blob: 6ba24c0e73c9b706905d9cd216a975b6d4ad7205 [file] [log] [blame]
Simon Glassfc3fe1c2013-04-03 11:07:16 +00001# Copyright (c) 2013 The Chromium OS Authors.
2#
Wolfgang Denk1a459662013-07-08 09:37:19 +02003# SPDX-License-Identifier: GPL-2.0+
Simon Glassfc3fe1c2013-04-03 11:07:16 +00004#
5
Simon Glass6eede342014-08-09 15:32:58 -06006(Please read 'How to change from MAKEALL' if you are used to that tool)
7
Simon Glassfc3fe1c2013-04-03 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 Glasse5a0e5d2014-08-09 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 Glassfc3fe1c2013-04-03 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 Glasscec83c32014-08-09 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 Glassfc3fe1c2013-04-03 11:07:16 +000079right one.
80
Simon Glasse5a0e5d2014-08-09 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
88If you just want to build the current source tree, leave off the -b flag.
89This will display results and errors as they happen. You can still look
90at them later using -s. Note that buildman will assume that the source
91has changed, and will build all specified boards in this case.
Simon Glassfc3fe1c2013-04-03 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 Warren8426d8b2013-10-10 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$' All boards with either Tegra20 or Tegra30 SoC
109* 'powerpc' All PowerPC boards
Simon Glassfc3fe1c2013-04-03 11:07:16 +0000110
111Buildman does not store intermediate object files. It optionally copies
112the binary output into a directory when a build is successful. Size
113information is always recorded. It needs a fair bit of disk space to work,
114typically 250MB per thread.
115
116
117Setting up
118==========
119
1201. Get the U-Boot source. You probably already have it, but if not these
121steps should get you started with a repo and some commits for testing.
122
123$ cd /path/to/u-boot
124$ git clone git://git.denx.de/u-boot.git .
125$ git checkout -b my-branch origin/master
126$ # Add some commits to the branch, reading for testing
127
1282. Create ~/.buildman to tell buildman where to find tool chains. As an
129example:
130
131# Buildman settings file
132
133[toolchain]
134root: /
135rest: /toolchains/*
136eldk: /opt/eldk-4.2
137
138[toolchain-alias]
139x86: i386
140blackfin: bfin
141sh: sh4
142nds32: nds32le
143openrisc: or32
144
145
146This selects the available toolchain paths. Add the base directory for
147each of your toolchains here. Buildman will search inside these directories
148and also in any '/usr' and '/usr/bin' subdirectories.
149
150Make sure the tags (here root: rest: and eldk:) are unique.
151
152The toolchain-alias section indicates that the i386 toolchain should be used
153to build x86 commits.
154
155
1562. Check the available toolchains
157
158Run this check to make sure that you have a toolchain for every architecture.
159
160$ ./tools/buildman/buildman --list-tool-chains
161Scanning for tool chains
162 - scanning path '/'
163 - looking in '/.'
164 - looking in '/bin'
165 - looking in '/usr/bin'
166 - found '/usr/bin/gcc'
167Tool chain test: OK
168 - found '/usr/bin/c89-gcc'
169Tool chain test: OK
170 - found '/usr/bin/c99-gcc'
171Tool chain test: OK
172 - found '/usr/bin/x86_64-linux-gnu-gcc'
173Tool chain test: OK
174 - scanning path '/toolchains/powerpc-linux'
175 - looking in '/toolchains/powerpc-linux/.'
176 - looking in '/toolchains/powerpc-linux/bin'
177 - found '/toolchains/powerpc-linux/bin/powerpc-linux-gcc'
178Tool chain test: OK
179 - looking in '/toolchains/powerpc-linux/usr/bin'
180 - scanning path '/toolchains/nds32le-linux-glibc-v1f'
181 - looking in '/toolchains/nds32le-linux-glibc-v1f/.'
182 - looking in '/toolchains/nds32le-linux-glibc-v1f/bin'
183 - found '/toolchains/nds32le-linux-glibc-v1f/bin/nds32le-linux-gcc'
184Tool chain test: OK
185 - looking in '/toolchains/nds32le-linux-glibc-v1f/usr/bin'
186 - scanning path '/toolchains/nios2'
187 - looking in '/toolchains/nios2/.'
