blob: bfb2f180c28c05717f3199f5a0ec2c0c6c909e61 [file] [log] [blame]
Simon Glassc05694f2013-04-03 11:07:16 +00001# Copyright (c) 2013 The Chromium OS Authors.
2#
Wolfgang Denkd79de1d2013-07-08 09:37:19 +02003# SPDX-License-Identifier: GPL-2.0+
Simon Glassc05694f2013-04-03 11:07:16 +00004#
5
Simon Glass3ec03d42014-08-09 15:32:58 -06006(Please read 'How to change from MAKEALL' if you are used to that tool)
7
Simon Glassc05694f2013-04-03 11:07:16 +00008What is this?
9=============
10
11This tool handles building U-Boot to check that you have not broken it
12with your patch series. It can build each individual commit and report
13which boards fail on which commits, and which errors come up. It aims
14to make full use of multi-processor machines.
15
16A key feature of buildman is its output summary, which allows warnings,
17errors or image size increases in a particular commit or board to be
18quickly identified and the offending commit pinpointed. This can be a big
19help for anyone working with >10 patches at a time.
20
21
22Caveats
23=======
24
25Buildman is still in its infancy. It is already a very useful tool, but
26expect to find problems and send patches.
27
28Buildman can be stopped and restarted, in which case it will continue
29where it left off. This should happen cleanly and without side-effects.
30If not, it is a bug, for which a patch would be welcome.
31
32Buildman gets so tied up in its work that it can ignore the outside world.
33You may need to press Ctrl-C several times to quit it. Also it will print
34out various exceptions when stopped.
35
36
37Theory of Operation
38===================
39
40(please read this section in full twice or you will be perpetually confused)
41
42Buildman is a builder. It is not make, although it runs make. It does not
43produce any useful output on the terminal while building, except for
Simon Glass78e418e2014-08-09 15:33:03 -060044progress information (except with -v, see below). All the output (errors,
45warnings and binaries if you are ask for them) is stored in output
46directories, which you can look at while the build is progressing, or when
47it is finished.
Simon Glassc05694f2013-04-03 11:07:16 +000048
49Buildman produces a concise summary of which boards succeeded and failed.
50It shows which commit introduced which board failure using a simple
51red/green colour coding. Full error information can be requested, in which
52case it is de-duped and displayed against the commit that introduced the
53error. An example workflow is below.
54
55Buildman stores image size information and can report changes in image size
56from commit to commit. An example of this is below.
57
58Buildman starts multiple threads, and each thread builds for one board at
59a time. A thread starts at the first commit, configures the source for your
60board and builds it. Then it checks out the next commit and does an
61incremental build. Eventually the thread reaches the last commit and stops.
62If errors or warnings are found along the way, the thread will reconfigure
63after every commit, and your build will be very slow. This is because a
64file that produces just a warning would not normally be rebuilt in an
65incremental build.
66
67Buildman works in an entirely separate place from your U-Boot repository.
68It creates a separate working directory for each thread, and puts the
69output files in the working directory, organised by commit name and board
70name, in a two-level hierarchy.
71
72Buildman is invoked in your U-Boot directory, the one with the .git
73directory. It clones this repository into a copy for each thread, and the
74threads do not affect the state of your git repository. Any checkouts done
75by the thread affect only the working directory for that thread.
76
Simon Glassfa1cfee2014-08-09 15:32:57 -060077Buildman automatically selects the correct tool chain for each board. You
78must supply suitable tool chains, but buildman takes care of selecting the
Simon Glassc05694f2013-04-03 11:07:16 +000079right one.
80
Simon Glass78e418e2014-08-09 15:33:03 -060081Buildman generally builds a branch (with the -b flag), and in this case
82builds the upstream commit as well, for comparison. It cannot build
83individual commits at present, unless (maybe) you point it at an empty
84branch. Put all your commits in a branch, set the branch's upstream to a
85valid value, and all will be well. Otherwise buildman will perform random
86actions. Use -n to check what the random actions might be.
87
Simon Glassd8158f92014-10-16 01:05:56 -060088If you just want to build the current source tree, leave off the -b flag
89and add -e. This will display results and errors as they happen. You can
90still look at them later using -se. Note that buildman will assume that the
91source has changed, and will build all specified boards in this case.
Simon Glassc05694f2013-04-03 11:07:16 +000092
93Buildman is optimised for building many commits at once, for many boards.
94On multi-core machines, Buildman is fast because it uses most of the
95available CPU power. When it gets to the end, or if you are building just
96a few commits or boards, it will be pretty slow. As a tip, if you don't
97plan to use your machine for anything else, you can use -T to increase the
98number of threads beyond the default.
