Buildman now uses worktrees when available, instead of doing a full clone.
This was done in this commit:
76de29fc4f buildman: Use git worktrees instead of git clones when possible
Drop the TODO.
Signed-off-by: Simon Glass <sjg@chromium.org>
The 'nm' tool can produce lines without a symbol, for example:
00000004 t
Silently skip these and anything else without three fields. Drop the
warning since there is nothing the user can do about it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Tom Rini <trini@konsulko.com>
This appears in boards.cfg but we want to remove it. Drop support for
generating it and reading it. Detect an old boards.cfg file that has
this field and regenerate it, to avoid problems.
Instead, add the config name in that place. This fixes a subtle bug in
the generation code, since it uses 'target' for the config name and then
overwrites the value in scan() by setting params['target'] to the name
of the defconfig. The defconfig name is not the same as the
SYS_CONFIG_NAME variable.
With this change, we still have the config name and it can be searched
by buildman, e.g. with:
buildman -nv sun5i
Signed-off-by: Simon Glass <sjg@chromium.org>
Reported-by: Tom Rini <trini@konsulko.com>
Bring this tool into buildman, so we don't have to run it separately. The
board.cfg file is still produced as part of the build, to save time when
doing another build in the same working directory. If it is out of date
with respect to the Kconfig, it is updated.
Time to regenerate on a recent single-thread machine is 4.6s (1.3s on a
32-thread machine), so we do need some sort of cache if we want buildman
to be useful on incremental builds. We could use Python's pickle format
but:
- it seems useful to allow boards.cfg to be regenerated, at least for a
while, in case other tools use it
- it is possible to grep the file easily, e.g. to find boards which use
a particular SoC (similar to 'buildman -nv <soc>'
Signed-off-by: Simon Glass <sjg@chromium.org>
Suggested-by: Tom Rini <trini@konsulko.com>
Use a separate file for the Boards class so that its name matches the
module name.
Fix up the function names to match the pylint style and fix some other
warnings.
Signed-off-by: Simon Glass <sjg@chromium.org>
We want to create a module called 'boards' so avoid use of this variable
name in this module. Change the global to be capitalised, as required by
Python style.
Signed-off-by: Simon Glass <sjg@chromium.org>
We want to create a module called 'boards' so avoid use of this variable
name in this module. Change the global to be capitalised, as required by
Python style.
Signed-off-by: Simon Glass <sjg@chromium.org>
We have a module called 'board'. Sometimes buildman uses 'brd' as an
instance variable but sometimes it uses 'board', which is confusing and
can mess with the module handling. Update the code to use 'brd'
consistently, making it easier for tools to determine when the module
is being referenced.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add a flag to allow buildman to behave properly for use from an IDE. This
shows error/warning output on stderr and drops all summary and progress
information.
This should normally only be used when building a single board.
Fix up a confusing comment for GetResultSummary() while we are here, since
we want to use the Outcome object to access the unprocessed error lines
from the build.
Signed-off-by: Simon Glass <sjg@chromium.org>
The python tools' test utilities handle printing test results, but the
output is quite bare compared to an ordinary unittest run. Delegate
printing the results to a unittest text runner, which gives us niceties
like clear separation between each test's result and how long it took to
run the test suite.
Unfortunately it does not print info for skipped tests by default, but
this can be handled later by a custom test result subclass. It also does
not print the tool name; manually print a heading that includes the
toolname so that the outputs of each tool's tests are distinguishable in
the CI output.
Signed-off-by: Alper Nebi Yasak <alpernebiyasak@gmail.com>
Reviewed-by: Simon Glass <sjg@chromium.org>
As removal of nds32 has been ack'd for the Linux kernel, remove support
here as well.
Cc: Rick Chen <rick@andestech.com>
Signed-off-by: Tom Rini <trini@konsulko.com>
Reviewed-by: Rick Chen <rick@andestech.com>
Fix documentation path in deprecated warning message about device
driver.
Signed-off-by: Johannes Krottmayer <krjdev@gmail.com>
Reviewed-by: Heinrich Schuchardt <heinrich.schuchardt@canonical.com>
At present the default .buildman file written by buildman does not specify
a default toolchain. Add an 'other' line so this works correctly and
sandbox builds run as expected.
Signed-off-by: Simon Glass <sjg@chromium.org>
This method has the same name as its class which is confusing. It is also
annoying when searching the code.
It builds a string with a colour, so rename it to build().
Signed-off-by: Simon Glass <sjg@chromium.org>
Rename this function so that when we convert it to snake case it will not
conflict with the built-in print() function.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add a -a option to specify changes to the config before the build
commences. For example
buildman -a ~CONFIG_CMDLINE
disables CONFIG_CMDLINE before doing the build.
This makes it easier to try things out as well as to write tests without
creating a new board or manually manging the .config file.
Signed-off-by: Simon Glass <sjg@chromium.org>
If a thread crashes it is helpful to try the operation again with
threading disabled. Add a hint about that.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present the only straightforward way to write tests that need a
slightly different configuration is to create a new board with its own
configuration. This is cumbersome.
