Based on kernel version 3.16. Page generated on 2014-08-06 21:36 EST.
1 2 Applying Patches To The Linux Kernel 3 ------------------------------------ 4 5 Original by: Jesper Juhl, August 2005 6 Last update: 2006-01-05 7 8 9 A frequently asked question on the Linux Kernel Mailing List is how to apply 10 a patch to the kernel or, more specifically, what base kernel a patch for 11 one of the many trees/branches should be applied to. Hopefully this document 12 will explain this to you. 13 14 In addition to explaining how to apply and revert patches, a brief 15 description of the different kernel trees (and examples of how to apply 16 their specific patches) is also provided. 17 18 19 What is a patch? 20 --- 21 A patch is a small text document containing a delta of changes between two 22 different versions of a source tree. Patches are created with the `diff' 23 program. 24 To correctly apply a patch you need to know what base it was generated from 25 and what new version the patch will change the source tree into. These 26 should both be present in the patch file metadata or be possible to deduce 27 from the filename. 28 29 30 How do I apply or revert a patch? 31 --- 32 You apply a patch with the `patch' program. The patch program reads a diff 33 (or patch) file and makes the changes to the source tree described in it. 34 35 Patches for the Linux kernel are generated relative to the parent directory 36 holding the kernel source dir. 37 38 This means that paths to files inside the patch file contain the name of the 39 kernel source directories it was generated against (or some other directory 40 names like "a/" and "b/"). 41 Since this is unlikely to match the name of the kernel source dir on your 42 local machine (but is often useful info to see what version an otherwise 43 unlabeled patch was generated against) you should change into your kernel 44 source directory and then strip the first element of the path from filenames 45 in the patch file when applying it (the -p1 argument to `patch' does this). 46 47 To revert a previously applied patch, use the -R argument to patch. 48 So, if you applied a patch like this: 49 patch -p1 < ../patch-x.y.z 50 51 You can revert (undo) it like this: 52 patch -R -p1 < ../patch-x.y.z 53 54 55 How do I feed a patch/diff file to `patch'? 56 --- 57 This (as usual with Linux and other UNIX like operating systems) can be 58 done in several different ways. 59 In all the examples below I feed the file (in uncompressed form) to patch 60 via stdin using the following syntax: 61 patch -p1 < path/to/patch-x.y.z 62 63 If you just want to be able to follow the examples below and don't want to 64 know of more than one way to use patch, then you can stop reading this 65 section here. 66 67 Patch can also get the name of the file to use via the -i argument, like 68 this: 69 patch -p1 -i path/to/patch-x.y.z 70 71 If your patch file is compressed with gzip or bzip2 and you don't want to 72 uncompress it before applying it, then you can feed it to patch like this 73 instead: 74 zcat path/to/patch-x.y.z.gz | patch -p1 75 bzcat path/to/patch-x.y.z.bz2 | patch -p1 76 77 If you wish to uncompress the patch file by hand first before applying it 78 (what I assume you've done in the examples below), then you simply run 79 gunzip or bunzip2 on the file -- like this: 80 gunzip patch-x.y.z.gz 81 bunzip2 patch-x.y.z.bz2 82 83 Which will leave you with a plain text patch-x.y.z file that you can feed to 84 patch via stdin or the -i argument, as you prefer. 85 86 A few other nice arguments for patch are -s which causes patch to be silent 87 except for errors which is nice to prevent errors from scrolling out of the 88 screen too fast, and --dry-run which causes patch to just print a listing of 89 what would happen, but doesn't actually make any changes. Finally --verbose 90 tells patch to print more information about the work being done. 91 92 93 Common errors when patching 94 --- 95 When patch applies a patch file it attempts to verify the sanity of the 96 file in different ways. 97 Checking that the file looks like a valid patch file & checking the code 98 around the bits being modified matches the context provided in the patch are 99 just two of the basic sanity checks patch does. 100 101 If patch encounters something that doesn't look quite right it has two 102 options. It can either refuse to apply the changes and abort or it can try 103 to find a way to make the patch apply with a few minor changes. 