99
Stephen Warren08447632013-10-10 10:00:20 -0600100Buildman lets you build all boards, or a subset. Specify the subset by passing
101command-line arguments that list the desired board name, architecture name,
102SOC name, or anything else in the boards.cfg file. Multiple arguments are
103allowed. Each argument will be interpreted as a regular expression, so
104behaviour 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$' All boards with either Tegra20 or Tegra30 SoC
109* 'powerpc' All PowerPC boards
Simon Glassc05694f2013-04-03 11:07:16 +0000110
Simon Glassaa40f9a2014-08-09 15:33:08 -0600111While the default is to OR the terms together, you can also make use of
112the '&' operator to limit the selection:
113
114* 'freescale & arm sandbox' All Freescale boards with ARM architecture,
115 plus sandbox
116
Simon Glass924c73a2014-08-28 09:43:41 -0600117You can also use -x to specifically exclude some boards. For example:
118
119 buildmand arm -x nvidia,freescale,.*ball$
120
121means to build all arm boards except nvidia, freescale and anything ending
122with 'ball'.
123
Simon Glassaa40f9a2014-08-09 15:33:08 -0600124It is convenient to use the -n option to see whaat will be built based on
125the subset given.
126
Simon Glassc05694f2013-04-03 11:07:16 +0000127Buildman does not store intermediate object files. It optionally copies
128the binary output into a directory when a build is successful. Size
129information is always recorded. It needs a fair bit of disk space to work,
130typically 250MB per thread.
131
132
133Setting up
134==========
135
1361. Get the U-Boot source. You probably already have it, but if not these
137steps 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
1442. Create ~/.buildman to tell buildman where to find tool chains. As an
145example:
146
147# Buildman settings file
148
149[toolchain]
150root: /
151rest: /toolchains/*
152eldk: /opt/eldk-4.2
Simon Glass9a016392014-08-09 15:33:07 -0600153arm: /opt/linaro/gcc-linaro-arm-linux-gnueabihf-4.8-2013.08_linux
154aarch64: /opt/linaro/gcc-linaro-aarch64-none-elf-4.8-2013.10_linux
Simon Glassc05694f2013-04-03 11:07:16 +0000155
156[toolchain-alias]
157x86: i386
158blackfin: bfin
159sh: sh4
160nds32: nds32le
161openrisc: or32
162
163
164This selects the available toolchain paths. Add the base directory for
165each of your toolchains here. Buildman will search inside these directories
166and also in any '/usr' and '/usr/bin' subdirectories.
167
168Make sure the tags (here root: rest: and eldk:) are unique.
169
170The toolchain-alias section indicates that the i386 toolchain should be used
171to build x86 commits.
172
173
1742. Check the available toolchains
175
176Run this check to make sure that you have a toolchain for every architecture.
177
178$ ./tools/buildman/buildman --list-tool-chains
179Scanning for tool chains
180 - scanning path '/'
181 - looking in '/.'
182 - looking in '/bin'
183 - looking in '/usr/bin'
184 - found '/usr/bin/gcc'
185Tool chain test: OK
186 - found '/usr/bin/c89-gcc'
187Tool chain test: OK
188 - found '/usr/bin/c99-gcc'
189Tool chain test: OK
190 - found '/usr/bin/x86_64-linux-gnu-gcc'
191Tool chain test: OK
192 - scanning path '/toolchains/powerpc-linux'
193 - looking in '/toolchains/powerpc-linux/.'
194 - looking in '/toolchains/powerpc-linux/bin'
195 - found '/toolchains/powerpc-linux/bin/powerpc-linux-gcc'
196Tool chain test: OK
197 - looking in '/toolchains/powerpc-linux/usr/bin'
198 - scanning path '/toolchains/nds32le-linux-glibc-v1f'
199 - looking in '/toolchains/nds32le-linux-glibc-v1f/.'
200 - looking in '/toolchains/nds32le-linux-glibc-v1f/bin'
201 - found '/toolchains/nds32le-linux-glibc-v1f/bin/nds32le-linux-gcc'
202Tool chain test: OK
203 - looking in '/toolchains/nds32le-linux-glibc-v1f/usr/bin'
204 - scanning path '/toolchains/nios2'
205 - looking in '/toolchains/nios2/.'
206 - looking in '/toolchains/nios2/bin'
207 - found '/toolchains/nios2/bin/nios2-linux-gcc'
208Tool chain test: OK
209 - found '/toolchains/nios2/bin/nios2-linux-uclibc-gcc'
210Tool chain test: OK
211 - looking in '/toolchains/nios2/usr/bin'
212 - found '/toolchains/nios2/usr/bin/nios2-linux-gcc'
213Tool chain test: OK
214 - found '/toolchains/nios2/usr/bin/nios2-linux-uclibc-gcc'
215Tool chain test: OK
216 - scanning path '/toolchains/microblaze-unknown-linux-gnu'
217 - looking in '/toolchains/microblaze-unknown-linux-gnu/.'
218 - looking in '/toolchains/microblaze-unknown-linux-gnu/bin'
219 - found '/toolchains/microblaze-unknown-linux-gnu/bin/microblaze-unknown-linux-gnu-gcc'
220Tool chain test: OK
221 - found '/toolchains/microblaze-unknown-linux-gnu/bin/mb-linux-gcc'
222Tool chain test: OK
223 - looking in '/toolchains/microblaze-unknown-linux-gnu/usr/bin'
224 - scanning path '/toolchains/mips-linux'
225 - looking in '/toolchains/mips-linux/.'
