Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 1 | # Copyright (c) 2013 The Chromium OS Authors. |
| 2 | # |
Wolfgang Denk | 1a45966 | 2013-07-08 09:37:19 +0200 | [diff] [blame] | 3 | # SPDX-License-Identifier: GPL-2.0+ |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 4 | # |
| 5 | |
Simon Glass | 6eede34 | 2014-08-09 15:32:58 -0600 | [diff] [blame] | 6 | (Please read 'How to change from MAKEALL' if you are used to that tool) |
| 7 | |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 8 | What is this? |
| 9 | ============= |
| 10 | |
| 11 | This tool handles building U-Boot to check that you have not broken it |
| 12 | with your patch series. It can build each individual commit and report |
| 13 | which boards fail on which commits, and which errors come up. It aims |
| 14 | to make full use of multi-processor machines. |
| 15 | |
| 16 | A key feature of buildman is its output summary, which allows warnings, |
| 17 | errors or image size increases in a particular commit or board to be |
| 18 | quickly identified and the offending commit pinpointed. This can be a big |
| 19 | help for anyone working with >10 patches at a time. |
| 20 | |
| 21 | |
| 22 | Caveats |
| 23 | ======= |
| 24 | |
| 25 | Buildman is still in its infancy. It is already a very useful tool, but |
| 26 | expect to find problems and send patches. |
| 27 | |
| 28 | Buildman can be stopped and restarted, in which case it will continue |
| 29 | where it left off. This should happen cleanly and without side-effects. |
| 30 | If not, it is a bug, for which a patch would be welcome. |
| 31 | |
| 32 | Buildman gets so tied up in its work that it can ignore the outside world. |
| 33 | You may need to press Ctrl-C several times to quit it. Also it will print |
| 34 | out various exceptions when stopped. |
| 35 | |
| 36 | |
| 37 | Theory of Operation |
| 38 | =================== |
| 39 | |
| 40 | (please read this section in full twice or you will be perpetually confused) |
| 41 | |
| 42 | Buildman is a builder. It is not make, although it runs make. It does not |
| 43 | produce any useful output on the terminal while building, except for |
Simon Glass | e5a0e5d | 2014-08-09 15:33:03 -0600 | [diff] [blame] | 44 | progress information (except with -v, see below). All the output (errors, |
Dirk Behme | 3e1ded1 | 2014-12-23 07:41:26 +0100 | [diff] [blame] | 45 | warnings and binaries if you ask for them) is stored in output |
Simon Glass | e5a0e5d | 2014-08-09 15:33:03 -0600 | [diff] [blame] | 46 | directories, which you can look at while the build is progressing, or when |
| 47 | it is finished. |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 48 | |
| 49 | Buildman produces a concise summary of which boards succeeded and failed. |
| 50 | It shows which commit introduced which board failure using a simple |
| 51 | red/green colour coding. Full error information can be requested, in which |
| 52 | case it is de-duped and displayed against the commit that introduced the |
| 53 | error. An example workflow is below. |
| 54 | |
| 55 | Buildman stores image size information and can report changes in image size |
| 56 | from commit to commit. An example of this is below. |
| 57 | |
| 58 | Buildman starts multiple threads, and each thread builds for one board at |
| 59 | a time. A thread starts at the first commit, configures the source for your |
| 60 | board and builds it. Then it checks out the next commit and does an |
| 61 | incremental build. Eventually the thread reaches the last commit and stops. |
| 62 | If errors or warnings are found along the way, the thread will reconfigure |
| 63 | after every commit, and your build will be very slow. This is because a |
| 64 | file that produces just a warning would not normally be rebuilt in an |
| 65 | incremental build. |
| 66 | |
| 67 | Buildman works in an entirely separate place from your U-Boot repository. |
| 68 | It creates a separate working directory for each thread, and puts the |
| 69 | output files in the working directory, organised by commit name and board |
| 70 | name, in a two-level hierarchy. |
| 71 | |
| 72 | Buildman is invoked in your U-Boot directory, the one with the .git |
| 73 | directory. It clones this repository into a copy for each thread, and the |
| 74 | threads do not affect the state of your git repository. Any checkouts done |
| 75 | by the thread affect only the working directory for that thread. |
| 76 | |
Simon Glass | cec83c3 | 2014-08-09 15:32:57 -0600 | [diff] [blame] | 77 | Buildman automatically selects the correct tool chain for each board. You |
| 78 | must supply suitable tool chains, but buildman takes care of selecting the |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 79 | right one. |
| 80 | |
Simon Glass | e5a0e5d | 2014-08-09 15:33:03 -0600 | [diff] [blame] | 81 | Buildman generally builds a branch (with the -b flag), and in this case |
| 82 | builds the upstream commit as well, for comparison. It cannot build |
| 83 | individual commits at present, unless (maybe) you point it at an empty |
| 84 | branch. Put all your commits in a branch, set the branch's upstream to a |
| 85 | valid value, and all will be well. Otherwise buildman will perform random |
| 86 | actions. Use -n to check what the random actions might be. |
| 87 | |
Simon Glass | 1d8104f | 2014-10-16 01:05:56 -0600 | [diff] [blame] | 88 | If you just want to build the current source tree, leave off the -b flag |
| 89 | and add -e. This will display results and errors as they happen. You can |
| 90 | still look at them later using -se. Note that buildman will assume that the |
| 91 | source has changed, and will build all specified boards in this case. |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 92 | |
| 93 | Buildman is optimised for building many commits at once, for many boards. |
| 94 | On multi-core machines, Buildman is fast because it uses most of the |
| 95 | available CPU power. When it gets to the end, or if you are building just |
| 96 | a few commits or boards, it will be pretty slow. As a tip, if you don't |
| 97 | plan to use your machine for anything else, you can use -T to increase the |
| 98 | number of threads beyond the default. |
| 99 | |
Stephen Warren | 8426d8b | 2013-10-10 10:00:20 -0600 | [diff] [blame] | 100 | Buildman lets you build all boards, or a subset. Specify the subset by passing |
| 101 | command-line arguments that list the desired board name, architecture name, |