104 105 One example of something that's not 'quite right' that patch will attempt to 106 fix up is if all the context matches, the lines being changed match, but the 107 line numbers are different. This can happen, for example, if the patch makes 108 a change in the middle of the file but for some reasons a few lines have 109 been added or removed near the beginning of the file. In that case 110 everything looks good it has just moved up or down a bit, and patch will 111 usually adjust the line numbers and apply the patch. 112 113 Whenever patch applies a patch that it had to modify a bit to make it fit 114 it'll tell you about it by saying the patch applied with 'fuzz'. 115 You should be wary of such changes since even though patch probably got it 116 right it doesn't /always/ get it right, and the result will sometimes be 117 wrong. 118 119 When patch encounters a change that it can't fix up with fuzz it rejects it 120 outright and leaves a file with a .rej extension (a reject file). You can 121 read this file to see exactly what change couldn't be applied, so you can 122 go fix it up by hand if you wish. 123 124 If you don't have any third-party patches applied to your kernel source, but 125 only patches from kernel.org and you apply the patches in the correct order, 126 and have made no modifications yourself to the source files, then you should 127 never see a fuzz or reject message from patch. If you do see such messages 128 anyway, then there's a high risk that either your local source tree or the 129 patch file is corrupted in some way. In that case you should probably try 130 re-downloading the patch and if things are still not OK then you'd be advised 131 to start with a fresh tree downloaded in full from kernel.org. 132 133 Let's look a bit more at some of the messages patch can produce. 134 135 If patch stops and presents a "File to patch:" prompt, then patch could not 136 find a file to be patched. Most likely you forgot to specify -p1 or you are 137 in the wrong directory. Less often, you'll find patches that need to be 138 applied with -p0 instead of -p1 (reading the patch file should reveal if 139 this is the case -- if so, then this is an error by the person who created 140 the patch but is not fatal). 141 142 If you get "Hunk #2 succeeded at 1887 with fuzz 2 (offset 7 lines)." or a 143 message similar to that, then it means that patch had to adjust the location 144 of the change (in this example it needed to move 7 lines from where it 145 expected to make the change to make it fit). 146 The resulting file may or may not be OK, depending on the reason the file 147 was different than expected. 148 This often happens if you try to apply a patch that was generated against a 149 different kernel version than the one you are trying to patch. 150 151 If you get a message like "Hunk #3 FAILED at 2387.", then it means that the 152 patch could not be applied correctly and the patch program was unable to 153 fuzz its way through. This will generate a .rej file with the change that 154 caused the patch to fail and also a .orig file showing you the original 155 content that couldn't be changed. 156 157 If you get "Reversed (or previously applied) patch detected! Assume -R? [n]" 158 then patch detected that the change contained in the patch seems to have 159 already been made. 160 If you actually did apply this patch previously and you just re-applied it 161 in error, then just say [n]o and abort this patch. If you applied this patch 162 previously and actually intended to revert it, but forgot to specify -R, 163 then you can say [y]es here to make patch revert it for you. 164 This can also happen if the creator of the patch reversed the source and 165 destination directories when creating the patch, and in that case reverting 166 the patch will in fact apply it. 167 168 A message similar to "patch: **** unexpected end of file in patch" or "patch 169 unexpectedly ends in middle of line" means that patch could make no sense of 170 the file you fed to it. Either your download is broken, you tried to feed 171 patch a compressed patch file without uncompressing it first, or the patch 172 file that you are using has been mangled by a mail client or mail transfer 173 agent along the way somewhere, e.g., by splitting a long line into two lines. 