226 - looking in '/toolchains/mips-linux/bin'
227 - found '/toolchains/mips-linux/bin/mips-linux-gcc'
228Tool chain test: OK
229 - looking in '/toolchains/mips-linux/usr/bin'
230 - scanning path '/toolchains/old'
231 - looking in '/toolchains/old/.'
232 - looking in '/toolchains/old/bin'
233 - looking in '/toolchains/old/usr/bin'
234 - scanning path '/toolchains/i386-linux'
235 - looking in '/toolchains/i386-linux/.'
236 - looking in '/toolchains/i386-linux/bin'
237 - found '/toolchains/i386-linux/bin/i386-linux-gcc'
238Tool chain test: OK
239 - looking in '/toolchains/i386-linux/usr/bin'
240 - scanning path '/toolchains/bfin-uclinux'
241 - looking in '/toolchains/bfin-uclinux/.'
242 - looking in '/toolchains/bfin-uclinux/bin'
243 - found '/toolchains/bfin-uclinux/bin/bfin-uclinux-gcc'
244Tool chain test: OK
245 - looking in '/toolchains/bfin-uclinux/usr/bin'
246 - scanning path '/toolchains/sparc-elf'
247 - looking in '/toolchains/sparc-elf/.'
248 - looking in '/toolchains/sparc-elf/bin'
249 - found '/toolchains/sparc-elf/bin/sparc-elf-gcc'
250Tool chain test: OK
251 - looking in '/toolchains/sparc-elf/usr/bin'
252 - scanning path '/toolchains/arm-2010q1'
253 - looking in '/toolchains/arm-2010q1/.'
254 - looking in '/toolchains/arm-2010q1/bin'
255 - found '/toolchains/arm-2010q1/bin/arm-none-linux-gnueabi-gcc'
256Tool chain test: OK
257 - looking in '/toolchains/arm-2010q1/usr/bin'
258 - scanning path '/toolchains/from'
259 - looking in '/toolchains/from/.'
260 - looking in '/toolchains/from/bin'
261 - looking in '/toolchains/from/usr/bin'
262 - scanning path '/toolchains/sh4-gentoo-linux-gnu'
263 - looking in '/toolchains/sh4-gentoo-linux-gnu/.'
264 - looking in '/toolchains/sh4-gentoo-linux-gnu/bin'
265 - found '/toolchains/sh4-gentoo-linux-gnu/bin/sh4-gentoo-linux-gnu-gcc'
266Tool chain test: OK
267 - looking in '/toolchains/sh4-gentoo-linux-gnu/usr/bin'
268 - scanning path '/toolchains/avr32-linux'
269 - looking in '/toolchains/avr32-linux/.'
270 - looking in '/toolchains/avr32-linux/bin'
271 - found '/toolchains/avr32-linux/bin/avr32-gcc'
272Tool chain test: OK
273 - looking in '/toolchains/avr32-linux/usr/bin'
274 - scanning path '/toolchains/m68k-linux'
275 - looking in '/toolchains/m68k-linux/.'
276 - looking in '/toolchains/m68k-linux/bin'
277 - found '/toolchains/m68k-linux/bin/m68k-linux-gcc'
278Tool chain test: OK
279 - looking in '/toolchains/m68k-linux/usr/bin'
280List of available toolchains (17):
281arm : /toolchains/arm-2010q1/bin/arm-none-linux-gnueabi-gcc
282avr32 : /toolchains/avr32-linux/bin/avr32-gcc
283bfin : /toolchains/bfin-uclinux/bin/bfin-uclinux-gcc
284c89 : /usr/bin/c89-gcc
285c99 : /usr/bin/c99-gcc
286i386 : /toolchains/i386-linux/bin/i386-linux-gcc
287m68k : /toolchains/m68k-linux/bin/m68k-linux-gcc
288mb : /toolchains/microblaze-unknown-linux-gnu/bin/mb-linux-gcc
289microblaze: /toolchains/microblaze-unknown-linux-gnu/bin/microblaze-unknown-linux-gnu-gcc
290mips : /toolchains/mips-linux/bin/mips-linux-gcc
291nds32le : /toolchains/nds32le-linux-glibc-v1f/bin/nds32le-linux-gcc
292nios2 : /toolchains/nios2/bin/nios2-linux-gcc
293powerpc : /toolchains/powerpc-linux/bin/powerpc-linux-gcc
294sandbox : /usr/bin/gcc
295sh4 : /toolchains/sh4-gentoo-linux-gnu/bin/sh4-gentoo-linux-gnu-gcc
296sparc : /toolchains/sparc-elf/bin/sparc-elf-gcc
297x86_64 : /usr/bin/x86_64-linux-gnu-gcc
298
299
300You can see that everything is covered, even some strange ones that won't
301be used (c88 and c99). This is a feature.