188 - looking in '/toolchains/nios2/bin'
189 - found '/toolchains/nios2/bin/nios2-linux-gcc'
190Tool chain test: OK
191 - found '/toolchains/nios2/bin/nios2-linux-uclibc-gcc'
192Tool chain test: OK
193 - looking in '/toolchains/nios2/usr/bin'
194 - found '/toolchains/nios2/usr/bin/nios2-linux-gcc'
195Tool chain test: OK
196 - found '/toolchains/nios2/usr/bin/nios2-linux-uclibc-gcc'
197Tool chain test: OK
198 - scanning path '/toolchains/microblaze-unknown-linux-gnu'
199 - looking in '/toolchains/microblaze-unknown-linux-gnu/.'
200 - looking in '/toolchains/microblaze-unknown-linux-gnu/bin'
201 - found '/toolchains/microblaze-unknown-linux-gnu/bin/microblaze-unknown-linux-gnu-gcc'
202Tool chain test: OK
203 - found '/toolchains/microblaze-unknown-linux-gnu/bin/mb-linux-gcc'
204Tool chain test: OK
205 - looking in '/toolchains/microblaze-unknown-linux-gnu/usr/bin'
206 - scanning path '/toolchains/mips-linux'
207 - looking in '/toolchains/mips-linux/.'
208 - looking in '/toolchains/mips-linux/bin'
209 - found '/toolchains/mips-linux/bin/mips-linux-gcc'
210Tool chain test: OK
211 - looking in '/toolchains/mips-linux/usr/bin'
212 - scanning path '/toolchains/old'
213 - looking in '/toolchains/old/.'
214 - looking in '/toolchains/old/bin'
215 - looking in '/toolchains/old/usr/bin'
216 - scanning path '/toolchains/i386-linux'
217 - looking in '/toolchains/i386-linux/.'
218 - looking in '/toolchains/i386-linux/bin'
219 - found '/toolchains/i386-linux/bin/i386-linux-gcc'
220Tool chain test: OK
221 - looking in '/toolchains/i386-linux/usr/bin'
222 - scanning path '/toolchains/bfin-uclinux'
223 - looking in '/toolchains/bfin-uclinux/.'
224 - looking in '/toolchains/bfin-uclinux/bin'
225 - found '/toolchains/bfin-uclinux/bin/bfin-uclinux-gcc'
226Tool chain test: OK
227 - looking in '/toolchains/bfin-uclinux/usr/bin'
228 - scanning path '/toolchains/sparc-elf'
229 - looking in '/toolchains/sparc-elf/.'
230 - looking in '/toolchains/sparc-elf/bin'
231 - found '/toolchains/sparc-elf/bin/sparc-elf-gcc'
232Tool chain test: OK
233 - looking in '/toolchains/sparc-elf/usr/bin'
234 - scanning path '/toolchains/arm-2010q1'
235 - looking in '/toolchains/arm-2010q1/.'
236 - looking in '/toolchains/arm-2010q1/bin'
237 - found '/toolchains/arm-2010q1/bin/arm-none-linux-gnueabi-gcc'
238Tool chain test: OK
239 - looking in '/toolchains/arm-2010q1/usr/bin'
240 - scanning path '/toolchains/from'
241 - looking in '/toolchains/from/.'
242 - looking in '/toolchains/from/bin'
243 - looking in '/toolchains/from/usr/bin'
244 - scanning path '/toolchains/sh4-gentoo-linux-gnu'
245 - looking in '/toolchains/sh4-gentoo-linux-gnu/.'
246 - looking in '/toolchains/sh4-gentoo-linux-gnu/bin'
247 - found '/toolchains/sh4-gentoo-linux-gnu/bin/sh4-gentoo-linux-gnu-gcc'
248Tool chain test: OK
249 - looking in '/toolchains/sh4-gentoo-linux-gnu/usr/bin'
250 - scanning path '/toolchains/avr32-linux'
251 - looking in '/toolchains/avr32-linux/.'
252 - looking in '/toolchains/avr32-linux/bin'
253 - found '/toolchains/avr32-linux/bin/avr32-gcc'
254Tool chain test: OK
255 - looking in '/toolchains/avr32-linux/usr/bin'
256 - scanning path '/toolchains/m68k-linux'
257 - looking in '/toolchains/m68k-linux/.'
