1.. _readme: 2 3Linux kernel release 5.x <http://kernel.org/> 4============================================= 5 6These are the release notes for Linux version 5. Read them carefully, 7as they tell you what this is all about, explain how to install the 8kernel, and what to do if something goes wrong. 9 10What is Linux? 11-------------- 12 13 Linux is a clone of the operating system Unix, written from scratch by 14 Linus Torvalds with assistance from a loosely-knit team of hackers across 15 the Net. It aims towards POSIX and Single UNIX Specification compliance. 16 17 It has all the features you would expect in a modern fully-fledged Unix, 18 including true multitasking, virtual memory, shared libraries, demand 19 loading, shared copy-on-write executables, proper memory management, 20 and multistack networking including IPv4 and IPv6. 21 22 It is distributed under the GNU General Public License v2 - see the 23 accompanying COPYING file for more details. 24 25On what hardware does it run? 26----------------------------- 27 28 Although originally developed first for 32-bit x86-based PCs (386 or higher), 29 today Linux also runs on (at least) the Compaq Alpha AXP, Sun SPARC and 30 UltraSPARC, Motorola 68000, PowerPC, PowerPC64, ARM, Hitachi SuperH, Cell, 31 IBM S/390, MIPS, HP PA-RISC, Intel IA-64, DEC VAX, AMD x86-64 Xtensa, and 32 ARC architectures. 33 34 Linux is easily portable to most general-purpose 32- or 64-bit architectures 35 as long as they have a paged memory management unit (PMMU) and a port of the 36 GNU C compiler (gcc) (part of The GNU Compiler Collection, GCC). Linux has 37 also been ported to a number of architectures without a PMMU, although 38 functionality is then obviously somewhat limited. 39 Linux has also been ported to itself. You can now run the kernel as a 40 userspace application - this is called UserMode Linux (UML). 41 42Documentation 43------------- 44 45 - There is a lot of documentation available both in electronic form on 46 the Internet and in books, both Linux-specific and pertaining to 47 general UNIX questions. I'd recommend looking into the documentation 48 subdirectories on any Linux FTP site for the LDP (Linux Documentation 49 Project) books. This README is not meant to be documentation on the 50 system: there are much better sources available. 51 52 - There are various README files in the Documentation/ subdirectory: 53 these typically contain kernel-specific installation notes for some 54 drivers for example. Please read the 55 :ref:`Documentation/process/changes.rst <changes>` file, as it 56 contains information about the problems, which may result by upgrading 57 your kernel. 58 59Installing the kernel source 60---------------------------- 61 62 - If you install the full sources, put the kernel tarball in a 63 directory where you have permissions (e.g. your home directory) and 64 unpack it:: 65 66 xz -cd linux-5.x.tar.xz | tar xvf - 67 68 Replace "X" with the version number of the latest kernel. 69 70 Do NOT use the /usr/src/linux area! This area has a (usually 71 incomplete) set of kernel headers that are used by the library header 72 files. They should match the library, and not get messed up by 73 whatever the kernel-du-jour happens to be. 74 75 - You can also upgrade between 5.x releases by patching. Patches are 76 distributed in the xz format. To install by patching, get all the 77 newer patch files, enter the top level directory of the kernel source 78 (linux-5.x) and execute:: 79 80 xz -cd ../patch-5.x.xz | patch -p1 81 82 Replace "x" for all versions bigger than the version "x" of your current 83 source tree, **in_order**, and you should be ok. You may want to remove 84 the backup files (some-file-name~ or some-file-name.orig), and make sure 85 that there are no failed patches (some-file-name# or some-file-name.rej). 86 If there are, either you or I have made a mistake. 87 88 Unlike patches for the 5.x kernels, patches for the 5.x.y kernels 89 (also known as the -stable kernels) are not incremental but instead apply 90 directly to the base 5.x kernel. For example, if your base kernel is 5.0 91 and you want to apply the 5.0.3 patch, you must not first apply the 5.0.1 92 and 5.0.2 patches. Similarly, if you are running kernel version 5.0.2 and 93 want to jump to 5.0.3, you must first reverse the 5.0.2 patch (that is, 94 patch -R) **before** applying the 5.0.3 patch. You can read more on this in 95 :ref:`Documentation/process/applying-patches.rst <applying_patches>`. 