302
303
304How to run it
305=============
306
307First do a dry run using the -n flag: (replace <branch> with a real, local
308branch with a valid upstream)
309
310$ ./tools/buildman/buildman -b <branch> -n
311
312If it can't detect the upstream branch, try checking out the branch, and
313doing something like 'git branch --set-upstream <branch> upstream/master'
314or something similar.
315
Simon Glassfa1cfee2014-08-09 15:32:57 -0600316As an example:
Simon Glassc05694f2013-04-03 11:07:16 +0000317
318Dry run, so not doing much. But I would do this:
319
320Building 18 commits for 1059 boards (4 threads, 1 job per thread)
321Build directory: ../lcd9b
322 5bb3505 Merge branch 'master' of git://git.denx.de/u-boot-arm
323 c18f1b4 tegra: Use const for pinmux_config_pingroup/table()
324 2f043ae tegra: Add display support to funcmux
325 e349900 tegra: fdt: Add pwm binding and node
326 424a5f0 tegra: fdt: Add LCD definitions for Tegra
327 0636ccf tegra: Add support for PWM
328 a994fe7 tegra: Add SOC support for display/lcd
329 fcd7350 tegra: Add LCD driver
330 4d46e9d tegra: Add LCD support to Nvidia boards
331 991bd48 arm: Add control over cachability of memory regions
332 54e8019 lcd: Add CONFIG_LCD_ALIGNMENT to select frame buffer alignment
333 d92aff7 lcd: Add support for flushing LCD fb from dcache after update
334 dbd0677 tegra: Align LCD frame buffer to section boundary
335 0cff9b8 tegra: Support control of cache settings for LCD
336 9c56900 tegra: fdt: Add LCD definitions for Seaboard
337 5cc29db lcd: Add CONFIG_CONSOLE_SCROLL_LINES option to speed console
338 cac5a23 tegra: Enable display/lcd support on Seaboard
339 49ff541 wip
340
341Total boards to build for each commit: 1059
342
343This shows that it will build all 1059 boards, using 4 threads (because
344we have a 4-core CPU). Each thread will run with -j1, meaning that each
345make job will use a single CPU. The list of commits to be built helps you
346confirm that things look about right. Notice that buildman has chosen a
347'base' directory for you, immediately above your source tree.
348
349Buildman works entirely inside the base directory, here ../lcd9b,
350creating a working directory for each thread, and creating output
351directories for each commit and board.
352
353
354Suggested Workflow
355==================
356
357To run the build for real, take off the -n:
358
359$ ./tools/buildman/buildman -b <branch>
360
361Buildman will set up some working directories, and get started. After a
362minute or so it will settle down to a steady pace, with a display like this:
363
364Building 18 commits for 1059 boards (4 threads, 1 job per thread)
365 528 36 124 /19062 1:13:30 : SIMPC8313_SP
366
367This means that it is building 19062 board/commit combinations. So far it
Simon Glassfa1cfee2014-08-09 15:32:57 -0600368has managed to successfully build 528. Another 36 have built with warnings,
Simon Glassc05694f2013-04-03 11:07:16 +0000369and 124 more didn't build at all. Buildman expects to complete the process
370in an hour and 15 minutes. Use this time to buy a faster computer.
371
372
373To find out how the build went, ask for a summary with -s. You can do this
374either before the build completes (presumably in another terminal) or or
375afterwards. Let's work through an example of how this is used:
376
377$ ./tools/buildman/buildman -b lcd9b -s
378...
37901: Merge branch 'master' of git://git.denx.de/u-boot-arm
380 powerpc: + galaxy5200_LOWBOOT
38102: tegra: Use const for pinmux_config_pingroup/table()
38203: tegra: Add display support to funcmux
38304: tegra: fdt: Add pwm binding and node
38405: tegra: fdt: Add LCD definitions for Tegra
38506: tegra: Add support for PWM
38607: tegra: Add SOC support for display/lcd
38708: tegra: Add LCD driver
38809: tegra: Add LCD support to Nvidia boards
38910: arm: Add control over cachability of memory regions
39011: lcd: Add CONFIG_LCD_ALIGNMENT to select frame buffer alignment
39112: lcd: Add support for flushing LCD fb from dcache after update
392 arm: + lubbock
39313: tegra: Align LCD frame buffer to section boundary
39414: tegra: Support control of cache settings for LCD
39515: tegra: fdt: Add LCD definitions for Seaboard
39616: lcd: Add CONFIG_CONSOLE_SCROLL_LINES option to speed console
39717: tegra: Enable display/lcd support on Seaboard
39818: wip
399
400This shows which commits have succeeded and which have failed. In this case
401the build is still in progress so many boards are not built yet (use -u to
402see which ones). But still we can see a few failures. The galaxy5200_LOWBOOT
403never builds correctly. This could be a problem with our toolchain, or it
404could be a bug in the upstream. The good news is that we probably don't need
405to blame our commits. The bad news is it isn't tested on that board.