174 Often these warnings can easily be fixed by joining (concatenating) the 175 two lines that had been split. 176 177 As I already mentioned above, these errors should never happen if you apply 178 a patch from kernel.org to the correct version of an unmodified source tree. 179 So if you get these errors with kernel.org patches then you should probably 180 assume that either your patch file or your tree is broken and I'd advise you 181 to start over with a fresh download of a full kernel tree and the patch you 182 wish to apply. 183 184 185 Are there any alternatives to `patch'? 186 --- 187 Yes there are alternatives. 188 189 You can use the `interdiff' program (http://cyberelk.net/tim/patchutils/) to 190 generate a patch representing the differences between two patches and then 191 apply the result. 192 This will let you move from something like 184.108.40.206 to 220.127.116.11 in a single 193 step. The -z flag to interdiff will even let you feed it patches in gzip or 194 bzip2 compressed form directly without the use of zcat or bzcat or manual 195 decompression. 196 197 Here's how you'd go from 18.104.22.168 to 22.214.171.124 in a single step: 198 interdiff -z ../patch-126.96.36.199.bz2 ../patch-188.8.131.52.gz | patch -p1 199 200 Although interdiff may save you a step or two you are generally advised to 201 do the additional steps since interdiff can get things wrong in some cases. 202 203 Another alternative is `ketchup', which is a python script for automatic 204 downloading and applying of patches (http://www.selenic.com/ketchup/). 205 206 Other nice tools are diffstat, which shows a summary of changes made by a 207 patch; lsdiff, which displays a short listing of affected files in a patch 208 file, along with (optionally) the line numbers of the start of each patch; 209 and grepdiff, which displays a list of the files modified by a patch where 210 the patch contains a given regular expression. 211 212 213 Where can I download the patches? 214 --- 215 The patches are available at http://kernel.org/ 216 Most recent patches are linked from the front page, but they also have 217 specific homes. 218 219 The 2.6.x.y (-stable) and 2.6.x patches live at 220 ftp://ftp.kernel.org/pub/linux/kernel/v2.6/ 221 222 The -rc patches live at 223 ftp://ftp.kernel.org/pub/linux/kernel/v2.6/testing/ 224 225 The -git patches live at 226 ftp://ftp.kernel.org/pub/linux/kernel/v2.6/snapshots/ 227 228 The -mm kernels live at 229 ftp://ftp.kernel.org/pub/linux/kernel/people/akpm/patches/2.6/ 230 231 In place of ftp.kernel.org you can use ftp.cc.kernel.org, where cc is a 232 country code. This way you'll be downloading from a mirror site that's most 233 likely geographically closer to you, resulting in faster downloads for you, 234 less bandwidth used globally and less load on the main kernel.org servers -- 235 these are good things, so do use mirrors when possible. 236 237 238 The 2.6.x kernels 239 --- 240 These are the base stable releases released by Linus. The highest numbered 241 release is the most recent. 242 243 If regressions or other serious flaws are found, then a -stable fix patch 244 will be released (see below) on top of this base. Once a new 2.6.x base 245 kernel is released, a patch is made available that is a delta between the 246 previous 2.6.x kernel and the new one. 247 248 To apply a patch moving from 2.6.11 to 2.6.12, you'd do the following (note 249 that such patches do *NOT* apply on top of 2.6.x.y kernels but on top of the 250 base 2.6.x kernel -- if you need to move from 2.6.x.y to 2.6.x+1 you need to 251 first revert the 2.6.x.y patch). 252 253 Here are some examples: 254 255 # moving from 2.6.11 to 2.6.12 256 $ cd ~/linux-2.6.11 # change to kernel source dir 257 $ patch -p1 < ../patch-2.6.12 # apply the 2.6.12 patch 258 $ cd .. 259 $ mv linux-2.6.11 linux-2.6.12 # rename source dir 260 261 # moving from 184.108.40.206 to 2.6.12 262 $ cd ~/linux-220.127.116.11 # change to kernel source dir 263 $ patch -p1 -R < ../patch-18.104.22.168 # revert the 22.214.171.124 patch 264 # source dir is now 2.6.11 265 $ patch -p1 < ../patch-2.6.12 # apply new 2.6.12 patch 266 $ cd .. 