96 97 Alternatively, the script patch-kernel can be used to automate this 98 process. It determines the current kernel version and applies any 99 patches found:: 100 101 linux/scripts/patch-kernel linux 102 103 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. 106 107 - Make sure you have no stale .o files and dependencies lying around:: 108 109 cd linux 110 make mrproper 111 112 You should now have the sources correctly installed. 113 114Software requirements 115--------------------- 116 117 Compiling and running the 5.x kernels requires up-to-date 118 versions of various software packages. Consult 119 :ref:`Documentation/process/changes.rst <changes>` for the minimum version numbers 120 required and how to get updates for these packages. Beware that using 121 excessively old versions of these packages can cause indirect 122 errors that are very difficult to track down, so don't assume that 123 you can just update packages when obvious problems arise during 124 build or operation. 125 126Build directory for the kernel 127------------------------------ 128 129 When compiling the kernel, all output files will per default be 130 stored together with the kernel source code. 131 Using the option ``make O=output/dir`` allows you to specify an alternate 132 place for the output files (including .config). 133 Example:: 134 135 kernel source code: /usr/src/linux-5.x 136 build directory: /home/name/build/kernel 137 138 To configure and build the kernel, use:: 139 140 cd /usr/src/linux-5.x 141 make O=/home/name/build/kernel menuconfig 142 make O=/home/name/build/kernel 143 sudo make O=/home/name/build/kernel modules_install install 144 145 Please note: If the ``O=output/dir`` option is used, then it must be 146 used for all invocations of make. 147 148Configuring the kernel 149---------------------- 150 151 Do not skip this step even if you are only upgrading one minor 152 version. New configuration options are added in each release, and 153 odd problems will turn up if the configuration files are not set up 154 as expected. If you want to carry your existing configuration to a 155 new version with minimal work, use ``make oldconfig``, which will 156 only ask you for the answers to new questions. 157 158 - Alternative configuration commands are:: 159 160 "make config" Plain text interface. 161 162 "make menuconfig" Text based color menus, radiolists & dialogs. 163 164 "make nconfig" Enhanced text based color menus. 165 166 "make xconfig" Qt based configuration tool. 167 168 "make gconfig" GTK+ based configuration tool. 169 170 "make oldconfig" Default all questions based on the contents of 171 your existing ./.config file and asking about 172 new config symbols. 173 174 "make olddefconfig" 175 Like above, but sets new symbols to their default 176 values without prompting. 177 178 "make defconfig" Create a ./.config file by using the default 179 symbol values from either arch/$ARCH/defconfig 180 or arch/$ARCH/configs/${PLATFORM}_defconfig, 181 depending on the architecture. 182 183 "make ${PLATFORM}_defconfig" 184 Create a ./.config file by using the default 185 symbol values from 186 arch/$ARCH/configs/${PLATFORM}_defconfig. 187 Use "make help" to get a list of all available 188 platforms of your architecture. 189 190 "make allyesconfig" 191 Create a ./.config file by setting symbol 192 values to 'y' as much as possible. 193 194 "make allmodconfig" 195 Create a ./.config file by setting symbol 196 values to 'm' as much as possible. 197 198 "make allnoconfig" Create a ./.config file by setting symbol 199 values to 'n' as much as possible. 200 201 "make randconfig" Create a ./.config file by setting symbol 202 values to random values. 203 204 "make localmodconfig" Create a config based on current config and 205 loaded modules (lsmod). Disables any module 206 option that is not needed for the loaded modules. 207 208 To create a localmodconfig for another machine, 209 store the lsmod of that machine into a file 210 and pass it in as a LSMOD parameter. 211 212 Also, you can preserve modules in certain folders 213 or kconfig files by specifying their paths in 214 parameter LMC_KEEP. 215 216 target$ lsmod > /tmp/mylsmod 217 target$ scp /tmp/mylsmod host:/tmp 218 219 host$ make LSMOD=/tmp/mylsmod \ 220 LMC_KEEP="drivers/usb:drivers/gpu:fs" \ 221 localmodconfig 222 223 The above also works when cross compiling. 224 225 "make localyesconfig" Similar to localmodconfig, except it will convert 226 all module options to built in (=y) options. You can 227 also preserve modules by LMC_KEEP. 228 229 "make kvmconfig" Enable additional options for kvm guest kernel support. 