406
407Commit 12 broke lubbock. That's what the '+ lubbock' means. The failure
408is never fixed by a later commit, or you would see lubbock again, in green,
409without the +.
410
411To see the actual error:
412
413$ ./tools/buildman/buildman -b <branch> -se lubbock
414...
41512: lcd: Add support for flushing LCD fb from dcache after update
416 arm: + lubbock
417+common/libcommon.o: In function `lcd_sync':
418+/u-boot/lcd9b/.bm-work/00/common/lcd.c:120: undefined reference to `flush_dcache_range'
419+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
420+make: *** [/u-boot/lcd9b/.bm-work/00/build/u-boot] Error 139
42113: tegra: Align LCD frame buffer to section boundary
42214: tegra: Support control of cache settings for LCD
42315: tegra: fdt: Add LCD definitions for Seaboard
42416: lcd: Add CONFIG_CONSOLE_SCROLL_LINES option to speed console
425-/u-boot/lcd9b/.bm-work/00/common/lcd.c:120: undefined reference to `flush_dcache_range'
426+/u-boot/lcd9b/.bm-work/00/common/lcd.c:125: undefined reference to `flush_dcache_range'
42717: tegra: Enable display/lcd support on Seaboard
42818: wip
429
430So the problem is in lcd.c, due to missing cache operations. This information
431should be enough to work out what that commit is doing to break these
432boards. (In this case pxa did not have cache operations defined).
433
434If you see error lines marked with - that means that the errors were fixed
435by that commit. Sometimes commits can be in the wrong order, so that a
436breakage is introduced for a few commits and fixed by later commits. This
437shows up clearly with buildman. You can then reorder the commits and try
438again.
439
440At commit 16, the error moves - you can see that the old error at line 120
441is fixed, but there is a new one at line 126. This is probably only because
Simon Glassfa1cfee2014-08-09 15:32:57 -0600442we added some code and moved the broken line father down the file.
Simon Glassc05694f2013-04-03 11:07:16 +0000443
444If many boards have the same error, then -e will display the error only
Simon Glass3394c9f2014-08-28 09:43:43 -0600445once. This makes the output as concise as possible. To see which boards have
446each error, use -l.
Simon Glassc05694f2013-04-03 11:07:16 +0000447
Simon Glass03749d42014-08-28 09:43:44 -0600448Buildman tries to distinguish warnings from errors, and shows warning lines
449separately with a 'w' prefix.
450
Simon Glassc05694f2013-04-03 11:07:16 +0000451The full build output in this case is available in:
452
453../lcd9b/12_of_18_gd92aff7_lcd--Add-support-for/lubbock/
454
455 done: Indicates the build was done, and holds the return code from make.
456 This is 0 for a good build, typically 2 for a failure.
457
458 err: Output from stderr, if any. Errors and warnings appear here.
459
460 log: Output from stdout. Normally there isn't any since buildman runs
461 in silent mode for now.
462
463 toolchain: Shows information about the toolchain used for the build.
464
465 sizes: Shows image size information.
466
467It is possible to get the build output there also. Use the -k option for
468this. In that case you will also see some output files, like:
469
470 System.map toolchain u-boot u-boot.bin u-boot.map autoconf.mk
471 (also SPL versions u-boot-spl and u-boot-spl.bin if available)
472
473
474Checking Image Sizes
475====================
476
477A key requirement for U-Boot is that you keep code/data size to a minimum.
478Where a new feature increases this noticeably it should normally be put
479behind a CONFIG flag so that boards can leave it off and keep the image
480size more or less the same with each new release.
481
482To check the impact of your commits on image size, use -S. For example:
483
484$ ./tools/buildman/buildman -b us-x86 -sS
485Summary of 10 commits for 1066 boards (4 threads, 1 job per thread)
48601: MAKEALL: add support for per architecture toolchains
48702: x86: Add function to get top of usable ram
488 x86: (for 1/3 boards) text -272.0 rodata +41.0
48903: x86: Add basic cache operations
49004: x86: Permit bootstage and timer data to be used prior to relocation
491 x86: (for 1/3 boards) data +16.0
49205: x86: Add an __end symbol to signal the end of the U-Boot binary
493 x86: (for 1/3 boards) text +76.0
49406: x86: Rearrange the output input to remove BSS
495 x86: (for 1/3 boards) bss -2140.0
49607: x86: Support relocation of FDT on start-up
497 x86: + coreboot-x86
49808: x86: Add error checking to x86 relocation code
49909: x86: Adjust link device tree include file
50010: x86: Enable CONFIG_OF_CONTROL on coreboot
501
502
503You can see that image size only changed on x86, which is good because this
504series is not supposed to change any other board. From commit 7 onwards the
505build fails so we don't get code size numbers. The numbers are fractional
506because they are an average of all boards for that architecture. The
507intention is to allow you to quickly find image size problems introduced by
508your commits.
509
510Note that the 'text' region and 'rodata' are split out. You should add the
511two together to get the total read-only size (reported as the first column
512in the output from binutil's 'size' utility).
