3 GNU Linux-libre <http://linux-libre.fsfla.org>
4 ==============================================
6 These are the release notes for GNU Linux-libre. Read them carefully,
7 as they tell you what this is all about, explain how to install the
8 kernel, and what to do if something goes wrong.
10 What is GNU Linux-libre?
11 ------------------------
13 GNU Linux-libre is a Free version of the kernel Linux (see below),
14 suitable for use with the GNU Operating System in 100% Free
15 GNU/Linux-libre System Distributions.
16 http://www.gnu.org/distros/
18 It removes non-Free components from Linux, that are disguised as
19 source code or distributed in separate files. It also disables
20 run-time requests for non-Free components, shipped separately or as
21 part of Linux, and documentation pointing to them, so as to avoid
22 (Free-)baiting users into the trap of non-Free Software.
23 http://www.fsfla.org/anuncio/2010-11-Linux-2.6.36-libre-debait
25 Linux-libre started within the gNewSense GNU/Linux distribution.
26 It was later adopted by Jeff Moe, who coined its name, and in 2008
27 it became a project maintained by FSF Latin America. In 2012, it
28 became part of the GNU Project.
30 The GNU Linux-libre project takes a minimal-changes approach to
31 cleaning up Linux, making no effort to substitute components that
32 need to be removed with functionally equivalent Free ones.
33 Nevertheless, we encourage and support efforts towards doing so.
34 http://libreplanet.org/wiki/LinuxLibre:Devices_that_require_non-free_firmware
36 Our mascot is Freedo, a light-blue penguin that has just come out
37 of the shower. Although we like penguins, GNU is a much greater
38 contribution to the entire system, so its mascot deserves more
39 promotion. See our web page for their images.
40 http://linux-libre.fsfla.org/
42 If you are the author of an awesome program and want to join us in
43 writing Free (libre) Software, please consider making it an official
44 GNU program and become a GNU Maintainer. You can find instructions
45 on how to do so at https://www.gnu.org/help/evaluation. We look
46 forward to hacking with you! :)
51 Linux is a clone of the Unix kernel, written from scratch by
52 Linus Torvalds with assistance from a loosely-knit team of hackers across
53 the Net. It aims towards POSIX and Single UNIX Specification compliance.
55 It has all the features you would expect in a modern fully-fledged Unix kernel,
56 including true multitasking, virtual memory, shared libraries, demand
57 loading, shared copy-on-write executables, proper memory management,
58 and multistack networking including IPv4 and IPv6.
60 It is distributed under the GNU General Public License v2 - see the
61 accompanying COPYING file for more details.
63 On what hardware does it run?
64 -----------------------------
66 Although originally developed first for 32-bit x86-based PCs (386 or higher),
67 today Linux also runs on (at least) the Compaq Alpha AXP, Sun SPARC and
68 UltraSPARC, Motorola 68000, PowerPC, PowerPC64, ARM, Hitachi SuperH, Cell,
69 IBM S/390, MIPS, HP PA-RISC, Intel IA-64, DEC VAX, AMD x86-64 Xtensa, and
72 Linux is easily portable to most general-purpose 32- or 64-bit architectures
73 as long as they have a paged memory management unit (PMMU) and a port of the
74 GNU C compiler (gcc) (part of The GNU Compiler Collection, GCC). Linux has
75 also been ported to a number of architectures without a PMMU, although
76 functionality is then obviously somewhat limited.
77 Linux has also been ported to itself. You can now run the kernel as a
78 userspace application - this is called UserMode Linux (UML).
83 - There is a lot of documentation available both in electronic form on
84 the Internet and in books, both Linux-specific and pertaining to
85 general UNIX questions. I'd recommend looking into the documentation
86 subdirectories on any Linux FTP site for the LDP (Linux Documentation
87 Project) books. This README is not meant to be documentation on the
88 system: there are much better sources available.
90 - There are various README files in the Documentation/ subdirectory:
91 these typically contain kernel-specific installation notes for some
92 drivers for example. Please read the
93 :ref:`Documentation/process/changes.rst <changes>` file, as it
94 contains information about the problems, which may result by upgrading
97 Installing the kernel source
98 ----------------------------
100 - If you install the full sources, put the kernel tarball in a
101 directory where you have permissions (e.g. your home directory) and
104 xz -cd linux-5.x.tar.xz | tar xvf -
106 Replace "X" with the version number of the latest kernel.
108 Do NOT use the /usr/src/linux area! This area has a (usually
109 incomplete) set of kernel headers that are used by the library header
110 files. They should match the library, and not get messed up by
111 whatever the kernel-du-jour happens to be.