| 102 | SOC name, or anything else in the boards.cfg file. Multiple arguments are |
| 103 | allowed. Each argument will be interpreted as a regular expression, so |
| 104 | behaviour is a superset of exact or substring matching. Examples are: |
| 105 | |
| 106 | * 'tegra20' All boards with a Tegra20 SoC |
| 107 | * 'tegra' All boards with any Tegra Soc (Tegra20, Tegra30, Tegra114...) |
| 108 | * '^tegra[23]0$' All boards with either Tegra20 or Tegra30 SoC |
| 109 | * 'powerpc' All PowerPC boards |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 110 | |
Simon Glass | 6131bea | 2014-08-09 15:33:08 -0600 | [diff] [blame] | 111 | While the default is to OR the terms together, you can also make use of |
| 112 | the '&' operator to limit the selection: |
| 113 | |
| 114 | * 'freescale & arm sandbox' All Freescale boards with ARM architecture, |
| 115 | plus sandbox |
| 116 | |
Simon Glass | 3cf4ae6 | 2014-08-28 09:43:41 -0600 | [diff] [blame] | 117 | You can also use -x to specifically exclude some boards. For example: |
| 118 | |
| 119 | buildmand arm -x nvidia,freescale,.*ball$ |
| 120 | |
| 121 | means to build all arm boards except nvidia, freescale and anything ending |
| 122 | with 'ball'. |
| 123 | |
Dirk Behme | 3e1ded1 | 2014-12-23 07:41:26 +0100 | [diff] [blame] | 124 | It is convenient to use the -n option to see what will be built based on |
Simon Glass | 6131bea | 2014-08-09 15:33:08 -0600 | [diff] [blame] | 125 | the subset given. |
| 126 | |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 127 | Buildman does not store intermediate object files. It optionally copies |
| 128 | the binary output into a directory when a build is successful. Size |
| 129 | information is always recorded. It needs a fair bit of disk space to work, |
| 130 | typically 250MB per thread. |
| 131 | |
| 132 | |
| 133 | Setting up |
| 134 | ========== |
| 135 | |
| 136 | 1. Get the U-Boot source. You probably already have it, but if not these |
| 137 | steps should get you started with a repo and some commits for testing. |
| 138 | |
| 139 | $ cd /path/to/u-boot |
| 140 | $ git clone git://git.denx.de/u-boot.git . |
| 141 | $ git checkout -b my-branch origin/master |
| 142 | $ # Add some commits to the branch, reading for testing |
| 143 | |
Simon Glass | 6200534 | 2014-12-01 17:34:02 -0700 | [diff] [blame] | 144 | 2. Create ~/.buildman to tell buildman where to find tool chains (see 'The |
| 145 | .buildman file' later for details). As an example: |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 146 | |
| 147 | # Buildman settings file |
| 148 | |
| 149 | [toolchain] |
| 150 | root: / |
| 151 | rest: /toolchains/* |
| 152 | eldk: /opt/eldk-4.2 |
Simon Glass | e956947 | 2014-08-09 15:33:07 -0600 | [diff] [blame] | 153 | arm: /opt/linaro/gcc-linaro-arm-linux-gnueabihf-4.8-2013.08_linux |
| 154 | aarch64: /opt/linaro/gcc-linaro-aarch64-none-elf-4.8-2013.10_linux |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 155 | |
| 156 | [toolchain-alias] |
| 157 | x86: i386 |
| 158 | blackfin: bfin |
| 159 | sh: sh4 |
| 160 | nds32: nds32le |
| 161 | openrisc: or32 |
| 162 | |
| 163 | |
| 164 | This selects the available toolchain paths. Add the base directory for |
| 165 | each of your toolchains here. Buildman will search inside these directories |
| 166 | and also in any '/usr' and '/usr/bin' subdirectories. |
| 167 | |
| 168 | Make sure the tags (here root: rest: and eldk:) are unique. |
| 169 | |
| 170 | The toolchain-alias section indicates that the i386 toolchain should be used |
| 171 | to build x86 commits. |
| 172 | |
| 173 | |
Simon Glass | 3469969 | 2014-12-01 17:34:01 -0700 | [diff] [blame] | 174 | 3. Make sure you have the require Python pre-requisites |
| 175 | |
Simon Glass | 827e37b | 2014-12-01 17:34:06 -0700 | [diff] [blame] | 176 | Buildman uses multiprocessing, Queue, shutil, StringIO, ConfigParser and |
| 177 | urllib2. These should normally be available, but if you get an error like |
| 178 | this then you will need to obtain those modules: |
Simon Glass | 3469969 | 2014-12-01 17:34:01 -0700 | [diff] [blame] | 179 | |
| 180 | ImportError: No module named multiprocessing |
| 181 | |
| 182 | |
| 183 | 4. Check the available toolchains |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 184 | |
| 185 | Run this check to make sure that you have a toolchain for every architecture. |
| 186 | |
| 187 | $ ./tools/buildman/buildman --list-tool-chains |
| 188 | Scanning for tool chains |
| 189 | - scanning path '/' |
| 190 | - looking in '/.' |
| 191 | - looking in '/bin' |
| 192 | - looking in '/usr/bin' |
| 193 | - found '/usr/bin/gcc' |
| 194 | Tool chain test: OK |
| 195 | - found '/usr/bin/c89-gcc' |
| 196 | Tool chain test: OK |
| 197 | - found '/usr/bin/c99-gcc' |
| 198 | Tool chain test: OK |
| 199 | - found '/usr/bin/x86_64-linux-gnu-gcc' |
| 200 | Tool chain test: OK |
| 201 | - scanning path '/toolchains/powerpc-linux' |
| 202 | - looking in '/toolchains/powerpc-linux/.' |
| 203 | - looking in '/toolchains/powerpc-linux/bin' |
| 204 | - found '/toolchains/powerpc-linux/bin/powerpc-linux-gcc' |
| 205 | Tool chain test: OK |
| 206 | - looking in '/toolchains/powerpc-linux/usr/bin' |
| 207 | - scanning path '/toolchains/nds32le-linux-glibc-v1f' |
| 208 | - looking in '/toolchains/nds32le-linux-glibc-v1f/.' |
| 209 | - looking in '/toolchains/nds32le-linux-glibc-v1f/bin' |
| 210 | - found '/toolchains/nds32le-linux-glibc-v1f/bin/nds32le-linux-gcc' |
| 211 | Tool chain test: OK |
| 212 | - looking in '/toolchains/nds32le-linux-glibc-v1f/usr/bin' |
| 213 | - scanning path '/toolchains/nios2' |
| 214 | - looking in '/toolchains/nios2/.' |
| 215 | - looking in '/toolchains/nios2/bin' |
| 216 | - found '/toolchains/nios2/bin/nios2-linux-gcc' |
| 217 | Tool chain test: OK |
| 218 | - found '/toolchains/nios2/bin/nios2-linux-uclibc-gcc' |
| 219 | Tool chain test: OK |
| 220 | - looking in '/toolchains/nios2/usr/bin' |
| 221 | - found '/toolchains/nios2/usr/bin/nios2-linux-gcc' |
| 222 | Tool chain test: OK |
| 223 | - found '/toolchains/nios2/usr/bin/nios2-linux-uclibc-gcc' |
| 224 | Tool chain test: OK |
| 225 | - scanning path '/toolchains/microblaze-unknown-linux-gnu' |
| 226 | - looking in '/toolchains/microblaze-unknown-linux-gnu/.' |
| 227 | - looking in '/toolchains/microblaze-unknown-linux-gnu/bin' |
| 228 | - found '/toolchains/microblaze-unknown-linux-gnu/bin/microblaze-unknown-linux-gnu-gcc' |
| 229 | Tool chain test: OK |