513
514A useful option is --step which lets you skip some commits. For example
515--step 2 will show the image sizes for only every 2nd commit (so it will
516compare the image sizes of the 1st, 3rd, 5th... commits). You can also use
517--step 0 which will compare only the first and last commits. This is useful
518for an overview of how your entire series affects code size.
519
520You can also use -d to see a detailed size breakdown for each board. This
521list is sorted in order from largest growth to largest reduction.
522
523It is possible to go a little further with the -B option (--bloat). This
Simon Glassfa1cfee2014-08-09 15:32:57 -0600524shows where U-Boot has bloated, breaking the size change down to the function
Simon Glassc05694f2013-04-03 11:07:16 +0000525level. Example output is below:
526
527$ ./tools/buildman/buildman -b us-mem4 -sSdB
528...
52919: Roll crc32 into hash infrastructure
530 arm: (for 10/10 boards) all -143.4 bss +1.2 data -4.8 rodata -48.2 text -91.6
531 paz00 : all +23 bss -4 rodata -29 text +56
532 u-boot: add: 1/0, grow: 3/-2 bytes: 168/-104 (64)
533 function old new delta
534 hash_command 80 160 +80
535 crc32_wd_buf - 56 +56
536 ext4fs_read_file 540 568 +28
537 insert_var_value_sub 688 692 +4
538 run_list_real 1996 1992 -4
539 do_mem_crc 168 68 -100
540 trimslice : all -9 bss +16 rodata -29 text +4
541 u-boot: add: 1/0, grow: 1/-3 bytes: 136/-124 (12)
542 function old new delta
543 hash_command 80 160 +80
544 crc32_wd_buf - 56 +56
545 ext4fs_iterate_dir 672 668 -4
546 ext4fs_read_file 568 548 -20
547 do_mem_crc 168 68 -100
548 whistler : all -9 bss +16 rodata -29 text +4
549 u-boot: add: 1/0, grow: 1/-3 bytes: 136/-124 (12)
550 function old new delta
551 hash_command 80 160 +80
552 crc32_wd_buf - 56 +56
553 ext4fs_iterate_dir 672 668 -4
554 ext4fs_read_file 568 548 -20
555 do_mem_crc 168 68 -100
556 seaboard : all -9 bss -28 rodata -29 text +48
557 u-boot: add: 1/0, grow: 3/-2 bytes: 160/-104 (56)
558 function old new delta
559 hash_command 80 160 +80
560 crc32_wd_buf - 56 +56
561 ext4fs_read_file 548 568 +20
562 run_list_real 1996 2000 +4
563 do_nandboot 760 756 -4
564 do_mem_crc 168 68 -100
565 colibri_t20_iris: all -9 rodata -29 text +20
566 u-boot: add: 1/0, grow: 2/-3 bytes: 140/-112 (28)
567 function old new delta
568 hash_command 80 160 +80
569 crc32_wd_buf - 56 +56
570 read_abs_bbt 204 208 +4
571 do_nandboot 760 756 -4
572 ext4fs_read_file 576 568 -8
573 do_mem_crc 168 68 -100
574 ventana : all -37 bss -12 rodata -29 text +4
575 u-boot: add: 1/0, grow: 1/-3 bytes: 136/-124 (12)
576 function old new delta
577 hash_command 80 160 +80
578 crc32_wd_buf - 56 +56
579 ext4fs_iterate_dir 672 668 -4
580 ext4fs_read_file 568 548 -20
581 do_mem_crc 168 68 -100
582 harmony : all -37 bss -16 rodata -29 text +8
583 u-boot: add: 1/0, grow: 2/-3 bytes: 140/-124 (16)
584 function old new delta
585 hash_command 80 160 +80
586 crc32_wd_buf - 56 +56
587 nand_write_oob_syndrome 428 432 +4
588 ext4fs_iterate_dir 672 668 -4
589 ext4fs_read_file 568 548 -20
590 do_mem_crc 168 68 -100
591 medcom-wide : all -417 bss +28 data -16 rodata -93 text -336
592 u-boot: add: 1/-1, grow: 1/-2 bytes: 88/-376 (-288)
593 function old new delta
594 crc32_wd_buf - 56 +56
595 do_fat_read_at 2872 2904 +32
596 hash_algo 16 - -16
597 do_mem_crc 168 68 -100
598 hash_command 420 160 -260
599 tec : all -449 bss -4 data -16 rodata -93 text -336
600 u-boot: add: 1/-1, grow: 1/-2 bytes: 88/-376 (-288)
601 function old new delta
602 crc32_wd_buf - 56 +56
603 do_fat_read_at 2872 2904 +32
604 hash_algo 16 - -16
605 do_mem_crc 168 68 -100
606 hash_command 420 160 -260
607 plutux : all -481 bss +16 data -16 rodata -93 text -388
608 u-boot: add: 1/-1, grow: 1/-3 bytes: 68/-408 (-340)
609 function old new delta
610 crc32_wd_buf - 56 +56
611 do_load_serial_bin 1688 1700 +12
612 hash_algo 16 - -16
613 do_fat_read_at 2904 2872 -32
614 do_mem_crc 168 68 -100
615 hash_command 420 160 -260
616 powerpc: (for 5/5 boards) all +37.4 data -3.2 rodata -41.8 text +82.4
617 MPC8610HPCD : all +55 rodata -29 text +84
618 u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80)
619 function old new delta
620 hash_command - 176 +176
621 do_mem_crc 184 88 -96
622 MPC8641HPCN : all +55 rodata -29 text +84
623 u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80)
624 function old new delta
625 hash_command - 176 +176
626 do_mem_crc 184 88 -96
627 MPC8641HPCN_36BIT: all +55 rodata -29 text +84
628 u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80)
629 function old new delta
630 hash_command - 176 +176
631 do_mem_crc 184 88 -96
632 sbc8641d : all +55 rodata -29 text +84
633 u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80)
634 function old new delta
635 hash_command - 176 +176
636 do_mem_crc 184 88 -96
637 xpedite517x : all -33 data -16 rodata -93 text +76
638 u-boot: add: 1/-1, grow: 0/-1 bytes: 176/-112 (64)
639 function old new delta
640 hash_command - 176 +176
641 hash_algo 16 - -16
642 do_mem_crc 184 88 -96
643...