113 - You can also upgrade between 5.x releases by patching. Patches are
114 distributed in the xz format. To install by patching, get all the
115 newer patch files, enter the top level directory of the kernel source
116 (linux-5.x) and execute::
118 xz -cd ../patch-5.x.xz | patch -p1
120 Replace "x" for all versions bigger than the version "x" of your current
121 source tree, **in_order**, and you should be ok. You may want to remove
122 the backup files (some-file-name~ or some-file-name.orig), and make sure
123 that there are no failed patches (some-file-name# or some-file-name.rej).
124 If there are, either you or I have made a mistake.
126 Unlike patches for the 5.x kernels, patches for the 5.x.y kernels
127 (also known as the -stable kernels) are not incremental but instead apply
128 directly to the base 5.x kernel. For example, if your base kernel is 5.0
129 and you want to apply the 5.0.3 patch, you must not first apply the 5.0.1
130 and 5.0.2 patches. Similarly, if you are running kernel version 5.0.2 and
131 want to jump to 5.0.3, you must first reverse the 5.0.2 patch (that is,
132 patch -R) **before** applying the 5.0.3 patch. You can read more on this in
133 :ref:`Documentation/process/applying-patches.rst <applying_patches>`.
135 Alternatively, the script patch-kernel can be used to automate this
136 process. It determines the current kernel version and applies any
139 linux/scripts/patch-kernel linux
141 The first argument in the command above is the location of the
142 kernel source. Patches are applied from the current directory, but
143 an alternative directory can be specified as the second argument.
145 - Make sure you have no stale .o files and dependencies lying around::
150 You should now have the sources correctly installed.
152 Software requirements
153 ---------------------
155 Compiling and running the 5.x kernels requires up-to-date
156 versions of various software packages. Consult
157 :ref:`Documentation/process/changes.rst <changes>` for the minimum version numbers
158 required and how to get updates for these packages. Beware that using
159 excessively old versions of these packages can cause indirect
160 errors that are very difficult to track down, so don't assume that
161 you can just update packages when obvious problems arise during
164 Build directory for the kernel
165 ------------------------------
167 When compiling the kernel, all output files will per default be
168 stored together with the kernel source code.
169 Using the option ``make O=output/dir`` allows you to specify an alternate
170 place for the output files (including .config).
173 kernel source code: /usr/src/linux-5.x
174 build directory: /home/name/build/kernel
176 To configure and build the kernel, use::
178 cd /usr/src/linux-5.x
179 make O=/home/name/build/kernel menuconfig
180 make O=/home/name/build/kernel
181 sudo make O=/home/name/build/kernel modules_install install
183 Please note: If the ``O=output/dir`` option is used, then it must be
184 used for all invocations of make.
186 Configuring the kernel
187 ----------------------
189 Do not skip this step even if you are only upgrading one minor
190 version. New configuration options are added in each release, and
191 odd problems will turn up if the configuration files are not set up
192 as expected. If you want to carry your existing configuration to a
193 new version with minimal work, use ``make oldconfig``, which will
194 only ask you for the answers to new questions.
196 - Alternative configuration commands are::
198 "make config" Plain text interface.
200 "make menuconfig" Text based color menus, radiolists & dialogs.
202 "make nconfig" Enhanced text based color menus.
204 "make xconfig" Qt based configuration tool.
206 "make gconfig" GTK+ based configuration tool.
208 "make oldconfig" Default all questions based on the contents of
209 your existing ./.config file and asking about
213 Like above, but sets new symbols to their default
214 values without prompting.
216 "make defconfig" Create a ./.config file by using the default
217 symbol values from either arch/$ARCH/defconfig
218 or arch/$ARCH/configs/${PLATFORM}_defconfig,
219 depending on the architecture.
221 "make ${PLATFORM}_defconfig"
222 Create a ./.config file by using the default
224 arch/$ARCH/configs/${PLATFORM}_defconfig.
225 Use "make help" to get a list of all available
226 platforms of your architecture.
229 Create a ./.config file by setting symbol
230 values to 'y' as much as possible.
233 Create a ./.config file by setting symbol
234 values to 'm' as much as possible.
236 "make allnoconfig" Create a ./.config file by setting symbol
237 values to 'n' as much as possible.