| 230 | - found '/toolchains/microblaze-unknown-linux-gnu/bin/mb-linux-gcc' |
| 231 | Tool chain test: OK |
| 232 | - looking in '/toolchains/microblaze-unknown-linux-gnu/usr/bin' |
| 233 | - scanning path '/toolchains/mips-linux' |
| 234 | - looking in '/toolchains/mips-linux/.' |
| 235 | - looking in '/toolchains/mips-linux/bin' |
| 236 | - found '/toolchains/mips-linux/bin/mips-linux-gcc' |
| 237 | Tool chain test: OK |
| 238 | - looking in '/toolchains/mips-linux/usr/bin' |
| 239 | - scanning path '/toolchains/old' |
| 240 | - looking in '/toolchains/old/.' |
| 241 | - looking in '/toolchains/old/bin' |
| 242 | - looking in '/toolchains/old/usr/bin' |
| 243 | - scanning path '/toolchains/i386-linux' |
| 244 | - looking in '/toolchains/i386-linux/.' |
| 245 | - looking in '/toolchains/i386-linux/bin' |
| 246 | - found '/toolchains/i386-linux/bin/i386-linux-gcc' |
| 247 | Tool chain test: OK |
| 248 | - looking in '/toolchains/i386-linux/usr/bin' |
| 249 | - scanning path '/toolchains/bfin-uclinux' |
| 250 | - looking in '/toolchains/bfin-uclinux/.' |
| 251 | - looking in '/toolchains/bfin-uclinux/bin' |
| 252 | - found '/toolchains/bfin-uclinux/bin/bfin-uclinux-gcc' |
| 253 | Tool chain test: OK |
| 254 | - looking in '/toolchains/bfin-uclinux/usr/bin' |
| 255 | - scanning path '/toolchains/sparc-elf' |
| 256 | - looking in '/toolchains/sparc-elf/.' |
| 257 | - looking in '/toolchains/sparc-elf/bin' |
| 258 | - found '/toolchains/sparc-elf/bin/sparc-elf-gcc' |
| 259 | Tool chain test: OK |
| 260 | - looking in '/toolchains/sparc-elf/usr/bin' |
| 261 | - scanning path '/toolchains/arm-2010q1' |
| 262 | - looking in '/toolchains/arm-2010q1/.' |
| 263 | - looking in '/toolchains/arm-2010q1/bin' |
| 264 | - found '/toolchains/arm-2010q1/bin/arm-none-linux-gnueabi-gcc' |
| 265 | Tool chain test: OK |
| 266 | - looking in '/toolchains/arm-2010q1/usr/bin' |
| 267 | - scanning path '/toolchains/from' |
| 268 | - looking in '/toolchains/from/.' |
| 269 | - looking in '/toolchains/from/bin' |
| 270 | - looking in '/toolchains/from/usr/bin' |
| 271 | - scanning path '/toolchains/sh4-gentoo-linux-gnu' |
| 272 | - looking in '/toolchains/sh4-gentoo-linux-gnu/.' |
| 273 | - looking in '/toolchains/sh4-gentoo-linux-gnu/bin' |
| 274 | - found '/toolchains/sh4-gentoo-linux-gnu/bin/sh4-gentoo-linux-gnu-gcc' |
| 275 | Tool chain test: OK |
| 276 | - looking in '/toolchains/sh4-gentoo-linux-gnu/usr/bin' |
| 277 | - scanning path '/toolchains/avr32-linux' |
| 278 | - looking in '/toolchains/avr32-linux/.' |
| 279 | - looking in '/toolchains/avr32-linux/bin' |
| 280 | - found '/toolchains/avr32-linux/bin/avr32-gcc' |
| 281 | Tool chain test: OK |
| 282 | - looking in '/toolchains/avr32-linux/usr/bin' |
| 283 | - scanning path '/toolchains/m68k-linux' |
| 284 | - looking in '/toolchains/m68k-linux/.' |
| 285 | - looking in '/toolchains/m68k-linux/bin' |
| 286 | - found '/toolchains/m68k-linux/bin/m68k-linux-gcc' |
| 287 | Tool chain test: OK |
| 288 | - looking in '/toolchains/m68k-linux/usr/bin' |
| 289 | List of available toolchains (17): |
| 290 | arm : /toolchains/arm-2010q1/bin/arm-none-linux-gnueabi-gcc |
| 291 | avr32 : /toolchains/avr32-linux/bin/avr32-gcc |
| 292 | bfin : /toolchains/bfin-uclinux/bin/bfin-uclinux-gcc |
| 293 | c89 : /usr/bin/c89-gcc |
| 294 | c99 : /usr/bin/c99-gcc |
| 295 | i386 : /toolchains/i386-linux/bin/i386-linux-gcc |
| 296 | m68k : /toolchains/m68k-linux/bin/m68k-linux-gcc |
| 297 | mb : /toolchains/microblaze-unknown-linux-gnu/bin/mb-linux-gcc |
| 298 | microblaze: /toolchains/microblaze-unknown-linux-gnu/bin/microblaze-unknown-linux-gnu-gcc |
| 299 | mips : /toolchains/mips-linux/bin/mips-linux-gcc |
| 300 | nds32le : /toolchains/nds32le-linux-glibc-v1f/bin/nds32le-linux-gcc |
| 301 | nios2 : /toolchains/nios2/bin/nios2-linux-gcc |
| 302 | powerpc : /toolchains/powerpc-linux/bin/powerpc-linux-gcc |
| 303 | sandbox : /usr/bin/gcc |
| 304 | sh4 : /toolchains/sh4-gentoo-linux-gnu/bin/sh4-gentoo-linux-gnu-gcc |
| 305 | sparc : /toolchains/sparc-elf/bin/sparc-elf-gcc |
| 306 | x86_64 : /usr/bin/x86_64-linux-gnu-gcc |
| 307 | |
| 308 | |
| 309 | You can see that everything is covered, even some strange ones that won't |
| 310 | be used (c88 and c99). This is a feature. |
| 311 | |
| 312 | |
Simon Glass | 827e37b | 2014-12-01 17:34:06 -0700 | [diff] [blame] | 313 | 5. Install new toolchains if needed |
| 314 | |
| 315 | You can download toolchains and update the [toolchain] section of the |
| 316 | settings file to find them. |
| 317 | |
| 318 | To make this easier, buildman can automatically download and install |
| 319 | toolchains from kernel.org. First list the available architectures: |
| 320 | |
Bin Meng | 9f244b2 | 2015-07-16 19:43:46 -0700 | [diff] [blame] | 321 | $ ./tools/buildman/buildman --fetch-arch list |
Simon Glass | 827e37b | 2014-12-01 17:34:06 -0700 | [diff] [blame] | 322 | Checking: https://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.6.3/ |
| 323 | Checking: https://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.6.2/ |
| 324 | Checking: https://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.5.1/ |
| 325 | Checking: https://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.2.4/ |
| 326 | Available architectures: alpha am33_2.0 arm avr32 bfin cris crisv32 frv h8300 |
| 327 | hppa hppa64 i386 ia64 m32r m68k mips mips64 or32 powerpc powerpc64 s390x sh4 |
| 328 | sparc sparc64 tilegx x86_64 xtensa |
| 329 | |
| 330 | Then pick one and download it: |
| 331 | |
Bin Meng | 9f244b2 | 2015-07-16 19:43:46 -0700 | [diff] [blame] | 332 | $ ./tools/buildman/buildman --fetch-arch or32 |
Simon Glass | 827e37b | 2014-12-01 17:34:06 -0700 | [diff] [blame] | 333 | Checking: https://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.6.3/ |
| 334 | Checking: https://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.6.2/ |
| 335 | Checking: https://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.5.1/ |
| 336 | Downloading: https://www.kernel.org/pub/tools/crosstool/files/bin/x86_64/4.5.1//x86_64-gcc-4.5.1-nolibc_or32-linux.tar.xz |
| 337 | Unpacking to: /home/sjg/.buildman-toolchains |
| 338 | Testing |
| 339 | - looking in '/home/sjg/.buildman-toolchains/gcc-4.5.1-nolibc/or32-linux/.' |
| 340 | - looking in '/home/sjg/.buildman-toolchains/gcc-4.5.1-nolibc/or32-linux/bin' |
| 341 | - found '/home/sjg/.buildman-toolchains/gcc-4.5.1-nolibc/or32-linux/bin/or32-linux-gcc' |
