higher than 386
127 will result in a kernel that does NOT work on a 386. The
128 kernel will detect this on bootup, and give up.
129 - A kernel with math-emulation compiled in will still use the
130 coprocessor if one is present: the math emulation will just
131 never get used in that case. The kernel will be slightly larger,
132 but will work on different machines regardless of whether they
133 have a math coprocessor or not.
134 - the "kernel hacking" configuration details usually result in a
135 bigger or slower kernel (or both), and can even make the kernel
136 less stable by configuring some routines to actively try to
137 break bad code to find kernel problems (kmalloc()). Thus you
138 should probably answer 'n' to the questions for
139 "development", "experimental", or "debugging" features.
141 - Check the top Makefile for further site-dependent configuration
142 (default SVGA mode etc).
144 - Finally, do a "make dep" to set up all the dependencies correctly.
146 COMPILING the kernel:
148 - Make sure you have gcc-2.7.2 or newer available. It seems older gcc
149 versions can have problems compiling newer versions of Linux. This
150 is mainly because the older compilers can only generate "a.out"-format
151 executables. As of Linux 2.1.0, the kernel must be compiled as an
152 "ELF" binary. If you upgrade your compiler, remember to get the new
153 binutils package too (for as/ld/nm and company).
155 Please note that you can still run a.out user programs with this
156 kernel.
158 - Do a "make zImage" to create a compressed kernel image. If you want
159 to make a boot disk (without root filesystem or LILO), insert a floppy
160 in your A: drive, and do a "make zdisk". It is also possible to do
161 "make zlilo" if you have lilo installed to suit the kernel makefiles,
162 but you may want to check your particular lilo setup first.
164 - If your kernel is too large for "make zImage", use "make bzImage"
165 instead.
167 - If you configured any of the parts of the kernel as `modules', you
168 will have to do "make modules" followed by "make modules_install".
169 Read Documentation/modules.txt for more information. For example,
170 an explanation of how to use the modules is included there.
172 - Keep a backup kernel handy in case something goes wrong. This is
173 especially true for the development releases, since each new release
174 contains new code which has not been debugged. Make sure you keep a
175 backup of the modules corresponding to that kernel, as well. If you
176 are installing a new kernel with the same version number as your
177 working kernel, make a backup of your modules directory before you
178 do a "make modules_install".
180 - In order to boot your new kernel, you'll need to copy the kernel
181 image (found in /usr/src/linux/arch/i386/boot/zImage after compilation)
182 to the place where your regular bootable kernel is found.
184 For some, this is on a floppy disk, in which case you can "cp
185 /usr/src/linux/arch/i386/boot/zImage /dev/fd0" to make a bootable
186 floppy. Please note that you can not boot a kernel by
187 directly dumping it to a 720k double-density 3.5" floppy. In this
188 case, it is highly recommended that you install LILO on your
189 double-density boot floppy or switch to high-density floppies.
191 If you boot Linux from the hard drive, chances are you use LILO which
192 uses the kernel image as specified in the file /etc/lilo.conf. The
193 kernel image file is usually /vmlinuz, or /zImage, or /etc/zImage.
194 To use the new kernel, save a copy of the old image and copy the new
195 image over the old one. Then, you MUST RERUN LILO to update the
196 loading map!! If you don't, you won't be able to boot the new kernel
197 image.
199 Reinstalling LILO is usually a matter of running /sbin/lilo.
200 You may wish to edit /etc/lilo.conf to specify an entry for your
201 old kernel image (say, /vmlinux.old) in case the new one does not
202 work. See the LILO docs for more information.
204 After reinstalling LILO, you should be all set. Shutdown the system,
205 reboot, and enjoy!
207 If you ever need to change the default root device, video mode,
208 ramdisk size, etc. in the kernel image, use the 'rdev' program (or
209 alternatively the LILO boot options when appropriate). No need to
210 recompile the kernel to change these parameters.