258 - looking in '/toolchains/m68k-linux/bin'
259 - found '/toolchains/m68k-linux/bin/m68k-linux-gcc'
260Tool chain test: OK
261 - looking in '/toolchains/m68k-linux/usr/bin'
262List of available toolchains (17):
263arm : /toolchains/arm-2010q1/bin/arm-none-linux-gnueabi-gcc
264avr32 : /toolchains/avr32-linux/bin/avr32-gcc
265bfin : /toolchains/bfin-uclinux/bin/bfin-uclinux-gcc
266c89 : /usr/bin/c89-gcc
267c99 : /usr/bin/c99-gcc
268i386 : /toolchains/i386-linux/bin/i386-linux-gcc
269m68k : /toolchains/m68k-linux/bin/m68k-linux-gcc
270mb : /toolchains/microblaze-unknown-linux-gnu/bin/mb-linux-gcc
271microblaze: /toolchains/microblaze-unknown-linux-gnu/bin/microblaze-unknown-linux-gnu-gcc
272mips : /toolchains/mips-linux/bin/mips-linux-gcc
273nds32le : /toolchains/nds32le-linux-glibc-v1f/bin/nds32le-linux-gcc
274nios2 : /toolchains/nios2/bin/nios2-linux-gcc
275powerpc : /toolchains/powerpc-linux/bin/powerpc-linux-gcc
276sandbox : /usr/bin/gcc
277sh4 : /toolchains/sh4-gentoo-linux-gnu/bin/sh4-gentoo-linux-gnu-gcc
278sparc : /toolchains/sparc-elf/bin/sparc-elf-gcc
279x86_64 : /usr/bin/x86_64-linux-gnu-gcc
280
281
282You can see that everything is covered, even some strange ones that won't
283be used (c88 and c99). This is a feature.
284
285
286How to run it
287=============
288
289First do a dry run using the -n flag: (replace <branch> with a real, local
290branch with a valid upstream)
291
292$ ./tools/buildman/buildman -b <branch> -n
293
294If it can't detect the upstream branch, try checking out the branch, and
295doing something like 'git branch --set-upstream <branch> upstream/master'
296or something similar.
297
Simon Glasscec83c32014-08-09 15:32:57 -0600298As an example:
Simon Glassfc3fe1c2013-04-03 11:07:16 +0000299
300Dry run, so not doing much. But I would do this:
301
302Building 18 commits for 1059 boards (4 threads, 1 job per thread)
303Build directory: ../lcd9b
304 5bb3505 Merge branch 'master' of git://git.denx.de/u-boot-arm
305 c18f1b4 tegra: Use const for pinmux_config_pingroup/table()
306 2f043ae tegra: Add display support to funcmux
307 e349900 tegra: fdt: Add pwm binding and node
308 424a5f0 tegra: fdt: Add LCD definitions for Tegra
309 0636ccf tegra: Add support for PWM
310 a994fe7 tegra: Add SOC support for display/lcd
311 fcd7350 tegra: Add LCD driver
312 4d46e9d tegra: Add LCD support to Nvidia boards
313 991bd48 arm: Add control over cachability of memory regions
314 54e8019 lcd: Add CONFIG_LCD_ALIGNMENT to select frame buffer alignment
315 d92aff7 lcd: Add support for flushing LCD fb from dcache after update
316 dbd0677 tegra: Align LCD frame buffer to section boundary
317 0cff9b8 tegra: Support control of cache settings for LCD
318 9c56900 tegra: fdt: Add LCD definitions for Seaboard
319 5cc29db lcd: Add CONFIG_CONSOLE_SCROLL_LINES option to speed console
320 cac5a23 tegra: Enable display/lcd support on Seaboard
321 49ff541 wip
322
323Total boards to build for each commit: 1059
324
325This shows that it will build all 1059 boards, using 4 threads (because
326we have a 4-core CPU). Each thread will run with -j1, meaning that each
327make job will use a single CPU. The list of commits to be built helps you
328confirm that things look about right. Notice that buildman has chosen a
329'base' directory for you, immediately above your source tree.
330
331Buildman works entirely inside the base directory, here ../lcd9b,
332creating a working directory for each thread, and creating output
333directories for each commit and board.