| 342 | Tool chain test: OK |
| 343 | |
| 344 | Buildman should now be set up to use your new toolchain. |
| 345 | |
| 346 | At the time of writing, U-Boot has these architectures: |
| 347 | |
| 348 | arc, arm, avr32, blackfin, m68k, microblaze, mips, nds32, nios2, openrisc |
| 349 | powerpc, sandbox, sh, sparc, x86 |
| 350 | |
Michal Simek | 1246231 | 2015-04-20 11:46:24 +0200 | [diff] [blame] | 351 | Of these, only arc and nds32 are not available at kernel.org.. |
Simon Glass | 827e37b | 2014-12-01 17:34:06 -0700 | [diff] [blame] | 352 | |
| 353 | |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 354 | How to run it |
| 355 | ============= |
| 356 | |
| 357 | First do a dry run using the -n flag: (replace <branch> with a real, local |
| 358 | branch with a valid upstream) |
| 359 | |
| 360 | $ ./tools/buildman/buildman -b <branch> -n |
| 361 | |
| 362 | If it can't detect the upstream branch, try checking out the branch, and |
Simon Glass | 2a9e2c6 | 2014-12-01 17:33:54 -0700 | [diff] [blame] | 363 | doing something like 'git branch --set-upstream-to upstream/master' |
| 364 | or something similar. Buildman will try to guess a suitable upstream branch |
| 365 | if it can't find one (you will see a message like" Guessing upstream as ...). |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 366 | |
Simon Glass | cec83c3 | 2014-08-09 15:32:57 -0600 | [diff] [blame] | 367 | As an example: |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 368 | |
| 369 | Dry run, so not doing much. But I would do this: |
| 370 | |
| 371 | Building 18 commits for 1059 boards (4 threads, 1 job per thread) |
| 372 | Build directory: ../lcd9b |
| 373 | 5bb3505 Merge branch 'master' of git://git.denx.de/u-boot-arm |
| 374 | c18f1b4 tegra: Use const for pinmux_config_pingroup/table() |
| 375 | 2f043ae tegra: Add display support to funcmux |
| 376 | e349900 tegra: fdt: Add pwm binding and node |
| 377 | 424a5f0 tegra: fdt: Add LCD definitions for Tegra |
| 378 | 0636ccf tegra: Add support for PWM |
| 379 | a994fe7 tegra: Add SOC support for display/lcd |
| 380 | fcd7350 tegra: Add LCD driver |
| 381 | 4d46e9d tegra: Add LCD support to Nvidia boards |
| 382 | 991bd48 arm: Add control over cachability of memory regions |
| 383 | 54e8019 lcd: Add CONFIG_LCD_ALIGNMENT to select frame buffer alignment |
| 384 | d92aff7 lcd: Add support for flushing LCD fb from dcache after update |
| 385 | dbd0677 tegra: Align LCD frame buffer to section boundary |
| 386 | 0cff9b8 tegra: Support control of cache settings for LCD |
| 387 | 9c56900 tegra: fdt: Add LCD definitions for Seaboard |
| 388 | 5cc29db lcd: Add CONFIG_CONSOLE_SCROLL_LINES option to speed console |
| 389 | cac5a23 tegra: Enable display/lcd support on Seaboard |
| 390 | 49ff541 wip |
| 391 | |
| 392 | Total boards to build for each commit: 1059 |
| 393 | |
| 394 | This shows that it will build all 1059 boards, using 4 threads (because |
| 395 | we have a 4-core CPU). Each thread will run with -j1, meaning that each |
| 396 | make job will use a single CPU. The list of commits to be built helps you |
| 397 | confirm that things look about right. Notice that buildman has chosen a |
| 398 | 'base' directory for you, immediately above your source tree. |
| 399 | |
| 400 | Buildman works entirely inside the base directory, here ../lcd9b, |
| 401 | creating a working directory for each thread, and creating output |
| 402 | directories for each commit and board. |
| 403 | |
| 404 | |
| 405 | Suggested Workflow |
| 406 | ================== |
| 407 | |
| 408 | To run the build for real, take off the -n: |
| 409 | |
| 410 | $ ./tools/buildman/buildman -b <branch> |
| 411 | |
| 412 | Buildman will set up some working directories, and get started. After a |
| 413 | minute or so it will settle down to a steady pace, with a display like this: |
| 414 | |
| 415 | Building 18 commits for 1059 boards (4 threads, 1 job per thread) |
| 416 | 528 36 124 /19062 1:13:30 : SIMPC8313_SP |
| 417 | |
| 418 | This means that it is building 19062 board/commit combinations. So far it |
Simon Glass | cec83c3 | 2014-08-09 15:32:57 -0600 | [diff] [blame] | 419 | has managed to successfully build 528. Another 36 have built with warnings, |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 420 | and 124 more didn't build at all. Buildman expects to complete the process |
| 421 | in an hour and 15 minutes. Use this time to buy a faster computer. |
| 422 | |
| 423 | |
| 424 | To find out how the build went, ask for a summary with -s. You can do this |
Dirk Behme | 3e1ded1 | 2014-12-23 07:41:26 +0100 | [diff] [blame] | 425 | either before the build completes (presumably in another terminal) or |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 426 | afterwards. Let's work through an example of how this is used: |
| 427 | |
| 428 | $ ./tools/buildman/buildman -b lcd9b -s |
| 429 | ... |
| 430 | 01: Merge branch 'master' of git://git.denx.de/u-boot-arm |
| 431 | powerpc: + galaxy5200_LOWBOOT |
| 432 | 02: tegra: Use const for pinmux_config_pingroup/table() |
| 433 | 03: tegra: Add display support to funcmux |
| 434 | 04: tegra: fdt: Add pwm binding and node |
| 435 | 05: tegra: fdt: Add LCD definitions for Tegra |
| 436 | 06: tegra: Add support for PWM |
| 437 | 07: tegra: Add SOC support for display/lcd |
| 438 | 08: tegra: Add LCD driver |
| 439 | 09: tegra: Add LCD support to Nvidia boards |
| 440 | 10: arm: Add control over cachability of memory regions |
| 441 | 11: lcd: Add CONFIG_LCD_ALIGNMENT to select frame buffer alignment |
| 442 | 12: lcd: Add support for flushing LCD fb from dcache after update |
| 443 | arm: + lubbock |
| 444 | 13: tegra: Align LCD frame buffer to section boundary |
| 445 | 14: tegra: Support control of cache settings for LCD |
| 446 | 15: tegra: fdt: Add LCD definitions for Seaboard |
| 447 | 16: lcd: Add CONFIG_CONSOLE_SCROLL_LINES option to speed console |
| 448 | 17: tegra: Enable display/lcd support on Seaboard |
| 449 | 18: wip |
| 450 | |
| 451 | This shows which commits have succeeded and which have failed. In this case |
| 452 | the build is still in progress so many boards are not built yet (use -u to |
| 453 | see which ones). But still we can see a few failures. The galaxy5200_LOWBOOT |
| 454 | never builds correctly. This could be a problem with our toolchain, or it |
| 455 | could be a bug in the upstream. The good news is that we probably don't need |
| 456 | to blame our commits. The bad news is it isn't tested on that board. |