644
645
646This shows that commit 19 has increased text size for arm (although only one
647board was built) and by 96 bytes for powerpc. This increase was offset in both
648cases by reductions in rodata and data/bss.
649
650Shown below the summary lines is the sizes for each board. Below each board
651is the sizes for each function. This information starts with:
652
653 add - number of functions added / removed
654 grow - number of functions which grew / shrunk
655 bytes - number of bytes of code added to / removed from all functions,
656 plus the total byte change in brackets
657
658The change seems to be that hash_command() has increased by more than the
659do_mem_crc() function has decreased. The function sizes typically add up to
660roughly the text area size, but note that every read-only section except
661rodata is included in 'text', so the function total does not exactly
662correspond.
663
664It is common when refactoring code for the rodata to decrease as the text size
665increases, and vice versa.
666
667
Simon Glasscc246fb2013-09-23 17:35:17 -0600668Providing 'make' flags
669======================
670
671U-Boot's build system supports a few flags (such as BUILD_TAG) which affect
672the build product. These flags can be specified in the buildman settings
673file. They can also be useful when building U-Boot against other open source
674software.
675
676[make-flags]
677at91-boards=ENABLE_AT91_TEST=1
678snapper9260=${at91-boards} BUILD_TAG=442
679snapper9g45=${at91-boards} BUILD_TAG=443
680
681This will use 'make ENABLE_AT91_TEST=1 BUILD_TAG=442' for snapper9260
Andreas Bießmannb5c56e32013-11-05 10:37:09 +0100682and 'make ENABLE_AT91_TEST=1 BUILD_TAG=443' for snapper9g45. A special
Simon Glasscc246fb2013-09-23 17:35:17 -0600683variable ${target} is available to access the target name (snapper9260 and
Simon Glass53e189d2014-08-28 09:43:40 -0600684snapper9g20 in this case). Variables are resolved recursively. Note that
685variables can only contain the characters A-Z, a-z, 0-9, hyphen (-) and
686underscore (_).
Simon Glasscc246fb2013-09-23 17:35:17 -0600687
688It is expected that any variables added are dealt with in U-Boot's
689config.mk file and documented in the README.
690
691
Simon Glass78e418e2014-08-09 15:33:03 -0600692Quick Sanity Check
693==================
694
695If you have made changes and want to do a quick sanity check of the
Simon Glassd8158f92014-10-16 01:05:56 -0600696currently checked-out source, run buildman without the -b flag. This will
697build the selected boards and display build status as it runs (i.e. -v is
698enabled automatically). Use -e to see errors/warnings as well.
Simon Glass78e418e2014-08-09 15:33:03 -0600699
700
Simon Glassc05694f2013-04-03 11:07:16 +0000701Other options
702=============
703
704Buildman has various other command line options. Try --help to see them.
705
Simon Glassc2f91072014-08-28 09:43:39 -0600706When doing builds, Buildman's return code will reflect the overall result:
707
708 0 (success) No errors or warnings found
709 128 Errors found
710 129 Warnings found
711
Simon Glassc05694f2013-04-03 11:07:16 +0000712
Simon Glass3ec03d42014-08-09 15:32:58 -0600713How to change from MAKEALL
714==========================
715
716Buildman includes most of the features of MAKEALL and is generally faster
717and easier to use. In particular it builds entire branches: if a particular
718commit introduces an error in a particular board, buildman can easily show
719you this, even if a later commit fixes that error.
720
721The reasons to deprecate MAKEALL are:
722- We don't want to maintain two build systems
723- Buildman is typically faster
724- Buildman has a lot more features
725
726But still, many people will be sad to lose MAKEALL. If you are used to
727MAKEALL, here are a few pointers.