334
335
336Suggested Workflow
337==================
338
339To run the build for real, take off the -n:
340
341$ ./tools/buildman/buildman -b <branch>
342
343Buildman will set up some working directories, and get started. After a
344minute or so it will settle down to a steady pace, with a display like this:
345
346Building 18 commits for 1059 boards (4 threads, 1 job per thread)
347 528 36 124 /19062 1:13:30 : SIMPC8313_SP
348
349This means that it is building 19062 board/commit combinations. So far it
Simon Glasscec83c32014-08-09 15:32:57 -0600350has managed to successfully build 528. Another 36 have built with warnings,
Simon Glassfc3fe1c2013-04-03 11:07:16 +0000351and 124 more didn't build at all. Buildman expects to complete the process
352in an hour and 15 minutes. Use this time to buy a faster computer.
353
354
355To find out how the build went, ask for a summary with -s. You can do this
356either before the build completes (presumably in another terminal) or or
357afterwards. Let's work through an example of how this is used:
358
359$ ./tools/buildman/buildman -b lcd9b -s
360...
36101: Merge branch 'master' of git://git.denx.de/u-boot-arm
362 powerpc: + galaxy5200_LOWBOOT
36302: tegra: Use const for pinmux_config_pingroup/table()
36403: tegra: Add display support to funcmux
36504: tegra: fdt: Add pwm binding and node
36605: tegra: fdt: Add LCD definitions for Tegra
36706: tegra: Add support for PWM
36807: tegra: Add SOC support for display/lcd
36908: tegra: Add LCD driver
37009: tegra: Add LCD support to Nvidia boards
37110: arm: Add control over cachability of memory regions
37211: lcd: Add CONFIG_LCD_ALIGNMENT to select frame buffer alignment
37312: lcd: Add support for flushing LCD fb from dcache after update
374 arm: + lubbock
37513: tegra: Align LCD frame buffer to section boundary
37614: tegra: Support control of cache settings for LCD
37715: tegra: fdt: Add LCD definitions for Seaboard
37816: lcd: Add CONFIG_CONSOLE_SCROLL_LINES option to speed console
37917: tegra: Enable display/lcd support on Seaboard
38018: wip
381
382This shows which commits have succeeded and which have failed. In this case
383the build is still in progress so many boards are not built yet (use -u to
384see which ones). But still we can see a few failures. The galaxy5200_LOWBOOT
385never builds correctly. This could be a problem with our toolchain, or it
386could be a bug in the upstream. The good news is that we probably don't need
387to blame our commits. The bad news is it isn't tested on that board.
388
389Commit 12 broke lubbock. That's what the '+ lubbock' means. The failure
390is never fixed by a later commit, or you would see lubbock again, in green,
391without the +.
392
393To see the actual error:
394
395$ ./tools/buildman/buildman -b <branch> -se lubbock
396...
39712: lcd: Add support for flushing LCD fb from dcache after update
398 arm: + lubbock
399+common/libcommon.o: In function `lcd_sync':
400+/u-boot/lcd9b/.bm-work/00/common/lcd.c:120: undefined reference to `flush_dcache_range'
401+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
402+make: *** [/u-boot/lcd9b/.bm-work/00/build/u-boot] Error 139
40313: tegra: Align LCD frame buffer to section boundary
40414: tegra: Support control of cache settings for LCD
40515: tegra: fdt: Add LCD definitions for Seaboard
40616: lcd: Add CONFIG_CONSOLE_SCROLL_LINES option to speed console
407-/u-boot/lcd9b/.bm-work/00/common/lcd.c:120: undefined reference to `flush_dcache_range'
408+/u-boot/lcd9b/.bm-work/00/common/lcd.c:125: undefined reference to `flush_dcache_range'
40917: tegra: Enable display/lcd support on Seaboard
41018: wip
411
412So the problem is in lcd.c, due to missing cache operations. This information
413should be enough to work out what that commit is doing to break these
414boards. (In this case pxa did not have cache operations defined).
415
416If you see error lines marked with - that means that the errors were fixed
417by that commit. Sometimes commits can be in the wrong order, so that a
418breakage is introduced for a few commits and fixed by later commits. This
419shows up clearly with buildman. You can then reorder the commits and try
420again.
