heads' kernel


  1. GNU Linux-libre <http://linux-libre.fsfla.org>
  2. These are the release notes for GNU Linux-libre. Read them carefully,
  3. as they tell you what this is all about, explain how to install the
  4. kernel, and what to do if something goes wrong.
  5. WHAT IS GNU Linux-libre?
  6. GNU Linux-libre is a Free version of the kernel Linux (see below),
  7. suitable for use with the GNU Operating System in 100% Free
  8. GNU/Linux-libre System Distributions.
  9. http://www.gnu.org/distros/
  10. It removes non-Free components from Linux, that are disguised as
  11. source code or distributed in separate files. It also disables
  12. run-time requests for non-Free components, shipped separately or as
  13. part of Linux, and documentation pointing to them, so as to avoid
  14. (Free-)baiting users into the trap of non-Free Software.
  15. http://www.fsfla.org/anuncio/2010-11-Linux-2.6.36-libre-debait
  16. Linux-libre started within the gNewSense GNU/Linux distribution.
  17. It was later adopted by Jeff Moe, who coined its name, and in 2008
  18. it became a project maintained by FSF Latin America. In 2012, it
  19. became part of the GNU Project.
  20. The GNU Linux-libre project takes a minimal-changes approach to
  21. cleaning up Linux, making no effort to substitute components that
  22. need to be removed with functionally equivalent Free ones.
  23. Nevertheless, we encourage and support efforts towards doing so.
  24. http://libreplanet.org/wiki/LinuxLibre:Devices_that_require_non-free_firmware
  25. Our mascot is Freedo, a light-blue penguin that has just come out
  26. of the shower. Although we like penguins, GNU is a much greater
  27. contribution to the entire system, so its mascot deserves more
  28. promotion. See our web page for their images.
  29. http://linux-libre.fsfla.org/
  31. Linux is a clone of the Unix kernel, written from scratch by
  32. Linus Torvalds with assistance from a loosely-knit team of hackers across
  33. the Net. It aims towards POSIX and Single UNIX Specification compliance.
  34. It has all the features you would expect in a modern fully-fledged Unix kernel,
  35. including true multitasking, virtual memory, shared libraries, demand
  36. loading, shared copy-on-write executables, proper memory management,
  37. and multistack networking including IPv4 and IPv6.
  38. It is distributed under the GNU General Public License - see the
  39. accompanying COPYING file for more details.
  41. Although originally developed first for 32-bit x86-based PCs (386 or higher),
  42. today Linux also runs on (at least) the Compaq Alpha AXP, Sun SPARC and
  43. UltraSPARC, Motorola 68000, PowerPC, PowerPC64, ARM, Hitachi SuperH, Cell,
  44. IBM S/390, MIPS, HP PA-RISC, Intel IA-64, DEC VAX, AMD x86-64, AXIS CRIS,
  45. Xtensa, Tilera TILE, AVR32, ARC and Renesas M32R architectures.
  46. Linux is easily portable to most general-purpose 32- or 64-bit architectures
  47. as long as they have a paged memory management unit (PMMU) and a port of the
  48. GNU C compiler (gcc) (part of The GNU Compiler Collection, GCC). Linux has
  49. also been ported to a number of architectures without a PMMU, although
  50. functionality is then obviously somewhat limited.
  51. Linux has also been ported to itself. You can now run the kernel as a
  52. userspace application - this is called UserMode Linux (UML).
  54. - There is a lot of documentation available both in electronic form on
  55. the Internet and in books, both Linux-specific and pertaining to
  56. general UNIX questions. I'd recommend looking into the documentation
  57. subdirectories on any Linux FTP site for the LDP (Linux Documentation
  58. Project) books. This README is not meant to be documentation on the
  59. system: there are much better sources available.
  60. - There are various README files in the Documentation/ subdirectory:
  61. these typically contain kernel-specific installation notes for some
  62. drivers for example. See Documentation/00-INDEX for a list of what
  63. is contained in each file. Please read the Changes file, as it
  64. contains information about the problems, which may result by upgrading
  65. your kernel.
  66. - The Documentation/DocBook/ subdirectory contains several guides for
  67. kernel developers and users. These guides can be rendered in a
  68. number of formats: PostScript (.ps), PDF, HTML, & man-pages, among others.
  69. After installation, "make psdocs", "make pdfdocs", "make htmldocs",
  70. or "make mandocs" will render the documentation in the requested format.
  71. INSTALLING the kernel source:
  72. - If you install the full sources, put the kernel tarball in a
  73. directory where you have permissions (eg. your home directory) and
  74. unpack it:
  75. xz -cd linux-4.X.tar.xz | tar xvf -
  76. Replace "X" with the version number of the latest kernel.
  77. Do NOT use the /usr/src/linux area! This area has a (usually
  78. incomplete) set of kernel headers that are used by the library header
  79. files. They should match the library, and not get messed up by
  80. whatever the kernel-du-jour happens to be.