267 $ mv linux-126.96.36.199 linux-2.6.12 # rename source dir 268 269 270 The 2.6.x.y kernels 271 --- 272 Kernels with 4-digit versions are -stable kernels. They contain small(ish) 273 critical fixes for security problems or significant regressions discovered 274 in a given 2.6.x kernel. 275 276 This is the recommended branch for users who want the most recent stable 277 kernel and are not interested in helping test development/experimental 278 versions. 279 280 If no 2.6.x.y kernel is available, then the highest numbered 2.6.x kernel is 281 the current stable kernel. 282 283 note: the -stable team usually do make incremental patches available as well 284 as patches against the latest mainline release, but I only cover the 285 non-incremental ones below. The incremental ones can be found at 286 ftp://ftp.kernel.org/pub/linux/kernel/v2.6/incr/ 287 288 These patches are not incremental, meaning that for example the 188.8.131.52 289 patch does not apply on top of the 184.108.40.206 kernel source, but rather on top 290 of the base 2.6.12 kernel source . 291 So, in order to apply the 220.127.116.11 patch to your existing 18.104.22.168 kernel 292 source you have to first back out the 22.214.171.124 patch (so you are left with a 293 base 2.6.12 kernel source) and then apply the new 126.96.36.199 patch. 294 295 Here's a small example: 296 297 $ cd ~/linux-188.8.131.52 # change into the kernel source dir 298 $ patch -p1 -R < ../patch-184.108.40.206 # revert the 220.127.116.11 patch 299 $ patch -p1 < ../patch-18.104.22.168 # apply the new 22.214.171.124 patch 300 $ cd .. 301 $ mv linux-126.96.36.199 linux-188.8.131.52 # rename the kernel source dir 302 303 304 The -rc kernels 305 --- 306 These are release-candidate kernels. These are development kernels released 307 by Linus whenever he deems the current git (the kernel's source management 308 tool) tree to be in a reasonably sane state adequate for testing. 309 310 These kernels are not stable and you should expect occasional breakage if 311 you intend to run them. This is however the most stable of the main 312 development branches and is also what will eventually turn into the next 313 stable kernel, so it is important that it be tested by as many people as 314 possible. 315 316 This is a good branch to run for people who want to help out testing 317 development kernels but do not want to run some of the really experimental 318 stuff (such people should see the sections about -git and -mm kernels below). 319 320 The -rc patches are not incremental, they apply to a base 2.6.x kernel, just 321 like the 2.6.x.y patches described above. The kernel version before the -rcN 322 suffix denotes the version of the kernel that this -rc kernel will eventually 323 turn into. 324 So, 2.6.13-rc5 means that this is the fifth release candidate for the 2.6.13 325 kernel and the patch should be applied on top of the 2.6.12 kernel source. 326 327 Here are 3 examples of how to apply these patches: 328 329 # first an example of moving from 2.6.12 to 2.6.13-rc3 330 $ cd ~/linux-2.6.12 # change into the 2.6.12 source dir 331 $ patch -p1 < ../patch-2.6.13-rc3 # apply the 2.6.13-rc3 patch 332 $ cd .. 333 $ mv linux-2.6.12 linux-2.6.13-rc3 # rename the source dir 334 335 # now let's move from 2.6.13-rc3 to 2.6.13-rc5 336 $ cd ~/linux-2.6.13-rc3 # change into the 2.6.13-rc3 dir 337 $ patch -p1 -R < ../patch-2.6.13-rc3 # revert the 2.6.13-rc3 patch 338 $ patch -p1 < ../patch-2.6.13-rc5 # apply the new 2.6.13-rc5 patch 339 $ cd .. 340 $ mv linux-2.6.13-rc3 linux-2.6.13-rc5 # rename the source dir 341 342 # finally let's try and move from 184.108.40.206 to 2.6.13-rc5 343 $ cd ~/linux-220.127.116.11 # change to the kernel source dir 344 $ patch -p1 -R < ../patch-18.104.22.168 # revert the 22.214.171.124 patch 345 $ patch -p1 < ../patch-2.6.13-rc5 # apply new 2.6.13-rc5 patch 346 $ cd .. 347 $ mv linux-126.96.36.199 linux-2.6.13-rc5 # rename the kernel source dir 348 349 350 The -git kernels 351 --- 352 These are daily snapshots of Linus' kernel tree (managed in a git 353 repository, hence the name). 354 355 These patches are usually released daily and represent the current state of 356 Linus's tree. They are more experimental than -rc kernels since they are 357 generated automatically without even a cursory glance to see if they are 358 sane. 359 360 -git patches are not incremental and apply either to a base 2.6.x kernel or 361 a base 2.6.x-rc kernel -- you can see which from their name. 362 A patch named 2.6.12-git1 applies to the 2.6.12 kernel source and a patch 363 named 2.6.13-rc3-git2 applies to the source of the 2.6.13-rc3 kernel. 