| 457 | |
| 458 | Commit 12 broke lubbock. That's what the '+ lubbock' means. The failure |
| 459 | is never fixed by a later commit, or you would see lubbock again, in green, |
| 460 | without the +. |
| 461 | |
| 462 | To see the actual error: |
| 463 | |
| 464 | $ ./tools/buildman/buildman -b <branch> -se lubbock |
| 465 | ... |
| 466 | 12: lcd: Add support for flushing LCD fb from dcache after update |
| 467 | arm: + lubbock |
| 468 | +common/libcommon.o: In function `lcd_sync': |
| 469 | +/u-boot/lcd9b/.bm-work/00/common/lcd.c:120: undefined reference to `flush_dcache_range' |
| 470 | +arm-none-linux-gnueabi-ld: BFD (Sourcery G++ Lite 2010q1-202) 2.19.51.20090709 assertion fail /scratch/julian/2010q1-release-linux-lite/obj/binutils-src-2010q1-202-arm-none-linux-gnueabi-i686-pc-linux-gnu/bfd/elf32-arm.c:12572 |
| 471 | +make: *** [/u-boot/lcd9b/.bm-work/00/build/u-boot] Error 139 |
| 472 | 13: tegra: Align LCD frame buffer to section boundary |
| 473 | 14: tegra: Support control of cache settings for LCD |
| 474 | 15: tegra: fdt: Add LCD definitions for Seaboard |
| 475 | 16: lcd: Add CONFIG_CONSOLE_SCROLL_LINES option to speed console |
| 476 | -/u-boot/lcd9b/.bm-work/00/common/lcd.c:120: undefined reference to `flush_dcache_range' |
| 477 | +/u-boot/lcd9b/.bm-work/00/common/lcd.c:125: undefined reference to `flush_dcache_range' |
| 478 | 17: tegra: Enable display/lcd support on Seaboard |
| 479 | 18: wip |
| 480 | |
| 481 | So the problem is in lcd.c, due to missing cache operations. This information |
| 482 | should be enough to work out what that commit is doing to break these |
| 483 | boards. (In this case pxa did not have cache operations defined). |
| 484 | |
| 485 | If you see error lines marked with - that means that the errors were fixed |
| 486 | by that commit. Sometimes commits can be in the wrong order, so that a |
| 487 | breakage is introduced for a few commits and fixed by later commits. This |
| 488 | shows up clearly with buildman. You can then reorder the commits and try |
| 489 | again. |
| 490 | |
| 491 | At commit 16, the error moves - you can see that the old error at line 120 |
| 492 | is fixed, but there is a new one at line 126. This is probably only because |
Dirk Behme | 3e1ded1 | 2014-12-23 07:41:26 +0100 | [diff] [blame] | 493 | we added some code and moved the broken line further down the file. |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 494 | |
| 495 | If many boards have the same error, then -e will display the error only |
Simon Glass | ed96665 | 2014-08-28 09:43:43 -0600 | [diff] [blame] | 496 | once. This makes the output as concise as possible. To see which boards have |
| 497 | each error, use -l. |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 498 | |
Simon Glass | e30965d | 2014-08-28 09:43:44 -0600 | [diff] [blame] | 499 | Buildman tries to distinguish warnings from errors, and shows warning lines |
| 500 | separately with a 'w' prefix. |
| 501 | |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 502 | The full build output in this case is available in: |
| 503 | |
| 504 | ../lcd9b/12_of_18_gd92aff7_lcd--Add-support-for/lubbock/ |
| 505 | |
| 506 | done: Indicates the build was done, and holds the return code from make. |
| 507 | This is 0 for a good build, typically 2 for a failure. |
| 508 | |
| 509 | err: Output from stderr, if any. Errors and warnings appear here. |
| 510 | |
| 511 | log: Output from stdout. Normally there isn't any since buildman runs |
| 512 | in silent mode for now. |
| 513 | |
| 514 | toolchain: Shows information about the toolchain used for the build. |
| 515 | |
| 516 | sizes: Shows image size information. |
| 517 | |
| 518 | It is possible to get the build output there also. Use the -k option for |
| 519 | this. In that case you will also see some output files, like: |
| 520 | |
| 521 | System.map toolchain u-boot u-boot.bin u-boot.map autoconf.mk |
| 522 | (also SPL versions u-boot-spl and u-boot-spl.bin if available) |
| 523 | |
| 524 | |
| 525 | Checking Image Sizes |
| 526 | ==================== |
| 527 | |
| 528 | A key requirement for U-Boot is that you keep code/data size to a minimum. |
| 529 | Where a new feature increases this noticeably it should normally be put |
| 530 | behind a CONFIG flag so that boards can leave it off and keep the image |
| 531 | size more or less the same with each new release. |
| 532 | |
| 533 | To check the impact of your commits on image size, use -S. For example: |
| 534 | |
| 535 | $ ./tools/buildman/buildman -b us-x86 -sS |
| 536 | Summary of 10 commits for 1066 boards (4 threads, 1 job per thread) |
| 537 | 01: MAKEALL: add support for per architecture toolchains |
| 538 | 02: x86: Add function to get top of usable ram |
| 539 | x86: (for 1/3 boards) text -272.0 rodata +41.0 |
| 540 | 03: x86: Add basic cache operations |
| 541 | 04: x86: Permit bootstage and timer data to be used prior to relocation |
| 542 | x86: (for 1/3 boards) data +16.0 |
| 543 | 05: x86: Add an __end symbol to signal the end of the U-Boot binary |
| 544 | x86: (for 1/3 boards) text +76.0 |
| 545 | 06: x86: Rearrange the output input to remove BSS |
| 546 | x86: (for 1/3 boards) bss -2140.0 |
| 547 | 07: x86: Support relocation of FDT on start-up |
| 548 | x86: + coreboot-x86 |
| 549 | 08: x86: Add error checking to x86 relocation code |
| 550 | 09: x86: Adjust link device tree include file |
| 551 | 10: x86: Enable CONFIG_OF_CONTROL on coreboot |
| 552 | |
| 553 | |
| 554 | You can see that image size only changed on x86, which is good because this |
| 555 | series is not supposed to change any other board. From commit 7 onwards the |
| 556 | build fails so we don't get code size numbers. The numbers are fractional |
| 557 | because they are an average of all boards for that architecture. The |
| 558 | intention is to allow you to quickly find image size problems introduced by |
| 559 | your commits. |
| 560 | |
| 561 | Note that the 'text' region and 'rodata' are split out. You should add the |
| 562 | two together to get the total read-only size (reported as the first column |
| 563 | in the output from binutil's 'size' utility). |
| 564 | |
| 565 | A useful option is --step which lets you skip some commits. For example |
| 566 | --step 2 will show the image sizes for only every 2nd commit (so it will |
| 567 | compare the image sizes of the 1st, 3rd, 5th... commits). You can also use |