239 "make randconfig" Create a ./.config file by setting symbol
240 values to random values.
242 "make localmodconfig" Create a config based on current config and
243 loaded modules (lsmod). Disables any module
244 option that is not needed for the loaded modules.
246 To create a localmodconfig for another machine,
247 store the lsmod of that machine into a file
248 and pass it in as a LSMOD parameter.
250 Also, you can preserve modules in certain folders
251 or kconfig files by specifying their paths in
254 target$ lsmod > /tmp/mylsmod
255 target$ scp /tmp/mylsmod host:/tmp
257 host$ make LSMOD=/tmp/mylsmod \
258 LMC_KEEP="drivers/usb:drivers/gpu:fs" \
261 The above also works when cross compiling.
263 "make localyesconfig" Similar to localmodconfig, except it will convert
264 all module options to built in (=y) options. You can
265 also preserve modules by LMC_KEEP.
267 "make kvmconfig" Enable additional options for kvm guest kernel support.
269 "make xenconfig" Enable additional options for xen dom0 guest kernel
272 "make tinyconfig" Configure the tiniest possible kernel.
274 You can find more information on using the Linux kernel config tools
275 in Documentation/kbuild/kconfig.rst.
277 - NOTES on ``make config``:
279 - Having unnecessary drivers will make the kernel bigger, and can
280 under some circumstances lead to problems: probing for a
281 nonexistent controller card may confuse your other controllers.
283 - A kernel with math-emulation compiled in will still use the
284 coprocessor if one is present: the math emulation will just
285 never get used in that case. The kernel will be slightly larger,
286 but will work on different machines regardless of whether they
287 have a math coprocessor or not.
289 - The "kernel hacking" configuration details usually result in a
290 bigger or slower kernel (or both), and can even make the kernel
291 less stable by configuring some routines to actively try to
292 break bad code to find kernel problems (kmalloc()). Thus you
293 should probably answer 'n' to the questions for "development",
294 "experimental", or "debugging" features.
299 - Make sure you have at least gcc 4.9 available.
300 For more information, refer to :ref:`Documentation/process/changes.rst <changes>`.
302 Please note that you can still run a.out user programs with this kernel.
304 - Do a ``make`` to create a compressed kernel image. It is also
305 possible to do ``make install`` if you have lilo installed to suit the
306 kernel makefiles, but you may want to check your particular lilo setup first.
308 To do the actual install, you have to be root, but none of the normal
309 build should require that. Don't take the name of root in vain.
311 - If you configured any of the parts of the kernel as ``modules``, you
312 will also have to do ``make modules_install``.
314 - Verbose kernel compile/build output:
316 Normally, the kernel build system runs in a fairly quiet mode (but not
317 totally silent). However, sometimes you or other kernel developers need
318 to see compile, link, or other commands exactly as they are executed.
319 For this, use "verbose" build mode. This is done by passing
320 ``V=1`` to the ``make`` command, e.g.::
324 To have the build system also tell the reason for the rebuild of each
325 target, use ``V=2``. The default is ``V=0``.
327 - Keep a backup kernel handy in case something goes wrong. This is
328 especially true for the development releases, since each new release
329 contains new code which has not been debugged. Make sure you keep a
330 backup of the modules corresponding to that kernel, as well. If you
331 are installing a new kernel with the same version number as your
332 working kernel, make a backup of your modules directory before you
333 do a ``make modules_install``.
335 Alternatively, before compiling, use the kernel config option
336 "LOCALVERSION" to append a unique suffix to the regular kernel version.
337 LOCALVERSION can be set in the "General Setup" menu.
339 - In order to boot your new kernel, you'll need to copy the kernel
340 image (e.g. .../linux/arch/x86/boot/bzImage after compilation)
341 to the place where your regular bootable kernel is found.
343 - Booting a kernel directly from a floppy without the assistance of a
344 bootloader such as LILO, is no longer supported.
346 If you boot Linux from the hard drive, chances are you use LILO, which
347 uses the kernel image as specified in the file /etc/lilo.conf. The
348 kernel image file is usually /vmlinuz, /boot/vmlinuz, /bzImage or
349 /boot/bzImage. To use the new kernel, save a copy of the old image
350 and copy the new image over the old one. Then, you MUST RERUN LILO
351 to update the loading map! If you don't, you won't be able to boot
352 the new kernel image.
354 Reinstalling LILO is usually a matter of running /sbin/lilo.
355 You may wish to edit /etc/lilo.conf to specify an entry for your
356 old kernel image (say, /vmlinux.old) in case the new one does not
357 work. See the LILO docs for more information.