230 231 "make xenconfig" Enable additional options for xen dom0 guest kernel 232 support. 233 234 "make tinyconfig" Configure the tiniest possible kernel. 235 236 You can find more information on using the Linux kernel config tools 237 in Documentation/kbuild/kconfig.rst. 238 239 - NOTES on ``make config``: 240 241 - Having unnecessary drivers will make the kernel bigger, and can 242 under some circumstances lead to problems: probing for a 243 nonexistent controller card may confuse your other controllers. 244 245 - A kernel with math-emulation compiled in will still use the 246 coprocessor if one is present: the math emulation will just 247 never get used in that case. The kernel will be slightly larger, 248 but will work on different machines regardless of whether they 249 have a math coprocessor or not. 250 251 - The "kernel hacking" configuration details usually result in a 252 bigger or slower kernel (or both), and can even make the kernel 253 less stable by configuring some routines to actively try to 254 break bad code to find kernel problems (kmalloc()). Thus you 255 should probably answer 'n' to the questions for "development", 256 "experimental", or "debugging" features. 257 258Compiling the kernel 259-------------------- 260 261 - Make sure you have at least gcc 4.9 available. 262 For more information, refer to :ref:`Documentation/process/changes.rst <changes>`. 263 264 Please note that you can still run a.out user programs with this kernel. 265 266 - Do a ``make`` to create a compressed kernel image. It is also 267 possible to do ``make install`` if you have lilo installed to suit the 268 kernel makefiles, but you may want to check your particular lilo setup first. 269 270 To do the actual install, you have to be root, but none of the normal 271 build should require that. Don't take the name of root in vain. 272 273 - If you configured any of the parts of the kernel as ``modules``, you 274 will also have to do ``make modules_install``. 275 276 - Verbose kernel compile/build output: 277 278 Normally, the kernel build system runs in a fairly quiet mode (but not 279 totally silent). However, sometimes you or other kernel developers need 280 to see compile, link, or other commands exactly as they are executed. 281 For this, use "verbose" build mode. This is done by passing 282 ``V=1`` to the ``make`` command, e.g.:: 283 284 make V=1 all 285 286 To have the build system also tell the reason for the rebuild of each 287 target, use ``V=2``. The default is ``V=0``. 288 289 - Keep a backup kernel handy in case something goes wrong. This is 290 especially true for the development releases, since each new release 291 contains new code which has not been debugged. Make sure you keep a 292 backup of the modules corresponding to that kernel, as well. If you 293 are installing a new kernel with the same version number as your 294 working kernel, make a backup of your modules directory before you 295 do a ``make modules_install``. 296 297 Alternatively, before compiling, use the kernel config option 298 "LOCALVERSION" to append a unique suffix to the regular kernel version. 299 LOCALVERSION can be set in the "General Setup" menu. 300 301 - In order to boot your new kernel, you'll need to copy the kernel 302 image (e.g. .../linux/arch/x86/boot/bzImage after compilation) 303 to the place where your regular bootable kernel is found. 304 305 - Booting a kernel directly from a floppy without the assistance of a 306 bootloader such as LILO, is no longer supported. 307 308 If you boot Linux from the hard drive, chances are you use LILO, which 309 uses the kernel image as specified in the file /etc/lilo.conf. The 310 kernel image file is usually /vmlinuz, /boot/vmlinuz, /bzImage or 311 /boot/bzImage. To use the new kernel, save a copy of the old image 312 and copy the new image over the old one. Then, you MUST RERUN LILO 313 to update the loading map! If you don't, you won't be able to boot 314 the new kernel image. 315 316 Reinstalling LILO is usually a matter of running /sbin/lilo. 317 You may wish to edit /etc/lilo.conf to specify an entry for your 318 old kernel image (say, /vmlinux.old) in case the new one does not 319 work. See the LILO docs for more information. 320 321 After reinstalling LILO, you should be all set. Shutdown the system, 322 reboot, and enjoy! 323 324 If you ever need to change the default root device, video mode, 325 etc. in the kernel image, use your bootloader's boot options 326 where appropriate. No need to recompile the kernel to change 327 these parameters. 