  81. - You can also upgrade between 4.x releases by patching. Patches are
  82. distributed in the xz format. To install by patching, get all the
  83. newer patch files, enter the top level directory of the kernel source
  84. (linux-4.X) and execute:
  85. xz -cd ../patch-4.x.xz | patch -p1
  86. Replace "x" for all versions bigger than the version "X" of your current
  87. source tree, _in_order_, and you should be ok. You may want to remove
  88. the backup files (some-file-name~ or some-file-name.orig), and make sure
  89. that there are no failed patches (some-file-name# or some-file-name.rej).
  90. If there are, either you or I have made a mistake.
  91. Unlike patches for the 4.x kernels, patches for the 4.x.y kernels
  92. (also known as the -stable kernels) are not incremental but instead apply
  93. directly to the base 4.x kernel. For example, if your base kernel is 4.0
  94. and you want to apply the 4.0.3 patch, you must not first apply the 4.0.1
  95. and 4.0.2 patches. Similarly, if you are running kernel version 4.0.2 and
  96. want to jump to 4.0.3, you must first reverse the 4.0.2 patch (that is,
  97. patch -R) _before_ applying the 4.0.3 patch. You can read more on this in
  98. Documentation/applying-patches.txt
  99. Alternatively, the script patch-kernel can be used to automate this
  100. process. It determines the current kernel version and applies any
  101. patches found.
  102. linux/scripts/patch-kernel linux
  103. The first argument in the command above is the location of the
  104. kernel source. Patches are applied from the current directory, but
  105. an alternative directory can be specified as the second argument.
  106. - Make sure you have no stale .o files and dependencies lying around:
  107. cd linux
  108. make mrproper
  109. You should now have the sources correctly installed.
  111. Compiling and running the 4.x kernels requires up-to-date
  112. versions of various software packages. Consult
  113. Documentation/Changes for the minimum version numbers required
  114. and how to get updates for these packages. Beware that using
  115. excessively old versions of these packages can cause indirect
  116. errors that are very difficult to track down, so don't assume that
  117. you can just update packages when obvious problems arise during
  118. build or operation.
  119. BUILD directory for the kernel:
  120. When compiling the kernel, all output files will per default be
  121. stored together with the kernel source code.
  122. Using the option "make O=output/dir" allow you to specify an alternate
  123. place for the output files (including .config).
  124. Example:
  125. kernel source code: /usr/src/linux-4.X
  126. build directory: /home/name/build/kernel
  127. To configure and build the kernel, use:
  128. cd /usr/src/linux-4.X
  129. make O=/home/name/build/kernel menuconfig
  130. make O=/home/name/build/kernel
  131. sudo make O=/home/name/build/kernel modules_install install
  132. Please note: If the 'O=output/dir' option is used, then it must be
  133. used for all invocations of make.
  134. CONFIGURING the kernel:
  135. Do not skip this step even if you are only upgrading one minor
  136. version. New configuration options are added in each release, and
  137. odd problems will turn up if the configuration files are not set up
  138. as expected. If you want to carry your existing configuration to a
  139. new version with minimal work, use "make oldconfig", which will
  140. only ask you for the answers to new questions.
  141. - Alternative configuration commands are:
  142. "make config" Plain text interface.
  143. "make menuconfig" Text based color menus, radiolists & dialogs.
  144. "make nconfig" Enhanced text based color menus.
  145. "make xconfig" X windows (Qt) based configuration tool.
  146. "make gconfig" X windows (GTK+) based configuration tool.
  147. "make oldconfig" Default all questions based on the contents of
  148. your existing ./.config file and asking about
  149. new config symbols.
  150. "make silentoldconfig"
  151. Like above, but avoids cluttering the screen
  152. with questions already answered.
  153. Additionally updates the dependencies.
  154. "make olddefconfig"
  155. Like above, but sets new symbols to their default
  156. values without prompting.
  157. "make defconfig" Create a ./.config file by using the default
  158. symbol values from either arch/$ARCH/defconfig
  159. or arch/$ARCH/configs/${PLATFORM}_defconfig,
  160. depending on the architecture.
  161. "make ${PLATFORM}_defconfig"
  162. Create a ./.config file by using the default
  163. symbol values from
  164. arch/$ARCH/configs/${PLATFORM}_defconfig.
  165. Use "make help" to get a list of all available
  166. platforms of your architecture.
  167. "make allyesconfig"
  168. Create a ./.config file by setting symbol
  169. values to 'y' as much as possible.