| 568 | --step 0 which will compare only the first and last commits. This is useful |
| 569 | for an overview of how your entire series affects code size. |
| 570 | |
| 571 | You can also use -d to see a detailed size breakdown for each board. This |
| 572 | list is sorted in order from largest growth to largest reduction. |
| 573 | |
| 574 | It is possible to go a little further with the -B option (--bloat). This |
Simon Glass | cec83c3 | 2014-08-09 15:32:57 -0600 | [diff] [blame] | 575 | shows where U-Boot has bloated, breaking the size change down to the function |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 576 | level. Example output is below: |
| 577 | |
| 578 | $ ./tools/buildman/buildman -b us-mem4 -sSdB |
| 579 | ... |
| 580 | 19: Roll crc32 into hash infrastructure |
| 581 | arm: (for 10/10 boards) all -143.4 bss +1.2 data -4.8 rodata -48.2 text -91.6 |
| 582 | paz00 : all +23 bss -4 rodata -29 text +56 |
| 583 | u-boot: add: 1/0, grow: 3/-2 bytes: 168/-104 (64) |
| 584 | function old new delta |
| 585 | hash_command 80 160 +80 |
| 586 | crc32_wd_buf - 56 +56 |
| 587 | ext4fs_read_file 540 568 +28 |
| 588 | insert_var_value_sub 688 692 +4 |
| 589 | run_list_real 1996 1992 -4 |
| 590 | do_mem_crc 168 68 -100 |
| 591 | trimslice : all -9 bss +16 rodata -29 text +4 |
| 592 | u-boot: add: 1/0, grow: 1/-3 bytes: 136/-124 (12) |
| 593 | function old new delta |
| 594 | hash_command 80 160 +80 |
| 595 | crc32_wd_buf - 56 +56 |
| 596 | ext4fs_iterate_dir 672 668 -4 |
| 597 | ext4fs_read_file 568 548 -20 |
| 598 | do_mem_crc 168 68 -100 |
| 599 | whistler : all -9 bss +16 rodata -29 text +4 |
| 600 | u-boot: add: 1/0, grow: 1/-3 bytes: 136/-124 (12) |
| 601 | function old new delta |
| 602 | hash_command 80 160 +80 |
| 603 | crc32_wd_buf - 56 +56 |
| 604 | ext4fs_iterate_dir 672 668 -4 |
| 605 | ext4fs_read_file 568 548 -20 |
| 606 | do_mem_crc 168 68 -100 |
| 607 | seaboard : all -9 bss -28 rodata -29 text +48 |
| 608 | u-boot: add: 1/0, grow: 3/-2 bytes: 160/-104 (56) |
| 609 | function old new delta |
| 610 | hash_command 80 160 +80 |
| 611 | crc32_wd_buf - 56 +56 |
| 612 | ext4fs_read_file 548 568 +20 |
| 613 | run_list_real 1996 2000 +4 |
| 614 | do_nandboot 760 756 -4 |
| 615 | do_mem_crc 168 68 -100 |
Marcel Ziswiler | e57c6e5 | 2015-03-26 01:31:54 +0100 | [diff] [blame] | 616 | colibri_t20 : all -9 rodata -29 text +20 |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 617 | u-boot: add: 1/0, grow: 2/-3 bytes: 140/-112 (28) |
| 618 | function old new delta |
| 619 | hash_command 80 160 +80 |
| 620 | crc32_wd_buf - 56 +56 |
| 621 | read_abs_bbt 204 208 +4 |
| 622 | do_nandboot 760 756 -4 |
| 623 | ext4fs_read_file 576 568 -8 |
| 624 | do_mem_crc 168 68 -100 |
| 625 | ventana : all -37 bss -12 rodata -29 text +4 |
| 626 | u-boot: add: 1/0, grow: 1/-3 bytes: 136/-124 (12) |
| 627 | function old new delta |
| 628 | hash_command 80 160 +80 |
| 629 | crc32_wd_buf - 56 +56 |
| 630 | ext4fs_iterate_dir 672 668 -4 |
| 631 | ext4fs_read_file 568 548 -20 |
| 632 | do_mem_crc 168 68 -100 |
| 633 | harmony : all -37 bss -16 rodata -29 text +8 |
| 634 | u-boot: add: 1/0, grow: 2/-3 bytes: 140/-124 (16) |
| 635 | function old new delta |
| 636 | hash_command 80 160 +80 |
| 637 | crc32_wd_buf - 56 +56 |
| 638 | nand_write_oob_syndrome 428 432 +4 |
| 639 | ext4fs_iterate_dir 672 668 -4 |
| 640 | ext4fs_read_file 568 548 -20 |
| 641 | do_mem_crc 168 68 -100 |
| 642 | medcom-wide : all -417 bss +28 data -16 rodata -93 text -336 |
| 643 | u-boot: add: 1/-1, grow: 1/-2 bytes: 88/-376 (-288) |
| 644 | function old new delta |
| 645 | crc32_wd_buf - 56 +56 |
| 646 | do_fat_read_at 2872 2904 +32 |
| 647 | hash_algo 16 - -16 |
| 648 | do_mem_crc 168 68 -100 |
| 649 | hash_command 420 160 -260 |
| 650 | tec : all -449 bss -4 data -16 rodata -93 text -336 |
| 651 | u-boot: add: 1/-1, grow: 1/-2 bytes: 88/-376 (-288) |
| 652 | function old new delta |
| 653 | crc32_wd_buf - 56 +56 |
| 654 | do_fat_read_at 2872 2904 +32 |
| 655 | hash_algo 16 - -16 |
| 656 | do_mem_crc 168 68 -100 |
| 657 | hash_command 420 160 -260 |
| 658 | plutux : all -481 bss +16 data -16 rodata -93 text -388 |
| 659 | u-boot: add: 1/-1, grow: 1/-3 bytes: 68/-408 (-340) |
| 660 | function old new delta |
| 661 | crc32_wd_buf - 56 +56 |
| 662 | do_load_serial_bin 1688 1700 +12 |
| 663 | hash_algo 16 - -16 |
| 664 | do_fat_read_at 2904 2872 -32 |
| 665 | do_mem_crc 168 68 -100 |
| 666 | hash_command 420 160 -260 |
| 667 | powerpc: (for 5/5 boards) all +37.4 data -3.2 rodata -41.8 text +82.4 |
| 668 | MPC8610HPCD : all +55 rodata -29 text +84 |
| 669 | u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80) |
| 670 | function old new delta |
| 671 | hash_command - 176 +176 |
| 672 | do_mem_crc 184 88 -96 |
| 673 | MPC8641HPCN : all +55 rodata -29 text +84 |
| 674 | u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80) |
| 675 | function old new delta |
| 676 | hash_command - 176 +176 |
| 677 | do_mem_crc 184 88 -96 |
| 678 | MPC8641HPCN_36BIT: all +55 rodata -29 text +84 |
| 679 | u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80) |
| 680 | function old new delta |
| 681 | hash_command - 176 +176 |
| 682 | do_mem_crc 184 88 -96 |
| 683 | sbc8641d : all +55 rodata -29 text +84 |
| 684 | u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80) |
| 685 | function old new delta |
| 686 | hash_command - 176 +176 |
| 687 | do_mem_crc 184 88 -96 |
| 688 | xpedite517x : all -33 data -16 rodata -93 text +76 |
| 689 | u-boot: add: 1/-1, grow: 0/-1 bytes: 176/-112 (64) |
| 690 | function old new delta |
| 691 | hash_command - 176 +176 |
| 692 | hash_algo 16 - -16 |
| 693 | do_mem_crc 184 88 -96 |
| 694 | ... |
| 695 | |
| 696 | |
| 697 | This shows that commit 19 has increased text size for arm (although only one |
| 698 | board was built) and by 96 bytes for powerpc. This increase was offset in both |
| 699 | cases by reductions in rodata and data/bss. |
| 700 | |
Dirk Behme | 3e1ded1 | 2014-12-23 07:41:26 +0100 | [diff] [blame] | 701 | Shown below the summary lines are the sizes for each board. Below each board |
| 702 | are the sizes for each function. This information starts with: |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 703 | |
| 704 | add - number of functions added / removed |
| 705 | grow - number of functions which grew / shrunk |
| 706 | bytes - number of bytes of code added to / removed from all functions, |
| 707 | plus the total byte change in brackets |
| 708 | |