364 365 Here are some examples of how to apply these patches: 366 367 # moving from 2.6.12 to 2.6.12-git1 368 $ cd ~/linux-2.6.12 # change to the kernel source dir 369 $ patch -p1 < ../patch-2.6.12-git1 # apply the 2.6.12-git1 patch 370 $ cd .. 371 $ mv linux-2.6.12 linux-2.6.12-git1 # rename the kernel source dir 372 373 # moving from 2.6.12-git1 to 2.6.13-rc2-git3 374 $ cd ~/linux-2.6.12-git1 # change to the kernel source dir 375 $ patch -p1 -R < ../patch-2.6.12-git1 # revert the 2.6.12-git1 patch 376 # we now have a 2.6.12 kernel 377 $ patch -p1 < ../patch-2.6.13-rc2 # apply the 2.6.13-rc2 patch 378 # the kernel is now 2.6.13-rc2 379 $ patch -p1 < ../patch-2.6.13-rc2-git3 # apply the 2.6.13-rc2-git3 patch 380 # the kernel is now 2.6.13-rc2-git3 381 $ cd .. 382 $ mv linux-2.6.12-git1 linux-2.6.13-rc2-git3 # rename source dir 383 384 385 The -mm kernels 386 --- 387 These are experimental kernels released by Andrew Morton. 388 389 The -mm tree serves as a sort of proving ground for new features and other 390 experimental patches. 391 Once a patch has proved its worth in -mm for a while Andrew pushes it on to 392 Linus for inclusion in mainline. 393 394 Although it's encouraged that patches flow to Linus via the -mm tree, this 395 is not always enforced. 396 Subsystem maintainers (or individuals) sometimes push their patches directly 397 to Linus, even though (or after) they have been merged and tested in -mm (or 398 sometimes even without prior testing in -mm). 399 400 You should generally strive to get your patches into mainline via -mm to 401 ensure maximum testing. 402 403 This branch is in constant flux and contains many experimental features, a 404 lot of debugging patches not appropriate for mainline etc., and is the most 405 experimental of the branches described in this document. 406 407 These kernels are not appropriate for use on systems that are supposed to be 408 stable and they are more risky to run than any of the other branches (make 409 sure you have up-to-date backups -- that goes for any experimental kernel but 410 even more so for -mm kernels). 411 412 These kernels in addition to all the other experimental patches they contain 413 usually also contain any changes in the mainline -git kernels available at 414 the time of release. 415 416 Testing of -mm kernels is greatly appreciated since the whole point of the 417 tree is to weed out regressions, crashes, data corruption bugs, build 418 breakage (and any other bug in general) before changes are merged into the 419 more stable mainline Linus tree. 420 But testers of -mm should be aware that breakage in this tree is more common 421 than in any other tree. 422 423 The -mm kernels are not released on a fixed schedule, but usually a few -mm 424 kernels are released in between each -rc kernel (1 to 3 is common). 425 The -mm kernels apply to either a base 2.6.x kernel (when no -rc kernels 426 have been released yet) or to a Linus -rc kernel. 427 428 Here are some examples of applying the -mm patches: 429 430 # moving from 2.6.12 to 2.6.12-mm1 431 $ cd ~/linux-2.6.12 # change to the 2.6.12 source dir 432 $ patch -p1 < ../2.6.12-mm1 # apply the 2.6.12-mm1 patch 433 $ cd .. 434 $ mv linux-2.6.12 linux-2.6.12-mm1 # rename the source appropriately 435 436 # moving from 2.6.12-mm1 to 2.6.13-rc3-mm3 437 $ cd ~/linux-2.6.12-mm1 438 $ patch -p1 -R < ../2.6.12-mm1 # revert the 2.6.12-mm1 patch 439 # we now have a 2.6.12 source 440 $ patch -p1 < ../patch-2.6.13-rc3 # apply the 2.6.13-rc3 patch 441 # we now have a 2.6.13-rc3 source 442 $ patch -p1 < ../2.6.13-rc3-mm3 # apply the 2.6.13-rc3-mm3 patch 443 $ cd .. 444 $ mv linux-2.6.12-mm1 linux-2.6.13-rc3-mm3 # rename the source dir 445 446 447 This concludes this list of explanations of the various kernel trees. 448 I hope you are now clear on how to apply the various patches and help testing 449 the kernel. 450 451 Thank you's to Randy Dunlap, Rolf Eike Beer, Linus Torvalds, Bodo Eggert, 452 Johannes Stezenbach, Grant Coady, Pavel Machek and others that I may have 453 forgotten for their reviews and contributions to this document.