212 - Reboot with the new kernel and enjoy.
214 IF SOMETHING GOES WRONG:
216 - If you have problems that seem to be due to kernel bugs, please check
217 the file MAINTAINERS to see if there is a particular person associated
218 with the part of the kernel that you are having trouble with. If there
219 isn't anyone listed there, then the second best thing is to mail
220 them to me (torvalds@transmeta.com), and possibly to any other
221 relevant mailing-list or to the newsgroup. The mailing-lists are
222 useful especially for SCSI and NETworking problems, as I can't test
223 either of those personally anyway.
225 - In all bug-reports, *please* tell what kernel you are talking about,
226 how to duplicate the problem, and what your setup is (use your common
227 sense). If the problem is new, tell me so, and if the problem is
228 old, please try to tell me when you first noticed it.
230 - If the bug results in a message like
232 unable to handle kernel paging request at address C0000010
233 Oops: 0002
234 EIP: 0010:XXXXXXXX
235 eax: xxxxxxxx ebx: xxxxxxxx ecx: xxxxxxxx edx: xxxxxxxx
236 esi: xxxxxxxx edi: xxxxxxxx ebp: xxxxxxxx
237 ds: xxxx es: xxxx fs: xxxx gs: xxxx
238 Pid: xx, process nr: xx
239 xx xx xx xx xx xx xx xx xx xx
241 or similar kernel debugging information on your screen or in your
242 system log, please duplicate it *exactly*. The dump may look
243 incomprehensible to you, but it does contain information that may
244 help debugging the problem. The text above the dump is also
245 important: it tells something about why the kernel dumped code (in
246 the above example it's due to a bad kernel pointer). More information
247 on making sense of the dump is in Documentation/oops-tracing.txt
249 - You can use the "ksymoops" program to make sense of the dump. Find
250 the C++ sources under the scripts/ directory to avoid having to do
251 the dump lookup by hand:
253 - In debugging dumps like the above, it helps enormously if you can
254 look up what the EIP value means. The hex value as such doesn't help
255 me or anybody else very much: it will depend on your particular
256 kernel setup. What you should do is take the hex value from the EIP
257 line (ignore the "0010:"), and look it up in the kernel namelist to
258 see which kernel function contains the offending address.
260 To find out the kernel function name, you'll need to find the system
261 binary associated with the kernel that exhibited the symptom. This is
262 the file 'linux/vmlinux'. To extract the namelist and match it against
263 the EIP from the kernel crash, do:
265 nm vmlinux | sort | less
267 This will give you a list of kernel addresses sorted in ascending
268 order, from which it is simple to find the function that contains the
269 offending address. Note that the address given by the kernel
270 debugging messages will not necessarily match exactly with the
271 function addresses (in fact, that is very unlikely), so you can't
272 just 'grep' the list: the list will, however, give you the starting
273 point of each kernel function, so by looking for the function that
274 has a starting address lower than the one you are searching for but
275 is followed by a function with a higher address you will find the one
276 you want. In fact, it may be a good idea to include a bit of
277 "context" in your problem report, giving a few lines around the
278 interesting one.
280 If you for some reason cannot do the above (you have a pre-compiled
281 kernel image or similar), telling me as much about your setup as
282 possible will help.
284 - Alternately, you can use gdb on a running kernel. (read-only; i.e. you
285 cannot change values or set break points.) To do this, first compile the
286 kernel with -g; edit arch/i386/Makefile appropriately, then do a "make
287 clean". You'll also need to enable CONFIG_PROC_FS (via "make config").
289 After you've rebooted with the new kernel, do "gdb vmlinux /proc/kcore".
290 You can now use all the usual gdb commands. The command to look up the
291 point where your system crashed is "l *0xXXXXXXXX". (Replace the XXXes
292 with the EIP value.)
294 gdb'ing a non-running kernel currently fails because gdb (wrongly)
295 disregards the starting offset for which the kernel is compiled.