It would be useful if buildman could adjust the configuration of a build
on the fly. In preparation for this, add a utility library which can
modify a .config file according to various parameters passed to it.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present the full horror of the Python traceback is shown by default. It
is normally only useful for debugging. Turn it off by default and add a
--debug flag to enable it.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reverse the order of the return tuple, so that the filename is first.
This seems more obvious than putting the temporary directory first.
Correct a bug that leaves a space on the final line.
Allow the caller to control the name of the temporary directory.
Signed-off-by: Simon Glass <sjg@chromium.org>
Add some empty __init__ files for binman, buildman and dtoc so that
pylint is able to recognise these as Python modules and produce more
useful pylint output.
Signed-off-by: Simon Glass <sjg@chromium.org>
While CI has been using gcc-11.1.0 for a long time, we have not updated
buildman to match. Correct this omission.
Signed-off-by: Tom Rini <trini@konsulko.com>
Hex and int Kconfig options are supposed to have defaults. This is so we
can configure U-Boot without having to enter particular values for the
items that don't have specific values in the board's defconfig file.
If this rule is not followed, then introducing a new Kconfig can produce
a loop like this:
Break things (BREAK_ME) [] (NEW)
Error in reading or end of file.
Break things (BREAK_ME) [] (NEW)
Error in reading or end of file.
The continues forever since buildman passes /dev/null to 'conf', and
the build system just tries again. Eventually there is so much output that
buildman runs out of memory.
We can detect this situation by looking for a symbol (like 'BREAK_ME')
which has no default (the '[]' above) and is marked as new. If this
appears multiple times in the output, we know something is wrong.
Add a filter function for the output which detects this situation. Allow
it to return True to terminate the process. Implement this termination in
cros_subprocess.
With this we get a nice message:
buildman --board sandbox -T0
Building current source for 1 boards (0 threads, 32 jobs per thread)
sandbox: w+ sandbox
+.config:66:warning: symbol value '' invalid for BREAK_ME
+
+Error in reading or end of file.
+make[3]: *** [scripts/kconfig/Makefile:75: syncconfig] Terminated
+make[2]: *** [Makefile:569: syncconfig] Terminated
+make: *** [Makefile:177: sub-make] Terminated
+(** did you define an int/hex Kconfig with no default? **)
Signed-off-by: Simon Glass <sjg@chromium.org>
At present buildman does not write any output (to the 'out' and 'err)
files if the build terminates with a fatal error. This is to avoid adding
lots of spam to the logs.
However there are times when this is actually useful, such as when the
build fails for an obscure reason such as a Kconfig loop.
Update the logic to always write the output, so that the user gets a clue
as to what is happening.
Signed-off-by: Simon Glass <sjg@chromium.org>
Collect the code for printing the full help message of patman, buildman
and binman into a single function in patman.tools.
Signed-off-by: Paul Barker <paul.barker@sancloud.com>
Rename these options so that CONFIG_IS_ENABLED can be used with them.
Signed-off-by: Simon Glass <sjg@chromium.org>
Reviewed-by: Jaehoon Chung <jh80.chung@samsung.com>
[trini: Fixup some incorrect renames]
Signed-off-by: Tom Rini <trini@konsulko.com>
At present we sometimes see problems in gitlab where the environment has
0x80 characters or sequences which are not valid UTF-8.
Avoid this by using bytes for the environment, both internal to buildman
and when writing out the 'env' file. Add a test to make sure this works
as expected.
Reported-by: Marek Vasut <marex@denx.de>
Fixes: e5fc79ea71 ("buildman: Write the environment out to an 'env' file")
Signed-off-by: Simon Glass <sjg@chromium.org>
There have been at least a few cases where an exception has occurred in a
thread and resulted in buildman hanging: running out of disk space and
getting a unicode error.
Handle these by collecting a list of exceptions, printing them out and
reporting failure if any are found. Add a test for this.
Signed-off-by: Simon Glass <sjg@chromium.org>
The environment may contain some unicode characters. At least that is what
seemed to happen on one commit:
Building current source for 1 boards (0 threads, 64 jobs per thread)
0 0 0 /1 -1 (starting)
Traceback (most recent call last):
File ".../tools/buildman/buildman", line 64, in <module>
ret_code = control.DoBuildman(options, args)
File "tools/buildman/control.py", line 372, in DoBuildman
options.keep_outputs, options.verbose)
File ".../tools/buildman/builder.py", line 1704, in BuildBoards
results = self._single_builder.RunJob(job)
File ".../tools/buildman/builderthread.py", line 526, in RunJob
self._WriteResult(result, job.keep_outputs, job.work_in_output)
File ".../tools//buildman/builderthread.py", line 349, in _WriteResult
print('%s="%s"' % (var, env[var]), file=fd)
UnicodeEncodeError: 'ascii' codec can't encode characters in position
311-312: ordinal not in range(128)
The problem defies repetition with any change at all to buildman. But
let's set an encoding in any case.
Signed-off-by: Simon Glass <sjg@chromium.org>
At present even if only a single thread is in use, buildman still uses
threading.
For some debugging it is helpful to do everything in the main process.
Allow -T0 to support this.
Signed-off-by: Simon Glass <sjg@chromium.org>