421
422At commit 16, the error moves - you can see that the old error at line 120
423is fixed, but there is a new one at line 126. This is probably only because
Simon Glasscec83c32014-08-09 15:32:57 -0600424we added some code and moved the broken line father down the file.
Simon Glassfc3fe1c2013-04-03 11:07:16 +0000425
426If many boards have the same error, then -e will display the error only
427once. This makes the output as concise as possible.
428
429The full build output in this case is available in:
430
431../lcd9b/12_of_18_gd92aff7_lcd--Add-support-for/lubbock/
432
433 done: Indicates the build was done, and holds the return code from make.
434 This is 0 for a good build, typically 2 for a failure.
435
436 err: Output from stderr, if any. Errors and warnings appear here.
437
438 log: Output from stdout. Normally there isn't any since buildman runs
439 in silent mode for now.
440
441 toolchain: Shows information about the toolchain used for the build.
442
443 sizes: Shows image size information.
444
445It is possible to get the build output there also. Use the -k option for
446this. In that case you will also see some output files, like:
447
448 System.map toolchain u-boot u-boot.bin u-boot.map autoconf.mk
449 (also SPL versions u-boot-spl and u-boot-spl.bin if available)
450
451
452Checking Image Sizes
453====================
454
455A key requirement for U-Boot is that you keep code/data size to a minimum.
456Where a new feature increases this noticeably it should normally be put
457behind a CONFIG flag so that boards can leave it off and keep the image
458size more or less the same with each new release.
459
460To check the impact of your commits on image size, use -S. For example:
461
462$ ./tools/buildman/buildman -b us-x86 -sS
463Summary of 10 commits for 1066 boards (4 threads, 1 job per thread)
46401: MAKEALL: add support for per architecture toolchains
46502: x86: Add function to get top of usable ram
466 x86: (for 1/3 boards) text -272.0 rodata +41.0
46703: x86: Add basic cache operations
46804: x86: Permit bootstage and timer data to be used prior to relocation
469 x86: (for 1/3 boards) data +16.0
47005: x86: Add an __end symbol to signal the end of the U-Boot binary
471 x86: (for 1/3 boards) text +76.0
47206: x86: Rearrange the output input to remove BSS
473 x86: (for 1/3 boards) bss -2140.0
47407: x86: Support relocation of FDT on start-up
475 x86: + coreboot-x86
47608: x86: Add error checking to x86 relocation code
47709: x86: Adjust link device tree include file
47810: x86: Enable CONFIG_OF_CONTROL on coreboot
479
480
481You can see that image size only changed on x86, which is good because this
482series is not supposed to change any other board. From commit 7 onwards the
483build fails so we don't get code size numbers. The numbers are fractional
484because they are an average of all boards for that architecture. The
485intention is to allow you to quickly find image size problems introduced by
486your commits.
487
488Note that the 'text' region and 'rodata' are split out. You should add the
489two together to get the total read-only size (reported as the first column
490in the output from binutil's 'size' utility).
491
492A useful option is --step which lets you skip some commits. For example
493--step 2 will show the image sizes for only every 2nd commit (so it will
494compare the image sizes of the 1st, 3rd, 5th... commits). You can also use
495--step 0 which will compare only the first and last commits. This is useful
496for an overview of how your entire series affects code size.
497
498You can also use -d to see a detailed size breakdown for each board. This
499list is sorted in order from largest growth to largest reduction.
500
501It is possible to go a little further with the -B option (--bloat). This
Simon Glasscec83c32014-08-09 15:32:57 -0600502shows where U-Boot has bloated, breaking the size change down to the function
Simon Glassfc3fe1c2013-04-03 11:07:16 +0000503level. Example output is below:
504
505$ ./tools/buildman/buildman -b us-mem4 -sSdB
506...