  170. "make allmodconfig"
  171. Create a ./.config file by setting symbol
  172. values to 'm' as much as possible.
  173. "make allnoconfig" Create a ./.config file by setting symbol
  174. values to 'n' as much as possible.
  175. "make randconfig" Create a ./.config file by setting symbol
  176. values to random values.
  177. "make localmodconfig" Create a config based on current config and
  178. loaded modules (lsmod). Disables any module
  179. option that is not needed for the loaded modules.
  180. To create a localmodconfig for another machine,
  181. store the lsmod of that machine into a file
  182. and pass it in as a LSMOD parameter.
  183. target$ lsmod > /tmp/mylsmod
  184. target$ scp /tmp/mylsmod host:/tmp
  185. host$ make LSMOD=/tmp/mylsmod localmodconfig
  186. The above also works when cross compiling.
  187. "make localyesconfig" Similar to localmodconfig, except it will convert
  188. all module options to built in (=y) options.
  189. You can find more information on using the Linux kernel config tools
  190. in Documentation/kbuild/kconfig.txt.
  191. - NOTES on "make config":
  192. - Having unnecessary drivers will make the kernel bigger, and can
  193. under some circumstances lead to problems: probing for a
  194. nonexistent controller card may confuse your other controllers
  195. - Compiling the kernel with "Processor type" set higher than 386
  196. will result in a kernel that does NOT work on a 386. The
  197. kernel will detect this on bootup, and give up.
  198. - A kernel with math-emulation compiled in will still use the
  199. coprocessor if one is present: the math emulation will just
  200. never get used in that case. The kernel will be slightly larger,
  201. but will work on different machines regardless of whether they
  202. have a math coprocessor or not.
  203. - The "kernel hacking" configuration details usually result in a
  204. bigger or slower kernel (or both), and can even make the kernel
  205. less stable by configuring some routines to actively try to
  206. break bad code to find kernel problems (kmalloc()). Thus you
  207. should probably answer 'n' to the questions for "development",
  208. "experimental", or "debugging" features.
  209. COMPILING the kernel:
  210. - Make sure you have at least gcc 3.2 available.
  211. For more information, refer to Documentation/Changes.
  212. Please note that you can still run a.out user programs with this kernel.
  213. - Do a "make" to create a compressed kernel image. It is also
  214. possible to do "make install" if you have lilo installed to suit the
  215. kernel makefiles, but you may want to check your particular lilo setup first.
  216. To do the actual install, you have to be root, but none of the normal
  217. build should require that. Don't take the name of root in vain.
  218. - If you configured any of the parts of the kernel as `modules', you
  219. will also have to do "make modules_install".
  220. - Verbose kernel compile/build output:
  221. Normally, the kernel build system runs in a fairly quiet mode (but not
  222. totally silent). However, sometimes you or other kernel developers need
  223. to see compile, link, or other commands exactly as they are executed.
  224. For this, use "verbose" build mode. This is done by inserting
  225. "V=1" in the "make" command. E.g.:
  226. make V=1 all
  227. To have the build system also tell the reason for the rebuild of each
  228. target, use "V=2". The default is "V=0".
  229. - Keep a backup kernel handy in case something goes wrong. This is
  230. especially true for the development releases, since each new release
  231. contains new code which has not been debugged. Make sure you keep a
  232. backup of the modules corresponding to that kernel, as well. If you
  233. are installing a new kernel with the same version number as your
  234. working kernel, make a backup of your modules directory before you
  235. do a "make modules_install".
  236. Alternatively, before compiling, use the kernel config option
  237. "LOCALVERSION" to append a unique suffix to the regular kernel version.
  238. LOCALVERSION can be set in the "General Setup" menu.
  239. - In order to boot your new kernel, you'll need to copy the kernel
  240. image (e.g. .../linux/arch/i386/boot/bzImage after compilation)
  241. to the place where your regular bootable kernel is found.
  242. - Booting a kernel directly from a floppy without the assistance of a
  243. bootloader such as LILO, is no longer supported.
  244. If you boot Linux from the hard drive, chances are you use LILO, which
  245. uses the kernel image as specified in the file /etc/lilo.conf. The
  246. kernel image file is usually /vmlinuz, /boot/vmlinuz, /bzImage or
  247. /boot/bzImage. To use the new kernel, save a copy of the old image
  248. and copy the new image over the old one. Then, you MUST RERUN LILO
  249. to update the loading map!! If you don't, you won't be able to boot
  250. the new kernel image.
  251. Reinstalling LILO is usually a matter of running /sbin/lilo.
  252. You may wish to edit /etc/lilo.conf to specify an entry for your
  253. old kernel image (say, /vmlinux.old) in case the new one does not
  254. work. See the LILO docs for more information.