| 709 | The change seems to be that hash_command() has increased by more than the |
| 710 | do_mem_crc() function has decreased. The function sizes typically add up to |
| 711 | roughly the text area size, but note that every read-only section except |
| 712 | rodata is included in 'text', so the function total does not exactly |
| 713 | correspond. |
| 714 | |
| 715 | It is common when refactoring code for the rodata to decrease as the text size |
| 716 | increases, and vice versa. |
| 717 | |
| 718 | |
Simon Glass | 6200534 | 2014-12-01 17:34:02 -0700 | [diff] [blame] | 719 | The .buildman file |
| 720 | ================== |
Simon Glass | 4281ad8 | 2013-09-23 17:35:17 -0600 | [diff] [blame] | 721 | |
Simon Glass | 6200534 | 2014-12-01 17:34:02 -0700 | [diff] [blame] | 722 | The .buildman file provides information about the available toolchains and |
| 723 | also allows build flags to be passed to 'make'. It consists of several |
| 724 | sections, with the section name in square brackets. Within each section are |
| 725 | a set of (tag, value) pairs. |
Simon Glass | 4281ad8 | 2013-09-23 17:35:17 -0600 | [diff] [blame] | 726 | |
Simon Glass | 6200534 | 2014-12-01 17:34:02 -0700 | [diff] [blame] | 727 | '[toolchain]' section |
Simon Glass | 4281ad8 | 2013-09-23 17:35:17 -0600 | [diff] [blame] | 728 | |
Simon Glass | 6200534 | 2014-12-01 17:34:02 -0700 | [diff] [blame] | 729 | This lists the available toolchains. The tag here doesn't matter, but |
| 730 | make sure it is unique. The value is the path to the toolchain. Buildman |
| 731 | will look in that path for a file ending in 'gcc'. It will then execute |
| 732 | it to check that it is a C compiler, passing only the --version flag to |
| 733 | it. If the return code is 0, buildman assumes that it is a valid C |
| 734 | compiler. It uses the first part of the name as the architecture and |
| 735 | strips off the last part when setting the CROSS_COMPILE environment |
| 736 | variable (parts are delimited with a hyphen). |
Simon Glass | 4281ad8 | 2013-09-23 17:35:17 -0600 | [diff] [blame] | 737 | |
Simon Glass | 6200534 | 2014-12-01 17:34:02 -0700 | [diff] [blame] | 738 | For example powerpc-linux-gcc will be noted as a toolchain for 'powerpc' |
| 739 | and CROSS_COMPILE will be set to powerpc-linux- when using it. |
| 740 | |
| 741 | '[toolchain-alias]' section |
| 742 | |
| 743 | This converts toolchain architecture names to U-Boot names. For example, |
| 744 | if an x86 toolchains is called i386-linux-gcc it will not normally be |
Simon Glass | 9b83bfd | 2014-12-01 17:34:05 -0700 | [diff] [blame] | 745 | used for architecture 'x86'. Adding 'x86: i386 x86_64' to this section |
| 746 | will tell buildman that the i386 and x86_64 toolchains can be used for |
| 747 | the x86 architecture. |
Simon Glass | 6200534 | 2014-12-01 17:34:02 -0700 | [diff] [blame] | 748 | |
| 749 | '[make-flags]' section |
| 750 | |
| 751 | U-Boot's build system supports a few flags (such as BUILD_TAG) which |
| 752 | affect the build product. These flags can be specified in the buildman |
| 753 | settings file. They can also be useful when building U-Boot against other |
| 754 | open source software. |
| 755 | |
| 756 | [make-flags] |
| 757 | at91-boards=ENABLE_AT91_TEST=1 |
| 758 | snapper9260=${at91-boards} BUILD_TAG=442 |
| 759 | snapper9g45=${at91-boards} BUILD_TAG=443 |
| 760 | |
| 761 | This will use 'make ENABLE_AT91_TEST=1 BUILD_TAG=442' for snapper9260 |
| 762 | and 'make ENABLE_AT91_TEST=1 BUILD_TAG=443' for snapper9g45. A special |
| 763 | variable ${target} is available to access the target name (snapper9260 |
| 764 | and snapper9g20 in this case). Variables are resolved recursively. Note |
| 765 | that variables can only contain the characters A-Z, a-z, 0-9, hyphen (-) |
| 766 | and underscore (_). |
| 767 | |
| 768 | It is expected that any variables added are dealt with in U-Boot's |
| 769 | config.mk file and documented in the README. |
| 770 | |
| 771 | Note that you can pass ad-hoc options to the build using environment |
| 772 | variables, for example: |
| 773 | |
| 774 | SOME_OPTION=1234 ./tools/buildman/buildman my_board |
Simon Glass | 4281ad8 | 2013-09-23 17:35:17 -0600 | [diff] [blame] | 775 | |
| 776 | |
Simon Glass | e5a0e5d | 2014-08-09 15:33:03 -0600 | [diff] [blame] | 777 | Quick Sanity Check |
| 778 | ================== |
| 779 | |
| 780 | If you have made changes and want to do a quick sanity check of the |
Simon Glass | 1d8104f | 2014-10-16 01:05:56 -0600 | [diff] [blame] | 781 | currently checked-out source, run buildman without the -b flag. This will |
| 782 | build the selected boards and display build status as it runs (i.e. -v is |
| 783 | enabled automatically). Use -e to see errors/warnings as well. |
Simon Glass | e5a0e5d | 2014-08-09 15:33:03 -0600 | [diff] [blame] | 784 | |
| 785 | |
Simon Glass | 5abab20 | 2014-12-01 17:33:57 -0700 | [diff] [blame] | 786 | Building Ranges |
| 787 | =============== |
| 788 | |
| 789 | You can build a range of commits by specifying a range instead of a branch |
| 790 | when using the -b flag. For example: |
| 791 | |
| 792 | upstream/master..us-buildman |
| 793 | |
| 794 | will build commits in us-buildman that are not in upstream/master. |
| 795 | |
| 796 | |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 797 | Other options |
| 798 | ============= |
| 799 | |
| 800 | Buildman has various other command line options. Try --help to see them. |
| 801 | |
Simon Glass | 2c3deb9 | 2014-08-28 09:43:39 -0600 | [diff] [blame] | 802 | When doing builds, Buildman's return code will reflect the overall result: |
| 803 | |
| 804 | 0 (success) No errors or warnings found |
| 805 | 128 Errors found |
| 806 | 129 Warnings found |
| 807 | |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 808 | |
Simon Glass | 6eede34 | 2014-08-09 15:32:58 -0600 | [diff] [blame] | 809 | How to change from MAKEALL |
| 810 | ========================== |
| 811 | |
| 812 | Buildman includes most of the features of MAKEALL and is generally faster |
| 813 | and easier to use. In particular it builds entire branches: if a particular |
| 814 | commit introduces an error in a particular board, buildman can easily show |
| 815 | you this, even if a later commit fixes that error. |
| 816 | |
| 817 | The reasons to deprecate MAKEALL are: |
| 818 | - We don't want to maintain two build systems |
| 819 | - Buildman is typically faster |
| 820 | - Buildman has a lot more features |