50719: Roll crc32 into hash infrastructure
508 arm: (for 10/10 boards) all -143.4 bss +1.2 data -4.8 rodata -48.2 text -91.6
509 paz00 : all +23 bss -4 rodata -29 text +56
510 u-boot: add: 1/0, grow: 3/-2 bytes: 168/-104 (64)
511 function old new delta
512 hash_command 80 160 +80
513 crc32_wd_buf - 56 +56
514 ext4fs_read_file 540 568 +28
515 insert_var_value_sub 688 692 +4
516 run_list_real 1996 1992 -4
517 do_mem_crc 168 68 -100
518 trimslice : all -9 bss +16 rodata -29 text +4
519 u-boot: add: 1/0, grow: 1/-3 bytes: 136/-124 (12)
520 function old new delta
521 hash_command 80 160 +80
522 crc32_wd_buf - 56 +56
523 ext4fs_iterate_dir 672 668 -4
524 ext4fs_read_file 568 548 -20
525 do_mem_crc 168 68 -100
526 whistler : all -9 bss +16 rodata -29 text +4
527 u-boot: add: 1/0, grow: 1/-3 bytes: 136/-124 (12)
528 function old new delta
529 hash_command 80 160 +80
530 crc32_wd_buf - 56 +56
531 ext4fs_iterate_dir 672 668 -4
532 ext4fs_read_file 568 548 -20
533 do_mem_crc 168 68 -100
534 seaboard : all -9 bss -28 rodata -29 text +48
535 u-boot: add: 1/0, grow: 3/-2 bytes: 160/-104 (56)
536 function old new delta
537 hash_command 80 160 +80
538 crc32_wd_buf - 56 +56
539 ext4fs_read_file 548 568 +20
540 run_list_real 1996 2000 +4
541 do_nandboot 760 756 -4
542 do_mem_crc 168 68 -100
543 colibri_t20_iris: all -9 rodata -29 text +20
544 u-boot: add: 1/0, grow: 2/-3 bytes: 140/-112 (28)
545 function old new delta
546 hash_command 80 160 +80
547 crc32_wd_buf - 56 +56
548 read_abs_bbt 204 208 +4
549 do_nandboot 760 756 -4
550 ext4fs_read_file 576 568 -8
551 do_mem_crc 168 68 -100
552 ventana : all -37 bss -12 rodata -29 text +4
553 u-boot: add: 1/0, grow: 1/-3 bytes: 136/-124 (12)
554 function old new delta
555 hash_command 80 160 +80
556 crc32_wd_buf - 56 +56
557 ext4fs_iterate_dir 672 668 -4
558 ext4fs_read_file 568 548 -20
559 do_mem_crc 168 68 -100
560 harmony : all -37 bss -16 rodata -29 text +8
561 u-boot: add: 1/0, grow: 2/-3 bytes: 140/-124 (16)
562 function old new delta
563 hash_command 80 160 +80
564 crc32_wd_buf - 56 +56
565 nand_write_oob_syndrome 428 432 +4
566 ext4fs_iterate_dir 672 668 -4
567 ext4fs_read_file 568 548 -20
568 do_mem_crc 168 68 -100
569 medcom-wide : all -417 bss +28 data -16 rodata -93 text -336
570 u-boot: add: 1/-1, grow: 1/-2 bytes: 88/-376 (-288)
571 function old new delta
572 crc32_wd_buf - 56 +56
573 do_fat_read_at 2872 2904 +32
574 hash_algo 16 - -16
575 do_mem_crc 168 68 -100
576 hash_command 420 160 -260
577 tec : all -449 bss -4 data -16 rodata -93 text -336
578 u-boot: add: 1/-1, grow: 1/-2 bytes: 88/-376 (-288)
579 function old new delta
580 crc32_wd_buf - 56 +56
581 do_fat_read_at 2872 2904 +32
582 hash_algo 16 - -16
583 do_mem_crc 168 68 -100
584 hash_command 420 160 -260
585 plutux : all -481 bss +16 data -16 rodata -93 text -388
586 u-boot: add: 1/-1, grow: 1/-3 bytes: 68/-408 (-340)
587 function old new delta
588 crc32_wd_buf - 56 +56
589 do_load_serial_bin 1688 1700 +12
590 hash_algo 16 - -16
591 do_fat_read_at 2904 2872 -32
592 do_mem_crc 168 68 -100
593 hash_command 420 160 -260
594 powerpc: (for 5/5 boards) all +37.4 data -3.2 rodata -41.8 text +82.4
595 MPC8610HPCD : all +55 rodata -29 text +84
596 u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80)
597 function old new delta
598 hash_command - 176 +176
599 do_mem_crc 184 88 -96
600 MPC8641HPCN : all +55 rodata -29 text +84
601 u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80)
602 function old new delta
603 hash_command - 176 +176
604 do_mem_crc 184 88 -96
605 MPC8641HPCN_36BIT: all +55 rodata -29 text +84
606 u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80)
607 function old new delta
608 hash_command - 176 +176
609 do_mem_crc 184 88 -96
610 sbc8641d : all +55 rodata -29 text +84
611 u-boot: add: 1/0, grow: 0/-1 bytes: 176/-96 (80)
612 function old new delta
613 hash_command - 176 +176
614 do_mem_crc 184 88 -96
615 xpedite517x : all -33 data -16 rodata -93 text +76
616 u-boot: add: 1/-1, grow: 0/-1 bytes: 176/-112 (64)
617 function old new delta
618 hash_command - 176 +176
619 hash_algo 16 - -16
620 do_mem_crc 184 88 -96
621...