359 After reinstalling LILO, you should be all set. Shutdown the system,
362 If you ever need to change the default root device, video mode,
363 etc. in the kernel image, use your bootloader's boot options
364 where appropriate. No need to recompile the kernel to change
367 - Reboot with the new kernel and enjoy.
369 If something goes wrong
370 -----------------------
372 - If you have problems that seem to be due to kernel bugs, please check
373 the file MAINTAINERS to see if there is a particular person associated
374 with the part of the kernel that you are having trouble with. If there
375 isn't anyone listed there, then the second best thing is to mail
376 them to me (torvalds@linux-foundation.org), and possibly to any other
377 relevant mailing-list or to the newsgroup.
379 - In all bug-reports, *please* tell what kernel you are talking about,
380 how to duplicate the problem, and what your setup is (use your common
381 sense). If the problem is new, tell me so, and if the problem is
382 old, please try to tell me when you first noticed it.
384 - If the bug results in a message like::
386 unable to handle kernel paging request at address C0000010
389 eax: xxxxxxxx ebx: xxxxxxxx ecx: xxxxxxxx edx: xxxxxxxx
390 esi: xxxxxxxx edi: xxxxxxxx ebp: xxxxxxxx
391 ds: xxxx es: xxxx fs: xxxx gs: xxxx
392 Pid: xx, process nr: xx
393 xx xx xx xx xx xx xx xx xx xx
395 or similar kernel debugging information on your screen or in your
396 system log, please duplicate it *exactly*. The dump may look
397 incomprehensible to you, but it does contain information that may
398 help debugging the problem. The text above the dump is also
399 important: it tells something about why the kernel dumped code (in
400 the above example, it's due to a bad kernel pointer). More information
401 on making sense of the dump is in Documentation/admin-guide/bug-hunting.rst
403 - If you compiled the kernel with CONFIG_KALLSYMS you can send the dump
404 as is, otherwise you will have to use the ``ksymoops`` program to make
405 sense of the dump (but compiling with CONFIG_KALLSYMS is usually preferred).
406 This utility can be downloaded from
407 https://www.kernel.org/pub/linux/utils/kernel/ksymoops/ .
408 Alternatively, you can do the dump lookup by hand:
410 - In debugging dumps like the above, it helps enormously if you can
411 look up what the EIP value means. The hex value as such doesn't help
412 me or anybody else very much: it will depend on your particular
413 kernel setup. What you should do is take the hex value from the EIP
414 line (ignore the ``0010:``), and look it up in the kernel namelist to
415 see which kernel function contains the offending address.
417 To find out the kernel function name, you'll need to find the system
418 binary associated with the kernel that exhibited the symptom. This is
419 the file 'linux/vmlinux'. To extract the namelist and match it against
420 the EIP from the kernel crash, do::
422 nm vmlinux | sort | less
424 This will give you a list of kernel addresses sorted in ascending
425 order, from which it is simple to find the function that contains the
426 offending address. Note that the address given by the kernel
427 debugging messages will not necessarily match exactly with the
428 function addresses (in fact, that is very unlikely), so you can't
429 just 'grep' the list: the list will, however, give you the starting
430 point of each kernel function, so by looking for the function that
431 has a starting address lower than the one you are searching for but
432 is followed by a function with a higher address you will find the one
433 you want. In fact, it may be a good idea to include a bit of
434 "context" in your problem report, giving a few lines around the
437 If you for some reason cannot do the above (you have a pre-compiled
438 kernel image or similar), telling me as much about your setup as
439 possible will help. Please read the :ref:`admin-guide/reporting-bugs.rst <reportingbugs>`
440 document for details.
442 - Alternatively, you can use gdb on a running kernel. (read-only; i.e. you
443 cannot change values or set break points.) To do this, first compile the
444 kernel with -g; edit arch/x86/Makefile appropriately, then do a ``make
445 clean``. You'll also need to enable CONFIG_PROC_FS (via ``make config``).
447 After you've rebooted with the new kernel, do ``gdb vmlinux /proc/kcore``.
448 You can now use all the usual gdb commands. The command to look up the
449 point where your system crashed is ``l *0xXXXXXXXX``. (Replace the XXXes
452 gdb'ing a non-running kernel currently fails because ``gdb`` (wrongly)
453 disregards the starting offset for which the kernel is compiled.