728
729First you need to set up your tool chains - see the 'Setting up' section
730for details. Once you have your required toolchain(s) detected then you are
731ready to go.
732
Simon Glass78e418e2014-08-09 15:33:03 -0600733To build the current source tree, run buildman without a -b flag:
734
735 ./tools/buildman/buildman <list of things to build>
736
737This will build the current source tree for the given boards and display
738the results and errors.
739
740However buildman usually works on entire branches, and for that you must
741specify a board flag:
Simon Glass3ec03d42014-08-09 15:32:58 -0600742
743 ./tools/buildman/buildman -b <branch_name> <list of things to build>
744
745followed by (afterwards, or perhaps concurrently in another terminal):
746
747 ./tools/buildman/buildman -b <branch_name> -s <list of things to build>
748
749to see the results of the build. Rather than showing you all the output,
750buildman just shows a summary, with red indicating that a commit introduced
751an error and green indicating that a commit fixed an error. Use the -e
Simon Glass3394c9f2014-08-28 09:43:43 -0600752flag to see the full errors and -l to see which boards caused which errors.
Simon Glass3ec03d42014-08-09 15:32:58 -0600753
Simon Glass78e418e2014-08-09 15:33:03 -0600754If you really want to see build results as they happen, use -v when doing a
Simon Glassd8158f92014-10-16 01:05:56 -0600755build (and -e to see the errors/warnings too).
Simon Glass78e418e2014-08-09 15:33:03 -0600756
Simon Glass3ec03d42014-08-09 15:32:58 -0600757You don't need to stick around on that branch while buildman is running. It
758checks out its own copy of the source code, so you can change branches,
759add commits, etc. without affecting the build in progress.
760
761The <list of things to build> can include board names, architectures or the
762like. There are no flags to disambiguate since ambiguities are rare. Using
763the examples from MAKEALL:
764
765Examples:
766 - build all Power Architecture boards:
767 MAKEALL -a powerpc
768 MAKEALL --arch powerpc
769 MAKEALL powerpc
770 ** buildman -b <branch> powerpc
771 - build all PowerPC boards manufactured by vendor "esd":
772 MAKEALL -a powerpc -v esd
773 ** buildman -b <branch> esd
774 - build all PowerPC boards manufactured either by "keymile" or "siemens":
775 MAKEALL -a powerpc -v keymile -v siemens
776 ** buildman -b <branch> keymile siemens
777 - build all Freescale boards with MPC83xx CPUs, plus all 4xx boards:
778 MAKEALL -c mpc83xx -v freescale 4xx
779 ** buildman -b <branch> mpc83xx freescale 4xx
780
781Buildman automatically tries to use all the CPUs in your machine. If you
782are building a lot of boards it will use one thread for every CPU core
783it detects in your machine. This is like MAKEALL's BUILD_NBUILDS option.
784You can use the -T flag to change the number of threads. If you are only
785building a few boards, buildman will automatically run make with the -j
786flag to increase the number of concurrent make tasks. It isn't normally
787that helpful to fiddle with this option, but if you use the BUILD_NCPUS
788option in MAKEALL then -j is the equivalent in buildman.
789
790Buildman puts its output in ../<branch_name> by default but you can change
791this with the -o option. Buildman normally does out-of-tree builds: use -i
792to disable that if you really want to. But be careful that once you have
793used -i you pollute buildman's copies of the source tree, and you will need
794to remove the build directory (normally ../<branch_name>) to run buildman
795in normal mode (without -i).
796
797Buildman doesn't keep the output result normally, but use the -k option to
798do this.
799
800Please read 'Theory of Operation' a few times as it will make a lot of
801things clearer.
802
803Some options you might like are:
804
805 -B shows which functions are growing/shrinking in which commit - great
806 for finding code bloat.
807 -S shows image sizes for each commit (just an overall summary)
808 -u shows boards that you haven't built yet
809 --step 0 will build just the upstream commit and the last commit of your
810 branch. This is often a quick sanity check that your branch doesn't
811 break anything. But note this does not check bisectability!
812
813
Simon Glassc05694f2013-04-03 11:07:16 +0000814TODO
815====
816
817This has mostly be written in my spare time as a response to my difficulties
818in testing large series of patches. Apart from tidying up there is quite a
Simon Glassd8158f92014-10-16 01:05:56 -0600819bit of scope for improvement. Things like better error diffs and easier
820access to log files. Also it would be nice it buildman could 'hunt' for
821problems, perhaps by building a few boards for each arch, or checking
822commits for changed files and building only boards which use those files.
Simon Glassc05694f2013-04-03 11:07:16 +0000823
824
825Credits
826=======
827
828Thanks to Grant Grundler <grundler@chromium.org> for his ideas for improving
829the build speed by building all commits for a board instead of the other
830way around.
831
832
Simon Glassc05694f2013-04-03 11:07:16 +0000833Simon Glass
834sjg@chromium.org
835Halloween 2012
836Updated 12-12-12
837Updated 23-02-13