622
623
624This shows that commit 19 has increased text size for arm (although only one
625board was built) and by 96 bytes for powerpc. This increase was offset in both
626cases by reductions in rodata and data/bss.
627
628Shown below the summary lines is the sizes for each board. Below each board
629is the sizes for each function. This information starts with:
630
631 add - number of functions added / removed
632 grow - number of functions which grew / shrunk
633 bytes - number of bytes of code added to / removed from all functions,
634 plus the total byte change in brackets
635
636The change seems to be that hash_command() has increased by more than the
637do_mem_crc() function has decreased. The function sizes typically add up to
638roughly the text area size, but note that every read-only section except
639rodata is included in 'text', so the function total does not exactly
640correspond.
641
642It is common when refactoring code for the rodata to decrease as the text size
643increases, and vice versa.
644
645
Simon Glass4281ad82013-09-23 17:35:17 -0600646Providing 'make' flags
647======================
648
649U-Boot's build system supports a few flags (such as BUILD_TAG) which affect
650the build product. These flags can be specified in the buildman settings
651file. They can also be useful when building U-Boot against other open source
652software.
653
654[make-flags]
655at91-boards=ENABLE_AT91_TEST=1
656snapper9260=${at91-boards} BUILD_TAG=442
657snapper9g45=${at91-boards} BUILD_TAG=443
658
659This will use 'make ENABLE_AT91_TEST=1 BUILD_TAG=442' for snapper9260
Andreas Bießmann61242ac2013-11-05 10:37:09 +0100660and 'make ENABLE_AT91_TEST=1 BUILD_TAG=443' for snapper9g45. A special
Simon Glass4281ad82013-09-23 17:35:17 -0600661variable ${target} is available to access the target name (snapper9260 and
662snapper9g20 in this case). Variables are resolved recursively.
663
664It is expected that any variables added are dealt with in U-Boot's
665config.mk file and documented in the README.
666
667
Simon Glasse5a0e5d2014-08-09 15:33:03 -0600668Quick Sanity Check
669==================
670
671If you have made changes and want to do a quick sanity check of the
672currently-checked-out source, run buildman without the -b flag. This will
673build the selected boards and display build status and errors as it runs
674(i.e. -v amd -e are enabled automatically).
675
676
Simon Glassfc3fe1c2013-04-03 11:07:16 +0000677Other options
678=============
679
680Buildman has various other command line options. Try --help to see them.
681
682
Simon Glass6eede342014-08-09 15:32:58 -0600683How to change from MAKEALL
684==========================
685
686Buildman includes most of the features of MAKEALL and is generally faster
687and easier to use. In particular it builds entire branches: if a particular
688commit introduces an error in a particular board, buildman can easily show
689you this, even if a later commit fixes that error.
690
691The reasons to deprecate MAKEALL are:
692- We don't want to maintain two build systems
693- Buildman is typically faster
694- Buildman has a lot more features
695
696But still, many people will be sad to lose MAKEALL. If you are used to
697MAKEALL, here are a few pointers.
698
699First you need to set up your tool chains - see the 'Setting up' section
700for details. Once you have your required toolchain(s) detected then you are
701ready to go.
702
Simon Glasse5a0e5d2014-08-09 15:33:03 -0600703To build the current source tree, run buildman without a -b flag:
704
705 ./tools/buildman/buildman <list of things to build>
706
707This will build the current source tree for the given boards and display
708the results and errors.