328 329 - Reboot with the new kernel and enjoy. 330 331If something goes wrong 332----------------------- 333 334 - If you have problems that seem to be due to kernel bugs, please check 335 the file MAINTAINERS to see if there is a particular person associated 336 with the part of the kernel that you are having trouble with. If there 337 isn't anyone listed there, then the second best thing is to mail 338 them to me (torvalds@linux-foundation.org), and possibly to any other 339 relevant mailing-list or to the newsgroup. 340 341 - In all bug-reports, *please* tell what kernel you are talking about, 342 how to duplicate the problem, and what your setup is (use your common 343 sense). If the problem is new, tell me so, and if the problem is 344 old, please try to tell me when you first noticed it. 345 346 - If the bug results in a message like:: 347 348 unable to handle kernel paging request at address C0000010 349 Oops: 0002 350 EIP: 0010:XXXXXXXX 351 eax: xxxxxxxx ebx: xxxxxxxx ecx: xxxxxxxx edx: xxxxxxxx 352 esi: xxxxxxxx edi: xxxxxxxx ebp: xxxxxxxx 353 ds: xxxx es: xxxx fs: xxxx gs: xxxx 354 Pid: xx, process nr: xx 355 xx xx xx xx xx xx xx xx xx xx 356 357 or similar kernel debugging information on your screen or in your 358 system log, please duplicate it *exactly*. The dump may look 359 incomprehensible to you, but it does contain information that may 360 help debugging the problem. The text above the dump is also 361 important: it tells something about why the kernel dumped code (in 362 the above example, it's due to a bad kernel pointer). More information 363 on making sense of the dump is in Documentation/admin-guide/bug-hunting.rst 364 365 - If you compiled the kernel with CONFIG_KALLSYMS you can send the dump 366 as is, otherwise you will have to use the ``ksymoops`` program to make 367 sense of the dump (but compiling with CONFIG_KALLSYMS is usually preferred). 368 This utility can be downloaded from 369 https://www.kernel.org/pub/linux/utils/kernel/ksymoops/ . 370 Alternatively, you can do the dump lookup by hand: 371 372 - In debugging dumps like the above, it helps enormously if you can 373 look up what the EIP value means. The hex value as such doesn't help 374 me or anybody else very much: it will depend on your particular 375 kernel setup. What you should do is take the hex value from the EIP 376 line (ignore the ``0010:``), and look it up in the kernel namelist to 377 see which kernel function contains the offending address. 378 379 To find out the kernel function name, you'll need to find the system 380 binary associated with the kernel that exhibited the symptom. This is 381 the file 'linux/vmlinux'. To extract the namelist and match it against 382 the EIP from the kernel crash, do:: 383 384 nm vmlinux | sort | less 385 386 This will give you a list of kernel addresses sorted in ascending 387 order, from which it is simple to find the function that contains the 388 offending address. Note that the address given by the kernel 389 debugging messages will not necessarily match exactly with the 390 function addresses (in fact, that is very unlikely), so you can't 391 just 'grep' the list: the list will, however, give you the starting 392 point of each kernel function, so by looking for the function that 393 has a starting address lower than the one you are searching for but 394 is followed by a function with a higher address you will find the one 395 you want. In fact, it may be a good idea to include a bit of 396 "context" in your problem report, giving a few lines around the 397 interesting one. 398 399 If you for some reason cannot do the above (you have a pre-compiled 400 kernel image or similar), telling me as much about your setup as 401 possible will help. Please read the :ref:`admin-guide/reporting-bugs.rst <reportingbugs>` 402 document for details. 403 404 - Alternatively, you can use gdb on a running kernel. (read-only; i.e. you 405 cannot change values or set break points.) To do this, first compile the 406 kernel with -g; edit arch/x86/Makefile appropriately, then do a ``make 407 clean``. You'll also need to enable CONFIG_PROC_FS (via ``make config``). 408 409 After you've rebooted with the new kernel, do ``gdb vmlinux /proc/kcore``. 410 You can now use all the usual gdb commands. The command to look up the 411 point where your system crashed is ``l *0xXXXXXXXX``. (Replace the XXXes 412 with the EIP value.) 413 414 gdb'ing a non-running kernel currently fails because ``gdb`` (wrongly) 415 disregards the starting offset for which the kernel is compiled. 416