  255. After reinstalling LILO, you should be all set. Shutdown the system,
  256. reboot, and enjoy!
  257. If you ever need to change the default root device, video mode,
  258. ramdisk size, etc. in the kernel image, use the 'rdev' program (or
  259. alternatively the LILO boot options when appropriate). No need to
  260. recompile the kernel to change these parameters.
  261. - Reboot with the new kernel and enjoy.
  263. - If you have problems that seem to be due to kernel bugs, please check
  264. the file MAINTAINERS to see if there is a particular person associated
  265. with the part of the kernel that you are having trouble with. If there
  266. isn't anyone listed there, then the second best thing is to mail
  267. them to me (torvalds@linux-foundation.org), and possibly to any other
  268. relevant mailing-list or to the newsgroup.
  269. - In all bug-reports, *please* tell what kernel you are talking about,
  270. how to duplicate the problem, and what your setup is (use your common
  271. sense). If the problem is new, tell me so, and if the problem is
  272. old, please try to tell me when you first noticed it.
  273. - If the bug results in a message like
  274. unable to handle kernel paging request at address C0000010
  275. Oops: 0002
  276. EIP: 0010:XXXXXXXX
  277. eax: xxxxxxxx ebx: xxxxxxxx ecx: xxxxxxxx edx: xxxxxxxx
  278. esi: xxxxxxxx edi: xxxxxxxx ebp: xxxxxxxx
  279. ds: xxxx es: xxxx fs: xxxx gs: xxxx
  280. Pid: xx, process nr: xx
  281. xx xx xx xx xx xx xx xx xx xx
  282. or similar kernel debugging information on your screen or in your
  283. system log, please duplicate it *exactly*. The dump may look
  284. incomprehensible to you, but it does contain information that may
  285. help debugging the problem. The text above the dump is also
  286. important: it tells something about why the kernel dumped code (in
  287. the above example, it's due to a bad kernel pointer). More information
  288. on making sense of the dump is in Documentation/oops-tracing.txt
  289. - If you compiled the kernel with CONFIG_KALLSYMS you can send the dump
  290. as is, otherwise you will have to use the "ksymoops" program to make
  291. sense of the dump (but compiling with CONFIG_KALLSYMS is usually preferred).
  292. This utility can be downloaded from
  293. ftp://ftp.<country>.kernel.org/pub/linux/utils/kernel/ksymoops/ .
  294. Alternatively, you can do the dump lookup by hand:
  295. - In debugging dumps like the above, it helps enormously if you can
  296. look up what the EIP value means. The hex value as such doesn't help
  297. me or anybody else very much: it will depend on your particular
  298. kernel setup. What you should do is take the hex value from the EIP
  299. line (ignore the "0010:"), and look it up in the kernel namelist to
  300. see which kernel function contains the offending address.
  301. To find out the kernel function name, you'll need to find the system
  302. binary associated with the kernel that exhibited the symptom. This is
  303. the file 'linux/vmlinux'. To extract the namelist and match it against
  304. the EIP from the kernel crash, do:
  305. nm vmlinux | sort | less
  306. This will give you a list of kernel addresses sorted in ascending
  307. order, from which it is simple to find the function that contains the
  308. offending address. Note that the address given by the kernel
  309. debugging messages will not necessarily match exactly with the
  310. function addresses (in fact, that is very unlikely), so you can't
  311. just 'grep' the list: the list will, however, give you the starting
  312. point of each kernel function, so by looking for the function that
  313. has a starting address lower than the one you are searching for but
  314. is followed by a function with a higher address you will find the one
  315. you want. In fact, it may be a good idea to include a bit of
  316. "context" in your problem report, giving a few lines around the
  317. interesting one.
  318. If you for some reason cannot do the above (you have a pre-compiled
  319. kernel image or similar), telling me as much about your setup as
  320. possible will help. Please read the REPORTING-BUGS document for details.
  321. - Alternatively, you can use gdb on a running kernel. (read-only; i.e. you
  322. cannot change values or set break points.) To do this, first compile the
  323. kernel with -g; edit arch/i386/Makefile appropriately, then do a "make
  324. clean". You'll also need to enable CONFIG_PROC_FS (via "make config").
  325. After you've rebooted with the new kernel, do "gdb vmlinux /proc/kcore".
  326. You can now use all the usual gdb commands. The command to look up the
  327. point where your system crashed is "l *0xXXXXXXXX". (Replace the XXXes
  328. with the EIP value.)
  329. gdb'ing a non-running kernel currently fails because gdb (wrongly)
  330. disregards the starting offset for which the kernel is compiled.