| 821 | |
| 822 | But still, many people will be sad to lose MAKEALL. If you are used to |
| 823 | MAKEALL, here are a few pointers. |
| 824 | |
| 825 | First you need to set up your tool chains - see the 'Setting up' section |
| 826 | for details. Once you have your required toolchain(s) detected then you are |
| 827 | ready to go. |
| 828 | |
Simon Glass | e5a0e5d | 2014-08-09 15:33:03 -0600 | [diff] [blame] | 829 | To build the current source tree, run buildman without a -b flag: |
| 830 | |
| 831 | ./tools/buildman/buildman <list of things to build> |
| 832 | |
| 833 | This will build the current source tree for the given boards and display |
| 834 | the results and errors. |
| 835 | |
| 836 | However buildman usually works on entire branches, and for that you must |
| 837 | specify a board flag: |
Simon Glass | 6eede34 | 2014-08-09 15:32:58 -0600 | [diff] [blame] | 838 | |
| 839 | ./tools/buildman/buildman -b <branch_name> <list of things to build> |
| 840 | |
| 841 | followed by (afterwards, or perhaps concurrently in another terminal): |
| 842 | |
| 843 | ./tools/buildman/buildman -b <branch_name> -s <list of things to build> |
| 844 | |
| 845 | to see the results of the build. Rather than showing you all the output, |
| 846 | buildman just shows a summary, with red indicating that a commit introduced |
| 847 | an error and green indicating that a commit fixed an error. Use the -e |
Simon Glass | ed96665 | 2014-08-28 09:43:43 -0600 | [diff] [blame] | 848 | flag to see the full errors and -l to see which boards caused which errors. |
Simon Glass | 6eede34 | 2014-08-09 15:32:58 -0600 | [diff] [blame] | 849 | |
Simon Glass | e5a0e5d | 2014-08-09 15:33:03 -0600 | [diff] [blame] | 850 | If you really want to see build results as they happen, use -v when doing a |
Simon Glass | 1d8104f | 2014-10-16 01:05:56 -0600 | [diff] [blame] | 851 | build (and -e to see the errors/warnings too). |
Simon Glass | e5a0e5d | 2014-08-09 15:33:03 -0600 | [diff] [blame] | 852 | |
Simon Glass | 6eede34 | 2014-08-09 15:32:58 -0600 | [diff] [blame] | 853 | You don't need to stick around on that branch while buildman is running. It |
| 854 | checks out its own copy of the source code, so you can change branches, |
| 855 | add commits, etc. without affecting the build in progress. |
| 856 | |
| 857 | The <list of things to build> can include board names, architectures or the |
| 858 | like. There are no flags to disambiguate since ambiguities are rare. Using |
| 859 | the examples from MAKEALL: |
| 860 | |
| 861 | Examples: |
| 862 | - build all Power Architecture boards: |
| 863 | MAKEALL -a powerpc |
| 864 | MAKEALL --arch powerpc |
| 865 | MAKEALL powerpc |
| 866 | ** buildman -b <branch> powerpc |
| 867 | - build all PowerPC boards manufactured by vendor "esd": |
| 868 | MAKEALL -a powerpc -v esd |
| 869 | ** buildman -b <branch> esd |
| 870 | - build all PowerPC boards manufactured either by "keymile" or "siemens": |
| 871 | MAKEALL -a powerpc -v keymile -v siemens |
| 872 | ** buildman -b <branch> keymile siemens |
| 873 | - build all Freescale boards with MPC83xx CPUs, plus all 4xx boards: |
| 874 | MAKEALL -c mpc83xx -v freescale 4xx |
| 875 | ** buildman -b <branch> mpc83xx freescale 4xx |
| 876 | |
| 877 | Buildman automatically tries to use all the CPUs in your machine. If you |
| 878 | are building a lot of boards it will use one thread for every CPU core |
| 879 | it detects in your machine. This is like MAKEALL's BUILD_NBUILDS option. |
| 880 | You can use the -T flag to change the number of threads. If you are only |
| 881 | building a few boards, buildman will automatically run make with the -j |
| 882 | flag to increase the number of concurrent make tasks. It isn't normally |
| 883 | that helpful to fiddle with this option, but if you use the BUILD_NCPUS |
| 884 | option in MAKEALL then -j is the equivalent in buildman. |
| 885 | |
| 886 | Buildman puts its output in ../<branch_name> by default but you can change |
| 887 | this with the -o option. Buildman normally does out-of-tree builds: use -i |
| 888 | to disable that if you really want to. But be careful that once you have |
| 889 | used -i you pollute buildman's copies of the source tree, and you will need |
| 890 | to remove the build directory (normally ../<branch_name>) to run buildman |
| 891 | in normal mode (without -i). |
| 892 | |
| 893 | Buildman doesn't keep the output result normally, but use the -k option to |
| 894 | do this. |
| 895 | |
| 896 | Please read 'Theory of Operation' a few times as it will make a lot of |
| 897 | things clearer. |
| 898 | |
| 899 | Some options you might like are: |
| 900 | |
| 901 | -B shows which functions are growing/shrinking in which commit - great |
| 902 | for finding code bloat. |
| 903 | -S shows image sizes for each commit (just an overall summary) |
| 904 | -u shows boards that you haven't built yet |
| 905 | --step 0 will build just the upstream commit and the last commit of your |
| 906 | branch. This is often a quick sanity check that your branch doesn't |
| 907 | break anything. But note this does not check bisectability! |
| 908 | |
| 909 | |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 910 | TODO |
| 911 | ==== |
| 912 | |
| 913 | This has mostly be written in my spare time as a response to my difficulties |
| 914 | in testing large series of patches. Apart from tidying up there is quite a |
Simon Glass | 1d8104f | 2014-10-16 01:05:56 -0600 | [diff] [blame] | 915 | bit of scope for improvement. Things like better error diffs and easier |
Dirk Behme | 3e1ded1 | 2014-12-23 07:41:26 +0100 | [diff] [blame] | 916 | access to log files. Also it would be nice if buildman could 'hunt' for |
Simon Glass | 1d8104f | 2014-10-16 01:05:56 -0600 | [diff] [blame] | 917 | problems, perhaps by building a few boards for each arch, or checking |
| 918 | commits for changed files and building only boards which use those files. |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 919 | |
| 920 | |
| 921 | Credits |
| 922 | ======= |
| 923 | |
| 924 | Thanks to Grant Grundler <grundler@chromium.org> for his ideas for improving |
| 925 | the build speed by building all commits for a board instead of the other |
| 926 | way around. |
| 927 | |
| 928 | |
Simon Glass | fc3fe1c | 2013-04-03 11:07:16 +0000 | [diff] [blame] | 929 | Simon Glass |
| 930 | sjg@chromium.org |
| 931 | Halloween 2012 |
| 932 | Updated 12-12-12 |
| 933 | Updated 23-02-13 |