709
710However buildman usually works on entire branches, and for that you must
711specify a board flag:
Simon Glass6eede342014-08-09 15:32:58 -0600712
713 ./tools/buildman/buildman -b <branch_name> <list of things to build>
714
715followed by (afterwards, or perhaps concurrently in another terminal):
716
717 ./tools/buildman/buildman -b <branch_name> -s <list of things to build>
718
719to see the results of the build. Rather than showing you all the output,
720buildman just shows a summary, with red indicating that a commit introduced
721an error and green indicating that a commit fixed an error. Use the -e
722flag to see the full errors.
723
Simon Glasse5a0e5d2014-08-09 15:33:03 -0600724If you really want to see build results as they happen, use -v when doing a
725build (and -e if you want to see errors as well).
726
Simon Glass6eede342014-08-09 15:32:58 -0600727You don't need to stick around on that branch while buildman is running. It
728checks out its own copy of the source code, so you can change branches,
729add commits, etc. without affecting the build in progress.
730
731The <list of things to build> can include board names, architectures or the
732like. There are no flags to disambiguate since ambiguities are rare. Using
733the examples from MAKEALL:
734
735Examples:
736 - build all Power Architecture boards:
737 MAKEALL -a powerpc
738 MAKEALL --arch powerpc
739 MAKEALL powerpc
740 ** buildman -b <branch> powerpc
741 - build all PowerPC boards manufactured by vendor "esd":
742 MAKEALL -a powerpc -v esd
743 ** buildman -b <branch> esd
744 - build all PowerPC boards manufactured either by "keymile" or "siemens":
745 MAKEALL -a powerpc -v keymile -v siemens
746 ** buildman -b <branch> keymile siemens
747 - build all Freescale boards with MPC83xx CPUs, plus all 4xx boards:
748 MAKEALL -c mpc83xx -v freescale 4xx
749 ** buildman -b <branch> mpc83xx freescale 4xx
750
751Buildman automatically tries to use all the CPUs in your machine. If you
752are building a lot of boards it will use one thread for every CPU core
753it detects in your machine. This is like MAKEALL's BUILD_NBUILDS option.
754You can use the -T flag to change the number of threads. If you are only
755building a few boards, buildman will automatically run make with the -j
756flag to increase the number of concurrent make tasks. It isn't normally
757that helpful to fiddle with this option, but if you use the BUILD_NCPUS
758option in MAKEALL then -j is the equivalent in buildman.
759
760Buildman puts its output in ../<branch_name> by default but you can change
761this with the -o option. Buildman normally does out-of-tree builds: use -i
762to disable that if you really want to. But be careful that once you have
763used -i you pollute buildman's copies of the source tree, and you will need
764to remove the build directory (normally ../<branch_name>) to run buildman
765in normal mode (without -i).
766
767Buildman doesn't keep the output result normally, but use the -k option to
768do this.
769
770Please read 'Theory of Operation' a few times as it will make a lot of
771things clearer.
772
773Some options you might like are:
774
775 -B shows which functions are growing/shrinking in which commit - great
776 for finding code bloat.
777 -S shows image sizes for each commit (just an overall summary)
778 -u shows boards that you haven't built yet
779 --step 0 will build just the upstream commit and the last commit of your
780 branch. This is often a quick sanity check that your branch doesn't
781 break anything. But note this does not check bisectability!
782
783
Simon Glassfc3fe1c2013-04-03 11:07:16 +0000784TODO
785====
786
787This has mostly be written in my spare time as a response to my difficulties
788in testing large series of patches. Apart from tidying up there is quite a
789bit of scope for improvement. Things like better error diffs, easier access
790to log files, error display while building. Also it would be nice it buildman
791could 'hunt' for problems, perhaps by building a few boards for each arch,
792or checking commits for changed files and building only boards which use
793those files.
794
795
796Credits
797=======
798
799Thanks to Grant Grundler <grundler@chromium.org> for his ideas for improving
800the build speed by building all commits for a board instead of the other
801way around.
802
803
Simon Glassfc3fe1c2013-04-03 11:07:16 +0000804Simon Glass
805sjg@chromium.org
806Halloween 2012
807Updated 12-12-12
808Updated 23-02-13